Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. Integrating Economics and Ecology: A Systems Approach to Sustainability in the Auckland Region A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy m Ecological Economics at Massey University, Palmerston North, New Zealand Garry McDonald 2005 III Abstract Urban sustainability has emerged as a central environmental and urban policy issue over the last decade, as our world becomes more urbanised. The purpose of this thesis is to operationalise systems modelling approaches (static, system dynamics) that will lead to improved understanding of urban sustainability in the Auckland Region. The first part of the thesis critically reviews and synthesises both the sustainability and urban development literatures. Consideration of the sustainability literature focuses on the economic, ecological, and thermodynamic interpretations of the sustainability concept, leading to the identification of eight principles used to guide the modelling process. The urban development literature revealed a significant schism between the anthropocentric approach of the social sciences and the more biophysical approach of the ecological sciences. Some suggestions are made on how to resolve this impasse. The static systems (input-output) analysis provided much structural detail about the Auckland Region economic system and its environmental system; more importantly, it also details the interdependencies between these systems. A significant achievement was the construction of a 48 industry physical (mass flows) input-output model of the Auckland Region economy, and how the economy depends on physical flows to and from the environmental system. This dependency of the economic system on natural capital and ecological services was further illustrated by an input-output analysis showing how the Auckland Region economy appropriates ecological services within the Auckland Region. This was supported by an ecological footprinting analysis that revealed how the Auckland Region economy depends on natural capital (land) from outside the Auckland Region economy. The system dynamics modelling extends the static systems analysis, to build the Auckland Region Dynamic Ecological-Economic Model (ARDEEM). This dynamic model is designed to simulate future development pathways for Auckland Region; consequently it contains a number of interconnected modules that represent components critical for achieving urban sustainability in Auckland Region: population, labour force, growth driver (based on an adjusted form of So low growth theory), economy (financial flows), economy (physical flows), and the economy­ environment interface (physical flows). The ARDEEM model's use is illustrated by generating 3 scenarios for the future development of Auckland Region: 'Business as Usual ' , 'Cornucopian Growth' and 'Prudent Pessimism' . IV Finally, several areas for future research are discussed. These should try to develop further the theory that underpins urban sustainability modelling, particularly regarding improved integration of disparate theories. The best prospects lie in the future development of ARDEEM, incorporating more sectoral detail (20 - 30 industries), spatial dynamics and ecological processes that were not originally included primarily due to the lack of data. v Acknowledgements I would like to begin by thanking my mother, Dianna McDonald ( 1 943 - 2005), who at great sacrifice in the final year of her life, relentlessly supported my completion of this thesis. I would also like to thank my wife Panjama Ampanthong for her patience, kindness, and endless support throughout my studies. I am indebted to Professor Murray Patterson for his supervision, guidance, insight and much­ appreciated enthusiasm and encouragement. I would also like to thank Murray, Ann and family for accommodating me on the many occasions I visited Palmerston North over the course of my studies. I am also indebted to Cathy MacGregor for her help in typing and proof reading this thesis, and particularly for her ongoing support and encouragement throughout the course of my studies. I would also like to thank Pete McGregor for his thorough proof reading of many of the Chapters of this thesis. I am particularly grateful to the Ministry of Research, Science and Technology, McDermott Fairgray Group Ltd, and Landcare Research Ltd for supporting my studies through the Bright Future's Scholarship scheme; without this funding this thesis would not have been possible. I am beholden to my co-directors in Market Economics Ltd, Dr Doug Fairgray and Greg Akehurst, for their patience and understanding during the last five months of this thesis. Finally, I would like to thank the following friends, family and acquaintances for their insight, support, encouragement, and laughter over the last four years. Kenneth McDonald; Janice Trotter; Professor Richard Le Heron (Auckland University); Dr Richard Gordon (Landcare Research Ltd); Dr Anthony Cole (Landcare Research Ltd); Dr Nigel J ollands (Landcare Research Ltd); Dr Beat Huser (Environment Waikato); Dr Philip McDermott; Vicky Forgie; Marie Cheong; Emy Perdanahari; Richard, Sarah and Lauren MacGregor and many unnamed others. Thanks also to the anonymous markers for taking the time to examine this thesis. Table of Contents Abstract Acknowledgements List of Figures L ist of Tables Chapter One Introduction 1 . 1 Issues of Growth in the Auckland Region 1 .2 1 .3 1 .4 1 .5 PART I Need for a Systems Approach to Urban Sustainability Research Aims and Objectives 1 .3 . 1 Overall Aim 1 .3 .2 Specific Objectives Methodological Approach Thesis Organisation THEORETICAL FRAMEWORKS: How DO REGIONS DEVELOP AND GROW SUSTAINABLY? Chapter Two What is Sustainability? 2 . 1 2.2 Economic Interpretations of Sustainability 2. 1 . 1 Classical Economic Perspectives 2 . 1 .2 Emerging Neo-Classical Economic Perspectives 2. 1 .3 Neo-Malthusian View and its Repudiation 2. 1 .4 Contemporary Neo-Classical Economic Perspectives 2 . 1 .4. 1 Exhaustible Natural Resources and Sustainability 2. 1 .4.2 Renewable Natural Resources and Sustainability 2 . 1 .4.3 Pollution and Sustainability 2 . 1 .4.4 Weak and Strong Sustainability Ecological Interpretations for Sustainability 2.2. 1 Sustainability of the Biosphere 2.2.2 Sustainability of Ecosystems, Communities and Populations 2.2.2 . 1 Equilibrium Ecology 2.2.2.2 Non-Equilibrium Ecology 2.2.3 Key Ideas in Ecological Sustainability iii v XVll xx 1 3 4 4 4 5 6 1 1 13 1 4 1 4 1 7 1 9 20 2 1 23 25 27 30 30 32 32 33 3 5 Vll VllI 2.3 2.2.3.1 Biological Diversity 2.2.3.2 Ecosystem Stability and Resilience 2.2.3.3 Carrying Capacity Thermodynamic Interpretations of Sustainability 2.3.1 Statement of the Laws of Thermodynamics 2.3.1.1 The First Law of Thermodynamics and Sustainability 35 36 37 38 38 42 2.3.1.2 The Second Law of Thermodynamics and Sustainability 44 2.4 Key Principles for Assessing Auckland Region's Sustainability Chapter Three Approaches to Urban Development and Sustainability 3.1 3.2 3.3 3.4 The Human Exemptionalism Paradigm 3.1.1 Assumptions of the Human Exemptionalism Paradigm 3.1.2 Classical Sociological Theory 3.1.3 Urban Ecology 3.1.4 3.1.5 3.1.6 3.1.7 Urban Geography Urban Psychology Urban Political Economy Brief Critique of the HEP-based Urban Schools of Thought The New Ecological Paradigm 3.2.1 Assumptions of the New Ecological Paradigm 3.2.2 Cities as Ecosystems 3.2.3 Ecological Footprinting Outstanding Theoretical Issues 3.3.1 Need for Maturation of the NEP-based Approaches to Urban Sustainability 3.3.2 Need to Integrate the HEP and NEP Approaches? Summary PART 11 ENVIRONMENT-ECONOMY INTERACTIONS IN AUCKLAND REGION: A STATIC SYSTEM ANALYSIS Chapter Four Methodological Framework for a Static Systems Model of the Auckland Region 4.1 4.2 Why Build Static Models? Input-Output Analysis as the Basis for an Integrated Environment-Economy 46 53 53 54 55 57 61 64 65 69 70 71 71 80 83 83 84 85 91 93 93 IX Systems Framework 95 4.3 Critical Review of Environmental Input-Output Modelling 96 4.3.1 Inter-Industry Environmental Input-Output Models 98 4.3.1.1 Cumberland Model 98 4.3.1.2 Daly Model 100 4.3.1.3 Ayres-Kneese Model 101 4.3.1.4 LeontiefModel 102 4.3.2 Commodity-by-Industry Environmental Input-Output Models 104 4.3.2.1 Isard Model 105 4.3.2.2 Victor Model 107 4.3.2.3 Physical Input-Output Tables 108 4.4 Static Systems Framework Used in this Research for the Auckland Region 110 4.4.1 Conceptualisation of the Environment-Economy System 110 4.4.1.1 System Boundaries 112 4.4.1.2 System Flows 113 4.4.1.3 Caveats to the Conceptualisation 121 4.4.2 Mathematical Description of Auckland Region's Environment-Economy System 122 4.4.2.1 Economic Input-Output Model 122 4.4.2.2 Physical Input-Output Model 127 4.4.3 Conversion to an Inter-Industry Framework 135 4.4.3.1 Commodity Technology and Industry Technology Assumptions 136 4.4.3.2 Generating an Inter-Industry Matrix Using the Industry Technology Assumption 137 Chapter Five Economic Input-Output Model: Financial Flows in the Auckland Region Economy 141 5.1 Generation of the Auckland Region Economic Input-Output Model 142 5.1.1 Previous Regional-Level Economic Input-Output Models 142 5.1.2 Methodological Process Used in the Auckland Region Study 143 5.1.3 Update of the New Zealand Input-Output Model 145 5.1.4 Regionalisation of the New Zealand Model to Generate an Auckland Region Model 148 5.1.5 Limitations of the Auckland Region Economic Input-Output Model 157 5.1.6 Accuracy ofthe Auckland Region Economic Input-Output Model 158 x 5.2 Structural Analysis of the Auckland Region Economy 5.2.1 Economic Production 5.2.2 Contribution to New Zealand GDP 5.2.3 5.2.4 5.2.5 Economic Specialisation and Comparative Advantage Balance of Trade Network Analysis of Financial Flows: Clusters of Comparative Advantage 5.2.5.1 Industries Driven by Export Demand 5.2.5.2 Support Industries Driven by Intermediate Demand 5.2.5.3 Service Industries Driven by Local Demand 5.2.6 Multiplier Analysis of Income and Employment Impacts 5.2.6.1 Output Multipliers 5.2.6.2 Value Added Multipliers 5.2.6.3 Employment Multipliers 5.2.6.4 A Final Note on Auckland Region's Structural Interdependencies 160 160 162 165 167 170 172 173 173 174 175 177 178 180 Chapter Six Physical Input-Output Model: Physical Flows in the Auckland Region Economy and Environment 6.1 6.2 6.3 Generation of an Auckland Region Physical Input-Output Model 6.1.1 Methodological Process in the Auckland Region Study 6.1.2 6.1.3 6.1.4 6.1.5 Derivation of a National Physical Input-Output Model Derivation of an Auckland Region Physical Input-Output Model Derivation of Raw Material and Residual Flows for the Auckland RegionlNational Economy Limitations of the Physical Input-Output Models Analysis of Physical Flows in Auckland Region's Economy 6.2.1 Overview of Physical Flows 6.2.2 Sectoral Physical Flows: Network Diagram 6.2.2.1 Explanation of the Network Diagram 183 184 185 186 198 200 206 208 208 211 211 6.2.2.2 Interpretation of the Network Diagram for Significant Industries 212 6.2.2.3 Key Inter linkages in the Network Diagram 216 6.2.3 Physical Balance of Trade 216 Ecological Multiplier Analysis: Combining Physical and Economic Flows 6.3.1 Rationale and Method 6.3.2 Ecological Multipliers for the Auckland Region 217 217 218 Xl 6.3.3 Cumulative Effects Indicator 220 Chapter Seven Extending the Input-Output Frameworks: Ecological Processes and Services in the Auckland Region 223 7.1 Input-Output Model of Ecological Processes in the Auckland Region 224 7.1.1 Previous Input-Output Models of Ecological Processes 224 7.1.2 Scope of the Auckland Region Input-Output Model of Ecological Processes 226 7.1.3 Auckland Region Carbon Cycle Module 230 7.1.4 Auckland Region Hydrological Cycle Module 235 7.1.5 Auckland Region Phosphorus Cycle Module 238 7.1.6 Auckland Region Sulphur Cycle Module 241 7.1.7 Auckland Region Nitrogen Cycle Module 244 7.1.8 Limitations and Caveats 247 7.2 Input-Output Model of Ecological Services Input into the Auckland Region Economy 248 7.2.1 Methodology 250 7.2.2 Direct Ecosystem Service Value by Industry 254 7.2.3 Direct and Indirect Ecosystem Services by Industry 257 Chapter Eight Auckland Region's Ecological Footprint and Environmental Interdependencies with Other Regions 261 8.1 The Ecological Footprint Concept 261 8.1.1 What is the Ecological Footprint? 261 8.1.2 History of the Ecological Footprint 262 8.1.3 How is the Ecological Footprint Calculated? 263 8.2 Critique of the Ecological Footprint Concept 265 8.2.1 Lack of Common Definitions and Methodologies 265 8.2.2 Why Use Land as the Numeraire? 266 8.2.3 Why Include Hypothetical Energy Land? 267 8.2.4 Is All Land the Same? 267 8.2.5 What Spatial Boundaries? 268 8.2.6 Dynamics - What About the Future? 268 8.2.7 Policy Relevance - A Policy Evaluation Tool? 269 XII 8.3 An Input-Output Method for Estimating Auckland Region's Ecological Footprint 8.3.1 Accounting Identity of the Component Parts of the Regional Ecological Footprint 8.3.2 Generation of Input-Output Tables 8.3.3 Calculation of the Land Appropriated Within the Study Region (a) 8.3.4 Calculation of the Land Appropriated from Other Regions (fJI + P2 + ... + Po-I) 8.3.5 Calculation of Land Appropriated from Other Countries (d) 8.3.6 Limitations of Using Input-Output Analysis 8.4 Ecological Footprint of the Auckland Region 8.4.1 Brief Description of the Auckland Region 8.4.2 Data Sources 8.4.3 Auckland Region's Ecological Footprint Disaggregated by Land Type 8.4.4 Auckland Region's Ecological Footprint Disaggregated by Economic Industry 8.4.5 Auckland Region's Ecological Balance of Trade 8.4.6 Comparing Auckland Region's Ecological Footprint with Other Regions 8.4.7 Comparing Auckland Region's Ecological Footprint with International Ecological Footprints PART III ENVIRONMENT-ECONOMY INTERACTIONS IN AUCKLAND REGION: A DYNAMIC SYSTEMS ANALYSIS Chapter Nine System Dynamics Approach to Modelling Ecological-Economic Systems 9.1 9.2 9.3 9.4 Why Build Dynamic Models? Why Build a System Dynamics Model? Brief History of System Dynamics Key Features of a System Dynamics Model 9.4.1 Dynamic System Behaviour 9.4.2 Building Blocks for System Dynamics Models 9.4.2.1 Stocks 9.4.2.2 Flows 269 270 270 271 271 274 274 276 276 276 277 278 280 283 286 289 291 291 292 293 294 295 297 298 298 9.5 9.4.2.3 Converters 9.4.2.4 Connectors 9.4.2.5 Bringing the Building Blocks Together 9.4.2.6 General Principles for Building Systems Models 9.4.2.7 Setting Delta Time 9.4.2.8 Hannon and Ruth's Four Model Set Modelling Process Chapter Ten Critical Review of Growth Theories 10.1 Classical Growth Theories 10.2 Neo-c1assical Growth Theories 10.2.1 Harrod-Domar Growth Model 10.2.2 The Solow Model 10.2.3 The So low Model with Technology 10.3 Endogenous or New Growth Theory 10.3.1 The Romer Model 10.3.2 Other Endogenous Growth Models 10.4 Critique of Growth Theories as Applied to Environmental and Regional Models 10.5 Natural Resources and Economic Growth 10.5.1 Modelling the Implications of Land Use 10.5.2 Modelling of the Depletion ofa Non-Renewable Resource X III 299 299 299 300 301 301 302 309 309 311 311 312 315 317 319 322 323 325 326 326 Chapter Eleven Auckland Region Dynamic Ecological-Economic Model 329 11.1 Structure of ARDEEM 11.2 ARDEEM Mathematical Nomenclature 11.3 Population Module 11.4 Labour Force Module 11.5 Growth Module 11.6 Economic Module 11.7 Economic Physical Flow Module 11.8 Environment-Economy Physical Flow Module 11.9 Validation and Verification of ARDEEM 11.9.1 Structural Validity of ARDEEM 11.9.2 Predictive Validity of ARDEEM 329 331 332 337 340 346 350 356 362 362 363 XIV 1 1 . 1 0 Scenario Analysis ILl 0 . 1 ARDEEM Scenarios 1 1 . 1 0.2 Simulation Results 1 1 . 1 1 Limitations of the ARDEEM 1 1 . 1 1 . 1 Extending ARDEEM to Include Ecological Processes Chapter Twelve Thesis Summary and Conclusions 1 2. 1 Thesis Contributions 1 2. 1 . 1 Theoretical Contributions 12. 1 .2 Methodological Contributions 1 2. 1 .3 Empirical and Knowledge Contributions 1 2.2 Limitations and Future Research 1 2.2. 1 Theoretical Analysis 1 2.2.2 Static Systems Analysis 1 2.2.3 Dynamic Systems Analysis List of PhD Outputs References APPENDICES Appendix A Input-Output Analysis: History, Mathematics and Assumptions A. 1 A.2 A.3 AA A.5 Brief History of Input-Output Modelling Input-Output Tables Technical Coefficients The Leontief Inverse Assumptions of Input-Output Modelling Appendix B Global Biogeochemical Cycling Model B . l B.2 Rationale for the Global Biogeochemical Cycling Model Data Sources B.2. 1 Global Fluxes 363 365 367 373 374 377 377 377 378 380 3 82 382 3 83 3 84 387 391 473 475 475 475 478 479 480 481 48 1 483 484 B.3 BA B.5 B.6 B.7 B.2.2 Global Reservoirs Description of Stocks, Flows and Converters in the GBCM GBCM Mathematical Description Steady State Analysis Limitations and Caveats of the GBCM Finite Difference Equations for the GBCM B.7 . 1 Atmosphere Stocks B.7 .2 Terrestrial Stocks B.7 .3 Marine Stocks B.7A Hydrosphere Stocks B.7 .5 Lithosphere Stocks B.7 .6 Oxygen Stock Appendix C Mathematical Description of the Interregional Trade 494 500 50 1 509 5 1 2 5 1 3 5 1 5 5 1 8 520 523 524 525 Flows Optimisation used in the Ecological Footprint Analysis 527 Appendix D Non-Survey Regionalisation Methodologies D. 1 D.2 Coefficient Reduction Methodologies D. 1 . 1 Location Quotient Methodologies D. 1 .2 The Commodity Balance Approaches D. 1 .3 Constrained Matrix Techniques Other Approaches D.2. 1 Regional Weights D.2.2 Representative Regional Coefficients Appendix E Industry Definitions and Concordances Appendix F Aggregated Commodity-by-Industry Economic Input-Output 529 530 530 53 1 532 534 534 535 537 Models For New Zealand and the Auckland Region, 1997-98 549 Appendix G Multiplier Analysis Methodology G. 1 Input-Output Multipliers G. 1 . 1 Output Multipliers G . 1 .2 Value Added Multipliers G . 1 .3 Employment Multipliers 555 555 557 557 558 xv XVI Appendix H Aggregated Commodity-by-Industry Physical Input-Output Models For New Zealand and the Auckland Region, 1997-98 559 Appendix I Raw Material and Residual Inputs/Outputs of the Auckland Region Economy, 1997-98 1.1 1.2 1.3 1.4 Raw Material Inputs Residual Inputs Raw Material Outputs Residual Outputs 565 565 567 568 568 AppendixJ Assessing the Value of Auckland Region's Ecosystem Services 573 J.l J.2 J.3 J.4 J.5 Valuation Approach Valuation Methods Methodological Sequence J.3.1 Auckland Region's Ecosystem Types and Services J.3.2 Estimates of the Direct Use-Value of Ecosystem Services J.3.3 J.3.4 J.3.5 Estimates of the Indirect Use-Value of Ecosystem Services Estimation of the Auckland Region's TEV Distribution of Auckland Region's TEV across 48 Input-Output Industries Theoretical and Methodological Issues The Auckland Region TEV J.5.1 By Terrestrial Ecosystem Type J.5.2 By Terrestrial Ecosystem Service J.5.3 Coastal Ecosystem Services Appendix K Environmental Flows of Land and Energy into the Auckland Region Economy Appendix L System Dynamics Model of Endogenous Growth Appendix M ARDEEM Regression Equations Appendix N CD-ROM Software 573 575 575 575 580 580 581 581 581 582 582 583 585 587 591 593 597 XVll List of Figures Figure 1.1 Interrelationships between Thesis Chapters 7 Figure 2.1 Non-Equilibrium and Equilibrium Points in Ecosystem Dynamics 34 Figure 2.2 Holling's Four Phase Model of Ecosystem Change and Resilience 37 Figure 2.3 Interrelationships between Key Principles of Sustainability 51 Figure 3.1 Urban Ecology Models of Urban Development 59 Figure 3.2 Urban Geography Models of Urban Development 62 Figure 3.3 Hierarchical and Spatial Arrangement of Central Places 64 Figure 3.4 Harvey's Model of the Circulation of Capital 67 Figure 3.5 Resource Inputs Consumed and Waste Outputs Discharged from Sydney, 1990 73 Figure 3.6 Extended Metabolism Model of Human Settlements 76 Figure 4.1 Cumberland Model 99 Figure 4.2 Daly Model 100 Figure 4.3 Ayres-Kneese Model 101 Figure 4.4 Leontief Model 103 Figure 4.5 Isard Model 106 Figure 4.6 V ictor Model 107 Figure 4.7 A Physical Input-Output Model 109 Figure 4.8 Auckland Region's Environment-Economy System and its Relationship with Other Systems 111 Figure 4.9 Auckland Region's Environment-Economy System With an Expanded Environmental System 116 Figure 4.10 Auckland Region's Environment-Economy System With an Expanded Economic System 118 Figure 5.1 Methodological Process for Generating an Auckland Region Economic (Commodity-by-Industry) Input-Output Model 144 Figure 5.2 Top Ranking Auckland Region Industries, 1997-98 (Percentage Contribution to New Zealand GDP) 164 Figure 5.3 Auckland Region's Clusters of Comparative Advantage, 1997-98 169 Figure 6.1 Methodological Process for Deriving New Zealand and Auckland Region Physical Input-Output Models 186 Figure 6.2 Auckland Region's Major Environment-Economy Physical Flows, 1997-98 210 Figure 6.3 Network Diagram of the Main Flows in the Auckland Region Economy, 1997-98 215 XVlll Figure 7.1 Structure of the U - V Matrix of Ecological Flows in the Auckland Region 229 Figure 7.2 Auckland Region's Carbon Cycle, 1997-98 232 Figure 7.3 Auckland Region's Hydrological Cycle, 1997-98 236 Figure 7.4 Auckland Region's Phosphorus Cycle, 1997-98 239 Figure 7.5 Auckland Region's Sulphur Cycle, 1997-98 242 Figure 7.6 Auckland Region's Nitrogen Cycle, 1997-98 245 Figure 7.7 Input-Output Model Accounting Framework 250 Figure 7.8 Methodological Process for Calculating Ecological Multipliers and Tree Diagrams 251 Figure 7.9 Total Embodied Ecosystem Services Appropriated by the Auckland Region Air Transport Industry, 1997-98 258 Figure 7.10 Total Embodied Ecosystem Services Appropriated by the Auckland Region Business Services Industry, 1997-98 259 Figure 8.1 Regional and International Origins of Auckland Region's Ecological Footprint, 1997-98 282 Figure 8.2 Ecological Footprints of New Zealand Regions, 1997-98 284 Figure 8.3 Comparison of Auckland Region Ecological Footprint Per Capita with Other Regions in New Zealand, 1997-98 285 Figure 8.4 Comparison of Auckland Region Ecological Footprint Per Capita with Other Nations 287 Figure 9.1 Common Types of Dynamic Behaviour 297 Figure 9.2 Hannon and Ruth's Four Model Set 301 Figure 9.3 Key Steps in the System Dynamics Modelling Process 302 Figure 9.4 Bull's Eye Diagram 304 Figure 10.1 The Solow Diagram 315 Figure 11.1 Module Linkages 331 Figure 11.2 Population Module Influence Diagram 334 Figure 11.3 Labour Force Module Influence Diagram 338 Figure 11.4 Growth Module Influence Diagram 343 Figure 11.5 Economic Module Influence Diagram 347 Figure 11.6 Economic Physical Flow Influence Diagram 352 Figure 11.7 Environment-Economy Physical Flow Influence Diagram 357 Figure 11.8 ARDEEM Scenario Analysis: Business As Usual, Cornucopian Growth and Prudent Pessimism 370 Figure A. l An Input-Output Table 477 Figure B . l Carbon Cycle Influence Diagram 504 Figure B.2 Figure B.3 Figure B.4 Figure B.5 Figure B.6 Figure C.1 Figure 1.1 Figure 1.2 Figure 1.3 Figure L . l Hydrological Cycle Influence Diagram Phosphorus Cycle Influence Diagram Sulphur Cycle Influence Diagram Nitrogen Cycle Influence Diagram Baseline Analysis of the Steady State Conditions of Critical Stocks in the GBCM, 2001-2051 Structure of the Interregional Trade Flows Optimisation Problem Estimation of the Consumers and Producers Surplus for a Substitutable Ecosystem Service Estimation of the Consumers and Producers Surplus for a Non­ Substitutable Ecosystem Service Methodological Sequence for the Estimation of the Total Economic Value (TEV) of Auckland Region's Ecosystem Services An Alternative Endogenous Growth Engine XIX 505 506 507 508 510 527 574 574 577 592 xx List of Tables Table 2 . 1 Theoretical Principles for Sustainable Development and their Application in this Thesis 47 Table 3 . 1 (a) Conceptual Foundations for HEP-based Urban Schools of Thought 87 Table 3 . 1 (b) Conceptual Foundations for NEP-based Urban Schools of Thought 88 Table 4. 1 Commodity-by-Industry Financial Flow Matrix 1 23 Table 4.2 Commodity-by-Industry Physical Flow Matrix 1 29 Table 5 . 1 Contribution to Auckland Region GRP 1 62 Table 5 .2 Location Quotients for the Auckland Region Economy 1 66 Table 5 .3 Auckland Region's Financial Balance of Trade, 1 997-98 1 69 Table 5.4 Output Multipliers for Auckland Region and New Zealand, 1 997-98 1 76 Table 5.5 Value Added Multipliers for Auckland Region and New Zealand, 1 997-98 1 78 Table 5.6 Employment Multipliers for Auckland Region and New Zealand, 1 997-98 1 80 Table 6. 1 International Exports as a Percentage of Commodity Output, 1 997-98 1 90 Table 6.2 International Imports as a Percentage of Commodity Input, 1 997-98 1 9 1 Table 6.3 Superior Data Inserted into the New Zealand Physical Input-Output Model, 1997-98 1 96 Table 6.4 Classification of Raw Materials and Residuals 201 Table 6.5 Data Sources for the Raw Material Inputs into the Auckland Region and New Zealand Economies, 1 995-98 203 Table 6.6 Data Sources for the Residual Outputs from the Auckland Region and New Zealand Economies, 1 995-98 204 Table 6.7 Physical Balance of Trade for the Auckland Region Disaggregated by Commodity Type, 1 997-98 (kt) 2 1 7 Table 6.8 Ecological Multipliers for Auckland Region's Economic Industries, 1 997-98 2 1 9 Table 6.9 Cumulative Effects Index for Auckland Region's Economic Industries, 1 997-98 222 Table 7. 1 World to Auckland Region Scalars Used to Generate the Prototype Input Output Model of Ecological Processes in the Auckland Region 227 Table 7 .2 Input-Output Model of the Carbon Cycle Processes for Auckland Region, 1997-98 223 Table 7 .3 Input-Output Model of the Hydrological Cycle Processes for Auckland Region, 1997-98 237 XXI Table 704 Input-Output Model of the Phosphorus Cycle Processes for Auckland Region, 1997-98 240 Table 7.5 Input-Output Model of the Sulphur Cycle Processes for Auckland Region, 1997-98 243 Table 7.6 Input-Output Model of the Nitrogen Cycle Processes for Auckland Region, 1997-98 246 Table 7.7 Summary of the Ecosystem Services Assessed 249 Table 7.8 Direct Value Derived from Auckland Region's Terrestrial Ecosystems, by Economic Industry, 1997-98 256 Table 8.1 Assumptions Made by Three Different Ecological Footprint Calculation Methods 266 Table 8.2 Illustrative Example of the Matrix T: Land Appropriated From Other Regions PI + P2 + . . . + Pn.1 274 Table 8.3 Auckland Region Ecological Footprint Disaggregated by Land Type, 1997-98 277 Table 804 Auckland Region Ecological Footprint Disaggregated by Economic Industry, 1997-98 279 Table 8.5 Ecological Balance of Trade for the Auckland Region Disaggregated by Land Type, 1997-98 280 Table 9.1 Four Systems Components and their Modelling Symbols 298 Table 11.1 Summary of Drivers under Each Scenario 367 Table B.l Biosphere Inputs Into Processes 486 Table B.2 Biosphere Outputs From Processes 490 Table B.3 Global Carbon Reservoirs 495 Table BA Global Hydrogen Reservoirs 496 Table B.5 Global Phosphorus Reservoirs 497 Table B.6 Global Sulphur Reservoirs 498 Table B.7 Global Nitrogen Reservoirs 499 Table B.8 Non-marker Arrayed Flows and Sets of Processes 514 Table D. l Hypothetical Transactions Table (Target Year) 532 Table D.2 Hypothetical Transactions Table (Base Year) 533 Table D.3 Hypothetical Transactions Table (Target Year) - 1 sI Iteration 533 Table DA Hypothetical Transactions Table (Target Year) - 2nd Iteration 533 Table D.5 Hypothetical Transactions Table (Target Year) - 3rd Iteration 534 Table D.6 Hypothetical Transactions Table (Target Year) - Final Iteration 534 Table E. l Industry Defmitions 537 Table E.2 Industry Definitions Concordance 539 xxii Table EJ Commodity Definitions 542 Table E.4 Commodity Definitions Concordance 545 Table F. l Aggregated Commodity-by-Industry Input-Output Model for New Zealand, 1 997-98 ($ mil) 550 Table F.2 Aggregated Commodity-by-Industry Input-Output Model for the Auckland Region, 1 997-98 ($ mil) 55 1 Table FJ Aggregated Industry-by-Industry Input-Output Model for New Zealand, 1997-98 ($ mil) 552 Table F.4 Aggregated Industry-by-Industry Input-Output Model for the Auckland Region, 1 997-98 ($ mil) 553 Table G. l Technical Coefficients Table for a Hypothetical Region 555 Table G.2 LeontiefMatrix (I-A) for a Hypothetical Region 555 Table GJ Leontief Inverse Matrix (I-Arl for a Hypothetical Region 556 Table G.4 Closed LeontiefMatrix (I-A *) for a Hypothetical Region 556 Table G.5 Closed Leontief Inverse Matrix (I-A *rl for a Hypothetical Region 556 Table H. l Aggregated Commodity-by-Industry Physical Input-Output Model for New Zealand, 1 997-98 ('OOOs t) 560 Table H.2 Aggregated Commodity-by-Industry Physical Input-Output Model for the Auckland Region, 1 997-98 (,OOOs t) 562 Table 1. 1 Raw Material Inputs by 48 Industries and Households in the Auckland Region Economy, 1 997-98 ('OOOs t) 566 Table 1.2 Residual Outputs by 48 Industries, Final Consumption, and Gross Capital Formation and Man-made Assets in the Auckland Region Economy, 1 997-98 ('OOOs t) 569 Table 1. 1 Direct and Indirect Use Value Derived from Auckland Region's Terrestrial Ecosystem Types, 1 997-98 583 Table 1.2 Direct and Indirect Use Value Derived from Auckland Region's Terrestrial Ecosystems, by Ecosystem Service, 1 997-98 585 Table K. l Summary of Methodologies Used to Estimate Physical Flows of Land and Energy 588 Table K.2 Energy and Land Inputs by 48 Industries and Households, 1 997-98 589 Table M . l ARDEEM Fertility Rate Regression Equations 593 Table M.2 ARDEEM Mortality Rate Regression Equations 594 Table MJ ARDEEM Labour Force Participation Rate Regression Equations 595 Table M.4 ARDEEM Employment by Industry Distribution Regression Equations 595 Table M.5 ARDEEM Depreciation Rate Regression Equations 595 Table M.6 Table M.7 Table M.8 XXlll ARDEEM Investment Rate Regression Equations 596 ARDEEM International Exports to Gross Output Regression Equation 596 ARDEEM Factors of Production Elasticities with Respect to Output 596 Chapter One Introduction 1.1 Issues of Growth in the Auckland Region 1 New Zealand is one of the most urbanised countries in the world, with more than 85 percent of its population living in urban areas. Since the 1 950s New Zealand's cities have grown rapidly outwards at relatively low densities (Ministry for the Environment, 2000). The pattern of urban growth has paralleled that of most developed nations. A key feature has been the development of cities geared toward transportation; road, rail, water and air networks have all strongly influenced the shape, form and density of our cities (Perkins et al. , 1 993). In Auckland Region, urban growth has largely been unconstrained with sprawl encroaching on the surrounding rural hinterland. The Auckland Region measures approximately 1 20 kilometres from north to south and spans 60 kilometres at its widest. With a land area of just 5,600 square kilometres, the region makes up only 2.0 percent of New Zealand's land area, but is densely populated with 29.5 percent of New Zealanders living there. The Auckland Region is made up of seven territorial local authorities (TLAs) - three districts (Rodney, Papakura and part of Franklin) and four cities (North Shore, Waitakere, Auckland and Manukau). Auckland Region is the largest and fastest growing region in New Zealand with a population of 1 .23 million at the 200 1 census. The population is projected to grow to 1 ,793,300 in 2026, a nearly 50 percent increase between 200 1 and 2026. This means that the Auckland Region is growing roughly by the population equivalent of a city the size of Dunedin (as at census 200 1 ) every five years. Five of Auckland Region's seven TLAs are predicted to be in the ten fastest growing TLAs nationwide over the next 25 years. Growth in the Auckland Region escalated in the mid 1 960s, partly as a response to post-World War II growth in manufacturing. This, together with an expansion in jobs and population, led to a concentration of industry in the Region and a preference by business to locate there. Ethnic diversity resulting from post-war migration of Maori and Polynesians to the Auckland Region as labour for the manufacturing sector, and more recently, encouragement by government of business migrants particularly from Asia, has made the Auckland Region relatively distinct from other regions in New Zealand. Population growth within the region and its main cities has continued to escalate, and attempts to limit the Region's growth and prevent urban sprawl have been largely unsuccessful. 2 Urban sprawl, coupled with increasing affluence, has led to significant pressures being placed on the biophysical environment (Newrnan, 1 999); examples include increased per capita demand for natural resources like land, water and energy, and increased assimilative demands on ecosystem services like CO2 absorption, biodegrading of waste, and cleansing of pollutants by fresh water systems and the ocean. A wider definition of environment also encapsulates losses of culture and heritage, pressures on amenity values, and adverse effects on wellbeing and health of people and communities. Furthermore, these impacts are not necessarily localised, but may be felt well beyond municipal limits, i.e. the appropriation of carrying capacity. Although its population is relatively modest in global terms, the Auckland Region urban area far exceeds that of most international cities. With the rapid increase in motor vehicle use in the 1950s and the introduction of motorways, the urban area expanded outward and congestion became a concern. Urban sprawl has meant Aucklanders depend heavily cars; as at the 1 996 census, 1 4. 1 percent of households owned three or more vehicles compared with 1 1 . 1 percent nationally. Traffic volume in the region is the greatest in the country; more than double that of the next closest region of Waikato, with travel to work being the primary generator of traffic flows. The problems caused by growth are also revealed in other areas e.g. water sewage and electricity services, which are all vulnerable and require substantial investment. With its rapidly growing population, environmental issues are critical to the Region. A growing population accompanied by significantly more economic throughput means more resource use and more waste generation. For example, Auckland Region's escalating traffic congestion results in increasing CO2 emissions. The Auckland Regional Council ( 1 998, p.2), in its Draft Annual Plan 1997198, states its role is to "protect the region's air, soil and water resources from pollution and to ensure their sustainable use as Auckland develops". To promote these goals, the Regional Growth Forum, comprising representatives of the local councils, has been established. The role of the Regional Growth Forum is to "develop sustainable growth strategies to accommodate anticipated popUlation growth while maintaining and improving the quality of life of Aucklanders" (Statistics New Zealand, 1 999, p.3). The distribution of the labour force reveals Auckland Region's commercial dominance. Auckland Region's status as an international market is reflected in the relatively high proportion of people employed in the wholesale trade industry ( 1 996 census): 9.0 percent compared with 6.2 percent nationally (Statistics New Zealand, 1 999). Service industries such as finance, insurance, property services and business services also employ a larger share of the population, with legislators, administrators and managers accounting for 1 4.9 percent of all employed people in the region, compared with 1 2.2 percent nationally (Statistics New Zealand, 1 999). 3 These occupations reflect the size of the urban area and the high number of company head offices found in the region. Not surprisingly, employment in the agriculture, forestry and fishing industries was about one-quarter of the national percentage, reflecting Auckland Region's predominantly urban nature (Statistics New Zealand, 1999). Growth of the tertiary sector rather than the primary or secondary sectors means that resource use per capita will be lower in the Auckland Region; however, this is offset by population growth which requires increasing resource inputs and produces more residual outputs. 1.2 Need for a Systems Approach to Urban Sustainability Urban areas such as the Auckland Region provide significant challenges for sustainable development due to the highly modified nature of the urban ecosystem; a facet of population density, intensive resource depletion and environmental degradation, and conflict between people/communities and their environmental, economic, social and cultural requirements (Parliamentary Commissioner for the Environment, 1998). Urban sustainability involves integrating the requirements of environmental management, social equity and economic opportunity into decision making. This is a process of change in which the use of resources; generation of wastes, emissions and pollutants; and technology and institutional factors are managed to meet the needs of current and future generations. Despite recognition of the importance of urban sustainabil ity and the need to adopt an integrated perspective on this issue the literature is devoid of substantive theory and method that give greater weight to the defmition of urban sustainability (Slocombe, 1993). Therefore, a challenge for this thesis is to develop a theoretical framework and associated analytical tools to give operational meaning to the concept, as it is applied to the Auckland Region. Without such endeavours, the concept of urban sustainabil ity will remain abstract and elusive, and decision makers and communities will be unable to make informed choices about the trade-offs inherent in the concept. There is widespread appreciation that urban sustainability can only be adequately dealt with through an integrated perspective that encapsulates the whole spectrum of economic, soc ial and ecological values. Without such a perspective, robust and enduring solutions to the urban sustainability problem will remain elusive. Systems theory and modelling are used in this thesis as one way of addressing the need for an integrated perspective. Because of the complex nature of urban sustainability, the traditional scientific method (reductionism) is not suitable. Instead, a systems methodology, which is 4 concerned with the functioning of the whole rather than the individual parts, is more suitable. This systems methodology has been developed over the last 40 years, gaining wider acceptance in the scientific community (Forrester, 1978; Checkland, 1981, Meadows et al., 1992). More specifically, systems modelling methods (static and dynamic) are applied in this thesis to address the issue of urban sustainability. Methods like system dynamics which is used in this thesis, attempt to model complex systems that are characterised by feedbacks, non-linearities, time lags and at a higher level emergent properties (Costanza and Voinov, 2004). System dynamics can be used to project future scenarios for Auckland Region, providing stakeholders with better information on the trade-offs and interconnections of the economic, social and ecological dimensions of urban sustainability. 1.3 Research Aims and Objectives 1.3.1 Overall Aim The overall thesis aim is to develop an operational systems modelling approach to address the issue of urban sustainability in Auckland Region. 1.3.2 Specific Objectives Specific objectives for this PhD project are to: (1) Provide a theoretical interpretation of the concept of urban sustainability by drawing upon the relevant literature from the social and biophysical sciences. From this analysis, to develop a set of principles for determining urban sustainability that can be used in the systems modelling. (2) Extend and further develop static modelling methods (input-output analysis) that have previously been used to understand economy-environment interactions but have not been widely used to understand urban sustainability. Such methods define the structural relationships in complex urban areas between the economy and the environment. (3) Develop an analytical framework for understanding urban dynamics and changes in urban sustainability in Auckland Region, drawing on theories of economic growth, spatial change, urban development and urban economics. 5 (4) Develop a dynamic simulation model of key interactions between population change, socio-economic trends and ecological flows (resources and residuals) in Auckland Region, then utilise this dynamic simulation model and associated indicators to project alternative scenarios of sustainable (or unsustainable) economic and social change in the Auckland Region. 1.4 Methodological Approach The methodology used in this thesis consists of four integrated components: ( 1 ) Critical reviews and synthesis of the relevant literature that relate to different dimensions of urban sustainability and change: (i) economic, ecological and thermodynamic interpretations of the sustainability concept; (ii) theories of urban change including those from urban ecology, urban geography, urban psychology, urban political economy, urban metabolism and ecological footprinting; and (iii) growth theories including neoclassical growth theories, endogenous or new growth theory, amongst others. The purpose of these l iterature reviews and syntheses is to provide a firm theoretical basis for systems modelling of the urban sustainability concept undertaken later in the thesis. (2) Development of a clear and conceptual framework for analysing urban sustainability in Auckland Region. Before the systems modelling can proceed, the different dimensions of urban sustainability in Auckland Region need to be conceptualised in an integrative fashion. A particular focus of this thesis is the interconnection between the economic system and the environmental system. The dynamics of both systems need to be understood as a basis for the systems modelling. A key challenge is to integrate the different perspectives on the issue of urban growth and change that became evident from the critical synthesis. (3) Development of Static Models of the economic and environmental systems as well as the interconnections between each system. These static (input-output) models provide detail on the structural interdependencies within and between the systems. There are several other reasons for building static models of the Auckland Region economy­ environment: (i) determining the state of these systems; (ii) setting the initial conditions for the dynamic models; (iii) to undertake comparative statics analysis; and (iv) analytical ease which allows for wider use and accountability of such models for end­ users. 6 (4) Development of Dynamic Models of the economic and environmental systems, as well as the interconnections between each system. A significant limitation of many models is that they are static, i .e. they only represent a snapshot of reality at a s ingle moment. Reality, however, is not static, but constantly changing. In this thesis, a dynamic model (ARDEEM) is built of the Auckland Region economy and its interaction with the environment system using the system dynamics modelling approach. This system dynamics model is used to model future scenarios of economic and environmental change in Auckland Region, developing the earlier work of Ryan ( 1 995) who built such models for New Zealand. As an adjunct (Appendix B) to this thesis, a model of ecological processes in the global environment system was operationalised. However, such a model proved too difficult to operationalise for the Auckland Region, primarily due to the lack of data. 1.5 Thesis Organisation This thesis is divided into three distinct yet related parts : Part I Theoretical Frameworks: How Do Regions Develop and Grow Sustainably Part II Environment-Economy Interactions in Auckland Region: A Static Systems Analysis Part III Environment-Economy Interactions in Auckland Region: A Dynamic Systems Analysis The interrelationships between these different parts of the thesis and the constituent Chapters are described by Figure 1 . 1 . Theoretical Chapter 2 What is Sustainability? Chapter 3 Approaches to Urban Development and Sustainability Static Analysis Chapter 4 Methodological Framework for a Static Systems Model of the Auckland Region Chapter 5 Economic Input-Output Model: Financial Flows in the Auckland Region Economy Chapter 6 Physical Input-Output Model: Physical Flows in the Aucklandl----l'---+-T-----� Region Economy and Environment Chapter 7 Dynamic Analysis Chapter 9 System Dynamics Approach to Modelling Ecological-Economic Systems Chapter 10 Critical Review of Growth Chapter 1 1 Auckland Region Dynamic Ecological-Economic Model Extending the I nput-Output Appendix B Frameworks: Ecological I-___ �:....---L-----� Global Biogeochemical Cycling Processes and Services in the Model Auckland Region Chapter 8 Auckland Region's Ecological Footprint and Environmental I nterdependencies with Other Regions Figure 1 .1 Interrelationships between Thesis Chapters. Note: Appendix B is dashed as it is a global, rather than Auckland Region specific, model. 8 Part I (Iheoretical) addresses the research questions of sustainability and growth in urban centres through a theoretical analysis of the literature. Chapter 2 reviews the sustainability literature to examine the economic, ecological and thermodynamic interpretations of sustainability. Chapter 3 then turns to the issue of urban growth and sustainability, as viewed through various disciplinary perspectives in both the social and biophysical sciences. Part 11 (Static Modelling) analyses the structural interdependencies within and between the Auckland Region economic and environment systems. Chapter 4 provides a conceptual and related mathematical framework for Part IT. Chapter 5 builds an economic input-output model of the Auckland Region economy and undertakes various analyses of the structure of this system. Chapter 6 builds a Physical Input-Output model of the Auckland Region economy that qualifies the flows of mass through the economy, as well as between the economy and environment. This Physical Input-Output model is important for understanding the biophysical and ecological aspects of urban sustainability, and represents a first attempt to operationalise such a model at the regional level. Chapter 7 extends the static (input-output) modelling framework to cover ecological processes and ecosystem services in the environment system. Some attempt is made to determine how these ecological processes and services support economic activity in the later part of Chapter 7. Chapter 8 uses ecological footprinting analysis to highlight the fact that urban areas such as Auckland Region depend on natural capital and ecosystem services drawn from other regions. In Auckland Region's case this is very significant, given its small natural capital base and hence its unavoidable dependence on other regions. Part 111 (Dynamic Modelling) builds on the static analysis to dynamically model the Auckland Region economy and its interaction with the environment. Chapter 9 delineates the key features of the system dynamics approach to modelling and its methodological underpinnings. Chapter 10 critically reviews growth theories and their possible application to the Auckland Region model. This review is important as it informs the selection of a growth driver in the dynamic model of the Auckland Region economy. Chapter 1 1 develops a dynamic model of the Auckland Region economy and its interactions with the environment system. This model contains several interrelated modules: population, labour force, growth (based on growth theory), economic, economic physical flow and an economic-environment physical flow. This model is used to produce data for three scenarios for Auckland Region: Scenario 1 'Business­ As-Usual ' , Scenario 2 'Cornucopian Growth', and Scenario 3 'Prudent Pessimism'. To conclude, Appendix B presents a dynamic global biogeochemical cycling model. ! I This model is a novel by-product of the thesis. 11 PART I THEORETICAL FRAMEWORKS How DO REGIONS DEVELOP AND GROW SUST AlNABLY? 1 3 Chapter Two What is Sustainability? The term sustainability is a popular buzzword of academic theory and public policy concerned with the environment. According to Caldwell ( 1990), the concept of sustainability was first acknowledged at the international policy level at the 1 972 United Nations Conference on the Human Environment. The term has since gained greater international political importance in publications such as World Conservation Strategy (IDCN, 1 980), Our Common Future (WCED, 1 987), Caring for the Earth: A Strategy for Sustainable Living (IDCN, 1 99 1 ), Rio Earth Summit Conference on Environment and Development (UNCED, 1 992), Kyoto Protocol to the United Nations Framework Convention on Climate Change (UNFCCC, 1 997), and the Johannesburg Conference Report of the World Summit on Sustainable Development (United Nations, 2002)? Nevertheless, it was the WCED ( 1 987, p.43) report that popularised the notion of sustainable development, defining it as "development that meets the needs of the present without compromising the ability of future generations to meet their own needs". Sustainability has however been critic ised as vague, unrealistic, contradictory, and politically expedient. Dialogue Consultants ( 1992), in a report to the New Zealand Ministry for the Environment, described sustainability as an 'absent referent' - a term of reference seldom defmed accurately. It is c lear from the burgeoning sustainability l iterature that there is no single defmition of sustainability; indeed different disciplines interpret the concept in fundamentally different ways. A comprehensive review of the sustainable development literature by Pezzoli ( 1 997 a, 1 997b) arranges these interpretations into eleven categories.3 Pezzoli' s classification has been criticised as prejudiced towards the social sciences and humanities, ignoring significant ecological and scientific l iterature. The focus of this Chapter is specifically on the economic, ecological, and thermodynamic interpretations of sustainability, i.e. those foundational disciplines of ecological economics.4 2 The international policy approaches to sustainability emphasise the social, institutional, economic and environmental aspects of sustainability within frameworks that seek to balance or integrate key sustainab ility factors (Patterson, 2002b). In particular, policy agencies have promoted perspectives that have attempted to integrate the social, economic and environmental dimensions of sustainability. 3 These categories include policy and p lanning; social conditions; environmental law; environmental sciences; eco-design and the environment; ecological economics; eco-philosophy; environmental values and ethics; environmental history and geography; utopianism, anarchism and bioregionalism; and political ecology. Pezzey ( 1992) provides a similar gallery of definitions. 4 It is beyond the scope of this thesis to review all theoretical interpretations of the sustainability concept. Similarly, coverage of the economic, ecological and thermodynamic interpretations of sustainability is limited to key contributions in these disciplines. 14 2.1 Economic Interpretations of Sustainability Early classical interpretations of sustained economic growth were largely pessimistic in their assessment of human ingenuity in overcoming issues of resource scarcity and environmental degradation. For example, Malthus (1864/1 798) envisaged an absolute scarcity effect with the finiteness of natural resources constraining growth. World events, such as World Wars I and IT, the Great Depression, the rise of Communism, and strong economic growth in Western countries following World War IT, saw thinking on issues of resource scarcity wane. Understanding how economic production and consumption patterns might grow, or at least be sustained, to ensure human social welfare intertemporally, has been a primary focus of contemporary economic theorists (Ramsey, 1928; Domar, 1946; Hicks, 1946; So low, 1956, 1986; Kaldor, 1 957; Arrow, 1962; Dixit and Stiglitz, 1977; Hartwick, 1977; Dasgupta and Heal, 1979; Romer, 1986; Lucas, 1988). In this way, economists recognise that two distinct forms of sustainability operate over time, namely sustainable economic growth (increasing per capita production and consumption) and sustainable economic development (non-declining per capita production and consumption). Moreover, economists argue that sustainability requires capital stocks5 to be increased, or at least maintained, from one generation to the next. A key tenet of the contemporary economic view on sustainability is that growth, as well as maintenance, of capital stocks may be achieved by substitution between the different forms of capital. More contentiously, economists are optimistic that technological change will mitigate any resource scarcity or any degradation in the assimilative capacity of the environment. 2.1.1 Classical Economic Perspectives The Physiocrats, led by Francois Quesnay, believed agriculture to be the basis for prosperity. They envisaged land as the only true factor of production, generating more wealth than is invested in its tillage by labouring farmers (pribram, 1983; Spiegel, 1991; Victor, 199 1). Quesnay argued that all economic activities of production by humans should be governed by the 'natural order' (Foley, 1973). This implies that all individuals have the right to enjoy the fruits of their own labour, provided that such enjoyment is compatible with the rights of others (Schumpeter, 1954; Screpanti and Zamagni, 1993). Embedded in this thinking are the notions of intra- and intergenerational equity (Neumayer, 2003) . 5 Herfindahl and Kneese ( 1 974, p.68) define capital as "anything which yields a flow of productive services over time and which is subject to control in production processes". Four forms of capital are typically recognised: ( 1 ) manufactured - as created by economic activity, (2) natural - as provided by the environment, (3) human - knowledge and skills, and (4) cultural - social and institutional structures. 1 5 In his 1 776 work, The Wealth of Nations, Adam Smith envisaged arrangements in an economy organised on the basis of competitive markets, through which the selfish behaviour of individuals could serve the collective interest of society (Raphael, 1 985; Common, 1 988; Kula, 1 994). Smith also distinguished between the 'natural price ' of a commodity, determined by the amount of embodied labour, and the market price of that good, determined by its market scarcity (Hollander, 1 973; Raphael, 1 985). While Smith emphasised that agricultural productivity6 could be increased through capital accumulation, he did not, as Malthus, Ricardo and Mill later did, conceive of diminishing marginal returns7 from land-based production (Barbier, 1 989). In 1 798 Malthus published An Essay on the Principle of Population, in which he argued that the human population increases geometrically, whilst land reserves increase only arithmetically (Barnett and Morse, 1 963 ; Tisdell, 1 990; Ekins, 2000). Malthus considered population growth unsustainable because it exceeded the carrying capacity of the natural resource base, in contrast to the Physiocrats and Adam Smith, who had emphasised nature's unlimited endowments (Paglin, 1 96 1 ; Fisher, 1 98 1 ; Barbier, 1 989; Coombs, 1 990; Tisdell, 1 990). Malthus foresaw an absolute scarcity effect where the finiteness of land or other natural resources would act as a binding constraint on population and economic growth. Only once the entire available stock of the natural resource is utilised, is the Malthusian scarcity effect evident, usually manifesting as rising production costs or market prices. At this point, rising production costs are ineffective in encouraging substitution amongst factors of production. Consequently, economic activity is abruptly halted with no possibility of mitigating the negative effects of diminishing returns on economic output (Barbier, 1 989; Pearce and Turner, 1990; Kula, 1 994; Ekins, 2000). David Ricardo' s ( 1 973/1 8 1 7) analysis of the land sustainability problem was more rigorous than that of Malthus. He suggested that land and, by implication, other resources would be subject to diminishing marginal returns because the best land is exploited first. Rather than a catastrophic decline in human population as it overshot its carrying capacity, Ricardo foresaw a slow and gradual reduction in economic returns as the economy reached a stationary state . He concluded it would be necessary to invest greater amounts of capital and labour into progressively less fertile land (Pearce and Turner, 1 990; Tisdell, 1 990; Kula, 1 994; Neumayer, 2003), with food 6 Smith argued that the welfare of a nation's people, and the competitiveness of that nation in trade, would be determined by economic growth. According to Smith, the primary factors that promote the growing stock of a nation's wealth are the division of labour and capital accumulation (Hollander, 1 973 ; Oser and Brue, 1 988). 7 The law of diminishing returns suggests that increasing amounts of a variable input, such as labour or capital, combined with a fixed amount of another input, such as land, leads to a decline in the additional amount produced for each successive increment of the variable input (Barbier, 1989 ; Rees, 1 990; Young, 1 992). 1 6 resources ultimately diminishing over time as the population increased (Schumpeter, 1 954; Samuelson and Nordhaus, 1 985; Lee, 1 989). Thus, Ricardo viewed scarcity as a relative effect. As progressively inferior resources are utilised, the cost of production rises through increasing price movements. The economic system might respond by substituting other factors of production, such as labour or capital, for the depleting natural resource. Without such substitution, increasing resource scarcity might impede economic growth, and the economy may reach an undesirable stationary state (Samuelson and Nordhaus, 1 985; Barbier, 1 989; Kula, 1 994). In his Principles of Political Economy published in 1 848, Mill reformulated the Malthusian and Ricardian doctrines of resource scarcity. He particularly emphasised the concept of stationary stateS, the point at which economic progress would cease, which is a key principle in modem sustainability l iterature. Like Ricardo, Mill accepted the notion that diminishing marginal returns from agriculture would eventually suppress economic development and population growth (Hollander, 1 985; Coombs, 1 990; Kula, 1 994). Nevertheless, Mill supported the idea that the emergence of the stationary state might be alleviated by technological progress (Robinson, 1989; Coombs, 1 990; Kula, 1 994). He argued that before the arrival of the stationary state, the 'progress of civilisation' would attain a standard of living comfortable for all humankind (Schwartz, 1972). In 1 867, Karl Marx, in his work Das Kapital, shared the classical economist's outlook of a dismal future for the labouring classes, and utilised a labour theory of value.9 Marx made no mention of a constraint on economic production imposed by natural resource scarcity (Barbier, 1 989; Lee, 1 989; Robinson, 1 989; Tisdell, 1 990); however, according to his theory, nothing apart from human labour, including natural resources, can have value (Georgescu-Roegen, 1 97 1 ; Gowdy, 1 984; Lee, 1 989). Marx asserted that only natural resources in the form of raw materials have value, since only these possess an embodiment of labour. Marx perceived science as humanising nature, turning the inherent value of natural resources, which he considered to be 'gratis' iO, into use value (Georgescu-Roegen, 1 97 1 ; Gowdy, 1 984; Sowell, 1 985 ; Barbier, 1 989). 8 In his writings concerned with the sustainability of the economic system and human welfare, Herman Daly draws upon Mill' s doctrine ofthe stationary state. However, Daly argues that thermodynamic limits dictate ( 1 ) the maximum sustainable rate of resource throughput in the economy; (2) the extent of technological progress, and (3) the extent to which manufactured capital can substitute for natural capital within the production process (Daiy, 1 973, 1 987, 1 99 1 a, 1 992, 1 994). 9 Smith, Malthus, Ricardo and Mill subscribed to a labour theory of value, i.e. that the price of a commodity was determined by the cost of its production, with the ultimate cost to production being labour (TisdeU, 1 990; Spiegel, 1 99 1 ; Hodgson, 1 993; Screpanti and Zamagni, 1 993) . 10 In Marxist analysis, the term 'gratis' refers to those objects that are granted or gifted by nature (Georgescu-Roegen, 1 97 1 ; Barbier, 1989). 2.1.2 Emerging Neo-Classical Economic Perspectives 1 7 In 1 87 1 , Menger, in Principles of Economics, argued the possibility and importance of substitution between factors of production (Christens en, 1 989; Screpanti and Zamagni, 1 993). In his analysis, Menger assumed that the inputs of land, capital and labour into the production process could be varied, and smoothly substituted for one another (Alter, 1 982). Using the example of agriculture, he argued that economic output need not necessarily be constrained by the scarcity of any one productive input, such as capital or labour, since more land or fertiliser could be employed (Christensen, 1 989). In his 1 874 work Elements of Pure Economics, the French economist Leon Walras conceptualised machinery and land as forms of capital that produce consumer income and flows of producer services (Walras, 1 9541 1 874; J affe, 1 976; Christensen, 1 989). Walras also proposed the concept of general economic equilibrium 11 in which, through a framework consisting of the basic price and output interrelationships for the entire economy, he demonstrated mathematically that all prices and quantities produced would adjust to mutually consistent levels (Schumpeter, 1 954; Oser and Brue, 1 988; Screpanti and Zamagni, 1 993; Ruth, 2002). In 1 890 Marshall published his Principles of Economics, in which he agreed with Ricardo's idea that any scarcity of fertile land would lead to increased market prices for agricultural goods (Barbier, 1 989; Robinson, 1 989). However, Marshall argued that any negative effects resulting from resource scarcity and diminishing returns in land based production would be offset by improvements in the organisation and knowledge of agriculture, as well as technological innovation (Marshall, 1 94911 890; Barbier, 1 989; Pearce and Turner, 1990; Neumayer, 2003). Marshall, like Mill, acknowledged that nature provides aesthetic services to humans which conform strictly to the Ricardian doctrine of diminishing returns. He argued, however, that the use of non-market environmental services by humans should command a rent because some environmental externalities are not captured by market prices (Marshall, 1 949/1 890; Barbier, 1 989). 1 1 Krelle ( 1984) defines economic equilibrium as a state where the rate of capital accumulation equals the growth rate of labour p lus increases in labour productivity. 1 8 William Stanley J evons, an early marginalist12, was concerned with the issue of natural resource scarcity. In his work The Coal Question, Jevons ( 1 9091 1 865) asserted that the inevitable exhaustion of British coal reserves posed a threat to the nation's sustained economic growth (Barbier, 1 989; Nentjes and Wiersema, 1 992; Ekins, 2000). Jevons was pessimistic about the coal exhaustibility issue, and was convinced that no other fuel, not even petroleum, could substitute for coal in the future (Common, 1988; Barbier, 1 989; Kula, 1 994; Neumayer, 2003) . Between 1 890 and the early 1 970s, neo-classical economics, with its belief in the capacity of the market to ensure the efficient allocation of resources (allocative efficiency), came to dominate Western economic thought (Rima, 1 986; Common, 1 988; Victor, 1 99 1 ; Screpanti and Zamagni, 1 993). With the exception of Hotelling ( 193 1 ) and Scott ( 1 955), issues of sustainability were largely overlooked for the best part of the twentieth century. The main world events of this period, such as World Wars I and 11, the Great Depression and the rise of Communism, had little to do with the environment or the concept of sustainability. During this period, public policy and government was dominated by social issues such as unemployment and economic reconstruction, resulting in a lapse in any serious academic and theoretical investigation of sustainability issues. During the earlier half of the twentieth century, the macro-economic ideas of John Maynard Keynes strongly influenced government policy, resulting in the adoption of a new form of welfare economics by Western governments. Economic growth became the overriding concern in the 1 950s and 1960s, and was viewed by economists as the most important determinant of human welfare (Daly, 1 973 ; Mishan, 1 977; Bartelmus, 1 994; Reid, 1 995). Over this period, the issue of the ecological sustainability of economic activity was overtaken by a belief that the power of technology, human ingenuity and market mechanisms could overcome any natural resource or environmental constraint. This worldview was typified and supported by the evidence of Barnett and Morse ( 1 963), who tested the implications of natural resource scarcity on extraction costs and market prices. They found that natural resources were steadily reducing in price relative to labour and were in fact becoming less scarce. Barnett and Morse ( 1 963) concluded that the diminishing returns encountered when employing progressively inferior natural resources would be overcome by development or discovery of alternative resources of equal, or even superior, economic quality to those replaced. Studies undertaken by Weinstein and Zeckhauser ( 1 975), Pindyck ( 1 978) and Heal and Barrow ( 1 980) have corroborated these results, while analysts such as Solow ( 1 974), Stiglitz ( 1 974), Kamien 12 Around 1 870, classical economics began to transform into neo-classical economics, a process which involved two theoretical changes: economists abandoned the embodied labour theory of value that had been the mainstay of classical economic analysis, and the predominant method of analysis shifted to marginal analysis. 1 9 and Schwartz (1978) and Dasgupta and Stiglitz ( 198 1 ) have concluded that it may be optimal to deplete an exhaustible resource completely, if future technologies and perfect substitutes exist. However, Chapman and Roberts ( 1 983), Hall et al. ( 1 986), Norgaard ( 1 990) and Ekins (2000), note that the Barnett and Morse work assumes technological progress will continue to mitigate any resource scarcities. Although their evidence strongly suggests this is true for the period covered by the study, this does not prove that the trend will continue ad injinitum.13 2.1 .3 Neo-Malthusian View14 and its Repudiation Throughout the 1 960s and 1 970s, the economic growth objective and economists' faith in the capacity of the market to ensure efficient resource allocation, came under attack from non­ economists. This scepticism coincided with the emergence of the conservation movement and the realisation of several environmental problems which the market had obviously failed to mitigate (Jacobs, 199 1 ; Ayres, 1 993a; Reid, 1 995), including soil erosion, deforestation, desertification, the salinisation of land and water, and species extinction. During this period, ecologists actively questioned the compatibility of the economic growth objective with the continued preservation of the environment. I S The Limits to Growth study (Meadows et al., 1 972), financed by the Club of Rome, utilised a system dynamics model of the world's population and economic system. Reflecting the emerging neo-Malthusian view, the study forecast the eventual collapse of the global economy sometime in the 2 1 st century, resulting from a growing human population and economy outstripping the environmental carrying capacity. Meadows et al. ( 1 972) recommended that economic growth be abandoned as a key economic policy in industrialised nations. The Limits to Growth study received a largely hostile and highly critical reception from mainstream economists who argued that the model failed to account for the capacity of the market to compensate for exhausting resources (Beckerman, 1 974; Lecomber, 1 975, 1 979; Neumayer, 2003). An extensive and passionate l iterature followed in which the economists dismissed the doomsday arguments of the ecologists and other neo-Malthusians. 16 13 There may, for example, be thermodynamic l imitations to technological progress. 14 Further elaboration on these neo-Malthusian ideas is contained in Section 2 .2 . 1 . 1 5 Notable publications included The Silent Spring (Carson, 1 962), The Economics of the Coming Spaceship Earth (Boulding, 1 966), Fam ine (Paddock and Paddock, 1 967), Population Bomb (Ehrlich, 1 968), Tragedy of the Commons (Hard in, 1 96 8), Small is Beautiful (Schumacher, 1 973), A Blueprint for Survival (Goldsmith et al., 1 972), The Closing Circle (Commoner, 1 972), and The Poverty of Power (Commoner, 1 976). 16 In a comprehensive critique of the Limits to Growth study, Sussex University's Science Policy Research Unit (Cole et al., 1 973) re-ran Meadows' model with different assumptions and produced more positive results. They replaced the key assumption of absolute limits to natural resource availability with 20 Maddox ( 1 972) refuted the peSSImIsm of the Limits to Growth study in The Doomsday Syndrome, claiming that human welfare was improving, and that natural resources were becoming more abundant. 1 7 He further argued that human ingenuity would solve the problem of energy resource scarcity, and that countries suffering famine would produce food surpluses by the end of the 1 970s. Similarly, Simon, in The Economics of Population Growth ( 1 977) and The Ultimate Resource ( 1 98 1 ), argued that population growth was beneficial, as it provided a continual source of human ingenuity with which to overcome problems of resource scarcity. Goeller and Weinberg ( 1 976) also emphasised human ingenuity and technological progress in their article The Age of Substitutability, in which they described how technical change had led to alternatives in mercury-based technologies. Similarly, in his book In Defence of Economic Growth, Beckerman ( 1974) suggested that future technological efficiency would enable society to continue to pursue economic growth. Solow ( 1 974, p. l l ) emphasised the importance of substitutability, "If it is very easy to substitute other factors for natural resources, then there is in principle no problem. The world can, in effect, get along without natural resources, so exhaustion is just an event, not a catastrophe". The optimistic view on substitutability is that "reproducible capital is a near perfect substitute for land and other exhaustible resources" (Nordhaus and Tobin, 1 972, p.204). 18 In The Next 200 Years, Kahn et al. ( 1 976) argued strongly against the neo-Malthusian position. He optimistically maintained that the economy could grow almost indefmitely in the future, unconstrained by natural resource scarcity, leading to an increased standard of living for all. Later, Simon and Kahn (1 984) argued that the Earth's endowments of natural resources were more than able to meet the needs of a growing economy and human population. 2.1.4 Contemporary Neo-Classical Economic Perspectives Neo-classical economists have investigated how economic activities might be sustained using exhaustible and renewable natural resource stocks. This perspective views the environment as a source of resource inputs for economic production. Neo-classical economists have also theorised about the effects of pollution emissions on economic activities, regarding the the assumption of exponential increases in available natural resources through discovery, recycling and improved pollution control (Cole et al., 1973 ; Ayres, 1 993a). 17 Hall et al. ( 1 986) suggest that resources as a percentage of US GNP have risen from 4 to 8 percent from 1 973 to 1 9 85 after remaining constant for the previous 70 years. 18 Nordhaus later changed his views on substitution, stating that: "There are simply no substitutes for many of today's uses of fossil fuels" (Nordhaus, 1 990, p.20). 2 1 environment as a sink for economic waste (Dasgupta, 1 982; Klaassen and Opschoor, 1 99 1 ; Jongeneel, 1 992). In this perspective, pollution is considered the inevitable result of economic extraction, production and consumption activities (Baumol and Oates, 1979; Krelle, 1 984; Klaassen and Opschoor, 1 99 1 ). The neo-classical economic approach to sustainability represents the mainstream economic approach and is inherently optimistic. According to this approach, economic activities can be sustained indefmitely because of ( 1 ) market mechanisms which augment technological progress, and (2) substitution between factors of production (Victor, 1 99 1 ; Ekins, 2000; Pearce and Barbier, 2000; Booth, 2004). 2.1.4.1 Exhaustible Natural Resources and Sustainability The Hotelling Rule The Hotelling rule (after Hotelling, 1 93 1 ) has provided the basis for much of the contemporary neo-classical literature on resource scarcity (Solow, 1 974, 1 986; Stigl itz, 1 979; Fisher, 1 9 8 1 ; Siebert, 1 98 1 ; Kemp and Long, 1 984; Hartwick and Olewiler, 1 986). Drawing on the Ricardian doctrine of relative scarcity, Hotelling argued that the market price of an exhaustible natural resource, less its extraction costs, rises over time at a rate of interest commensurate with the increasing relative scarcity of that resource (Hotelling, 1 93 1 ; Ruth, 1 993, 2002; Common, 1 995; Ekins, 2000; Neumayer, 2003). The Hotelling rule, as recorded by Common and Perrings ( 1 992), can be written as, Pi(t! ) = r with Pi(t2 ) "* 0 , Pi(t2 ) (2. 1 ) where r is the interest rate or discount rate, and Pi(t) the in situ price of the ith resource at time t. The Hotelling rule suggests that as the relative scarcity of an exhaustible resource increases, its market price will also increase, with a corresponding fall in demand for that resource. According to Hotelling ( 193 1 ), the optimum rate of depletion!9 occurs when, simultaneously, a resource is exhausted and demand falls to zero, at which time the use of that resource in production ceases (Hotelling, 1 93 1 ; Barbier, 1 989; Ruth, 1 993, 2002). Combined with the work of Ramsey ( 1 928), whose approach required a pattern of investment in manufactured capital, the Hotelling rule may be used to sustain social welfare intertemporall o (So low, 1 974, 1 986; 19 Known as the ' rate of efficient intertemporal allocation' . 20 For a mathematical description of the Hotelling rule operating with the Ramsey rule, refer to Neumayer (2003 , pp.201 -208) . 22 Dasgupta and Heal, 1 979; Hartwick and Olewiler, 1 986). Furthermore, the Hotelling rule has been used by neo-classical economists in the search for empirical evidence of the market's ability to mitigate natural resource scarcity in the long term, and to find substitutes for exhaustible resources in the production process (Goeller and Weinberg, 1 976; Frank and Babunovic, 1 984). Authors like Smith ( 1 98 1 ), Bretschger ( 1 999), Ekins (2000) and Neumayer (2003) have questioned the validity of the Hotelling rule. For example, Smith ( 1 98 1 ) examined the price movements of twelve non-renewable resources (four fossil fuels and eight metals) and found that only two of the twelve supported the rule. He concluded that extraction costs, new discoveries, changes in market structure and institutional effects also play an important role in explaining price change. Bretschger ( 1 999) added lack of long-term ownership (shortening the time horizon for optimal use of the resource) and the existence of backstop technologies (which may substitute for the resource) as further reasons for price variation. Neumayer (2003) argued that the difficulty in measuring resource rent over time has been a major obstacle in empirical attempts to validate the Hotelling rule. The Hartwick Rule Hartwick's rule (after Hartwick, 1 977, 1978) states that the welfare of future generations may be satisfied by investing rents derived over time from the use of exhaustible natural resources into the acquisition of manufactured capital, ensuring the total capital stock remains unchanged. Dasgupta and Heal ( 1 979) model the Hartwick rule by employing the Cobb-Douglas production function2 1 : Y = f(K,R) = Ka1Ra2 with a I , a2 > 0 and a 1 + a2 < 1 , (2.2) where Y, K and R respectively represent an output of manufactured capital, an input of manufactured capital, and an input of an exhaustible natural resource. The f is used to denote a production function, and a the elasticities of substitution between inputs. This production function has two noteworthy properties. First, an input of natural resources is held to be essential for production, i.e. without such an input no output could be produced. Second, there are perfect substitution possibilities between all inputs into the production process, i.e. the elasticities of substitution, a1 and a2, always sum to greater than zero. 21 A production function states the functional relationships between inputs and outputs in the production process (Blaug, 1 985 ; Baumol et al., 1 991) . 23 It is the second property of the production function that is most contentious. Dasgupta and Heal ( 1979, p.200) debate the possibilities of such substitution, stating that the "crucial question is whether or not al>a2 . . . since these two parameters represent the elasticities of substitution with respect to manufactured capital and the exhaustible resource". If al>a2 then, despite any increasing scarcity of the natural resource R, manufactured capital can be considered sufficiently important in production to allow the possibility that the output level Y could be maintained permanently. However, if a2>al , i .e. the exhaustible resource is essential to production, output must eventually fall to zero. Dasgupta and Heal ( 1 979) argue optimistically that, given the technological progress of the last 300 years, they would expect that al>a2, possibly even four times greater?2 The Hartwick rule and the neo-classical position on economic sustainability rest upon two contestable assumptions: fIrst, to enable the production process to be sustained, manufactured capital can substitute for natural resources as a production input, and second, ongoing technological progress will at least keep pace with the depletion of exhaustible natural resources (Victor, 1 99 1 ; Munasinghe and McNeely, 1 995; Reid, 1 995; Neumayer, 2003). In the words of Dasgupta and Heal ( 1979, p.205), "even in the absence of technological progress, exhaustible resources do not pose a fundamental problem" if reproducible capital is sufficiently substitutable for natural resources. They further conclude that technological progress improves possibilities for sustainable increases in output. 2.1 .4.2 Renewable Natural Resources and Sustainability Early studies on optimal rates of renewable resource harvesting focused on forestry and fIshery depletion (refer to Gordon, 1 954; Scott, 1955; Faustmann, 1 968). These, and many later studies, looked at additional factors not normally considered in determining optimal rates of exhaustible resource depletion, such as natural (biological) growth rates, reproduction and mortality . Maximum Sustainable Yield 22 Victor ( 1 99 1 ) has argued that the elasticities of output formation might be used as indicators of sustainability since, according to neo-classical economists, it is these parameters that will alleviate, through substitution effects, any future resource scarcities or environmental degradation. Using mathematical models, he demonstrates several critical problems with elasticities of output as possible indicators. For example, he argues that assumptions of perfect markets are unrealistic. If markets are perfect, and natural resources are being allocated efficiently over time, i.e. without policy intervention, then sustainability indicators are unnecessary. However, markets are generally acknowledged as imperfect; this results in policy intervention and sustainability indicators are essential. Nevertheless, sustainability indicators based on economic valuation cannot be implemented as ultimately they must rest on the assumption of perfect markets - a prerequisite position that has already been shown to be wrong. 24 In neo-classical economiC analysis concerned with sustainability, a natural resource regeneration function is used to relate the growth rate of a renewable resource to the size of the resource stock. Klaassen and Opschoor ( 1 99 1 ) state that this regeneration function typically takes the form, S = dR = aR(I - R ) with C ;;t. 0 , dt C (2.3) where S denotes the increase in the stock of the renewable resource over time, a the coefficient of natural growth rate, R the resource stock, and C the environmental carrying capacity. Assuming that the renewable resource is essential for production and that human population growth is zero, neo-classical economists conclude there is a maximum level of resource harvest that can be sustained indefinitely, termed the Maximum Sustainable Yield (MSYi3 (Clark, 1 976; Baumol and Oates, 1 979; Conrad and Clark, 1 987; Daly, 1 994; Costanza and Patten, 1 995). MSY depends on the growth rate, a, and the size, R, of the resource stock (Clark et al., 1 985; Conrad and Clark, 1 987). If MSY exceeds present levels of harvest, then the resource stock will grow to an optimal point where the regeneration rate is equal to the rate of harvest (Turner, 1 988). If economic extraction exceeds MSY, then resource saving technical change or substitution of manufactured capital for the resource, would allow production and consumption to be sustained and resource scarcity to be mitigated, even if the resource was completely depleted (Herfindahl and Kneese, 1 974; Baumol and Oates, 1 979; Dasgupta and Heal, 1 979; Kemp and Long, 1 984). Sustainable Economic Welfare Siebert ( 1 98 1 , 1 995/4 uses the fol lowing welfare function to model sustainable economic development with renewable natural resources : W = U(Ct, R)e-r1dt , (2.4) 23 MSY is an ecological term that represents the theoretical point at which the size of a population is such as to produce a maximum rate of increase. 24 Siebert ( 1 98 1 , 1 995) was amongst the first to formalise this model of sustainable economic development - recent developments in the field are discussed by Pezzey ( 1 997). Other authors have developed similar indices. Daly and Cobb ( 1 989), for example, based on the work of Nordhaus and Tobin ( 1 972), advocate the use of an Index of Sustainable Economic Welfare (ISEW). The ISEW has subsequently been reformulated as the Genuine Progress Indicator (GPI) by including additional costs and benefits. 25 where W represents economic welfare, U utility, C consumption, R a renewable resource stock and rt a social discount rate. In Siebert's model resource regeneration, R', is a function of the natural growth rate less total extraction, Xt (the sum of extraction, X, and consumption, C), while net capital accumulation, K, is a function of production less consumption, K=fCK, X)-C. Using these functions, Siebert ( 1 98 1 ) investigates the potential for sustainable economic growth given the possible use of the renewable resource as ( 1 ) a consumer good and (2) a factor of production. He concludes that, in the absence of technical progress or human population growth, the rate of use of a renewable resource as a consumer good is optimal when the rate of resource regeneration equals the rate of social discount, R'=rt. If the social discount rate25 exceeds the maximum rate of regeneration, rt>R', the renewable resource will become depleted, resulting in a decline in consumption over time. If the discount rate is less than the rate of resource regeneration, rt 0 if S > (DM + DN J with Y :;t: 0 , Y Y Y (2 . 7) where Z is the sustainability index for a nation, S the savings or the accumulation of capital, DM the value of depreciation on manufactured capital, DN the value of depreciation on natural capital, and Y income. Utilising the above savings rule, Pearce and Atkinson ( 1 993) formulate a weak sustainability indicator for a nation's economy as: S DM DN . Z = - -- - - WIth Y :;t: O . Y Y Y (2.8) Using the same notation as Equation 2.7, Z produces a measure for deviation from marginal sustainability for a nation. The larger any negative value of Z, the greater the effort required to return the economy to a sustainable path relative to national income (Pearce and Atkinson, 1 993). Strong Sustainability Pearce and Turner ( 1 990) of the London Environmental Economics Centre (LEEC) present a strong sustainability interpretation, arguing that a prerequisite for achieving sustainable economic development is the maintenance of a constant natural capital stock.29 They challenge the weak sustainability argument, presenting five reasons why the substitution of manufactured capital for natural capital within production processes is often impossible: ( 1 ) manufactured capital is not independent of natural capital because the latter is often needed to create the former; (2) natural capital fulfils other economic functions including basic life support, and is multifunctional to an extent not shared by manufactured capital; (3) natural capital cushions the economy against environmental shocks; (4) natural capital stock should be maintained over time to ensure intergenerational equity; and (5) conservation of the natural capital stock recognises economy can only be weakly sustainable if GSP2• Energy released in the transformation between PI and P2 takes the form of heat transferred, Ql2, and work done, W12, with the subscripts representing transfers between states 1 and 2 (Burness et al. , 1 980; Ruth, 1 993). Heat and work represent the flow of energy, while internal energy represents a stock (Kneese et al., 1 970; Georgescu-Roegen, 197 1 ; Ayres and Nair, 1 984). Given the conversion process described above, according to Burness et al. ( 1980), the first law establishes conservation of energy in a system as,43 (2. 1 2) The first law states that when a system undergoes a process, the energy transferred across the system boundary as either heat or work, is equivalent to the net change in the internal energy of the system. Thus, energy in any transformation process is not lost, but conserved, i .e. its form is simply altered (Ayres and Nair, 1 984; Ruth, 1 993). The Second Law of Thermodynamics has been termed the 'entropy law' (Georgescu-Roegen, 1 97 1 ). In order to use the second law in a quantitative sense, thermodynamicists introduce a working variable named 'entropy'. The second law has been expressed and utilised in different ways by different academic disciplines44 (O'Connor, 1 99 1 ) . According to classical 42 Three types of system are recognised in thermodynamic analysis: ( 1 ) an isolated system, in which neither energy nor matter crosses the system boundary, (2) a closed system in which only energy, but not matter, crosses the system boundary, and (3) an open system, in which both matter and energy cross the boundary of the system (Faber et al., 1 987; Binswanger, 1 993 ; Ruth, 1 993). 43 The changes in the coal's kinetic and potential energy are considered to be negligible. 44 Binswanger ( 1 993) lists five bodies of theory that utilise the concept of entropy. These include: ( 1 ) classical thermodynamic theory, which uses measurable macroscopic variables such as volume and temperature to describe isolated systems in a state of thermodynamic equilibrium, (2) statistical mechanics which describes the probabilistic behaviour of an ideal gas in an isolated system at a 40 thermodynamic theory, the entropy law describes, for an isolated system, the irreversible dissipation of (Gibbs free) energy - the result of which is an increase in the system's entropy (Prigogine, 1 967; Georgescu-Roegen, 1 97 1 ; Denbigh and Denbigh, 1 985; Binswanger, 1 993; Ruth, 1 993). That is, if a system is initially in a low-entropy (ordered) state, its condition will tend to move involuntarily toward a state of maximum entropy (disorder). Under the fIrst law, energy transfers across a system's boundary are represented as heat and/or work. Nothing prevents a process from being undone or reversed. The second law however dictates that every process that a system may undergo, be it physical, biological, technological or otherwise, can go in only one direction, and that the opposite process, which would return the system and its surrounds to their original states, is impossible. According to classical thermodynamics, an increase in entropy, ds, of an isolated system between two states, SI and S2, may be written as, ds = Q r ' (2. 1 3) where Q represents incremental and infmitesimal heat transfer, and T represents the absolute temperature of the system when Q is absorbed. Summation of all infInitesimal entropy increases, ds, allows for the specification of the total increase in system entropy, S, as, (2. 1 4) In classical theory, the entropy law measures the exchange of heat45 between systems that are considered to be in a state of thermodynamic equilibrium. Under a classical thermodynamic expression of the entropy term, reversible processes46 may exist - characterised by infinitesimal changes in macroscopic variables which occur infInitely slowly (Denbigh and Denbigh, 1 985; Faber, 1 985; Binswanger, 1 993). Nevertheless, all macroscopic processes operating in open microscopic level, (3) information or negentropy theories, concerned with the information content of messages and systems (Shannon and Weaver, 1 949; Brillouin, 1 962), (4) the theory of dissipative structures, which draws upon classical thermodynam ics and statistical mechanics to describe the evolution of open systems toward higher complexity, in a state far from thermodynamic equilibrium (Prigogine, 1 967), and (5) thermodynamic evolutionary theories in biology and ecology, that utilise both classical thermodynamics (Lotka, 1 922), and the theory of dissipative structures (Odum, 1 969). 45 The presence of a temperature gradient, heat flow from a hotter to a colder body, is considered to be a prerequisite for the performance of work. Temperature, T, is a qualitative dimension of heat flow, Q (Denbigh and Denbigh, 1985 ; Binswanger, 1 993). 46 This implies that any transformations of a system during a process, measured as changes in macroscopic thermodynamic variables, can be completely undone (Binswanger, 1 993). Reversible processes within isolated systems are theoretical ideals, ignoring friction, heat transfer across finite temperature boundaries, mixing, inelastic deformation or free expansion (Ruth, 1 993). 41 and closed economic and ecological systems are irreversible. Several theorists therefore consider it inappropriate to use a classical thermodynamic expression of the second law to quantify entropy (Faber, 1 985; Khalil, 1 990; B inswanger, 1993 ; Ruth, 1 993). Ruth ( 1 993), for example, proposes that the entropy law for such systems, involving irreversible processes, must be written as, s = Q with S > 0, T (2. 1 5) where S denotes a change in system entropy, Q heat transfer to, or from, the system, and T the absolute temperature of the part of the system to which heat Q is transferred. Prigogine ( 1 967, 1 973) and Prigogine et at. ( 1 972) have reformulated and extended the classical thermodynamic, and statistical mechanics47, conceptions of entropy. Prigogine's work provides an explanation for the irreversible processes associated with open systems. According to Prigogine's ( 1 967) theory of non-equilibrium thermodynamics, entropy change, ds, may be reformulated as, (2. 1 6) where dSe represents the flow of entropy resulting from exchanges with the environment of the system48, and dSj the production of entropy as a consequence of irreversible processes inside the system. The term dSe may be either positive or negative in value - when negative, entropy inside the system, dsj, will decrease (prigogine, 1 967; Binswanger, 1 993). In such a case, according to Binswanger ( 1 993), one of three distinct forms of entropic interaction between the system and the environment may exist: ( 1 ) entropy is increasing inside the system at a rate faster than the flow of negative entropy to the environment i.e. -dse < dSj with ds > 0, (2) the flow of negative entropy to the environment equals the production of entropy within the system, so the entropy level of the system remains constant i.e. -dse = dSj with ds > 0, or (3) the entropy of the system is decreasing, but the flow of negative entropy to the environment exceeds the production of entropy within the system, resulting in a system moving away from 47 Statistical mechanics suggests that systems do not tend to go into states that are less probable than those they are already in (Binswanger, 1 993). Statistical mechanics, unlike classical thermodynamics, acknowledges that systems may be in non-equilibrium states, providing a microscopic basis for irreversibility (Prigogine, 1 973 ; Harrison, 1 975; Wicken, 1 988). 48 Under Prigogine's theory this flow may, in the case of an isolated system, be zero i.e. dse=O. Thus ds would not only equal dsj, but would be positive in value, or zero if the system is in thermodynamic equilibrium (Prigogine, 1 967; Binswanger, \ 993). 42 thermodynamic equilibrium i.e. -dse > ds; with ds < O. In this way, Prigogine ( 1 967) defines the second law for irreversible processes, including open systems, as ds; > 0 . It is the third form of entropic interaction noted above that is often used in explaining the evolution of open systems towards a state of higher complexity (Prigogine, 1 967; Binswanger, 1993 ; Ruth, 1 993). Open systems, far off thermodynamic equilibrium, may only establish and sustain their low entropy states by creating flows of negative entropy to their environment via the dissipation of energy and matter (Prigogine, 1 967; Glansdorff and Prigogine, 1 97 1 ; Prigogine et al., 1 972). Such systems, following Schrodinger ( 1 944), are termed 'dissipative structures , .49 The flow of negative entropy from a dissipative structure always results in an increase in the entropy of the system's environment. The Third Law of Thermodynamics further constrains the interplay between heat, temperature and entropy. The third law states that as absolute zero (approximately -273 °C) is approached, the extraction of energy from a system or its environment becomes increasingly more difficult. 2.3.1 . 1 The First Law of Thermodynamics and Sustainability During the 1 960s and 1 970s, environmental economists employed the first law (and the mass conservation principle) to characterise the relationship between the scale of economic activity and environmental quality5o,5 1 (Barbier, 1 989; Common, 1 995). Ayres and Kneese ( 1 969), for example, utilise the so-called 'materials balance principle' in their interpretation of the first law. This principle implies that, barring accumulation in the production process, all materials extracted or harvested from the environment for use in economic activity, must ultimately, in mass terms, be returned to the environment in the form of unwanted products and wastes. Material and energy residues generated by the economic system and emitted into the environment as pollution or waste must therefore (by assuming no accumulation) be equal to those initially extracted from the environment (Ayres and Kneese, 1 969; Kneese et al., 1 970; Victor, 1 972a). 49 Thermodynamic potentials between systems and their environment, as measured by temperature, concentration or pressure gradients, must be sufficiently large to permit flows of negative entropy, -dse, to the system environment. When such thermodynamic potentials are exceeded, the system may move to a new state further from thermodynamic equilibrium, where new structures may evolve to dissipate higher levels of negative entropy (Prigogine et ai., 1 972; Prigogine and Stengers, 1 984; B inswanger, 1 993). 50 Strictly speaking, the First Law of Thermodynamics only applies to the conservative nature of energy transformations. In Section 2 .3 . 1 . 1 , however, the term 'first law' is used to collectively refer to both the energy and mass conservation principles. 51 Including studies by Ayres and Kneese ( 1 969), Kneese et al. ( 1970), Converse ( 197 1 ), Victor ( 1 972), Cumberland and Korbach ( 1 973), d'Arge and Kogiku ( 1 973), Nijkamp ( 1 977), Ayres and Noble ( 1 978), Kneese and Bower ( 1 979), and Iohnson and Bennett ( 1 98 1). 43 From these first law interpretations (energy and mass conservation), arose a conceptualisation of the economy used by many ecological economists. The economy is seen as an open system embedded within the global biophysical system - Gilliland ( 1977) was amongst the first to formally propose this theoretical schema. The economic system is therefore viewed as being nested within the greater ecological system, ultimately transforming inputs of low entropy energy (e.g. fossil fuels) and matter (e.g. minerals) into outputs of highly degraded entropy, like manufactured goods and emissions, that flow into the environment (Colby, 1 99 1 ; Ekins, 1 994; Reid, 1 995; Wetzel, 1 995). The biophysical (thermodynamic) VIew of sustainability leads to an appreciation of three important constraints on the sustainability of the economic system, as identified by Patterson (2002b) among others: • Resource (input) constraints. The physical growth of the economic system depends on the continuous flow of materials and energy inputs into the system from the biosphere. Many resources are c learly fmite given first law considerations and are therefore depletable, e.g. fossil fuels, minerals. If the stock of these resources is depleted or degraded, economic growth cannot be sustained indefmitely; • Waste/residual (output) constraints. In a physical sense, sustainability of the economic system depends on the ability of the biophysical environment to absorb and purify wastes/residuals produced by the economy, i.e. the economy relies on the sink functions of the biophysical environment, such as efficient purification and absorption of wastes and emissions. Critical thresholds exist, however, beyond which the environment may not cope with ever increasing wastes, which lead to local scale impacts like eutrophication through to global impacts like c limate change;52 • Size/scale of the economic subsystem. In a physical sense the embedded economic subsystem cannot exceed the size of the biosphere space which it occupies. For example, Vitousek et al. (1 986) estimate that the economy has appropriated 40 percent of the net primary productivity of the terrestrial biosphere. The ultimate physical limit cannot exceed 1 00 percent, and to provide a safety margin, it has been argued that a 52 One possibility to reduce flows of wastes/residuals into the biophysical environment from the economy is to recycle materials (Biancardi et al., 1 993, 1 994, 1 996; Khalil, 1 994; Mansson, 1 994). Authors such as Georgescu-Roegen ( 1 976), Kummel ( 1 994) and Converse ( 1 996) however argue that 1 00 percent recycling is a physical impossibility. The Second Law of Thermodynamics, for example, tells us that energy cannot be re-used or recycled, in the sense that once, say, a piece of coal is burnt, the same amount of 'useful' energy cannot be extracted from that piece of coal. Furthermore, the costs of recycling material are likely to become prohibitively high as the recycling rate tends to 1 00 percent (Neumayer, 2003). 44 limit of 80 percent is more realistic. Any increase in scale, as measured in terms of economic production and consumption, will also result in a corresponding increase in the required quantity of environmental material and energy inputs. In turn, increased throughput of materials and energy resources through the economy will lead to a corresponding increase in the pollution loading placed on the environment (Barbier, 1 989; Klaassen and Opschoor, 1 99 1 ; Victor, 1 99 1 ). In summary, the capacity of the environmental system to provide resources and assimilate waste governs the maximum sustainable physical scale of the economy. As Daly ( 1 994) argues, according to the first law, economic growth can only occur at the expense of the environment. Any increase in the physical dimensions of the world economy, as a subsystem of physical Earth, therefore implies a corresponding decrease in the physical size of the environment, s ince the Earth has only a finite mass. Accordingly, Kamien and Schwartz ( 1 982) have argued that the first law thus imposes an upper limit on the extent to which manufactured capital can substitute for natural capital as a factor of production. Ecological economists, such as Goodland and Daly ( 1 993), therefore suggest an economic policy goal ?f minimisation of energy and material throughput in order to maintain the physical sustainability of the economic system. 2.3.1.2 The Second Law of Thermodynamics and Sustainability While many economists interested in thermodynamics have concentrated on the first law, theorists such as Georgescu-Roegen ( 1 97 1 ), Daly ( 1 977) and Perrings ( 1 987) have insisted that it is the second law which is of most significance. Georgescu-Roegen ( 1 97 1 , 1 976, 1 977a) has argued that the entropy law can be applied to the transformations of energy and matter53 that occur within the economic system. The argument follows that such transformations always involve the degradation of high quality energy and matter (low entropy) forms of higher economic value (highly organised materials and energy), to low quality (high entropy) forms of lower economic value (highly disorganised materials and energy) (Georgescu-Roegen, 1 97 1 ; Daly, 1 973, 1 99 1 a; Perrings, 1 987; Barbier, 1 989; Victor, 1 99 1 ). The second law implies that complete recycling of economic wastes and residuals is a physical impossibility (Georgescu- 53 Georgescu-Roegen ( 1 977a, 1 977b, 1 977c, 1 979b) has formulated a ' fourth law' of thermodynamics, which he argues governs all economic activity. He argues that "matter matters too" ( 1 979, p . l 039) and therefore in a closed system, such as biosphere, "material entropy must ultimately reach a maximum" ( 1 977b, p.269). The notion of material entropy has however been hotly debated (see, for example, Bianciardi et al. ( 1 993, 1 994, 1 996), Khalil ( 1 994), Kummel ( 1 994), Mansson (1 994), Converse ( 1996), and Coming (2002» . At the crux of the debate is the assumed isomorphism between energetic order and physical order. 45 Roegen, 1 97 1 , 1 976; Booth, 2004).54 At each successive stage of an economic process, the entropy of matter and energy engaged within that process increases in an irreversible manner, making that matter and energy progressively less useful in future economic activity55 (Daly, 1 987; Barbier, 1 989, 1 990; Klaassen and Opschoor, 1 99 1 ; Victor, 1 99 1 ). Taking the longest of long views, the second law acts as the ultimate regulator of all activity, be it economic or otherwise; thus sustainability is inherently unachievable, even theoretically meaningless (Victor, 1 99 1 ). Daly ( 1 99 1 a, 1 992, 1994) takes a more pragmatic approach in his application of the second law to sustainability issues. He argues for a steady state economic system based on the thermodynamic constraints that the first and second laws impose on the overall sustainable scale of macro-economic activity. Daly ( 1 973, p.98) defmes the desirable steady state economy as "an economy in which the total population and the total stock of physical wealth are maintained constant at some desired levels by a minimal rate of maintenance throughput (Le. by birth and death rates that are equal at the lowest feasible level, and by physical production and consumption rates that are equal at the lowest feasible level)". Daly ( 1 987, 1 99 1 a, 1992, 1 994) advocates that the achievement of economic sustainability will require a decrease in the current rate of matter and energy throughput into the economic system, such that the overall scale of the economy corresponds with the carrying capacity of the global environment. Descaling the economy will require zero economic growth and near zero human population growth (Daly, 1 99 1 a, 1 992; Turner, 1 993 ; Ekins, 2000). To this end, Daly ( 1 99 1 b) provides four operational principles for economic sustainability: ( 1 ) limit the scale of human, including population and macro-economic, activity to a level which is within the carrying capacity of the Earth; (2) technological progress should be efficiency increasing rather than throughput increasing; (3) the rate of harvest of renewable resources should not exceed the regeneration rate of those resources, and the generation of economic waste emissions should not exceed the waste assimilating capacity of the environment; and (4) non-renewable resources should be exploited, but at a rate equal to the creation of renewable substitutes. 54 Georgescu-Roegen (197 1 , 1976) asserts that complete recycling of matter is impossible in a closed system. This assertion has been challenged, as part of the debate on Georgescu-Roegen's proposed fourth law of thermodynamics, by several authors (see, for example, Bianciardi et al. ( 1 993, 1 994, 1996), Khalil ( 1994), Mansson ( 1 994» . B ianciardi et al. (1 993, 1 994, 1996) and Khalil ( 1994, 1 995, 1 997), for example, argue that Georgescu-Roegen' s assertion cannot be theoretically grounded as a physical principle. Nevertheless, they also point out that complete recycling is practically impossible, "[complete recycling] would involve a tremendous increase in the entropy of the environment, which would not be sustainable for the biosphere" (Bianciardi et al., 1 996, p . l95). 55 Many neo-classical economists assume the possibility of continual resource recycling in their analysis of sustainable economic growth (see, for example, Maler ( 1 974» . 46 Norgaard ( 1986) is critical of Daly's argument in favour of a low throughput economy. Norgaard ( 1 986) argues that since increasing entropy is not simply just an i l l effect of human activity, but also a natural process, there is no critical level below which global entropy will not increase. Daly ( 1 987, 1 99 1a, 1 992) ignores or glosses over the Earth's annual influx of solar energy, which potentially could fix energy into Earth matter, countering any entropic degradation brought on by natural or economic processes (Norgaard, 1 986; Klaassen and Opschoor, 1 99 1 ). Norgaard ( 1 986) argues that we could make far more efficient use of solar flux than we do now, offsetting any entropy increases effected by economic activity. 2.4 Key Principles for Assessing Auckland Region's Sustainability This Section presents eight theoretical principles for progressing toward sustainable development. The principles, which are described in detail in Table 2. 1 , are: • Principle 1 Maintain total capital stock within the limits of critical natural capital ('Weak Sustainability ') • Principle 2 Substitute non-renewable with renewable natural capital • Principle 3 Maintain human activity within the carrying capacity of the environment • Principle 4 Maintain environmental life support services • Principle 5 Maintain the assimilative capacity of the environment • Principle 6 Use natural capital efficiently Principle 7 Minimise material and energy throughput through the economy • Principle 8 Maintain the resilience of the ecological-economic system These principles are based on the economic, ecological, and thermodynamic interpretations of sustainability outlined in the preceding Sections. They advocate the theoretical means by which the concept of sustainability might be operationalised or implemented in the Auckland Region. A rationale for each principle is provided, along with supporting l iterature and theorists, and a guide to its application in relevant subsequent thesis Chapters and Sections. Connections and interdependencies between the principles are described by Figure 2.3 . --- Table 2.1 Definition Rationale Key Supporting References Application of the Principle in this Thesis Notes: Theoretical Principles for Sustainable Development and their Application in this Thesis Principle 1: Malntaln1 total capital stock within the limits of critical natural capital ('Weak Principle 2: Substitute non·renewable with renewable natural capital Sustainabllity') The total capital stock, including natural, manufactured and social2 capital must be maintained inter-generationally. Substitutability between different forms of capital is permissible . .. The provision of a constant stock of capital ensures that the economic welfare of current and future generations is met. Solow (1 974, 1 986), Hartwick (1 977, 1 978), Holling (1 986, 1 995), Perrings (1 987, 1 99 1 ) , Pearce and Turner (1 990), Daly (1991a, 1 994), Victor ( 1 99 1 ) , Ehrlich (1 994), Jansson and Jansson (1 994), Costanza and Patten ( 1 995), Munasinghe and McNeely (1 995), Ekins (2000, 2003), Pearce and Barbier (2000), Neumayer (2003) • Chapter 4: Establish framework for static flow analysis of economic and biophysical flows, including depletion and formation of manufactured and natural capital stocks .. Chapter 5: Establish flow estimates of the depletion and formation of manufactured capital stocks .. Chapters 6 & 7: Establish flow estimates of the depletion and formation of natural capital stocks e.g. water, energy, land, ecosystem services, minerals, forestry, fisheries, etc . • Chapter 1 0 : Identify key drivers of change that influence Auckland Region's manufactured and natural capital stocks .. Chapter 1 1 : Conceptualise the dynamics of the interdependencies between Auckland's manufactured and natural capital stocks. Scenario modelling of policy options that influence natural capital stocks In economic processes, substitute non-renewable resources with renewable resources. The greater use of renewables: • inherently involves the continuous recycling of mass, which has two key advantages: ( 1 ) wastesfresiduals are reused , thereby reducing the environmental loading b y these wastesfresiduals, (2) the stocks of renewable resources are not depleted, as they are being continuously regenerated by these recycling processes. This is not the case with renewable resources • reduces the pressure placed on the depletion of non-renewable resources Clark (1 976), Baumol and Oates ( 1 988), Pearce and Turner ( 1 990), Turner (1 993), Pearce and Barbier (2000), Booth (2004) • Chapter 1 1 : Dynamic modelling of Auckland Region's ecological-economic interactions, including the possibility of substituting non-renewable for renewable resources. Scenario modelling of policy options, including the implications of substituting non-renewable with renewable natural capital 1 . The term 'maintain', as used in the first five principles, refers to maintenance of both quali tative and quantitative aspects. 2. While it is acknowledged that the total capital stock includes social ( i .e . human, cultural, etc.) capital, these forms of capital receive only cursory consideration in the remainder of this thesis. Table 2.1 Theoretical Principles for Sustainable Development and their Application in this Thesis (Continued) Definition Rationale Key Supporting References Principle 3: Maintain human activity within the carrying capacity of the environment Human activity must be maintained within the supportable bounds of the environment. Carrying capacity refers to the maximum number of people that the environment can support, without reducing its ability to support future generations. Carrying capacity implicitly allows for the continued co-existence of other life forms. The unique ability of humans to appropriate resources from elsewhere, and smooth out economic or environmental fluctuations, must also be taken into consideration when determining carrying capacity. Maintenance of human activity within the carrying capacity of the environment • recognises the finite capacity of the environment in providing natural resources and ecosystem services, assimilating waste residuals, absorbing stochastic environmental shocks or disturbances, and in providing life support services • recognises that exceeding carrying capacity could irreversibly degrade the stock of natural capital Holdren and Ehrlich (1 974), Ehrlich and Ehrlich (1981) , Daily and Ehrlich (1992), Holdren et al . (1 993), Ehrlich (1 994), Wackernagel and Rees (1 996), Wackernagel et al . ( 1999, 2002), Rees (2000) Application of • Chapter 8: Estimate Auckland Region's Ecological Footprint using an input-output approach. the Principle This includes estimation of land, energy and CO2 resource/residual footprints for the region in this Thesis Principle 4: Maintain environmental life support services Environmental life support services must be maintained. Such services encompass those biogeochemical processes and ecosystem services functions essential for al l forms of life. Environmental life support services • provide refuge and reproduction habitats that contribute to the in situ conservation of biological, genetiC and evolutionary processes (de Groot et a/ . , 2002) • provide production functions that enable conversion of biomass and energy into carbohydrate through the process of photosynthesis, in turn providing food for human consumption (de Groot et a/ . , 2002) • provide opportunities for reflection, spiritual enrichment, cognitive development, recreation and aesthetic experience (de Groot et a/ . , 2002) • protect the biosphere from risk of collapse or decline • protect the biosphere's self-organising and species-supporting capacity • cannot be replicated, or substituted for, by other forms of capital • possess intrinsic value in their own right Barbier (1 989, 1 990), Folke et al . (1 989), Folke (1991) , de Groot (1 992), Berkes and Folke (1 994), Costanza et al . (1 997), Folke et al . (1 997), Ekins (2000, 2003), de Groot et al . (2002) • Chapter 4: Establish framework for analysis of ecological interdependencies between biosphere stocks and processes that operate within the Auckland Region • Chapter 6: Estimate biophysical waste/residual fluxes associated with economic activity on an industry- and economy-wide basis • Chapter 7: Assess the value that ecosystem services, including processes providing life support, make to the Auckland regional economy using a TEV approach. Estimate the mass flux associated with biogeochemical cycing in the Auckland Region • Chapter 1 0: Identify the key drivers of human controlled change that influence Auckland Region's environmental life supporting capabilities • Chapter 1 1 : Dynamic modelling of biophysical waste/residual fluxes associated with economic activity • Appendix B: Dynamic modelling of global biogeochemical processes Table 2.1 Definition Rationale Key Supporting References Application of the Principle in this Thesis Theoretical PrinCiples for Sustainable Development and their Application in this Thesis (Continued) Principle 5: Maintain the assimilative capacity of the environment The assimilative capacity of the environment must be maintained. Assimilative capacity refers to those biogeochemical processes and ecosystem services that ameliorate the impacts of residuals (i.e. pollution, wastes and emissions) generated from human activities. Cleansing of sudden stochastic events, decomposition of biological materials and the like are similarly included. The assimilative capacity of the environment • provides regeneration functions that maintain a healthy environment providing clear air, water and soil, and biological control services • contributes to the continued regeneration of renewable resources • protects the biosphere from risk of collapse or decline de Groot (1 992), den Elzen et al . (1 995), Costanza et al . (1 997), Ekins (2000, 2003), de Groot et al . (2002) • Chapter 7: Assess the value that ecosystem services make to the Auckland Region using a TEV approach, including processes assimilating residuals and absorbing natural shocks/perturbations. Estimate the mass flux associated with biogeochemical cycing in the Auckland Region • Chapter 1 0: Identify the key drivers of human induced change that influence Auckland Region's assimilative capacity • Appendix B: Dynamic modelling of key biogeochemical cycles, including assimilative processes PrinCiple 6: Use natural capital effic iently The efficient use of natural capital in economic production requires minimisation of resource input per unit of output. Similarly, the efficient use of natural capital in human consumption activities requires maximisation of economic welfare per unit of natural resource input. The most efficient processes may involve a utocatalytic feed backs, giving the maximum power per unit of energy input (Odum, 1 996). Using natural capital effiCiently • preserves stocks of natural capital for future generations • provides opportunities for enlargement of the natural capital stock • reduces economic waste associated with the use of natural capital, resulting in a cleaner environment and higher quality of life support systems • reduces the potential impact on the regeneration rates of other forms of natural capital Herfindahl and Kneese (1 974), Dasgupta and Heal ( 1 979). Baumol and Oates (1 988), Young (1 992), Turner (1 993), Bartelmus (1 994), Pearce and Barbier (2000), Booth (2004) • Chapter 4: Establish framework for static flow analysis of economic and biophysical flows • Chapter 5: Estimate gross output, gdp and bot on an industry basis. This information will be used in conjunction with biophysical information to estimate eco-efficiency indicators • Chapter 6: Estimate biophysical flows associated with economic activity on an industry­ and economy-wide basis • Chapter 6: In conjunction with the economic and biophysical fluxes established above, derive eco-efficiency on an industry- and economy-wide basis e.g. resource input per $ outpuVgdp, residual outputs per $ outpuVgdp, total embodied resource requirements per total embodied dollars generated, total embodied residuals output per total embodied dollars generated • Chapter 8: Estimate Ecological Footprint of the Auckland Region and establish level of interdependence with other New Zealand regions Table 2.1 Definition Rationale Key Supporting References Application of the Principle in this Thesis Theoretical Principles for Sustainable Development and their Application in this Thesis (Continued) Principle 7: Minimise material and energy throughput through the economy Economic throughput of materials and energy must be curtailed inter-generationally. Throughput requires the extractionlharvest of natural resources, manufacture or consumption of an economic product and the generation of residualslwastes. Minimisation of material and energy throughput • provides a buffer of natural capital for protection against stochastic environmental change • ensures that materials and energy are available over and above those required for carrying capacity, life support and so on. This provides a reserve for currently unanticipated future requirements Le. is prudently pessimistic, or precautionary, to accommodate unforeseen applications of the various forms of capital • ensures that the scale of human activity remains not only within the bounds dictated by the finiteness of the biosphere, but also provides a reserve • recognises, in accordance with the first law of thermodynamics, that all material and energy inputs into the economy must ultimately (Le. assuming no storage) equate to the q uantity of residual wastes 'spat out' into the environment • recognises, in accordance with the second law of thermodynamiCS, that increased throughput accelerates the entropic degradation of useful matter and energy Daly (1973, 1 99 1 a , 1 992), Ekins (2000, 2003), Pearce and Barbier (2000) • Chapter 4: Establish framework for static flow analysis of economic and biophysical flows • Chapter 6: Benchmark material throughput, in mass terms, using an Auckland Region PlOT • Chapter 1 1 : Dynamic modelling of Auckland Region material throughput. Scenario modelling of policy options, including the implications on material throughput Principle 8: Maintain the resilience of the ecological-economic system The resilience of the ecological-economic system must be maintained so that it can absorb stresses and shocks without fundamental change Le. the organisational structure, function and patterns of behaviour of the system must be retained. The resilience of the environment • protects the biosphere from risk of stochastic collapse or decline • protects the biosphere from risk of human-induced perturbations that may result in collapse or decline • protects the economy by allowing recovery from the negative effects of environmental and human-induced shocks Holling ( 1 973, 1 986), Common and Perrings (1 992), Odum ( 1 996), Levin et al . (1 998) • Chapter 1 1 : Test out the resilience (i.e. risk of collapse) of the Auckland ecological­ economic system to shocks resulting from possible future human-induced growth scenarios • Appendix B: Test out the resilience (Le. risk of collapse) of the global ecological-economic system to stochastic natural shocks by the incorporation of random environmental change into the modelling framework Principle 6 Use natural capital efficiently Principle 3 Maintain human activity within the carrying capacity of the environment Principle 7 Minimise material and energy throughput through the economy Principle 1 Maintain total capital stock within the l imits of with renewable natural capital Principle 4 Maintain environmental / � I l ife support services of the ecological­ economic system Figure 2.3 Interrelationships between Key Principles of Sustainability 53 Chapter Three Approaches to Urban Development and Sustainability This Chapter narrows the focus of the thesis to urban sustainability and the way different disciplines conceptualise urban growth and development. The fIrst part of this Chapter critically reviews the traditional approaches to the question of urban dynamics, and argues that these approaches reflect the 'Human Exemptionalism Paradigm' (HEP)56 which dominates the social sciences and humanities. Under this paradigm it is argued that human beings, by virtue of culture and human ingenuity will overcome all social and environmental problems confronting humankind. The remainder of this Chapter outlines new and emerging ecological approaches to urban sustainability and development which reflect the so-called 'New Ecological Paradigm' (NEp).57 These approaches acknowledge the ecological and the thermodynamic realities of (urban) growth as alluded to in Chapter 2, i.e. that cities depend on the biophysical environment for their survival and functioning, cities are ecosystems in the sense that they are complex networks of energy and material flows, and cities are subject to ecological limits like any ecological system. 3.1 The Human Exemptionalism Paradigm Emerging environmental problems in the late 1 960s and early 1 970s focused society's attention on the reality of biophysical constraints to human progress. Consequently, many sociologists began to examine the relevance of the environment to sociology (Buttel, 1 997; Dunlap, 2002). In seeking to understand whether environmental sociology was sufficiently distinctive to be considered a new fIeld, two sociologists in the 1 970s, William Catton and Riley Dunlap, noticed that despite the apparent diversity in the main competing theoretical sociological perspectives (e.g. functionalism, conflict theory, symbolic interactionism and so on) a fundamental anthropocentrism underpinned them all (Catton and Dunlap, 1 978; Buttel et al., 2002).58 Catton and Dunlap ( 1 978) argued that mainstream sociology had largely ignored the biophysical 56 Catton and Dunlap ( 1 978) originally termed this perspective the 'human exceptional ism paradigm' as it emphasises the exceptional characteristics of the human species by virtue of culture, language and technology. In order, however, to acknowledge that they were not questioning these characteristics, but rather the assumption that these characteristics exempted humans from ecological constraints, they later changed the term to 'human exemptionalism paradigm' (Dunlap, 1 997, 2002). 57 Catton and Dunlap ( 1 978) originally labeled this paradigm the 'new environmental paradigm' , but later revised this to the 'new ecological paradigm' as it seeks to emphasise the ecological base of human society (Dun lap, 1 997, 2002). 58 This also applies to the sociological perspectives that have since followed or were at the time in their infancy i.e. rational choice and exchange theory, ethnomethodology, phenomethodology, feminism, post­ structuralism and post-modernism. 54 environment. Drawing on Thomas Kuhn's ( 1 962) theory of scientific revolutions, as well as sociologists like Ritzer ( 1 975), they argued that certain sociological traditions and assumptions constituted a paradigm which they called the Human Exemptionalism Paradigm (Catton and Dunlap, 1 978 ; Dunlap, 1 997, 2002; Bell, 2004).59 3.1.1 Assumptions of the Human Exemptionalism Paradigm Catton and Dunlap ( 1 978) and Dunlap (2002) outline four primary assumptions that they believe not only blinkered contemporary sociologists to the social implications of ecological issues, but predisposed them to the optimism of the prevailing worldview, i .e . the assumption that there were no resource scarcities or other biophysical constraints on society's infinite growth and progress. These HEP assumptions are ( 1 ) humans, by virtue of their cultural heritage, are unique among all animal species; (2) social and cultural factors (including technology) can vary almost infinitely, changing far more rapidly than biological traits, and are therefore the primary determinants of human affairs. Many human differences are socially conditioned rather than genetic and can therefore be changed or eliminated if undesirable; (3) the crucial context for human affairs is the social and cultural environment, rendering the ecological environment largely inconsequential; and (4) cultural accumulation means that technological and social progress can continue without limit, making all social problems u ltimately solvable. Catton and Dunlap ( 1 978) argue that this optimistic worldview has been undoubtedly fostered by Western thought, i .e . that human ingenuity, in particular the creation of technology, creates unlimited opportunities to continue human progress. Accordingly, sociologists have not perceived the possibility of a future of genuine scarcity, but have instead ignored concepts such as carrying capacity, scientific laws such as conservation of mass and energy, and the principle of entropy (Catton and Dunlap, 1 978). In the remainder of this Section five major schools of thought on urban environments are discussed. Because of their inherent anthropocentrism all five may be characterised as belonging to the HEP worldview. Urban sociology examines social order in the urban environment, urban ecology studies how people arrange themselves within an urban environment, urban geography looks at the physical location of urban areas, urban psychology examines people's experiences of the urban environment, and urban political economy 59 Ritzer ( 1975, p.7) described a paradigm as "a fundamental image of the subject matter" and "the broadest unit of consensus" within a discipline. Warner ( 1 997, p . l 93) defines a paradigm as "a way of seeing the world, a representation, picture or narrative of the fundamental properties of reality" . Dunlap (2002) believes that the REP represented the largely consensual opinion of the sociologists of the time, i.e. that a consideration of society's ecological base is irrelevant to an understanding of modem industrial societies, and for this reason the REP constitutes a paradigm. 55 examines the urban dynamics resulting from political and economic decisions and trends.60 All examine the urban environment from diverse perspectives, but share a common underlying anthropocentrism. 3.1.2 Classical Sociological Theory Late nineteenth century sociologists were particularly int�rested in the urban upheaval resulting from the capitalistic processes of the Industrial Revolution. They were pessimistic about urban life, envisaging the city as a dangerous, dirty and undesirable place where the traditional values of social life disintegrated. European sociologists such as Karl Marx ( 1 8 1 8 - 1 883), Friedrich Engels ( 1 820 - 1 895), Ferdinand Tonnies ( 1 855 - 1 936), Emile Durkheim ( 1 858 - 1 9 1 7), Georg S immel ( 1 858 - 1 9 1 8) and Max Weber ( 1 864 - 1 920) typified this thinking. Karl Marx and Friedrich Engels collaborated closely on their work examining the conflict between the social classes during the height of the Industrial Revolution (Frey and Zimmer, 200 1 ; Kleniewski, 2002). They argued that the social institutions of family, religion and the political system were founded on the economic structure of society. Marx and Engels viewed the rise of urban settlements as signalling a transition from barbarism to civilisation, arguing that urban problems such as poverty and unemployment are the fault, not of the individual, but of the capitalist structure (Kleniewski, 2002; Thorns, 2002; Bounds, 2004). They contended that the social evolution of humans would not be complete until revolution transformed into socialism. In 1 887, Ferdinand Tonnies published GemeinschaJt and GesellschaJt, in which he described two contrasting types of social life (Tonnies, 1 9631 1 887; Flanagan, 1 993 ; Saunders, 200 1 ; Paddison, 200 I ; Kleniewski, 2002; Thorns, 2002; Bounds, 2004). Gemeinschaft, or 'community' , is characteristic of the country village where people work together for the common good, united by ties of family and neighbourhood. Gesellschaft, or 'association', is characteristic of urban settlements, where social life is a 'mechanical aggregate' distinguished by disunity, individualism and selfishness. The meaning of existence shifts from the group to the individual with family and neighbourhood ties of little significance. Tonnies believed that, 60 These schools of thought encompass Western thinking on the city as recorded over the last 160 years. By no means is this coverage presented here complete, nor are the summaries of the contributions made by key authors; rather the focus is on capturing the central themes of each school. Other viewpoints on the city, as based on the REP, include urban planning (e.g. Ebenezer Roward ( 1 965), Jane Jacobs ( 1 96 1 , 1 970, 1 984), Rerbert Gans ( 1 962, 1 982)), urban design and form (e.g. Le Corbusier ( 1 929[ 1 924], 1 967[1 935]), Frank Lloyd Wright ( 1958), Paolo Soleri ( 1969)) and the history of the city (e.g. Lewis Mumford ( 1 96 1 )). 56 when viewed over time, urbanisation in European history revealed a gradual replacement of Gemeinschaft by Gesellschaft as the dominant way of life (Macionis and Parrillo, 1 998). Emile Durkheim ( 1 9641 1 893), in The Division of Labor in Society, considered urban social structure, developing a model of two contrasting types of social order, namely 'mechanical solidarity' and 'organic solidarity' (Flanagan, 1 993 ; K leniewski, 2002; Thorns, 2002; Bounds, 2004). Mechanical solidarity describes the social unity that is brought about by a commonality of beliefs, customs, rituals and symbols, and is characteristic of many rural and primitive self­ sufficient societies. By contrast, organic solidarity refers to the social order of modem industrial societies, which are constructed on individual differences and an advanced division of labour - distinguished by specialisation of occupation. While Tonnies viewed the urban settlement as undermining of the very fabric of society, Durkheim viewed such settlements as giving rise to greater freedom of choice for the individual, creating a liberating form of social cohesion based on mutual interdependence (Macionis and Parrillo, 1 998). Georg S immel ( 1 964/ 1905) focused on the urban experience, in particular the social psychology of urban dwellers. Simmel viewed the urban experience as a constant bombardment of nervous stimuli with the individual continually discriminating between stimuli to avoid being overwhelmed. As a result, he believed that urban dwellers became more rational, calculating, emotionally reserved and intellectual (Flanagan, 1 993 ; Kleniewski, 2002; Popenoe and Michelson, 2002; Bounds, 2004). Simmel, like Durkheim, acknowledged the existence of an advanced economic division of labour, with social order based on the interplay of specialists. He argued that money played a pivotal role as a universal means of exchange, driving the shift from Gemeinschaft to Gesellschaft (Macionis and Parrillo, 1 998). He further believed that the urban environment held the potential for great personal fulfilment, but feared that feelings of alienation and loss of individuality may ultimately override this potential (Flanagan, 1 993 ; Wirth-Nesher, 200 1 ). Max Weber ( 1966) undertook a survey of various urban areas in Europe, the Middle East, India and China (Macionis and Parrillo, 1 998; Orum and Chen, 2003). Weber argued that any theory that studied the sociology of urban settlements in only one part of the world, at any one point in time, was inadequate. In 1 92 1 Weber published Die Stadt ( 'The City') in which he defmed the requirement of a full urban community: a predominance of trade or commercial relations, such as a market; a court and at least partially autonomous law of its own; at least partial political autonomy; a protective fortification or military self-sufficiency; and a means of association or social participation for individuals (Macionis and Parrillo, 1 998; Bounds, 2004). Weber viewed the urban environment as a potentially positive and liberating force in human life, but felt that 57 the Industrial Revolution had brought about a loss of military and political autonomy, such that people no longer felt a sense of allegiance to the urban community (Flanagan, 1 993). As a result, Weber did not hold out much hope for future urban environments. From Classical Theory to Modern Research The major contribution of the early urban theorists was their identification of the urban environment as an object worthy of study (Macionis and Parrillo, 1 998). All viewed urban settlements as increasing the potential for self-actualisation, and they emphasised rationality, individuality and an advanced division of labour. T6nnies, Simmel and Durkheim emphasised social order, social cohesion, community ties and social differentiation, while Marx, Engels and Weber stressed social, economic and political power relations as prime determinants of the urban experience (Kleniewski, 2002). There are several key criticisms of the early sociologists' thinking on the city. Firstly, they tended to focus on the sociological extremes of urban settlements, proposing dichotomous perspectives with the perceived future endpoints dominating their discourse (Thorns, 2002). Secondly, they all had an implicit bias in favour of rural l ife, with a predominant view of the urban environment as a place where community was lost and feelings of alienation developed (Thorns, 2002). The idea that urban relationships were impersonal, lonely, indifferent and anonymous has s ince been refuted by numerous studies which have revealed that city life does not preclude the existence of strong family, neighbourhood and friendship ties (Whyte, 1 943; Bell and Boat, 1 957; Greer, 1 962; Bruce, 1970; Suttles, 1 972; Fischer, 1975 ; Street et al., 1 978). Thirdly, the classical theorists' hypothesis of urban malaise has been questioned by subsequent research (Verbrugge and Taylor, 1 980) which suggests that the increased social accessibility of higher density city life may have positive effects on the psyche. Finally, a hallmark theme of the classical theory was its anthropocentrism. Catton and Dunlap ( 1 978), for example, note that Durkheim's insistence that social phenomena must be explained only by other social phenomena, and Weber's insistence that the methods and concepts of the social sciences must be kept separate from those of the biological sciences, has led to a deliberate omission of biophysical considerations and their relations to human society. 3. 1.3 Urban Ecology While the early sociologists focused mainly on the urban experience, urban ecologists began to question why cities took on a particular form. The principle concern of urban ecology is how people choose to spread out and arrange themselves within the urban environment (Paddison, 5 8 200 1) . Key early contributors, who were influenced by Tonnies, Simmel and Durkheim (Kleniewski, 2002), include inter alia Robert Park ( 1 864 - 1 944), Ernest Burgess ( 1 886 - 1 966) and Louis Wirth ( 1 897 - 1 952). During the 1 920s, Robert Park established the first urban studies centre in the sociology department at the University of Chicago. As the founding father of 'human ecology,6\ Park metaphorically applied biological processes and concepts to the social world (Kleniewski, 2002; Thorns, 2002; Orum and Chen, 2003; Bounds, 2004). His theory of urban ecology encapsulates forces similar to those of Darwinian evolution, such as competition (between classes and ethnic groups), dominance (by one group at the expense of others), succession (a dominant group taking over from a defeated group), and invasion (by a defeated group, initiating the process again), which were observed in the struggle for scarce urban resources, especially land (Saunders, 200 1 ; Buttel and Humphrey, 2002). This led ultimately to the division of urban space into specific 'ecological niches' in which the occupants shared similar social characteristics through enduring the same 'ecological pressures' (Flanagan, 1 993 ; Kleniewski, 2002). Park's greatest contribution was arguably his departure from purely theoretical study, and his insistence that it was crucial to 'get out there' and observe how urban environments actually worked (Kleniewski, 2002). Ernest Burgess of the Chicago School suggested that urban settlements develop by growing outwards in a series of concentric rings over time (Flanagan, 1 993 ; Kleniewski, 2002; Thorns, 2002; Orum and Chen, 2003 ; Bounds, 2004). Burgess ' s concentric zone model consisted of five main zones: central business district (CBD); c ircling the CBD, an area in transition containing business and l ight manufacture; residential area for industry workers who had moved from zones 1 and 2; residential area of higher class single fami ly dwellings or apartments; and a commuter zone of suburban areas or satellite urban areas (Figure 3 . 1 (a)) . A lthough his model was inspired solely by his study of Chicago, he believed it to be representative of any urban settlement. 6 1 Human ecology seeks to isolate the forces at work within an urban community which facilitate ordered groupings of people and institutions, and describe the typical clusters or constellations of persons and institutions brought about by the cooperation of these forces (Park, 1 967[19 16]). Figure 3.1 59 1 . Central business district 2. Wholesale light manufacturing 3. low-ciass residential 4. Medium-class residential 5. High-class residential (a) (b) 1 . Central business district 2. Wholesale light manufacturing 3. low-class residential 4. Medium-class residential 5. High-class residential 6. Heavy manufacturing 7. Outlying business district 8. Residential suburb 9. Industrial suburb (c) Urban Ecology Models of Urban Development Ca) Burgess's Concentric Zones Model, Cb) Hoyt's Sector Model, and Cc) Harris and Ullrnan's Multiple Nuclei Model. Source: Kleniewski (2002, pp.3 l-33). In 1 938, Louis Wirth of the Chicago school combined theory with empirics in his essay Urbanism as a Way of Life, by synthesising the insights of previous urban sociologists, with an empirical focus on urban lifestyle. Wirth drew up a set of universal social characteristics of the urban environment, and analysed the effects of these on urban social l ife. He viewed the urban community as a large, dense, permanent settlement comprising socially and culturally heterogeneous individuals - a function of population size, density and heterogeneity (Macionis and Parrillo, 1 998). Wirth saw the urban environment as an acid that over time dissolved traditional values and undermined the formation of meaningful relationships. In . his view a humane urban environment could only be created through the intense efforts of urban planning. 60 While the early European sociologists produced much theory, the American urban ecologists conducted actual research, publishing many descriptive studies. Criticisms of the Chicago School have, however, called into question the applicability of Park's ecological theory and Burgess 's concentric zone model . Firstly, apart from older U.S. cities, further examples of Burgess 's model are not readily found (Bounds, 2004). Secondly, the appropriateness of using ecological concepts such as competition, succession and invasion to explain human behaviours has been hotly debated (Berry and Kasarda, 1 977). Thirdly, Burgess's assumption that settlements are primarily based on transportation systems has been rejected. And, finally, the critical factor of choice through human intellect makes a comparison between ecology and the urban environment unsatisfactory. Despite these limitations, numerous alternative urban ecology models and theories have developed, taking into account more complex urban forms. Other notable theories, which attempted to address the deficiencies of the early Chicago School theorists, include the sector theory (after Hoyt ( 1 939)), the multiple nuclei theory (after Harris and Ullman ( 1 945)), social area analysis (after Shevky and Williams ( 1949) and Shevky and Bell ( 1 955)), and factorial ecology (after Pederson ( 1 967) and Johnston ( 1 976)). The sector and multiple nuclei theories are depicted respectively in Figures 3 . 1 (b) and 3 . 1 (c). Hoyt ( 1 939) made a block-by-block study of residential patterns in 1 42 cities between 1 900 and 1 936 to form his sector theory (Flanagan, 1 993; Kleniewski, 2002; Thorns, 2002). His main findings were: upmarket areas formed different sized sectors; many sectors were pie-shaped rather than ring-shaped; poorer areas were frequently adjacent to, or surrounded, more upmarket areas; over time, sectors exhibited a tendency to move out of the city radially; and over time, popular areas could be found in two or three different places, being influenced by not only competition and population movement, but also other factors.62 In 1 945, Harris and Ullman extended Hoyt's residential study, arguing that urban growth is not predictable or inevitable, but diversifies into many distinct sectors of activity in the form of multiple nuclei (Kleniewski, 2002). Shevky and Williams ( 1 949) and Shevky and Bell ( 1 955) followed a new direction by analysing urban land use in terms of the social characteristics of the inhabitants. They believed that variations among the three easily measured characteristics of economic status, namely, family type, income, and ethnicity formed the basis of the urban social structure, with the population of a particular sector sharing a high degree of similarity in these characteristics (Flanagan, 1 993 ; Kleniewski, 2002). They termed these homogeneous sectors ' social areas' . Social area analysis 62 Hoyt observed, for example, that wealthier residents opt for neighbourhoods on higher ground, indicating that they see themselves as above poorer residents. 6 1 has, however, been strongly criticised as offering only limited descriptive insight into urb� land use, but has no inherent predictive or explanatory value (Hawley and Duncan, 1 957). Factorial ecology employs computer techniques to analyse the social traits of an urban population. It differs from social area analysis in that it analyses clustered social traits that might influence urban form, rather than a few selected social traits (Flanagan, 1 993; Thorns, 2002). Several characteristics may be grouped together into a single factor - socioeconomic status and family status, for example, have been used by researchers such as Pedersen ( 1 967) and Johnston ( 1 976) to explain specific residential land use patterns. Berry and Rees ( 1 969) found that factorial ecology, rather than discrediting the concentric zone, sector and multiple nuclei theories, has shown that these approaches all exhibit a degree of partial correctness, but none captured the full complexity of urban land use due to its isolated focus on one or only a few social characteristics . Factorial ecology has thus shown that residential land use patterns are the combined influence of multiple factors. Factorial ecology, although remaining an accepted approach in explaining urban land use, has been criticised on several fronts. Firstly, the method relies on census tracts, which are not always socially defined districts and are known to contain coding and data errors. Secondly, the method has only been applied to a selected few cities around the world -further case studies are required to test its validity. Finally, many urban geographers and sociologists believe that urban land use patterns are also the result of capitalist market forces which are largely downplayed in factorial ecology. 3.1.4 Urban Geography The focus of urban geography is on the significance of urban location, and in particular availability of natural resources and landscapes (Macionis and Parrillo, 1 998; Pacione, 200 1 ). An urban community has physical needs such as a hospitable environment, and access to adequate supplies of food, water and building materials. Furthermore, social reasons for a particular location may exist, namely: natural crossroads or 'break-of-bulk' po ints63 , access to valuable raw materials, amenity, administrative or political functions of government, strategic military capability, and religious or education reasons (Macionis and Parrillo, 1 998). A combination of suitable environmental and social conditions will faci litate urban development, with an urban region' s importance typically determined by the degree to which these physical requirements are met. 63 These are locations where bulk goods are transferred from one type of transportation to another. 62 The form or physical shape taken by an emerging urban settlement reflects not only its physical environment, but also its social and economic functions. Figures 3 .2(a) and 3 .2(b) depict two patterns of a city's physical development, namely the radiocentric and gridiron city (Macionis and Parrillo, 1 998; Pacione, 200 1 ). Radiocentric cities, which radiate out from a common centre, originated in preindustrial times out of a need for protection. Other reasons for radiocentric formation are equal proximity to the city centre as the city grows, and the shortest possible access routes to the centre - these take the form of radiating lines, like spokes in a wheel. The radiocentric design is the exception in North America, where many downtown areas of well-known cities are gridiron-shaped. This form, composed of straight streets crossing at right angles to form city blocks, is typical of post-Industrial Revolution design. It facilitates economic activity, such as movement of people and products, and the subdivision of land for real estate. Figure 3.2 (a) , DDDDD I DDDDD DDCJDDD DDDDDD DDDDDD DDG;i1,DDD DD "\� DDD DDDDDD DDDDDD DDCJDDD DDDDr?CJ DDDDbd DDDDDD�= DDDDDD DDDDDDD DD�DDDD DD�DDDD DDDDDDDD DDDDDDDD I DDDDDDDD D DODO ' D DDDDD Delaware River ------------------------------ (b) Urban Geography Models of Urban Development (a) The Radiocentric City: Baghdad circa 1 46-763 C.E., and (b) The Gridiron City: Philadelphia, 1 682. Source: Macionis and Parrillo ( 1998, pp. 1 68, 172). 63 Urban geographers provide three pathways for the physical expansion of cities: horizontally, vertically, and interstitially (toward a greater density) (Blumenfeld, 1 949). The dominant pattern depends on available technology. Before motorised transport such as elevators, cars, buses or trains, and construction materials such as steel frames, growth was mainly interstitial with intensification of buildings taking up all available land. With the advent of buses and trains, the city spread out along transport lines and clustered in suburbs around transport nodes such as railway stations. Once the automobile was in use, the spaces between these nodes filled up, although still less densely settled than the city centre. Similarly, the advent of steel framing and the Otis elevator permitted the city to grow skywards at increasing densities. Urban geographers have also paid attention to how urban settlements form in relation to one another. Settlements exist because selected activities can be performed more efficiently if they are clustered together rather than dispersed. Most towns develop a service function for their surrounding hinterland - such settlements have been termed 'central places' (Pacione, 200 1 ). Urban geographers argue that the location of central places reflects the general population distribution i .e. an evenly spread population will result in an even spread of central places, while an unevenly spread population will result in central places being typically located where most accessible. Central places that serve large populations typically offer more specialised services and often grow progressively larger. A growth pattern emerges with various grades of central places distinguished by population size and zones of influence. This has become known as Central Place Theory. One of the most commonly cited applications of Central Place Theory is the 1 930s work undertaken by Christaller ( 1 9661 1933) in southern Germany and later reformulated by Losch ( 1 9541 1 943). This fundamentally economic approach to analysing central places predicts how, through competition for space, an optimal pattern of settlement will emerge (Pacione, 200 1 ). According to Christaller ( 1 966/ 1933) each central place has a circular trade area - ideally this would be hexagonal for maximum efficiency. Settlements with the lowest levels of specialisation would be equally spaced and surrounded by hexagonal shaped hinterlands. For every six low specialisation settlements there would be a larger more specialised central place, say, township, which, in turn, would be equally distant from other townships, and so on (Figure 3 .3). Christaller 's theory has been criticised on various grounds including its limited applicability, economic determinism, and static formulation. 64 : I ' , I " - - - - ., A ,.. - - - -\ '" " '" " "'r.\ ' · ... \ .. � _ _ _ f/' 'c;y' '" \ T I \ T I , , ' , I ' \ I I :� - -I- - ': 0 ,� - -.- -': , , : I " , I " .. 0.. ,� - - - - .. , .. 0.. '" ..... , \, ..... '" - - - - �/ .... 'er ' \� - - - - ..... '-, I ,t' Higher-order market area Middle-order market area Lower-order market area o Higher-order centre o Middle-order centre <:> Lower-order centre Lowest-order centre Figure 3.3 H ierarchical and Spatial Arrangement of Central Places. Source: Pacione (200 1 , p. 1 l 8) Diffusion theories provide an alternative to Central Place Theory by means of analysing the processes by which settlements spread from the point of initial colonisation (Pacione, 200 1 ). The idea behind diffusion models proposed by inter alia Bylund ( 1 960), Morrill ( 1 963) and Hudson ( 1 969) is that human behaviour, as applied to the formation and growth of settlements, occurs gradually over time and may be described as chaotic/random within certain limiting conditions. The elements of time and indeterminacy of behaviour are the central tenets of diffusion theory. In his application of diffusion theory Morrill ( 1 963) applied a historical-predictive approach based on Monte Carlo analysis64 to describe the location of settlements. Hudson ( 1 969) attempted to integrate diffusion theory with central place theory by drawing on the ecological principles of colonisation, dispersal and competition to illustrate the spatial pattern of settlements. Although diffusion theory possesses analytical explanatory power, the great variety of settlement forms and distributions has meant that more meaningful interpretations were required, particularly with reference to the influence of political and social forces operating at different spatial scales. 3.1.5 U rban Psychology Human reactions to the physical and social urban environment represent an urban social psychology. Ordering urban environments through mechanisms such as mental mapping, developing social codes of behaviour and establishing urban networks, aids people in the 64 A stochastic modelling technique in which behavioural choice is governed by a set of probabilities. 65 formation of a sense of urban Gemeinschaft. This provides people with emotional security, making urban society more meaningful and enjoyable. In the 1 960s and 1970s, Kevin Lynch became concerned with how people make sense of the urban environment. In The Image a/the City, Lynch ( 1960) presents the concept of ' legibility ' , or mental mapping, of the urban environment. He discovered that people based their own mental image of the urban community on the five elements of paths, edges, districts, nodes and landmarks (Wirth-Nesher, 200 1 ; Popenoe and Michelson, 2002). Lynch found that urban environments differed markedly in ' imageability' , with strong imageability heightening the potential for an intense human experience of urban life. In a study of New Y orkers and Parisians, psychologist Stanley Milgram found that people's mental maps are based on personal experience, their interests, and their understanding of socially acknowledged important areas of the urban environment (Milgram, 1972; Duncan, 1 977; Thill and Sui, 1 993; Kulhavy and Stock, 1 996). Milgram also found that this mental image keeps on changing, and that no one is able to recreate the complexity of the entire urban area. Urban environments are thus a dynamic, creative, continual mixture of experiences. In Tonnies' Gesellschaft society, individuals coped with vast numbers of people and the accompanying anonymity by observing an intricate set of social rules. Social and psychological security in urban society rests on interpersonal relationships formed through urban networks. Studies by Suttles ( 1968), Howel l ( 1 973), Fischer et al. ( 1 977) and Gans ( 1982) revealed that many Gemeinschaft relationships exist in an urban neighbourhood, with strong ties forged with family, neighbours and friends. Lofland ( 1 973) proposed that, since not everyone has a traditional location such as a neighbourhood for developing interpersonal relationships, many people transform areas into private or semi-private space (Flanagan, 1 993; Popenoe and Michelson, 2002). Irwin ( 1 977) termed such places 'scenes' - typical scenes include bars, c lubs, or urban areas taken over by a particular group. 3.1.6 Urban Political Economy Toward the end of the 1 960s, social scientists found that the theories of urban sociology, ecology, geography and psychology were inadequate in explaining contemporary developments in urban life. Walton (198 1 ) asserts that the inadequacies/deficiencies of the early sociological theorists and of the urban ecology model in not only explaining prevailing social conditions, but in anticipating conflict and change, provided the impetus for the emergence in the early 1 970s of a 'new urban sociology' or 'urban political economy' . This approach directs attention to how 66 social conflict, inequality and change affect urban settlements globally (Gottdiener and Feagin, 1 988; Flanagan, 1 993; Gottdiener, 1 994; Kleniewski, 2002). Urban political economy argues that urban dynamics are heavily influenced by investment decisions and economic trends, in particular: urban settlements emerge within the larger political structures of county, state, nation and the rest of the world; local economies do not operate in isolation, but are linked together to form state, national and international economic networks; and political and economic institutions such as governments, international corporations and banks, and their investment decisions, are critical in shaping urban life. Under the influence of Marx, Engels and Weber (Kleniewski, 2002), authors such as Henri Lefebvre, David Harvey, Manuel Caste lis, Alien Scott, and Logan and Molotch drew heavily on political economic and Marxist theory in their attempt to understand urban form, in particular the social structures and processes of change that privilege some to the detriment of others. Henri Lefebvre, a French philosopher, extended the ideas of Marx and Engels by applying socioeconomic concepts to an understanding of the unevenness of urban development (Macionis and Parrillo, 1 998). Lefebvre suggested that urban development, particularly as manifested by differences in economic growth, was a product of the capitalist economic system. He identified three core influences: two circuits of capital (i .e. the primary circuit being investment capital for industry, and the second circuit real estate investment); space as a part of social organisation (i .e. the construction of space to meet needs is closely linked to behaviour); and the role of government in managing space (i.e. government influences land use patterns through urban development decisions such as roading, zoning, funding, taxes and so on). Lefebvre classified space as either abstract space - the environment envisaged by business, investors and government in terms of size, location and profit; or social space - the environment envisaged by the individuals who live, work and play there (Orum and Chen, 2003). These two perspectives result in a conflict along the lines of Marx's class conflict. Lefebvre' s work is often considered a seminal contribution to the study of urban development. David Harvey, a prominent English geographer in the 1 970s, used Lefebvre's ideas on the second circuit of capital to illustrate how capitalist real estate investment directly shaped social inequality in Baltimore (Flanagan, 1993 ; Kleniewski, 2002; Orum and Chen, 2003 ; Bounds, 2004). Selective capitalist investment in the housing market, combined with government intervention serving capitalist interests, resulted in highly uneven urban development. In this way, Harvey il lustrated the role of fmance capital, rather than industrial capital, in determining a city's use of space (Figure 3 .4). Productivity of labour Innovation Figure 3.4 Fixed capital , producer durables 2 built environment Cred� and money creation Debt repayment Production of values and surplus value Intermediate inputs Transfers ... .. Consumption fund , oonsumer durables 2 built environment Consumption of commodities and reproduction of labour power Social expenditures (education, health, welfare, ideology, police, military, etc.) Harvey's Model of the Circulation of Capital. Source: Pacione (200 1 , p. 143) 67 Manuel Caste lis highlighted the conflict between local government and the working class arising from local administration of various social welfare programmes. Castells viewed welfare capitalism as an effort by government to extend capitalism, resulting in new urban struggles and patterns of conflict affecting urban life, He also introduced the concept of 'mode of development' .65 Caste lis suggested developing new forms and sources of information were a key element in today's informational mode of development (Orum and Chen, 2003 ; Bounds, 2004). Thus, corporate location no longer relied on proximity to sources of raw materials and labour, but was increasingly found in suburban or peripheral districts. In the 1 980s the English geographer Allen Scott studied the relationship between urban growth and economic globalisation. He analysed the impact of changes in the production process on urban space and suggested that urban growth patterns were determined by powerful transnational corporations (TNCs) rather than territorial competition as per the urban ecology model (Macionis and Parrillo, 1 998). Two major contributions by Scott were the ideas of ( 1) horizontal integration - companies in the past were small entities that centralised all their functions in one location, but gradually, through absorption and consolidation of competitors, expanded their operations. Headquarters were maintained in a central location, but production plants and distribution centres were established in other more advantageous locations; (2) vertical disintegration - from the 1 970s, companies began divesting themselves of their production companies, instead awarding contracts to suppliers through a process of competitive bidding, Many of the new production companies were located where labour and energy costs 65 This was an adaptation of Marx's ' mode of production' concept. Marx saw the discovery and application of new sources of energy as a key element in the industrial mode of production, 68 were low, allowing large companies to conduct competitive business on a global scale, becoming multi- or transnational. Logan and Molotch ( 1 987) use political economic theory to identify central decision makers on urban growth in North America. They view urban development in terms of Lefebvre's categories of 'abstract space' and 'social space', with local conflicts arising between pro-growth and anti-growth factions (Boyle and Rogerson, 200 1 ; Orum and Chen, 2003). They presented their theory of the 'growth machine' , a coalition of entrepreneurs and urban politicians who favour increased economic development at the expense of smaller private sector entities, neighbourhood residents and other vulnerable stakeholders (Kleniewski, 2002; Orum and Chen, 2003). The growth machine's focus on the high profits that often accompany urban growth only extends quality of life to abstract space, failing to take into account the 'social space' ideas of local people. The Global Economy: Cities as Consumers A central tenet of the political economy approach is that urban change is linked to the development of a global economy. Regions of the world are increasingly being drawn into a single economic and political system (Wallerstein, 1 979) which operates as a hierarchy, with countries at various stages of development constituting the 'core' (most economically developed), 'semi-periphery' (those with close ties to the core) or 'periphery' (the poorer less developed countries). The decision-making headquarters of TNCs, and their well-paid professionals, are typically situated in large cities in core countries. On the other hand, the manufacturing and distribution sectors, and their low-waged labourers, are located in the semi­ periphery or periphery countries, where the impact of globalisation can be more negative than positive (Macionis and Parrillo, 1 998; Orum and Chen, 2003). Under globalisation, a 'post-industrial' city results; this is essentially a product of economic forces. Dramatic change has occurred as urban areas have moved away from manufacturing to become service centres, with a new focus on advertising, management, finance, and other business services as required to oversee investment activity in a global economy (Sassen, 1 99 1 ; Kleniewski, 2002). Consequently, two labour markets have emerged: well-paid white collar professionals and low-paid service workers. Berry ( 1 985, p.69) is pessimistic about this dichotomy, believing that the inequalities between the two will increase over time, resulting in cities with "islands of renewal in seas of decay". 69 A key consequence of the shift away from manufacturing to service provision is that globalised cities have become sites of consumption (Saunders, 1 98 1 ; Orum and Chen, 2003 ; Miles and Miles, 2004). The role of consumption rather than production as a key influence on urban shape and form began to receive greater prominence in the later 1 980s. Consumption in the urban context is both individual and collective. Individual consumption has significantly influenced a dispersed low density spatial structure, supporting development of a private transport­ orientated, owner-occupied city. Collective consumption has also resulted in urban change, with a shift away from government provision of services such as healthcare, education, transport, and urban open space in favour of market provision of these services (Thorns, 2002). Key Principles of the Urban Political Economy Perspective Although urban political economists have emphasised different aspects of economic activity, general agreement exists on four foundational principles for analysing urban l ife : ( 1 ) urban development is not shaped by natural processes, but by human decisions made by those that control wealth and resources; (2) urban social arrangements reflect conflicts between rich and poor, powerful and powerless, and business and local communities over the distribution of resources; (3) government plays a key role in shaping urban life by allocating resources, mediating conflicts, and regulating economic activity; and (4) economic restructuring as a result of a global economy has significantly altered urban growth patterns (Macionis and Parrillo, 1 998). 3.1.7 Brief Critique of the HEP-based Urban Schools of Thought All the urban schools of thought so far reviewed assume no biophysical constraints to development of urban settlements. The inherently western idea of continued progress has helped to foster and reinforce the REP assumption that all urban social and environmental problems can ultimately be solved through unlimited human ingenuity (Boyden et al., 1 98 1 ). The early European sociologists, for example, focused on the city's potential for higher levels of rationality, specialisation and individuality by virtue of culture. Durkheim stressed that the causes of social phenomena should only be explained by social phenomena. Weber focused exclusively on historical and institutional considerations to the exclusion of the biophysical environment. Tonnies' Gemeinschaft-Gesellschaft typology similarly omits any consideration of ecological or biophysical constraints, despite the obvious shift in spatial configuration from one form of social organisation to another. 70 In urban psychology, the belief that the urban experience is very much subjective, that urban dwellers may reach higher levels of self-actualisation and sophistication, and that city life can be made more pleasant simply by cultural adaptation, reinforce the REP assumption that cultural traits are more important than biological traits. Urban ecology, while appearing to acknowledge the biophysical environment, merely draws an analogy between ecological and urban environments, but fails to integrate these in any way. Similarly, the work on sector, multiple nuclei, and social area analysis, factorial ecology and urban geography utilised spatial units to explain urban form, but saw the spatial structure as primarily a manifestation of social processes (Michelson and Van Vliet, 2002). Urban political economy studies the influence of human political and economic systems in shaping urban environments, but makes little acknowledgement of the city's dependence on ecological resources, or the assimilative capacity of the environment to detoxifY wastes, pollutants and emissions. 3.2 The New Ecological Paradigm Evidence of serious environmental problems escalated throughout the 1 970's and has since continued relentlessly in the form of major issues such as global warming, ozone depletion, acid rain, energy crises, and environmental disasters such as Bhopal 1 984, Chernobyl 1 986 and the Exxon Valdez 1 989 (Dun lap, 1 997, 2002). Environmental threats in the 1 970s were problematic to adherents of the REP, as they highlighted the interdependencies between the welfare of human societies and the biophysical environment. Sociologists such as Schnaiberg ( 1975), Anderson ( 1 976) and Catton ( 1 976a, 1 976b) began analysing the causes of environmental degradation, and the impacts of pollution and resource scarcity on society. In doing so, they implicitly rejected the REP assumption that human beings are exempt from ecological constraints (Dunlap, 2002). Catton and Dunlap ( 1 978) argued this rejection supported the emergence of an alternative worldview which they termed the New Ecological Paradigm (NEP). Whereas the REP represents the environment as something humans control for their own ends, the NEP presents the environment as critical for human life, as potentially fragile and limited in resources, and as imposing constraints on the achievement of unlimited human objectives (Catton and Dunlap, 1 978; Dunlap, 1 997, 2002; Dunlap et al., 2002; Buttel et al., 2002; Bell, 2004).66 66 Note that the REPINEP dichotomy does not represent the poles of an anthropocentric-ecocentric continuum. Like the REP, the NEP is inherently anthropocentric - it must be so to be considered a sociological paradigm, but it differs from the REP by acknowledging humankind's critical dependence on the environment. 3.2.1 Assumptions of the New Ecological Paradigm 7 1 Catton and Dunlap ( 1 978) and Dunlap (2002) outline four fundamental assumptions of the NEP, extracted from the writings of early environmental sociologists such as Burch ( 1 97 1 , 1 976), Schnaiberg ( 1 972, 1 975), Anderson ( 1 976), Catton ( 1 976a, 1 976b) and Morrison ( 1 976). Firstly, despite their exceptional characteristics of culture, language and technology, humans are only one species among many that are interdependently involved in the global ecosystem. Secondly, human affairs are determined not only by social and cultural elements, but also by the complex cause, effect and feedback linkages in the web of nature, which produce unintended outcomes from purposive human actions. Thirdly, a finite globe means that physical and biological constraints restrict economic growth, social progress, and other human affairs. And finally, although human ingenuity may appear to extend carrying capacity, ecological laws (e.g. the laws of thermodynamics) cannot be revoked. The essence of the NEP is society's critical dependence on the biophysical environment (Dunlap, 1997, 2002). The natural sciences, social sciences and humanities have typically studied human dependence on the environment in isolation of each other. According to Boyden et al. ( 1 98 1 ), this excessive compartmentalisation, fragmentation and specialisation in human thinking has lies at the root of many social and environmental crises facing modern society. C layton and Radcliffe ( 1 997) argue that any strategy which attempts to instigate relatively unconnected changes to society, economy and environment is less likely to succeed compared with a systematic attempt to build integrated socio-economic and ecological systems. Understanding human activity and its implications requires an approach that focuses on the interrelationships between humans and their surrounding environment. A systems approach offers one such pathway for achieving this. 3.2.2 Cities as Ecosystems In an article entitled "Cities are ecosystems ! : new trend to study urban areas" in the journal Ecological Economics, Breslav et al. (2000, p .337) announced that "educators and scientists are joining forces to build a more comprehensive, interdisciplinary understanding of cities as ecological systems". This view is supported by Roseland ( 1 992), Tjallingii ( 1 993) and Newman ( 1 999) who all believe the key to solving environmental problems is to view the city as an ecosystem67,68, using inputs such as energy and materials (minerals, biomass, fossil fuels, land 67 The term ecosystem first appeared in a 1 93 5 publication by British ecologist Arthur Tansley. It was however originally coined in 1 930 by Roy Clapham, a colleague of Tansley's. Notable definitions of the concept have been made by Tansley ( 1935), Lindeman ( 1 942), Evans ( 1 956), Odum ( 1 97 1a, 1 983), King ( 1 993) and Kay and Schneider ( 1994). Evans ( 1956), for example, defines an ecosystem as an organisational unit, comprising one or more living entities, through which energy and matter are 72 and so on), and producing outputs such as liveability (commodities, transportation, social networks and so on) and residuals (solid waste, water pollution, gaseous emissions and so on). The view of the city as an ecosystem is most aptly summarised by Tjallingii ( 1 993, p .7), "The city is [now] conceived as a dynamic and complex ecosystem. This is not a metaphor, but a concept of a real city. The social, economic and cultural systems cannot escape the rules of abiotic and biotic nature. Guidelines for action will have to be geared to these rules". Girardet ( 1992) proposes that an understanding of this urban biophysical functioning is crucial to sustainability. Newman ( 1 999, p.220) asserts that the view of a city as an ecosystem is "one of the strongest themes running through the l iterature on urban sustainability". Baccini ( 1 996), Nijkamp and Pepping ( 1 998) and Decker et al. (2002) contend that the focus on sustainability is because cities are the major consumers of natural resources and producers of wastes. Implicitly, these justifications necessitate a transformation of the city from a less to a more sustainable form. Chapter 2 outlines the key principles for monitoring the progress of this transformation toward sustainability. As an ecosystem, the city is a system, typified by resource inputs (e.g. land, water, fuels, foods, building materials and so on) and residual outputs (e.g. solid waste, pollution, emissions, toxins, waste heat and so on). F igure 3 .5 , for example, depicts Sydney' s key resource inputs and residual outputs. To faci litate analysis of the c ity ecosystem, these fluxes are typically measured in material (i.e. mass) and energy terms. Furthermore, by considering the city as a whole, it is possible to conceive of management structures and technologies aimed at mimicking efficient natural processes, increasing eco-efficiency, recycling wastes, and reducing material and energy throughput. Girardet ( 1 996, p.23) advocates a circular metabol ism for cities where "every output can also be used as an input into the production system".69 processed and transferred - a description that could arguably be applied to a city. S imilarly, King ( 1 993, p.24) defines an ecosystem as a system of "interacting biota and environment of some time-space domain". Odum (1 983) provides further insight into where the ecosystem concept rests as an organisational unit by noting that the inclusion of the physical environment differentiates an ecosystem from a community. 68 It is important not to confuse the treatment of the city as an ecosystem with the urban ecology school of thought. The former is typically concerned with a city's consumption of material and energy resources and production of waste outputs (Girardet, 1 996; Newman et al., 1 996; Newman, 1 999 ; Breslav et al., 2000; Decker et al., 2002). The latter simply uses the analogy of Darwinian processes of organisation within ecosystems to explain how people spread out and arrange themselves within the city (Flanagan, 1 993 ; Macionis and Parrillo, 1998; Thorns, 2002; Orum and Chen, 2003), but fails to acknowledge any relationship between the city and the biophysical environment to which it is bound. 69 Using the second law of thermodynamics Georgescu-Roegen ( 1 971 ) has argued that complete recycling is physically impossible. Although others disagree with Georgescu-Roegen' s ( 1 97 1 ) assertion on theoretical grounds, they all agree that in practical terms complete recycling is impossible. Refer to Chapter 2 for further details on this topic. Figure 3.5 14,017,131 tonnea LMJ Oil Coal Gas (18'1. 75'1. 7'1.) �--@- �'."'.284_ -", Tunber 3.839.325 cu. m Products Air Waste c:::::Q..sn,670.ooo tonnes � Water miD 2,130,000 tonnee SoUdWaste .CoUNil -Sl'1. Cornmen:ial/Indust -34'1. Demolition - 15'1. . ����� C17,703,780,oooMJ Waste Heat 73 Resource Inputs Consumed and Waste Outputs Discharged from Sydney, 1 990. Source: Newman ( 1996). In studying the city as an ecosystem, several approaches have been pursued including urban metabolism and extended urban metabolism, energy analysis and emergy analysis. A brief discussion of these approaches follows (for further details refer to Wolman ( 1 965), Newcombe ( 1 975a, 1975b), Newcombe et al. ( 1 978), and Boyden et al. ( 1 98 1 ) - on urban metabolism; Newman et al. ( 1 996), and Newman ( 1 999) - on extended urban metabolism; Hannon ( 1973a, 1 982), Bullard and Herendeen ( l 975a), and Brown and Herendeen ( 1 996) - on urban energy analysis; and Huang ( 1998) - on urban emergy analysis) . Urban Metabolism Furthering the 'city as an ecosystem' approach, the concept of urban metabolism views the city as an organism, utilising the metaphor of biological metabolism (Le. the chemical process within an organism involving intake of resources, their transformation into more or less complex forms, and the subsequent excretion of wastes) to describe human processes (e.g. production and consumption) undertaken within cities. Urban metabolism provides a holistic framework for analysing a city's input-output relationships with its surrounding biophysical environment. Although urban metabolism studies are a relatively recent phenomenon (Girardet, 1 992, 1 996; Tjallingii, 1 993 ; Newman et al., 1 996; Newman, 1 999; peE, 1 998), their antecedents may be traced back much further. 74 Marx and Engels were the fIrst to apply the term 'metabolism' to society to describe the material exchange between humans and nature in their critique of the capitalist mode of production (Fischer-Kowalski, 2002). Marx's use of the term was not however metaphorical, but referred specifIcally to plant nutrient cycles (Martinez-Alier, 1 987). Spencer ( 1 862) noted that societal progress is based on energy surplus, and that the amount of available energy explained differences in stages of advancement among societies. Nevertheless, it was not until Patrick Geddes ( 1 854 -1932), arguably the founding father of town p lanning, that the key ideas underpinning urban metabolism were laid down (Martinez-Alier, 1 987). Geddes' ( 1 885) thoughts on urban ism arose out of the birth and growth of vast urban areas in Great Britain - a transformation made largely possible through the burning of coal. The prevailing worldview of the time was a belief in unlimited industrial progress, but Geddes undertook an ecological critique of urbanisation, recognising that the availability of energy and materials imposed strict constraints on modem industrial activity. This included establishing an urban energetic and material budget in physical input-output terms - inspired by Quesnay's Tableau Economique. Geddes' table consisted of the sources of energy and materials transformed into products in three stages: extraction of fuels and raw materials; the manufacture, and transport and exchange. The table also included intermediary products used for manufacture or transport of the fInal products; calculation of energy losses between each of the three stages; and the resultant fInal product - which was often surprisingly small, in material terms, compared with its overall material inputs (Geddes, 1 885 ; Martinez-Alier, 1 987; Fischer­ Kowalski, 2002). Through this analysis, Geddes proved to be ahead of his time as the fIrst scientist to attempt an empirical description of urban metabolism on a macroeconomic scale (Fischer-Kowalski, 2002). Using the idea of society as a machine, he suggested the possibility of complete quantifIcation of the way in which all matter and energy is integrated (and/or disintegrated) by transformation and by dissipation (Martinez-Alier, 1 987).70 Unfortunately Geddes' ideas ran counter to the prevailing attitudes of his time and, thus, it was not until the mid 1 960s that the concept of urban metabolism was revisited. Abel Wolman ( 1 965) used urban metabolism in his study of a typical American city. He described the city in terms of the through-flow of energy and materials, and captured the salient features of the ecosystem approach, i.e. the city as a consumer of materials/energy and a producer of wastes. Wolman's work was not alone - seminal contributions were made by Newcombe ( 1 975a, 1 975b, 1 976, 1 977a, 1 977b), Boyden ( 1977), Newcombe et al. ( 1 978), and Boyden et al. ( 1 98 1 ) 70 Authors like Martinez-Alier ( 1987) and Fischer-Kowalski (2002) note that Geddes' urban metabolism, as based on flows of energy and materials, is far closer to the study of ecology than the misnamed 'urban ecology' . Martinez-Alier ( 1 987) even suggests that the new environmental sociology o f Catton and Dunlap ( 1 978) and Humphrey and Buttel ( 1 982) should adopt Geddes as a founding father. 75 in their studies of Hong Kong.7 1 The complete and comprehensive nature of that latter work deserves further praise as the first attempt "to study and describe a human settlement in a comprehensive and integrative way, taking into account physio-chemical, biotic, societal and cultural components . . . and considering the dynamic interrelationships between them . . . as they relate to human health and wellbeing and to the life-supporting properties of the biosphere" (Boyden et al., 198 1 , p.xv). Extended Urban Metabolism Alberti ( 1 996), Newman et al. ( 1 996), PCE ( 1 998), Newman ( 1 999), Newman and Kenworthy ( 1 999) and Newton (200 1 ) advocate an ' extended metabolism' model to study the city. Figure 3 . 6 depicts Newman and Kenworthy's conceptualisation of the extended metabolism model. This model specifies not only the physical and biological basis of the city, but also its human basis. The physical and biological processes convert resource inputs into products and finally waste outputs, in a manner analogous to biotic metabolic processes. Underlying these processes are physical laws of nature, such as the laws of thermodynamics, where anything entering the system must pass through and ultimately exit in some form. The amount of waste exiting the system is therefore dependent on the amount of inputs required. In this way, all inputs and outputs can be accounted for in a balance sheet format. In this extended urban metabolism model, the economic and social aspects of sustainability are integrated by the inclusion of liveability (i.e. the human need for social amenity, health and well-being). By acknowledging human dependence on the environment the extended metabolism model belongs clearly to the NEP. 71 Warren-Rhodes and Koenig (200 1 ) have since published an update of the Newcombe inter alia work on Hong Kong's urban metabolism. 76 Figure 3.6 umd Water Food Energy Building ma terial Other resources Transportation prioritil!S Economic priorities Cultural prioritil!S Health Employment Income Education Housing l£isu Tt ac ti vities ' Accessibility Urban design quality Community Solid waste Liquid waste Toxies Sewage Air pollutants Greenhouse gases Waste heat Noise Extended Metabolism Model of Human Settlements. Source: Newman and Kenworthy ( 1 999, p.8). Energy Analysis of Urban Settlements Energy analysis, a tool based upon thermodynamics, entails the determination of the energy required both directly and indirectly by a system (usually an economic system) in the process of producing a good or service (IFIAS, 1974; Hendtlass et al., 1 988). The motivation behind energy analysis is the quantification of the connection between human activities and the demand for energy (Brown and Herendeen, 1 996). Energy analysis attempts to make explicit not only the direct use of energy in production, but also indirect (i .e. as appropriated through production chains) energy requirements. The realisation that energy is a scarce and essential resource for all production processes has been a key impetus behind the focus on energy (Chapman, 1 977). Notable energy analysis studies at a national level have been carried out by inter alia Bullard and Hannon ( 1976) and Bullard et al. ( 1 978) for the US economy, Denton ( 1 975) for the Federal Republic of Germany and, in the New Zealand context, by inter alia Pe et ( 1 986), Hendtlass et al. ( 1988), Cocklin et al. ( 1989) and McDonald et al. ( 1 999). Although there are fewer examples of energy analysis at the regional or urban level (Brown, 1 98 1 ), studies include Odum and Brown ( 1 975) and Browder et al. ( 1976) for the region of South Florida, J ansson and Zucchetto ( 1 978) and Zucchetto and Jansson ( 1 979) for Gotland, Sweden, and, in the New Zealand context, McDonald et al. ( 1 999) for Auckland Region, Hamilton City and Whangarei City. 77 Emergy Analysis of Urban Settlements Huang ( 1 998), in a case study of Taipei, seeks to understand urban ecosystems, explaining the hierarchical spatial organisation of cities using methods of analysis based on Odum's ( 1 986) concepts of emergy72 (i.e. embodied solar energy) and transformity73 . He argues that since transformity increases in a system with the number of energy transformations, and that in urban settlements energy transformations typically increase when moving from geographically dispersed to more centralised areas (i.e. energy flows from the surrounding landscapes to converge on urban centres), it therefore fol lows that an urban energy hierarchy must exist. For Taipei he identifies a hierarchy of five zones (ranked from highest to lowest): a mixed use urban core, a high density urban residential district, a service and manufacturing district, an agricultural district, and a natural area, each drawing resources from the next zone outward.74 He noted that over time, zones of higher order develop successively as a result of energy convergence, with each zone evolving and being forced, in response to external changes, to adapt its internal structure. Urbanisation may therefore be seen as the process of change in the type of energy sources utilised. Moreover, this process of change will be reflected in the urban ecosystem's internal self-organisation.75 Limitations of the Urban Ecosystem Approach Although the study of the city as an ecosystem is a useful tool for understanding of the functioning and sustainability of urban systems, the city ecosystem differs from natural or biological ecosystems in several important ways: • Ecological versus social variables. Biological metabolic processes differ from those of the urban setting in that natural ecosystems do not exhibit the social variables which divide, mobilise and powerfully influence the actions of human beings76 (AUICK, 1 994). As such, 72 Emergy is defined by Odum ( 1 996, p.288) as "all the available energy that was used in the work of making a product and expressed in units of one type of energy". 73 Huang ( 1 998, p.5 0 l ) defines transformity as "the ratio of energy of one type required to produce a unit of energy of another type". 74 The spatial arrangement of these zones would appear to favour a concentric formation where zones with higher energy hierarchy are located in the centre, with energy hierarchy of zones decreasing outwards. However, rapid urban sprawl results in uneven overlap and encroachment of adjacent zones when, for example, patches of the natural zone are developed and converted to suburban residential districts (Huang, 1 998). 75 This represents an alternative to the earlier zonation theories of such as Burgess's Concentric Zone Model (Park et al., 1 925), Hoyt's ( 1 939) Sector theory and Harris and Ullman's ( 1945) Multiple Nuclei theory. 76 The extended metabolism model of inter alia Newman et al. ( 1 996) attempts to address this by the inclusion of social aspects of sustainability into the model to achieve ' livability' in urban settlements. 78 the interlinkages between elements and flows within the urban ecosystem are not as tight as those within a biological ecosystem (Marcotullio and Boyle, 2003) . • Proximity. In biological ecosystems, metabolic efficiency and particularly recycling are made possible by the physical proximity of producers and decomposers, thus minimising energy losses in material transport. In an urban setting, however, recycling efficiencies are reduced by the physical separation of producers from consumers ( linked through long­ distance trade exchanges), and of consumers from recycling facilities (Gasson, 2002). • Open versus closed systems. Biological ecosystems are energetically open but materially closed systems characterised by cyclical metabolisms. The balance between their metabolic demands and environmental regenerative and assimilative capacity is brought about through negative feedback processes, making them sustainable. Urban ecosystems, however, are both energetically and materially open systems. They are driven by positive feedback processes such as population and economic growth, causing an imbalance between their metabolic demands and the regenerative and assimilative capacity of the surrounding environment - a situation that is ecologically unsustainable in the long term (Hughes, 1 974; Husar, 1 994; Gasson, 2002). Toolkit of Methods for Furthering the Urban Ecosystem Approach An array of operational methods developed over the last two decades that may be used to help understand 'cities as ecosystems ' : ( 1 ) industrial ecology, (2) materials flow analysis, and (3) physical input-output analysis. Generally these methods seek to understand the biophysical functioning of systems through the lens of energy and mass transformation that takes place. Industrial Ecology. Industrial ecology sees human economic activity as an integral part of the larger ecosystems that support it, especially in terms of resource supply and waste assimilation. Ecological concepts such as carrying capacity and ecological resilience are used to examine the extent to which human economic activity is undermining environmental services critical to humanity (Graedel, 1 994; Lifset and Graedel, 2002). Graedel and Allenby ( 1 995, cited in Allenby, 1 999, p .40), encapsulated the essence of the field as follows: Industrial ecology is the means by which humanity can deliberately and rational ly approach and maintain a desirable carrying capacity, given continued economic, cultural and technological evolution. The concept requires that an industrial system be viewed not in isolation from its surrounding systems, but in concert with them. It is a systems view in which one seeks to optimise the total materials cycle from virgin material, to finished material, to component, to product, to obsolete product, and to ultimate disposal. Factors to be optimised include resources, energy, and capital. 79 A core element of the field is the use of a systems perspective. Lifset and Graedel (2002) outline several different forms which reflect this systems orientation: ( 1 ) lifecycle perspective - this is reflected in the use of formal methods such as lifecycle assessment (LCA) where the impacts on the environment of products, processes and services from resource extraction through production, consumption and fmally disposal, are taken into acc6unt; (2) materials and energy flow analysis, including physical input-output tables - this involves tracing the ' industrial metabolism' , or flux of materials and energy through the various economic processes from extraction to disposal at various scales. Compliance with the First Law of Thermodynamics is a critical component of research in this area, involving the mathematics of budgets, cycles, stocks and flows77; (3) systems modelling - formal methods such as dynamic modelling reveal the complex interactions and feedbacks between system components driving the behaviour of the system under study; and (4) use of techniques and insights from multiple disciplines. Materials Flow Analysis. Bringezu and Moriguchi (2002, p .79) define materials flow analysis (MF A) as "the analysis of the throughput of process chains comprising extraction of harvest, chemical transformation, manufacturing, consumption, recycling and disposal of materials". MF A is based on a system of accounting, where the inputs and outputs of these processes are quantified in physical units, usually mass. These inputs and outputs may be chemical substances (e.g. carbon or carbon dioxide) or natural or technical compounds (e.g. coal or wood). Material flow accounting is only one of several steps in MF A. Different strategies for a sustainable industrial metabolism have been pursued, but an underlying concept common to all is that of the embeddedness of the industrial system and its societal interactions in the biophysical system. One of these strategies is detoxification of the industrial metabolism - mitigation of release of polluting substances to the environment. A second complimentary strategy is dematerialization of the industrial metabolism - a reduction in the throughput of the economy as a whole by increasing efficiency. Schmidt-Bleek ( 1 994a, 1 994b, 1 994c) introduced the factor concept with MIPS Factor 1 0, and Weizsacker et al. ( 1997) with Factor 4, which, in consideration of current resource use by industrial economies, proposed a four to ten fold increase in resource efficiency. This eco-efficiency concept is not only concerned with inputs such as materials, energy, water and land, but also outputs to the environment such as pollutant emissions and waste, relating them to the products or services 77 Ayres et at. ( 1994) and Socolow ( 1994) cite applications in the study of toxic chemicals, resource depletion, environmental degradation resulting from residual wastes, and the perturbation of biogeochemical life-supporting cycles. 80 produced (OEeD, 1 998; EEA, 1 999; Verfaillie and Bidwell, 2000; Bringezu and Moriguchi, 2002). Physical Input-Output Analysis. A physical input-output table (PlOT) is a comprehensive and detailed physical accounting system based on macroeconomic activity (Strassert, 2002).78 A PlOT traces not only the commodity flows of the traditional input-output table in physical units, but also the material flows across the environment-economy interface . While MF A is typically focused on economy-wide fluxes of mass and energy, PlOTs are structurally detailed, recording mass and energy exchanges across numerous industries. In this way complete materials balance can be achieved for the various economic activities (Stahmer et al. , 1 997). A key feature of physical input-output analysis is that all repercussionary effects across by industry within an economy may be evaluated. The PlOT is conceptually consistent with the ideas of authors such as Boulding ( 1966) and Daly ( 1 99 1 ) who view the economic production system as an open subsystem of the finite and non-growing natural environment. This economic subsystem imports low entropy matter-energy in the form of raw materials and exports high entropy matter energy, or waste (Daly, 1 99 1 ; Strassert, 2002). This one-way flow, beginning with resources and ending with wastes, is analogous to a digestive tract in an open biosystem, connected to the environment at each end (Daly, 1 995). Through incorporation of materials balance, the PlOT overcomes the major shortcoming of conventional national accounting, namely that the economic process is viewed as a closed circular flow from firms to households and back to firms, with no inlets or outlets (Daly, 1 995; Strassert, 2002). 3.2.3 Ecological Footprinting Wackemagel and Rees ( 1 996) pioneered the concept of Ecological Footprinting. The Ecological Footprint (EF) is the area of land required to produce the resources consumed and assimilate the wastes produced by a given population (Rees, 2000). The EF is a sustainability indicator for two main reasons: ( l ) it measures the total ecological cost of supplying goods and services for a population both directly (actual land for agriculture, housing and so on), and indirectly (embodied land in products consumed); and (2) it invokes the ecological concept of carrying capacity (the maximum population that can be supported indefinitely by a given land area). Where a region's popUlation 'overshoots' its carrying capacity, it is said to be in 78 Geddes ( 1 885) was ahead of his time in his attempt to develop a unified calculus based on energy and material flows, capable of providing a coherent framework for all economic and social activity. He developed a type of economic input-output table in physical terms, with the first column containing sources of energy and materials transformed into products. His work represents an embryonic form of an empirical description of societal metabolism on a macroeconomic scale (Martinez-Alier, 1 987; Fischer­ Kowalski, 2002). 81 'ecological deficit', using more productive land than it has available within its borders. This population can be said to be appropriating carrying capacity from elsewhere (Vitousek et al., 1986; Wackemagel, 1991 ; Rees, 1992). Ecological Footprinting highlights the fact that far from being geographically discrete, most of the land occupied or appropriated by urban areas lies beyond metropolitan boundaries (Rees, 1 992). In this way, Ecological Footprinting reconnects the urban centre, as a focus of consumption, with its hinterland, a focus of production, through trade and natural flows of ecological goods and services. In an ecological sense, the city as a node of consumption can be seen as a parasite existing on a vast external resource base, not only importing carrying capacity, but also exporting ecological degradation through environmental and economic exploitation of distant locations. The effect of urbanisation and trade is thus to physically and psychologically distance urban populations from the ecosystems that sustain them (Rees, 1992). The interdependencies which are created between the urban region and distant locations may not be ecologically sustainable in the long term. B icknell et al. (1998) and Ferng (2001 ), among others, have promoted input-output analysis as a systematic and standardised method for calculating EFs. McDonald and Patterson (2003c, 2004) have employed this approach to highlight the importance of ecological interdependencies at a regional level in New Zealand, stressing in particular what matters is not so much the size of the EF, but the location from which it is appropriated. Numerous studies using Ecological Footprinting in an urban context have been carried out. Wackernagel and Rees (1996) found in a study on Vancouver, Canada, that the Vancouver Regional District of 1 .6 million inhabitants and a land area of 293,000 ha, has an EF of 6,720,000 ha (or 23 times its geographic area). This equates to a per capita EF of 4.2 ha. Other researchers have reached similar fmdings. In a study of 29 cities in the Baltic Sea drainage basin, Folke et al. (1994) found that urban consumption of wood, paper, fibre and food was sustained by an ecosystem area 200 times larger than the urban area itself. EF studies of Canberra, Australia (Close and Foran, 1 998; Lenzen, 2004) have revealed a 1998-99 EF of 1 ,790,575 ha which, with a population of 312,300, equates to about 5.7 ha per capita. Regional EF studies have been carried out in New Zealand at the national level by Bicknell et al. (1998), and at the regionaVurban level by McDonald and Patterson (2002b, 2003c, 2004), including a number of urban areas. Auckland Region, for example, was found to have a 1997-98 per capita EF of 5.68 ha (adjusted for international comparison). These urban per capita footprints can be 82 placed in context when it is considered that Wackernagel et al. ( 1 999) estimate the average 'earthshare,79 (considered the maximum sustainable EF allowance) to be l .9 ha per capita. Limitations of the Ecological Footprinting Approach While the EF provides a valuable heuristic and pedagogic tool that captures current human resource use in a way that is readily understood (Costanza, 2000; Moffatt, 2000), the methodology does have several weaknesses and limitations, as discussed below: A lack of common definitions and methodologies for calculating the EF. A paucity of international conventions has led to ambiguities in interpreting the results of various EF studies. • Land as the numeraire. Since land is not the only scarce natural resource, it is valid to question the use of 'embodied land' as the sole numeraire for the calculation of a sustainability indicator. Many have argued (Slesser, 1 973 ; Gilliland, 1 975; Costanza, 1 980; H.T Odum, 1 983 ; Herendeen, 1 998) that a more appropriate numeraire might be 'embodied energy' or 'embodied solar energy ' . • Hypothetical energy land required for sequestration of CO2 emissions often constitutes a disproportionately large part of the EF. Other emissions and pollutants have ecological consequences yet they are ignored. Critics have also questioned the use of afforestation as the preferred option for CO2 sequestering. Commensuration of different land types. The use of 'equivalence factors' (adjustments made to take into account variations in biological productivities) is contentious. The focus on biological productivity ignores other relevant factors such as cultural values, social preferences or relative scarcity. • Dynamics - future scenarios? EFs provide a snapshot of current population requirements, but fail to inform of likely future scenarios. EFs are thus always 'yesterday's news' . Key dynamic components of the sustainability equation such as intergenerational equity, technological change, and adaptability of social systems are overlooked, as are non-linearities, feedback loops, and thresholds (Holling, 1 973; Levin, 1 998) which are characteristic of complex adaptive systems. • Policy relevance. The claim that EFs can evaluate potential strategies for avoiding ecological overshoot (Wackernagel and Rees, 1 996; Wackernagel and S ilverstein, 2000) is highly contested (Ayres, 2000; Moffatt, 2000). While the EF may not be sufficient for targeted policy action, it does however provide a high level indicator of 79 'Earthshare' is estimated by dividing the total amount of global productive land by the global popUlation (Wackemagel et ai., 1 999). 83 ecological impact. Its key feature is its ability to stimulate public awareness of the far­ reaching ecological effects of human activities. 3.3 Outstanding Theoretical Issues 3.3.1 Need for Maturation of the NEP-based Approaches to Urban Sustainability Urban theory has a long and rich history in the social sciences, spanning sociology, geography, psychology, and political science. While urban sociologists were concerned with the urban experience, fearing a breakdown of community ties and a subsequent negative effect on the human psyche, urban ecologists and geographers were more interested in how and where urban settlements arose and the factors that influence their internal arrangement. Urban psychologists focused on human reactions in adapting to an urban lifestyle, while urban political economists turned to social conflict, inequality and the greater political structures and processes of change in their search for an explanation of urban dynamics. The common thread in all of these schools of thought is their focus on the human condition with little or no regard for ecological issues. In spite of the increased environmental awareness of the last 20 to 30 years in academia, it is not surprising that most students of urban planning and studies are taught from an anthropocentric worldview that has dominated the social sciences over the last two centuries. Consequently, city and town planning practitioners and academics mostly operate within the HEP that underpins these disciplines. On the other hand, NEP theoretical approaches to urban development are comparatively new and underdeveloped, while the NEP literature on urban issues is relatively thin and dominated by only a few authors.8o Urban issues still seem to be largely the domain of the social sciences, rather than the natural sciences, in spite of the early bridging work of the father of town planning Patrick Geddes. The uptake of the NEP is more evident in non-urban areas of public policy such as catchment management, natural resource management and environmental conservation. Consequently, there is a clear a need to build and mature the NEP approaches to urban sustainability beyond the current urban ecosystem/metabolism and ecological footprinting approaches outlined in this Chapter. This process of building and maturation could involve: 80 Only very recently, for example, was the journal of Urban Ecosystems established (March 1 997). 84 • Empirical studies. A growth of empirical studies focusing on the ecosystem processes of cities and their ecological footprints. To date, there are few substantive studies (e.g. urban metabolism of Hong Kong) with most being of a more general nature. Without detailed and comparative empirical studies, the basis for developing theory and methods based on the NEP view of urban sustainability remains weak; • New ecological ideas and terms. Development of specific ecological ideas uniquely applicable to cities and urban spaces. There is a tendency to draw ideas and analogies directly from biological ecology and apply them to cities. At the very least, we should be careful when applying ideas such as carrying capacity to urban situations; • Stronger links between HEP and NEP research. Forging stronger links between the established REP research and the more recent NEP thinking. For example, there are very few studies such as Huang ( 1 998) that attempt to explain the ecological determinants of urban phenomena (e.g. spatial zonation) that has long been observed in the REP literature. The two schools of thought most often operate in complete isolation of each other; and • Institutionalisation. Institutionalising the NEP view of urban sustainability and building a critical mass of research activity in this area. Institutionally the NEP-based field of urban sustainability is weak with no strong international community of scholars or teaching institutions in this field. REP scholars from the social sciences dominate the field, while NEP scholars in urban areas are often marginalised. 3.3.2 Need to Integrate the HEP and NEP Approaches? The REP and NEP approaches to urban sustainability and development have so far been presented in this thesis as dichotomous choices in that they are based on fundamentally different views of the world. REP is based on the view that humans are unique among all species in that they have the 'know-how' to overcome any environmental problems, while NEP is based on the view that humans, like all species, are essentially constrained by ecological limits. Given these s ignificantly opposing views, is it therefore feasible and/or desirable to attempt to integrate these two approaches? One response to this question is that the REP theories that focus on human behaviour, institutions and political economy are not so much wrong as incomplete. They simply fail to take account of biophysical constraints, thresholds and discontinuities. If, however, they could acknowledge biophysical limitations, the resulting theory would be all the more rich. The understanding in political economy theory, for example, of cities as nodes of consumerism and globalisation is insightful, but could be further enhanced by integration with the ecological 85 footprint concept that links these consumerismlglobalisation processes to their ecological consequences. Such integrational theorising could lead to some useful outcomes, with both the REP and NEP perspectives being enriched. At some point, however, conflict between the assumptions that drive each paradigm might conceivably impose limits on such integration, e.g. the two perspectives are unlikely to accept similar assumptions regarding the ability of humans to overcome biophysical limits through technological advancement or environmental manipulation. A second response therefore is to accept that integration of the REP and NEP is mostly not feasible because the two perspectives have fundamentally different views of the world. In this case, 'methodological pluralism', rather than 'methodological integration' , would appear to be the best way forward (Norgaard, 1989). In other words, both the REP and NEP approaches can be pursued independently but in open dialogue with each other. The NEP scholars could continue to use ecological principles and ideas to explore issues of urban sustainability, which may provide useful insights into the public debate on urban policy issues as well as challenge REP-based perceptions of urban development. Similarly, the REP scholars would similarly explore issues of urban sustainability, but from their perspective. Dialogue and debate would ensue, and c ity planning practitioners would draw as appropriately from both the REP and NEP streams when formulating plans and policies. The approach taken in this thesis is to focus on the NEP-based concepts relating to urban sustainability, emphasising those ecological and thermodynamic sustainability principles outlined in Chapter 2. Rather than operating entirely in the ecological realm, however, an attempt will be made to link ecological factors with economic factors and, to a lesser extent, social factors. 3.4 Summary A summary of the REP and NEP worldviews and their different schools of thought is presented in Tables 3. l (a) and (b) respectively. The dominant anthropocentrism of the REP contrasts clearly with the consideration of environmental constraints under the NEP, particularly in their different views of the nature of urban problems, and their proposed solutions to these problems. Furthermore, the compartrnentalised approach of the REP to urban settlements contrasts with the more holistic systems approach of the NEP. It is important to note that energy and emergy analysis differ substantially (centre column of Table 3.1(b)), particularly in respect of conceptual underpinnings and accounting procedures 86 (Brown and Herendeen, 1 996). Energy analysis involves determining the direct and indirect (embodied) energy requirements needed to produce a good or service (lFIAS, 1 974). A key component of this analysis is calculation of indirect effects using an economic technique with strong similarities to that of input-output analysis (Bul lard and Herendeen, 1 975b). Emergy analysis uses principles of energetics (Lotka, 1 922, 1 925, 1 945), system theory (von Bertalanffy, 1968) and systems ecology (H.T. Odum, 1 975, 1 983, 1 988, 1 99 1 ), to determine the value of resources, goods and services in terms of the solar energy it took to make them (called solar ' emergy') . While emergy analysis has pioneered the concept of 'energy form' (all energies are not of the same quality and are thus expressed in the equivalent solar energy units), energy analysis does not recognise differences in energy quality. For a more complete comparative view between embodied energy analysis and emergy analysis, refer to Brown and Herendeen ( 1 996). 87 Table 3.1 (a) Conceptual Foundations for HEP-based Urban Schools of Thought Human Exemptionalism Paradigm Classical Urban Sociology Urban Ecology c.1920- Urban Geography c.1940- Urban Psychology c.1960- Urban Political Economy c.1850-1920' 1975 1980 1 980 c.1975- Scope of Area of . Social order, cohesion & Study ! Themes community ties · Arrangement of people within urban environment · Analogies drawn from ecology: competition, dominance, succession & invasion · Significance of urban . Human reac\ions to location & physical shape urbanisation Relationships Examples of Contributing Authors · Social, economic & political power relations social ! economic social social Marx, Engels, Tonnies, Park and Burgess (1967), Blumenfeld (1949), Losch Durkheim, Simmel, Weber Wirth (1938), Hoy! (1939), (1954), Christaller (1966), Harris and Ullman (1945), Byland (1960), Morrill Shevky and Bell (1955), (1963), Hudson (1969) Pederson ( 1967) · How people make sense of the urban environment social Lynch (1 960), Suttles (1 968), Milgram ( 1972), Lofland (1 973), Irwin (1 977), Gans ( 1982) · Political structures · I nvestment decisions · Economic trends · Trade pattems · Globalisation socio-political ! economic Walton (1981), Lefebvre (1991), Harvey (1978, 1 985), Castells (1977), Logan and Molotch (1987) Dominant Methodological Approach' Structuralism (Marx, Engels, Weber): Mainly POSitivism, ea�ier Positivism work Environmentalism Behaviouralism and Structuralism, to a lesser Positivism, more recently extent Managerialism & Environmentalism (T6nnies, Durkheim, Simmel) Humanism Postmodemism Nature of Urban Problems · Breakdown of community · Dominance of particular ties & relationships groups at the expense of · Feelings of alienation others · ExplOitation of proletariat . Traditional community by capitalist bourgeoisie values undermined · Meaningful social relationships difficult Solutions to . Overthrow of capitalism . Urban planning Urban Problems through revolution Drivers of Urban . Capitalism Change . Social, economic & political power relations Determinants of Spatial Differentiation · Class structure · Complex division of labour Future Prognosis . Pessimistic - for Cities ! Limits diSintegration of social life to Cities & relationships · Urban malaise Potential Pros of . Self-actualisation Cities . Individuality · Rationality · Competition between classes ! social groups · Technology e.g. transport, elevators, construction materials · Clustering of social factors e.g socio­ economic status, ethnicity · Class structure · Technology e.g. transport, elevators, construction materials · Other social factors e.g. socio-economic status, ethnicity · Urban planning · Pessimistic - if planning was unable to mitigate potential problems · Efficient organisation of urban space into 'ecological niches' · Unplanned urban growth . Loneliness pattems e.g. sprawi . Urban malaise · Exhaustion or . Alienation contamination of the . Lack of community resource which had initially led to the city's development · Urban planning · Technology e.g. transport, elevators, construction materials . Trade · Technology e.g. transport, elevators, construction materials · Trade · Urban planning · Social codes of behaviour · Urban networks · Establishment of Gemeinschaft-like relationships · Planning for livability · Desire for heightened urban experience · Urban networks & relationships · City needs to make the . More meaningful most of its strategic experience of urban life location to its advantage through adaptive · Spatial ! biophYSical mechanisms constraints govem . Urban life becomes a phYSical growth and shape dynamic creative mixture · Nexus of activity benefiting both the city and its surrounding hinterland · Economies of scale of experiences · Higher levels of self­ actualisation & sophistication Potential Cons of . Loss of sense of · Conflict between social . Uncontrolled growth e.g. · Failure to adapt may result in loneliness & isolation Cities community · Alienation groups sprawl · Societal segmentation as a result of shared 'ecological' pressures Note: l . This Table considers only contributions made during the indicated periods. 2 . Refer to Pacione (200 1 , pp,27-32) for definitions of these methodological approaches. · Institutional · Social conflict · Inequality · Uneven spatial development · Policy change · Political mobilisation · Redirection of investment · Global economy e.g. more open trade, TNC's · Investment decisions · Political structures · Govemmental policy · Uneven investment · Real estate investment · Informational mode of investment · Employment opportunities · Increased inequity · Sites of consumption rather than production · Growth opportunities as result of integration into global network · Greater opportunities through integration of material & information flows across the glob le · Economies of scale · Marginalisation & alienation of those disadvantaged by globalisation 88 Table 3.1 (b) Scope of Area of Study I Themes Relationships Examples of Contributing Authors Methodological Approach Nature of Urban Problems Solutions to Urban Problems Drivers of Urban Change Determinants of Spatial Differentiation Future Prognosis for Cities I Limits to Cities Potential Pros of Cities Potential Cons of Cities Conceptual Foundations for NEP-based Urban Schools of Thought New Ecological Paradigm Ecosystems View of U rban Settlements Urban Metabolism & Extended Urban Metabolism c. 1 965- · Metabolism metaphor used to describe city as an organism ecological I economic Wolman (1 965), Newcombe (1975a, 1 975b), Newcombe et al . (1 978), Boyden et al . ( 1981) , Newman et al . (1 996), Newton (2001 ) Energy/Emergy Analysis c. 1 970- · Determination of energy requirements for urban functioning · Urbanisation seen as process of change in type of energy sources ecological I economic Hannon ( 1973), Bullard and Herendeen (1 975), Brown ( 1 981 ) , Peet ( 1 986), Hendtlass et al . (1 988), Odum (1 996), Huang (1998) Ecological Footprinting c. 1 990- · Evaluates land-based carrying capacity of a given population ecological I economic Rees ( 1992), Wackernagel and Rees ( 1996), Bicknell et al. ( 1 998), Ferng (2001) , McDonald and Patterson (2003b, 2004) Systems Theory; Empirical (energy Systems Theory; Empirical (energy Systems Theory; Empirical and materials flow analysis) accounting); Thermodynamics (embodied land analysis) · Cities encountering . Thermodynamic limits to energy thermodynamic & biophysical limits availability · Linear metabolism . Convergence of energy flows on urban centres · Circular metabolism · Mimicking of efficient natural processes · Reduction of material & energy throughput · Increased eco-efficiency · Industry growth · Material affluence · Recycling and energy policies · Waste management practices · Relationship with rural hinterland · Natural resource and energy availabil ity · Urban sprawl & degradation of surrounding natural landscapes · Optimum efficiency for maximum power (emergy analysis) · Autocatalytic feed backs to reinforce production (emergy analysis) · Maximise the flow of useful energy · Thermodynamics · Maximum Power Principle (for emergy analysis) · Autocatalytic feed backs (for emergy analysis) · Energy availability · Zones with higher energy requirements tend to encroach on those with lower energy requirements · Humans exceeding carrying capacity · Critical dependence on surrounding hinterland and other nations · Appropriation of resources from localities with sustainable land management practices · Reduced and more selective consumption · Movement toward being self­ sustaining · Consumption & production patterns · Interregional and international trade · Population growth · Relationship with surrounding hinterland and other nations · Access to natural resources · Urban growth outstrips material and energy resource base · Slowdown of economic and . Large urban ecological footprints societal progress in urban centres are unsustainable, as long as there · Urban growth produces unsustainable levels of wastes due to thenmodynamic limits is an overshoot of · Decline of cities and urban international/national carrying · Urban growth is unsustainable unless 'dematerialisation' policies are adopted activities that are energy intensive capacities · Efficient use of natural resources . Financial and control activities due to economies of scale and are appropriately located in urban spatial concentration of activities in centres, as this makes sense from urban areas an emergy hierarchy view · Resource scarcity . Uncontrolled encroachment of · Environmental degradation lower energy zones by higher through waste, pollution, emissions energy zones · High infrastructural/technology . Ecological degradation costs to cleanse natural . Reduced flows of useful energy environment · Decline of cities over rural activities · Cities have low ecological footprints per capita et rural footprints · Overshoot of carrying capacity is inevitable with all cities PART 11 ENVIRONMENT-ECONOMY INTERACTIONS IN AUCKLAND RE GION A STATIC SYSTEMS ANALYSIS 9 1 Chapter Four Methodological Framework for a Static Systems Model of the Auckland Region 93 In this Chapter a static systems modelling framework of the Auckland Region economy, and how it interacts with its biophysical environment, is outlined in detail. This static systems framework is based on two extensions of conventional input-output analysis. Firstly, the analysis is extended to take account of the use of natural resource and ecosystem service inputs into the economy and the production of residuals (wastes) by the economy. Secondly, a physical input-output accounting system of the Auckland Region economy and its interaction with the biophysical economy is put forward, as a complementary framework to the conventional fmancial accounting framework used in input-output analysis. This physical input-output framework is important as it directly assists the understanding of the biophysical sustainability of the Auckland Region economy. This methodological framework outlined in this Chapter is then operationalised in Chapters 5 and 6, and to a lesser extent in Chapters 7 and 8. The framework not only provides detailed information on the structure of the environment-economy system in Auckland Region, but also provides the starting basis for the dynamic modelling undertaken in Chapter 1 1 and Appendix B. 4.1 Why Build Static Models? A key characteristic of many formal models is that they are static in nature. Static models are normally used to understand the functioning of a system at a particular point in time.81 For example, a static model would be used to answer questions like: 'how many people were employed directly and indirectly by the business services industry in Auckland Region in 200 1 ? ' or 'what was the level of residential water use in Auckland Region in 1 998?' In this way, static models are independent models in their own right (Hicks, 1 985 ; Ruth and Hannon, 1 997). There are several reasons why we would want to build a static model of the Auckland Region: 81 When considering time, economists crudely characterise models as static, comparative static, dynamic and comparative dynamic. 94 • Determining system state. A static model can yield theorems about "the values of . . . variables . . . in a state of rest" (Kuenne, 1 963 , p. 1 4). Thus, static models focus on state rather than on the process of change. • Setting initial conditions. By capturing state, static models may be used to set initial conditions for dynamic models (Ford, 1 999). Similarly, static models may be used to establish baselines or benchmarks, or validate dynamic models. • Comparative statics. Static models may be extended to formulate theorems about changes in the values of variables "between two states of rest" (Kuenne, 1 963, p. 1 4) or before and after change82 (Baumol, 1 970; Fisher, 1 983). • Analytical ease of use. Static models yield single solution vectors, compared with dynamic systems where "a set of such vectors is linked in a path through time" (Kuenne, 1 963, p . 1 4). • Structural analysis. The main advantage of static models is that they provide detail on the structure of the system being considered (see below). Static analysis, and in particular structural analysis, of economic and ecological systems is not however a new idea. Regional economists, for example, have used input-output analysis (Richardson, 1 972; O'Connor and Henry, 1 975 ; Leontief, 1 985; Lonergan and Cocklin, 1 985; Miller and Blair, 1 985) to study in detail the structural make up of their economies including identification of key economic interdependencies, the economic consequences of financial injection (e.g. infrastructure construction, major tourist event etc.), patterns of production and consumption, and trade analysis . Similarly, ecologists have studied structural relationships using methods such as network analysis (Hannon, 1 973 b), ascendency indices (Ulanowicz, 199 1 ), energy analysis (Hannon, 1 973a, 1 979, 1 982, 1 99 1 ; Hannon et al. , 1 984), and environs analysis (Patten, 1 98 1 , 1 982). Moreover, authors such as Isard ( 1 972, 1 975) and Victor ( 1 972a, 1972b) have extended input-output analysis to study the structural make up of integrated ecological-economic systems. Careful consideration and caution must, however, be given to the findings of any static analysis. In a temporal sense, static models are closed, permitting the functioning of a system to be examined without any reference to that which is going on outside it (Hicks, 1 985). Interpretation of static analysis therefore requires one to assume that the system is in, or near, equilibrium (Baumol, 1 970; Fisher, 1 983). Of course, this means that no planning has taken place, or possibly could take place, within the system that could influence any possible future 82 For example, a comparative static model of Auckland Region's economy could answer questions such as 'what was the annual trend of employment in business services between 1 987 and 2004?' or 'how much land was used by the mining and quarrying industry in 200 1 and 2003? ' / 95 state of the system. As Fisher ( 1 983, p.3) puts it, "convergence to equilibrium must be sufficiently rapid that the system, reacting to a given parameter shift, must get close to the predicted new equilibrium before parameters shift once more". In other words, for static models to be useful, the temporal variables that have been assumed to be static with the period of study must also remain relatively static in the real world. Many real world systems, including the Auckland Region environment-economy system, are characterised by constant change resulting from the presence of feedbacks, time lags and the like associated with investment, construction, climatic conditions and so forth. In the context of understanding how the Auckland Region environment-economy system functions, temporal implications associated with human perturbation or stochastic environment change are dealt with in Chapters 1 1 and Appendix B of this thesis. 4.2 Input-Output Analysis as the Basis for an Integrated Environment-Economy Systems Framework The pathway pursued here in the development of an integrated environment-economy systems framework for the Auckland Region is based on input-output analysis. Input-output analysis, developed by Wassily Leontief during the 1 930s, provides a comprehensive snapshot of the structure of inter-industry linkages in an economy. Most developed nations prepare input­ output tables at regular intervals. Generally speaking, an input-output table of a nation is conceptually reconcilable with its system of national accounts (SNA). In addition, input-output tables adopt internationally recognised systems of commodity/industry classification (e.g. the Central Product Classification (CPC), International Standard Industrial Classification (ISIC), and the Harmonised System (HS)). This facilitates comparison across space and through time. Although input-output tables (models)83 are usually presented in financial terms, authors such as Daly ( 1 968), Isard ( 1 968), Kneese et al. ( 1 970), Leontief ( 1 970) and Victor ( 1 972a) inter alia have demonstrated that biophysical information on resource use and generation of residuals (i .e. waste, pollution, emissions etc. ) may also be placed in an input-output framework. More recently, analysts such as Stahmer et al. ( 1 996, 1997, 1 998) and Gravgard ( 1 998) have 83 An input-output table is no more than a numerical description of the flows in an economy, in tabular form. These numbers (in the 'Table') can be readily converted into: (i) mathematical matrices (and vectors). Matrices are mathematically rectangular arrays of elements, set out in rows and columns, that facilitate the solution of mathematical problems; (ii) mathematical models which are systems of simultaneous linear equations of the interdependencies (flows) in the economy. The author has attempted to distinguish between input-output 'tables ' , ' matrices' and 'models' in this thesis, but sometimes the terms can be used interchangeably. Appendix A contains a description of how input-output tables can be converted to corresponding ' matrices' and 'models' . 96 generated national physical input-output tables. Along these lines, an input-output approach may be used to assess Auckland Region' s environment-economy system. Moreover, there are several reasons for the adoption of an input-output approach: • It is a comprehensive method. It provides a detailed decomposition of the structural relationships in an entire economy. This includes complete coverage, albeit through aggregation, of the types of commodities produced, of the production processes used, and offmal consumption within an economy. • It is a systematic method. Input-output matrices provide a convenient checklist to ensure that all flows are taken into account. The conservation principle (i .e. inputs equalling outputs) of input-output accounting further ensures that there are no I unintentional blind spots. • It avoids common methodological pitfalls. Input-output analysis avoids double counting, particularly when dealing with complicated networks of indirect flows that have significant feedbacks, and joint production problems when allocating multiple outputs of commodities from a single economic process. • • It is a mathematically rigorous method. The use of matrix algebra is not only efficient in dealing with large computationally complex datasets, but also enables analysis to be undertaken in an internally consistent mathematical framework. Using input-output mathematics, it is possible to calculate first, second, third . . . nth round (i .e. infinite regress) effects accurately and comprehensively. It is easily extended. Input-output models may also be utilised in short run comparative static analysis, for example, to study the implications of policy change, economic growth, population change, capital investment, trade and so on. Furthermore, the distributional impacts of change may also be traced with extensions to the basic framework, including socio-economic effects (e.g. Economic Impact Assessment, Social Accounting Matrices) and price change (as an input into Computable General Equilibrium Modelling). A brief description of the mathematics of input-output analysis along with its most critical assumptions is provided in Appendix A - full details may be found in Richardson ( 1 972), O'Connor and Henry ( 1 975), Leontief ( l985) and Miller and Blair ( 1 985). 4.3 Critical Review of Environmental Input-Output Modelling Environmental input-output models modify and extend the conventional input-output framework to include resource use and residual (Le. waste, pollutant and emission) generation. 97 A key feature of environmental input-output modelling is that it is principally concerned with the environment-economy interface; in particular, how changes in the economy might impact on the environment (e.g. resource provision/scarcity, residual generation, and the costs of substitutes/abatement) or vice versa. Authors such as Miller and Blair ( 1 985) have tentatively grouped environmental input-output frameworks into three categories : • Generalised or augmented input-output models. These are typically formed by adding rows and columns, representing pollution generation and abatement activities84, to a technical coefficients matrix. A matrix of pollution or abatement coefficients, P, is defmed where each element, Pkj, represents pollutant k generated per dollar of output of industry j. Multiplying P by the Leontief Inverse yields the direct and indirect pollution, P*, produced per unit of final demand generated in industry j i.e. p* = P(1 - Arl Y. Although simple, this approach can provide valuable insight into the magnitude of the indirect environmental impacts associated with changes in economic activity. • Inter-industry input-output tables . These models extend the basic inter-industry framework to include an environment sector, under which the use and production of ecological commodities is recorded (Miller and Blair, 1 985). Several examples of inter­ industry economic-ecological models, as discussed below, include Cumberland ( 1 966), Daly ( 1 968), Ayres and Kneese ( 1 969) and Leontief ( 1 970). • Commodity-by-industry models. Such models treat resource use and residuals production as commodities in the form of a commodity-by-industry framework. The key difference between a commodity-by-industry and an industry-by-industry model is data is compiled at a commodity level rather than industry level which aggregates to form homogeneous outputs. Thus, multiple outputs per industry are permitted. Examples of commodity-by-industry models, as discussed below, include Isard ( 1 968, 1 972, 1 975), Victor ( 1972a, 1 972b) and the recent Physical Input-Output Tables (Models) developed by Stahmer et al. ( 1 996, 1 997, 1 998) and Gravgard (1 998). 84 Resource requirements have also been determined e.g. for energy (Gilliland, 1 975; Hannon, 1 979; Costanza, 1 980; Giampietro and Pimentel, 1 99 1 ), water (Hite and Laurent, 1 97 1 , 1 972; McDonald and Patterson, 1 998) and land (Bicknel l et al. 1 998; Ferng, 200 1 ; McDonald and Patterson, 2004). 98 4.3.1 Inter-Industry Environmental Input-Output Models 4.3.1 . 1 Cumberland Model85 Cumberland ( 1 966) is generally acknowledged as the first to develop an environmental extension of input-output analysis.86 In order to incorporate production of externalities into the analysis of alternative regional development strategies, Cumberland ( 1 966) argued that additional rows and columns could be augmented to a conventional input-output matrix to accommodate environmental costs and benefits associated with economic activity. The Cumberland model is shown in Figure 4. 1 . Rows Q and C respectively record in financial terms the environmental benefits and costs associated with each industry and fmal demand category for a given development programme. As a minimum, Cumberland suggests that each development program cover water, air, open space and possibly aesthetics and personal safety. The net environmental benefit row R records the difference between rows Q and C, while the column vector B measures the costs, by industry, of restoring the environment to its pre­ development state. 85 This Section outlines a various environmental input-output 'models' . Strictly speaking, these descriptions are of 'tables' as generally there is no matrix or equation structure specified - the exception being Section 4.3 . 1 .4 (for the Leontiefmodel). However, the author uses the term 'models' as there is an implicit l inear equation structure even though it is not explicitly specified. 86 Herfindahl and Kneese ( 1 965) had however previously described the resource/environment interface from an economic perspective in a framework comparable with an input-output table. 99 c:: ca c:: .2 .2 0. � 0. E '" c:: '0 E ::J c:: 'S CtI ::J '" CtI (ij '" c:: E 0. co c:: 0 'S 0 () Q) 0 (ij () c 0 C . .." (ij '" � c: '0 Q) '" Q) (ij '0 E c:: (ij e E � � � '0 .c E u:: '" '0 C> c:: Cii Cii Cii l- Q) t: I- e '" Q) Cii 0 (ij ::J ::J ::J .0 ::J > .c 0. .0 .s; '0 '0 '0 ::J 0 0 0 )( ::J '0 c:: .E .E .E m J: C> UJ m I- UJ Industry 1 Industry . . . i . . . Quadrant I Quadrant 1 1 8 / Industry n Sub Total Labour Value Added Quadrant I I I Quadrant IV Other Primary Inputs Imports Sub Total Total Gross Input Environmental Benefits (+) Q Environmental Costs (-) C Environmental Balance R=(Q-C ) Rj Figure 4.1 Cumberland Model. Adapted from Cumberland ( 1966, p.68). One criticism of this approach is difficulties exist with the non-market valuation87 of environmental costs (e.g. restoration) and benefits. A further deficiency of the Cumberland model is that it ignores the flows from the environment into the economy and vice versa. Richardson ( 1972, pp.2 1 8-2 1 9) suggests that this is because "Cumberland intends his extended input-output model to be used as an aid to a goal-oriented regional policy or development programme and not as a general inter-industry analytical tool". Richardson ( 1 972) also asserts that the Cumberland model more closely resembles a cost benefit evaluation than an input­ output table. Therefore it can be argued that the Cumberland model suffers from all of the shortcomings of cost benefit analysis as outlined by authors such as Blarney and Common ( 1 994), Norton ( 1 995), More et al. ( 1 996) and Kahn (2005). 87 This includes the well known limitations of non-market valuation techniques such as contingent valuation, hedonic pricing, willingness to pay, replacement cost methods and willingness to accept compensation (refer to Kahn (2005, pp. 92-1 28) for further details). 1 00 4.3.1.2 Daly Model Daly ( 1 968) proposed a highly aggregated model of the environment-economy interface based on an industry-by-industry framework (Figure 4.2). The model is divided into two domains: human and non-human. Conventional economic activities, such as agriculture, industry and households, are categorised under the human domain, while biophysical/ecological processes are classified within the non-human domain. The biophysical processes are further subdivided into living (animal, plant and bacteria) and non-living (atmosphere, hydrosphere and lithosphere) transformers of matter-energy. Interdependence between processes within the human and non-human spheres is portrayed respectively in Quadrants I and IV. Quadrant III represents the reverse flow of 'free goods' (e.g. resource inputs) while Quadrant II depicts the flow of externalities (e.g. residuals) between the human and non-human spheres. Mixed financial and physical units are utilised to describe the flows. Human Non-Human '2 .Q c.. E '2 ::J rJl 0 e: a 0 u E re ::J rJl e: e: !e. ID Q) 0 .... Q) u Q) '0 .... Q) Qj .... '0 ID � re .3 � C1l � a. � e: 'S � re .� a. rJl a. !e. en Q) rJl 0 rJl U rJl E C 0 .... 0 re ·c ::J ::J U E '0 £ .><: Cl '0 0 'c C1l C1l >- e: 0 -< E I -< 0:: CD ::c I :.:J U5 f0- e: Agriculture C1l E Industry Quadrant I Quadrant 1 1 ::J I Household (final consumption) Animal Plant e: Bacteria C1l E ::J Atmosphere Quadrant I I I Quadrant IV I C 0 Hydrosphere z Lithosphere Sun (primary services) Figure 4.2 Daly Model. Adapted from Daly ( 1968, p.402). Not satisfied with a purely descriptive tool, Daly proposed the calculation of technical coefficients by dividing each row element by its corresponding row total. This approach has however been criticised on the grounds that the economic and ecological commodities cannot be totalled as they are expressed in different units. According to Victor ( 1 972a, pAl ) these row 1 0 1 totals are meaningless, "despite Daly's unsubstantiated claim that 'there appear to be no theoretical problems in extending the input-output model in this way"'. Daly's adoption of an industry-by-industry framework for analysing environment-economy interactions results in a number of additional complications. Firstly, the homogeneity assumption is i llogical when transferred to the non-human (ecological) domain because aggregation of different ecological commodities is not possible due to the absence of a consistent numeraire. Secondly, in the adoption of non-comparable units the model tries to commensurate ecological commodities, which have no price, with economic commodities which do. Thirdly, the linearity assumption converts many non-linear ecological interdependencies to a linear nature. And finally, the assumption of fixed proportions of inputs is not necessarily valid when considering ecological interrelationships. 4.3.1 .3 Ayres-Kneese Model Ayres and Kneese ( 1969) developed an extended inter-industry model incorporating resource use, residuals, pollutant abatement and recycling. A key feature of the model is that it invokes the 'materials balance principle' i.e. mass and energy must be conserved across the model. The Ayres-Kneese model is depicted in Figure 4.3 . Coefficients in the extraction matrix I, and production matrix II, form a conventional inter-industry input-output matrix. These coefficients represent the fractional inputs per unit of output, as measured in pecuniary terms, of each industry. The coefficients in the abatement matrix III represent the actual costs of abatement. Additionally, matrices representing resource inputs, R, and residual outputs, W, are also incorporated. R Resource Input Matrices IV V VI ---------------------------- ---------------- ------------------.--------- ---------------- en en en ... CS ... 0 0 1:3 1:3 1:3 Conventional Input-Output ID ID ID en en en Table ID I c: 1 1 C I I I > 0 ID plus "Abatement" Sector t5 TI E � ::J ID '0 co X 0 w ... .J:l a.. u "0 UJ .EO A 0 H c' e' s R "0 c: co E Ql 0 '" m m c: (5 u:: t- B c e K I 0' p Ecological Outputs "0 c: co ...J G F q' Figure 4.6 Victor Model. Adapted from Victor ( 1972a, p.56). Quadrants with no symbols in them represent null matrices i.e. matrices containing only zero elements. The accounting framework used by Victor is essentially a commodity-by-industry matrix, in Stone's (1961, 1966) supply-use format, appended with additional rows and columns respectively representing ecological inputs and outputs. Economic transactions are represented in fmancial terms, while entries in ecological sectors are expressed in physical units. The 1 08 ecological commodities that constitute the ecological sector are classified under three headings: land, air and water. In addition to the conventional input-output accounting identities, Victor defmes several ecological accounting identities based on the materials balance principle.91 Using this framework Victor developed a series of analytical models relating economic production to effects on the environment in terms of resource use and residual generation. First, he created a set of ecological impact matrices, with and without import leakages. Second, by using shadow prices to represent the social valuation of ecological commodities, he outlined a procedure for using the impact matrices to derive estimates of ecological costs of producing and consuming economic commodities. And third, he disaggregated the estimates of ecological inputs and outputs by province, adding a valuable spatial dimension to his model. 4.3.2.3 Physical Input-Output Tables Physical Input-Output Tables (PlOTs) not only trace the physical flow of commodities through the environment, but also between the environment and the economy and vice versa (Stahmer et al.,. 1 997). A cornerstone of the PlOT framework is adherence to fundamental physical principles, particularly materials and energy balance as required by the first law of thermodynamics. PlOT accounting is a recent phenomenon.92 Katterl and Kratena ( 1 990) are credited with pioneering the first PlOT - a partially complete table of the 1 983 Austrian economy (Strassert, 2000). Old Uinder, a PlOT of the 1 990 West German economy, was the first complete and official table to be constructed (refer to Stahmer et al. ( 1 996, 1 997, 1 998)). Several other PlOTs have followed, including an official table for the 1 990 Danish economy (Gravgard, 1 998), and less ambitious unofficial efforts for Italy (Nebbia, 1 999) and the United States (Acosta, 2000).93 A PlOT is typically presented in a tabular commodity-by-industry format with production processes (industries) described by their material inputs and outputs in physical units i .e. tonnes (Figure 4.7). Each input is described by its industrial origin (or as imports), while each output is 91 This assumes that the model is a closed economy and there is no accumulation of mass in the economy itself (Victor, 1 972a). 92 The roots of physical accounting can be traced back much further. Strassert (2000) identifies two main analytical strata, namely, production theory and national accounting. The former stratum is based on the physical economy-environment work of Georgescu-Roegen ( 1 97 1 , 1 979a, 1984) and Perrings ( 1 987), and the latter stratum on Stahmer ( 1 988, 1993), the United Nations ( 1 993a, 1 993b), Radermacher and Stahmer ( 1 996) and Stahmer et al. ( 1 996, 1 997, 1 998). Underpinning these strata is the earlier materiaVenergy balance work of Ayres and Kneese ( 1 969), Kneese et al. ( 1 970), Ayres ( 1 978, 1 993b) and more recently the Material F low Accounting efforts of inter alia Bringezu (2000). 93 The only regional attempt to develop a PlOT appears to have been undertaken by Baden-WUrttemberg ( 1 990, cited in Strassert (2000)) for Bundesland in Germany. 1 09 explained by its destination i.e. industry, final consumption or exports (Strassert, 2000). Strassert (2000), in his description of the German PlOT, uses five matrices to describe physical flow. Matrix I, the intermediate production matrix, describes physical flow within the economic system. Matrix HA describes the fmal consumption of physical commodities by households etc. , while any residuals (Le. waste, pollutants and emissions) produced in production or consumption are described in Matrix IIB. Similarly, Matrices IIIA and IIIB respectively describe the use or conservation of material funds94, and the use of natural resources (i.e. renewable, non-renewable and recycled) supplied by the environment as an input into the production process. Intermediate Production Il i A Primary I nput A Use and Conservation of Funds II I B Primary Input B Natural Resources Solid Fluid Gaseous I I A Final Production A Final Consumption 1 1 B Final Production B Residuals Solid Fluid Gaseous Figure 4.7 A Physical I nput-Output Model. Adapted from Strassert (2000, p.3). Unlike a conventional input-output model which focuses on the structural nature of economic transactions (Matrix I), fmal consumption by households and exports (Matrix HA), and particularly the contribution made by each production process to value added (Matrix IIIA), a PlOT tends to focus instead on the structural nature of economic transactions (Matrix I), resource use (Matrix IIIB), residual production (Matrix IIB) and particularly on the completeness of materials balance. Moreover, the PlOT is conceptually consistent with the ideas of authors such as Boulding ( 1966) and Daly ( 1 99 1 ) who view economic production as a 94 Value added components such as labour and capital are conceived of as funds or agents transforming the flow of natural resources into flows of products (Daly, 1 995). 1 1 0 subsystem encapsulated within a finite and non-growing environment. This conceptualisation implicitly captures the role played by economics in extractinglharvesting low entropy matter­ energy and ultimately producing high entropy matter-energy. Consequently, this one-way flow beginning with resources and ending with residuals can be thought of as the digestive tract of an open biological system connected by the environment at both ends (Daly, 1 995). 4.4 Static Systems Framework Used in this Research for the Auckland Region The focus of the remainder of this Chapter is on developing a static systems environment­ economy framework for the Auckland Region. This framework provides a comprehensive, consistent and robust analytical mechanism for quantifying the linkages between the Region's economic activity and its biophysical environment. The detailed methodological steps used in development of this framework, and the analytical fmdings related to Auckland Region, are described in Chapters 5, 6 and 7. 4.4.1 Conceptualisation of the Environment-Economy System In its simplest form Auckland Region's environment-economy system may be conceptualised as an economy connected to its surrounding environment via material and energy flows. The Auckland Region does not however exist in isolation, but rather interfaces with various other environment-economy systems. The Auckland Region environment-economy system, and its relationship with other systems, is depicted in Figure 4.8. radiation Figure 4.8 Assimilative capacity Biogeochemical' cyc'iing -1 , ... _ - _ ... , ' Assimilative capacity Biogeochemical cycling t " , , , , Assimilative capacit}\ Biogeochemical cycling -1 , , ... ... ... ... � ' , , " , Auckland Region's Environment-Economy System and its Relationship with Other Systems World Environmental System Outgoing radiation (reflected by clouds, dust & earth's surface; & degraded heat) 1 12 4.4.1 .1 System Boundaries System diagrams are used to defme the Auckland Region environment-economy system, its relationship with other systems, and to identify the key flows and influences between these systems. Oblongs are used to delineate environmental system boundaries, while an elliptical interface is used to separate an economy from its environment. These system boundaries are based on an ecological-economic world view.95 Environmental Systems (Denoted by light shading on Figure 4.8) A nested hierarchy is used to represent the Auckland Region, New Zealand and World environmental systems. The Auckland Region environmental system is nested within the New Zealand environmental system, which is in turn nested within the World environmental system. These environmental systems encompass all physical flows not covered by the flow of economic commodities. The geospatial boundaries of Auckland Region and New Zealand are aligned to Statistics New Zealand's regional boundary definitions.96 This does not include any part of New Zealand' s Exclusive Economic Zone (EEZ). Economic Systems (Denoted by dark shading on Figure 4.8) Each economic system is, according to the adopted biophysical world view, seen as a subsystem of the wider environmental system in which it resides, i.e. the Auckland Region, rest of New Zealand and rest of World economies are located respectively within the Auckland Region, New Zealand and World environmental systems. The flows contained within the economic systems are, where possible, defined according to the United Nations (1993b) SNA. This includes the flow of commodities and services produced for production97 (i.e. intermediate demand) and to 95 System boundaries are almost always an artificial or abstract construct governed by a particular world view. Underpinning any world view is a set of values and beliefs. Although the framework developed here can help us to understand the implications of human-induced or stochastic change on the Auckland Region environment-economy system, it has limited ability to help us resolve any consequential ethical dilemmas that may arise. 96 Much has been written about the appropriate selection of spatial boundaries. In ecological footprinting, for example, Wackernagel and Silverstein (2000) argue for political or cultural boundaries as they represent the level at which environmental policy or decision making most often occurs. By contrast, van den Bergh and Verbruggen ( 1 999) dispute such boundaries on the grounds that they have no environmental meaning, favouring instead hydrological, climate zone or larger connected ecosystem boundaries. In this thesis, Statistics New Zealand's Auckland Regional Council boundary is used, which reflects both political (i.e. as the principal agency involved in regional environmental governance) and �eographic (i.e. river catchment) boundaries. 7 Industries which perform this production are made up of resident establishments. In some cases these establishments may, for example, be located within the Auckland Region, but operate outside the region and vice versa. This may result in coding difficulties. 1 1 3 satisfy fmal demand (i.e. domestic household consumption and exports), temporarily stored in physical stocks, or used as capital for production (i.e. accumulation). Patterson (2002b, p.21 ) argues, from a similar biophysical worldview, that several important sustainability issues arise from the above conceptualisation: • The extent to which the economy occupies the biosphere space. Estimates by Vitousek et al. ( 1 986) indicate that the economy has appropriated 40 percent of the Net Primary Productivity of the terrestrial biosphere. The ultimate physical limit cannot exceed 100 percent and certainly, to have a safety margin, it has been argued that realistically this l imit could be more like 80 percent. • The extent to which the sustainability of an economic system depends on its surrounding environmental system as a source of raw material inputs. Many raw material inputs are clearly finite and depletable e.g. fossil fuels, minerals, land and so on. Economic growth carmot be sustained indefinitely if these resources are depleted or degraded. • The extent to which the sustainability of an economic system depends on the sink functions of its surrounding environmental system. Often the environmental system is relied on to efficiently purify and absorb economic residuals. There are however critical thresholds beyond which the environmental system cannot cope e.g. eutrophication of a lake, or at a global level the warming resulting from the environmental system's inability to absorb greenhouse gas emissions. • Critical limits exist that make complete recycling of material residuals a physical impossibility. The Second Law of Thermodynamics tells us that degraded energy outputs can never be recycled - it has also been argued that the degree to which we may recycle materials is also limited.98 4.4.1 .2 System Flows A critical step in conceptualising Auckland Region's environment-economy system is to establish clearly the types of flows that exist within the various economic and environment systems, and between these systems. Nevertheless, before these flows are established it is useful to identify the types of media through which these flows take place. 98 Refer to Chapter 2 for further details. 1 14 Flow media There are four pnmary media99 through which flows take place III the Auckland Region environment-economy framework: • Raw material inputs . 100 These comprise natural resources 101 (e.g. mineral, energy, soil, water and biota) and inputs from ecosystems102 (e.g. gases needed for combustion). Natural resources traded on a market become commodities - see below. • Residuals. These include solid waste, liquid pollution and gaseous emissions. Residuals may, or may not, c irculate in an economy (i.e. through recycling or reuse) for some time, but are ultimately, at some point in time, discarded to the environment. 103 • Commodities . 104 These comprise goods and services that are traded on financial markets. Commodities are grouped according in the Australia New Zealand Standard Commodity Classification (ANZSCC). • Ecosystem services. These are services provided by the environment that are critical to our very existence. They include services that provide environmental regulation functions such as biogeochemical cycles and other biosphere processes (providing clean air, water, soil and biological control), habitat functions (e.g. refuge and reproduction habitat), and information functions (e.g. opportunities for reflection, spiritual enrichment, and cognitive development). Environmental System Flows The 'white space' labelled 'Biogeochemical cycling' and denoted by the broken lines represents a complex mix of biogeochemical processes that cycle materials and energy (refer to Figure 4.9 for an expanded view of Auckland Region's environmental system). Such feedbacks may cleanse, purify and assimilate both natural materials and economic residuals . Over long time periods (often measured in geological time) they may also transform waste residuals into useful 99 Some environment resources (e.g. land) are only used in situ and thus have no associated physical flow. Nevertheless, a media type may embody resources which are used in situ. It is therefore possible to think of a commodity flow, say for example the purchase of a burger bun, in embodied land terms. 100 The definitions of raw material inputs and residuals provided here are based on the United Nations ( 1 993a, 2003) SEEA system. 101 Some natural resource flows simply represent displacement such as mining lag or water for electricity generation. 102 It is important that ecosystem inputs be seen as distinct from ecosystem services. Ecosystem services encompass a much wider sphere of influence including b iogeochemical cycling, the assimilative capacity of the environment, the provision of biodiversity and so on. Auckland Region's ecosystem services are assessed in Chapter 7 and Appendix J of this thesis. 103 Careful classification of residuals is essential if double counting is to be avoided. Incinerated landfill waste may cause, for example, methane emissions resulting in a classification overlap. 104 The definition provided here is based on the United Nations ( l 993b) SNA system. 1 1 5 material inputs. Nevertheless, perturbation of these cycles through economic activity may lead to more immediate consequences (e.g. eutrophication of lakes or global warming through the release of greenhouse gases) which may, in turn, affect economic systems (den Elzen et al., 1 995). Because these cycles are characterised by dynamic feedback loops, non-linearities and time lags the consequences of perturbing them is often unforeseen. A crude static snapshot of biogeochemical flows occurring within Auckland Region, within New Zealand and for the globe is generated in Chapter 7 and Appendix B of this thesis. Incoming solar radiation �, Imports from other regions & nations e.g. commodities, services, knowledge & recycled goods ($, mass & energy) Mass & energy fromlto other regions & nations � Figure 4.9 " " . . .. -- .. - .. ... - .. , - - - hydrological cycle , , , , , , , , , , , , , " ptlosphorus Cycle , " '" " - - ... - .. ... ... .. - - - World Environmental System Auckl.and Region Environmental System � - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Assimilative capacity .. .. .. .. .. .. .. .. .. -'" - - -.�- v - - .. .. .. ... ... _ - - - .. ... ... .. .. .... .. .. - .. .. .. .. - " - _ .. -. .. ... " , sulphur cycle . " .. ... ,.... , carbon cycle " .- , , , , , ' " " , ... , ,,, ,,' , .. " " Auckland Region's Environment-Economy System With an Expanded Environmental System " ... .... ... .... ... ,r Exports to other regions & nations e.g. commodities, services, knowledge & recycled goods ($, mass & energy) Outgoing radiation (reflected by clouds, dust & earth's surface; & degraded heat) " � 1 1 7 Economic System Flows The Auckland Region environment-economy system is expanded in Figure 4. 1 0 to show clearly the types of flows that exist within the economic sub-system. Based on a conventional economic perspective, the within economic system flows may be described as follows. People provide labour, investment capital and knowledge (e.g. technology) to manufacture commodities and provide services, and are in turn rewarded with income (payments for labour/knowledge, dividends from investments) for their efforts. They use this income to purchase commodities and services (i.e. consume) or to reinvest in the economic process (i.e. capital formation). The production process involves numerous inter-industry exchanges until a [mal commodity/service is delivered to consumers. In this way, the economic flows are characterised by a closed loop between consumption and production (Ryan, 1 995). Raw material inputs and any residuals generated receive only cursory consideration from an economic perspective. An economic snapshot of Auckland Region's and New Zealand's within economy flows is generated for the study year using a fmancial input-output model in Chapter 5 of this thesis. lOS 105 Financial input-output models, both commodity-by-industry and industry-by-industry, were also developed for all of New Zealand's 74 Territorial Local Authorities and 1 6 Regional Councils. Incoming solar radiation Imports from other regions & nations e.g. commodities, services, knowledge & recycled goods ($, mass & energy) Mass & energy from/to other regions & nations Figure 4.1 0 World Environmental System Auckland .R�gion Environmental System Auckland Region's Environment-Economy System with an Expanded Economic System Exports to other regions & nations e.g. commodities, services, knowledge & recycled goods ($, mass & energy) Outgoing radiation (reflected by clouds, d ust & earth's surface; & degraded heat) 1 1 9 Daly ( 1 99 1 , p. 1 96) is particular scathing of the above view of an economic system: "Studying an economy in terms of the circular flow without considering the throughput is like studying physiology in terms of the circulatory system without ever mentioning the digestive tract". Greater understanding of the environment-economy system may be obtained by studying Figure 4. 1 0 not only from a conventional economic perspective, but also from a biophysical perspective. The biophysical view of an economic system is as follows. Within an economic system low entropy energy inputs (e.g. fossil fuels) and low entropy matter from the environment (e.g. minerals, biomass, water) are transformed through various production and consumption processes, sometimes stored or recycled, but ultimately degraded into high entropy residuals that flow back into the biophysical environment. A biophysical snapshot of Auckland Region's and New Zealand's within economy flows is generated for the study year using a PlOT in Chapter 6 of this thesis. Flows across System Interfaces Figures 4.8 to 4. 1 0 also reveal a number of important economic and biophysical flows that occur between systems. It is beyond the scope of this thesis to model all of these flows - some are modelled, some not at all, some only partially. These flows include (the degree to which they are modelled in this thesis is also indicated): • Auckland Region Economic System from/to Auckland Region Environmental System. This includes the use of raw material inputs (extraction/harvest of natural resources e.g. minerals, energy, soil, water and biota. Inputs from ecosystems e.g. gases for combustion), dumping of residuals (e.g. solid waste, water pollution, gaseous emissions) and release of waste heat. A comprehensive biophysical snapshot of these flows (measured in tonnes) is generated for the study year using a PlOT in Chapter 6 of this thesis. • Auckland Region Economic System from/to Other New Zealand Regional Economic Systems. These flows represent import/exports of economic commodities and services. An interregional trade optimisation model is developed in Chapter 8 and Appendix C that traces import/export flows between Auckland Region and each of New Zealand's fifteen remaining Regional Councils for the study year. The flows in this model are recorded in pecuniary ($), embodied land (ha), and embodied CO2 (tonne) terms. • Auckland Region Economic System from/to the Rest of World Economic System. These flows represent imports/exports of economic commodities and services. A comprehensive snapshot of these flows (measured in $ and tonnes) is generated for the study year using a financial/physical input-output model in Chapter S/Chapter 6 1 20 respectively of this thesis. Crude estimates of embodied land (ha) and CO2 (tonne) are also generated for the study year in Chapters 7 and 8 of this thesis. Other interface flows not involving Auckland Region include: • Other New Zealand Regional Economic Systems from/to the Rest of World Economic System. Financial estimates of these trade flows are generated in Chapter 5 of this thesis for the study year. Estimates of land (ha) and CO2 (tonnes) embodied in these trade flows are modelled for the study year using the interregional trade optimisation model developed in Chapter 8 and Appendix C of this thesis. • Other New Zealand Regional Economic Systems from/to the Rest of New Zealand Environmental System. This includes the use of raw material inputs, dumping of residuals and release of waste heat. A comprehensive biophysical snapshot of these flows (measured in tonnes) is generated for the study year using a PlOT in Chapter 6 of this thesis. • Rest of World Economic System from/to the Rest of World Environmental System. This includes the use of raw material inputs, dumping of residuals and release of waste heat. These flows are not modelled in this thesis. • Incoming solar radiation106 and outgoing radiation107• Solar energy is a critical factor in the functioning of many biogeochemical processes - without it life could not exist. Radiation is also lost to space through the reflection by c louds, dust and the Earth's surface, and also as degraded heat. Solar insolation estimates taken directly from scientific literature (Landcare Research, 2003 ; Patterson, 2005) are used to scale known biogeochemical fluxes at the global level to New Zealand and Auckland Region. A Note on the Physi E " c: ::J m ., E c: 0 ., ., u a ., � ., " (ij ., 0 c: ., .s::; u: ., 11> 'C ::> E in ., t:: E ., G; &. S ::J ::J 5 0 " 0 Lll 0 u .= 1: � 1 . . , B 1 .. , r 1 . . , f!. 1 .. , e 1 . . , 11 U S T X 0 Commodities 1 .. , B Use Household Other Final Exports Gross Commodity Consumption Demands Outputs · V 13 Industries 1 .. , r Supply Gross Industry · Outputs · · · · · R · Z · Value Added 1 . . , n W Value Added to ! Value Added to 1) 11> Value Added Household ! Other Final Value Added :; a. Consumption ! Demands E i!' ------------------------ -----------.-------- ------------------- -------------------�-------------------t------------------- ----------.-------- m E '" Y Q. Imports 1 .. , Cl> · · Imports · · · · · · · · · · 0' 13' ( � · · · Total Supply Gross Commodity Gross Industry Household Other Final · · · Inputs Inputs Consumption Demand · · · · Component Matrices and Vectors Commodity accounts are comprised of matrices U (B x r), S, T, X and vector a, An element UijE U represents, in basic prices, the value of commodity i used by industry} (i = 1 " . B;) = 1 . . . r) within a given flllanc ial year. Each column in matrix U shows the inputs used by industries classified according to the type of commodity used, while each row shows the inputs of each commodity according to the industries that use it. Matrix U is commonly referred to as the 'use', ' industry' or 'absorption' matrix. Element SijE S represents, in basic prices, the consumption of commodity i by household category } within a given flllancial year. This includes consumption of consumer durables and non-marketable governmental services1 l3 by households. 1 1 3 This includes the value of the goods and services provided by the producers of government services for consumption by the community e.g. benefits and pensions, primary and secondary school education, and public health care. A convention is adopted that the government itself is the consumer on behalf of the community. 1 24 Element fij E T denotes, in basic prices, the consumption of commodity i by other fmal demand category j within a given financial year. The following other final demand categories are covered by matrix T: gross fixed capital formation and changes in inventories. Element XijE X denotes, in free on board (fob) terms, the consumption of commodity i by export region j within a given financial year. Depending on their destination, exports are grouped as international (i .e. heading abroad) or interregional (i.e. heading to other New Zealand regions). Together matrices S, T and X represent a complete set of final demand categories. Element ai E (l gives the total value of commodity i output as supplied to all industries and final demand categories. Vector (l is referred to as 'gross commodity output' . It is calculated by summing the elements of commodity i in matrices V, S, T and X. Thus, r I:; E> 1\ ai = IUij + ISij + Itij + IXij ' \;j i,i = 1 . . . B. )=1 )=1 )=1 )=1 (4.7) Letting i denote an appropriately dimensioned column-summing vector, Equation 4.7 may be rewritten as, Vi + Si + Ti + Xi == (l. (4.8) The production relationships within the Auckland Region economy are captured in matrix V. An element, VijE V, represents, in basic prices, the output of commodity j produced by domestic industry i within a given fmancial year. 1 14 In this way, matrix V describes the sources of supply of products to the economy. Each row i shows the production of a particular industry classified according to the type of commodity produced, while each columnj shows the production of a commodity according to the industries that produced it. This matrix is commonly referred to as the 'supply' , 'commodity', 'production' or 'make' matrix. Element Yij E Y denotes, in cost, insurance and freight (cif) terms, imports of commodity j from region i. Depending on their origin, imports are grouped as international (i.e. from abroad) or interregional (i.e. from other New Zealand regions). 1 14 Taxes on products and margins are removed from the Table so that all entries may be expressed in basic prices. 1 25 By summing column j of matrices V and Y the total domestic commodity j output, aj, may be derived, while summing row i of matrix V produces the total domestic industry i output, Pi (know as 'gross industry output'). Hence, i 'V + i'Y = u' (4.9) and Vi = f}. (4. 1 0) Together Equations 4.8 and 4.9 fulfil the first key principle in balancing supply and use tables - the supply of a commodity, ai, must be equal to the use of that commodity, a'j, where i = j. The components of value added are recorded in matrix W. Value added components include compensation of employees, operating surplus, consumption of fixed capital, taxes on production and subsidies. The element WijE W denotes the value added by component i to the economy in producing column j's industry output within a given financial year. Summing all commodity inputs made to an industry, i'U, with the primary inputs made to that same industry, i 'w, derives an estimate of P' (known as 'gross industry input'), i 'U + i'W = W · (4. 1 1 ) This fulfils the second key principle in balancing supply and use matrices - that the total output of an industry, Pi, must be equal to its cost of production, P'j, where i = j. Equations 4.7 to 4 . l 1 form the key economic flow accounting identities of the Auckland Region commodity-by­ industry model. Definitions for matrices R and Z, and the remaining three vectors a, E and �, are however required to complete the economic flow model . Element rijE R represents the expenditure on value added component i by household consumption category j within a given fmancial year.1 l5 This includes non-market transfers such as benefits and pensions. E lement ZijE Z denotes the expenditure on value added component i by final demand category j within a given financial year. Matrix Z is a sparsely populated matrix consisting of commodity and non-commodity indirect taxes on products sold directly to capital formation or stored in stocks. Closely associated with matrices R and Z are vectors a, E and �. l iS This includes commodities imported by wholesalers/retailers who add a margin and then on-sell to households. 1 26 Element cri E 0 represents the total value of value added component i supplied to all industries and final demand categories, thus, Wi + Ri + Zi = o. (4. 1 2) Element cj E E shows the total expenditure on commodities and value added components by household category j, hence, i 'S + i 'R = E. (4. 1 3) S imilarly, element t; E � gives the total expenditure on commodities of value added components by other final demand category j, thus, i 'T + i 'Z = �. (4. 1 4) Gross Domestic Product and Expenditure An accounting identity equating Gross Domestic Product (GDP) with Gross Domestic Expenditure (GDE) may also be formed from Table 4. 1 . GDP may be derived by summing the elements of matrices W, R and Z1 l6, thus, To derive GDE we must first augment several matrices together. Using I to denote the horizontal augmentation operator, household consumption matrix, S, other [mal demands matrix, T, exports matrix, X, and a transposed and negated imports matrix, -Y' , are augmented together to form a new matrix, T . Hence, T = S I T I X I -Y' . (4. 1 6) Element [ij E T denotes the value of commodity i sold to final demand category j. In turn, GDE may be derived by summing the elements of matrices T , R and Z, thus, n r 1 1 6 Double summations, such as L L W ij , are summarised here a s Lij W ij . i=1 j=i 1 27 (4 . 1 7) 4.4.2.2 Physical Input-Output Model The Auckland Region economy is inherently dependent on natural resources from the environment and, in turn, on the assimilative capacity of the environment to absorb unwanted by-products. One approach that may be used to capture these dependencies is to measure these flows in physical terms - this, however, is not a trivial task. The United Nations Statistical Commission (2002, pJ-5) has noted that "a complete implementation of [physical] accounts is very ambitious". Physical accounting of the flows identified in Figures 4.8, 4.9 and 4. 1 0 would, at least, require ( 1 ) comprehensive accounting, and (2) compliance with fundamental physical laws (i.e. the laws of thermodynamics). It would also be advantageous, from a replication perspective, if the framework made the most of existing accounting practices, definitions, classifications, standards and so on. Additionally, compatibility with the commodity-by­ industry financial flow model already developed would permit, to varying degrees, comparisons to be made between physical and fmancial flow matrices. Developing a physical accounting framework aligned to Figures 4.8, 4.9 and 4. 1 0 would, however, not be without limitations. It would be difficult, say, for such a framework to capture the key functional relationships of system fluxes characterised by feedback loops, time lags, non-linearities and the like. It is also likely that physical accounting would provide little insight into the qualitative attributes or risk profiles of the flows being measured.ll7 Additionally, no assessment is made of 'ecosystem services' which are not only essential for the continuation of life, but also for the enjoyment of life. l l8 Nevertheless, despite these limitations, a commodity­ by-industry physical flow model is developed here to analyse Auckland Region's environment­ economy system. Structure of the Physical Flow Model Physical fluxes within the economy, across the environment-economy interface and vice versa, are captured by the commodity-by-industry physical flow matrix depicted in Table 4.2. This matrix provides a detailed description of the physical flows within the economy (matrices U , S , T , V , X and Y ), from the environment to either industry for production (matrices :K 1 1 7 Steurer ( 1 996), for example, argues that the toxicity of a physical material is often negatively correlated with its mass. 1 18 The importance of ecosystem services to Auckland Region's environment-economy system is investigated further in Chapter 7 and Appendix J. 1 28 and E ) or households/capital for final consumption (matrices H , G and D ), and ultimately, - - from industry or fmal consumption back to the environment (matrices Q , P , 0 and L ). Furthermore, physical flows may also be exchanged with other regions and nations (matrices X , Y , if and C and vector � ) or accumulated by households, 7i , in capital, p , or within the local environment, B and \jI . Algebraically, the matrix may therefore be described by seventeen sub-matrices and thirteen vectors. The � situated above a matrix or vector is used to indicate measurement in physical units (e.g. tonnes). Table 4.2 Commodity-by-Industry Physical Flow Matrix � " E � E E 0 0 � .� "ii � , ... B ... r Commodities ... B U Use Industries . . . r V Supply Household . . . tJ. Consumption .§ I ______________________________ .. __________________ _ � � I Capital o � � I------------------------------.. ------------------- Imports ••• <1> Raw Materials Inputs (Natural Resources . . , '" and Ecosystem Inputs) Residual Inputs " (} Gross Supply y Imports K Raw Materials to Industry E Residuals to Industry Q' I P' Commodity Inputs Industry Inputs c 0 � � � � 0 0 I 0 , . . . tJ. 5 Household Consumption Final Consumption i 0 , ... � T Capl'.1 Consumption ! ! ! . {! ! .. , 11 i Exports -------------------r-------------------r·------------------- : I I i ! I -------------------f-------------------t-------------------: I i I ! I i I : I ! I - I _ H I G Raw Materials to ! Raw Materials to Household I Capital Consumption ! Consumption : .' Household Consumption Inpu1s (} Residuals to CapH.1 Consumption �' Capital Consumption Inputs Raw Materials Exported C Reslduals Exported Exports .. '5 .. % � S o � E ! � � Q. '5 0 !! a:: � ; � � -;;- � � � � l .� er . . , '" , . . , (} Q P Raw Materials Reslduals Generated by Generated by Industry Industry 0 Residuals Generated by Household Consumption --.----------------- L Reslduals Generated by Capital Consumption ------------------- Biogeochemic.' Cycling Model (included for conceptual comp$eteness, but not Included in the Chapter 6 physical Input-output modeQ A' I �' Material Outputs Residual Outputs 1 1 1 1 g � ID .. � " 10 ::> Tf Net AccumulaUon by Househokj Consumption ----------------.-- p Net Accumulation by Capital Consumption ------------------ Net Accumulation by Imports a Net Accumulation of Raw Materials • Net Accumulation of Reslduals 1 1 � :::> .. e Cl 1 1 a Commodity Outputs I l lndust: Outputs i Household Consumption Ou1pu1S ------------------- q Capital Consumption Outputs ------------------- i Imports A Material Inputs � Residual Inputs 1 30 Component Matrices and Vectors Matrices U , S , T , X , V and Y are physical equivalents (measured in tonnes) of their fmancial counterparts described in Table 4. 1 . An element uij E U therefore represents the physical quantity of commodity i used by industry j (i = 1 . . . B; j = 1 . . . r) within a given financial year. Each column in matrix U shows the physical inputs used by industries classified according to the type of commodity used, while each row shows the inputs of each commodity according to the industries that use it. Apart from the inclusion of matrices catering for raw material inputs/outputs and residual inputs/outputs, Table 4.2 incorporates two subtle, but important changes from Table 4. 1 : ( 1 ) the value added matrices W, R and Z are omitted as they have no associated physical flows, and (2) household consumption and capital formation are mirrored on both axes. The environmental input-output matrices in the bottom left and top right hand corners may be read in an analogous manner to the following example. Row entries in the household consumption row matrix, 0 , describe the origin of residual outputs from household consumption (i.e. household consumption category i is the origin of residual j), while column entries in the household consumption column matrix, H , denote the destination of raw materials to household consumption (i.e. household consumption category j is the destination of raw material i). In other words, the environmental input matrices (bottom left) describe the inputs of raw materials and residuals used by household consumption, capital formation and exports, while the environmental output matrices (top right) describe the raw material and residual outputs generated by industries, household consumption, capital formation and imports. � � � The matrix for manufactured commodities comprise sub-matrices U , S , T and X and vector a . 1 19 In a similar manner to Equation 4.8, the vector a may be defmed as, Vi + Si + Ti + Xi == a , (4. 1 8) while the production relationships are captured in physical terms in the regional supply, V , and imports, Y , matrices. By summing the columns of V and Y , the gross physical mass of commodity j output, a') , may be derived, thus, 1 1 9 To establish mass balance identities between the economic and environmental flow matrices requires that all transactions be expressed in mass terms. For all earthly purposes mass and weight may be considered to be equivalent. 131 (4.19) Together Equations 4.18 and 4.19 formulate the first key principle in balancing the physical supply and use matrices - the physical supply of a commodity, aj , must be equal to the use of that commodity, a'} , where i = j. � � Summing row i of matrices V , Q and P defines the gross physical mass of domestic industry i output, fJi ' thus, Vi + Qi + Pi = P . (4.20) An element 'ifij E Q represents the physical output of raw material j generated by industry i within a given fmancial year. This is a sparsely populated matrix made up of livestock, trees and some plant species. Similarly, element P ij E P denotes the physical output of residual j generated by industry i within a given financial year. Unl ike matrix Q , matrix P is a densely populated matrix capturing the solid wastes, air emissions and water pollutants generated by industry. Industries, like commodities, must be in materials balance. Thus, i' U + i' K + i ' E == pt , (4.21) where an element kij E K denotes raw material input i into industry j. Matrix K includes natural resource inputs and ecosystem inputs. Similarly, element eij E E represents residual input i into industry j. Matrix E is a sparsely populated matrix including economic residuals recycled from nature by industry. Taken together, Equations 4.20 and 4.21 fulfil the second key principle of materials balance in the commodity-by-industry physical flow matrix - gross physical industry output, fJi ' must equal gross physical industry input, fJ'} , where i = j. Equations 4.18 to 4.21 form the key physical flow identities associated with economic production. 1 32 The remaining matrices of Table 4.2 focus on fluxes between the environment and fmal consumption (i.e. the use of raw materials and residuals by final consumption and as denoted by ,....,., ,....,., ,.....,., ......, matrices H , G , F , D and C ), conversely, fluxes between final demand and the environment (i.e. the generation of raw materials and residuals by final consumption as denoted by matrices 6 and L ), and various materials balance vectors (1t , p , � , (i and (j) ). Element hi) E H denotes the raw material input i into household consumption category j. This includes, for example, household extraction of water for domestic supply, oxygen for combustion in private motor vehicles, and the private harvesting of vegetables and fruits and so on. Summing the columns of matrices S and H gives gross physical household consumption of inputs, E' , (4.22) while summing the rows of matrix 0 , household residuals produced, and vector 1t , net accumulation of household durables, a mass balance may be established for household consumption, E , ( 4.23) where element oi) E 6 denotes residual} generated by household consumption category i, and element ifi E 1t represents net accumulation of durables by household consumption category i. Matrix 0 includes residuals, pollutants and the like discharged directly by households into the environment, while vector 1t comprises the accumulation of consumer durables (manufactured commodities) by households. Similarly, a mass balance may also be defined for capital formation by summing the columns of � � � matrices T , G and D , (4.24) 1 33 where element g ij E G denotes raw material i used in forming capital category j, element dij E D represents residual input i used in forming capital category j, and if'} E 11' gives gross physical inputs into capital formation. Matrix D consists of, for example, recycled or reused solid waste from controlled landfills. Accordingly, summing the rows of matrix L , residuals generated in forming capital, and vector p, net accumulation of manufactured capital, establishes a mass balance for capital formation, (4.25) with '4 E L denoting residual j produced by capital formation category i, and Pi E P the net accumulation of manufactured capital category i. Empirically, vectors 7i' and p' may be algebraically formulated; for example, i' is found by summing the rows of matrix 0 , transposing the result, and subtracting this from vector E' , thus, -, -, 0- ·' 1t = E - I (4.26) Over and above the economic trade physical flows, captured in matrices X and Y , raw materials and residuals (e.g. recycled commodities) may also be exchanged with other regions and nations. While non-economic flows may be captured in detail by separating domestic flows from those originating from, or destined to, other regions and nations 120, the approach taken here is simply to attribute these flows to matrices F and C . 1 2 1 Element �} E F denotes the export of raw material i to export region j, and similarly, element cij E C represents the export of residual i to export region j. Matrix F is a sparsely populated matrix capturing, for example, fish resources harvested by foreign regions and nations. Matrix C includes oxygen and water taken by ocean-going vessels from other regions and nations. 120 For example, matrix K could be separated into two sub-matrices, the first representing the utilisation of domestic raw materials, and the second traded raw materials. 121 The implications of tourist activity are particularly difficult to demarcate. Carbon dioxide emissions from international air travel, for example, could be coded in several ways - each resulting in inclusion in a different matrix in Table 4.2 e.g. to the region of origin, to the region of destination, or on a pro-rata basis weighted by distance within a region's airspace and so on. 1 34 The gross export of physical mass, � , may be defmed by summing the columns of matrices X , F and C , thus, � � i' X + i' F + i'C == � ' , ( 4.27) and a physical balance of trade by subtracting a transposed row summation of matrix Y , hence, �, r · ' Y� � = ., - I . (4.28) If � is positive then the region has a net physical trade surplus, else the region has a net physical trade deficit. Hence, the vector () may be determined as the sum of the row entries in matrix Y and vector � , therefore (4.29) where Bi E () denotes gross physical imports by import category i. The final materials balance equations that may be established in Table 4.2 relate to the accumulation of raw materials and residuals within the environment. Summing the column entries of matrix Q gives A' , and subtracting a transposed summation of matrices K , H , G and F provides an estimate of (J , the net accumulation of raw materials within the environment, (J = i'-(:[(i + Hi + Gi + Fi)' . (4.30) Similarly, summing the column entries of matrices P , 0 and L gives ji' and, in turn, subtracting a transposed summation of matrices E , D and C gives (j) , the net accumulation of residuals within the environment, (4.3 1 ) 1 3 5 Finally, the conversion of residuals outputs to raw material inputs, irrespective of whether they are produced within the economy or environment, takes place through various biogeochemical cycles. Phosphorus leached from soils may, via runoff and river transportation, fmd its way into the sea. Over time it is possible, through the processes of sedimentation, sedimentary rock formation and tectonic uplift, that the phosphorus may once again exist within soil - thereby completing the cycle. These transformations are encapsulated in the block labelled 'Biogeochemical Cycling Model' . Since they occur over time periods both much shorter, or typically much longer, than a fmancial year and also tend to be dynamic and non-linear rather than static, their inclusion in the accounting framework is considered inappropriate. Instead, they are modelled separately in Chapter 7 and Appendix B . A Note on Spatial Location If the spatial location of the raw materials utilised (or generated) by the economy, or alternatively, the residuals generated (or utilised) by the economy, are to be captured in the framework, then Table 4.2 requires modification. Since environmental problems are often spatially explicit, particularly air and water pollution, the identification of, say, ecologically sensitive areas, water catchments, densely populated localities or the like within Auckland Region would undoubtedly improve the framework's usefulness. Spatial disaggregation of the matrices in Table 4.2 may be achieved either by assigning each raw material/residual type a spatial location, or by assigning each spatial location a complete set of raw materiaVresidual types. As both approaches are simply rearrangements of equivalent information, the selection of one approach over the other is only a matter of presentation. 4.4.3 Conversion to an Inter-Industry Framework The commodity-by-industry accounting framework described in Section 4.4.2. 1 has the advantage of being able to account for multiple outputs per industry, rather than assuming there is one homogeneous output. However, for the purposes of multiplier analysis (refer to Chapters 5, 6, 7 and 8, and Appendix G), using commodity-by-industry models are often problematic as: • Analytical ease of use. Commodity-by-industry models are often rectangular matrices which means that straightforward matrix algebra (based on square matrices) cannot be used; and • Negative coefficients. Commodity-by-industry models usually have multiple outputs per industry, which inevitably generates negative coefficients. As others (e.g. Almon, 136 2000) have pointed out these negative coefficients are problematic as they make no economic sense. In order, therefore, that the derivation of economic multipliers (Chapter 5, Appendix G), ecological multipliers (Chapter 6, Chapter 7) and ecological footprints (Chapter 8) could be successfully undertaken, the procedure described below which converts a commodity-by­ industry economic model (outlined in Section 4.4.2. 1 ) into an industry-by-industry economic model was used. 4.4.3. 1 Commodity Technology and Industry Technology Assumptions Under ideal circumstances an inter-industry matrix would be derived from source data describing the input structure of every type of activity producing a single commodity. This ensures homogeneity except in those cases where secondary products (i.e. by-products and joint products) are intrinsic to the production process, e.g. it is physically impossible to separate meat (main product) and offal (by-product) production. Most statistical agencies, however, can only derive inter-industry matrices according to enterprise definitions which, in turn, may be complicated with many secondary products. To minimise the bias resulting from the presence of jointlby-products, statistical agencies carefully craft business census questionnaires to utilise itemised cost accounting definitions which aid in separating secondary products. Alternatively, input-output compilers may utilise, in combination with census questionnaires, approaches such as the 'redefinition method', 'negative transfer method', 'aggregation' or ' positive transfer method' to deal with secondary products (United Nations, 1 999). Data availability, confidentiality, cost/time constraints and so on all prohibit the use of such procedures in this thesis. Crude inter-industry matrices may however be generated mechanically by assuming one of two possible production pathways : commodity-based or industry-based technology. 122 Under the commodity technology assumption industries produce commodities in fixed proportions (Jackson, 1 998; United Nations, 1 999). A given commodity therefore has the same input structure irrespective of the industry that produces it. By contrast, under the industry technology assumption, inputs are consumed in the same fixed proportions independently of the commodity being produced by an industry (Jackson, 1 998; United Nations, 1 999). Primary and secondary products are therefore assumed to be produced using the same technology. 122 Recently Konijn and Steenge ( 1 995) have proposed, and convincingly argued for, the use of an activity-based technology assumption. l37 Various input-output analysts have critiqued each assumption's pros and cons - for example, refer to ten Raa et al. ( 1984), Kop Jansen and ten Raa ( 1990), Konijn and Steenge ( 1 995) and the United Nations ( 1 999). Kop Jansen and ten Raa ( 1 990, p.2 1 4) argue that selection of one assumption over another is simply a "matter of judgement or taste". The United Nations ( 1 999) provides strong theoretical justifications for the adoption of a commodity technology assumption123, but acknowledges that due to the presence of jointlby-products negative coefficients may be produced in multiplier calculations124, or when key balance identities are not adhered to. Given that the industry technology assumption guarantees positive inter­ industry coefficients and has been widely utilised125, it is selected here as the approach for generating inter-industry matrices. 4.4.3.2 Generating an Inter-Industry Matrix Using the Industry Technology Assumption Under an industry technology assumption, the relationship between commodity input and industry output, in matrix V, provides the basis for a transformation from commodity to industry space.1 26 Such a transformation, and the subsequent generation of an inter-industry matrix, requires several steps. Step I Estimate Gross Commodity Outputs Required/or Domestic Purposes We begin by estimating gross commodity outputs required for domestic purposes, a , as a=a-(i'Y) ' , (4.32) where i denotes an appropriately dimensioned column-summing vector. 123 The United Nations ( 1 999) argues that the industry technology assumption is "economically nonsensical" (p.99) as commodities are produced with the same input structures irrespective of the industry they are produced in. If two commodities have the same costs, because they are produced by the same technology (Le. in a perfectly competitive market where prices equal costs), it then follows that they cannot have different prices. But this is what is required by the industry technology assumption. Furthermore, this assumption requires that market shares remain constant over time - which is also implausible. 1 24 The United Nations ( 1 999) puts forward a suite of ad hoc methods for dealing with negative coefficients. Almon (2000) has also proposed a methodology for generating non-negative multipl iers under the commodity technology assumption. 125 The United Nations ( 1 968) SNA handbook advocated the adoption of an industry technology assumption in constructing inter-industry accounts. 126 By corollary, the inverse ofF may be used to transform a matrix from industry to commodity space. 1 3 8 Step 2 Calculate Domestic Industry Production o/Commodities/or Export This requires construction of a matrix of commodity-by-industry direct requirements coefficients, B, by defming a fixed relationship between commodity input and industry output values. Thus, B = V &-1 , (4.33) followed by estimation of a column vector, Xi, of total exports, y = Xi . (4.34) Diagonalising y and then premultiplying it by B yields domestic production of commodities for export, (4.35) Step 3 Calculate Total Supply for Domestic Use Subtracting M from V gives the domestic supply by industry, and augmenting this with a row for imports gives total supply for domestic use, N = (V - M J . i' Y (4.36) The values of matrix N represent the commodities, domestically produced and imported, that are required to support intermediate demand and fmal domestic demand. Standardising N yields a matrix, N , that shows the composition of industry and import sources of commodity production for domestic demand with columns summing to one. 127 1 27 A . . . is p laced above matrix N to indicate that it has been standardised. Mathematically, --1 standardisation is achieved by the fol lowing formula: N(i'N) . 139 Step 4 Transform U I T from Commodity to Industry Space Postmultiplying N by U I T completes the transformation to industry space. Matrices W, R and Z may then be inserted - no mathematical manipulation of these matrices is required as they are already in the required form. In this way, the entire inter-industry framework may be generated, (4.37) Chapter Five Economic Input-Output Model: Financial Flows in the Auckland Region Economy 1 4 1 In Chapter 5 an economic input-output model of the Auckland Region economy i s constructed, using the commodity-by-industry framework (Table 4. 1 ) mathematically defined in Chapter 4. The reasons for constructing this economic input-output model of the Auckland Region economy are several: • Structural relationships. To understand the issues concerning the sustainability and growth of the Auckland Region, it is essential to appreciate the structural relationships in the Auckland Region economy, as the economy is arguably the 'engine room' and 'driver' of human activity in the Region. These structural relationships are conventionally analysed by using an economic input-output model such as the one developed in this Chapter; • Basis for physical input-output model. The economic input-output model is a starting point for constructing the physical input-output model which is in turn required to gain an appreciation of the biophysical (urban metabolism) aspects of the sustainabi lity of the Auckland Region; and • Basis for ecological footprinting. In this thesis, the economic input-output model, along with a multi-regional economic input-output model, is the core analytical tool used to calculate the ecological footprint of the Auckland Region - i .e . how, in structural terms, does Auckland Region ecologically depend on other regions and, for that matter, other countries? • Basis for dynamic modelling. The static economic input-output model developed in this Chapter is the starting point for constructing the dynamic input-output model in Chapter 1 1 of this thesis; A commodity-by-industry economic input-output model is constructed for Auckland Region as the preferred framework, as it allows for multiple outputs per industry, thereby providing for a more realistic representation of the economy. This is particularly important in the construction of the physical input-output model, where the aggregation of all outputs into one homogeneous industry output becomes particularly troublesome. The regionalisation of commodity-by­ industry models has, however, only been described in the l iterature only recently (Jackson, 1 998 and Lahr, 200 1 ) with no reported actual operationalisation of such methods. This thesis 1 42 proposes and then operationalises such a cornrnodity-by-industry regionalisation method so that an Auckland Region model can be constructed from the New Zealand model, drawing on and extending the methods of Jackson ( 1 998) and Lahr (200 1) . 5.1 Generation of the Auckland Region Economic Input-Output Model 5.1 .1 Previous Regional-Level Economic Input-Output Models Since the 1970s input-output analysis has been the method of choice for analysing regional economic activity. Consequently, there has been a great deal of interest in methods for constructing regional economic input-output models (for reviews, see Round ( 1 983), Miller and Blair ( 1 985), Hewings and Jensen ( 1 986) and Jensen ( 1 990)). Most of this work has focused on methods for deriving regional inter-industry models from national equivalents. 128 However, this traditional focus on industries has been questioned on theoretical grounds (Madsen and Jensen­ Butler, 1 998, 1 999) as neglecting the key attributes of the regional cornrnodity-by-industry (functional) accounting framework. Apart from the analytical advantages of the commodity-by­ industry model, it is a more reasonable form for data collection and handles secondary or joint products more effectively (St Louis, 1989). It is not only capable of answering the same questions as the inter-industry or Leontief format, albeit with mathematical transformation, but also others that the latter cannot. A researcher may, for example, construct commodity-by­ commodity, commodity-by-industry, industry-by-commodity or inter-industry models depending on the research interest. Furthermore, as Oosterhaven ( 1 984) points out, it is particularly useful for interregional or multi-regional models where trade flows are typically in terms of commodities rather than industries. The statistical agencies of several nations, including the Netherlands, Finland, Denmark and Canada, have created regional level commodity-by-industry accounting frameworks using surveys e.g. Oosterhaven ( 1 984), Boomsma and Oosterhaven ( 1 992), S iddiqi and Salem ( 1 995), Eding and Oosterhaven ( 1996), Eding et al. ( 1 998), Piispala ( 1 998, 2000) and Madsen and Jensen-Butler ( 1 998, 1 999). Between late 2002 and mid 2003 Statistics New Zealand (SNZ) investigated the possibility of developing survey-based regional input-output models in New Zealand. This feasibility study assessed user requirements, reviewed existing New Zealand and 1 28 lackson ( 1 998) attributes this focus to several factors including inertia, industry-based statistical reporting by government agencies, and because economic impact assessment, arguably the most common application of input-output analysis, requires only inter-industry matrices. Regional inter-industry tables have been developed in New Zealand by Hubbard and Brown ( 1 98 1 ) , Moore ( 1 9 8 1 ), Butcher ( 1 985), Kerr, Sharp and Gough ( 1 986) and McDonald ( 1 997, 1 999a). 1 43 international methodologies 129, evaluated data sources and provided recommendation for a development plan (Statistics New Zealand, 2002, 2003a, 2003b, 2003c). It was concluded that if official regional input-output tables were to be developed that these would adopt the national commodity-by-industry framework, however, "limited data availabi lity restricts any official development" (Statistics New Zealand, 2003a, p.8). When time and cost constraints prohibit the development of survey-based regional tables, as is the case here, input-output practitioners have typically turned to 'non-survey' methodologies that mechanically reduce national coefficients to regional equivalents. Czamanski and Malizia ( 1 969), Schaffer and Chu ( 1969, 1 97 1 ) and Smith and Morrison ( 1 974), compared survey-based industry-by-industry matrices and synthetic equivalents produced by the most commonly used non-survey techniques (refer to Appendix D for a review of these methods). Although non­ survey coefficients were consistently larger than their survey counterparts, the Simple Location Quotient (SLQ) method and RAS constrained matrix technique generated reasonable estimates (refer to Appendix D for further explanation). Despite the considerable research effort directed at industry-by-industry matrix regionalisation using non-survey methodologies, literature on the regionalisation of commodity-by-industry matrices using non-survey methods remains scarce ­ notable exceptions include lackson ( 1 998) and Lahr (200 1 ). Statistics New Zealand's (2003a, p.8) feasibility study concluded that the development of regional input-output tables "would begin with a simple non-survey-based methodology and move toward more complex survey­ based methods over time". 5. 1.2 Methodological Process Used in the Auckland Region Study In Sections 5 . 1 .3 and 5 . 1 .4, a non-survey method is developed that generates a regional commodity-by-industry model for Auckland Region through a series of 1 1 mechanical steps using a modicum of Auckland Region-specific data (Figure 5 . 1 ). Section 5 . 1 .3 (steps 1 to 4) updates the national use and supply matrices from 1 996 to 1 998 for volume, price and productivity changes, while Section 5 . 1 .4 (steps 5 to 1 1 ) regionalises the updated New Zealand commodity-by-industry model using the SLQ technique. This technique has successfully been applied to inter-industry regionalisation in New Zealand by analysts such as Hubbard and Brown ( 1 98 1 ), Butcher ( 1985), Kerr et al. ( 1 986), McDonald and Patterson ( 1994, 1 995a, 1 995b, 1 995c), Patters on and McDonald ( 1996), Ministry of Agriculture and Forestry ( 1 997) and McDonald ( 1 999a). A key feature of the regionalisation procedure is that it provides significant opportunities for the insertion of superior data. Figure 5 . 1 provides an overview of 129 This included a review of the three most widely utilised regionalisation methodologies within New Zealand, including one developed by the author. 1 44 the methodological steps undertaken in the generation of the Auckland Region Economic Input­ Output Model (commodity-by-industry). Update of the National Input-Output Model (Commodity-by­ Industry) 1 996-1 998 Step 1 : Pre-update Aggregation Step 2: Volume, Price and Productivity Update Step 3: Primary Inputs and Final Demands Update Step 4: Matrix Balancing ---��� Section 5. 1 .3 Regionalisation of the NZ Input-Output Model (Commodity-by-___ �� Section 5. 1 .4 Industry) to Generate an Auckland Model Step 5: Estimation of Regional Output Step 6: Derivation of the Regional Supply Matrix Step 7: Estimation of Regional Value Added Step 8: Derivation of the Regional Use Matrix Step 9: Estimation of the Regional Final Demand Step 1 0: Estimation of Regional I nflow and Outflow Figure 5.1 Step 1 1 : Final Adjustments Methodological Process for Generating an Auckland Region Economic (Commodity-by-Industry) Input-Output Model The fol lowing statistical data sources were employed in the construction of the commodity-by­ industry model for the Auckland Region economy: • 1 996 Inter-industry Study of the New Zealand Economy (Statistics New Zealand, 200 1g). 1 45 • 1 996, 1 998 and 200 1 full time equivalent (FTE) employment at the five-digit level as extracted from the Business Directory (Statistics New Zealand, 1 996b, 1 998c, 200 1 b) • 1 998 Labour Force Survey FTE Employment (Statistics New Zealand, 1 998g) • 1 996 and 2001 Usually Resident Population estimates as extracted respectively from the 1 996 and 2001 Censuses of Population and Dwellings (Statistics New Zealand, 1996c, 200 1 c) • 1 998 Sub-National Usually Resident Population estimates as extracted from Hot off the Press (Statistics New Zealand, 1 998k) • 1 996, 1 998 and 200 1 Labour Cost Index (Statistics New Zealand, 1 996f, 1 998f, 200 1f) • 1 996, 1 998 and 200 1 Producer's Price Index Outputs as extracted from Key Statistics (Statistics New Zealand, 1 996f; 1 998f, 200 1f) • 1 996, 1 998 and 2001 GDP Series for 25 industries from INFOS (Statistics New Zealand, 1 996e; 1998d, 200 1 d) • 1 998 and 2001 International trade statistics as extracted from the New Zealand Harmonised System Classification 1 996 (Statistics New Zealand, 1 998e, 200 1 e) • 1996 and 1 999 Agricultural Statistics (Statistics New Zealand, 1 998a, 200 1 a) 5.1.3 Update of the New Zealand Input-Output Model In this Section the national 1 996 inter-industry study of the New Zealand economy (Statistics New Zealand, 200 1 g) is updated to the year ending 3 1 March 1 998. The national supply and use tables are updated for volume, price and productivity changes. Superior data for the 1 998 year covering the primary input and final demand categories are also inserted. Step 1 Pre-update Aggregation In mid 200 1 , SNZ's National Accounts division produced an interim release of the 1 996 Inter­ industry Study of the New Zealand Economy (Statistics New Zealand, 200 1g). This study covered 1 26 economic industries and 2 1 0 economic commodities, along with standard primary inputs and final demands. Industry definitions were based on the Australia New Zealand Standard Industrial Classification (ANZSIC), while commodity definitions were based on the Australia New Zealand Commodity Classification (ANZCC). The base tables were aggregated from 1 26 industries to 1 23 industries to ensure that each industry was built up from a unique combination of ANZSIC definitions which, in turn, permitted matching with other statistical datasets (e.g. Business Directory, Capital Accounts, Household Economics Survey, and Harmonised System) and pre-existing environmental accounts. The commodity definitions required no modifications. 1 46 Step 2 Volume, Price and Productivity Update Input-output studies, including those produced by SNZ, are invariably out of date due to the significant time needed for processing surveys. Consequently, updating is usually required for at least volume, price and productivity changes (Jensen et aI. , 1 979; Butcher, 1 985 ; Kerr et al. , 1 986; J ensen and West, 1 988). Given a base year of 1 998 for the static analysis, it was necessary to update tables from 1996 to 1998. The key step in updating the national tables for volume, price and productivity changes is a s imple transformation mapping the national gross output vector from Jl 1996 to Jl 1998. Mathematically, the update can be expressed as : ,...----- 1996 ,...---.. 1996 �1996 emp; * 0 , gdp; * 0 and ppi; * 0 , ,.....--... (5 . 1 ) where emp ( r x 1 ) represents FTE employment, gdp ( r x 1 ) represents GDP ( in constant � prices), and ppi (r x 1 ) the Producer's Price Index for Outputs for industry i (the overbar is used to denote a single mathematical term). The superscripts differentiate the 1 996 and 1 998 financial years. The first two terms on the right hand side of Equation 5 . 1 update gross output for volume changes, assuming that changes in FTE employment are a reasonable proxy of actual volume change. The third term is a simple productivity estimate based on the ratio of GDP contribution per FTE in 1 996 to its equivalent in 1 998. The fmal term is a price inflator which converts the results of the first three terms from 1 996 into 1 998 dollars. Step 3 Primary Inputs and Final Demands Update A common theme in the input-output regionalisation l iterature is that superior data must be inserted wherever possible (Jensen and West, 1 988; Jackson, 1 998; Lahr, 200 1 ). Updating the primary inputs, W, final demand, S and T, and trade, X and Y, components of the accounting framework provided an opportunity for the insertion of superior data. Beginning with trade, X and Y, Harmonised System commodity imports and exports for 1 998 were directly matched to 1 47 the 1 23 industry and 2 1 0 commodity definitions of the accounting framework usmg concordances provided by SNZ's (B. Voice, pers. comm., 2 1 Nov 200 1 ). The 'taxes on products' primary input essentially represents import duties and was updated by assuming that the ratio between imports and duties in each industry remained constant between 1 996 and 1 998. The updating of the primary inputs matrix, W, was more cumbersome. Although SNZ regularly produces estimates of primary inputs in its Key Statistics publication, these are only released at a highly aggregated industry level. It was therefore not possible to use these data directly in the updating process: however it was used to cross-check results, and in some cases to make ad hoc adjustments. 130 Given that wages and salaries represented approximately 40 percent of total primary inputs across all industries, and are a critical determinant of induced (consumer­ spending) effects in Type IT multipliers, it was considered essential that this category be updated. Specifically, wages and salaries, Wws ' in each industry j, were updated in the following manner, 1998 �1998 , WwS,j Id j ' �1996 "'---.1996 [ 1996 ]( ,-..1998 J Wws,j = emp j �1 996 -----:1 996 , emp) :;t: 0 and lel j :;t: 0 , empj , ICI) , (5.2) where lci (r x 1) represents SNZ's Labour Cost Index. Each remaining primary input, Wopi (e.g. operating surplus, consumption of fixed capital, other taxes on production and subsidies), in each industry j, was updated by assuming that only relative change had occurred between 1 996 and 1 998, [ ",---,,1998 ] [ �1998 ] 1998 1996 gdp) , ppij ' ,.---..1996 �1996 Wopi,j=Wopi,j ,..---..1996 �1996 , gdp) :;t: 0 and ppI j :;t: 0 . gdp) , PPlj , (5.3) Unlike the primary input by industry data produced by SNZ, no equivalent dataset is produced for final demand. Fortunately, SNZ does however regularly produce total estimates for the final demand categories. This permitted only a simple pro-rata scaling of each final demand category - the approach used in this thesis. In the context of estimating regional final demands, Stevens et al. ( 1983), Treyz and Stevens ( 1 985) and Treyz and Petraglia (2001 ) all discuss an alternative econometric approach which could be adopted to determine each component of final 1 30 Despite the highly aggregated nature of this data, a few industries had one-to-one definitional matches with the 1 23 industries, enabling direct insertion ofthe superior data. 1 48 demand at the national level . This method relies on estimates of disposable income, intermediate activity and other factors. Step 4 Matrix Balancing Once volume and price updates for gross output totals, primary inputs, final demand and trade totals had been established, the national use and supply tables were balanced using the bi- proportional R A S technique (refer to Appendix D for a worked example). It was assumed that the supply matrix coefficients remained the same between 1 996 and 1 998, allowing the use matrix to be balanced using the R A S technique. Several coefficients in the national use and � 13 \ supply tables were known accurately and were thus left out of the RA S update. 5.1.4 Regionalisation of the New Zealand Model to Generate an Auckland Region Model Lahr (200 1) asserts that the regional economics literature, with the exception of Jackson ( 1 998), is "devoid of any blow-by-blow narrative of the development of regional accounts in a commodity-by-industry setting". Perhaps the most significant l imiting factor in the development of commodity-by-industry matrices has been the paucity of specific regional data. This is certainly the true for Auckland Region where only a modicum of regional statistical data is available. It is therefore not surprising that the methodologies developed to date (e.g. Jackson ( 1 998) and Lahr (200 1 » have tended to be pragmatic. Similarly, in this study, emphasis is placed on applications of the resulting matrices rather than on the creation of data itself. Several guiding principles have emerged for the regionalisation of industry-by-industry models. These principles are equally applicable to the regionalisation of commodity-by-industry models : ( 1 ) use as much sectoral detail as is available to minimise the impact of aggregation bias132 (Sawyer and Miller, 1 983; Stevens and Lahr, 1 993; Lahr, 200 1) ; (2) assume technology is spatially invariant within a nation thereby enabling direct application of raw use coefficients, albeit with some adjustments (Isard, 1 95 1 ; Lahr, 200 1 ); and (3) regionalisation should be performed on domesticated national accounts133 (Sawyer and Miller, 1 983; Jackson, 1 998). 13 1 Updating after removing known or troublesome coefficients is not a new approach; it has been used at least since Paelinck and Waelbroeck ( 1 963), who found the approach produced significantly better tables than the unadjusted R A S. 132 Aggregation bias manifests itself through improper specification of regional industry mix when aggregating. 133 Most regionalisation methodologies are unable to account for international trade (Lahr, 200 1 ). Thus, regionalisation is only undertaken on domestic coefficients and international trade must be estimated independently. 149 These principles are applied in the generation of the Auckland Region commodity-by-industry model. Step 5 Estimation of Regional Output The core task in any input-output regionalisation methodology is estimating a region' s ability to supply its own requirements. Ultimately this information contributes to the estimation of regional imports and exports. Once coefficients have been estimated, regional commodity and industry output must be derived in order to generate commodity/industry balance equations. As with many national statistical agencies, SNZ does not produce sub-national output figures for industries or commodity groupings therefore these must be estimated. 134 Sectoral estimates of regional output are often produced by assuming productivity does not vary among sub-national regions. Simple estimates, based on the Kendricks-Jaycox proportional scaling, have been used here i .e. regional output in each industry i is estimated by scaling the national output, Pi, based on the region's share of national employment for the fmancial year ending 1 998, denoted ,..---..r emp; E emp; . Thus, h h · . 135 w ere t e superscnpt r represents a reglOn. productivity exist these may also be incorporated, emp . r--.r pr = ,..---..' prod; Pi ' empi * 0 , emPi (5 .4) If reliable estimates of region-to-nation (5 .5) 134 One promising approach is the use of scaled national coefficients based on SNZ's recently released regional GST sales and purchases figures. Current limitations of this approach include how to deal with: ( 1 ) businesses that earn under $40,000 per annum, the minimum turnover required for GST registration; (2) sales and purchases of businesses, which are currently included in the GST figures; (3) limited industry disaggregation; and (4) accounting for the zero-rated GST sales of service industries. 135 The exception is the owner-occupied dwellings industry, which represents the imputed rental value of owner-occupied dwellings, and therefore has no associated employment. This industry'S output was �1998 amr estimated by mUltiplying the number of owner-occupied dwellings by the following scalar, �1996 ' amr where ;;;;;;. represents the average market rental ($ per week) for Auckland Region. 1 50 ..---..r where prod (r x 1 ) is a regional productivity vector. The ratio of regional to national labour productivity (measured in income per worker), on an industry basis, was the only productivity adjustment made here. Step 6 Derivation of the Regional Supply Matrix The national supply matrix, V, is regionalised by applying a row-only reduction scalar based on modified SLQs. This assumes, among other things, that the mix of commodities produced by an industry is spatially invariant, an approach that parallels Isard' s ( 1 95 1 ) industry technology assumption. Analytically, the first step in regionalising the supply matrix is to construct a direct requirements coefficient matrix, B. This matrix is analogous to the relationship given by a conventional inter-industry technical coefficients matrix. The formal representation, as determined in Equation 4.33, is The coefficient representing the direct requirements for national supply may be adjusted for both competitive and non-competitive imports using the SLQ approach. F irstly, this requires that row coefficients be set to zero in industries where production does not occur i.e. for non- �r competitive imports, if emp; =0 then B; =0. Secondly, row coefficients for industries assumed to be capable of satisfying local demand are left unchanged, i .e. if ';Iq; >= 1 then B; = B; . Finally, the remaining row coefficients of matrix B, those assumed incapable of satisfying local ,.---... demand (i.e. slq; < 1 ), are reduced to a regional level using, where slq; is the simple location quotient for industry i. 136 (5 .6) Ideally, output estimates would be used to calculate the SLQs. As noted previously, regional output estimates are not readily available in New Zealand. Employment was therefore used as a proxy for output. Employment estimates, measured in FTEs, were derived from SNZ's 1 36 The difference between bij and b� when ';0; < 1 is attributed to regional imports in step 1 0. 1 5 1 Business Directory for each of the 1 23 industries. 137 To improve the accuracy of the regional output estimates, management structures were excluded from non-service based industries in the calculation of SLQs. This facilitated the elimination of anomalies resulting from the presence of head office management structures. Applying unadjusted Auckland Region forestry FTEs in SLQ calculations would, for example, result in erroneous output estimates due to the presence of head office management. 138 The use of employment as a surrogate for output however introduces inaccuracies. One major drawback of using employment figures as surrogates, for example, is that productivity differences (i.e. output per FTE) between the region and nation are overlooked. A further limitation is that because the Business Directory represents annualised FTEs the seasonal trends which affect agricultural industries and, through flow-on effects, their associated processing industries are misrepresented. Where possible, ad hoc adjustments were made to the Business Directory FTEs to minimise the impact of this discrepancy using Agriculture Statistics (Statistics New Zealand, 1 998a, 200 1a), Labour Force Survey (Statistics New Zealand, 1 998g) and Census of Population and Dwellings (Statistics New Zealand, 1 996c, 200 1 c) data. Over the past three decades, a substantial body of literature has hotly debated the costs and benefits of using SLQs to regionalise national input-output models (refer to inter alia Richardson ( 1 972), Morrison and Smith ( 1974), McMenamin and Haring ( 1 974), Round ( 1 983), Sawyer and Miller ( 1 983), Jensen et al. ( 1 979) and Stevens et al. ( 1 989)). On the one hand, analysts have found that coefficients adjusted by location quotients compare favourably with actual coefficients (Schaffer and Chu, 1969; Morrison and Smith, 1 974; J ensen et al. , 1 979; Sawyer and Mil ler, 1 983). On the other hand, analysts have criticised the application of location quotients on theoretical grounds, in particular, on the ability of such a simple construct to adequately reflect the complex interrelationships that exist in an economy (Miernyk, 1 968, 1969, 1 976; Round, 1983; Stevens et al. 1 989). i39 The SLQ technique has been applied here in the absence of data suitable for any of the other approaches reviewed in Appendix D. !37 For owner-occupied dwellings, national coefficients were used as regional substitutes. 138 A recent survey by Deloitte and Touche Consulting Group ( 1 997) ranked the top 200 New Zealand businesses according to actual turnover; it found that 96 of those businesses were located within the Auckland Region. Of these, nearly all represented head offices, many of which have their principal activities located elsewhere in New Zealand. 139 Mayer and Pleeter ( 1 975) offer theoretical justifications for the use of location quotients. 1 52 Step 7 Estimation of Regional Value Added Crude estimates of regional value added may be derived by scaling each industry in the national value added matrix by its corresponding share of national output. W -r lj fJr l fJ l 0 Wij = fJi I i ' i i= . (5 .7) Note that the concave over-bar - in the left-hand term of Equation 5 .7 s ignifies that, for the time being, it is only an estimate. Sawyer and Miller ( 1 983) have identified that a large percentage of error in non-survey tables is attributable to inaccuracies in estimating value added coefficients, partiCUlarly wages, salaries and proprietor's income. Jensen and West ( 1 980), among others, have shown that the accuracy of large coefficients, such as wage and salary coefficients, in a direct requirements matrix, is a critical determinant of the accuracy of the matrix's Leontief inverse. Moreover, wages and salaries are required for the calculation of induced economic impacts brought on by consumer spending. Accurate estimation of this component is therefore particularly crucial. Fortunately, SNZ's Census of Population and Dwellings (Statistics New Zealand, 1 996c, 200 1 c) provides estimates of mean wage/salary earnings per worker (FTE) by disaggregated industry and spatial location. Implementation is as follows, r - • X (....-------r r-----. r ) Wws- IDCpW emp , ....-------r (5 . 8) where incpw is a r x 1 vector of regional labour income per worker (as measured in $ per FTE). 140 No reliable estimates of the remaining value added categories exist at the regional level. Instead the remaining components are recalibrated as suggested by Lahr (200 1 ) based on the wages and salaries row inserted in Equation 5 .8 . Wages and salaries are first subtracted from value added, and then the difference is redistributed to the remaining components of value added, keeping fixed each component's share of non-wage and salary value added, 140 For simplicity, in the context of vectors, the usual algebraic representation of a ratio is used to represent a Hadamard scalar (element-by-corresponding-element) ratio. S imilarly, the multiplication symbol, x , is used to denote Hadamard scalar (element-by-corresponding-element) multiplication. ( � ) W' . � � W' .= i 'W' -W' X op. i 'W' - W' ;I:. 0 op. ws ( i 'W' -W;s ) " WS ' 1 53 (5 .9) where the subscript opi denotes a given non-wage and salary value added row. Thus, regional value added, W', is a con.catenation of Equations 5 .8 and 5.9. Note that the concave overbar has been removed from the left hand side of Equation 5 .9 signifying that it is no longer an estimate. Step 8 Derivation of the Regional Use Matrix Once regional output and value added has been estimated, the region's use matrix may be derived. However, the so-called 'fabrication adjustment' must first be applied to the national use table in direct requirements coefficient format i.e. E= Up-l . This requires that regional output, pr, less regional value added, W', equals regional intermediate use for each industry, i.e. i'E' + [( i I W' ) Ip"] = 1, where If ' ;1:. o . This is performed by rescaling the columns oL Ii to account for known regional output and value added, (5 . 1 0) The concave over-bar once again reminds us that U' is only an estimate and that superior data could be inserted where available. Step 9 Estimation of Regional Final Demand The national final demand matrices, S and T, must also be modified to approximate regional equivalents, S' and T'. Treyz and Stevens ( 1985) argue that the coefficients of regional [mal demand may be estimated using an econometric approach dependent on regional disposable income, intermediate demand, and several other factors. The lack of specific Auckland Region data makes any attempt to derive regional final demand estimates using this approach almost futile. F inal consumption expenditures, excluding household consumption (private non-profit institutions serving households, central government and local government) were considered dependent on domestic supply and, moreover, a function of regional population size. Thus, the final consumption expenditure columns of matrices S and T were reduced to the regional level by mUltiplying each matrix element by the ratio of regional-to-national population 1 54 (5 . 1 1 ) where K represents S and T augmented i.e. SIT, pop denotes a population scalar, and old represents e ither expenditure on private non-profit institutions serving households, central government expenditure, or local government expenditure. Similarly, household consumption was considered a function of the number of households, hhlds , and concomitantly household income, ;;J;i . 14 1 Again, where superior data were known, they were inserted in place of the scaled entries. Thus, K�e was determined as follows, [[ �r ) [ �r )l r hhlds ahi � � Ki he = Ki he -=- -===" , hhlds 7:- 0 and ahl 7:- 0 , . . hhlds ahi (5 . 1 2) where the subscript he represents the household consumption final demand column. The most difficult final demand activities to regionalise were gross fixed capital formation and change in inventories. It is likely that a significant proportion of regional capital investment lies in construction expenditure; however, other forms of investment demand may also exist. Despite the fact that changes in inventories may be tracked by industry at a national level, no such data exists to undertake this at a regional level. Instead, changes in inventories, K;i ' are calculated as, (5 . 1 3) where the subscript ci denotes the changes in inventories final demand column. In this way, changes in inventories are estimated by assuming that the national ratio of changes in inventories to the industries in the use matrix is equivalent to that of the region. 141 A more conceptually appealing adjustment based on SNZ's Household Economic Survey (HES), which matched regional household types with household spend by commodity group, was also pursued. This approach parallels SNZ's own method of compiling household consumption at the national level; however this approach was stymied as SNZ would not release a concordance matching HES commodity definitions to input-output commodity definitions. 1 55 To complete the regionalisation of the fmal demand, estimates must be calculated for the value added into final demand matrices, R and Z. To faci litate this calculation, R and Z are first augmented to form ,lJ; i.e. ,lJ; = R I Z. Value added inputs into household consumption, ,lJ;'he, are generated in a similar marmer to commodity inputs into household consumption (as per Equation 5 . 1 2), [[ ----' ][""-"' ]1 ' hhlds ahi "-----' � ,lJ; i,he = ,lJ;/,he -===- � , hhlds * ° and ahl * ° , hhlds ahl (5 . 1 4) where the value added inputs into the remaining final demand categories, ,lJ;'nhe, were estimated by assuming that for each value added category the ratio of regional to national value added into non-household consumption final demand is equivalent, for each corresponding category, to that of regional to national value added, where nhc represents a given non-household fmal demand category. Step 10 Estimation of Regional Inflow and Outflow (5 . 1 5) International imports into Auckland Region are partly a function of interregional demand for each commodity, While it may be assumed that Auckland Region utilises international imports in proportion to its share of non-export demand, local production and demand patterns should also be taken into consideration. However, a paucity of regional specific data again prohibited consideration of these patterns. It was instead assumed that international imports by each regional industry did not vary relative to their national equivalents, thus, [ � ' J emp . ' � *' _ * } Y ij -Y ij =- , emPi * 0 , empj ' (5. 1 6) where Y* ij is an element in a B x r international import matrix, recording commodity i imports into industry j. Matrix y* was produced as part of Statistics New Zealand's (200 1g) interim release of the New Zealand inter-industry study. Note that the vector Yintimp = (Y*i)', where 1 56 intimp denotes the international imports row. International exports from Auckland Region, x'io intexp where the subscript intexp denotes international exports, were calculated in a similar manner to changes in inventories and gross fixed capital formation (refer to Equation 5 . 1 3). By comparison, regional economies are generally more open than their national counterparts (Richardson, 1 972; Jensen et al. 1979; Lahr 200 1 ). To account for interregional trade, inflows and outflows of commodities must also be evaluated. Ideally, this would be achieved by surveying of the origins and destinations of traded commodities within the nation. Time constraints precluded a survey; instead the industry-based rows-only location quotient approach, as employed in deriving the regional supply matrix, was used to generate crude estimates of interregional trade. 142 Specifically, the difference between the regional direct requirements coefficient, b:'i ' as calculated by Equation 5 .6, and the national direct requirements coefficient, bi,i ' provides a measure of the size of the interregional import coefficient, for those industries unable to satisfy local demand (i.e. slqi < 1 ). This assumes that the regional and national direct requirements coefficients differ only by the size of interregional importS. 143 In turn, the coefficient may be converted to a transaction value ($) by multiplying by a; , i .e. Y - (" B " B' ) " ,egimp I - I a . (5 . 1 7) This approach has been widely used by input-output analysts in non-survey regionalisation of industry-by-industry matrices (Richardson, 1 972; Jensen et al. , 1 979; Jensen and West, 1 980; Butcher, 1 985; Kerr et al. , 1 986; Jensen, 1 990), and is adapted here for use with commodity-by­ industry matrices. Step 1 1 Final Adjustments With estimates by commodity of regional output, a' , fmal demand, K' , regional outflows, X', and inflows, Y' , and by industry of regional output, P' , and value added, W' , the regional use matrix, U', was re-estimated using the MS technique to ensure table balance. The MS technique is widely utilised by national statistical agencies, including SNZ (200 1 g), as a final step in balancing input-output tables. 142 Chapter 8 and Appendix C investigate this problem further. 143 By corollary, if the slqi> 1 then the contribution made in each element of row i to interregional exports may be derived as ((hi x � ) - b . . ) f3 ' . • j I I.} ) 1 57 Once the fmal matrices have been constructed two additional tasks are commonly performed to faci litate further analys is. Firstly, a industry-by-industry (institutional) matrix for the region may be derived using the algebraic steps outlined in Section 4.4.3 .2 . Secondly, industries and commodities may be aggregated to form matrices that are significantly easier to manipulate analytically and, more importantly, these matrices are easier to interpret.l44 Several collapsible resolutions of mutually compatible defmitions were developed, allowing for a variety of intended purposes. Specifically, matrices were generated for 1 23, 48, 35 and 23 industry classifications and 2 1 0, 48 and 35 commodity classifications. Concordances showing the relationships between the various aggregate definitions and the ANZSIC and ANZCC statistical defmitions are provided in Appendix E. The completed commodity-by-industry models for New Zealand and Auckland Region for the year ending 3 1 March 1 998 are available as two Excel files ('New Zealand Commodity-by­ Industry Model.xls' and 'Auckland Region Commodity-by-Industry Model.xls ') in the Chapter 5 directory of the accompanying CD-ROM. These have been aggregated from 1 23 industries by 2 1 0 commodities to 48 industries by 48 commodities. Also included on the CD-ROM are standard Leontief input-output tables for New Zealand and Auckland Region ('New Zealand Input-Output Model.xls' and 'Auckland Region Input-Output Model.xls'), converted from 123 industries by 2 1 0 commodities to 1 23 industries by 1 23 industries under the industry-based technology assumption described in Section 4.4.3 .2, and then aggregated to 48 industries by 48 industries. The commodity-by-industry matrices, further aggregated to 3 commodities by 3 industries, and the standard Leontief input-output tables for New Zealand and Auckland Region, aggregated to 3 industries by 3 industries, are presented in Appendix F. 5.1.5 Limitations of the Auckland Region Economic Input-Output Model The methodology for deriving the commodity-by-industry model has, among other things, incorporated superior data, particularly: ( 1 ) modified regional output estimates based on the exclusion of management structures from non-service based industries, (2) the inclusion of regional productivity estimates, and (3) the inclusion of region specific ad hoc data. Despite these improvements, several major l imitations of the national update and regionalisation procedure remain leading to imprecise calculation of Auckland Region's commodity-by­ industry model: 144 Aggregation must be undertaken with care to avoid unnecessary loss of data. Minimally aggregation should consider ( 1 ) the intended purpose of the tables, (2) peculiarities in regional economic make up, including physical and political boundaries or impediments to trade, and (3) the user's ability to interpret meaningfully the information embedded within the tables. 1 58 • Cross-hauling. The use of SLQs permits only estimation of net outflows in a region. This assumption is naIve and underestimates gross imports and exports from other regions resulting in overestimates of interdependence, and as a corollary, inflated multipliers. • Multi-region consistency. The method aims to construct of a single set of supply, use and inter-industry matrices for one particular region. Additional issues, such as consistency across a multi-regional framework, require further research effort. • Self-sufficiency. The SLQs applied in the regionalisation assume maximum self­ sufficiency in each intermediate demand industry; this, in turn, leads to overestimation of coefficients and inflated multipliers. • Industry technologies. The national supply table is based on industry production technologies that represent an average for all regions in the nation. Regional production technologies can however vary considerably due to factors such as age of capital stock, relative prices and resource availability. • Consumption and investment patterns. National and regional consumption patterns may differ substantially, particularly in consumer preferences and household income. Moreover, capital investment patterns are inherently volatile, and changes in inventories are often due to factors exogenous to the regional economy, e.g. international investment decisions. 5.1.6 Accuracy of the Auckland Region Economic Input-Output Model The overarching issue in assessing the performance of any regionalisation method is how to evaluate accuracy. Partitive assessment of accuracy would require a cell-by-cell comparison with survey-based tables. 145 Cell-by-cel l accuracy has been assessed using various statistical techniques including Chi-square, correlation, regression analysis, and Theil coefficients (Schaffer and Chu, 1 969, 1 97 1 ). The absence of any survey-based table in Auckland Region however makes this infeasible. Holistic accuracy, first coined by Jensen ( 1 980), is perhaps a better measure of the accuracy of the input-output matrices created here. A holistic approach suggests that table accuracy may be viewed as a portrait, which reflects characteristics of the economy in question, both in terms of structure and function (Jensen, 1 980; West et al. , 1 980; Jensen et al. , 1988). Moreover, it attempts to derive comprehensive 'whole table' concepts from the overwhelming data contained within input-output tables. Such concepts include ( 1 ) 145 However, survey-based tables are not without their own problems, including ( 1 ) survey design errors, (2) incorrect coding at the firm level, (3) misreporting of sale and purchase figures, (4) data handling errors, and (5) in the case of compiling industry-by-industry model, errors through the use of commodity, industry or hybrid technology assumptions. 1 59 the contribution of the various constituent matrices to the whole, (2) emphasis on "the main features of the economy in terms of size and structure, with the analytically-less-important features treated as background" (Jensen, 1 980, p. 1 42), (3) the movement from fme-grained disaggregated to course-grained aggregated tables, and (4) new methods for table classification based on the development of a typology of economies. Two tests were conducted on the Auckland Region cornmodity-by-industry framework for holistic accuracy. In the fIrst test, key table indicators such as gross output, GRP balance of trade, and Type I and Type IT value added and employment multipliers were compared with independent estimates generated by Butcher Partners Limited (BPL). 146 BPL is acknowledged as the main provider of regional input-output tables in New Zealand. BPL utilises the GRIT methodology, developed by Jensen et al. ( 1979), Jensen and West ( 1 988), and West ( 1 990) at the University of Queensland, to produce regional input-output tables. A full description of the national update and regionalisation procedure used by BPL is available in Butcher ( 1985). Multiplier comparisons were made for the 48 aggregated industries147 for the updated national and Auckland Region input-output models. 148 At the national leve.1, comparisons of economy-wide gross output and GDP compared favourably. Total gross output in the New Zealand economy was estimated by BPL to be $392,8 1 6 million, while the methodology developed here produced an estimate of $388,647 million - a difference of 1 .06 percent. Estimates of national GDP were also close at $ 1 13 ,078 million (BPL) and $ 1 1 2,3 1 5 million (this study), a difference of $762 million or 0.67 percent. A comparison of Type I and Type IT value added multipliers produced less convincing, albeit still comparable, results. It was found that 73 percent of Type I value added multipliers, for example, were within 20 percent of the BPL estimate, and that the average difference (in absolute terms) was 1 6 percent. Of the Type IT value added mUltipliers, 69 percent (33 industries) were within 20 percent of the BPL estimates, that 8 percent (4 industries) recorded differences of between 50 and 70 percentl49, and that the average percentage difference (in absolute terms) was 17 percent. A correlation analysis of the Type I and n employment 146 A necessary prerequisite for this task was the conversion of the cornrnodity-by-industry matrices to an industry-by-industry form. This was achieved using an industry technology assumption as outlined in Section 4.4.3 .2 . 147 Slight differences in industry definitions meant that two industries (Construction [29] and E lectricity [26]) were not strictly comparable. 148 The comparisons recorded here are for the year ending 3 1 March 200 1 , rather than the 1 997-98 study year. Nevertheless, the methodology used in construction of the 2000-0 1 industry-by-industry table is the same. Using a 2000-01 table for comparison is likely to lead to inaccuracies in the national update phase of the methodology, as greater structural change will have occurred between 1 995-96 and 2000-0 1 , than between 1 995-96 and 1 997-98. 149 This included Construction [29], one of the two industries not strictly comparable with the BPL industries. 1 60 multipliers generated by BPL and through this study (across the industries) produced correlation coefficients at 0.84 (Type D and 0.87 (Type II). Thus, differences in the Type I and II employment multipliers were less dramatic than for the value added multipliers, with 81 percent of both Types within 20 percent of the BPL equivalents. In the regional comparison, gross output and GRP for the economy also compared well, differing from BPL estimates respectively by 0.5 percent ($694 million) and -2.0 percent (-$754 million). Interregional trade also showed a small discrepancy with imports differing by 2.2 percent and exports by -3.2 percent. The comparison of value added and employment multipliers with BPL equivalents compared well, showing a high degree of correspondence. Type I and Type II value added multipliers differed slightly, with 7 1 percent of both Types falling within 20 percent of BPL estimates, and only one industry differing by between 50 and 60 percent. Similarly, 79 percent of Type I and 75 percent of Type IT employment multipliers varied from BPL figures by 20 percent or less, generating correlation coefficients of 0.69 and 0.68 respectively. In the second test, large coefficients in the commodity-by-industry direct requirements matrices were identified, and those that had been estimated by superior data were tagged. Of the top 1 00 largest coefficients, it was found that 73 percent had been derived from superior data. Jensen ( 1 980) and Sawyer and Miller ( 1 983) inter alia, have argued that obtaining the highest possible accuracy for large coefficients is essential because these coefficients that contribute most to regional interdependency. 5.2 Structural Analysis of the Auckland Region Economy 5.2 .1 Economic Production The Auckland Region GRP for the year ending 3 1 March 1 998 was estimated to be $33 .2 billion, or 34.2 percent of national GDP. The contribution to GRP by the 20 most important industries (out of an aggregated 48) is shown in Table 5 . 1 . The difference in industry rankings between the nation and Auckland Region is also shown. Basic metals manufacture (ranked 28 out of 48), for example, is ranked 12 places higher in the contribution it makes to the Auckland Region economy compared to its national counterpart (ranked 40 out of 48). Regional industries that ranked well above their national equivalents included Rubber, plastic and other chemical product manufacturing [ 19] ; Beverages, malt and tobacco manufacturing [ 1 3 ] ; and Non-metallic mineral manufacturing [20] . Despite the lowering of trade barriers in 1 6 1 the 1 980s, and a reorientation toward provision of services, these findings corroborate earlier research which identified the region's strength in import substitution, i .e. the repackaging or processing of imported products and their subsequent redistribution to other regions. Given Auckland Region's dominance as the commercial heart of New Zealand, it is also not surprising that primary industries such as Forestry and logging [6], Dairy cattle farming [3], Livestock and cropping farming [2], and downstream Meat and meat product manufacturing [ 1 0] , are ranked well below national equivalents. 1 62 Table 5.1 Contribution to Auckland Reg ion GRP Industry 30 Wholesale trade 42 Business services 41 Ownership of owner-occupied dwellings 31 Retail trade 40 Real estate 29 Construction 37 Finance 46 Health and community services 35 Air transport, services to transport and storage 36 Communication services 45 Education 47 Cultural and recreational services 1 7 Printing, publishing and recorded media 43 Central government administration, defence, public order and safety services 1 9 Rubber, plastic and other chemical product manufacturing 24 Machinery and equipment manufacturing 33 Road transport 22 Structural, sheet, and fabricated metal product manufacturing 1 2 Other food manufacturing 32 Accommodation, restaurants and bars 48 Personal and other community services 38 Insurance 16 Paper and paper product manufacturing 14 Textile and apparel manufacturing 26 Electricity generation and supply 39 Services to finance and investment 44 Local government administration services and civil defence 21 Basic metal manufacturing 1 3 Beverage, malt and tobacco manufacturing 23 Transport equipment manufacturing 20 Non-metallic mineral product manufacturing 25 Furniture and other manufacturing 18 Petroleum and industrial chemical manufacturing 34 Water and rail transport 15 Wood product manufacturing 1 Horticulture and fruit growing 6 Forestry and logging 1 1 Dairy product manufacturing 10 Meat and meat product manufacturing 3 Dairy cattle farming 2 Livestock and cropping farming 8 Mining and quarrying 28 Water supply 27 Gas supply 7 Fishing 4 Other farming 5 Services to agriculture, hunting and trapping 9 Oil and gas exploration and extraction 5.2.2 Contribution to New Zealand GDP Contribution Difference in to GRP Rank Auckland Region cf NZ $ billion 3,231 2 3,058 -1 2,589 -1 1 ,973 0 1 ,833 1 ,442 1 1 ,350 2 1 ,309 -3 1 ,223 3 1 ,2 1 7 0 1 ,062 -3 745 5 676 6 673 -3 607 1 0 598 2 525 -2 506 9 444 2 389 0 384 5 372 8 363 5 354 7 3 1 7 - 1 1 3 1 0 8 285 -3 280 1 2 274 1 2 274 5 267 8 264 1 0 204 4 1 99 -5 1 53 -3 1 23 -3 1 1 4 - 1 4 1 06 -2 99 - 1 7 90 -27 69 -25 63 2 59 5 58 3 35 34 -1 25 -4 8 - 1 0 Figure 5 .2 shows the 20 most significant industries in Auckland Region's economy arranged in terms of the percentage contribution they make to their equivalent industry in the national economy, as measured in value added terms. This clearly shows Auckland Region's strategic 1 63 advantage as the main entry point into New Zealand with Air transport, including services to transport and storage [35] , comprising more than 55 percent of the national figure. 150 This, coupled with a strong emphasis on light manufacturing, retailing, commerce and provision of services, confirms Auckland Region as the dominant economic region within New Zealand (Deloitte and Touche Consulting Group, 1 997). S ignificant population growth over the last two decades has fuelled several residential property booms, the effects of which are captured in Figure 5 .2 with the contributions made by the Real estate [40] and Construction [29] industries. This is reinforced by SNZ ( 1999b) who note that Auckland Region ranked third out of the nation's sixteen regions in terms of the proportion of businesses involved in the Construction [29] industry. 150 Because Auckland Region is the main international gateway into New Zealand, its accommodation captures a significant portion of total guest nights; for the 1 997 calendar year, this figure was estimated to exceed 3 .6 million ( 1 6 .5 percent of the national total) (Statistics New Zealand, 1 999b). " c.Q' c .., CD ?' N -I 0 " ::0 III ::I � ::I cc l:- e (") ... iii" ::I Q, ::0 CD cc 0' ::I ::I Q, c 1/1 .. .., (ij' 1/1 � to to "" I to QC) -"'C CD .., :;-(") CD Q, ::I C .. (JI III -cc -< CD 0 0 ::I .. ::l, e' C .. 0' ::I .. 0 Z CD =e N CD III iii" ::I Q, G') C � Air trns, srvcs trns & strg Rub, plstc & chm prdct mnf Wholesale trade Print, pub & rec media Strctrl, sht & fb mtl prdct Cultural & rec srvcs Business srvcs Real estate Communication srvcs Mach inery & equip manuf Ownership 000 Finance Construction Retail trade Other food manuf Road transport Health & community srvcs Accom, restaurants & bars Education Central government % Contribution to national industry 0 en o 5.2.3 Economic Specialisation and Comparative Advantage 1 65 The industries in which Auckland Region has a comparative advantage compared with the rest of New Zealand, can be identified using the SLQs generated in Section 5 .2.2. The SLQ approach is generally the most commonly used method of determining the size of a region's basic and non-basic industries1 51 (McCann, 200 1 , p. 1 48) and of measuring the comparative strength or weakness of an industry in a region, relative to the national s ituation. Accordingly, industries with a SLQ > 1 are, relative to the nation, capable of supplying local demand, and are therefore deemed to be strong industries. Those industries with a SLQ < 1 are, relative to the nation, incapable of supplying local demand and deemed weak industries. Of 48 industries in the Auckland Region economy, 25 have a comparative advantage over the nation (Table 5 .2). Again the importance of Auckland Region as the key entry point into New Zealand is highlighted, along with light manufacturing and wholesaling aligned with import substitution, and the general associating of business services with all parts of the economy. Although the location quotient may be used to specify industries with a comparative advantage over their national counterparts it is also important to consider the size of the industry, as measured by contribution to value added. An analysis of both comparative advantage and size reveals that not only does Auckland Region have 25 industries with comparative advantage, but also 23 of these industries, as a percentage of their national counterparts, contribute more to GRP than the overall Auckland Region economy average of 34.2 percent. Nevertheless, the percentage contribution to overall GRP is small for several industries including Water supply [28] ($59 million or 0. 1 8 percent of GRP), Petroleum and industrial chemicals manufacture [ 1 8] ($204 million, 0.62 percent) and Furniture and other manufacturing [25] ($264 million, 0.80 percent). Hence, those industries with a comparative advantage also make a s ignificant value added contribution. 15 1 Basic industries tend to operate, be owned, and make decisions at the national or international level. They are often large scale and, in the case of business services, concentrated in high quality office space with skilled staff. Non-basic industries tend to serve local demand and are usually concentrated throughout cities. 1 66 Table 5.2 Location Quotients for the Auckland Region Economy Industry 35 Air trns, srvcs trns & strg 1 9 Rub, plstc & chm prdct mnf 25 Furniture & other manuf 30 Wholesale trade 21 Basic metal manuf 1 3 Bev, malt & tobacco manuf 1 7 Print, pub & rec media 22 Strctrl, sht & fb mtl prdct 39 Srvcs to fnnce & invstmnt 20 Nn-mtllc mnrl prdct manuf 40 Real estate 42 Business srvcs 28 Water supply 38 Insurance 14 Textile & apparel manuf 36 Communication srvcs 2 3 Transport equip manuf 24 Machinery & equip manuf 47 Cultural & rec srvcs 1 8 Ptrlm & ind chem manuf 37 Finance 29 Construction 1 6 Paper & paper prdct manuf 48 Personal & other comm srvcs 31 Retail trade 1 2 Other food manuf 27 Gas supply 3 3 Road transport 46 Health & community srvcs 3 2 Accom, restaurants & bars 4 5 Education 44 Local government 4 3 Central government 34 Water & rail trans 1 5 Wood prdct manuf 2 6 Elctrcty gnrtn & spply 1 1 Dairy prdct manuf 1 Hort & fruit growing 8 Mining & Quarrying 7 Fishing 4 Other farming 1 0 Meat & meat prdct manuf 6 Forestry & logging 5 Srvcs to ag, hnt & trppng 3 Dairy cattle farming 2 Livestock & cropping 9 Oil & gas explr & extrc 4 1 Ownership 0001 Note: Simple Location Quotient 1 . 72 1 .62 1 . 54 1 . 54 1 . 53 1 . 53 1 . 5 1 1 .42 1 . 3 1 1 .3 1 1 . 30 1 . 30 1 .28 1 .27 1 .24 1 .23 1 .23 1 .22 1 . 1 7 1 . 1 4 1 . 1 0 1 .08 1 .07 1 . 05 1 .00 0.99 0.98 0.98 0.90 0.88 0.87 0 . 7 1 0.70 0.65 0 . 57 0 . 56 0.48 0.48 0.40 0.33 0.30 0.23 0. 1 6 0. 1 5 0. 1 3 0. 1 3 0 . 06 N/A 1 . The output of this industry represents the imputed rental value of owner-occupied dwellings. As this industry has no employment a location quotient cannot be calculated. 1 67 5.2.4 Balance of Trade Auckland Region's combined interregional and international balance of trade was estimated to be $957 million for the year ending 3 1 March 1 998 (Table 5 .3). The region was therefore a net exporter of goods and services. By contrast, regional expenditure on international imports, $9,65 1 million, exceeded revenue generated from international exports, $9, 1 74 million, resulting in a trade deficit of $477 million. l 52 Some 33 .8 percent (or $3, 1 05 million) of the value of Auckland Region's international exports comprised so called re-exports i.e. goods and services imported from other regions/abroad for export elsewhere. 1 53 This is not a surprising finding given that Ports of Auckland is the nation's largest cargo port, handling 3 .4 million tonnes of exports for the year ending 30 June 1 996 (Statistics New Zealand, 1 999b). The sizeable export earnings generated from air transport and its associated services IS a consequence of the presence of Auckland International Airport, the most active gateway in and out of New Zealand. Close to 22 percent ($2, 1 20 million) of international imports are purchased by the region's Households [49] . This includes goods imported by local wholesalers/retailers and, in turn, on­ sold with an additional markup to households e.g. motor vehicles and computers. The importations of Gross fixed capital [53], particularly plant and machinery, is also significant at $ 1 ,57 1 million. Auckland Region's historical role in import substitution, through final processing, repackaging and redistribution to other New Zealand centres, particularly in the import contributions made by Wholesale trade [30] and the Rubber, plastic and chemical product manufacturing [ 19] industry. Again, Auckland International Airport's influence is apparent. Interregional trade is of a similar magnitude to international trade. Importation of international exports for re-export accounts for 33 .7 percent of the $9, 1 92 million of interregional imports. Households [49], Central government [5 1 ] , and Wholesale trade [30] consume $2,501 million (27.2 percent) of goods and services imported from other regions. By comparison, interregional exports amounted to $10,626 million. Principal exporting industries were the Wholesale trade [30], Business services [42] and Air transport etc. [35] industries; these account for 43 . 1 percent (or $4,580 million) of interregional exports. These findings substantiate Auckland Region' s claim as New Zealand's commercial hub and key gateway for international travellers, and 152 New Zealand as a whole recorded a trade surplus of$343 million for the same period. 1 53 Some commentators have noted that it is advantageous for cargo ships destined for Australia to first berth in New Zealand, and in so doing, gain cheaper port handling fees on their arrival in Australia due to the bilateral Closer Economic Relations (CER) agreement between the two nations. 1 68 emphasise its national role in import substitution and redistribution. Overall, Auckland Region recorded a significant interregional trade surplus of $ 1 ,434 million. Table 5.3 Auckland Region's Financial Balance of Trade, 1 997-98 Exports Value Imports Value Balance of Trade $ mil l ion $ mill ion $ million Interregional exports Interregional imports 30 Wholesale trade 2,269 56 International exports 3, 1 05 42 Business services 1 , 270 51 Central government 983 35 Air transport, services to transport and storage 1 , 041 49 Households 954 1 9 Rubber, plastic and other chemical product manufacturing 635 30 Wholesale trade 564 29 Construction 444 1 1 Dairy product manufacturing 5 1 9 Others 4,968 Others 3,068 Sub-total 1 0, 626 Sub-total 9, 1 92 1 , 434 International exports International imports 55 Interregional imports 3, 1 05 49 Households 2, 1 20 1 1 Dairy product manufacturing 643 53 Gross fixed capital formation 1 , 57 1 35 Air transport, services to transport and storage 562 35 Air transport, services to transport and storage 6 1 6 24 Machinery and equipment manufacturing 472 30 Wholesale trade 5 1 1 14 Textile and apparel manufacturing 426 1 9 Rubber, plastic and other chemical product manufacturing 403 Others 3, 966 Others 4,431 Sub-total 9, 1 74 SUb-total 9,651 -477 Total exports 1 9, 800 Total imports 1 8, 843 957 1 70 5.2.5 Network Analysis of Financial Flows: Clusters of Comparative Advantage An analysis of the major financial flows between the top twenty industries with a comparative advantage, as measured in SLQ terms, provides an overall understanding of how the Auckland Region economy is structured (Figure 5.3). The two largest financial inputs for each industry are recorded, along with each industry's gross output (within the boxes). 1 54 Moreover, the industries are broadly grouped: industries driven by local demand, industries driven by interregional or international export demand, and industries driven by intermediate demand in supporting roles. Placement in each group was determined by calculating the share of each industry's output consumed by local demand, export demand and by other industries. Cross boundary industries are also identified. Several key clusters of comparative advantage are described below. 1 54 Only inter-industry flows are considered. Primary inputs, such as wages and salaries, and imports, are not considered - these inputs often constitute 30 to 40 percent of the financial value of total inputs into an industry. Figure 5.3 Industries driven by Intermediate Demand 99m Srvcs 10 fnnce & invstmnl 643m Industries driven by Local Demand 1 68m 33m Insurance 699m 103m Rub, plstc L-__________ -\ ________ ��� & �m prdct mnf 1 ,761 m Transport �_5_9_m--__>.,------------------------__Y equip manuf 987m 84m Auckland Region's Clusters of Comparative Advantage, 1 997-98 tobacco Air tms, srvcs tms & strg 3,1 15m E M ... N Basic melal � manuf Elctrcty gnrtn 1 , 1 1 9m & spply 95m Industries driven by International and Other Region Demand 1 72 5.2.5.1 Industries Driven by Export Demand A feature of the industries driven by export demand is that, with the exception of Air transport and storage [35] and Wholesale trade [30], they are all involved in manufacturing. Furthermore, the emphasis is primarily on light manufacturing such as Beverages, malt and tobacco [ 1 3 ] ; Textile and apparel [ 14]; Rubber, plastics and chemical products [ 1 9] ; and Furniture and other manufacturing [25]. Basic metal manufacture [2 1 ] , primarily steel manufacture at Glenbrook, is the exception. Glenbrook produces steel using titanomagnetite sand as an ore and low quality coal as the redundant. It can produce around 700,000 tonnes per annum, half of which is generally exported, with the remainder satisfying about half New Zealand's steel requirements (Statistics New Zealand, 1 999b). There are no major interdependencies highlighted between the light manufacturing industries driven by export demand. It is however important to note that the Wholesale trade [30] industry plays a critical role in the distribution of commodities between these industries. Approximately three-quarters (73 .8 percent) of the Wholesale trade industry's inputsl55 are sourced from intermediate demand industries, with the remainder from international and interregional imports. Thus, the Wholesale trade industry acts as the key distributor of domestically produced light manufacturing commodities within the Auckland Region economy. The input mixes of the light manufacturing industries with comparative advantage reveal two distinct groupings. On the one hand, the input mixes of the Textile and apparel [ 14] and Furniture and other manufacturing [25] industries depend upon significant inputs of primary products sourced from other regions in New Zealand. On the other hand, the input mixes of the Beverage, malt and tobacco [ 1 3 ] and Rubber, plastic and chemical product [ 1 9] manufacturing industries largely depend on products sourced from other industries within the Auckland Region economy. Air transport and its associated services [35] is also identified as a key industry driven by export demand. Auckland International Airport is the country's main gateway handling the bulk of airfreighted exports. More than 7.5 million domestic and international passengers arrived and departed through Auckland airport for the year ending June 1 997, a figure almost twice the then resident population of New Zealand. Auckland International Airport is a critical factor in the strength of businesses and their associated financial services. Of the top 200 largest companies in New Ze�land, an estimated 96 companies had their head offices situated in Auckland Region, 155 Not including primary inputs such as wages and salaries, subsidies, depreciation of fixed capital and so on. 1 73 at least in partly because of Auckland International Airport (Deloitte and Touche Consulting Group, 1 997). 5.2.5.2 Support Industries Driven by Intermediate Demand Two key c lusters of comparative advantage are apparent among industries driven by intermediate demand in Auckland Region, namely provision of Business services [42] and Wholesale trade [30]. Business services [42] are provided to industries driven by both local and export demand, as well as to the region's financial and insurance industries. Only 1 4 percent of Business services [42] outputs are provided to comparative advantage industries driven purely by intermediate demand. All of these industries are also service industries e.g. Services to finance and investment [39] and Communication services [36] . This is not surprising as, relative to the national economy, business and fmancial services provide a larger proportion of jobs, a finding aligned with Auckland Region's importance as a commercial hub. The remaining Business services [42] outputs are to industries driven by some combination of intermediate, local or export demand, and are utilised by service industries (e.g. Cultural and recreational services [47]) and light manufacturing industries (e.g. Rubber, plastic and chemical manufacturing [ 1 9]). Another industry with comparative advantage is Printing, publishing and recorded media [ 1 7] which provides large inputs into the Business services [42] industry. The fastest growing areas of the Auckland Region economy over the past decade have been fmance, real estate and business services, all of which appear as industries with comparative advantage in the Auckland Region (Auckland Regional Council, 2003, 2004). Another area that has experienced growth is the distribution sector with Wholesale trade [30] playing a pivotal role in servicing light manufacturing industries, in particular Machinery and other equipment [24], Transport equipment [23] , Structural, sheet and fabricated metal [22], Rubber, plastic and chemical [ 1 9] and Furniture and other [25] manufacturing. The manufacturing industry, although historically significant to the Auckland Region economy, has declined in relative importance in recent years, with a move towards import substitution. The central role of Wholesale trade [30] in servicing light manufacturing highlights this emphasis on processing, repackaging and redistribution of commodities. 5.2.5.3 Service Industries Driven by Local Demand Within Auckland Region, no industries with a comparative advantage are driven purely by local demand. In fact, education, health or community, social or personal industries are notably absent as drivers of local demand. The under representation of these industries in Auckland 1 74 Region as compared to the nation may be more a factor of economies of scale or the ability to provide such services in a more spatially centralised way than in other regions in New Zealand. A particularly notable exception is the absence of education as driven by both local and international demand, although this industry has grown substantially s ince the 1 997-98 static snapshot analysed here. Nevertheless, several industries are driven by either a combination of local, interregional and international demand (e.g. Beverage, malt and tobacco manufacturing [ 1 3] , Furniture and other manufacturing [25], and Wholesale trade [30]), a combination of local and intermediate demand (e.g. Real estate [40] and Insurance [38]) or a combination of local, export and intermediate demand (e.g. Furniture and other manufacturing [25] and Wholesale trade [30]). 5.2.6 Multiplier Analysis of Income and Employment Impacts Perhaps the major reason for deriving industry-by-industry model from their commodity-by­ industry model is that conventional input-output multipliers may be calculated. Input-output multipliers show the relationship between an additional unit of spending (final demand) and changes in output, income, value added and employment within the economy. They capture not only the direct effects of additional spending captured, but also the indirect effects resulting from the interdependencies that exist between industries within the economy. Say, for example, a group of investors decide to invest in the development of a high-rise tower block. The major direct impacts reSUlting from this investment would primarily be felt within the construction industry e.g. payments by developers for architectural services, project managers, earthworks, carpentry services and so on. In turn, this would affect associated expenditure, e.g. purchases of steel, concrete and timber from other industries. In this way, economic impacts beyond the initial change in final demand may be initiated in the economy. Moreover, additional impact may be induced through household expenditure generated because more people earn wages and salaries. Input-output practitioners summarise direct and indirect economic impact using Type I multipliers, and direct, indirect and induced economic impact using Type 11 multipliers. Although input-output multipliers are not a modem development, having been first conceived of by Kahn ( 1 93 1 ) and popularised by Keynes ( 1 936), they are a useful measure of structural interdependence within an economy. Moreover, they are not only used below to summarise the structural fmancial interdependencies of the Auckland Region economy in terms of economy output, income, value added and employment, but when coupled with the physical equivalents outlined in Chapter 6, are also used to construct new measures of the impact of structural 1 75 interdependence. The calculation of output, income, value added and employment mUltipliers is described in detail in Appendix G. 5.2.6.1 Output Multipliers Output multipliers relate a unit of spending to an increase in output in the economy. Industries with large Type I output multipliers in the Auckland Region (Table 5 .4) included Water supply [28], Construction [29], Structural, sheet and fabricated metal product manufacture [22] and Machinery and equipment manufacture [24] . These results corroborate existing research findings; for example, the Auckland Regional Council (2003, 2004) estimates that construction, particularly residential housing growth fuelled by an increasing population, was responsible for more than 30 percent of Auckland Region's economic growth during the past decade. The strong linkage between Structural, sheet and fabricated metal product [22] and Machinery and equipment [24] manufacturing is aligned to the construction industry and the presence of the nationally significant Glenbrook steel mill. The provision of reticulated water is a critical operational ingredient in many manufacturing industries; this was critically highlighted during the region's 1 994 water crises, 1 56 and acted as a catalyst for constructing the Waikato pipeline. By contrast, the Finance [7] and Education [45] industries exhibit very weak interlinkages within the region'S economy. Since the mid 1 990s, Auckland Region has witnessed considerable growth in 'export-education' ( i .e . foreign student registrations) spurred by favourable foreign exchange rates and international perceptions of New Zealand as a politically stable, clean-green, and safe place to live. While only minimal indirect effects will be stimulated by education expenditure, more significant economic impact will be generated through foreign student expenditure or Retail trade [3 1 ] and Personal and other community services [48] . 1 56 At the time the New Zealand Treasury speculated that the drought would cost $800 million in real terms (National Business Review, 1 994). 1 76 Table 5.4 Output Multipliers for Auckland Region and New Zealand, 1 997-98 Auckland Region New Zealand Type I Type I I Type I Type I I Type I Type 1 1 Output Output Output Output Output Output Multiplier Multiplier Multiplier Multiplier Multiplier Multiplier Rank Rank Rank Rank 1 Hort & fruit growing 1 .59 2.44 31 27 1 8 1 2 2 Livestock & cropping 1 .54 2.21 36 34 1 3 1 6 3 Dairy cattle farming 1 .48 2 .05 41 42 33 40 4 Other farming 1 .65 2.21 24 35 22 33 5 Srvcs to ag, hnt & trppng 1 .62 2.52 27 21 23 1 5 6 Forestry & logging 1 . 53 1 .95 38 45 26 38 7 Fishing 1 .35 1 .8 1 4 8 46 42 46 8 Mining & Quarrying 1 .57 2 . 1 8 3 3 37 2 1 31 9 Oil & gas explr & extrc 1 .60 2.07 30 4 1 2 0 39 10 Meat & meat prdct manuf 1 .58 2 . 1 9 32 36 3 1 1 Dairy prdct manuf 1 .36 1 .64 46 47 2 4 12 Other food manuf 1 .83 2.55 7 1 8 9 20 1 3 Bev, malt & tobacco manuf 1 .92 2.57 5 1 7 8 25 14 Textile & apparel manuf 1 .64 2.42 25 28 1 1 1 1 1 5 Wood prdct manuf 1 .60 2.28 29 31 4 5 1 6 Paper & paper prdct manuf 1 .53 2 . 1 2 37 38 36 41 17 Print, pub & rec media 1 .71 2.60 1 6 1 6 30 30 1 8 Ptrlm & ind chem manuf 1 .76 2.39 1 2 30 34 43 19 Rub, plstc & chm prdct mnf 1 .67 2.40 22 29 32 35 20 Nn-mtllc mnrl prdct manuf 1 .76 2.51 1 3 22 1 7 22 2 1 Basic metal manuf 1 .83 2.50 6 23 7 1 4 22 Strctrl, sht & fb mtl prdct 1 .96 2.89 3 5 1 0 1 0 23 Transport equip manuf 1 .71 2.45 1 7 25 35 36 24 Machinery & equip manuf 1 .93 2.80 4 8 1 6 1 8 25 Furniture & other manuf 1 .76 2.69 1 4 1 2 1 4 9 26 Elctrcty gnrtn & spply 1 .68 2 . 1 0 2 1 40 5 27 27 Gas supply 1 .54 2 .01 35 43 1 2 32 28 Water supply 2.34 2 .90 1 4 1 6 29 Construction 1 .98 2.87 2 6 6 7 30 Wholesale trade 1 .69 2.53 20 20 24 26 31 Retail trade 1 .60 2.67 28 1 3 38 24 32 Accom, restaurants & bars 1 .77 2.62 1 1 1 5 1 5 1 3 3 3 Road transport 1 .71 2.72 1 8 1 1 31 23 34 Water & rail trans 1 .36 2.21 45 33 44 42 35 Air trns, srvcs trns & strg 1 .66 2.44 23 26 37 37 36 Communication srvcs 1 .49 2 . 1 2 40 39 43 45 37 Finance 1 .37 2.22 44 32 48 44 38 Insurance 1 .64 2.49 26 24 40 34 39 Srvcs to fnnce & invstmnt 1 .81 2.81 9 7 27 2 1 4 0 Real estate 1 . 52 1 .96 39 44 41 47 41 Ownership 000 1 . 39 1 .59 43 48 46 48 42 Business srvcs 1 . 76 2.78 1 5 1 0 29 1 9 43 Central government 1 . 55 3.1 9 34 39 2 44 Local government 1 .82 3.02 8 2 1 9 3 45 Education 1 . 35 2.80 47 9 47 1 7 46 Health & community srvcs 1 .45 2.63 42 1 4 45 29 47 Cultural & rec srvcs 1 .70 2.55 1 9 1 9 28 28 48 Personal & other comm srvcs 1 .81 2.99 1 0 3 25 8 Once induced impacts, initiated through consumer spending, are considered (as recorded by Type II multipliers), the ranking of industries which are most strongly interconnected in the economy changes. Both Central [43] and Local [44] government display a high level of interdependence within the regional economy. Since the fourth Labour Government in 1 984, central government policy has focused on economic deregulation, breaking of trade tariffs and, in the case of sub-national governance, devolution to local authorities. Nevertheless, lack of 1 77 leadership and direction at the national level on major issues within Auckland Region (e.g. traffic congestion, energy and water supply) has led to more interventionist policy since the mid 1 990s e.g. the appointment of a minister to assist the Prime Minister with Auckland Region issues, the diversion of national petrol taxes revenue to help alleviate Auckland Region's roading issues, and the legislation of the Waikato Pipeline Act. At the local government level, significant effort has been devoted in recent years to an Auckland-wide growth policy (e.g. the Growth Forum, the Auckland Regional Land Transport Strategy (ARLTS), and Regional Economic Development Strategy (AREDS» and infrastructure funding and planning (e.g. the establishment of Infrastructure Auckland, and its successors Auckland Regional Holdings (ARH) and Auckland Regional Transport Authority (ARTA» . Thus, taken together, central and local government have a significant effect on interlinkages within the Auckland Region economy, not only in terms of expenditure patterns (as captured by multipliers), but also through the circumlocutory effects of policy. 5.2.6.2 Value Added Multipliers Value added multipliers show the relationship between an additional unit of spending and changes in the level of value added generated in an economy. Underpinning these value added multipliers is the argument that if an industry's output changes, there will be an associated change in value added and, in turn, final demand. Value added mUltipliers for Auckland Region are shown in Table 5.5 . Industries are ranked according to multiplier size in a descending order. Columns 5 and 6 of the Table records each industry's equivalent rank in the national economy e.g. the Auckland Region Water supply [28] industry value added multiplier ranks as the third largest multiplier regionally, while nationally the Water supply industry ranks fourth. The rank ordering of the five largest Type I and Type IT value added industry mUltipliers reveals similar rankings across industries, irrespective of the introduction of induced effects. These industries include the Beverage, malt and tobacco industry [ 1 3 ] which has historically been a dominant local industry (Statistics New Zealand, 1 999b), and Machinery and equipment manufacture [24] which delivers and consumes the outputs of the region's light manufacturing industries. Low ranking industries in terms of value added flow-on effects include Real estate [40], Finance [37], Education [45] and Ownership of owner occupied dwellings [4 1 ] . In this respect, Auckland Region is similar to the nation. The flow on impacts of these industries tend to be low because these industries have low operating overheads and limited interconnections with manufacturing and primary industries. 1 78 Table 5.5 Value Added Mu ltipl iers for Auckland Region and New Zealand, 1 997-98 Auckland Region New Zealand Type I Type 1 1 Type I Type 11 Type I Type 1 1 Value Value Value Value Value Value Added Added Added Added Added Added Multiplier Multiplier Multiplier Multiplier Multiplier Multiplier Rank Rank Rank Rank 1 Hort & fruit growing 1 . 51 1 .90 36 36 26 26 2 Livestock & cropping 1 .46 1 .76 40 40 27 33 3 Dairy cattle farming 1 .60 1 .99 32 34 25 27 4 Other farming 1 .82 2.20 1 6 25 1 8 1 9 5 Srvcs to ag, hnt & trppng 1 .58 2.03 35 32 30 29 6 Forestry & logging 2 .96 3.73 2 2 5 5 7 Fishing 1 .40 1 .69 42 43 38 41 8 Mining & Quarrying 1 .59 1 .95 34 35 23 28 9 Oil & gas explr & extrc 1 .91 2.34 1 3 1 5 1 5 1 7 1 0 Meat & meat prdct manuf 1 .97 2.54 1 1 1 0 2 2 1 1 Dairy prdct manuf 1 .79 2 . 1 7 1 9 28 1 1 12 Other food manuf 2.21 2 .83 5 5 7 7 1 3 Bev, malt & tobacco manuf 3.00 3.81 1 3 3 14 Textile & apparel manuf 1 .77 2.30 2 1 1 8 1 2 1 1 1 5 Wood prdct manuf 1 .74 2.24 25 23 9 9 16 Paper & paper prdct manuf 1 .62 2.03 31 31 31 32 1 7 Print, pub & rec media 1 .68 2 . 1 6 30 29 33 35 18 Ptrlm & ind chem manuf 2.02 2.50 1 0 1 1 1 4 1 6 1 9 Rub, plstc & chm prdct mnf 1 . 81 2.31 1 7 1 6 29 25 20 Nn-mtllc mml prdct manuf 1 .92 2.44 1 2 1 3 1 6 1 5 21 Basic metal manuf 2.08 2.60 8 9 8 1 0 22 Strctrl, sht & fb mtl prdct 2.04 2.65 9 7 1 7 1 4 2 3 Transport equip manuf 1 .90 2.46 14 1 2 22 22 24 Machinery & equip manuf 2.22 2 .89 4 4 1 3 1 2 25 Furniture & other manuf 1 .74 2 .26 26 21 21 1 8 26 Elctrcty gnrtn & spply 2 . 1 3 2.61 7 8 6 6 27 Gas supply 1 .75 2 . 1 3 24 30 1 1 1 3 28 Water supply 2.79 3.45 3 3 4 4 29 Construction 2 . 1 8 2.83 6 6 1 0 8 30 Wholesale trade 1 .78 2.29 20 1 9 1 9 21 31 Retail trade 1 .43 1 .85 41 38 41 39 32 Accom, restaurants & bars 1 .75 2.27 23 20 20 20 33 Road transport 1 .59 1 .99 33 33 37 37 34 Water & rail trans 1 .29 1 .65 46 46 44 43 35 Air trns, srvcs trns & strg 1 .76 2.25 22 22 32 30 36 Communication srvcs 1 .51 1 .88 37 37 40 40 37 Finance 1 .30 1 .66 45 45 47 46 38 Insurance 1 .70 2 . 1 8 28 27 36 36 39 Srvcs to fnnce & invstmnt 1 .84 2 . 37 1 5 1 4 28 24 40 Real estate 1 .47 1 .68 39 44 42 45 41 Ownership 000 1 .39 1 .50 43 48 43 48 42 Business srvcs 1 .72 2.22 27 24 34 34 43 Central government 1 .27 1 .69 47 42 45 42 44 Local government 1 .47 1 .85 38 39 39 38 45 Education 1 . 1 7 1 .53 48 47 48 47 46 Health & community srvcs 1 .33 1 .75 44 41 46 44 47 Cultural & rec srvcs 1 .80 2.31 1 8 1 7 24 23 48 Personal & other comm srvcs 1 .69 2.20 29 26 35 31 5.2.6.3 Employment Multipliers Employment multipliers show the relationship between an additional unit of spending and changes in the level of employment in an economy. Employment multipliers for the Auckland Region are depicted in Table 5 .6, using the same layout as Table 5 .5 . 1 79 Industries with large Type I and Type IT employment multipliers in the Auckland Region include Oil and gas exploration [9] , Beverage, malt and tobacco manufacture [ 1 3], Forestry and logging [6] and the Electricity generation [26] and Water supply [28] utilities. By comparison, these industries are ranked, respectively, flrst, third, tenth, fourth and seventh at the national level. These similarities in the rankings of the employment multipliers at the regional and national levels are not unexpected. Commentators such as the Auckland Regional Council (2003, 2004) argue that the region'S business cycle closely resembles the nation's, while others simply note that, with the exception of farming activities, Auckland Region is the dominant manufacturing and service hub in New Zealand. Several similarities exist between the industries with large employment multipliers within the Auckland Region. The most obvious similarity is that they are primarily utilities or manufacturing industries with significant backward l inkages or upstream effects associated with their operation, typically these linkages are to primary resource extraction industries e.g. water and beverage manufacture. A second similarity is that it is principally the management structures of the industries that are located within Auckland Region, and these management structures are purchasing manufactured goods (particularly machinery and equipment) that are mostly imported, finished and redistributed from Auckland Region. A final similarity is that these industries have typically been the mainstay of the Auckland Region economy. 1 80 Table 5.6 Employment Multipl iers for Auckland Reg ion and New Zealand 1 997-98 Auckland Region New Zealand Type I Type 1 1 Type I Type 1 1 Type I Type 1 1 Employment Employment Employment Employment Employment Employment Multiplier Multiplier Multiplier Multiplier Multiplier Multiplier Rank Rank Rank Rank 1 Hort & fruit growing 1 .27 1 .46 45 45 41 43 2 Livestock & cropping 1 .38 1 .61 40 40 28 36 3 Dairy cattle farming 1 .35 1 . 55 42 42 38 42 4 Other farming 1 .46 1 .67 38 39 34 39 5 Srvcs to ag, hnt & trppng 1 .35 1 .59 41 41 40 41 6 Forestry & logging 3.31 4. 1 3 3 5 9 1 0 7 Fishing 1 .47 1 .79 37 38 36 38 8 Mining & Quarrying 2.22 2.95 1 1 1 1 1 1 1 1 9 Oil & gas explr & extrc 7.30 1 0.04 1 1 1 1 0 Meat & meat prdct manuf 1 .92 2.40 1 9 20 6 6 1 1 Dairy prdct manuf 2.80 3.48 8 9 2 2 1 2 Other food manuf 2.22 2.80 1 2 1 2 1 2 1 3 1 3 Bev, malt & tobacco manuf 4.25 5.45 2 2 3 3 1 4 Textile & apparel manuf 1 .62 2.01 30 32 1 7 20 1 5 Wood prdct manuf 1 .61 2.00 31 33 1 5 1 6 1 6 Paper & paper prdct manuf 1 .96 2.63 1 6 1 5 1 6 1 5 1 7 Print, pub & rec media 1 .65 2 . 1 3 28 28 31 29 18 Ptrlm & ind chem manuf 2.82 3.70 7 7 8 8 1 9 RUb, plstc & chm prdct mnf 1 .94 2.52 1 7 1 7 2 1 1 9 2 0 Nn-mtllc mnrl prdct manuf 2.09 2.74 1 3 1 3 1 4 1 4 2 1 Basic metal manuf 2.72 3.59 9 8 1 0 9 22 Strctn , sht & fb mtl prdct 1 .89 2.39 22 21 22 22 23 Transport equip manuf 1 .89 2.38 21 22 24 25 24 Machinery & equip manuf 2.00 2.51 1 5 1 8 1 9 2 1 25 Furniture & other manuf 1 . 56 1 .94 33 35 30 32 26 Elctrcty gnrtn & spply 3.29 4.35 4 3 4 4 27 Gas supply 3.00 4. 1 5 6 4 5 5 28 Water supply 3.09 3.95 5 6 7 7 29 Construction 1 .89 2.31 20 23 20 23 30 Wholesale trade 1 .71 2. 1 5 26 27 25 26 31 Retail trade 1 .28 1 . 54 43 43 45 44 32 Accom, restaurants & bars 1 .28 1 .45 44 46 44 45 33 Road transport 1 .61 1 .99 32 34 35 37 34 Water & rail trans 1 .48 2 . 1 0 35 29 37 28 35 Air trns, srvcs trns & strg 2.02 2.67 14 1 4 1 8 1 7 3 6 Communication srvcs 1 .69 2 . 1 6 27 25 32 30 37 Finance 1 .48 2.04 36 31 43 34 38 Insurance 1 .93 2.57 1 8 1 6 2 3 1 8 3 9 Srvcs to fnnce & invstmnt 1 .87 2.41 23 1 9 26 24 40 Real estate 1 .81 2.21 24 24 27 27 41 Ownership OOD 0.00 0.00 48 48 48 48 42 Business srvcs 1 .64 2.06 29 30 33 35 43 Central government 1 .39 1 .93 39 36 42 33 44 Local government 2.36 3.29 10 1 0 1 3 1 2 4 5 Education 1 . 1 5 1 .43 47 47 47 47 46 Health & community srvcs 1 .25 1 .53 46 44 46 46 47 Cultural & rec srvcs 1 .73 2 . 1 6 25 26 29 3 1 4 8 Personal & other comm srvcs 1 .50 1 .83 34 37 39 40 5.2.6.4 A Final Note on Auckland Region's Structural Interdependencies To complete the analys is of Auckland Region's mUltiplier impacts two further insights, albeit less related to the core theme of mUltipliers, require additional discussion. F irstly, structural interdependencies in the Auckland Region economy have weakened over time. Work by Le Heron and McDonald (2005) has replicated the clusters of comparative advantage analysis, 1 8 1 undertaken in Section 5 .4.5, and the multiplier analysis work performed here, for the 1986-87, 1 989-90, 1 992-93, 1 995-96, 1 997-98, and 2000-0 1 financial years. Although the principal theme of this comparative static study is identification and explanation of Auckland Region' s drivers of structural change, the paper shows that interlinkages within the region' s economy have been weakening in favour of a more open Pacific-based economy. Supporting this argument is an increased gap between the region's import dependence and GRP growth i.e. regional imports are growing at a greater rate than regional GRP. Secondly, the impact of tourism on the Auckland Region economy is apparently ignored. This is surprising given that the tourism industry has grown substantially over the last two decades (McDermott, 1 998; Goh and Fairgray, 1999a, 1 999b; Statistics New Zealand, 2000), and that Auckland Region is the key gateway in and out of the nation. The apparent omission of tourism effects is however more an artefact of industry (ANZSIC) classification than actual occurrence. This reveals a significant limitation of the way statistics are collected in New Zealand, namely, the impacts of tourism are ostensibly omitted because coding by the ANZSIC system which has no tourism industry per se, but instead represents tourism through the partial output of several industries, e.g. Retail trade [3 1 ] , Accommodation, restaurants and bars [32] , Air transport and and associated services [35] . This defmitional problem is not limited to tourism alone; other emergmg industries are also not recorded uniquely e.g. the amalgamation of telecommunications, radio and television. Ideally, with these emerging industries typically focused on consumption rather than production, an aligned movement in the way industries are c lassified seems compelling. Chapter Six Physical Input-Output Model: Physical Flows in the Auckland Region Economy and Environment 1 83 In this Chapter a physical input-output model ' 57 of the Auckland Region economy is constructed, using the commodity-by-industry framework (Table 4.2) as was previously mathematically defined in Chapter 4. To the author's knowledge, this is the first regional-level physical input-output model that has been constructed, with previous efforts focusing on the national level. l58 The most comprehensive models developed, for example, are for Germany and Denmark, both of which are at the national level. This model has two features that can be considered to be extensions to the economic input­ output model : ( 1 ) it quantifies the flow of commodities between industries in the Auckland Region economy in physical (mass' 59) terms rather than in financial terms, and (2) it quantifies the physical flows between industries in the economy and the biophysical environment. This includes mainly raw material inputs (e.g. soil, water, air) used by economic industries and residuals (e.g. solid wastes, air emissions) produced by these industries. The construction of physical input-output models is important for understanding the biophysical fimctioning of the economy, in terms of theoretical schemas advocated by the early ecological economists such as Boulding ( 1 966), Georgescu-Roegen ( 1 97 1 ) and Daly ( 1 973), which emphasised the criticality of energy and mass flows in sustaining the economy. Without such an analytical tool, it is difficult to determine how the economy is performing in physical sustainabil ity terms, e.g. how much waste is being recycled, how industries directly and 1 57 The term PlOT (Physical Input-Output Table) is often used in the l iterature. In this Chapter, the author prefers to use the term physical input-output model or matrix for the reason stated in Chapter 4. Although the description in this Chapter relies on the specification and manipUlation of matrices, these can be directly converted to an equation-based model (e.g. for the type of multiplier analysis carried out in Section 6 .3) . For this reason primarily, the author has tended to refer to a physical input-output model rather than matrix. 1 58 Several other nations have developed economy wide material flow accounts (see, for example, Adriaanse et.a!' ( \ 997) and Steurer and SchUtz (2000)), national resource accounts (e.g. France, Norway and Canada), SEEA accounts or MlPSlFactor 1 0-type (Schmidt-Bleek, 1 994a, 1 994b, 1 994c, 1 997) analyses. These approaches are however not considered comprehensive in the context of this thesis, as they tend to focus on either the economy as an aggregate, or on selected industries within the economy. This differs from the framework utilised here which comprehensively evaluates, on an industry-by­ industry basis, physical flows. 1 59 The physical flows are measured in mass (tonnes), even for energy resources such as coal. This was done in order to be consistent with other PlOT research. Data on energy inputs and land inputs are also collected in this exercise although will not be reported in this Chapter - refer to Appendix K. This latter data (energy, land) is important in understanding the biophysical functioning of the economy and was used in the calculation of Auckland Region's ecological footprint in Chapter 8 . 1 84 indirectly depend on fossil fuel inputs, what the critical connectivity's in the economy are in mass and energy terms, and so forth. Furthermore, because Chapter 6 focuses on an urban region, the model can be seen to be the first operationalisation of the type of city-level physical input-output model first proposed by Geddes ( 1 885) in which he imagined energy and material flows through cities could be depicted and quantified. The structure of the physical input-output model is similar but not the same as the German model developed by Stahmer et al. ( 1 996, 1 997, 1 998). Most notably the model presented in this Chapter of the thesis differs from the German model in that: ( 1 ) a 'materials balance' column is added, and (2) to ensure direct comparability with the economic input-output model, some raw materials that have a market price (e.g. coal), are classified as commodities rather than raw materials as in the German model. 6.1 Generation of an Auckland Region Physical Input-Output Model The methodology used to construct the New Zealand and Auckland Region physical input­ output model builds on the input-output accounting framework conceptualised in Chapter 4 and constructed in financial terms in Chapter 5. The same industry and commodity definitions are employed so that comparisons can be made between the results of the financial and the physical input-output models. The physical input-output model is initially constructed for New Zealand using a combination of data from the financial input-output model, commodity prices, and the insertion of superior ad hoc data. Development of the Auckland Region physical input-output model is based on the prices established in the construction of the national input-output model, supplemented where possible with region specific ad hoc data. With limited access to national statistical agency data, the New Zealand and Auckland Region physical input-output models can be considered to be only tentative prototypes. A significant amount of research funding has been granted by the Foundation of Science, Research and Technology (FoRST), the New Zealand Government's principal science funder, to improve the physical flow estimates presented here at both the national and regional level. The national work falls under the 'Ecological Footprint Plus' Programme (WROX0305), while the regional work falls under the 'Sustainable Pathways' Programme (MAUX0306). These projects consist of representatives from several governmental research institutes and private sector companies within New Zealand, including the New Zealand Centre of Ecological Economics, Landcare Research Ltd, Canesis Network Ltd, Forestry Research Institute Ltd, Massey University, WelNetworks Ltd, Market Economics Ltd and various others. Researchers 1 85 from these organisations will provide superior data, verify existing calculations, and expand the number of commodities evaluated within the framework. 6.1 . 1 Methodological Process in the Auckland Region Study In this Section a methodology is developed that generates the within economy physical flow matrices for New Zealand and, in turn, for Auckland Region (i .e. matrices S , T , U , V , X and Y of Table 4.2). The methodological process consists of a series of 9 steps (Figure 6. 1 ). In steps 1 to 4, the national financial supply-use matrices constructed in Chapter 5 are transformed into the physical input-output model using prices generated from trade statistics. In step 5, superior data from ad hoc sources including Statistics New Zealand, producer boards, published industry reports, industry representatives, and so on is inserted. Calculation of service industry physical flow in step 6 completes construction of the prototype New Zealand physical input-output model . Transformation from the nation to Auckland Region is performed in steps 7 and 8. In step 9 industry and commodity definitions are aggregated to faci litate analytical manipulation and reporting. 1 86 Derivation of the National Physical Input-Output Model Step 1 : Estimation of I nternational Trade Physical Flows Step 2: Selection of I nternational Trade Prices as Surrogates for Domestic Supply-Use Prices Step 3: Estimation of Unknown Domestic Supply-Use Prices Step 4: Derivation of Domestic Supply-Use Physical Flows Step 5: I nsertion of Superior National Data Step 6: Estimation of Service Sector Physical Flows Derivation of the Regional Physical I nput-Output Model Step 7 : Derivation of Regional Physical Flows Step 8: Insertion of Superior Regional Data Step 9: Aggregation of I ndustries and Commodities for Analytical Purposes ---�,... Section 6. 1 .2 ---....,,. .. Section 6. 1 . 3 Figure 6.1 Methodolog ical Process for Deriving New Zealand and Auckland Reg ion Physical Input-Output Models 6. 1 .2 Derivation of a National Physical Input-Output Model Step 1 Estimation of International Trade Physical Flows The conversion of financial flows of internationally traded commodities in the New Zealand economy to physical equivalents was undertaken using prices expressed in tonnesl 60 per NZ 1 998 $ as obtained from Statistics New Zealand's Harmonised System Classification (NZHSC) 1 996. The NZHSC 1 996 is a subset of the internationally recognised Harmonised System Commodity Description and Coding System. The HSC faci litates statistical and administrative comparability of trade information between nations. Traded commodities are c lass ified under approximately 5 ,000 headings in a broad 6-digit structure. The NZHSC disaggregates this structure to a 1 0-digit level comprising just over 1 3 ,000 commodity groups. The fol lowing 160 These are tonnes in net weight terms, i .e . without packaging. All further occurrences of the word 'tonne' in this Chapter represent 'net weight tonnes' - unless stated otherwise. 1 87 information for each NZHSC code is available: ( 1 ) formal descriptor; (2) estimated fmancial value in New Zealand dollars, with exports valued free on board (fob - the value of goods in New Zealand ports before export) and imports valued in both value for duty (vfd) and cost including insurance and freight (Cif) 161 ; (3) gross weight162, and (4) net weight or counts. Prices per tonne were calculated for both imports and exports for approximately 1 3 0 out of the 2 1 0 commodity deflnitions of the commodity-by-industry accounting framework. 163 This required aggregating net weights and traded values (with imports measured c ir64 and exports fob) of the NZHSC 1 0-digit commodity groupings to a 5 -digit level . These, in turn, were matched to the 1 3 0 commodities of the commodity-by-industry accounting framework based on a concordance supplied by Statistics New Zealand. Export prices, px; , expressed in net weight tonnes per NZ 1 998 $, were calculated as px=Xi/Xi , Xi ::;:. 0 , � and import prices, PY j ' in net weight tonnes per NZ 1 998 $, as Step 2 py=i'Y li'Y , i'Y ::;:. O . Selection of International Trade Prices as Surrogates for Domestic Supply-Use Prices (6. 1 ) (6.2) Once import and export prices per tonne had been established for the 1 3 0 commodities, it was then possible to derive estimates of physical mass for each commodity group. This was achieved by assuming price invariance between traded and domestic commodities, i .e. import/export commodity prices per net weight tonne were used as surrogates for their domestic supply/use counterparts. The international export price per tonne for pipfruit, for example, was used as a surrogate for its domestic use equivalent. The implic it assumption of equivalence however requires additional thought. 161 These values are converted from foreign currency when import documents are processed by the New Zealand Customs service. The exchange rates used are set by the New Zealand Customs Service on a fortnightly basis. 1 62 Gross weight includes cargo and packaging, but excludes the weight of re-usable cargo containers. 1 63 The remaining 80 commodity definitions relate predominantly to the provision of services - the physical flows of which are negligible. Furthermore, the physical flows associated with these services may be computed indirectly from their physical inputs (refer to step 5). 164 Imports were measured in cif rather than vfd terms as it is argued that cif costs would need to be equal to, or less than, domestic costs to provide an overseas supplier with sufficient incentive to import goods into New Zealand. 1 88 Question 1 : What justifications exist for assuming equivalence between a trade price and a domestic equivalent? Consider, for example, an importer and a domestic supplier producing a competing commodity. Under perfect market conditions, where all other things remain equal, competition between the price paid for an imported/exported commodity and the price paid for the same commodity suppl ied/used domestically, would exhibit one of the following relationships: • Import price per tonne > domestic supply price per tonne. Consumers of the commodity wil l purchase it from the domestic supplier rather than from the importer. If the importer is to be competitive a reduction in the price to at least the domestic supply price will be necessary. • Import price per tonne < domestic supply price per tonne. Consumers of the commodity will purchase it from the importer rather than the domestic supplier. If the domestic supplier is to be competitive a reduction in the price to at least the import price will be necessary. • Import price per tonne = domestic supply price per tonne. Both importer and domestic supplier operate without price competition. Combating a downward spiral in price is critical to the economic survival of both the domestic supplier and importer. If, for example, the importer sets too Iow a price then they may cease being financially viable and ultimately exit the market. A s imi lar set of behaviours can be established for the competitive relationship between an exporter and domestic use. The relationships outlined here may only arise under perfect market conditions. Markets are however often far from perfect - due at least in part to: • • Tariffs. To protect domestic suppliers, many governments place tariffs on incoming commodities from other nations. The highly deregulated nature of the New Zealand economy, where almost all tariffs have been removed from commodity imports, means that this is unlikely to be a significant factor in the relationship between supply and use price. Nevertheless for New Zealand exporters, where markets are control led by other governments, this will certainly be a s ignificant factor in the setting of price. Seasonality and climate. Seasonal fluctuations in c l imate have a significant impact on trade and domestic prices, particularly for primary producers and, through interdependencies, their associated processing industries. Although technological innovation such as refrigeration can minimise the impact of seasonality and c limatic fluctuations, some industries remain susceptible to seasonal and c limatic influences. In 1 89 the context of this thesis, these impacts wil l be averaged out over a year. Furthermore, the more urban economic profile of Auckland Region means that the impact on local primary producers is not considered to be great, although import substitution by processing/manufacturing may be greatly affected. • Business decision making, size and ownership. Business decisions, such as purchasing commodities while prices are low or in advance through pre-ordering, can result in oscillation around a commodity's optimal price. Similar oscillation may also result from time lags between purchase and use of a commodity. The size of a business may also be a factor in price setting - a substantial bulk-buy may warrant a lower than average price as risk and uncertainty over future price fluctuations can be reduced. Ownership patterns may also affect price as parent industries may purchase from subsidiaries at higher prices simply because of the security of knowing that orders can be filled and production time l ines adhered to. There are also several potentially significant operational weaknesses associated with the price invariance assumption. These are considered below by exploring three further questions: Question 2: What if exports in a particular commodity grouping exhibit a significantly different mix of sub-commodities than those used domestically? Differences in the mix of sub-commodities (i.e. at a disaggregation below the 1 30 commodity groups) between traded and domestic commodities certainly exist. Unfortunately, a paucity of data on domestic commodity consumption and production patterns restricts any analysis of the implications of these differences . Nevertheless, the relationship between NZHSC 5 -digit headings and the input-output commodity groups was analysed. Of a possible 1 28 commodity groups with export prices per tonne only 29 commodity groups were used as surrogates for their domestic use equivalents. Of these 29 commodity groups, one-to-one matches existed in 3 cases ( 1 0 .3 percent) between NZHSC 5 -digit headings and the commodity group definitions, and a further 1 8 cases (62. 1 percent) were dominated (> 50 percent of commodity group financial value) by a single NZHSC 5-digit heading. Similarly, of a possible 1 26 commodity groups with import prices per tonne, some 61 commodity group prices were used as surrogates for their domestic supply equivalents. Of the 6 1 commodity groups one-to-one matches existed in 3 cases (4.9 percent) between NZHSC 5-digit headings and commodity group definitions, with a further 24 cases (39.3 percent) being dominated by a s ingle NZHSC 5-digit heading. Overall, these results are considered favourable. Further research is, however, required at the sub-commodity group level on domestic consumption and production patterns to improve these findings. 190 Question 3: How much of each commodity 's output (input) is destined for export (import)? Generally speaking the greater the contribution made to commodity output by exports, the more likely the domestic use price per tonne will reflect the export price per tonne. In those cases where, say, an export price per tonne for a commodity was used as a surrogate for its domestic use equivalent, the share that exports made up of commodity output, a , was calculated (Table 6. 1 ) . The Table shows that 14 commodities (48 .3 percent of all commodities) were based on surrogate prices per tonne where exports made up 40 percent or more of total commodity output, and 23 commodities (79.3 percent) with 20 percent or more of total commodity output. Table 6.1 International Exports as a Percentage of Commod ity Output, 1 997-98 Commod ity Group 1 8 Other livestock 48 Other dairy products 49 Prepared fish 2 Pipfruit 42 Meat and meat products 64 Natural textiles 72 Tanned skins and leather 27 Crustaceans 90 Other chemical products 3 Kiwifruit 24 Other forestry products 65 Cotton textiles 45 Hides and skins 6 Plants, flowers, seeds 53 Oils and fats 21 Forestry and logging 47 Yoghurt, buttermilk, icecream 89 Industrial chemicals 84 Non metal wastes and scraps 103 Metal wastes 1 2 1 Machinery for food production 52 Prepared fruit and nuts 79 Veneer sheets and plywood 70 Carpets 7 Raw vegetable materials 1 9 Other animal products 80 Builders joi nery 62 Spirits, wines, beer, tobacco 26 Fish Note: 1 . Values are 1 998 New Zealand dollars. International Exports as a Percentage of Total Price Use % $/tonne' 92.6 73,640 8 1 .9 3 , 1 1 8 78.9 2,648 77 . 1 1 , 1 26 68.6 3,437 67.9 4,580 67.2 7,653 65.0 22,365 58.6 2 ,870 58.2 1 ,576 48.7 1 2,354 48.0 8,046 45.5 7,296 40.7 9,374 39.6 927 35.4 1 22 32. 1 2,439 3 1 .5 273 28.2 221 27.5 901 25.8 1 5,324 25.8 2 , 100 22.7 1 ,336 1 6.6 8 ,916 1 4.5 2,902 8.7 5,841 8.2 2,509 7 .2 1 ,9 1 5 6 .5 7,091 If an import price per tonne for a commodity was used as a surrogate for its domestic supply equivalent, the share that imports made up of commodity input, a I , was estimated (Table 6.2). The Table shows that 34 commodity groups (55 .7 percent) were based on surrogate prices where imports made up 40 percent of more of commodity input, and 49 commodity groups (80.3 percent) with 20 percent or more of commodity input. 1 9 1 Table 6.2 I nternational Imports as a Percentage of Commodity Input, 1 997-98 International Imports as Commodity Group a Percentage of Total Price Supply 0/0 $/tonne' 1 3 Unmanufactured tobacco 1 00.0 6 ,876 22 Natural gums 1 00.0 2,900 35 Chemical and fertilizer minerals 1 00 .0 1 1 0 1 26 Audio and video records and tapes 99.1 47,553 12 Beverage and spice crops 95.8 4,303 1 34 Games and toys 94.8 1 4,643 1 1 0 Engines 88.7 38,782 1 08 Steam generators 88.0 7 ,432 1 1 3 Ships 85.5 40, 1 25 1 27 Watches and clocks 84.2 75,932 1 1 6 Aircraft and parts 82.0 1 , 1 49,852 1 29 Photographic and scientific equipment 75.6 46,921 1 24 Computers and parts 75.2 96,084 5 Oil seeds 74.9 1 , 576 1 32 Musical instruments 74.3 26,306 68 Woven fabriCS 68.2 1 3, 1 39 1 20 Machinery for mining 65.5 7,791 66 Man-made fibres and textiles 63.2 4, 732 86 Prepared printing plates 62.4 1 8,601 1 1 7 General industrial machinery 61 .6 1 8 , 1 23 1 33 Sports goods 57. 0 1 0,790 76 Footwear 54.5 1 8,020 75 Handbags and articles of leather 53.7 1 5,569 95 Pharmaceutical products 52.0 32,399 1 1 1 Motor vehicles and parts 5 1 . 0 8,599 4 Other fruit and nuts 49.1 1 ,242 92 Rubber 47.7 4,945 1 28 Medical equipment 44.2 1 07,730 1 1 5 Other transport equipment 43.6 9 , 2 1 6 1 23 Office equipment 43.6 29,666 1 02 Other mineral products 43.3 2,546 98 Pesticides 43.0 1 0,386 136 Other manufactured articles 42.6 1 7,796 9 1 Plastics 42.0 3 , 0 1 3 1 25 Electric equipment 39.3 26,471 96 Soap and perfumes 36.7 4 ,879 69 Other textiles 34.5 1 0,563 99 Glass and glass products 34.4 1 , 1 27 1 3 1 Jewellery 33.5 1 ,392,689 74 Clothing 30 9 26, 506 1 1 2 Coachwork 30.6 5,334 1 09 Other fabricated metal products 29 .2 6,2 1 3 1 1 9 Agricultural and forestry equipment 28.8 1 0, 1 97 1 06 Structural metal products 26. 1 1 , 458 93 Rubber tyres 25.2 4,81 6 82 Other wood products 25. 1 5,765 61 Other food products 24.0 4,743 94 Paints 23.2 7 ,080 88 Petroleum products 2 1 .4 380 1 07 Tanks, reservoirs and containers 1 8.8 5 ,089 1 1 4 Pleasure and sporting boats 1 5.6 20,383 1 22 Domestic appliances 1 4.8 1 0,590 85 Books and stationery 14 .4 1 3 ,274 1 30 Furniture 1 1 .7 5,045 1 76 Computer software and services 1 1 .0 3 1 ,221 8 1 Wood containers 9.7 8,335 37 Precious metals and stones 9.4 431 71 Twine, rope, netting 9.0 7 ,689 87 Newspapers and journals 8.8 1 6,598 1 35 Prefabricated buildings 8.6 2,7 1 7 1 04 I ron and steel 0 .8 757 Note: 1 . Values are 1998 New Zealand dollars, 1 92 Question 4: Is it reasonable to assume price invariance across all industries that supply or use a commodity? The use of trade prices per tonne as domestic supply/use surrogates assumes price invariance across all industries either purchasing a commodity, in the case of domestic use, or selling a commodity, in the case of domestic supply. This assumption is, however, certainly incorrect. It is likely, for example, that a company making apple juice concentrate wil l bulk-buy apples for pulping, at a heavily reduced price, while a local superette (cornershop) will pay a higher premium. In these cases, where the discrepancy is thought to be great, superior data (refer to step 5 ) should be inserted and prices in the remaining industries adjusted accordingly (refer to steps 3 and 4). Overall, the trade prices utilised here represent annual average and, as such, price differences between imported and domestic commodities will , to some degree, have been ironed out. Moreover, steps 4 and 8 of the methodological sequence provide significant opportunities for the insertion of superior data to alleviate as many of the discrepancies above as possible. Step 3 Estimation o/Unknown Domestic Supply-Use Prices Once it was decided whether an international import or export price per tonne would be used to convert domestic supply or use to a physical mass, it was then a matter of algebra to determine the remaining domestic supply or use price per tonne. If, for example, domestic supply had been determined from an export price per tonne then domestic use was algebraically determined so as to ensure that supply and use of a commodity equated, both in financial and physical terms. 165 To compute the unknown domestic supply or use price per tonne using surrogate trade prices per tonne, it is useful to set out the fol lowing equations: my=i'Y , (6.3) where my j represents the physical mass of the internationally imported commodity j, mv=i'V, (6.4) 1 65 Equations 4.9 and 4 . 10 have previously established that in financial terms the supply of a commodity, ai, must be equal to the use of that commodity, ai' where i=j. Physical equivalence is similarly establ ished in Equations 4 . 1 8 and 4 . 1 9 . 1 93 where mv J denotes the physical mass of domestically supplied commodity j, mx=Xi , (6. 5) � where ;;:;;, is the physical mass of internationally exported commodity i, and mu ;::: (UISIT)i , (6.6) r--. where ;; , describes the physical mass of domestically used commodity i. Together Equations 6.3, 6.4, 6. 5 and 6.6 may be used to formulate physical mass balance for commodity i and), where i =), (6. 7) Similarly financial balance (Equation 6 .10) may be established with the use of Equations 6. 1 and 6.2 and the following two price equations, pv=i'(V)/i' (V) , i'(V) *' 0 , (6.8) r----. where pVJ is the domestic supply price per tonne for commodity), pu :::: (UIS IT)/(UIS IT) , (VISIT) 7= 0 , (6 .9) r-o. where pu, represents the domestic use price per tonne for commodity j, and (6. l 0) Having establ ished financial and physical mass balance respectively in Equations 6. 1 0 and 6.7, and by substitution of the domestic supply price per tonne for commodity) with its international import equivalent, J; J :::: Py J ' the physical mass of this commodity may be derived, 1 94 � . '(V) I � mv = -=-, py *- 0 , py ,........, as may the domestic use price per tonne for commodity i, pui , where i=j, PYi (mYj + mVj ) = PXi mXi + pU, rnu, . By rearrangement, � = PYj (mYj + mvj ) - px/ mXi --- PUj = � , mu, *- 0 , mU j ___ r---"o.. ,---... """""-"" and by substitution of mu, for my j + mv j - ;;;;i as per Equation 6.7, � = Py/mYj + mvj ) - pxi mxi � � � pu, = � ,........, ,........, , my j + mv j - mXi *- 0 . my j + mv j - mXi (6. 1 1 ) (6. 1 2) (6. 1 3 ) Of course, an unknown domestic supply pnce per tonne for commodity j, pv j ' may be determined by substitution of the domestic use price per tonne for commodity i, pu i ' with its international export equivalent, Pxi ' using a similar set of equations. 1 66 Step 4 Derivation of Domestic Supply-Use Physical Flows The physical mass of commodities supplied and used domestically may be established using the prices per tonne generated in step 3 . For example, the physical mass of each coefficient in the domestic supply matrix, V , may be derived in the following manner: (6. 1 4) 166 In some circumstances the import, export, domestic supply or domestic use price, or any combination of these, could be zero. Equations may be adjusted accordingly. 195 S imilarly, the physical mass of each domestic use coefficient, as recorded in matrices S, T and U, may also be ascertained. The physical mass of each element, U u , in the domestic use matrix may, for example, be calculated as Step 5 � V � VI} = � , pu, 7: 0 . pu, Insertion of Superior National Data (6. 1 5) Trade-based prices per tonne were not solely used to determine domestic physical flow. Some 46 commodity groups, consisting predominantly of the larger physical flows in the New Zealand economy, were determined from the insertion of superior ad hoc data taken from various sources including Statistics New Zealand, producer boards, industry and academic publications. These superior data insertions are summarised in Table 6.3. It is worth noting that rather than being based on prices per tonne of commodity, these sources tended s imply to record total tonnage produced. Adjustments were required in some cases to take account of already determined import and export tonnages, any moisture content that had been removed from the commodities, and other lesser issues. Furthermore, data was used in some cases to verify physical flows calculated from the trade-based price per tonne estimates generated in steps 2 and 3 . Table 6.3 Superior Data Inserted into the New Zealand Physical Input-Output Model, 1 997-98 Commodity Group 6 Sheep 9 Cattle 1 0 Wool 1 1 Grain and other crops 14 Raw milk 15 Pigs 16 Poultry 1 7 Deer 23 Standing timber 30 Coal 31 Metal ores 32 Building stone 33 Gypsum and limestone 34 Sand, pebbles, gravel, clay 36 Salt 40 Crude petroleum and natural gas 43 Poultry products 44 Bacon, ham and smallgood products 46 Processed milk and cream Price Data Sources $/tonne' 1 ,8 1 7 SNZ ( 1 998d, p.42) Burtt ( 1 999, pp.A-1 2 to A-15) 2 , 1 32 Meat and Wool Economic Service of New Zealand ( 1 999, pp. 1 9,27) 4,469 New Zealand Wool Group (n.d.) 252 Food and Agriculture Organisation of the United Nations (n.d.) Burtt ( 1999, pp. A-55 to A-64) 406 Fonterra Co-operative Group (n.d.) Burtt ( 1999, p.A-30) 2,700 Burtt ( 1999, p.A-52) SNZ ( 1 999d, p.42) 4,250 Poultry Industry Association of New Zealand Inc. (n.d.) Burtt ( 1999, p.C-28) 5,764 Ministry of Agriculture and Forestry ( 1 998) 39 Ministry of Agriculture and Forestry ( 1999, 2000) 92 Dang ( 1 999, p.3 1 ) 1 3 Christie, Brathwaite and Thompson ( 1993, pp.20-23) Ministry of Economic Development ( 1998) New Zealand Mining ( 1 998, p.1 1 ) 246 Christie, Douch, Winfield and Thompson (2000, pp.1 5-25) New Zealand Mining ( 1998, p. 1 1 ) 1 3 Christie, Thompson and Brathwaite (2001a, pp.6-25) New Zealand Mining ( 1998, p. 1 1 ) 16 Christie, Thompson and Brathwaite (2000, pp.26-43) Christie, Thompson and Brathwaite (2001 b, pp.6-26) New Zealand Mining ( 1 998, p. 1 1 ) 332 Hogan and Williamson ( 1 999, p.287) 257 SNZ Infos Series NRGASCR7ZM and NRGA.SGP3AM 3, 1 1 2 SNZ (2000, p. 1 8) 5,784 SNZ Infos Series PRPH.SAACD, PRPH.SAACE, and PRPH.SAACF 1 , 359 SNZ (2000, p. 1 8) Notes Weighted average $ per tonne for ewes, rams, wethers and lambs Weighted average $ per tonne for beef cattle, breeding cows, dairy cows and heifers Weighted average $ per tonne for barley, wheat, maize and oats Milk solids were multipl ied by a conversion factor of 8.42 to convert into wholemilk (M.G. Patterson, pers. comm. , November 26, 2002) Weighted average $ per tonne for weaners, stores, porkers, sows and choppers Weighted average $ per tonne for weaners, velvet stags and breeding hinds Change in standing timber stock. Conversion from m3 to tonnes required an adjustment for the density of Pinus radiata based on the U.S. Department of Agriculture ( 1974) and Patterson ( 1 980) Weighted average $ per tonne for ironsand and smaller quantities of various other metal ores for 1 996. Prices adjusted from 1 996 to 1 998 using SNZ's Mining PPI (Outputs) Price adjusted from 1 996 to 1996 using SNZ's Mining PPI (Outputs) Weighted average $ per tonne for limestone and dolomite for agriculture, roading and industrial use Weighted average $ per tonne for sand, gravel, clay and rock for various activities including bricks, ceramics, pottery, roading, reclamation and industrial use. Price adjusted from 1 996 to 1998 using SNZ's Mining PPI (Outputs) Natural gas converted from T J to tonnes using a ratio of 1 9.9 metric tonnes C per T J (M.G. Patterson , pers. comm. , November 26, 2002) Table 6.3 Superior Data I nserted into the New Zealand Physical Input-Output Model, 1 997-98 (Continued) Commodity Group 50 Prepared vegetables 51 Fruit juices 54 Grain products 55 Starches 56 Animal feedings 57 Bakery products 56 Sugar 59 Confectionery 60 Macaroni and noodles 63 Soft drinks, bottled water 67 Yarn and thread 73 Knitted fabrics 77 Wood 78 Panels and boards 63 Pulp, paper and paperboard 97 Fertilisers 1 0 1 Articles of concrete and stone 1 40 Water Notes: I . 1 998 New Zealand Dol lars. 2. Census of Manufacturing. Price Data Sources $ltonne' 1 , 1 32 New Zealand Vegetable and Potato Growers' Federation Inc. (n.d . ) 1 ,670 Statistics New Zealand ( 1 996a) 1 ,234 Department of Statistics ( 1 96 1 ) 1 ,077 Department o f Statistics ( 1 96 1 ) 566 Department of Statistics ( 1 96 1 ) 2,665 Department of Statistics ( 1 96 1 ) 761 Department of Statistics ( 1 96 1 ) 6,373 Department of Statistics ( 1 96 1 ) 2,596 Department o f Statistics ( 1 96 1 ) 2,1 1 1 Statistics New Zealand Infos Series SEPQ.SABBA, SEPQ.SABBC, SEPQ.SABBD, SEPQ.SABBE and SEPQ.SABBF 1 7, 1 60 Statistics New Zealand Infos Series SEPA.SYZ 1 2,064 Statistics New Zealand Infos Series SEPA.SATTD 4 1 5 Statistics New Zealand Infos Series FL TA. SBEA3 563 Statistics New Zealand Infos Series FL TA.SMAA, FL TA.SMBA and FLTA.SBDA 1 ,022 Statistics New Zealand Infos Series FL TA.SPFA and FL TA. SPGA 254 Statistics New Zealand Infos Series MAGQ.SAB 221 Statistics New Zealand Infos Series SEPA.SAFRZM 0.77 McDermott Fairgray Group Ltd (1 996) McDermott Fairgray Group Ltd and Massey University ( 1 999) Patterson and McDonald (2004) Notes Weighted average $ per tonne for various fruit juices including orange, apple, currant, grape and combinations Weighted average $ per tonne for various bakery products recorded in the 1 976-79 CoM2. Prices adjusted from 1979 to 1 996 using SNZ's CPI for Grain Products Weighted average $ per tonne for various bakery prod ucts recorded in the 1 976-79 CaM. Prices adjusted from 1979 to 1 996 using SNZ's CPI for Sugars and Sweeteners Weighted average $ per tonne for various animal feeds recorded In the 1976-79 CaM. Prices adjusted from 1979 to 1 996 using SNZ's Food, Beverage and Tobacco PPI (Outputs) Weighted average $ per tonne for various bakery products recorded in the 1 976-79 CaM. Prices adjusted from 1979 to 1 996 using SNZ's CPI for Bakery Products Weighted average $ per tonne for refined sugar and molasses as recorded in the 1 976-79 CaM. Prices adjusted from 1979 to 1996 using SNZ CPI for Sugars and Sweeteners Weighted average $ per tonne for various confectionary products recorded in the 1 976-79 CaM. Prices adjusted from 1979 to 1 996 using SNZ CPI for Sugars and Sweeteners Weighted average $ per tonne for noodles, macaroni, spaghetti and vermicelli as recorded in the 1 976-79 CaM. Prices adjusted from 1 979 to 1 99B using SNZ CPI for Pasta and Pastry Conversion from m' to tonnes required an adjustment for the dens�y of sawn timber based on the U.S. Department of Agriculture (1 974) and Patterson (1 960) Conversion from m' to tonnes required an adjustment for the density of plywood and panels based on the U.S. Department of Agriculture ( 1 974) and Patterson ( 1960) Reticulated household water use calculated as weighted average m' per person per day for Waitakere, North Shore, Auckland and Manukau cities. Scaled from c�les to Ihe nation based on population estimates. Reticulated industrial use calculated m' per FTE for Northland, Auckland and Waikato region industries. Scaled from regions 10 the nation based on FTE estimates 198 Step 6 Estimation of Service Sector Physical Flows Of a possible 2 1 0 commodities, some 80 represent service commodities. Generally speaking, the provision and use of service commodities results in negligible physical flows - l imited mainly to paper flow. Several notable exceptions do however exist inc luding takeaways, computer software and photographic services. Such physical flows were estimated form ad hoc data obtained from various statistical, industry and academic sources. The physical flow of takeaways, for example, was determined by taking a weighted average price per tonne of commonly consumed fast foods. Paper flow associated with the delivery or use of a service commodity is typically informational (e.g. written reports), administrative/accountancy based (e.g. invoices, receipts, purchase orders) or advertising/packaging related (e.g. envelopes, cardboard boxes). Thus, paper flow is a by­ product of service provision or use. Crude estimates of service industry paper flow were generated by assuming that a relationship exists between the mass of the paper and paperboard products purchased by the service industry, on the one hand, and the fmal paper outputs delivered, on the other hand. Paper inputs were transformed to paper outputs by multiplying paper inputs by a wastage percentage obtained from ad hoc sources. Further research into other possible approaches or minimally surveying of service industry paper waste flows is, however, required to substantiate/improve the estimates generated. Commodity 2 1 0 of the national financial supply-use framework represents direct purchases abroad by residents. It comprises primarily the purchases made by New Zealanders whi le travelling abroad. The physical mass of such purchases was crudely estimated by assuming industry purchase patterns overseas mirrored purchase patterns domestically. In this way, the physical mass of commodity 2 1 0 purchased by a particular industry was determined by spreading the financial value, on a pro-rata basis, over all other purchases made by the industry, and applying their price per tonne figures. 6.1.3 Derivation of an Auckland Region Physical Input-Output Model Step 7 Derivation of Regional Physical Flows Crude estimates of physical flows in the Auckland Region economy were generated by assuming that national prices per tonne, both for domestic consumed/produced and traded commodities, were spatially invariant. In other words, Auckland Region prices per tonne for a commodity were set to be equivalent to national prices per tonne. In mathematical terms: px, = px" PY) = PYj ..-...r ,.--... r 199 and either pUi was assumed to be equal to PU" or pVj equal to pv) . In the case of interregional trade, the price per tonne for interregional imports was assumed to be equal ,r-.. � to pv) , while the price per tonne for interregional exports was assumed to equate to pu, . The ,..---.... r ,..--.." r remaining unknown pUt or pVj were calculated at the regional level in an analogous manner to the procedure outlined in step 3 . In this way, the Auckland Region supply and use matrices developed in Chapter 5 were transformed into physical equivalents by dividing financial flows by price per tonne figures. Step 8 Insertion of Superior Regional Data Once again an opportunity was provided for the capture of any ad hoc superior data that existed at the regional level. Under ideal conditions superior regional data should be inserted when the regional supply or use price per tonne differs significantly from its national equivalent. Such situations may arise with differences in business management practices (e.g. a business may prefer to purchase commodity inputs from a subsidiary at a higher price than a local producer), business size or scale (e.g. the presence, or absence, of economies/diseconomies of scale), business location (e.g. proximity to consumer markets, production networks, transportation hubs, natural resources), and so on. In the few instances where primary data actually existed for Auckland Region, this was inserted into the physical supply-use matrices. Estimates of reticulated water use by Auckland Region businesses and residents, as obtained from metered recordings, for example, replaced the less accurate national price per tonne figures. Time constraints however prohibited a comprehensive l iterature search, surveying or interviewing of Auckland Region businesses. Despite a paucity of region-specific data, the estimates generated enable tentative results to be obtained. Step 9 Aggregation of Industries and Commodities for Analytical Purposes The final step in constructing the New Zealand and Auckland Region commodity-by-industry physical input-output model was the aggregation of the commodities and industries. Aggregation was undertaken for reasons of ease of analytical manipulation and reporting. The physical commodity-by-industry matrices were aggregated from 2 1 0 commodities by 1 23 200 industries to 48 commodities by 48 industries, and are available as Excel files ( 'New Zealand Physical Commodity-by-Industry Model.xls' and 'Auckland Region Physical Commodity-by­ Industry Model.xls ') in the Chapter 6 directory of the accompanying CD-ROM. These matrices were further aggregated to 3 commodities by 3 industries and are presented in Appendix H. 6.1.4 Derivation of Raw Material and Residual Flows for the Auckland RegionlNational Economy The methodology created above to estimate the physical mass flows within the New Zealand and Auckland Region economies assures commodity mass balance, i .e . the mass of each commodity supplied equal the mass of the commodities used (or stored). At an industry level, however, mass balances must also exist, i.e. the mass of commodities inputs into an industry must equal the mass of commodities outputs (included those placed in storage) produced by that industry. This requires consideration of commodities not normally valued in economic production, namely : ( 1 ) raw materials that flow from the environment to the economyl67 (e.g. nutrients, water, oxygen and carbon dioxide), and (2) residuals that flow from the economy to the environmentl68 (e.g. wastes, pollutants, emissions). 1 69 These physical flows are represented - - - "" - - -- "..., - - in Table 4.2 by matrices C, D, E, F, G, H, K , L, 0, P and Q . There are currently no internationally recognised c lass ifications systems or guidel ines governing the recording of the physical flow of materials/residual associated with economic activity. The SEEA system is perhaps the most advanced, but is not comprehensive in its coverage or implementation. Other possibilities include systems utilised in economy-wide material flow accounting (e.g. Adriaanse et al. ( 1 997) and Steurer and Shtitz (2000)), national resource accounting (e.g. the French Patrimony Accounts, Norwegian Natural Resource Accounts), MIPSlFactor 1 0 type analyses (Schmidt-Bleek, 1 994a, 1 994b, 1 994c, 1 997), and GermanlDanish PlOT models. Given that the physical flow accounting framework presented in Chapter 4 is comprehensively similar to the German PlOT, the major categories of the latter are adopted to c lassify the physical flows of Auckland Region's environment-economy interface (Table 6.4). 1 67 Residuals, particularly construction or demolition waste, may also flow from the environment to the economy. 168 Raw materials, such as livestock, may also be considered to flow from the economy to the environment. 1 69 Raw material and residual inputs flowing from the environment to the economy are sometimes recorded as 'free goods of nature' . This terminology is however considered inappropriate in this context as raw materials/residuals are included irrespective of whether they are provided free by nature or traded through markets. Coal, for example, requires two commodities to describes its physical flow: ( \ ) in­ ground coal (which has no price), and (2) mined coal (which has a price). Although subtle this distinction is important. Table 6.4 Classification of Raw Materials and Residuals Major Category Raw Materials Soil excavation for structures Water raised Oxygen Carbon dioxide Other air components (nitrogen etc.) Residuals Wastes for economic re-use Wastes for treatment, storage Category Sub-category Solid waste Recycled solid waste Solid waste Landfill solid waste Item Water tal n); (2) the ' U - V' matrix structure means that inputs are represented by negatives and outputs by positives, unlike the Leontief formulation; (3) the column totals represent the net outputs or net inputs from the Auckland Region system, enabling us to determine whether there is a flow of a particular quantity across the systems boundary and in what direction; (4) the row totals (with respect to energy and mass) should be zero if there is energy mass balance in a process, as there must be according to First Law of Thermodynamics and the Mass Conservation principle (refer to Chapter 2). That is, the mass of reactants must equal the mass of the products for each of the biogeochemical processes 229 in the input-output model - refer to Appendix B; and (5) unlike many biogeochemical models, the biogeochemical cycles (carbon, hydrogen, phosphorus, sulphur and nitrogen) are coupled and interdependent as invariably at least one of the reactants and/or products is from different cycles. Quantities '" '" '" '" '" "0 :J "O "0 "0 c: � c: c: c: :J o :J :J c: :J c: 0 .c. 0 '5 0 Cl> 0 >. o a. 0. 0. .c. a. "' a. Q:; e> -e E :g E .9- E e E � Cl> '" 0 .c. 0 :J 0 ;� 0 c: U U a.. u (fl U Z U w Carbon Cycle oxidation of land humus volcanic action land plant respiration formation of limestone weathering of limestone igneous rock formation Phosphorus Cycle weathering of sedimentary P rock soil to ocean P sediments P runoff : dissolution of atmospheric P dust particles P sedimentation sedimentary P rock formation Sulphur Cycle release of H25 '" ocean spray of 504 Cl> '" uptake of S02 by ocean '" Cl> u e a.. sedimentary 5 rock formation uplift and weathering of sedimentary 5 rock H25 dissolution Nitrogen Cycle reduction of N20 oxidation of N2 marine dentrification N runoff sedimentary N rock formation weathering of sedimentary N rock Hydrological Cycle transpiration evaporation from ocean precipitation to ocean : uptake of H2O photooxidation of H2 river discharge Net InpuUOutput2 Figure 7.1 Structure of the U - V Matrix of Ecological Flows in the Auckland Region. In the matrix, a negative represents a process input, and a positive represents a process output. A column total represents the net input or output into Auckland Region in the reference year. 230 For ease of presentation, even though each cycle is connected with the others, there is a separate treatment for each of the five major cyc les in Sections 7 . 1 .3 to 7 . 1 .7 of this thesis. Each cycle, as outlined in these Sections, is accompanied by a diagrammatic (Figures 7 .2 to 7.6) and a numerical representation (Tables 7.2 to 7 .6). In the Figures, the arrows represent processes and the rectangular boxes represent quantities. Each Table describes a biogeochemical cycle in a numerical format which is compatible with the accompanying Figure for that cycle. As Table elements are expressed in net terms (i .e . outputs - inputs) it is possible that a Table cell may appear blank, but in fact be populated with input and output values that cancel each other out. Mass balance is reported in the Table row labelled 'Total ' ; as inputs must equate to outputs this will be zero. The Figures and the Tables differ in one further respect: the values in the Figures represent only the fluxes of the 'marker' element (refer to Appendix B for further details), while the values in the Tables represent total mass flux. 7. 1 .3 Auckland Region Carbon Cycle Module The carbon cycle for the Auckland Region is depicted by Figure 7.2 . Table 7.2 describes the total elemental fluxes of the carbon cycle in a numerical format which is compatible with that outlined by Figure 7.2 . Slow Carbon Cycle Biogeochemists such as Bolin et al. ( 1 979), Revel le ( 1 982), Degen et al. ( 1 984), Schlesinger ( 1 99 1 ) and Holman ( 1 992) have separated the carbon cycle into s low and fast sub-components. The slow carbon cyc le is primarily inorganical ly control led. Reservoirs of sedimentary rocks, such as limestone or calc ite, are augmented through a series of processes that convert CO2 released into the atmosphere ultimately into sedimentary rock. Dissolved calc ium, magnesium and bicarbonate ions are transported via freshwater systems to the oceans (Ayres, 1 992, 1 996), and are prec ipitated through ocean layers or extracted and temporarily incorporated by marine organisms into their skeletons or shells. Over time the inorganic precipitates and shells/skeletons, comprising 34 kt C per year in the Auckland Region, settle on the ocean floor as sediments and, under pressure, utilising 1 1 kt C, are converted to sedimentary rocks. In reverse, the carbon in the sedimentary rocks may be released through volcanic action or tectonic uplift and, once again over time, broken down, releasing 27 kt of C per year through chemical weathering. The geochemical processes of the s low carbon cycle are however too slow to account for the hypothesised relationship between CO2 production and global warming. 23 1 Fast Carbon Cycle The fast carbon cycle is mainly controlled by biological processes - it also incorporates the largest carbon fluxes. Through the process of photosynthesis, CO2 is b iologically transformed into carbohydrates i.e. gross land/marine production, utilising an estimatedl8 1 1 7,368 kt C and 1 ,252 kt C respectively. Operating in reverse is the aerobic process of land/marine respiration, releasing 3 , 1 39 kt C and 368 kt C respectively . 182 Photosynthesis, and sorption by the ocean of 2, 1 7 1 kt C, are the major processes involved in removing CO2 from the atmosphere. I S3 Other lesser land-based fluxes include soil humus formation, partially by way of consumption of land producers l 84, uti lising 1 ,573 kt C. Subsequent oxidation of soil humus releases CO2 containing 1 ,402 kt C into the atmosphere. Soil humus may also be transferred by freshwater systems to the ocean and accounts for 1 2 kt C . S imilarly, lesser marine-based fluxes encompass marine plant consumption of 1 2 kt C, formation of marine biomass accounting for 8 kt C (consumers and decomposers) and 724 kt C (producers), and in turn, its oxidation, releasing 645 kt C. Of course, the fast carbon cycle is not solely limited to these fluxes . A small reservoir of organic carbon exists in kerogen, which through heat, geological and biological processes is transformed into coal (0.26 kt C), o i l shale and petroleum (Ayres, 1 996; Smil, 2002). Human combustion of fossil fuels is primarily responsible for the re-release of atmospheric CO2 from these sources . 181 As discussed in Section 7 . 1 .2, all the data in Sections 7. 1 .3 - 7. 1 .7 are crude estimates based on scaling down global data to the regional level. Therefore, all the numbers quoted in these Sections of the text should be treated as such, even though the author has desisted from explicitly stating this for the sake of brevity. 1 82 The net product of gross production and respiration is referred to as net photosynthesis, while net photosynthesis less respiration of non-photosynthetic p lants is known as ' net primary production' (den E lzen et al. , 1 995 ; Ayres, 1 996). 183 Atmospheric CO2 content exhibits a significant seasonal oscillation, due mainly to differences in ( 1 ) the uptake o f CO2 by producers, and (2) human combustion offossil fuels. 1 84 Consumers and decomposers typically have a combined b iomass of less than 1 percent of producers, but are able to reoxidise more than 99 percent of net primary production into CO2 or bicarbonates (den Elzen et al. , 1 995 ; Ayres, 1 996). absorption of CO 2 17 1 kt C 2 t ! � tropospheric oxidation of CO 47 kt C I � oxidation of CH. 49 kt c Atmosphere CO2 0 Atmosphere CH. Atmosphere NEC :;;! 0 N C 0 I 0 0 :Si :Si "" '" u '" � � 0 '" .0. .... "'-(/) :ll c 0 cS � � 13 C -0 C :J U ro o c 0 -0 j)l"oduction of C O 3 kt c 0 Cl. s: ro � e - () I ID -0 '" (/) :J a. (/) c '" (/) -0 ro () ro �- E? E? ID - '" � :;;! 0 Ol a. 0 :;;! 0 marine consumption 12 kt C • N :Si c 0 Land Consumers and .... Marine Consumers � 0 "- Land Plants Marine Plants U Decomposers 0 and Decomposers (/) land consumption 34 kt C U 0 ro :J :;;! () E 0 '" ·c :J I c I N rl .<:: c (/) .... -0 0 .2 marine humus formation :J � (5 .� e- E .90 () 0 > c land humus formation :J c l< :l< � E 0 (c&d) 8 kt C 0 0 (c&d) 14 kt c (/) .<:: 0 co :;;! N .E 0) :;:::-' :ll '" c ro " "'-c (/) 0 ·c E c '" .2 :J :Si ro � 0 0 c ro E '" E .E � 0 -0 :J .... U U ·x .<:: '" .0. ro :J 0 -0 �- (/) U -0 C � QJ � 0 e � .90 c C a. Land Humus Marine Humus .� c transfer of land humus 12 kt C ro 0 ID E 0 a. U c 0 :;;! ID :J .� '" C -0 E I T c ;g .� e (/) 0 a. (/) 'iii (/) E e coal formation 0 kt C kerogen formation -0 kt C :J -0 E Ol 'x :J weathering of o .r: limestone 27 kt C - Igneous Rock igneous rock Fossil Fuels Sedimentary Rock Ocean NEC formation 0 kt C formation of l imestone 1 1 kt C Figure 7.2 Auckland Region's Carbon Cycle, 1 997-98 Table 7.2 Input-Output Model of the Carbon Cycle Processes for Auckland Region, 1 997-98 Carbon Processes 0 u c c E c 0 c :::> 0 � 0 0 .r: :;::; cS '13 I � C "0 e "0 :::> 0 I E g c c .0. ·x cS U "0 U U C. .!!! 0 :::> c co C '0 0 c 0 c 0 0 E � .!!! ti 0 c U :::> 0 � ·c 0. :::> � "0 U "0 a. 0 > .!!! co Cl) a. 0 0 0 N ;;; ""' N C ",-ro c Land Plants Cl> ro transpiration 1 04.64 Mt H t> Cl> 0 t> 0 E .8 0 I .):: C C 0 0 � � .0. 0 ·u Cl. � ro > Cl. Cl> uptake of H20 104.64 Mt H Freshwater H2O Ocean H2O -" river discharge '" 0 Mt H � Figure 7.3 Auckland Region's Hydrological Cycle, 1 997-98 Table 7.3 Input-Output Model of the Hydrological Cycle Processes for Auckland Region, 1 997-98 Hydrological Processes c co C "0 m u co c "" 0 m "0 � I c E u � E .... 0 0 0 .8 .8 0 c m ..... .l:: 0 - .Q 0) c c C N ..... C c I 10 co .2 0 0 0 0 10 . � � � .... "0 ..c � 0 ·x u ..... (; '"6. Cl) .0. .0. 0 m 0 '6 � Cl) 0. ·0 ·0 0. .><: 0 co (5 Cii c co � � co C. c co > > ..c > ::J .!:; m 0. n. m ::J n. ·c Hydrological Compounds Water Mt ( 1 ) 1 . 1 4 Hydrogen (Atmosphere) Mt Hydrogen Mt -0. 1 3 Hydroxide Mt Other Oxygen Mt - 1 .01 Other Mt Total Mt Energy (2) (2) (2) (2) (2) (2) (2) Note: the same notes apply as for Table 7.2. 238 7.1 .5 Auckland Region Phosphorus Cycle Module The phosphorus cycle module for the Auckland Region is depicted in Figure 7.4 with total elemental flux described in numerical format in Table 7.4. Phosphorous is another important element playing an essential role in many biogeochemical processes. Phosphorous, like nitrogen, is a key limiting factor in terrestrial and oceanic plant growth. Controversy exists as to which of P or is the more limiting factor, particularly in the ocean. lahnke ( 1 992) argues that in the long term P is more likely to be the l imiting factor - if only because the atmosphere contains an abundant supply of N for marine fixation. Previous studies of the global P cyc le have included Stumm ( 1 973), Lerman et al. ( 1 975), Schlesinger ( 1 99 1 ), Wollast et al. ( 1 993), den Elzen et al. ( 1 995), Tiessen ( 1 995) and Smil (2000, 2002). The phosphorous cyc le is unique among the cycles presented here for two reasons, and this is demonstrated in the Auckland Region input output model . 188 Firstly, the P cycle lacks any gaseous fluxes (Jahnke, 1 992; den Elzen et aI. , 1 995 ; Smil, 2002). The atmosphere plays a relatively minor role in P transportation in dust particles, seaspray, rain and cloud droplets (Jahnke, 1 992). In the Auckland Region this includes formation of atmospheric dust particles containing 0. 1 0 kt P, dissolution of these particles to soil releasing 0.07 kt P, deposition of 0.02 kt of soluble P in the surface ocean and, in reverse, evaporation of 0 .0 1 kt of soluble P by the atmosphere, and the deposition of 0.02 kt of insoluble P to sediments. Secondly, microbial and oxidation-reduction reactions are of little influence in control ling the distribution of P terrestrially, i .e. P cycling is largely inorganic in nature (Jahnke, 1 992). Weathering and transportation of calcium phosphate rock, particularly apatite CaS(P04)30H, is instead the major terrestrial flux (Mackenzie et al., 1 993; Tiessen, 1 995; Smil, 2000), releasing 4 .01 kt P into Auckland Region soils. Transportation of 0.65 kt P is typically through stream/river transfer of soil to marine sediment. Oceanic upweUing and downwelling also transports P, with eventual deposition as sediment. Over geological time sedimentary rocks are formed by compaction, pressure and heat, accounting for 0.66 kt P. 188 A possible third reason is that the estimation of natural P fluxes is extremely difficult (Schlesinger, 1 99 1 ; Jahnke, 1 992; den Elzen et aI. , 1 995) . There are primarily two reasons for this: ( 1 ) P is mostly present in particulate, rather than biological, form, and (2) the introduction of P from anthropogenic sources obscures natural fluxes. £l. :l;! ""' o :t: o C :::J ..... 0... r-- Figure 7.4 fl l' f at h . ellc p Atmosphere NEC dissolution of atmospheric P dust particles 0.07 kt P £l. soil to ocean P :l;! N sediments 0.65 kt P 0 Soil 0 0... � .c :::J 0 � (fl (IJ .S C '+-Q) 0 E c 0 "0 . .." Q) 'Ui r.f) 0 '+- £l. Cl. o :;;:: Q) O> � "0 c 0 ·c ...f Q) x £ () (IJ 0 Q) ..... 3: 0... Sedimentary Rock sedimentary P rock Marine Sediments formation 0.66 kt P Auckland Region's Phosphorus Cycle, 1 997-98 - £l. :l;! N C> 0 £l. 0... :l;! Q) 0 :0 Marine Plants 0 :::J 0... 0 (fl Q) a :0 :::J C 0 0 (fl � £l. '+- 'Ui :;;:: 0 0 0 c Cl. 11 .2 Q) x "0 m :::J <+= ..... 0 2 Cl. (IJ (IJ > :i Q) () t (IJ Cl. - P sedimentation =0 kt P Ocean NEC j Table 7.4 Input-Output Model of the Phosphorus Cycle Processes for Auckland Region, 1 997-98 Phosphorus Processes m m ID ID U U t t (1) (1) a. .:>£ a. Ul 0 Ul e ::J ::J "0 0.. "0 C 0.. 0 � m 0.. 0 � (1) C 0.. 0.. 0 ·c C ID 0.. ID .� ID E ID :0 ID :0 .c .E ID E .c a. E (1) :0 (1) a. m "0 ::J (1) ::J m 0 .:>£ '6 ID "0 ::J "0 0 E 0 ID m m "0 m x E c e m 0.. c ::J ro 0 '+- m <+= ro 0.. '+- 0 '+- � 0 c '+- '+- ID 0 C 0 0 '+- � en (1) +' 0 C C ID 0 C C � C (1) .� 0 '.;J 0 0 ::J C 0 ID C 0 � � :;:; :;:; 0 0 S .� 0 '00 '00 � ID .c .8 c 0 t "0 "0 E .� ro a. 0 0 2 (1) a. a. (1) E m ID '6 a. m C ID '0 > ID ID I- m ID ::J � m 0.. ID "0 "0 0.. .E '6 0.. m Phosphorus Compounds Calcium phosphate kt 0.05 -0.08 0.53 -0.36 Phosphate kt -0.30 0 . 2 1 Monohydrogen phosphate kt -0.03 ( 1 ) 0.05 Other Water kt 0.00 -0.01 Oxygen kt Other kt -0.03 0.04 -0.23 0 . 1 6 Total kt Note: the same notes apply as for Table 7.2. 7. 1 .6 Auckland Region Sulphur Cycle Module 24 1 The Auckland Region sulphur cycle module is depicted in diagrammatic format in F igure 7 .5 , while the accompanying Table 7 .5 represents total elemental flux for the sulphur cyc le in numerical. Sulphur is a key nutrient to life providing structural integrity to protein-based tissue (Charlson et al., 1 992a). In its fully oxidised state sulphur exists as sulphate, SO/, an abundant anion in both freshwater and seawater. Sulphate is the major cause of acidity in rainwater, and consequently plays a fundamental role in chemical weathering, often in the form of acid rain. The S cycle has been studied extensively by inter alia Freney et al. ( 1 983), Galloway ( 1 985), Schlesinger ( 1 99 1 ), Charlson et al. ( 1 992a, 1 992b), Mackenzie et al. ( 1 993), Wollast et al. ( 1 993), Jones et al. ( 1 994), den Elzen et al. ( 1 995), Ayres ( 1 996) and Smil (2002). Figure 7 . 5 depicts the major annual fluxes of the S cycle in the Auckland Region. Significant fluxes of S entering the atmosphere are the release of H2S by marine plants containing 1 .69 kt S, and sea spray containing l .3 5 kt S, which is promptly deposited onto land. B igg et al. ( 1 9 84) have identified sulphate as the dominant component of cloud condensing nuclei (CCN) - which is postulated to have a significant impact on global radiation. 1 89 B iological processes result in the release of reduced gases to the atmosphere. Ocean-based emissions are typically of dimethylsulphide (CH3SCH3 or 'DMS') and carbonyl sulphide (COS) 190 from the reduction of seawater sulphate by phytoplankton, while hydrogen sulphide (H2S) is emitted by terrestrial soils and p lants. Sulphates in soil and seawater are taken up by plants and then reduced to organic sulphur compounds (e.g. amino acids such as cysteine). Decaying matter retains some S, but is mostly converted to H2S by decomposers. Volcanism may also sporadically inject sulphate into the atmosphere. 1 89 Charlson et al. ( 1 987, 1992a), inter alia, have at the global level proposed the existence of a homeostatic feedback loop which may regulate the Earth's climate. Charlson et al. ( 1 987) point out that because cloud albedo is partially a function of CCN density, any increase in atmospheric S may lead to greater cloud cover, reducing solar radiation, and lowering phytoplankton photosynthesis. Consequently, lowering phytoplankton photosynthesis lessens the amount of CCN and, in turn, increases solar radiation. Watson and Liss ( 1 998) and Smil (2002) are however sceptical of this homeostatic climate control, questioning both its magnitude and direction. 190 COS is the most abundant atmospheric species in the troposphere. Because of its inert chemical behaviour, and consequently long residence time, it may enter the stratosphere. In turn, it may be photochemically oxidised to sulphate (Mackenzie et al., 1 993) H2S dissolution 14 . 14 kt S uptake of S02 by soil 4.05 kt S Atmosphere NEC , release of H2S 1 .69 kt S en en :;;: :;;: '" '" M 0> \ N " c 0 Cl! Cl) QJ ...... U 0 - 0 >-Soil >- Cl! Marine Plants .0 L-a. S runoff 3.76 kt S N en 0 c Cl) Cl! ...... QJ 0 U en QJ 0 :;;: � ...... 0 Cl! o ex> 15.. 0) 00 � C � ·c U QJ 0 � L-ro Cl) QJ i::' � Cl! ocean spray to land 5.77 kt S "0 ...... c c Cl! QJ � .§ = "0 a. QJ � en I Sedimentary Rock Marine Sediments , Ocean NEC � sedimentary S S sedimentation 1 .69 kt S rock formation 1 .69 kt S Figure 7.5 Auckland Region's Sulphur Cycle, 1 997-98 243 S ignificant sources of S in soil are obtained from uptake of S02 containing 4.05 kt S, ocean spray depositing 5 .77 kt S onto land, and uplift and weathering of sedimentary S rock, supplying 8.80 kt S per year to soil in the Auckland Region. Transportation of soil via runoff accounts for 3 . 76 kt of S entering the ocean, with 1 .69 kt S per year undergoing sedimentation to form S rock. Two further notable fluxes of S into the ocean are H2S dissolution releasing 1 4. 1 4 kt S, and oceanic uptake of S02, accounting for 2.95 kt S . The cycling of sulphur from gas to sulphate particles and subsequent wet deposition (via rainwater) on the Earth ' s surface is a rapid process. The impacts of the atmospheric S cycle are thus typically localised or regional, with most S being distributed c lose to its source. Moreover, the S cycle is arguably the most perturbed by human activity of all the cycles through mainly fossil fuel combustion and biomass burning, with anthropogenic emissions equating to 40 percent of total surface-atmosphere flux (Andreae, 1985). The most common manifestation of this perturbation is acid rain - an event that is best studied on a regional basis. 1 9 1 Table 7.5 Sulphur Compounds Hydrogen sulphide Sulphate Sulphur dioxide Sulphur Other Water Oxygen Other Total I nput-Output Model of the Sulphur Cycle Processes for Auckland Region, 1 997-98 Sulphur Processes -"" u e m � ell E - >- � c e oD oD .s:: m a '" .B g m (il c 0 .2 -£ >- 0 >- tU � Q) S ell m m ell E s: a 0- .... 0- Ol 0 0 a (l) E -0 '" Q) '" '" lI:: E Q) c '" Q) (l) 0 Ol '" C -" -"" c c 15 E 15 2 ell ell '" E (l) Q) Ol Ol Q) 2 Q) 15 m ·c � u 0. 0. u 0 :l ::> 0 m m '" :l kt -1 5 .0 Kt B.B 12 . 1 -5 . 1 26.4 42.4 kt -5.9 (1 ) -B . 1 Kt 1 .7 -B.B kt -31 .B kt -2.9 -4.0 3 .4 -1 7.6 kt 4.4 kt Note: the same notes apply as for Table 7.2. 19 1 Charlson et al. ( 1 992) and den Elzen et al. ( 1 995) note that the calculation of a global S budget requires the painstaking process of a myriad of measurements over a wide range of regions, in turn adjusted for seasonality, to produce reasonable averages. 244 7.1.7 Auckland Region Nitrogen Cycle Module The nitrogen cycle module for the Auckland Region is depicted In F igure 7 .6 with total e lemental fluxes described in numerical format in Table 7 .6 . Despite its abundance in the atmosphere, nitrogen is a relatively rare e lement in the terrestrial biosphere (Soderland and Svenson, 1 976; McElroy et al. , 1 977; Jaffe, 1 992; Bart et al., 1 993 ; Wol last et al., 1 993; Ayres et al., 1 994). The availability of nitrogen is a critical/limiting factor in numerous biogeochemical processes. Nitrogen l imits the rate of net primary production (Vitousek and Howarth, 1 99 1 ), is a critical component in enzymes, hormones, chlorophyl l and genes that control biogeochemical reactions that reduce/oxidise carbons (den Elzen et al. , 1 995), and plays an important role in a number of environmental issues e.g. ac idification, surface water pollutionleutrophication, stratospheric ozone depletion, photochemical smog and so on. Figure 7.6 reveals the complex set of biotic and abiotic fluxes in the Auckland Region that comprise the n itrogen cycle. These processes may be broadly grouped as : ( 1 ) organic atmospheric. Denitrification by bacteria (mainly of genus Rhi=obium) of decaying organic materials and by plants, which release 2 and N20 into the atmosphere (i .e. m icrobial production of inert N2 and N20, producing 8.46 kt N per year), and release of 2 and N20 by marine plants, accounting for 1 .23 kt N per year. In reverse, biological fixation by bacteria and phytoplankton of 1 .23 kt N convert 2 to NH3, which is converted in turn through bacterial n itrification to N}-4, and ultimately to nitrites (N02) and nitrates (N03). S ince plants cannot util ise elemental nitrogen, al l l ife therefore depends on biological fixation - not surprisingly 95 percent of all n itrogen fluxes thus involve biological processes (Jaffe, 1 992); (2) inorganic atmospheric from inter alia the formation of NOx by l ightning (0.32 kt N), wet deposition of NOx (0.48 kt N) and acid rain (0.45 kt N); and (3) oceanic . Weathering and formation of sedimentary rock uti l ising 0 .94 kt N, uptake and oxidation of NH3 using 4.29 kt N, and receipt of non-agricultural runoff containing 0.90 kt N. The complex nature of the nitrogen cycle means that quantification of these fluxes is stil l very m uch in its infancy. z :tL (") (0 o c o � u <;::: � c Q) "0 Q) C '': co E � Figure 7.6 oxidation of N2 0.48 kt N It r Atmosphere N2 acid rain 0.45 kt N NOx formation by liqhtninq 0.32 kt N I wet deposition of NOx 0.48 kt N uptake of NH3 by ocean 4.29 kt N Land Humus Atmosphere NEC c o U ::J "0 Z e .x 0. (") _ m CO N :0 '" O I o Z 'E '0 Z :tL (0 (0 o rJl > D M I Z -o Q) .><: CO i5.. ::J reduction of N20 0.96 kt N photochemical oxidation of N20 0. 1 6 kt N ( Z § -'" U � ::J (0 "0 0 o N Ci z (ij "O ._ C D co e N .!:? z E 'O T uptake of N2 and N20 6.44 kt N 1 Soil N runoff 0.90 kt N Ocean NEC 1 Auckland Region's N itrogen Cycle, 1 997-98 weathering of sedi­ mentary N rock 0.94 kt N sedimentary N rock formation 0.22 kt N l' Atmosphere N20 T Z :tL (") N o N Z "0 C co N Z '0 Q) rJl co Q) � Marine Plants Z :tL (") N c c o 0 � :t:: co c � � - a. co 0 u ...... .- > 0l .J::. o a. 0 > :O D Sedimentary Rock Table 7.6 Input-Output Model of the Nitrogen Cycle Processes for Auckland Region, 1 997-98 Nitrogen Processes c 0 0 z -"" � 0 (.) C "0 e co c Z c 0. co r: 0 z Cl 0 Z '0 � � c >. c z E 2 E .<:: - .Q - c 0. 0 0 0 0 0 co .E c .s:::. iii Q) Q) c ,g> >- z c '" c '0 (.) .£ 0 .0 0 Z :2 Z 0 - "0 >- >- e 0 .0 0 c 0 0 Q) (.) :::> c :::> .0 .0 Z Z � co "0 >< 0 co co 0 "0 co co 2 iii "0 'x co 0 0 Q) (.) 15. ID (.) 15. 15. u c � E :0 � 'E .s:::. � Ti 'E Q) Q) :::> 0 Z :::> 0. co :::> :::> Z "0 c: u °E o c: o u UJ If) Ql U o� Ql (/) ro u 1 Horticulture and fru it growing 2 Livestock and cropping farming 3 Dairy cattle farming 46 Health and community services 47 Cultural and recreational services 48 Personal and other community services 1 Gas regulation 2 Climate regulation 3 Disturbance regulation 08' 15 Genetic resources "8 16 Recreation UJ 1 7 Cultural "0 c: ro � ::> :: Cl ::> c: u 0_ 0- 3: � e I Cl Cl c: 00. 0.. e u "0 c: ro .:£ .9 g' � 'E > ... � � N Economic I ndustries � c: ::> E E o U "0 c: ro en E � ro o� Ql o E !!? E f) the subsequent analysis would require the use of generalised or pseudo matrices. The major drawback of moving to multiple outputs per industry is that negative ecological multipliers may be generated. This is a long-standing problem in input-output analysis that is yet to be satisfactorily resolved. 252 Ib (n - Jhrl = 3 (7.3) The solution vector 3, the so-called ecological multiplier, represents the total (direct and indirect) embodied resource inputs (or residual outputs) measured in physical terms per unit of net output. Step 3 Calculation of First Round Impacts Greater insight into the distribution of ecological multiplier impacts may be gained by decomposition of the multiplier into flrst, second and subsequent round impacts. F irst round resource inputs (or residual outputs) requirements for every industry may be determined by diagonalising 3, and in turn, postmultiplying by the inverse matrix (n - Jhrl to obtain an 'evaluated' matrix, )K (B x f), (7.4) The first round impacts for industry 1 are presented in the first column of the evaluated matrix, )K. Simi larly, the first round impacts for industry 2 are presented in the second column of the evaluated matrix, )K, and so forth. Step 4 Calculation of 2nd to nth Round Impacts In this step the second round impacts for a given co lumn industry of matrix )K are generated. 196 To trace second round impacts we must flrst select the given column industry (industry 1 column in this example) from the evaluated matrix )K for further analysis. The second round impacts resulting from the first round input of commodity 2 into industry 1 , for example, may be computed as, * 1> - R 1 2.2, 1 - .,2,2, 1 (7 .5) where *1 is a column vector (B x 1) of column industry 1 of the evaluated matrix )K; �2,2, 1 is a vector (B x l ) of the second round impact (denoted by the flrst subscript), stemming from the first round input of commodity 2 (denoted by second subscript) into industry 1 (denoted by third subscript); and b is a scalar (1 x 1) ratio of commodity 2 input into industry 1 from the 1 96 This column tells us the total embodied resource/residual requirements (a positive element) and the indirect embodied resource/residual requirements (negative elements). An additional row recording the direct resource/residual requirements (positive elements) may be augmented to matrix )K to complete the analysis. 253 evaluated matrix )1(, divided by the 'net output of industry 2 ' from the evaluated matrix )1(. The same subscripting system is applied to 1>2,2, 1 as for �2,2, 1 ' The same procedure may be used to calculate the second round impacts of other commodity inputs in industry 1 . In turn, the subsequent rounds of impact may be calculated in an analogous manner. Ultimately the so called ' infinite regress ' situation arises, as the individual indirect contributions of elements in P become progressively smaller with subsequent rounds. The number of branches in the tree diagram increase exponentially with each subsequent round. The number of branches, It, in any given round, rp, is given by, The total number of branches across all rounds, rp), is given by, Id) Id) I 'tJ = I [er - l )P;] , I�J I�) (7 .6) (7 .7) where i represents rounds ° to j. The number of branches generated over a small number of rounds with a relatively small number of industries may be substantial. A complete lifecyc1e assessment of 48 industries over 5 rounds would, for example, generate 234,330,767 branches ! Diagrammatic representation of the lifecycle assessment chains for even a single industry requires that selection criteria be applied to restrict the analysis to only the main embodied flows. Step 5 Determination o/the Direct and Total Multipliers Matrices The direct multipliers matrix may be determined by dividing the physical amount of resource input (residual output), the elements in vector lb, by the net output for each industry, the elements on the diagonal of matrix Tt. Matrix $I is used to denote the result. These multipliers represent the resource input (residual output) in physical units per $ net output. Total impact (direct plus indirect) may be measured relative to direct impact to produce an ecological multiplier analogous to the Type I economic multiplier used frequently in economic impact by economists. This matrix, M, may be derived by element-wise division of matrix 3 by matrix $I. 254 In this Section, tree diagrams are used to i l lustrate the embodied appropriation of ecosystem services associated with producing a commodity. 1 97 Ecological multipliers, as calculated using the methodology presented in Section 7.2 . 1 , may be used to create tree diagrams that show the conservative nature of embodied ecosystem service flows - viz, for a particular industry, ecosystem service inputs taken directly from the environment + ecosystem service inputs appropriated from the environment through economic production chains = the total ecosystem services embodied in any economic outputs produced by the industry. Moreover, for those ecosystem services appropriated through production chains it is possible to calculate the ecosystem service value appropriated at each point in the chain. 7.2.2 Direct Ecosystem Service Value by Industry In Table 7.8, the direct value derived from Auckland Region's terrestrial ecosystems is disaggregated by 48 economic industries and households. In total this accounts for $ 1 .03 bi llion, which represents 3 . l percent of the region' s GRP ($33 .2 billion) . Compared with other regions (such as Canterbury $7 .8 billion or 63 percent of GRP - refer to McDonald and Patterson (2003)) this represents a low level of appropriation. This result is however not surprising given: ( 1 ) the relatively small land area of the region (i .e. a larger land areas generally mean larger ecosystem service contributions); (2) the presence of a large urban land cover (i .e. the ecosystem services contribution by urban areas on a per hectare basis is substantially lower than other land cover types); and (3) the lack of consideration of ecosystem services embodied in goods and services purchased from other regionsl98• This contribution is likely to be very high given that Auckland Region is a significant appropriator of interregional and international land - refer to Chapter 8 for further details. Additionally, the ecosystem services provided by the Auckland Region marine ecosystem have been tentatively valued at $ l .29 billion. A more detailed discussion of the contribution made by ecosystem services to the Auckland Region is given at the end of Appendix J by ( 1 ) ecosystem type, and (2) ecosystem service. Households [49] rank highest in their appropriation of ecosystem services at $24 l . 8 million (23.4 percent of the total). The primary industries of l ivestock and cropping farmirIg [2] 197 Lifecycle assessment procedures have been formalised by the Society of Environmental Chemistry and Toxicology ( 1 993). Four formal steps are recognised in l ifecycie assessment: ( 1 ) goal and scope definition; (2) inventory analysis, to quantify the resource inputs and pollutant outputs at each step in the lifecycle; (3) impact assessment, to quantify the impacts, often using CO2 equivalents, acidification equivalents or the like; and (4) interpretation of results and recommendations to reduce environmental impacts. 1 98 The contribution made by the marine system is estimated to be $ 1 .29 billion - refer to Appendix J for further details. Overall, 255 ($200.6 million), mining and quarrying [8J ($ 1 34 .8 million), dairy cattle fanning [3J ($ 1 33 .9 million), forestry and logging [6J ($49 .0 million), other farming [4] ($46.2 million) and horticulture and fruit growing [ 1 J ($20. 5 million) rank in the top eight appropriators of ecosystem service value. Ranked third is the water supply industry [28J at $ 1 89.6 million with its heavy reliance on water supply services. In contrast, the lowest rankings tend to be for service industries such as finance [3 7], insurance [3 8], services to finance and investment [39] and real estate [40J. 256 Table 7.8 Direct Value Derived from Auckland Region's Terrestrial Ecosystems, by Economic Industry, 1 997-98 Industry 1 Horticu ltu re and fru it growing 2 Livestock and cropping farming 3 Dairy cattle farming 4 Other farming 5 Services to agricu lture, hunting and trapping 6 Forestry and logging 7 Fishing 8 Mining and quarrying 9 Oi l and gas exploration and extraction 1 0 Meat and meat product manufacturing 1 1 Dairy product manufactu ring 1 2 Other food manufacturing 1 3 Beverage, ma lt and tobacco manufacturing 14 Textile and apparel manufacturing 15 Wood product manufacturing 1 6 Paper and paper product manufacturing 1 7 Printing , publ ishing and recorded media 1 8 Petroleum and industria l chemical manufacturing 19 Rubber, plastic and other chemical p roduct manufacturing 20 Non-metall ic mineral product manufacturing 21 Basic metal manufacturing 22 Structural , sheet, and fabricated metal product manufacturing 23 Transport equipment manufacturing 24 Machinery and equ ipment manufacturing 25 Furn iture and other manufacturing 26 Electricity generation and supply 27 Gas supply 28 Water supply 29 Construction 30 Wholesale trade 31 Retail trade 32 Accommodation , restaurants and bars 33 Road transport 34 Water and rail transport 35 Air transport, services to transport and storage 36 Communication services 37 Finance 38 Insurance 39 Services to finance and investment 40 Real estate 41 Ownership of owner-occupied dwel l ings 42 Business services 43 Central government administration , defence, publ ic o rder and safety service� 44 Local government admin istration services and civil defence 45 Education 46 Health and commun ity services 47 Cultural and recreational services 48 Personal and other commun ity services 49 Households Total Direct Direct Value Value ( in rank order) $ m il l ion 20.52 8 200.55 2 1 33.85 5 46.21 7 0 .02 30 49 02 6 0.02 32 1 34.77 4 0.00 48 0.02 29 0.02 36 0.05 20 0 .01 38 0.04 23 0 .05 22 0 .02 28 0 .02 35 0 .01 4 1 0 .03 26 0.03 25 0 .02 33 0.02 34 0 .01 39 0 .02 3 1 0.02 37 0 .01 42 0 .00 47 1 89 .64 3 0.06 1 8 0.05 21 0.06 1 9 0.53 1 4 0. 1 5 1 7 0 .01 40 0 .47 1 6 0.03 24 0 .00 44 0.00 46 0.00 45 0 .00 43 0.00 48 0.03 27 0.83 1 3 1 .88 1 0 1 .2 1 1 1 0.49 1 5 1 0 .20 9 1 . 1 6 1 2 241 .80 1 1 ,033.96 7.2.3 Direct and Indirect Ecosystem Services by Industry 257 F igure 7.9 and Figure 7. 10 depicted the total appropriation of embodied ecosystem services by two key Auckland Region industries, namely, air transport, services to transport and storage [35 ] and business services (42] . 199 These industries draw only very small amounts of ecosystem services directly from the environment, respective $472,500 (4.66 percent of their total embodied ecosystem service appropriation) and $30,200 (0.07 percent). By corollary, these industries must appropriate most of their ecosystem service requirements through their production chains. The air transport, services to transport and storage [35] industry, for example, appropriates ecosystem services indirectly to the value of $9.7 million (1 .0 percent of total ecosystem service appropriation by Auckland Region industries2oo), while the business services [42] industry indirectly appropriates $4 1 .0 million (4. 1 percent) . Indirect appropriation through production chains of ecosystem services by the air transport, services to transport and storage (35] industry can be traced back primarily to direct appropriation from the environment by the cultural and recreation services [47J industry (> $3 .0 million), water supply [28] industry (> $ 1 .9 million) and the mining and quarrying [81 industry (> $0.9 million). This includes the indirect appropriation through production chains of primarily aesthetic, artistic, educational, spiritual and scientific values through the cultural and recreation services [47] industry, water through the water supply [28] industry, and raw materials through the mining and quarrying [8] industry. Lesser contributions are also made through the livestock and cropping (2J industry by way of erosion control, food production, waste treatment, raw materials and nutrient cycling services. The indirect appropriation of ecosystem services through production chains by the business services [42] industry is by comparison dominated by aesthetic, artistic, educational, spiritual and scientific values obtained via the cultural and recreational services (47J industry ($25 . 1 million or 6 1 .2 percent of the industry's total ecosystem service appropriation). Other significant embodied ecosystem services include water supply appropriated indirectly through the real estate [40] industry (> $3.2 million); paper products, c limate regulation, and waste treatment through the forestry and logging [6] industry; and erosion control, food production, raw materials (fibre), and nutrient cycling services indirectly through the wholesale trade (30] industry. 199 This refers only to within region ecosystem service appropriation, i .e. it excludes the appropriation of ecosystem services embodied in imported goods and services. 200 This does not include the ecosystem service appropriation by households. Figure 7.9 1st Round Effects 2nd Round Effects 3rd Round Effects $1 .0m '----------- $0 4m $1 .0m 4th Round Effects KEY Total Output .-__ Direct Input =-.... �,. ..... $0 3m . Indirect Input from Other Industries (2] Livestock and Cropping [a] Mining and Quarrying ••• $O.9m (23] Transport Equipment Manufacturing (28] Water S u pply $ 1 1 m [30] Wholesale Trade [3 1 ] Reta il Trade [35] Air Transport. SeNices to Transport and Storage [40] Real Estate (42] Business SeNlces (47] Cultural and Recreattonal Services Total Embodied Ecosystem Services Appropriated by the Auckland Region Air Transport Industry, 1 997-98 $41.0m [42) Figure 7 .10 ---- 1 st Round Effects 2nd Round Effects 3rd Round Effects 4th Round Effects ,r--- $1 7m � [6) [ 17( ' $0.8m .., 4 $1 Om ,p [20( �" [8( � '-�---- ��;::::::::==-===� [30( [2) rt $1 Om 51 4m � � [31 ] .... :- Sl 1m $25 1m �---------- $0 4m $4 4m $3.2m KEY Direct Input Tolal Output Indirect Inpul from Other Industries [2) lives lock and Cropping Fanmlng [6] Forestry and Logging (8) Mining and Quarrying ( 17J Printrng. Publishing and Recorded Media (20) Non,Metallic Mineral Product Manufacturing [28J Water Supply (301 Wholesale Trade [31 J Retail Trade (40J Real Estate (42J Business Services (47J Cultural and Recreational Services Total Embodied Ecosystem Services Appropriated by the Auckland Region Business Services Industry, 1 997-98 Chapter Eight Auckland Region's Ecological Footprint and Environmental Interdependencies with Other Regions201 261 This thesis has thus far focused on structural analyses of the interdependencies in and between the Auckland Region economic system and the Auckland Region environmental system (refer to Figure 4.9). In Chapter 8, through the use of the ecological footprinting method, the analysis is extended to quantify how the Auckland Region economic system depends on natural capital and ecosystem services drawn from other regions and countries?02 This analysis is important as an urban region, like Auckland Region, is arguably more dependent on natural capital and ecosystem services obtained from other regions/countries than on those obtained within its own regional boundaries. Specifically, this Chapter extends Bicknell et al. 's ( 1 998) input-output based ecological footprinting method, to investigate the ecological interdependencies between Auckland Region and other regions203 in New Zealand. This required the construction of 1 6 regional input-output matrices by regionalising the 1 997-98 New Zealand input-output table using Jensen et al. ' s ( 1 979) GRlT (Generating Regional Input-Output Tables) methodology. From these regional input-output matrices, interregional flows of commodities were established using an optimisation model, which in turn enabled the ecological interdependencies to be tracked and quantified. 8.1 The Ecological Footprint Concept 8. 1 . 1 What is the Ecological Footprint? The Ecological Footprint is defined by Rees (2000, p.3 7 1 ) as the "area of productive land and water ecosystems require to produce the resources that a popUlation consumes and assimilate the wastes that the population produces, wherever on Earth that land and water may be located". 201 A copy of most of this Chapter appears in the journal Ecological Economics, 2004, Volume 50, pp. 49-67. The introductory Section of this Chapter differs from that contained in the journal article, and the journal article contains a summary Section which does not appear in this Chapter. 202 The ecological footprint uses embodied land as a proxy for the amount of natural capital consumed by a population or economy. As discussed in Section 8.2.2, there are other numeraires (apart from land) that can be used in this context. 203The method also quantifies how Auckland Region's ecological footprint is drawn from other countries. Due to lack of data, it is assumed that products imported from overseas have exactly the same embodied land per $ ratio as products made in New Zealand. 262 It can be seen as a sustainabi lity indicator in two senses. Firstly, it measures the ecological cost -"Cin land area) of supplying all of the goods and services to a human population. This recognises that people not only directly require land for agricultural production, roads, buildings and so forth, but that land is indirectly embodied in the goods and services that people consume. For example, the indirect (or embodied) land required to produce a kilogram of butter includes not only the land used directly in manufacturing, but all land embodied in the inputs that went into producing the butter - dairy farm land, land required to produce the packaging and so forth. In this sense, the footprint can be used to make visible the hidden ecological cost of an activity or population. A second, and more controvers ial interpretation of the Ecological Footprint as a sustainability indicator, invokes the idea of carrying capacity. Carrying capacity in ecology is the maximum population a given land area can support indefinitely. The idea is relatively straightforward when appl ied to well-defined biological populations - e.g., a certain number of hectares are required to support a herd of deer. If the number of deer exceeds their carrying capacity then the population is said to be in 'overshoot' . Resources (mainly food) will become scarce and population die-back wil l occur. This idea is more controversial when applied to human populations, as in the Limits to Growth study, which projected a decline in global human population as it overshot its carrying capacity (Meadows et al., 1 972; Meadows et al., 1 992). Some proponents of footprinting argue that the total embodied land area required by a population should not overshoot its' biocapacity - e .g. , Loh (2000) argues that the Ecological Footprint of the Netherlands at 92.9 mil lion ha considerably overshoots its biocapacity of 37 .4 million ha. Less dogmatically, it can be concluded that the Netherlands is in 'ecological deficit' , in the sense that it is using more biologically productive land than is available within its borders. 8. 1.2 History o� the Ecological Footprint The University of British Columbia's School of Community and Regional P lanning developed the Ecological Footprint in the early 1 990s. The concept was popularised by Wackemagel and Rees ( 1 996) in the publication Our Ecological Footprint - Reducing Human Impact on th� Earth. Wackernagel et al. ( 1 999) acknowledge Vitousek's et al. ( 1 986) study on the human j appropriation of photosynthesis products as the inte l lectual predecessor to the footprint concept. However, its antecedents can be traced back a lot further. In the eighteenth century the Physiocrats argued that the embodied land content of a commodity determined its value. For the Physiocrats, all value was derived from the land (nature), and in 263 this sense agriculture was the only productive industry in the economy with the manufacturing and service industries considered sterile. Classical economists, although not subscribing to an 'embodied land theory of value' did emphasise the idea of carrying capacity. Both Thomas Malthus ( 1 766- 1 834) and David Ricardo ( 1 772- 1 823) saw population being constrained by the carrying capacity imposed by land availabil ity. Malthus argued that population growth wasn't sustainable in the long run, as it grew according to a geometric progression and it would eventually overshoot food supply from land that grew arithmetically. Ricardo didn't foresee an overshoot like Malthus did, but instead suggested that popUlation growth would gradually approach its carrying capac ity as food production was forced to use less fertile land. In the modern era, Borgstrom ( 1 967, 1 973) developed the concept of 'ghost acreage ' which is s imilar to the idea of the Ecological Footprint. This idea was further promoted by soc iologist Carton ( 1982) in his book Overshoot: The Ecological Basis of Revolutionary Change. Ghost acreage is the additional land a nation needs in order to supply the net amount of food and fuel from sources outside the nation. The appropriation of ecosystem areas and services has also been a central theme in other approaches e .g., Folke et al. ( 1 997) and Brown and Ulgiati ( 1 998). 8. 1 .3 How is the Ecological Footprint Calculated? Several methods have been advanced for calculation of Ecological Footprints - e.g., Wackernagel and Rees ( 1 996), Folke et al. ( 1 997), Bicknell et al. ( 1 998), Wackernagel et al. ( 1 999), Loh (2000), van Vuuren and Smeets (2000), and Ferng (200 1 ) . Although each of these methods has its own peculiarities and insights, many have their roots in the work of Wackernagel and Rees ( 1 996). Wackernagel and Rees Method The Wackernagel and Rees calculation method begins with the construction of a ' consumption by land use' matrix for a given population. The consumption dimension covers food, housing, transport, consumer goods and services, while the land use dimension encompasses built-up areas (supporting roads, housing and other infrastructure), crop land and pasture (for production of food and other goods), managed forest (for production of wood products), and energy land (for sequestering carbon dioxide emissions resulting from the burning of fossil fuels). 264 Population data, together with consumption data (mainly in physical units), for each land use category are used to derive an average annual consumption per person (physical units per capita). Consumption is calculated by adding imports to domestic production and excluding exports. The land area utilised by each consumption category is then determined for each land use category. This requires dividing consumption in each land use category by a relevant global average yield to obtain land area. Global average yields are used so that comparisons can be made between the footprints of different nations and with the globe. The land appropriated for energy consumption is treated separately, primarily due to the size of the contribution it makes. Wackernagel et al. ( 1 999) distinguishes between five types of energy, namely: gas fossil, liquid fossil, solid fossil, firewood and hydropower. Nuclear power is treated as a fossil fuel. Energy land is calculated by assessing the amount of planted forest land required to absorb the CO2 emissions released in the burning of fossil fuels. The role played by the oceans in CO2 sequestration is also acknowledged. The oceans are assumed to absorb some 35 percent of CO2 emissions at the global level. Once again, correction for trade is required as energy is utilised in the production of exported goods and services and conversely embodied in imports. Aggregating the land area appropriated by each land use category generates the Ecological Footprint. Prior to aggregation each category is multiplied by an 'equivalence factor' to take account of differences in biological productivity. The Ecological Footprint may also be expressed in per capita terms, which permits the comparison between different nations, regions or populations. Input-Output Method The Ecological Footprint can also be calculated by using input-output analysis to track the flow of embodied land. This method of analysis which was fITst developed by Bicknell et al. ( 1 998) and refilled by Ferng (200 1 ) and others, has not to date been widely used. It should, however, be noted that the calculation of embodied resources using input-output analysis has been widely undertaken since the early 1 970's by analysts such as Hite and Laurent ( 1 97 1 ), Herendeen ( 1 972) and Wright ( 1 975). The input-output method of calculating the Ecological Footprint attempts to s ituate the analysis in a rigorous mathematical framework, but draws upon many of the ideas and principles of the Wackernagel and Rees method. Readers should refer to Section 8 .3 .6 of this Chapter for a discussion of the limitations of input-output analysis in ecological footprinting. 8.2 Critique of the Ecological Footprint Concept 265 Costanza (2000) and Moffatt (2000) argue that the key feature of the Ecological Footprint is that it provides an effective heuristic and pedagogic tool that captures current human resource use in an easily digestible form. In this way, footprinting frequently promotes discussion on issues directly relevant to sustainable development - viz., issues such as: ( 1 ) the fmite dimensions of human activity; (2) the key resources and ecosystem functions for sustainable development; (3) the role played by trade in distributing ecological resources and pressures; (4) the selection of indicators for monitoring progress toward sustainable development and so forth. The Ecological Footprint methodology does, however, have a number of well-known weaknesses and l imitations that are described below. 8.2.1 Lack of Common Definitions and Methodologies There is no accepted methodology for calculating the Ecological Footprint. The Ecological Footprint is not, for example, constructed according to widely accepted international conventions such as that used in the United Nations System of National Accounts (UNSNA). This has led to ambiguities in interpreting the results of various Ecological Footprint studies. For instance, estimates of New Zealand's Ecological Footprint range between 3 .49 and 9.6 ha per capita (Bicknell et al., 1 998; Wackernagel et al., 1 999; Loh, 2000). Investigation of these studies reveals that differences result largely from the assumptions made concerning biological productivity, the use of equivalence factors, and the calculation of energy land. To avoid misinterpretation in this Chapter, and to allow comparison w ith earlier footprint estimates, differences in assumptions between three different calculation methods are outlined in Table 8. 1 . 266 Table 8.1 Assumptions Made by Three Different Ecological Footprint Calculation Methods Bicknell et al . (1 998) Applies local yields for pasture, arable and forest land Loh (2000) Applies global average yields for pasture, arable and forest landa This paper Applies local yields for pasture, arable and forest land Does not apply equivalence factors Applies equivalence factors when Does not apply equivalence factors aggregating land typesa Applies an international energy-to- Applies a world average CO2 Applies a CO2 absorption factor for land ratio obtained from absorption factor' New Zealand Pinus radiata C Wackernagel and Rees ( 1 996) I gnores CO2 absorption by oceans Assumes oceans absorb 35 percent Ignores CO2 absorption by oceans of CO2 emissions Excludes sea space Considers ecological interdependencies between regions as an aggregate (total imports) Based on input-output analysis Notes: a Discussed further in Section 8.2.4 b Discussed further in Section 8.2.3 c Discussed further in Section 8.2.2 I ncludes sea space, estimated to be Excludes sea space 0 . 1 ha per capita for NZ Considers ecological Makes explicit ecological interdependencies between regions interdependencies between as an aggregate (total imports) regionsd Based on work of Wackernagel and Based on input-output analysis Rees ( 1 996) d Discussed further in Sections 8.2.5 and 8.3 8.2.2 Why Use Land as the Numeraire? Why should 'embodied land' be used as the numeraire for a sustainability indicator? Others have argued (Slesser, 1 973; Gilli land, 1 975; Costanza, 1 980; H.T. Odum, 1 983 ; Herendeen, 1 998) that 'embodied energy' or ' embodied solar energy ' is a more appropriate numeraire. Land isn't the only scarce natural resource, so why should it be the only resource entered into the calculation of a sustainability indicator? Arguments alluding to the non-substitutability of land are not compelling, as it could be argued that other natural resources also don't have substitutes - e.g., solar energy. By using input-output analysis to calculate footprints, as is applied in this thesis, the ecological consequences of human activity on other key resources are easily determined. Energy Analysis, for example, has been widely applied in estimating energy embodied in human activities204. The focus of this Chapter is however on the appropriation of biologically productive land. 204 Examples include Gilliland ( 1 975), Hannon ( 1 979), Costanza ( 1 980), and Giampietro and Pimentel ( 1 99 1 ). 8.2.3 Why Include Hypothetical Energy Land? 267 The hypothetical land required to absorb atmospheric CO2 emissions resulting from the burning of fossil fuels, often constitutes more than 5 0 percent of the Ecological Footprint. Critics such as Ayres (2000) find this result questionable. According to them, it assumes that afforestation is the preferred option for CO2 sequestering. However, the use of renewable energy sources such as wind power and energy efficiency initiatives are realistic alternatives (apart from afforestation) for reducing CO2 emissions. Alternatives such as liquefying CO2 and pumping it into the ocean depths or into oil and gas fields replacing the fuel extracted also exist. Planting production forest to sequester CO2 is arguably only a temporary measure. The forests will die, be harvested as products that will eventually decompose, or be used as a fuel source, all of which wil l result in CO2 being re-released back into the atmosphere. Another critical issue with the Ecological Footprint is that it exclusively focuses on energy related CO2 emissions, neglecting the ecological consequences caused by other emissions - e .g., the depletion of ozone by CFCs, or acidification caused by S02 and NOx• More importantly, the Ecological Footprint as currently formulated overlooks pol lution and wastes generated by other unsustainable practices, such as the disposal of non-biodegradable consumer wastes (e.g., p lastics, metals) and persistent toxins (e.g., rubbish leachate). These issues are not addressed in this Chapter of the thesis, although it is recognised that they are important issues that need to be addressed in the further development of the footprint indicator. 8.2.4 Is All Land the Same? The use of equivalence factors during the aggregation of Ecological Footprint components (built-up land, arable land, forest land etc.) is contentious. These equivalence factors recognise that adjustments need to be made to land areas (ha) to take into account variations in biological productivities. For example, fertile flood plains may have a biological productivity several times that of mountainous land, and adjustments need to be made to reflect this difference. It can be argued that this narrow focus on biological productivity ignores other factors that determine the relative value of different types of land - e.g., cultural values, social preferences or relative scarcity. International companson of Ecological Footprints requires consideration of differences in biological productivity. Such differences are primarily due to environmental factors - i .e . , solar flux, soil type, c limatic conditions and type of vegetation cover. This issue is addressed in Ecological Footprint calculations by relating consumption to global average yields rather than 268 local yields205. Such an approach is problematical as it produces results that are not comparable with the actual land area occupied by the appropriating population. At a national or sub­ national level, it is often desirable to be able to examine ecological consequences in terms of actual occupied land area - a unit of measurement famil iar to the resident communities. In this Chapter of the thesis, neither global average yields nor equivalence factors are used, except when international comparisons are made in Section 8.4.7. 8.2.5 What Spatial Boundaries? The selection of appropriate spatial boundaries is a critical issue in footprinting. For example, footprints can be calculated at global, national, regional and local (c ity) scales. Wackernagel and S i lverste in (2000) argue for political or cultural boundaries, as they represent the level at which environmental policy and decis ion-making is most often made. By contrast, van den Bergh and Verbruggen ( 1 999) dispute the use of such boundaries on the grounds that they have no environmental meaning, favouring instead hydrological, climate zone, or larger connected ecosystem boundaries. In this Chapter, New Zealand Regional Counc il areas are used which reflect both political and environmental boundaries. C losely assoc iated with the selection of appropriate spatial boundaries are the ecological impl ications of trade. Rees ( 1 992) argues that trade has the effect of physically and psychologically distanc ing populations from the ecosystems that sustain them. From a regional perspective, information is required not only on footprint size (and on its component shares - e.g., agricultural, arable, forest, built-up and energy land), but also on the origins of contributions made by each imported component and how sustainable it is. For this reason the Ecological Footprint methodology is extended in this Chapter of the thesis to inc lude an analysis of the ecological interdependenc ies of New Zealand regions, in order to cons ider not only the footprint from the consumption (end-use) perspective, but also the production (source) perspective. 8.2.6 Dynamics - What About the Future? The Ecological Footprint provides a snapshot of a population's environmental requirements using current technology under prevailing management practices and social values. Even i f the 205 Global (Loh, 2000) average milk yields may differ substantially from local yields. In New Zealand, for example, the local yield for milk production is 1 ,759 kg ha- I , which compares with a global average yield of 336 kg ha-I . Therefore, applying a global average milk production yield results in a domestic Ecological Footprint contribution that is 5 .24 times the actual land area used for milk production. 269 footprint for a particular population is calculated at regular intervals, the results are always out of date - in this respect the footprinting only tells us 'yesterday's news' . Key dynamic components of the sustainability equation such as intergenerational equity, technological change, and the adaptability of social systems are simply overlooked. Moreover, nature is characterised by complex adaptive systems with non-linearities, feedback loops, and thresholds (Ho I I ing, 1 973; Levin, 1 998). By ignoring such dynamics the Ecological Footprint cannot inform us on the ecological consequences of likely futures, or even possible scenarios. This Chapter makes no attempt to address these issues. 8.2.7 Policy Relevance - A Policy Evaluation Tool? Proponents of the Ecological Footprint (e.g., Wackernagel and Rees ( 1 996), Wackernagel and Silverstein (2000) advocate that the footprint can evaluate potential strategies for avoiding ecological overshoot. The Ecological Footprint is seen as an instrument that provides decision­ makers with "a physical criterion for ranking policy, project or technological options accounting for their ecological impacts" (Wackernagel and Rees, 1 996, p .27). This claim has, however, been hotly contested. Ayres (2000) asserts that footprinting provides no meaningful rank ordering, and even less so any value for policy evaluation. This view is shared by Moffatt (2000, p.360) who notes "it offers no policy suggestions apart from either including more land, reducing population, or reducing consumption per head". Although it is agreed that the policy instruments or actions required to counteract overshoot cannot be implied from the Ecological Footprint method, it is argued here that the footprinting does provide a broad level measurement of ecological impact. In this way, the Ecological Footprint may be used to signal the relative ecological cost of different policy options. Careful consideration of the components of the footprint may also help to evaluate the relative ecological cost of various human activities, enabling policy analysts to identifY 'hotspots' for pol icy action. By far the greatest contribution that footprinting can make to policy and decision-making is as an educative tool to stimulate thinking about the far-reaching nature of the indirect ecological effects of human activities. 8.3 An Input-Output Method for Estimating Auckland Region's Ecological Footprint Much of the Ecological Footprint work undertaken to date is based on methodology that lacks formal structure. Some approaches may even be considered to be ad hoc. A major limitation of such methods is that they may lead to results that are not easily reproduced, either through time or across space. In turn, this restricts comparability or leads to inconsistencies that are more an 270 artefact of the method rather than actual differences. Such concerns led Bicknell et al. ( 1 998) to develop an alternative formulation of the Ecological Footprint based on input-output analysis. 8.3.1 Accounting Identity of the Component Parts of the Regional Ecological Footprint In this Chapter of the thesis, B icknell et al. 's ( 1 998) input-output approach is extended to formally permit calculation of regional ecological footprints and to make explicit interregional appropriation of land. Essentially, the regional ecological footprint is defined by the following accounting identity: EF == a + (fJI + P2 + . . . + Po-I) + 15 (8. 1 ) where: a = land appropriated from within the study region; PI + P2 + . . . + Pn-I = land appropriated from other regions ( l . . . n- 1 ); and 15 = land appropriated from other countries. Fully worked examples of how to calculate each of the three components of regional ecological footprints is outlined in a recently published report by McDonald and Patterson (2003b). In Section 8 . 3 .4 of this Chapter, we only outline how to calculate the land appropriated/rom other regions component (fJI + P2 + . . . + Pn-I), and we refer readers to Bicknell et al. ( 1 998) on how to calculate both the land appropriated from within the study region (a) and the land appropriated from other countries (15). 8.3.2 Generation of Input-Output Tables Regional input-output matrices need to be calculated for the study region and the other regions in the nation. These regional input-output matrices and the data contained in them are then subsequently used in the calculation of each of the main components in the regional ecological footprint identity (refer to Equation 8. 1 ) . In this Chapter of the thesis, these matrices were derived using the GRIT method, which was developed by Jensen et al. ( 1 979) and West et al. ( 1980)206. This method consists of a series of mechanical steps that reduce national input-output coefficients to sub-national (regional) equivalents, while providing opportunities for the insertion of superior data. Such non-survey based methods of generating regional input-output matrices are frequently utilised, as in this 206 Studies that have applied the GRIT method in New Zealand include Butcher ( 1 985), Kerr et at. ( 1 986) and the Ministry of Agriculture ( 1 997). 271 thesis, when time, cost and data constraints preclude generation of matrices based on survey data. 8.3.3 Calculation of the Land Appropriated Within the Study Region (a) The land appropriated from industries within the study region is calculated using the Bickne ll et al. ( 1 998) method. Readers should refer to that paper for full details. Instead of using a national input-output matrix (as did Bicknell et al., 1 998), we used a regional input-output matrix. 8.3.4 Calculation of the Land Appropriated from Other Regions (fJ1 + P2 + . . . + Po-I) The fol lowing five-step process calculates the land appropriated by the study region from other regions: Step 1 Determination of the Regional Imports ($) Matrix, Gr, for Region 1 Each industry in the study region purchases commodities ($) from various regions in the nation. For a given industry in New Zealand, as in most countries, it is not known exactly from which region these commodities originate. This is estimated in this thesis by solving an optimisation problem. It is assumed that each industry within a region will seek to source commodities from supplier regions closest to them in terms of travel time. Thus, minimisation of travel time is to set the objective function, while known levels of industry imports (and exports) are used as the binding constraints207. Fuller details of this optimisation problem are contained in Appendix C. Solving the optimisation problem enables matrices Gr to be defined for each region that exports commodities to the study region - in our study there are 1 5 non-study regions. Matrices Gr define the exact quantities of commodities being imported into each study region industry from each of the non-study regions. 207 This assumes that transport operators will only minimise their travel times, whereas in actuality other factors may also come into play. Nevertheless, analytical tests for the results reported in Section 8 .4 show that the optimisation problem is relatively constrained with a small feasibility space i .e . , differences between optimal and actual flows will be minimal. 272 Step 2 Determination of the Land (ha) Embodied in Regional Imports, Kr, for Region 1 The imports matrix Gr (from Step One) quantifies the imports of commodities into a given industry (in the study region) from industries in Region 1 . These quantities of commodities are enumerated in monetary ($) terms. They are converted to embodied land terms (ha) by: (8 .2) where: Kr = embodied land matrix [i X (j + J)] , describing the land embodied in imports into industries} and into final demandfderived from industries i in Region 1 ; Gr = imports matrix [i X (j + J)], describing the $ imports from industries i from Region 1 into industry} and into final demand f; and Hr = inverse Leontief matrix (i + i), describing the direct plus indirect land requirements from industries i needed to generate an additional unit of output ($) in industry} in Region 1 . Step 3 Determination of the Land Supporting Domestic Consumption, Mr, ha, for Region 1 In the footprint analysis we are only concerned with the Land required to support domestic consumption - not the portion of land that passes out of the study region as land embodied in exports. This is calculated by: (8 .3) where: L r = final demand matrix [(j + J) + U + J)] , describing on the diagonal the fraction of final demand consumed in the study region; and Mr = domestic land consumption matrix [i X U + J)] , describing the land embodied in industry } output and fmal demand f which supports domestic consumption in the study region. The data for i r is calculated from the study region's input-output matrix generated by the GRIT process referred to in Section 8 .3 .2 . Step 4 Repeat Steps One to Three for the Calculation of Energy Land, Region 1 Energy land represents the area of planted forest needed to sequester CO2 emissions resulting from the burning of fossil fuels. The approach used to calculate the energy land appropriated from Region 1 is analogous to that used to calculate the land appropriated from Region 1 . This means steps one to three are now 273 repeated with the exception that matrix Hr is replaced with an inverse Leontief matrix for energy land. Step S Repeat Steps One to Four for all Regions Steps One to Four described the process of calculating the land and energy land appropriated from other regions. This needs to be repeated for all other regions in the nation - in our study this included 1 5 other regions. Once the calculations have been repeated for all other regions, the data then needs to be compiled into one matrix T. Matrix T represents the total land appropriated by each industry (column) according the industry-region combinations (rows). An illustrative example of matrix T is provided by Table 8 .2 . Each of the components of the Equation 8 . 1 expression PI + P2 + . . . + pO-I (land appropriated from other regions) can now be directly abstracted from the Matrix T. For example, PI is the grand total for Region 1 (in Table 8.2) - it is the sum of the column totals (or row totals) which is 2,70 1 hectares. By summing these grand totals for each of the non-study regions PI + pz + . . . + PR-l> the overall land appropriated from all other regions is determined. 274 Table 8.2 I l lustrative Example of the Matrix T: Land Appropriated From Other Reg io ns /31 + /32 +" . + /3n-1 Imports from Study Region Sectors and Final Demand Sectors in Other Regions Agriculture Manufacturing Services Region 1 Agriculture 1 4 653 Manufacturing 0 4 Services 1 3 Sub-total 1 5 670 Region 2 Agriculture 75 2,007 Manufacturing 0 2 Services 4 30 SUb-total 79 2,039 Region 3 Agriculture 591 14 ,465 Manufacturing 0 4 Services 31 Sub-total 592 1 4,500 Total 686 1 7,209 Notes: I . All values are in ha per year unless otherwise stated. 607 5 81 693 1 ,053 2 230 1 ,285 1 ,875 3 8 1 ,886 3,864 Final demand Total 1 ,079 2 ,353 8 1 7 237 332 1 , 324 2 ,702 2 ,094 5,229 5 9 693 957 2 ,792 6, 1 95 1 ,042 1 7 ,973 3 1 0 7 47 1 ,052 1 8,030 5 , 1 68 26,927 2 . Reading the Table: The land embodied in interregional imports into the study region industries and final demand are obtained by reading down the column e.g., the land embodied in agricultural sector imports (from Region I ) into the study region's manufacturing sector is 607 ha, and so forth for the other interregional imports reading down the column. 3 . The values (not including sub-totals and totals) for Region I are the elements i n the matrix that results from adding the matrices Mr (for land) and Mr (for energy land) for that region. The same arithmetic applies for the values for Regions 2 and 3 . 4 . P I = 2,702 ha (total for Region I ), P2 = 6, 195 ha (total for Region 2), p 3 = 1 8,030 ha (total for Region 3 ) . The sum of P I + �2 + p3, which is 26,927 ha in this example, is the total land appropriated from other regions. 8.3.5 Calculation of Land Appropriated from Other Countries (15) The land appropriated from other countries IS calculated using the Bicknell et al. ( 1 998) method. Readers should refer to that paper for full details. Essentially, the Bicknell et al. ( 1 998) method assumes that products imported from overseas have exactly the same embodied land per $ ratio as products made in New Zealand. This assumption is necessary due to the lack of such data for overseas countries, although superior data can be substituted if it is available. 8.3.6 Limitations of Using Input-Output Analysis There are a number of critical assumptions that underpin the method presented in this Chapter which stem directly from using input-output analysis. Rather than recite these assumptions of input-output analysis, which are well documented elsewhere (e.g., Richardson ( 1 972» , we instead focus on those assumptions we consider most relevant in applying input-output analysis 275 to footprinting. We also refer readers to Bicknell et al. ( 1 998) for a more detailed discussion of the limitations of applying input-output analysis to ecological footprinting. In Leontief-based input-output analysis, as described in this Chapter, the homogeneity assumptions require that only one commodity be produced per industry. This does not always occur in reality as industries are often involved in 'joint production' - e.g., a dairy farm may use land to produce not only milk-fat but also lesser amounts of beef or horticultural product. There are, however, input-output methods (e.g., Costanza and Hannon, 1 989) that can be applied to deal with this joint production problem, which could be adapted for use in ecological footprinting. The inter-industry linkages in an economy generally represent flows of physical goods. In input-output analysis such flows are usually summarised in a transactions table denominated in monetary units. However, the use of monetised tables can lead to problems if the price per tonne paid for a given product differs across purchasing industries. If industry 1 , for example, purchases 1 0 kg of goods at 0.20 $kg-1 and industry 2 purchases 1 0 kg of goods at 0. 1 0 $kg-1 from the same industry, then both industries receive the same physical quantity of goods ( l Okg), but spend different amounts of $2 .00 and $ 1 .00 respectively. This implies, from a monetary transaction perspective, that the land embodied in industry 1 purchases is twice that of industry 2 purchases - whereas, from a physical perspective, both industries are purchasing the same physical quantity of goods. This effect may result in both under-and-overestimation of industry contributions made to the Ecological Footprint. The input-output method as presented in this Chapter makes two particular assumptions concerning imported commodities. Firstly, it assumes that imported commodities are essentially final or finished goods. This implies that only backward l inkages through the economy in the region of origin are measured. If, however, there are imported commodities requiring further processing in the study region, then forward linkages may also need to be estimated. Secondly, the method assumes that imported commodities have the same embodied land-to-output ($) ratio as in the domestic economy - this is probably a reasonable assumption in most cases. Ideally, the actual land-to-output ($) ratios for imported commodities should be used in the analysis, but unfortunately such data is rarely available thereby necessitating the use of these surrogate values. 276 8.4 Ecological Footprint of the Auckland Region The methodology described in Section 8.3 was used to calculate Auckland Region's Ecological Footprint and to identify its source-of-origin. Al l calculations are in terms of actual biological productive land areas needed to satisfy domestic final demand, based on local yie lds. Similarly, no adjustments are made for differences in biological productivity between land types when aggregating - i .e . , no equivalence factors are applied. The results presented here are aggregated to facilitate comparison with earlier studies by Bicknel l et al. ( 1 998) and McDonald and Patterson (2003 b). 8.4.1 Brief Description of the Auckland Region The Auckland Region is New Zealand's largest and fastest growing region, with a population of 1 , 1 59,400 in 1 998 (Statistics New Zealand, 1 998h) . Nearly 30 percent of New Zealanders l ive in the Auckland Region. Most of the region's residents l ive in the Auckland metropolitan area. The metropolitan area itself is a sprawling c ity of largely detached single storey dwellings. Geographically the Auckland metropolitan area is located on an isthmus between two natural harbours. From north to south the region measures 1 20 ki lometres and at its widest point is 60 kilometres. The land area of the region is 560,000 ha (2 percent of New Zealand's land area). Auckland Region is New Zealand's commercial hub having the highest proportion of people employed in finance, insurance, property, wholesale trade and business service industries (Statistics New Zealand, 1 998b). The traditional economic base of Auckland Region has been manufacturing, but this has experienced some decline in recent years due to impacts of trade l iberalization and globalisation. 8.4.2 Data Sources Input-output tables for New Zealand's 1 6 Regional Counci l areas are derived from the national inter-industry table produced by Statistics New Zealand using the GRIT method ( 1 99 1 , 1 998i, 1 998f, 1 999c). Each input-output matrix covers 23 industries. Estimates of land use data by economic industry are based on data gathered from Quotable Value New Zealand ( 1 998), Statistics New Zealand (1 998h, 1 998a), Ministry of Agriculture and Forestry ( 1 999), and Works Consultancy Services Ltd ( 1 996). These estimates exclude national parks, lakes, rivers and the marine environment. Energy related CO2 emissions by economic industry were obtained from the Energy Efficiency Conservation Authority ( 1997). The conversion of CO2 emissions into 277 energy land is based on sequestration data obtained from Hol linger et al. ( 1 993). They estimate that an average hectare of Pinus radiata in New Zealand absorbs 3 . 6 t of e, which equates to 0.07 58 ha per t of CO/os. Appendix K describes the methodological processes used in calculating the environmental flows of land and energy into the Auckland Region - a more detailed analysis these flows is also given. Population statistics are based on sub-national estimates produced by Statistics New Zealand ( 1 998k). 8.4.3 Auckland Region's Ecological Footprint Disaggregated by Land Type Agricultural land consists of land used for sheep and beef, dairy, mixed livestock, other farming and horticulture . Auckland Region appropriates 1 ,525,000 ha of agricultural land for domestic use or 1 .32 ha per capita (Table 8.3) . Over half of this land, 805,000 ha, is embodied in agricultural products imported from other New Zealand regions. Only 1 68,000 ha or 1 1 .0 percent of agricultural land was appropriated from within the region, particularly from l ivestock farms in the south of the region. S ignificant amounts of land are also planted in horticultural crops such as onions, spinach, capsicum and Asian vegetables, with lesser amounts in strawberries and persimmons. Table 8.3 Auckland Region Ecolog ical Footprint Disaggregated by Land T�pe, 1 997-98 Land type Within Land from Land from Land type ha per % of land region land other NZ other total capita type total regions nations Agricultural land 1 68 , 000 805,000 552,000 1 , 525,000 1 .32 65. 7 Forest land 6 , 000 45,000 50, 000 1 0 1 ,000 0.09 4.4 Built-up land 96,000 1 7,000 32, 000 1 44,000 0. 1 2 6.2 Energy land 355,000 1 8 ,000 1 77,000 550,000 0.47 23.7 Total 624,000 885,000 81 0,000 2 , 320,000 2 . 00 1 00.0 Note: All values are In ha per year unless otherwise stated. Figures may not add up to the stated totals due to rounding. Forest land refers to forest plantings used for commercial gain. It does not inc lude the hypothetical forest planted to sequester CO2 emissions. Some 94.5 percent of forest land appropriated by domestic final demand is imported into the region. Only 6,000 ha is appropriated from within the region, made up almost entirely of Pin us radiata. On a per capita basis forest land appropriation amounts to 0.09 ha or 4.4 percent of the region's footprint. 208 It is worth noting that these figures may vary considerably between regions depending on plantation age, soil type, climatic conditions and so on. The possibility of planting indigenous forest to sequester CO2 emissions is also ignored (refer to Hall and Hollinger ( 1 997) for further debate concerning this issue). 278 Built-up land represents built-up areas that host human settlements. This includes land used for housing, commercial and governmental purposes209• It accounts for 6.2 percent of the region's footprint and equates to 0. 1 2 ha per capita. Some 96,000 ha or 66 .4 percent i s appropriated from within the region, including 3 8,000 ha for housing. It also captures the region's road network, which exceeds 7,500km or 8 percent of total national road length (Works Consultancy Services, 1 996) . Energy land is a measure of the hypothetical planted forest needed to sequester CO2 emissions. It accounts for 550,000 ha or 23 .7 percent of the region' s footprint. This is relatively low when compared to most developed nations. Loh (2000), for example, estimates Canada' s energy land contribution to be 47.0 percent, Australia's to be 56.4 percent and the United States' at 60.8 percent. Auckland Region's relatively low energy land component can be explained by: ( 1 ) the Pinus radiata CO2 sequestration factor - which is significantly higher than the global average used by Loh (2000), (2) the structure of the Auckland regional economy, which is dominated by less energy intensive light manufacturing and service sectors, rather than more energy intensive heavy manufacturing sectors, (3) the low electricity CO2 emission rate - a result of the h igh proportion of electricity (64.5 percent) drawn from hydro sources. 8.4.4 Auckland Region's Ecological Footprint Disaggregated by Economic Industry The appropriation of land embodied in goods and services purchased by the region's primary sectors (agriculture, forestry, fishing and hunting, and mining and quarrying) accounts for 1 68,000 ha which represents 7 .2 percent of the region's footprint (Table 8.4). The agriculture sector appropriates most of this land in the form of internal transactions e.g., sales and purchases of stock from within the agriculture sector. 209 Large commercial and governmental land users include office and retail space, warehouses, amusement parks, holiday parks, car parks, schools, hospitals, prisons and military bases. 279 Table 8.4 Auckland Region Ecolog ical Footprint Disaggregated by Economic Indust!1. 1 997-98 Economic sector Within Land from Land from Economic ha per % of region land other NZ other sector total capita economic regions nations sector total Agriculture 1 05,000 37,000 5,000 1 47,000 0 . 1 3 6.3 Forestry 2,000 1 8, 000 0 20,000 0.02 0.8 Fishing and hunting 0 0 0 0 0 . 00 0.0 Mining and quarrying 0 0 0 1 , 000 0.00 0.0 Manufacturing 1 09 , 000 794,000 224,000 1 , 1 27,000 0.97 48.6 Utilities and construction 44,000 1 6,000 35,000 96,000 0 .08 4 . 1 Services 1 95,000 7 , 000 1 66,000 369,000 0 . 32 1 5 .9 Domestic final demand 1 69,000 1 3, 000 379, 000 562 ,000 0.48 24.2 Total 624,000 885, 000 8 1 0,000 2,320,000 2 . 00 1 00.0 ote: All values are in ha per year unless otherwise stated. F igures may not add up to the stated totals due to rounding. On a sectoral basis, the largest appropriation of land is by the manufacturing sector ( 1 , 1 27,000 ha), accounting for 48.6 percent of the region's footprint. Of this amount, 794,000 ha are embodied in imports from other New Zealand regions. These imports largely represent backward linkage purchases of land embodied in agricultural products. This is not surprising given that the region's agriculture sector is unable to satisfy local demand2 1 0. Of the 1 09,000 ha appropriated locally by the manufacturing sector, most is used in the manufacture of food and beverages, with lesser amounts in the production of textiles and clothing. Auckland Region's service sector appropriates 369,000 ha or 1 5 .9 percent of the region's footprint. Interestingly, this is almost 8.0 times greater than the actual land area occupied by the service sector. This is because service industries are typically at the end of the value-added chain and therefore they tend to have s ignificant backward linkages, many of which appropriate land. Domestic final demand, although not strictly an industry, is included in Table 8.4 for completeness. This category encompasses land occupied by housing, land embodied in goods and services purchased by households, and smaller amounts of land embodied in non­ marketable governmental goods and services consumed by the region's communities. Households appropriate 562,000 ha of biologically productive land, which is almost one quarter of the region's footprint. Most of this land is embodied in consumer goods purchased from 210 The respective employment location quotients for agriculture and forestry relative to the national are estimated to be 0.24 and 0.28 . 280 other nations. This includes goods that are imported directly by wholesalers and retailers and then resold to households without further processing. 8.4.5 Auckland Region's Ecological Balance of Trade An input-output formulation of the Ecological Footprint permits an 'ecological balance of trade' to be calculated on a systematic basis. Auckland Region's ecological balance of trade by land type is depicted in Table 8 .5 . This Table indicates that 2,509,000 ha of land were appropriated from outside the region (this includes the 1 ,696,000 ha supporting domestic consumption), while 1 ,089,000 ha were embodied in exported goods and services. Thus, the Auckland Region is a significant net importer ( 1 ,420,000 ha) of biologically productive land. Table 8.5 Ecological Balance of Trade for the Auckland Region Disaggregated by Land Type, 1 997-98 Land type Interregional trade Agricultural land Forest land Bui lt-up land Energy land Interregional balance of trade Intemational trade Agricultural land Forest land Bui lt-up land Energy land International balance of trade Balance of trade Note: Land embodied in Land embodied in Balance of trade imports exports 1 ,293,000 1 4 ,000 - 1 , 2 78, 000 1 24,000 5 ,000 - 1 1 8 , 000 1 9 ,000 1 ,000 - 1 8 ,000 28,000 9 , 000 -20,000 1 ,464,000 29,000 - 1 ,434,000 7 1 1 ,000 72 1 ,000 1 0 , 000 63,000 94,000 32 ,000 59,000 2 1 ,000 -38 ,000 2 1 3,000 224,000 1 1 ,000 1 ,046,000 1 ,060,000 1 4 ,000 2,509, 000 1 ,089,000 - 1 ,420,000 All values are in ha per year. F igures may not add up to the stated totals due to rounding. As previously noted, it is not only important to evaluate solely the total quantity of land appropriated from outside the region, but also to consider where this land originates. At its source-of-origin, the land may be farmed unsustainably (e.g., overcropping or accelerated erosion caused by sub-standard farming practices) or it could be used sustainably (e.g., by using nutrient recycling or minimising fossil fuel inputs). Or perhaps the land could be drawn from a source nation where land is scarce and there is much poverty . In terms of sustainabi l ity (and equity) arguments, these factors relating to the source-of-origin of land could be more important than just considering the total quantity of appropriated land - viz, it is not just the s ize of the 28 1 footprint that counts, but also where and how the footprint falls, and what impact it has on the environment. Figure 8. 1 graphically depicts the origin of biologically productive land imported into Auckland Region. Approximately 1 5 .5 percent (227,000 ha) of the land appropriated from other regions is embodied in products imported from the Waikato region. Otago2 1 l , Northland and Southland regions also make significant interregional contributions. Auckland Region's manufacturing sector is the greatest appropriator, accounting for some 87.6 percent of the land embodied in interregional imports. The size of the contribution made by Northland and Waikato regions is not surprising given that they are Auckland Region's c losest neighbours. More influential, however, is the role these regions play in providing agricultural product to the Auckland Region food manufacturing industries. The rich pastures of the Waikato region support intensive farming, with 75 . 3 percent ( 1 ,500,000 ha) of its bio logically available land in grazing, arable or fodder use. Similarly, Northland region is a major producer of sheep, beef and horticultural products. 21 1 The size of the contribution is due primarily to the low productivity of much of the region's agricultural land. That is, agricultural products appropriated from Otago supporting Auckland Region 's domestic consumption embodied far more land than similar products purchased from other regions. 282 Figure 8 . 1 t LAND (ha) .""� • 150 000 .. 100 000 ... 50 000 - 10 000 5000 1000 100 200 km Regional and International Origins of Auckland Reg ion's Ecological Footprint, 1 997-98 The ongms of forest land embodied in interregional imports reflects the spatial location of major planted forests in the North Island. The Waikato and Bay of Plenty regions form part of the largest planted forest area in New Zealand, mostly Pinus radiata although small plantings of Douglas fir and other varieties do exist. Lesser, but stil l significant, forest p lantations also exist 283 in the Northland and Gisbome regions. Auckland Region' s s izable construction, printing and publishing, and other manufacturing sectors drive the demand for forest products from these hinterland regions. 8.4.6 Comparing Auckland Region's Ecological Footprint with Other Regions The Auckland Region' s Ecological Footprint is compared with other New Zealand regions in F igure 8.2. It shows in relative terms each region's actual land area (on the left) alongside its corresponding appropriated footprint area (on the right). Overall , Auckland Region has the largest footprint of any region, in excess of 1 .3 times Canterbury, the next largest region. N + Actual land Figure 8.2 Ecological Footprints of New Zealand Regions, 1 997-98 New Zea la nd Regions III Auckland � Bay of P len ty .. Canterbury � Gisborne � Hawke 's Bay � Manawatu-Wangan u i Marlborough • Nelson ij j North l an d Otago � South land mm Taran aki Tasman Waikato Welli ngton � West Coast 1 00 0 1 00 200 300 Ki lometers - - - - - 285 Auckland (2.00 ha per person) along with Wellington (2.40 ha per person), and Nelson ( 1 .86 ha per person) are among the lowest per capita footprints in New Zealand (Figure 8.3). These are the three most urban regions in New Zealand, and this seems to be the main determinant of their low footprints. Urban settlement and consumption patterns are more efficient in their use of land - viz, land requirements per capita for retailing, housing, infrastructure and transport are considerably lower in urban areas compared with rural areas. There is also some evidence that urban transport requirements are relatively low, thereby reducing the size of the energy component of the ecological footprint for these - this is particularly so for Wellington that has an energy efficient public transport system based to a large extent on l ight rail. 6 00 5.50 5 00 4.50 4.00 !! 'g. 3.50 " lI. 3.00 .. � � 2 50 " .r::: 2.00 1 50 1 . 00 0.50 0.00 5.41 o en cS Figure 8.3 , 4 1 3 3 92 3 80 3.70 3.57 3 33 3.08 3 03 i 2.87 i 2.63 2.59 I 2 40 2 . 19 2 08 2.00 1 86 , I -' ·s '" � '0 '0 '" 0 .,.. � c X C '0 c c '" c c E � '" c 0 '" '" c g '" 0 fl '" '" 0 iD " � c E '" C> u :c - « u '" " " I '" ro z iD " '" c '" :2 Comparison of Auckland Region Ecological Footprint Per Capita with Other Regions in New Zealand, 1 997-98 Otago has the highest footprint per capita of any region in New Zealand, at 5.4 1 ha per person. This can be mainly attributed to the low productivity of Otago land, which is the 2nd to lowest of any region in the country. This means that Otago requires significantly more land to produce the same amount of output as other regions. Marlborough has the second highest footprint at 4. 1 3 ha per person, again largely attributable to the region'S low land productivity. Both of these regions also have relatively low population densities meaning that the spread-out nature of their settlement increases travel distances and hence the size of their footprints. Southland (3.92 ha per person) and Manawatu-Wanganui (3 .80 ha per person) rank 3rd and 4th in tenus of the size of their per capita footprint. While parts of these reglons are highly 286 productive, other parts, particularly mountainous areas with harsh climates, have extremely low productivities. This is the main reason why these regions have reasonably h igh per capita footprints. Southland also has the highest per capita Energy Land footprint component of any region due to its colder climate. West Coast (3 .70 ha per person), Canterbury (3 .57 ha per person), Northland (3 .33 ha per person), Gisborne (3 .03 ha per person) and Hawke's Bay (2.63 ha per person), all have footprints around the New Zealand average which would be expected on the basis of their land productivities. Waikato (2 .87 ha per person), Bay of Plenty (2. 59 ha per person), Taranaki (2. 1 9 ha per person), and Tasman (2 .08 ha per person) all have per capita footprints below the New Zealand average. These regions have relatively h igh land productivities (ranked 1 st to 4th). It is therefore not surprising that the per capita footprints of these regions are among the lowest. The spread-out nature of the Waikato and Bay of Plenty settlements, which are less urban than some other regions, may explain why the footprints of these regions are not lower than stated. 8.4.7 Comparing Auckland Region's Ecological Footprint with International Ecological Footprints Auckland Region's Ecological Footprint can be compared with 1 996 international footprint estimates produced by Loh (2000). This requires that Auckland Region's footprint be adjusted for: ( 1 ) global average yields21 2; (2) b iological equivalence factors2 1 3 and (3) the application of a global average CO2 sequestration factor214 . On this basis, Auckland Region's footprint of 5 .68 ha per capita was found to be significantly smaller than both New Zealand (8 . 3 5 ha) and Australia (8.5 0 ha), slightly smaller than Japan (5.90 ha), but larger than South Africa (4.04 ha) and Argentina (3 .80 ha) (refer to Figure 8.4). 212 Loh (2000) estimates New Zealand's average pasture yield factor to be 5 .24, with the average yield factors for arable and forest land estimated to be 2 .09 and 0 .6 1 respectively. In the case of built-up land the average arable yield factor is applied. 213 The fOllowing equivalence factors based on Loh (2000) were applied: for energy land 1 .78, for arable land 3 . 1 6, for forest land 1 .78 and for pasture land 0.39. The equivalence factor for arable land was used as a proxy for built-up land. 214 Loh (2000) estimates the world average carbon absorption (including roots) to be 0.956 t ha·l • In accordance with Loh (2000) oceans are also assumed to take up 35 percent of CO2 emissions. 13 0 12.25 12 0 1 1 0 10.0 9.0 i 8.0 .. " � " 7.0 a. I 6 0 " 5.0 "" 4 0 3.0 2 0 1 .0 0 0 Figure 8.4 10.51 11.53 8.50 8.35 7.68 7.30 7 .14 '-..,- 0> c 0 '" '" " :i' 6.26 6.26 5.98 5.90 '-..,- '-.....- L..,-- '-..,- ,.., E '" c " 0 " '" '" u c a. E '" '" '!l c " '" '" Cl s: .<:: '" <> 2 z C ::> 287 I 5 68 4 04 3.80 3.68 ., i'l � .E '" < � - '" '5 0> "" 0 "' U (f) � .2 '" '" '" '" c vn then plans to increase the capital-output ratio will be less than plans to save and growth will drop away. By contrast, if s < vn then plans to increase the capital-output ratio will exceed plans to save and 222 In retrospect the imminent doom predicted by Malthus and Ricardo did not eventuate - primarily due to their underestimation of the influence of technology and substitution effects. Of course, Malthus and Ricardo are not alone in making this error - the highly popularised Lim its to Growth study (Meadows et aI., 1 972) also fell short in this respect. Unfortunately, this has led many economists to assume that backstop technologies will always exist to overcome neo-Malthusian predictions - this is despite binding �hysical constraints imposed by the laws of thermodynamics. 23 The work of Ramsey ( 1 928) on a mathematical theory of saving is considered by some commentators to be a key impetus behind Harrod's ( 1 939) efforts - refer to Asimakopulos and Weldon ( 1965). 3 1 2 growth may run away.224 Thus, economies appeared to be balanced on a 'knife-edge' (Bretschger, 1 999; Thirlwall, 2002; Panico, 2003). The major criticism of the Harrod-Domar model is that the predicted instability facing many economies never eventuated - in fact, empirical studies have shown that many western economies exhibit near steady state growth (So low, 2000). This is because s , v and n are not simple constants as proposed by Harrod, but variables characterised by continuous change.225 If it is assumed that changes to n result only in scale effects, it may then be argued that economic growth/decline is a result of either: ( 1 ) changes to the savings rate, S , or, (2) changes to the capital-output ratio, v . In a debate that would last many years, arguably without resolution, Keynesian economists such as Nicholas Kaldor226, Joan Robinson, Richard Kahn and Luigi Pasinetti argued in favour of the savings rate, while neo-classical economists such as Robert Solow, Paul Samuelson and Franco Modigliani supported the capital-output ratio (So low, 1 988, 2000; Bretschger, 1 999; Thirlwall, 2002). Given its dominant and widespread use today, the remainder of this Section focuses on the neo-classical growth model.227 10.2.2 The Solow Model In 1 956, Robert Solow published A Contribution to the Theory of Economic Growth - this was to become a landmark paper on economic growth (Jones, 1 998, 2002). For this work, and subsequent contributions to the understanding of economic growth, he was awarded the economics Nobel Prize in 1 987. Solow's models focused on the importance of technology rather than the savings rate as the key determinant of economic growth. Solow ( 1 956) and 224 This condition, together with the preceding one, represents the so-called 'Harrod instability problem' (So low, 2000; Thirlwall, 2002). 225 The Harrod-Domar model assumes that s , v and n exist independently of each other i.e. possible dynamic feedbacks are overlooked. This criticism also applies to the neo-c1assical models that are shortly described. 226 Nicholas Kaldor, a prominent Keynesian economist, has argued that it is difficult to understand economic growth without taking a sectoral approach (Kaldor, 1 96 1 ; Thirlwall 2002; Panico, 2003). He emphasises the unique role p layed by manufacturing industries in generating economic growth through increasing returns, but also acknowledges that this growth can only occur through the physical transformation of raw materials, typically provided by farming and mining industries with diminishing returns to scale. Kaldor ( 1 96 1 ) argues that agriculture provides the initial impetus for economic growth in industry. A fine balance ensures that industrial growth is neither supply-constrained due to agricultural prices being too high relative to industrial prices, or demand-constrained because they are too low. It is argued that through time the importance of agriculture as a market for industry products will decline in favour of export based markets. 227 Developments in growth theory are not solely limited to the contributions of Keynesian and neo­ classical economists. Other theories have also been developed. Nelson and Winter ( 1 982) have, for example, developed an 'evolutionary economics' model of economic growth (S immie, 200 1 ; Loeschel, 2002). Of course, a comprehensive review of all contributions made in economic growth theory would be a monumental task - well beyond the scope of this thesis. Readers are instead referred to publications such as Aghion and Howitt ( 1 998), Solow (2000), Foster and Metcalfe (200 1 ), Salvadori (2003), Helpman (2004), and Aghion and Griffith (2005). 3 1 3 Swan ( 1 956) were also the first to emphasise long run economIC growth (Panico, 2003). Solow's models are built on two equations: ( 1 ) a production function, and (2) a capital accumulation equation . The production function, presented i n Equation 1 0. 1 , i s o f a Cobb-Douglas228 form, ( l 0. 1 ) where capital (e.g. tools, machinery and factories), K , and labour, L , combine to produce output, Y . It is assumed that the e lasticities of output with respect to capital, a , and with respect to labour, 1 - a , exhibit constant returns to scale (if factor inputs double then output will also double) i .e. a + (1 - a) ::::: 1 . The production function of Equation 1 0 . 1 is typically y rewritten in terms of output per worker (i.e. as a measure of labour productivity), Y = - , and L K capital per worker (i.e. as a measure of capital productivity), k = - , i .e., L C l 0 .2) Given more capital a worker will therefore produce more output. But there are diminishing returns to capital per worker i .e. each additional unit of capital given to a worker increases the output of that worker by less and less (Solow, 1 956, 1 95 7, 2000; Jones, 1 998, 2002; Helpman, 2004). Capital accumulation is the focus of the second equation of the Solow model, K = sY - dK , ( 1 0.3) where the change in the capital stock over time, K , equates to gross investment, s Y , less depreciation, dK .229 It is assumed that workers save a constant fraction, s , of their income230 228 Cobb and Douglas ( 1 928) proposed this production function while undertaking an analysis of US manufacturing (Thirlwall, 2002). They found that a value of Y. for a mirrored empirical findings - without taking into account technological progress. 229 'Dot' notation is used to denote a derivative with respect to time i.e. k := dK . dt 3 14 and that this saving is completely invested within the economy to accumulate capital23 1 • Workers rent this capital to ftrms for use in production. It is further assumed that the capital stock depreciates by a constant fraction, d , every period.232 Neo-classical economists typically rewrite the capital accumulation equation in terms of capital per worker, le = sy - (n + d)k , ( l OA) where the population growth rate, n , is assumed to be a constant.233,234 This equation te lls us that the change in capital per worker is a function of three terms: ( 1 ) investment per worker, sy , increases k ; (2) depreciation per worker, dk , reduces k ; and (3) the population growth rate, n , also reduces k . To study economic growth neo-classical economists track how output per worker changes over time. A so-called 'Solow' diagram is often used for this purpose (Jones, 1 998, 2002; Thirlwall, 2002). The So low diagram depicted in Figure 1 0. 1 consists of two curves drawn as functions of the capital-labour ratio, k : ( 1 ) investment per worker, sy , and (2) the amount of new investment required to keep capital per worker constant, (n + d)k (Jones, 1 998, 2002). The transformation in capital per worker over time may be described using the difference between the curves. If sy > (n + d)k , as at point ko , then the economy is increasing its capital per worker or 'capital deepening, .235 This continues until the amount of capital per worker remains constant i .e. k = k. . Economists label this point the steady state. 230 Workers earn income from wages and salaries, w , and rental income from capital investments, r , such that across the entire economy Y = wL + rK . The goal of firms is to profit maximise output, Y , after making payments for both labour, wL , and rents for capital, rK . 23 1 It is assumed here that the economy is closed i .e. the economy does not trade with other economies. 232 Most neo-classical growth models assign d a value of between 0.05 and 0 . 1 0 (Jones, 1 998, 2002). 233 Two mathematical steps are required to rewrite the capital accumulation equation in terms of capital per worker: ( 1 ) economists ' take the logs and then derivatives' of the capital per worker equation, and then (2) combine this result with the capital accumulation equation. For full details refer to Iones ( 1 998, p . 23 and p . 1 68). 234 The population growth rate is used as a proxy for the labour force growth rate - it is also assumed that the labour force participation rate remains constant over time. 235 If ko was positioned to the right of k (i.e. sy < (n + d)k ) then investment per worker would be less than the amount required to keep capital per worker constant. This situation is known as 'capital widening' . 3 1 5 (n + d k sy k' I< Figure 1 0.1 The Solow Diagram. Source: lanes ( l998, p.2S) Careful consideration of this diagram reveals several key findings on economic growth. Firstly, an increase in the investment rate will only raise capital per worker temporarily. The economy will however be richer. Secondly, an increase in the population growth (or capital depreciation) rate wil l only temporarily lower capital per worker, but the economy as a whole wil l be poorer. F inally, a corollary of the preceding two findings, as currently formulated, is that it is impossible to derive continuous per worker growth in the model - a consequence of the diminishing returns exhibited by the individual factors of production. In other words, an economy may grow for a while, but not forever (Jones, 1 998, 2002). Neo-classical economists argue that ongoing growth may only be achieved through technological progress. 10.2.3 The Solow Model with Technology Technological progress is introduced into the Solow model to offset the diminishing returns to capital accumulation i.e. enabling labour productivity (output per worker) to grow ad infinitum. Through technology the amount of output generated from a given set of factor inputs may be increased. It is argued therefore that the long term growth rate of the economy is thus equal to the rate of technological progress. To account for technological progress a technology variable, A , is added to the production function of Equation 1 0. 1 236,237: ( l 0.5) 236 The technology variable added in this way is known as 'labour-augmenting' . Other possibilities include: 'capital-augmenting' (i.e. Y = (AKt i-a ) and ' Solow-neutral' (i.e. Y = A(Ka LI-a ) ). 237 Representing technological change as a single variable is very rudimentary. Authors such as Perrings ( 1 987) and O'Connor ( 1 993) have rigorously defined technology using a matrix translating factor inputs into outputs across several industries. 3 1 6 Technological progress occurs when the technology variable, A , increases over time. It i s here that the key assumption of the Solow model is revealed. Technological progress is considered to be an exogenously determined constant i .e . it is assumed to be independent of the functioning of the rest of the economy. Or, put alternatively, there are no feedbacks between economic activity and technological advancement. Of course, this has been hotly contested in the literature - in fact, attempts to endogenise technology into models of economic growth provided the impetus behind the emergence of so-called 'New Growth Theory' (refer to Section 1 0.3 below). Incorporation of technological progress into the Solow model is undertaken as fol lows (for full mathematical details refer to Jones ( 1 998, pp.32-35» . Firstly, we make the assumption that A is growing at a constant rate, ( 1 0.6) where g denotes the growth rate of technology. Secondly, since k is no longer constant in the long term due to the presence of technological change, we must defme another variable to consider steady state conditions. Neo-classica1 economists typically select the ratio of capital per worker to technology (known as the 'capital technology' ratio), for this purpose i .e. k = K . Finally, equivalents of Equations 1 0.2 and 1 0.4 must be generated as a consequence AL of using the capital technology ratio. The production function becomes, ( 1 0.7) h - y - h w ere y = - . The variable y is known as t e 'output-technology' ratio. The updated AL capital accumulation equation becomes, k = sji - (n + g + d)k . ( 1 0.8) Once again the transition dynamics of, say, a change in investment pattern through capital injection, or of a policy change which results in positive net migration, may be traced using a Solow diagram. Of course, the effect on long run growth is the same as occurred in the Solow 3 1 7 model without technology. Changes brought about by investment or policy may only temporarily increase the economic growth rate i.e. unti l a new steady state is reached. Investment and policy changes result only in level effects - a raising or lowering of the output per worker, but not in continued growth. In the long term, the economic growth rate of the economy in the So low model with technology will approximate the rate of technological progress. A Note on Growth Accounting In 1 957 Solow published Technical Change and the Aggregate Production Function.238 Solow demonstrated an accounting exercise which disintegrated growth in output into capital based growth, labour based growth and growth attributable to technological change (Jones, 2002; Helpman, 2004). The rate of growth of output that is given to a particular input equals the input's share of GDP multiplied by the rate of growth of the input. In this way, the contribution of all inputs is the weighted average of the growth rates of the inputs, in which the weight of every input equals its share of GDP (Helpman, 2004). Since Solow ( 1 957) the work has been refined by authors such as Jorgenson and Gril iches ( 1 967) and more recently, Young ( 1 995) and Jorgenson and Yip (200 1 ). Despite growth accounting being useful in decomposing growth, it unfortunately reveals little about the causes of growth. 1 0.3 Endogenous or New Growth Theory Unlike the neo-c lassical growth models of Solow where the emphasis is on the mere effect of technological change (i.e. through the use of an exogenous variable), endogenous or ' new growth theory' attempts to understand how technological progress occurs (Grubb et al., 1 995 ; J ones, 1 998, 2002; Thirlwall , 2002). Endogenous growth theory builds on the recognition that technological innovation is in its own right an economic activity - arising from the efforts of profit-maximising activities (Loesche1, 2002). Research on endogenous growth theory is very much in its infancy (Loeschel, 2002; Thirlwall, 2002). Notable contributions to date have included the work of Romer ( 1 986, 1 990), Lucas ( 1 988), Grossman and Helpman ( 1 99 1 , 1994), Aghion and Howitt ( 1 992) and Young ( 1 993). The focus of these contributions has been on explaining technological change from four closely re lated perspectives239: 238 Abramovitz ( 1 956) had previously alluded to the possibil ity of growth accounting. Nevertheless, it was Solow ( 1 957) who provided the analytical framework for its operationalisation. 239 Economists often label the flow-on effects of technology as a positive externality. An externality is a consequence of an activity that is not fully accounted for in a market (Bannock et al., 1 992). 3 1 8 • Research and development. Romer ( 1 986, 1 990) was among the first to suggest research and development as a key driver of technological change.24o Robert Lucas ( 1 988), another Nobel Prize winner, has focused on the effects of human capital formation through education.24 1 Corporate investment in research and development is often in response to market conditions. • Spillovers. Grossman and Helpman ( 1 99 1 , 1 994) have concentrated on the technological spi llovers from trade and investment. Spillovers occur when knowledge is created in one economic activity that may be of use, or indirectly (often without payment) leads to technological improvements, in other economic activities. Gril iches ( 1 992) has demonstrated the empirical significance of spil lovers. Spillovers may occur within industries or between industries. At a regional level spillovers may occur within a region, between regions, and with other nations. • Creative destruction. Faber et al. ( 1 990) and Aghion and Howitt ( 1 992) have generated models of economic growth based on Schumpeter's ( 1 942) notion of 'creative destruction' . Schumpeter ( 1942) distinguishes three stages of the process of technological change ( 1 ) invention, (2) innovation - the transformation of the invention into a commercial product, and (3) diffusion - the process of gradual adoption of the innovation from niche market to widespread use. • Technology learning. Young ( 1 993) builds on the earlier work of Arrow ( 1 962) on ' learning-by-doing' . Empirical studies have shown that the cost of producing a commodity decreases as a function of cumulative production i .e. we learn how to do things at less cost the more often we do it (Ryan, 1 995; Loeschel, 2002; Castlenuovo et al. 2005). Technology learning, or learning-by-using, is a more recent derivative of learning-by-doing - it refers to worker uptake of new technologies through their use. Learning-by-doing occurs most often in the innovation phase of technological change, while learning-by-using occurs during the diffusion phase (Rosenberg, 1 982). 240 Romer ( 1 986) and Lucas ( 1 988) are often credited with sparking the so-called 'first wave' of research into endogenous growth theory, while Romer ( 1 990) is credited with initiating endogenous growth theory's ' second-wave' . The key difference between Romer's two papers, and between the first and second waves of endogenous growth theory, lies in the movement from explaining growth using knowledge accumulation in an aggregate macro economy, to explaining growth using the detailed micro mechanics of ideas through the interaction between a research sector, an intermediate capital manufacturing sector and a final goods producing sector. This weaving together of macro and micro economies is arguably the greatest contribution made by the new growth theorists. 241 Uzawa ( 1 965) had however previously developed a human capital driven model of productivity improvement. 3 1 9 10.3. 1 The Romer Model The essence of new growth theory is perhaps best captured by considering one of its principle models. The Romer ( 1 990) model is se lected here for two key reasons: ( 1 ) it is one of the first endogenous growth models to be developed; and (2) the model 's salient features have appeared in many subsequent models. Beginning in the 1 980s, Paul Romer, over a series of papers, formalised the relationship between the economics of ideas and the economics of growth (Romer, 1 986; Jones, 1 998, 2002; Helpman, 2004). A ful l account of these developments is provided in Romer's ( 1 990) paper entitled Endogenous Technological Change. Romer's ( 1 990) endogenous growth theory is founded on two key observations. Firstly, ideas are inherently nonrivalrous i .e . once conceived anyone can apply them. By contrast, most economic commodities are rivalrous - the use of a commodity by one person prohibits its use by another. Secondly, ideas are at least partially excludable. A commodity is excludable if its owner can prevent others from using it. Excludable commodities allow their producers to capture the benefits they produce, while unexcludable cornrnodities often have spillover benefits not captured by their producers.242 Copyrights and patents are often used to control the degree of excludabil ity of ideas?43 Ideas are only partially excludable because they often create benefits that the owner cannot completely capture. The nonrivalry and partial excludability of ideas necessitates that their transformation into commodities may only occur in an imperfect market with increasing returns to scale. This is because the incentive to create a new idea is typically dependent upon the inventor earning profits from it. The transformation from idea to commodity involves a fixed cost of production, but zero marginal costs. Of course, nonrivalrous ideas are typically embodied in manufactured rivalrous cornrnodities for sale purposes - which are characterised by marginal costs. Nevertheless, the important point here is that an idea will generally be transformed into a commodity if the private benefits of a commodity exceed its one-time invention costs. This necessitates the presence of increasing returns to scale and imperfect markets. 242 Commodities with positive spillovers tend to be underproduced by markets (government intervention is often required here to provide such commodities e.g. education, healthcare, infrastructure provision), while commodities with negative spillovers are often overproduced/utilised by markets (government intervention is required here to avoid the 'tragedy of the commons' problem). 243 Douglass North ( 1 98 1), another Nobel Prize laureate, noted that until the advent of robust property rights there had been little incentive for inventors to develop new methods of production or advanced consumer commodities. It can therefore be argued that an idea with a high social rate of return will not be transformed into a commodity unless legally protected private benefits exist. 320 In the Romer model technological progress is driven by the research and development process - in particular, the search for new ideas by researchers seeking profit from their inventions (Jones, 1 998, 2000; Romer, 1 990). The Romer model consists of two main elements: ( 1 ) a production function, and (2) a set of equations describing how factor inputs evolve over time. Unlike the Solow model however, the aggregate production function describes how the stocks of capital, K , and labour devoted to producing economic output, Ly , combine to produce output based on a stock of ideas, A . The production function is modelled as: ( 1 0 .9) The production function exhibits increasing returns to scale as a result of the inclusion of the stock of ideas, A. The increasing returns to scale are a consequence of the nonrivalrous nature of ideas (Jones, 1 998, 2002). The equations describing how capital and labour evolve over time are simi lar to those of the So low model. Firstly, the capital accumulation equation, ( 1 0. 1 0) where Sk denotes a given savings rate. Secondly, the labour accumulation equation, 1 - = n L ' ( 1 0. 1 1 ) which assumes that labour grows at the exogenous population rate, n . The Romer model, unlike the Solow model, endogenises technological progress. To complete the model an equation is therefore required to describe how the stock of ideas, A , changes over time, ( 1 0. 1 2) where La is the labour devoted to producing new ideas244 and "8 is that rate at which new ideas are discovered. Of course, it is unlikely that 5 is constant. Careful consideration of how ideas 244 The total labour force, L , assuming full employment, equates to L plus L . y Q 32 1 are discovered reveals that 8 i s probably a function of the stock of ideas (Romer, 1 990). If, for example, the pool of ideas in the past raises the productivity of current researchers then 8 would be an increasing function of A (Jones, 1 998, 2002). Or alternatively, if the most obvious ideas are invented first and subsequent ideas are increasingly more difficult to discover then 8 would be a decreasing function of A (Jones, 1 998, 2002). This line of thinking suggests, ;5 = 8ArP , ( 1 0. 1 3 ) where rjJ and 8 are constants. The term rjJ indicates the productivity of the research i .e. if rjJ > 1 then the productivity of researchers i s an increasing function of A - a positive knowledge spillover exists245; if rjJ < 1 then the productivity of researchers is a decreasing function of A - the discovery of new ideas becomes harder over time?46 Romer ( 1 990) also points out that the productivity of researchers may also depend on the number of people engaged in idea discovery. Duplication of ideas, for example, may occur with increasing numbers of researchers. Or, alternatively, the formation of research teams may accelerate the creation of ideas. To this end, Romer ( 1 990) modifies La to include a parameter, A , for idea creativity - where A is a constant between 0 and 1 . We may now combine La A. with Equations 1 0. 1 2 and 1 0. 1 3 to obtain: ( 1 0. 1 4) Equation 1 0 . 1 4 completes the Romer model . The key feature of this equation, and of the Romer model, is the feedback loop between the change in the stock of ideas over time and the stock of ideas itself. It is this feature that endogenises technological change within the model . Overall, the long run growth rate in the Romer model is determined by Equation 1 0 . 14, which in itself is ultimately a function of population growth rate - in order to generate growth the number of ideas must be expanding over time, which can only occur with population growth.247 Changes 245 In reference to Isaac Newton's famous statement, "If I have seen farther than others, it is because I was standing on the shoulders of giants", Jones ( 1 998, p .93) refers to this positive knowledge spillover as the "standing on the shoulders of giants" effect. 246 If I/J = 1 then the positive knowledge spillovers exactly offset the increase in difficulty of discovering new ideas. 247 If the productivity of research is proportional to the existing stock of ideas the long-run growth may be sustained. Under this situation the productivity of research grows over time despite the number of 322 in the investment rate or policy initiatives aimed at subsidising research, as with the Solow model, result only in level effects i.e. growth occurs along a transition path until a new steady state is reached. It should be noted that during the mid-to-late 1 960s economists such as Uzawa ( 1 965), Phelps ( 1 966), Shell ( 1 967) and Nordhaus ( 1 969) developed s imilar growth models. The distinguishing feature of Romer' s ( 1 990) contribution was to explain how technological progress could be modelled through a path involving the creation of new designs, the legal patent of these designs, sale of the designs to manufacturers, production by manufacturers, and ultimately final consumption. Navigating this path was according to Jones ( 1 998, 2002) only possible however once economists such as Spence ( 1 976) and Dixit and Stiglitz ( 1 977) better understood the microeconomic basis for imperfect competition in a general equi librium setting. 10.3.2 Other Endogenous Growth Models Several other endogenous growth models have also been developed. It is beyond the scope of this thesis to investigate all such models. Nevertheless, two further commonly cited endogenous growth theory models require mention. First, the so-called 'AK' model . In this model the production function for output generation takes the following form, Y = AK , ( 1 0 . 1 5) where A is some positive constant i .e. a = 1 . Under this model Y is l inear in K - the key property of the AK model . I f we assume that capital accumulates as per Equation 1 0.4 then the growth rate of the economy under the AK model wil l be an increasing function of the investment rate. Pol ic ies that increase the investment rate will therefore increase the growth rate of the economy. Of course, this is the case where the transition dynamics never end. Unfortunately, no theoretical basis can be given for setting a to 1 . Moreover, empirical studies certainly do not corroborate a l inear model. The second model, created by Robert E. Lucas Jr (the 1 995 Nobel laureate in economics), is based on human capital. The Lucas ( 1 988) model assumes a production function very similar to Equation 1 0. 1 , researchers being constant. This improbable case produces a model similar to the AK model described below - it is also characterised by similar limitations. 323 where h is a measure of 'human capital ' per worker. Human capital evolves over time according to the following differential equation, h == (1 - u)h , ( l 0. 1 7) where u is time spent working, and 1 - u is time spent accumulating skills. Growth in human capital is therefore a function of time invested in generating skills. By corollary, economic growth must thus also be a function of time invested in generating skills. The effect of policy changes or interventions on growth in the Lucas ( 1 988) model will be as per the Solow model with technology - as a consequence of the labour-augmenting nature of technology/human capital. The Romer ( 1 990) model, and AK and Lucas ( 1 988) models, typity the two main pathways that neo-classical economists have pursued in endogenous growth theory. On the one hand, the Romer-like models have generated increasing returns through imperfect competition brought about by the intentional efforts of inventors seeking profits. On the other hand, the AK- and Lucas-like models have generated increasing returns by maintaining perfect competition and accumulating knowledge by some by-product of the economy such as human capital formation. 10.4 Critique of Growth Theories as Applied to E nvironmental and Regional Models The neo-classical growth models presented above have several shortcomings, over and above those already alluded to. These shortcomings may lead to system instabilities, including delayed feedbacks which may result in irreversible change. These shortcomings include: • Cobb-Douglas production functions . Authors such as Benhaim and Schembri ( 1994), Victor ( 1 99 1 ), Ayres (200 1 ), and Ayres and van den Bergh (2005) inter alia have identified several shortcomings with using Cobb-Douglas production functions. Victor ( 1 99 1 , p. 1 96), for example, states that "the potential for additional substitution never diminishes" in a Cobb-Douglas production function. Ayres (2001 ) and Ayres and van den Bergh (2005) argue that many economists have overlooked the critical role p layed by minor factor inputs in the Cobb-Douglas production function - in particular, the commodities (and their physical characteristics i .e. embodied mass and energy) used in intermediate consumption. 324 • Only one homogeneous output. Many neo-classical growth models assume that only a single homogeneous output is produced by the economy. Such models make no allowance for unique characteristics of the different commodities produced within the economy. • Unique role of manufacturing industries. The unique role played by manufacturing industries in an economy, in particular the creation of capital, is largely ignored. • Path dependence, uncertainty, discontinuity and heterogeneity of investment decisions . Weyant and Olavson ( 1 999) note that aggregative models ignore path dependence (e.g. the need to process intermediate goods and delays in capital formation), uncertainty in major innovations, discontinuity in the process of technological change, and the heterogeneity of firm behaviour and investment decisions. • No international or interregional trade. Another major limitation is that the role played by trade is often ignored. Thirwall (2002) suggests that this is also a factor of the very supply (production) orientated nature of the neo-classical models. L ittle credence is given to an economy's balance of payments, in particular the demand for exports e lsewhere. • The process of technological change. Consideration of how the key determinants of technological change are interconnected, respond in relation to one another, respond in relation to influences within the wider economic system, whether multip le trajectories or pathways exist - such queries indicate that theory underpinning the process of technological change is still very much in its infancy (Bretschger, 2005). Quantifying the impacts, for example, of price induced technological change poses a major theoretical problem. • Convergence. A principle focus of growth theory has been on the convergence between rich and poor countries (lones, 1 998, 2 002; Ayres, 200 1 ; Ayres and van den Bergh, 2005). Much of the growth theory l iterature has been developed w ith the intention of investigating solely this issue. Growth models, like other models, are context sensitive - thus, taking a growth model designed to reveal insights on the convergence debate, and in turn, expecting it to explain environmental issues, may be entirely inappropriate. • Nationfocus. The vast majority of growth models have been constructed at the national level. There are however significant differences between national and regional economies e.g. the existence of interregional trade, the greater openness of the economy, and ease of labour force exchange. Many of the working assumptions of neo­ classical growth theory could never be justified in a world which recognises space (Richardson, 1 973) . Space is incompatible with a s ingle industry aggregative economy, perfect competition, complete certainty, marginal adjustments in prices, evenness in the 325 spread of technologies, zero transportation costs and other lesser conditions of neo­ classical economics (Richardson, 1 973). • Biophysical constraints. A s ignificant weakness of neo-classical growth theory is the myopic focus on technological change as the key determinant of long term growth. Growth theory pays little respect to well-known b iophysical constraints. Many growth theorists and economists (e.g. Barnett and Morse ( 1 963), S imon ( 1 9 8 1 ) and Romer ( 1 990» , for example, assume that new ideas or designs will offset any diminishing returns resulting from resource scarcity etc., but to be useful such designs must take on a physical form constrained by, at least, the laws of thermodynamics?48 Although technological change has to date offset diminishing returns from resource scarcity etc., it is highly uncertain whether this trend will continue, particularly in the long term (Faber et al. 1 990; Kaufman, 1 995; Ryan, 1 995; Ayres, 200 1 ; Ayres and van der Bergh, 2005 ; Bretschger, 2005; Vollebergh and Kemfert, 2005) 10.5 Natural Resou rces and Economic Growth In this Section theoretical growth models are developed that incorporate the depletion and degradation of natural resources. The model builds on the neo-classical models described earlier in this Chapter, but incorporates features that address some, but not all of the biophysical constraints identified in the previous Section. At the core of the model is a race between the increasing returns to scale induced through technological change and possibilities of substitution, and diminishing returns to scale resulting from resource scarcity and environmental degradation (e.g. pollution). Samuelson and Nordhaus ( 1 989, p. 859) note that, "In the race between diminishing returns and advancing technology, technology has [to date] won by several lengths".249 Nordhaus ( 1 992, p.2) however warns that, " . . . to dimiss today's ecological concerns out of hand would be reckless. Because boys have mistakenly cried 'wolf in the past does not mean that the woods are safe." Authors like Faber et al. ( 1 990), Kaufman ( 1 995), Ryan ( 1 995), and Ayres ( 1 995, 2002) also support this argument. 248 Authors such as Chapman and Roberts ( 1 983) and Ruth and Cleveland (1 994) argue that the laws of thermodynamics play a critical role in understanding the longer term consequences of technological change. They show that as metals become harder to extract, the amount and cost of energy required increases - thus becoming a key determinant in economic growth (Ryan, 1 995). 249 Jones (2002) has analysed the factor share of energy in the V.S . economy for the period 1 949- 1 999. He concludes that " . . . the energy share shows a general decline, apart from the sharp spike associated with the oil-price shocks of the 1 970s" (Jones, 2002, p . l 84). He attributes this decline to new discoveries of fossil fuels and to new ways of tapping into reserves previously considered inaccessible. Because the energy factor share is declining he argues that the use of a Cobb-Douglas production function to model natural resource depletion may be inappropriate; instead he favours the use of a constant elasticity of substitution (CES) production function. Time constraints imposed on the completion of this thesis prohibited exploration of this alternative. 326 For ease of mathematical presentation the model developed below treats the economy as an aggregated whole i .e. it is assumed that the economy produces only one homogeneous output. A further simplification is that the economy is assumed to be dependent upon only a single resource. 1 0.5. 1 Modelling the I mplications of Land Use Jones (2002) has modified the Solow model with technology to include land. He assumes that only a fixed amount of land, T, is available in each time period; with output generated according to the following production function, ( 1 0. 1 8) This function exhibits diminishing returns to capital and labour i .e . as the economy expands it gets less productive at the margin. This impact may however be offset by increasing returns to all inputs brought about by technological change, A. By reformulating the model in endogenous terms the implications of population growth could be studied more fully i .e. the trade off between having more people to create ideas, and increasingly less space to support each person?50 10.5.2 Modelling of the Depletion of a Non-Renewable Resource Unlike land, which is assumed to be in fixed supply, non-renewable resources such as coal, fossil fuels, minerals and the l ike represent finite resources that may over time be entirely depleted. The depletion of a non-renewable resource may be included in the model as follows, ( 1 0. 1 9) where once again A augments the entire production function, and E represents depletion of the non-renewable resource. The behaviour of the non-renewable resource stock, R, over time may be described by the following differential equation, R = -E . ( l 0.20) 250 S imon ( 1 98 1 ) is a very strong advocate of the 'more people, more ideas' positIOn. Others are skeptical, for example, Dasgupta and Heal ( 1 979, p . 1 94) question this position, " . . . if only for reasons of space". 327 If we assume that a constant fraction of the non-renewable stock is depleted during each period25 1 , i .e . s = E , then the behaviour of the stock over time may be described in further R detail by the following differential equation. R - = -s R ' ( 1 0.2 1 ) Based on Equation 1 0.2 1 it can be mathematically shown that the non-renewable stock, R, exhibits negative exponential growth at the rate, s, over time Under this model an increase in the depletion rate, s, will reduce the long-term growth rate of the economy i .e. the non­ renewable resource is used up faster, leaving less of the resource for production of future output. Or, put alternatively, by reducing the current depletion rate, s, it is therefore possible to increase the long-term growth rate of the economy. Of course, this impact may be offset by the increasing returns that may be generated through technological change or via substitution between factor inputs. The key finding of the preceding two Sections is that the presence of natural resources ( including land) reduces the long run growth rate of the economy. The production functions exhibit diminishing, rather than constant returns to capital and labour. These diminishing returns are the result of: ( l ) population pressure on finite resources reduces growth in proportion to the population growth rate, and (2) the depletion of non-renewable resources slows growth in proportion to the share of these resources in production (Jones, 2002). Nevertheless, these impacts may be offset through technological change or via substitution between factor inputs. However, this offset, as pointed out by Ryan (1 995), is constrained by the laws of thermodynamics. It is therefore impossible to know if technological change will continue in the future to compensate for the diminishing returns brought on from depleting fixed or finite resources. 251 Economists typically use prices to describe the depletion variable, E. A standard result from such analyses is that in the long run, a constant fraction of the remaining stock of resource is depleted in each time period (Jones, 2002). 329 Chapter Eleven Auckland Region Dynamic Ecological-Economic Model In this Chapter a system dynamics model of Auckland Region's environment-economy interactions is developed. The model, known as the Auckland Region Dynamic Environment­ Economy Model (ARDEEM), bui lds on the static monetary and physical flow models developed in Chapters 4, 5 and 6 and the (theoretical) models of economic growth discussed in Chapter 1 0. The model is characterised by positive and negative non-l inear feedbacks between its component modules. The purpose of the model is not to predict Auckland Region's economic future, but instead to highlight possible physical and economic consequences under various scenarios. A key reason for the adoption of a system dynamics modelling framework is that it allows a great deal of flexibility in setting the scenarios that may be investigated. The scenarios themselves are designed to capture not only the 'business as usual' situation, but also the dynamic physical and economic consequences resulting from more extreme change. The ARDEEM system dynamics model is presented in Vensim® DSS format in the Chapter 1 1 directory of the accompanying CD-ROM. The directory also contains a full program l isting for the ARDEEM model - labelled ARDEEM.txt. 1 1 . 1 Structure o f ARDEEM ARDEEM is a novel system dynamics model designed to simulate the combined environmental and economic implications of change in the Auckland Region between 1 998 and 205 l . The focus of ARDEEM is therefore on the medium to long term (i.e. 30 to 70 years) consequences of change in the Auckland Region. ARDEEM cannot therefore be expected to capture short term fluctuations in economic activity such as those arising from cyc l ical commodity price fluctuations.252 The ARDEEM model consists of the following integrated modules: • Population module. This module simulates population growth by age-sex cohort. The population module provides inputs directly for the labour force, economic flow and physical flow modules, and indirectly for the growth module. It is also used in the generation of several key indicators, including resource use per capita, GRP per capita and so on. 252 Other modelling frameworks such as Computable General Equilibrium (CGE), optimisation models (e.g. MARKAL), and some econometric models are better suited for this purpose. 330 • Labour force module. This module takes outputs from the population module by age­ sex cohort and generates estimates of total labour force, employment and unemployment by industry. • Growth module. This module generates estimates of economic output by industry. The cornerstone of the growth module is a production function with constant returns to scale. The production function has the following factor inputs : employment (as generated by the labour force module), commodity imports and use (from the economic flow module), and manufactured capital stocks. The production function is augmented with indices representing technological change and natural capital depletion/degradation. The output estimates generated by this module feed into the economic flow and economic physical flow modules. • Economic flow module. This module describes the financial flow of commodities within the Auckland Region economy. This includes commodity supply, use, imports and exports. The module provides inputs for the growth and economic physical flow modules and generates key economic aggregates including value added (regional GRP), balance of trade, labour productivity, capital productivity and so on. • Economic phys ical flow module. This module describes the Auckland Region economy in physical (mass) flow terms, including commodity supply, use, imports and exports, and is c losely related to the economic flow module. The focus of the module is on the within economy physical flows. Monetary estimates of commodity supply and use from the economic flow module are converted into physical equivalents based on price ($ per tonne) and indices of improvements in eco-efficiency. • Environment-economy physical flow module. This module describes the physical flow of raw materials and residuals associated with economic activity in the Auckland Region. The focal point of this module is the physical flow of ecological commodities not conventionally measured in economic markets. The module draws on the output by industry estimates of the growth module, exogenous estimates of raw material use/residual generation per $ output, and indices of improvements in eco-effic iency to generate its estimates of the physical flow of raw materials and residuals. The l inkages between the various modules are described in F igure 1 1 . 1 . Sections 1 1 .3 to 1 1 . 8 fully describe ARDEEM. Verification and validation of the model is conducted in Section 1 l .9 . In Section 1 1 . 1 0 three scenarios are developed and simulated: ( 1 ) 'business as usual ' , (2) 'cornucopian growth' , and (3) 'prudent pessimism' . The fmal Section of this Chapter outlines the major l imitations of ARDEEM model including the identification of key areas for future development. Population Module Q: � .g Population Commodity Imports, Use Commodities Economic Physical Flow Module Labour Force Module Growth Module Economic Module 33 1 output by ind, final demand Environment - Economy Physical Flow Module Figure 1 1 .1 ARDEEM Module Linkages. Note: The italicised key variables pass information between the modules. 1 1 .2 Brief Description of ARDEEM's Mathematical Nomenclature This Section provides a brief description of ARDEEM's mathematical nomenclature and naming conventions. Specifically, this includes: • Upper case stocks. All stocks begin with a capital letter. • Lower case flows and converters . All flows and converters begin with a lower case letter. • Subscripted arrays. Variables with multiple dimensions are arrayed. A population stock, for example, may have two dimensions, namely: age and sex. In ARDEEM variables with arrayed dimensions are denoted by variable names with suffix subscripts. Vensim's® array functionality substantially reduces ( 1 ) the visual c lutter of influence diagrams, and (2) the time required to program equations. • Full variable names. To aid in the comprehension of Vensim® system dynamics influence diagrams and mathematical formulae, variables names are presented in full . A complete l ist of ARDEEM arrays and their elements is presented below: 332 a = 0 to 4 yrs, 5 to 9 yrs, 1 0 to 1 4 yrs, 1 5 to 1 9 yrs, 20 to 24 yrs, 25 to 29 yrs, 30 to 34 yrs, 35 to 39 yrs, 40 to 44 yrs, 45 to 49 yrs, 50 to 54 yrs, 5 5 to 59 yrs, 60 to 64 yrs, 65 to 69 yrs, 70 to 74 yrs, 75 to 79 yrs, 80 yrs and over c = ComO 1 , Com02, Com03 f = HhldCons, OFD, IntRegExp, IntNatExp = IndO 1 , Ind02, Ind03 imp = Interregional, International p = 0 to 9 yrs, 1 0 to 1 9 yrs, 20 to 29 yrs, 30 to 39 yrs, 40 to 49 yrs, 50 to 59 yrs, 60 to 69 yrs, 70 to 79 yrs, 80+ yrs s = male, female rm = Rm0 1 , Rm02, Rm03, RJn04, Rm05 r = Res0 1 , Res02, Res03, Res04, Res05, Res06 1 1 .3 Population Module In this Section, a population module is developed that disaggregates Auckland Region's population by sex and five year age cohorts (i.e. 0 to 4 years, 5 to 9 years . . . 75 to 79 years, and 80 years and over). Sub-national population projections from Statistics New Zealand (2004) suggest that Auckland Region will grow from a 200 1 base popUlation of 1 ,2 1 6,900 to between 1 ,624,400 (low projection series) and 1 ,926,500 (high proj ection series) by 2026. This represents total popUlation growth of between 33 . 5 percent (low) and 5 8.3 percent (high) over the 25 year period. Over two-thirds of New Zealand's total population growth between 200 1 and 2026 is projected to be in the Auckland Region (Statistics New Zealand, 2004). By 2026 Auckland Region is projected to be home to more than 37 percent of New Zealand's total usually resident population, compared with 3 1 percent as at the 2001 Census. The implications of this growth cannot be understated: • Changes in the types of infrastructure required. Although Aucklanders are relatively young, when compared with other New Zealanders, the average age has been steadily rising (Statistics New Zealand, 1 998). Changes in the age structure of Aucklanders could potentially affect b irth rates, housing requirements, health and education requirements, consumption patterns, and the nature of the labour force. • Pressure on existing infrastructure. Much of Auckland Region's infrastructure is at capacity or the end of its life, or needs to meet higher environmental standards (Auckland Regional Council, 1 999). Of particular concern is the pressure being placed 333 on transportation networks, water supply services, waste water treatment, and energy . . fr 253 generatIOn 10 astructure. • Demand for new irifrastructure. This includes demands for power stations, transportation networks2S\ social and community services (i.e. hospitals, schools, l ibraries, museums, recreational facilities), open space (i.e. neighbourhood reserves, parklands and sports grounds) and additional housing255,256. • Structural mix of the economy. Community, social and personal services play a more s ignificant role in the Auckland Region economy than elsewhere in New Zealand. It can be argued that this role may be exacerbated through growth in population based services such as health and education. Export education, for example, has over recent years become a substantial industry in the Auckland Region economy. The ARDEEM population module is shown as a Vensim® system dynamics influence diagram in Figure 1 1 .2. Note how the age-sex cohort structure of the model is captured using Vensim's@array functionality, rather than by building multiple population stocks with inflows and outflows for each age-sex cohort. 253 The pressure of population growth on Auckland Region's infrastructure may arguably be seen through a number of local crises and associated policy responses including: the 1994 energy b lackouts (resulting from a poorly maintained and ageing energy supply network), 1 998 water shortage (resulting in the construction of the so-called 'Waikato pipeline'), and ongoing traffic congestion ( leading to substantial local and central government expenditure on roading projects). 254 Household trends in car ownership and energy consumption during the 1 990s have exacerbated these demands by growing at rates substantially higher than the population growth rate (Auckland Regional Council, 1 999). 255 The average home occupancy rate in Auckland Region has been steadily rising (Statistics New Zealand, 1 998; Auckland Region Council, 1 999) Although this trend may to some extent dampen the demand for additional housing, it is insufficient to offset the likelihood of substantial future housing requirements, By contrast, the New Zealand home occupancy rate has been steadily declining. 256 Over the last two decades Auckland Region territorial local authorities, supported by the Auckland Regional Council, have through initiatives such as the Auckland Regional Growth Forum advocated a more compact urban form, resulting in greater numbers of apartments, terraced housing and infil l housing, Although trends for traditional housing have persisted, there has been a significant increase in h igher density living. 334 o sex at b i rth tot b i rths o b irths o i n it ia l p op -- _ f\ deaths ----�---'� o� w W======5�===l1ll-i Popu l ation tot d eaths into cohort �1I====,,*===�:::,:j;�-----+--'-- out of coho rt O�netmig rati on �O tot net m ig ration ------iIo.. _ ) / � share exiti ng cohort <.T irn�� net m ig rati on flux < 40 yrs tot pop tot pop Figure 1 1 .2 Population Module Influence Diagram The population module may be described using the following mathematical equations257: Stocks Populationa,!,J + dt) = Populationa,s(t) + (birthsa,s + net migrationa,s + into cohorta,s - deaths a,s - out of cohorta,s) x dt. As measured in number of people. where: Initial Populationa,s Inflows birthsa,s = initial pOPa,s for the 1 998 base year (no. people). = I a,s (( Populationait) x fertility ratea.s / 1 000 ) x sex at births ) 258 As measured in number of people. 257 It is important to note that variables are defined only once, at first use, to avoid unnecessary duplication. 258 Double summations, I a,s ( Popuiationa,s(t». such as n I I ( Popuiationa,s(t» , are summarised here as a:O s:j,m where: fertility ratea,s fr consta,s fr xcoeila,s sex at births net migrationa,s where: net migrationjlux <45 yrs tot pop into cohorta,s 335 == (fr xcoel!a,s x LN((t) - 197 1 )) + fr consta,s == the constant of a logarithmic time series regression equation describing the Auckland Region fertility rate of a particular age-sex cohort between 1 97 1 and 2000.259 If s = male thenfr xcoef!a,s is set to zero. S imilarly, fr xcoeila,s for females aged under 1 0, and over 50, years of age is set to zero. == the 'x ' coefficient of a logarithmic time series regression equation describing the Auckland Region ferti lity rate of a particular age-sex cohort between 1 97 1 and 2000. Once again, if s == male, or s == female and a < 1 0 or a >= 50 years, thenfr consta,s is set to zero. == shares of sex at birth. It is assumed that likelihood of a male or female being born is the same. = net migration flux x (Popuiationa,s(t) / < 45 yrs tot pop) == a graph showing annual net migration into/from Auckland Region. These estimates are taken directly from Statistics New Zealand's (2004) sub-national population projections (medium series ). 45 = L I ( Populationa,s (t) . Total population under 45 years a�O s�f,m of age. = out of cohorta,s' If a represents the 5 to 9 age cohort then it is assumed that one fifth of the 0 to 4 age cohort moves into the 5 to 9 age cohort each year. A s imilar pattern applies to other age cohorts. As measured in number of people. 259 Linear and logarithmic time series regressions are utilised throughout this Chapter to account for the changing nature of exogenous variables. The pros and cons of using time series regression in this way is given at the end of Section 1 1 .3 . The use of regression equations is easily identified according to the 'xcoeff' and ' const' endings of variable names. Appendix M provides full details of all variables determined using time series regression. For each time series regression this includes: the type of regression (i.e. linear or logarithmic), time series period (e.g. from 1971 to 2000), the regression equation, R-squared value, and data source. In most cases the p-values, F-ratios and Durban-Watson statistics of the regression equations were found to be acceptable. All regression equations were 'sense' checked by graph equations against observed data; with slope, direction and residual outliers being investigated. 3 3 6 Outflows deathsa,s where: mortality ratea,s mr consta.s mr xcoefla,s out of cohorta,s where: share exiting cohort Reporting variables pop by sexs tot births tot deaths tot net migration tot pop pop pyramido 10 9,s = Populationa.sCt) x (mortality ratea,s / 1 000). As measured in number of people. = (mr xcoefla.s x LN« t) - 1 97 1 )) + mr consta.s = the constant of a linear or logarithmic time series regression equation describing the mortality rate of a particular age-sex cohort between 1 97 1 and 1 99 5 . = the 'x' coeffic ient o f a linear o r logarithmic time series regression equation describing the mortal ity rate of a particular age-sex cohort between 1 97 1 and 1 995. = Populationa,s(t) x share exiting cohort. As measured In number of people. = the share of population in each age-sex cohort exiting the cohort in each fu ll time step. It is assumed that the number of people in each year of age in a cohort is the same i.e. one fifth of the age cohort moves into the next cohort each year. n I ( Populationa,s{t)). Total population by sex (no . of a=O people). = I a,s ( birthsa,s). Total births (no. of peop le). = I a,s ( deathsa,s). Total deaths (no. of people). = I a / net migrationa,s))' Total net migration (no . of people). = I a,s ( Populationa.s 1 5 years of age), adjust these estimates for unemployment to derive FTE employment and, in turn, distribute this employment to economic industries. The employment by industry estimates are a critical factor input into the economic growth module of Section 1 1 .5 . The Vensim® system dynamics influence diagram for the labour force module is depicted in Figure 1 1 .3 . The mathematics of the module is given below: 260 'Curve fitting' approaches have also been extensively used by the Resource Futures Group at the CSIRO in Canberra. This group, led by Dr Barney Foran, has developed the Australian Stocks and F lows Model (AS FM) to simulate the resource requirements necessary to sustain the Australian economy to 2 1 00 under particular policy driven scenarios. 338 Figure 1 1 .3 Converters o eb i d xcoeff emp by i nd d istrib tot unemp labour force by sex Ifpr xcoeff Labour Force Module Influence Diagram lab force part ratea,s = (lfpr xcoef!a,s x LN((t) - lfpr base yeara,s» + lfpr consta,s' Labour force participation rates for those under 1 5 years of age are set to zero. where: lfpr base year a,s lfpr cons ta,s = the base year of a logarithmic time series regression equation describing labour force participation of a particular age-sex cohort. A 1 986 base year was used for a < 60, and a 1 993 base year for a >= 60. = the constant of a logarithmic time series regression equation describing labour force participation of a particular age-sex cohort from the base year. /fpr xcoeila,s labour jorcea,s where: unemp ratea,s unempa,s emp by ind, where: emp by ind distrib, ebid const, ebid xcoeff, Reporting variables labour jorce by sexs tot unemp 339 == the 'x' coefficient of a logarithm ic time series regression equation describing labour force participation of a particular age-sex cohort from the base year. == Populationa,s(t) x lab force part ratea,s' As measured in full­ time equivalents (FTEs). == labour jorcea,s x Cl - unemp rate a,s) ' As measured in FTEs . == a graph depicting annual unemployment rates for New Zealand as taken from Statistics New Zealand. It is assumed that Auckland Region unemployment rates in each age-sex cohort are similar to those of the nation. Post-200S unemployment rates for each age-sex cohort were derived using a moving average of the preceding 6 years. == labour jorcea,5 x unemp ratea.s. As measured in FTEs. n == I I ( empa.s ) x emp by ind distrib,. As measured III a=1 5 s=J,m FTEs. = (ebid xcoefl, x LN« t) - 1 987)) + ebid const, = the constant of a logarithmic time series regression equation describing the distribution of employment (FTEs) across economic industries between 1 987 and 2003 . = the 'x' coefficient of a logarithmic time series regresSIOn equation describing the distribution of employment (FTEs) across economic industries between 1 987 and 2003 , n L ( labour jorcea,s)' Total labour force by sex (FTEs). a=1 5 n = I I ( unempa,s)' Total unemployment (FTEs). a=1 5 s=J,m 340 n unemp by sexs I ( unempa.s). Total unemployment by sex (FTEs). a=15 tot labour force emp by sexs tot emp 1 1 .5 Growth Module n = I I ( tot labour forcea,s)' Total labour force (FTEs). a=15 s=/,m n = I ( empa,s)' Total employment by sex (FTEs). a=1 5 n = I I ( empa.s). Total employment (FTEs). a=15 s=/,m In this Section a growth model for ARDEEM is developed. The model builds on the economic growth theories critiqued in Chapter 1 0. Although several alternative growth models were operationalised and tested using hypothetical data, a severe paucity of actual data261 , along with time constraints, prohibited fuller implementations. One or two of these alternatives could arguably be considered to be more conceptually appealing than the actual model implemented below. One such alternative, an endogenous growth model, is depicted in Appendix L using a Vensim® system dynamics influence diagram. At the core of the growth model is a production function control ling the estimation of future output by industry within the Auckland Region economy (Figure 1 1 .4). The production function is comprised of factor inputs (manufactured capital, natural capital, labour, domestic commodity use, commodity imports, and technological change) which are determined through a number of dynamic feedback loops. The factor inputs and their loops are considered further below: • Capital. This represents the stock of manufactured capital (covering intangible assets, plant and machinery, transport equipment, other construction, non-residential buildings, and residential buildings) utilised in producing economic output in the economy. Capital stock estimates for the base year were derived by scaling down national estimates to the Auckland Region based on FTE employment.262 The national estimates were obtained from Statistics New Zealand (2000). Capital formation depends on 261 An alternative engine based on endogenous growth theory, for example, required estimates of knowledge stocks, knowledge creation/duplication rates, and so on for the modelling of the ' stepping on toes' and 'standing on the shoulders of giants' technological spillover effects described in Chapter 1 0 . A further complication, relevant to this example, was the necessity to build not only dynamics for knowledge creation occurring within the Auckland Region, but also for the rest of the world. 262 It is assumed that the mix of capital used by each worker is spatially invariant across New Zealand. 341 economic output and an exogenously set investment rate, while capital depreciation depends on the size of the capital stock and an exogenously set depreciation rate. Capital investment and depreciation rates were developed by applying regression analysis to national time series obtained from Statistics New Zealand's INFOS database. Again, there is no reason why future patterns of investment and depreciation should reflect past trends. Furthermore, capital investment and the production of economic output are interdependent activities. The economic output of an industry includes wage, salary and dividend payments made to employees, which in turn, provides the fuel for further investment. Data constraints prohibited the explicit modelling of this feedback. 263 • Labour. Labour inputs are included in ARDEEM through the estimation of the number of human hours worked annually in each industry. These estimates were generated by multip lying for each industry employment estimates by occupation (FTEs), by the number of hours typically worked in each week within each occupation (hours), and then scaling these to produce annual estimates. Measurement in human-hours accounts for productivity changes brought about by working more hours per day. Labour factor payments (i.e. wages and salaries) also play a critical role in ARDEEM, through namely: ( 1 ) investment in the formation of capital - as discussed above, and (2) commodity consumption - as captured in the positive feedback between the Economic and Growth modules involving the use, variable. • Commodity use. The criticality of minor factor inputs in generating an industry's output along with path dependencies are captured in the model by consideration of commodities used in intermediate consumption. Currently commodity inputs ill ARDEEM are only considered in aggregate; it is envisaged that future versions of the model will consider more carefully the role p layed at a detailed commodity level. • Commodity imports. Commodity imports are essential to the Auckland Region economy (refer to Chapters 5, 6 and 8 for further details).264 Auckland Region' s traditional role in import substitution was identified in Chapter 5 , as was the increasing trade openness of the economy; particularly for l ight manufacturing industries. If local supply is unable to satisfy local demand for a particular commodity it is l ikely that the market response wil l be to import this commodity. Furthermore, if a locally provided 263 Separation of domestic and foreign capital investment at a disaggregated sectoral level was the main constraint. 264 This critical dependence has been further investigated by the author and Professor Le Heron of the School of Geography and Environmental Science at University of Auckland. Based on an analysis of changes in Auckland Region value added and employment multipliers between 1 987 and 2003 it was found that economic interdependencies between industries had substantially weakened, while a compensatory growth in trade, particularly with neighbouring regions and Australia, had eventuated. Given the globalisation of international markets this is perhaps not surprising. 342 non-renewable resource becomes scarce, and cannot easily be substituted for, then importation of the resource wi l l be critical for continued economic activity. Allowing for the possible simulation of substitution of domestic commodities for imported equivalents is therefore considered paramount. It is envisaged that in future versions of ARDEEM consideration wi l l also be given to the demand for exports occurring elsewhere (refer to Appendix C for a detailed analysis of interregional trade flows). • Technology index. This stock represents technological change over time via the positive feedback loop between the Technology Index, stock and the technology formation; flow. The formation rate is controlled by the exogenously determined technology rate;. The technology rate for each industry was set equal to the 1 998 to 2002 geometric annual average total factor productivity (TFP) rate as obtained from Black et al. (2003). Since the TFP covers all factor inputs the technology index must augment the entire production function. It should be noted that if each industry's TFP is set to zero then the reporting variables output per worker; and capital per worker; will tend toward a steady-state over the long term i .e . there will be no productivity growth and the growth rate of the Auckland Region economy wil l simply mirror the population growth rate. Again, it is important to note that future trends in TFP may not reflect historical trends. • Elasticities of output with respect to factor inputs (a" b;, g;, and d;) . These elasticities were estimated by taking a 1 987 to 2003 time series of the logs of the factor inputs ( i.e. Capital" emp by ind" use" and imports,) and performing a constrained regression such that the coefficients of the dependent variables of the regression equation (i .e. a;, b" g" and d;) summed to 1 (i .e. exhibited constant returns to scale). This approach is commonly used by economists to derive the elasticities of factor inputs with respect to output. It is important to note that the regression analysis is used only to establ ish the initial values of ai, bi, gi, and di, i .e . it does not in any way mean that these elasticities wil l remain the same over the next 30 to 70 years. Furthermore, no assumptions have been made as to how one factor input may substitute for another; instead these may be tested explicitly under various simulations. o tech n o l o gy rate ·!,tut ernp> tot capital per wolker o < i m ports> 343 d r const o utput p e r wolker Q=======9�====� tech nology formation Figure 1 1 .4 Growth Modu le Influence Diagram Capital Stock Capital; (t + dt) where: Initial Capital; Inflows capital formation; where: investment rate, where: ir xocefJ; ir const; = Capital,{t) + (capital formation, - capital depreciation;) x dt. The total available manufactured capital stock ($ mili65 utilised by industry i. = initial capital; ($ mil) for the 1 998 base year. = output; x investment rate,. As measured in $ mil. = (ir xcoeff; x LN((t) - 1 987)) + ir const; = the 'x' coefficient of a linear or logarithmic time senes regression equation describing the rate of capital investment by industry i between 1 987 and 2003. = the constant of a linear or logarithmic time series regression equation describing the rate of capital investment by industry i between 1 987 and 2003 . 265 All financial values are in $ 1995 unless stated otherwise. 344 output; where: a; g; imports;mp,; Outflows capital depreciation; where: depreciation rate; where: dr xcoefJi dr const; Technology Index Stock Technology Index; (t + dt) = ((Technology Index;(t) x Capital;(t» Q; ) x ((Technology n Index;(t) x emp by ind;) b; ) X (( I ( imports;mp, i) x ;mp=l n Technology Index; (t» g; ) X (( I ( usec,;) x Technology c=l Index;(t» d; ) . A production function estimating total output ($ mil) in each industry i. The production function assumes constant returns to scale (i.e. a; + b; + gi + d; = 1 ) . = the elasticity of output with respect t o capital utilised by industry i. = the elasticity of output with respect to employment utilised by industry i. = the elasticity of output with respect to total imports utilised by industry i. = the elasticity of output with respect to total intermediate commodity use by industry i. = total imports ($ mil) used by industry i. = total commodities ($ mil) used by industry i. = Capital,{t) x depreciation rate; As measured in $ mil. = dr xcoefJi x LN((t) - 1 972) + dr const; = the 'x' coefficient of a linear or logarithmic time senes . regression equation describing the rate of capital depreciation by industry i between 1 972 and 2003 . = the constant of a linear or logarithmic time series regression equation describing the rate of capital depreciation by industry i between 1 972 and 2003 . = Technology Index;(t) + (technology formation;) x dt where: Initial Technology Index Inflows technology formationi where: technology rate, Reporting variables tot cap form tot cap dep tot capital tot capital per worker tot output per worker tot output capital per workeri output per worker, 345 = 1 for the 1 998 base year. = Technology Index,(t) x technology rate, = the geometric rate of annual technological change for industry i. Black et al. (2003) have estimated total factor productivity by industry in the New Zealand economy over the period 1 988 to 2002. These estimates are used here as a proxy for the rate of technological change in the Auckland Region economy. n = L ( capital formation, ) . Total capital formation ($ mil). ,=\ n L ( capital depreciation, ) . Total capital depreciation ($ , =\ mil). n = L ( Capitallt» . Total capital ($ mil). ,=\ = IF THEN ELSE(tot emp = 0, 0, tot capital / tot emp). As measured in $ mil. = IF THEN ELSE(tot emp = 0, 0, tot output / tot emp). As measured in $ mil. n = L ( output,). Total output ($ mil). ,=1 = IF THEN ELSE(emp by ind, = 0, 0, Capitallt) / emp by ind,). Total capital ($ mil) by industry i. = IF THEN ELSE( emp by ind, = 0, 0, output, / emp by indi). Total output ($ mil) by industry i. 346 1 1 .6 Economic Module The economic module consists of a commodity by industry input-output economIC system (compatible with Table 4. l ). This module describes the circular flow of commodities supplied both domestically and internationally, and their corresponding use and final consumption (Figure 1 1 .5) . The module is linked with the growth module through a number of positive (reinforcing) feedbacks. On the one hand, it provides key inputs into the growth module by generating estimates of ( 1 ) commodity imports required to satisfy both intermediate and final demand, and (2) intermediate demand commodity use. On the other hand, it uti lises estimates of output and capital formation in the calculation of the interregional exports, international exports and other final demands (capital formation). There are several key features of the economic module. Firstly, utilising the input-output model conceptualised in Chapter 4, and implemented in Chapter 5, allows the interrelationships between economic industries to be simulated over time. If, for example, households consume more dairy products, then the model would simulate not only a resultant increase in dairy product manufacture, but also an increase in dairy cattle farming.266 Secondly, the input-output model is created in a commodity-by-industry format which records joint production. Although data constraints will typically restrict the s imulation to less than 5 0 industries, the number of commodities wil l be far less restricted; the supply and use of hundreds of commodities could be s imulated without difficulty. Thirdly, this detailed consideration of industries and their commodities potentially enables the unique role played by manufacturing in capital formation to be directly incorporated in the growth module production function. It also permits consideration of minor, but limiting or critical commodity factor inputs, to be incorporated in the production function. Fourthly, the adoption of a financial commodity-by-industry framework ensures comparabi l ity and the straightforward translation into physical equivalents (see Section 1 1 .7 below). F inally, the commodity-by-industry format permits the computation of economic and ecological multipliers (and by corollary ecological footprints) at each time step. Overall, the economic module combines the detailed commodity-by-industry input-output data with the flexibil ity of dynamic simulation. 266 These relationships are evaluated at each time step within the model. It should be noted however that the input mix of commodities (i.e. purchase pattern) utilised by each industry is assumed to remain constant over time. A more complete implementation of the model would allow this mix to change over time. Duchin and Szyld ( 1 985) and Leontief and Duchin ( 1 986) have, for example, performed time series regression on input-output technical coefficients to assess the future impact of automation on workers. This approach, while beyond the scope of this thesis, provides a possible pathway for the future development of the economic module. Figure 1 1 .5 Economic Module Influence Diagram Use Commodities Stock 347 Use Commoditiesc,i (t + dt) = Use Commoditiesc, ;(t) + (form oJ corn Jor usec,i - usec,;) x dt. As measured in $ mil. where: Initial Use Commoditiesc,; Jorm oJ corn Jor usec,i where: = init usec,i ($ mil) for the 1 998 base year. = usec,; x use growth scalars; As measured in $ mil. 348 use growth scalarslndOI use growth scalarslnd02 use growth scalarslnd03 diag B less use invc, i diag B less usec,i diag Bc, i gross com inputs Bc tot final demandc final demandc,HhldCons final demandc,OFD final demandc,intRegExp final demandC,lntNatExp intnat exp to go etor const etor xcoeff = « diag B less use invComOl,lndOJ x tot final demandComOJ) + (diag B less use invComOJ,Ind02 x tot final demandcom02) + (diag B less use invComOJ,lnd03 x tot final demandCom03))267 = « diag B less use invCom02,lndOJ x tot final demandComOJ) + (diag B less use invCom02,lnd02 x tot final demandcom02) + (diag B less use inVCom02,lnd03 x tot final demandcom03)) = « diag B less use invCom03,lndOJ x tot final demandComOJ) + (diag B less use invCom03,lnd02 x tot final demandcom02) + (diag B less use inVCom03,lnd03 x tot final demandcom03)) = INVERT MATRIX(diag B less useC, i, 3) = diag Bc, i- usec, i = gross com inputsc. As measured in $ mil. n = I ( suppIYi,c) + i;\ n I ( imp of comimp,c), imp;\ commodity inputs ($ mil) , n Total gross = I (final demandcj), Total fmal demand ($ mil) by /;\ commodity c. = hhld cons per capitac x tot pop, As measured in $ mil . n = init fd coejJsc,OFD x I ( capital formationi), As measured in i;\ $ mil . = init fd coejJsC,lntRegExp x tot output x intreg exp to go. As measured in $ mil. = init fd coejJsC,lntNatExp x tot output x intnat exp to go, As measured in $ mil . = (etor xcoeffx LN« t) - 1 987)) + etor const = the constant of a logarithmic time series regression equation describing the ratio of international exports to gross output between 1 987 and 2003 , = the 'x' coefficient of a logarithmic time series regression equation describing the ratio of international exports to gross output between 1 987 and 2003 , 267 This equation along with the two that follow represents a matrix multiplication of diag B less use inv by tot final demand, intreg exp to go in it fd coe./fic! init final demandc! hhlds cons per capitac Outflow usec, / imports /mps, / cnvsrn to ind sp/mps,c Supply Commodities Stock 349 == the ratio of interregional exports to gross output for the 1 998 year. n == init final demandc! I L ( in it final demandc!) c=1 == final demand consumption by commodity c across final demand f for the 1 998 base year ($ mil). == init final demandc,HhldCons I La,s ( initial papa,s). As measured in $ mil. == Use Commoditiesc, ,(t). As measured in $ mil. == (cnvrsn to ind sp/mp,ComOl X uSeComOl,/) + (cnvrsn to ind sP/mp,Com02 X uSeCOm02, 1) + (cnvrsn to ind sP/mp,Com03 X uSeCom03, /) ' Repeat for c == 1 to n. As measured in $ mil. == a matrix for converting imports from commodity to industry space for the 1 998 base year. This matrix was derived from the Auckland Region input-output model developed in Chapter 5 by applying the procedure outlined in Section 4.4.3 .2 . Supply Commodities;,c ( t + dt) == Supply Commoditiesjt) + (form of cam for supply/, c - supply/,c) x dt. As measured in $ mil. where: Initial Supply Commodities/,c form of cam for supply/,c where: cam growth scalarsc Outflow supply/,c = init supply;,c for the 1 998 base year ($ mil). == supply/,c x cam growth scalarsc. As measured in $ mil. = use growth scalars/ == Supply Commodities;,c(t). As measured in $ mil. 3 50 Com modity I mports Stock Commodity Imports;mp,c U + dt) == Commodity Imports,mp,cU) + (form of corn for imp;mp,c - imp of com;mp,c) x dt. As measured in $ mil. where: Initial Commodity Imports,mp,c == init imports/mp,c for the 1 998 base year ($ mil). form of corn for imp,mp,c Outflow imp of com'mp.c Reporting variables output check output by ind, value added, tot value added labour productivity; capital productivity; == imp of com,mp,c x corn growth scalarsc. As measured in $ mil. == Commodity Imports,mp,cU), As measured in $ mil. == L ; ,c (form of corn for supply" c). Total output check ($ mil). n == L (form of corn for supply;,c). Total output by industry i ($ c=l mil). n == output by ind; - L ( usec, ;). Total value added by industry i c=l ($ mil). n == L ( value added;) . Total value added ($ mil) . ;=1 == IF THEN ELSE( emp by ind; == 0, 0, value added, / emp by ind;) . Labour productivity as measured in $ mil. == IF THEN ELSE(Capital,U) == 0, 0, value added; / Capital;{t)). Capital productivity as measured in $ mil. 1 1 .7 Economic Physical Flow Module The economic physical flow module is the physical equivalent of the economic flow module. It describes the Auckland Region economy in physical (mass) flow terms, including commodity 3 5 1 supply, use, imports and exports (Figure 1 1 .6). The module focuses purely on the within economy physical flows. F inancial estimates of commodity supply, use, imports and exports are converted to physical equivalents based on price ($ per tonne) and eco-efficiency indices which allow for technological improvements.268 The module utilises within economy data from the financial input-output model conceptualised in Table 4. 1 and implemented in Chapter 5, and the physical input-output model conceptualised in Table 4.2 and implemented in Chapter 6. 268 It is assumed that these technological improvements occur at a constant compounding rate. This simplifYing assumption has been adopted to demonstrate how technological change might be incorporated within ARDEEM, but is considered questionable given long-run thermodynamic constraints. Figure 1 1 .6 � tot P hys use by co m o tot p hys use ()======��====�� o fo rm of e co- effi ci e nt tech fo r dom use eco-eff imp imprv rate ()======��====�� form of eco-efficie nt tech for imports tot phys imp by com Economic Physical Flow Influence Diagram Import E co-effi ci e n cy I ndex o tot phys supp ly by com EXllort & F ina l Demand E co- effi ci e n cy I n dex tot phys supp ly eco- eff dom supp ly imprv rate �======�======�() fo rm of e co- effi ci e nt tech for d om supp ly eco-eff exp & fd imprv rate form of eco-effici ent tech for exports & fi na l demand tot p hys exp & fd by com " f" � tot ro by r tot ro for fd tot ro for i nd 358 Raw Material Inputs Eco-Efficiency Index Stock Raw Material Inputs Eco-Efficiency Indexrm (t + dt) = Raw Material Inputs Eco-Efficiency Indexrm(t) + (form of eco-elf tech for rmirm) x dt where: Initial Raw Material Inputs Eco-Efficiency Indexrm = I for the 1 998 base year. form of eco-elf tech for rmirm eco-elf rmi imprv raterm rmi for indrm,i init rmi coe./fsrm,i rmi for fdrmJ init rmifor fd coe./fsrmJ fin dem by catf = Raw Material Inputs Eco-Efficiency Indexrm(t) x eco-elf rmi imprv raterm = the rate of eco-efficiency improvements in the use of raw material input rm. This rate is assumed to compound over time through technological change.27 1 = init rmi coe./fsrm,i x output by indi x Raw Material Inputs Eco-Efficiency Indexrm(t). As measured in tonnes. = the 1 998 physical input of raw material rm (tonnes) required to produce $ of output in industry i. = init rmi for fd coe./fsrmJ x fin dem by catf x Raw Material Inputs Eco-Efficiency lndexrm(t). As measured in tonnes. = the 1 998 physical input of raw material rm (tonnes) required for consumption of $ of output in fmal demand category f n I (final demandcJ) . Total flnal demand by c=l category f ($ mi l). 271 This simplifying assumption has been adopted to demonstrate how eco-efficiency improvements might be included within ARDEEM, but is considered questionable given long-run thermodynamic limits to technological change. This assumption also applies to the following variables within this module: eco-eff rmo imprv raterm, form of eco-eff tech for rir and eco-eff ro imprv rater. 359 Raw Material O utputs Eco-Efficiency Index Stock Raw Material Outputs Eco-Efficiency Indexrm (t + dt) == Raw Material Outputs Eco-Ef iciency Indexrm(t) + (form of eco-eff tech for rmorm) x dt where: Initial Material Outputs Eco-Efficiency Indexrm = 1 for the 1 998 base year. form of eco-efftechJor rmOrm eco-eff rmo imprv raterm rmo for indvm inif rmo coejJs{,rm rmo for fdjrm init rmo for fd coejJsjrm = Raw Material Outputs Eco-Ef iciency Indexrm x eco-eff rmo imprv rate rm(t) = the rate of eco-efficiency improvements in the supply of raw material output rm. = init rmo coejJsl,rm x output by ind{ x Raw Material Outputs Eco-Ef iciency Indexrm(t). As measured in tonnes. = the 1 998 physical output of raw material rm (tonnes) generated in producing $ of output in industry i. = init rmo for fd coejJsjrm x fin dem by catf x Raw Material Outputs Eco-Efficiency Indexrm(t). As measured in tonnes. = the 1 998 physical output of raw material rm (tonnes) generated in consuming $ of output in final demand category f Residual Inputs Eco-Efficiency Index Stock Residual Inputs Eco-Efficiency Indexr (t + dt) = Residual Inputs Eco-Ef iciency Indexr(t)+ (form of eco-efftechfor rir) x dt where: Initial Residual Inputs Eco-Efficiency Indexr == 1 for the 1 998 base year. 360 Inflow form of eco-eff tech for ri, eco-eff ri imprv rate, ri for ind'. i init ri coef s r.i rifor jd,j init rifor fd coeffs,j = Residual Inputs Eco-Efjiciency Index,(t) x eco­ effri imprv rate, = the rate of eco-efficiency improvements in the use of residual input r. = init ri coef s,, i x output by ind, x Residual Inputs Eco-Efjiciency Index,(t) . As measured in $ mil . = the 1 998 physical input of residual r (tonnes) required to produce $ of output in industry i. = init rifor fd coef s,j x fin dem by catl x Residual Inputs Eco-Efjiciency Index,(t). As measured in tonnes. = the 1 998 physical input of residual r (tonnes) required for consumption of $ of output in final demand category f Residual Outputs Eco-Efficiency I ndex Stock Residual Outputs Eco-Efjiciency Index, (t + dt) = Residual Outputs Eco-Efjiciency Indexr(t) + (form of eco-eff tech for ro,) x dt where: Initial Residual Outputs Eco-Efjiciency Index, = 1 for the 1 998 base year. form of eco-eff tech for ro, eco-eff ro imprv rate, ro for indi., init ro coeffsi.r = Residual Outputs Eco-Efjiciency Index,(t) x eco­ effro imprv rate, = the rate of eco-efficiency improvements in the supply of residual output r. = init ro coeffsi., x output by indi x Residual Outputs Eco-Efjiciency Index,(t). As measured in tonnes. = the 1 998 physical output of residual r (tonnes) generated in producing $ of output in industry i. rofor fdf,r init ro for fd coeffsj,r Reporting variables tot rmi by rmrm tot rmo by rmrm tot ri by rr tot ro by rr tot rmi by ind, tot rmi by fdl tot rmo for fdl 36 1 = init ro for fd coeffsj,r x fin dem by catf x Residual Outputs Eco-Efficiency lndex,(t). As measured in tonnes. = the 1 998 physical output of residual r (tonnes) generated in consuming $ of output in final demand category f n n I ( rmi for fdrmj) + I ( rmi for indrmJ. Total /=1 <=1 physical input of raw material rm (tonnes) into the economy. n n = I ( rmo for fdfrm) + L ( rmo for indt,rm). Total 1=1 1= 1 physical output of raw material rm (tonnes) from the economy. n n = I ( ri for Idrj) + L ( ri for indr, ,). Total 1=1 1=1 physical input of residual r (tonnes) into the economy. n n = I ( ro for Idfr) + I ( ro for ind" r). Total 1=1 1=1 physical output of residual r (tonnes) from the economy. n = L ( rmi for indrm,,). Total physical input of raw rm=1 materials into industry i (tonnes). n = L ( rmi for fdf,rm). Total physical input of raw rm=1 materials into final demand category f(tonnes). n = L ( rmo for fdj,rm). Total physical output of raw rm=1 materials from final demand category f(tonnes). 362 n tot rmo by ind; I ( rmo for indi,rm). Total physical output of tot rifor ind; tot rifor fd! tot ro for fd! tot ro for ind; rm=1 raw materials from industry i (tonnes). n = I ( ri for indrJ). Total physical input of r=1 residuals into industry i (tonnes). n = I ( ri for fdrj). Total physical input of residuals r=1 into fmal demand category f(tonnes). n = I ( ro for fd!.r). Total physical output of r= 1 residuals from final demand category f(tonnes). n = I ( ro for ind,.r). Total physical output of r=1 residuals from industry i (tonnes). 1 1 .9 Validation and Verification of ARDEEM Several steps were undertaken during the model ling process to ensure that the results generated by ARDEEM were as valid as possible. These are considered below in terms of structural and predictive validity. 1 1 .9.1 Structural Validity of ARDEEM Structural validity refers to the logic, consistency and accuracy of the model 's internal structure i .e . its equations, interrelationships, and units of measurement. The structural val idity of ARDEEM was evaluated by: • Creation of 1998 reference mode. Simulation results generated for the 1 998 base year were compared with actuals or estimates generated independently in Microsoft Excel®; particular emphasis was placed on the validity of endogenous variables. • Independent peer review. The relationships within the model were independently peer reviewed by Professor Murray Patterson (School of People, Environment and Planning, Massey University), Professor Richard Le Heron (School of Geography and Environmental Science, University of Auckland), Dr Doug Fairgray (Economist, Market Economics Ltd) and Mr Geoff Butcher (Economist, Butcher Partners Ltd). In 363 l ight of these peer reviews several changes were made to the conceptualisation of ARDEEM. 1 1 .9.2 Predictive Validity of ARDEEM Predictivity validity refers to the model 's ability to adequately imitate the behaviour of the real system. Predictive validity is however of only limited usefulness as a model may produce results which provide an extremely good historical data fit, but may in no way reflect future outcomes. The predictive ability of ARDEEM was evaluated by : • Backcasting. The model was backcasr72 so as to produce results for the period 1 980 to 1 998. Graphs of key variables (Population, Capital, Commodity Use, Commodity Supply, capital investment, labour force participation, employment and so on) were plotted against actuals. Given the use of time series regression to 'curve fit' historical trends, it is perhaps not surprising that the results generated reflected actuals. • Comparison with Statistics New Zealand projections . In the case of the Population, births, deaths, net migration and labour force variables it was possible to compare ARDEEM simulation results, under a Business as Usual Scenario, with SNZ projections. Overall, it is important to remember that complete validation of a model by comparison with the real world is not possible, as ARDEEM only captures a selected number of components and behaves purely in response to its internal relationships. 1 1 . 10 Scenario Analysis There are several reasons why pol icy and decision makers need to look into the future. This includes p lanning for possible futures, deciding between competing alternatives, making provisions for new infrastructure, and so on. Underpinning all of these reasons is arguably a desire to manage complexity and m in imise risk (Shearer, 1 994). While it is impossible for us to predict the future, it is however useful for us to understand what 'might' happen in the future. This forces us to consider the implications of our proposed trajectories; reducing uncertainty 272 Several simulations were required for this purpose; with appropriate corrections to the conceptualisation of ARDEEM being made following each simulation. 364 and avoiding possible pitfalls. Scenario modelling is one approach that may be used to help us simulate possible futures and their impl ications (Wilson, 1 978; Schwarz, 1 987).273 Scenarios have been defined by Kahn and Wiener ( 1 967) as "a hypothetical sequence of events constructed for the purpose of focusing attention on causal processes and decision points" (Wilson, 1 978, p. 225). SRl/CSSP ( 1 975), Boshier et al. ( 1 986) and Schaar ( 1 987) have identified several key advantages of the scenario approach, including: ( 1 ) suitable for long-run projections where uncertainty is h igh and historical relationships have been characterised by dynamic feedbacks, non-linearities, time lags and the like; (2) help us to see the future in totality, rather than piecemeal; (3) allows us to trace people ' s behaviour in the face of perturbation; and (4) may provide common ground for communication between diverse interest groups or backgrounds. Scenario development has several important methodological considerations, including: ( 1 ) how many scenarios? Despite the lack of agreement within the l iterature, there seems to be a consensus for three scenarios (Linneman and Klein, 1 979). Two scenarios are likely to be categorised as 'good and bad' , while the simulation of more than three scenarios often becomes uncontrol lable; (2) what time horizon? Most analysts agree that scenario analysis is best suited for long-run s imulation (Wilson, 1 978; Schnaars, 1 987; Armstrong, 1 978; Linneman and K lein, 1 979; van der Heijden, 1 996) and (3) what is the process for constructing and writing scenarios? The development of consistent and comprehensive scenarios typically involves the following steps (Wilson, 1 978; van der Heijden, 1 996): • Step 1 : Selection of scenario themes. This wil l involve consideration of possible future changes in cause and effect, development of internal consistency, avoidance of contradictory sub-themes, and relevance to the issues facing the c lient or stakeholders most interested in the simulation. • Step 2: Carefully detailed, plausible and informative story lines. The story line should ideally be formulated in the form of a qualitative and contextual narrative, and be underpinned by careful documented assumptions that ensure diversity and generate plausible and rich scenarios. A central tenet of story writing is the development of a 'gestalt' or integrated narrative, rather than a disintegrated or piecemeal one. • Step 3: Setting of initial driver values. All initial values should be carefully specified as it is these values which are the main determinants of each scenario. 273 Forecasting is the major alternative approach. It is typically quantitative, relying on historical trends in key system variables to proj ect futures. It is often undertaken with only limited understanding of how a system operates; particularly the consequences of dynamic feedbacks between key system variables. For this reason forecasting is better suited to projecting short-or-medium term futures. 365 • Step 4: Simulation and generation of indicator variables for each scenario. These indicators should encompass variables that may be used to assess ( 1 ) the validity of the model 's structures and behaviours (refer to Section 1 1 .9), and (2) the model ling results. Under ideal circumstances interest and stakeholder groups should be involved ill assessing the model l ing results. Their opinions, views and inputs are useful ill evaluating model results. Refinements may include rewriting of the narrative, resetting of driver values, development or redevelopment of indicators, and improvements to the model 's internal structure. • Step 5: Reporting of results. This includes presenting results to clients and stakeholder groups, and also often analysing the possible pol icy/investment impl ications of each scenario. Comparison of the scenarios is critical as this provides insight into the strengths, weaknesses and tradeoffs of each scenario. This will aid decision makers in selecting the best, or most appropriate, actions given the scenario results. 1 1 .10 .1 ARDEEM Scenarios Three scenarios are developed for ARDEEM below. These scenarios are developed to demonstrate the usefulness of ARDEEM, but require signficiant further work - in particular, further peer review and, in turn, redevelopment.274 • Scenario 1 : Business As Usual(BAU). The 'business as usual' scenario assumes that the trends experienced over the last 1 0 to 20 years will continue to prevail over the next 50 years. These trends are captured in the regression equations used throughout this Chapter to initialise ARDEEM's exogenous variables. Given that these trends are discussed in depth in earlier sections of this Chapter no further discussion is presented here. • Scenario 2: Cornucopian Growth(CG). Under the cornucopian growth scenario market orthodoxy holds sway. This is a world where the ideology of economic rationalism, l iberalism and consumption hold a monopoly of power. Key features of the scenario are ( 1 ) an increased instensification of economic interdependence with other economies, and (2) a desire for increased levels of material wealth. Resource constraints are disputed because technological substitutes are readily avai lable. 274 To this end a series of workshops is scheduled under the Sustainable Pathways FRST contract. These workshops will focus on 'what makes Auckland tick' from the viewpoint of key actors within the Auckland Region, namely: central and local government politicians, central and local government policy makers, infrastructure provides, developers, iwi, business and the public at large. These workshops will be jointly prepared and presented by the author and Professor Richard Le Heron of the School of Geography and Environmental Science, University of Auckland. 366 • Scenario 3: Prudent Pessimism(PP). Aucklanders adopt a communal philosophy of self-sufficiency. Global geopolitical instabi lity and cultural social change override the incentives of economic globalisation. Aucklanders develop a strong and mutual sense of purpose including a shared national desire for sustainable l iving. Underpinning this desire is the bel ief that current material consumption cannot be sustained without future implications i.e. conservation and maintenance of critical natural capital for future generations is seen as paramount. The key exogenous drivers of change in the 'Cornucopian Growth' and 'Prudent Pessimism' scenarios are specified in ful l in Table 1 1 . 1 below. Table 1 1 .1 Summary of D rivers under Each Scenario Fertility rate Mortality rates Net migration Labour force participation Unemployment rates Employment distribution industry Investment rates Depreciation rates Technology rates Substitution effects I nternational exports Interregional exports Ecc-efficiency improvements Cornucopian Growth Woman defer having children until their mid 30's, focusing instead on gaining material wealth. Fertility rates for under 29 year olds are 0.03 percent below the BAU scenario, while fertility rates for over 30's increase marginally at 0.01 percent above the BAU scenario. Reflect past trends. Growth in the economy necessitates skilled and semi-skilled employment opportunities which cannot be fullfilled locally. A more open immigration policy is therefore pursued to avoid possible skill shortages. Net migration numbers grow at 1 2.5 percent above the BAU scenario. Reflect past trends, except for those aged over 60 who engage at a rate 2.5 percent above the BAU scenario; a consequence of a desire for higher levels of material wealth. Reflect past trends. by . The distribution of employment in the primary and secondary industries reflects past trends. More people are however involved in services; in particular retailing and wholesaling. Services thus grow at a rate 1 percent above the BAU scenario. The desire for greater material wealth results in increased i nvestment in manufacturing and service industries at a rate 1 .5 percent above the BAU scenario. Reflect past trends. Technological solutions result in substantial increasing returns in all industries within the economy; at a rate 5 percent above the BAU scenario. Technology continues to offset environmental degradation. Although the use of domestically supplied commodities reflects past trends, the desire for more luxurious commodities results in the substitution of domestically produced commodities for imported commodities. This occurs at a rate 2.5 percent above the BAU scenario. All sectors grow exports at rate 1 percent above the BAU scenario. All sectors grow exports at rate 1 percent above the BAU scenario. Little regard is given to improving the eco­ efficiency of commodities. Consequently, eco­ effiCiency improvements decline at a rate 1 percent belOW the BAU scenario. 367 Prudent Pessimism Past fertility trends prevail for woman under 30 years of age. A marginal decrease in fertility rates (0.01 percent below the equivalent BAU rate) occurs for woman over 30; a consequence of lower material wealth. Reflect past trends. A very tight immigration policy is adopted in an attempt to avoid overexploitation of the nation's natural resources. Immigrants are selected that have skills which will make New Zealand more self-sufficient. Overall, the number of immigrants drops at a rate 5 percent below the BAU scenario. Reflect past trends, except for those over 60 who engage at a rate 0.5 percent above the BAU scenario. This is a result of a desire to retain skilled labour as long as possible in the workforce. Reflect past trends. A trend toward a more self-sufficient economy requires that more people are employed in primary and secondary industries; at a rate 1 .5 percent above the BAU scenario. Primary industry investment rates increase with the desire to be self-sufficient; this occurs at a rate 1 percent above the BAU scenario. Incentives are introduced by government to maintain high quality capital stocks for longer periods within the economy. This results in a depreciation rate 1 percent lower than the BAU scenario. Technological change is felt most in the primary industry at a rate of 5 percent above the BAU scenario, the secondary and tertiary industries however experience less technological innovation and cannot completely offset environmental degradation (growing at a rate 2.5 percent below the BAU scenario). Domestically supplied commodities are substituted for imported goods, respectively growing at 2 percent above and -3.5 percent below the BAU scenario. This is a response to a desire to minimise transportation costs to the environment to encourage local production of commonly consumed commodities. There is movement away from international export as a result of the environmental implications of transportation; at a rate 1 .5 percent below the BAU scenario. Interregional exports grow at a rate of 2 percent above the BAU scenario. Conservation and maintenance of natural capital is pursued both in relation to commodities consumed within the economy, extracted directly from the environment or released back into the environment after use. Overall, ecc-efficiency rates vary between 1 and 3 percent above the BAU scenario. Note: All rates are annualised geometric averages for the 2001 to 205 1 period. 1 1 . 10.2 Simulation Results The results presented in this Section are preliminary and are meant only to i l lustrate the potential value of ARDEEM. 368 There is very little difference between the three scenarios for growth in total population (Figure l 1 .8a) and total employment (Figure 1 1 .8b) between 200 1 and 205 1 . Under the Cornucopian Orowth (CO) scenario, population is projected to grow to only 1 00,000 or so higher than under the Business As Usual (BAU) and Prudent Pessimism (PP) scenarios. There is overall steady population growth from around 1 .2 million in 200 1 to about 2 mil l ion by 205 1 for all three scenarios. Total employment (FTEs) mirrors population growth, with minor differences between the three scenarios. Overall growth in total employment is projected at about 400,000 FTEs between 200 1 and 205 1 . Productivity gains are evident however in projected total output per worker (Figure 1 1 .8c), with output tripling under the CO to $450,000 in 205 1 , one and a half times more than under the BAU, and three times more than under PP. PP shows only a 40 percent growth in total output per worker over 50 years, while CO indicates a 200 percent growth over the same period. Total capital per worker (Figure 1 1 . 8d) shows an initial decl ine under all three scenarios. This decline occurs because capital investment rates were being outstripped by capital depreciation rates. Under CO, total capital per worker begins an upward trend around 20 1 3 while under PP this only occurs 30 years later, around 2033 . Under CO this variable grows rapidly to $650,000 per worker (an 85 percent increase between 200 1 and 205 1 ), while BAU shows a growth of about 25 percent. There is an overall decrease of about 20 percent under PP over the study period with the variable not recovering its 200 1 level by 205 1 . Under CO, total industry output (Figure l 1 .8e) escalates fairly rapidly to nearly five times its 200 1 value over the study period, with a difference of about $200,000 million between CO and BAU at 205 1 . Under PP, total industry output grows relatively modestly, doubling over the 50 year study period. Total industry ODP (Figure 1 1 . 8f) mirrors total industry output, with CO 50 percent h igher at $ 1 50,000 mill ion, and PP 30 percent lower at $70,000 mil l ion, than BAU of $ 1 00,000 mil l ion at 205 1 . BAU total physical supply (Figure 1 1 .8g) grows by 200 mill ion t over the study period, while growth of $380 mill ion t under CO makes it double that under PP at 205 1 . S imi larly, total physical use (Figure 1 1 . 8h) grows by three and a half times under CO, but only doubles under PP in 50 years. Under CO, total physical imports (Figure l 1 . 8 i) rapidly increases nearly threefold to 75 mi ll ion t over the 50 year period, 35 mill ion t and 45 million t higher than under BAU and PP respectively. Under all three scenarios, total physical exports and final demand (Figure 1 1 . 8j) shows relatively s lower growth from about 70 mil l ion t in 200 1 to between 1 20 million t (PP) and 1 70 mi llion t (CO), i.e. a 70 to 140 percent increase. Similarly to total physical imports, CO shows total physical exports and final demand to be 35 mi llion t higher than BAU at 205 1 . Under CO, total raw material inputs (Figure 1 1 . 8k) quadrupled, a requirement nearly 60 percent greater than BAD. In comparison, this variable under PP doubled over the 5 0 years. Similarly, 369 total residual outputs under CG almost triples, while under pp it doubles from 200 1 to 205 1 . Under CG, the economy produces nearly 40 percent more residual outputs (370,000 t) than BAU, and nearly 70 percent more than pp (220,000 t) by 205 1 . Figure 1 1 .8 Total POlmlatioll. 2001-2051 4 M 3 M 2 M 1 M � 0 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) BAU • Total Population Co:> · Total Population pp . Total Population (a) Tot.)1 Olltl)llt I)er Workel . 2001-2051 (1995 $ millioll) 0.6 0.40 0.3 0.15 l====�==�=======:::::::=�---�-i o L-______________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) BAU · Total Output per Wolker Co:> · T otal Output p e r Wolker --------------­ pp . Total Output p e r Wolker ------------- (c) Total Eml)loYlIlellt. 2001-2051 (FTEs) 2 M Hi M 1 M 500,000 - 0 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Ye a r) BAU · Total Employment CG · Total Employment pp . Total Employment (b) Total Cal)it.)I I)er Worker , 2001-2051 (1995 $ millioll) 0.8 0.6 0.4 0.2 o 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) BAU · Total Ca pital per Wolker Co:> · Total Capital per Wolker pp . Total C a pital per Wolker (d) ARDEEM Scenario Analysis: Business As Usual, Cornucopian Growth and Prudent Pessimism Figure 1 1 .8 Tohll lndustry Outl)!n. 2001-2051 (1995 $ lIlillion) 600,000 460,000 300,000 o L-______________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) 8AU · Total Industry Output CC:; · T otal Industry Output ---------------­ pp . Total Industry Output (e) Total Physical SU!)l)Iy. 2001-2051 (tonnes) 600 M 460 M 300 M o L-________________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 T ime (Year) 8AU - Total Physical Supply --------------­ CG · Total Physical Supply -------------­ pp . Total Physical Supply (9) Tot<1l lndustry GDP, 2001-2051 (1995 $ million) 200,000 150,000 100,000 50,000 b=:::=::�=������==::::::==----,....---l o L-____________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) BAU · Total Industry GDP CG · T otal Industry GDP ----------------­ pp . Total Industry GDP (f) Total Physical Use. 2001-2051 (tonnes) 400 M 300 M 200 M o L-________________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) BAU · Total Physical Use CC:; · Total Physical Use pp . Total Physical Use ----------------- (h) ARDEEM Scenario Analysis: Business As Usual, Cornucopian Growth and Prudent Pessimism (Continued) Figure 1 1 .8 Tot,,1 Physic.,l lnlJ)ons. 2001·2051 (tonnes) 80 M 60 M 20 M o 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) eAU · Total Physical Imports CG · T otal Physical Imports ---------------­ pp . Total P hysical Imports -------------- (i) Tot, 1 R"w M,lteri.,l lnllllts, 2001·2051 (tonnes I 600,000 460,000 300,000 150,000 b:=;::::=������::::::::==::=:::.--�--- o 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 T ime (Year) eAU · Total Raw Material lnpuls -------------­ CG · Total Ra"" Material l npuls -----------­ pp . Total Raw Material lnpuls (k) Total Physical EXIIOnS & Final Demand, 2001·2051 (tonnes) 200 M 150 M 50 M o L-______________________________________ � 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 T ime (Year) eAU · Total Physical Exports & F i na l Demand CG · Tota l Physical Exports & F ina l Demand ---------­ pp . Total P hysical Exports & F ina l Demand Total Residual OlltlllltS, 2001·2051 (tonnes) 'lOO ,000 300,000 200,000 L �:;::;�����;;;:::::::::::::::::: :"'---� 100,000 o 2001 2006 201 1 2016 2021 2026 2031 2036 2041 2046 2051 Time (Year) eAU · Total Residual Outpuls --------------­ CG · Total Residual Outpuls -------------­ pp . Total Residua l Outpuls -------------- (I) ARDEEM Scenario Analysis: Business As Usual, Cornucopian Growth and Prudent Pessimism (Continued) 1 1 . 1 1 Limitations of the ARDEEM 373 The ARDEEM, l ike all other mathematical models, is underpinned by a number of assumptions. Often the degree of influence of these assumptions depends on the worldview or belief system of the user analysing the modelling results. It is therefore possible that ARDEEM simulations may, in the eyes of different users, produce results ranging from totally plausible and likely, to completely implausible and unlikely. The purpose of ARDEEM is not however to predict futures, but instead through simulation to investigate the possible dynamic implications of change in the Auckland Region environment-economy system. ARDEEM's major limitations include: • Neglect of natural resource constraints and critical environmental processes. A significant shortcoming of the model is the omission of possible diminishing returns associated with resource scarcity and the degradation of environmental processes. Both international and domestic resource scarcity/environmental degradation will effect the Region's economic growth. Model ling such consequences requires not only data for Auckland Region, but also for the globe. An attempt was made to construct a biogeochemical cycling model of Auckland Region; this however proved too difficult primarily due to data paucity. Nevertheless, a prototype model was constructed for the globe - refer to the Global Biogeochemical Cycling Model (GBCM) presented m Appendix B. This limitation is discussed further in Section 1 1 . 1 1 . 1 below. • Price and substitution effects. The ARDEEM, l ike the Limits to Growth model, may be criticised for the lack of consideration of price effects which might lead to substitution between factor inputs. It is argued that if we know the price elasticity of a commodity then changes in the commodity 's supply and demand may be predicted. The ARDEEM, l ike other simulation models, may only be used to investigate scenarios i .e . it cannot predict the future. It is however possible to test out a scenario with price change and substitution. • Number of industries and commodities. The ARDEEM is currently only a prototype covering three industries and three commodities - a result of time constraints imposed on the completion of this thesis . Only minimal additional system dynamics modelling is however required to extend ARDEEM to, say, 20 or 30 industries, and in the case of commodities to, say, 200 plus commodities. Data constraints will however impose restrictions on industry/commodity coverage in future versions of ARDEEM. • Spatial dynamics. Many of the sustainabi lity issues facing Auckland Region are localised or spatially specific in nature and thus not suitable for simulation in ARDEEM. Consideration of the spatial dynamics i s however beyond the scope of this 374 thesis. With further research it may however be possible to interface ARDEEM with static spatial models already in existence e .g. Auckland Regional Council's Auckland Strategic Planning (ASP) and Auckland Regional Transport (ART) models.275 1 1 . 1 1.1 Extending ARDEEM to Include Ecological Processes The original intention was to build a module of ARDEEM that dynamically modelled the ecological processes in the Auckland Region, building on the static model outlined in Chapter 7 and the growth theory including natural resources developed at the end of Chapter 1 0. Although this could not be achieved within the time constraints of this thesis, a Global Biogeochemical Cycling Model (GCBM) was constructed which is broadly indicative of the type of model that could be developed. The GCBM consists of ( 1 ) 73 'marker element' flows grouped according to the carbon, sulphur, nitrogen, phosphorus, and hydrological cycles, (2) 42 'non-marker element' flows, (3) 2 1 biogeochemical stocks broadly categorised into atmosphere, terrestrial biosphere, marine biosphere, hydrosphere and l ithosphere, and (4) converters that govern the flow of the marker elements out of donor stocks. Initial runs of this global model indicated that the model could reliably simulate a number of anthropogenic perturbations, moving typically to new steady-state values. The establishment of an Auckland Region model would require consideration of a number of additional factors, which time constraints did not permit. Firstly, the global model was based on a closed system, whereas the Auckland Region system is clearly an open system. For example, global emissions of CO2 will affect ecological processes in Auckland Region due to the inflow of CO2 emissions from elsewhere in the world. This openness of the Auckland Region environmental system, particularly for atmospheric processes, is difficult to model due to both data and methodological issues that need to be addressed. Secondly, there is very l ittle reliable data on the nature and extent of biogeochemical processes in the Auckland Region. Very indicative flow data was produced in Chapter 7, but this is inadequate for a dynamic model. Stock data is particularly difficult to obtain. It is estimated that it would take several person years to reliably estimate such data for the Auckland Region. Thirdly, as noted in Chapter 7, there are spatial effects that are important in modelling ecological processes in Auckland Region. The operationalisation of the global model assumed that for each process the spatial 275 This possibil ity is currently being explored under the Sustainable Pathways (MAUX) FoRST contract. Moreover, the final two years of this contract (i .e. 2008 and 2009) wil l be directed at developing a spatially explicit version of ARDEEM. 375 effects are irrelevant. For a regional system, such as Auckland Region, this isn' t the case. For example, point source pollutants at relatively low levels of total emission, have highly localised effects. The same amount of total emissions spread across a larger area, may have little or no effect because of relatively low concentrations which fal l below a biological threshold. 377 Chapter Twelve Thesis Summary and Conclusions This thesis has focused on the integration of economics and ecology in the study of Auckland Region's sustainable development. This integration has been demonstrated not only through the extension of several existing methodologies, and the creation and implementation of new methodologies, but also by critically analysing the underpinning theory concerning sustainable urban development. The purpose of this Chapter is to : ( 1 ) explicitly identify the key contribution of this thesis, in particular, how these contributions extend the body of knowledge beyond that which previously existed; and (2) identify areas for further research and development. 12.1 Thesis Contributions The key contribution of this thesis has been the integration of economics and ecology in the study of Auckland Region's progress toward sustainable development. Underpinning this integration has been the adoption of a systems framework which has considered not only critical interdependencies occurring within the Auckland Region economy, but across the economy­ environment interface and within the natural environment itself. These interdependencies have been studied through a variety of theories (e.g. urban and urban sustainabi l ity perspectives), static modell ing (e.g. monetary input-output models, physical input-output models, lifecyc le assessment, ecological footprinting, materials flow analysis, ecosystem service valuation) and dynamic models (e.g. the ARDEEM and GBCM system dynamics models) lenses. The major theoretical, methodological and empirical contributions of the thesis are considered below. 12. 1 . 1 Theoretical Contributions The theoretical dimensions of the question of urban sustainability and growth formed a critical core of this thesis, and the basis for the modell ing that fol lowed. The main theoretical contributions of this thesis are : • Establishment of a set of key theoretical principles for sustainable development for an integrated environment-economy system. These principles were developed through a comprehensive critique of economic, ecological and thermodynamic interpretations of sustainability theory, and they integrate these various interpretations. 3 78 • Critical analysis of the major theoretical approaches to urban development and sustainability. Based on an environmental sociology viewpoint, the thesis consolidates the major schools of thought regarding urban development and sustainabi l ity. This work extends the earlier contributions of sociologists such as Catton and Dunlap ( 1 978) by: ( 1 ) focusing purely on the urban system, and (2) incorporating theory about cities as ecosystems into the emerging, but currently underdeveloped, NEP perspective. C ities, l ike other ecosystems, function through the metabol ism of mass and energy. This involves complex dynamics typified by time lags, feedbacks, and exchanges with other systems, and is regulated by biophysical constraints (e.g. the laws of thermodynamics). • Critical analysis of growth theory as applied to sustainability. The main domains of growth theory as developed in economics are critically surveyed, and their applicabil ity to the issue of model ling sustainability options in Auckland Region is examined. The salient features of growth theory are revealed and it is argued that much of the research effort conducted to date has proceeded w ithout any consideration of b iophysical or thermodynamic constraints. When these concerns are addressed, growth models exhibit diminishing returns to labour and capital, although (within biophysical l imits) technological progress has potential to offset these effects. 12.1 .2 Methodological Contributions A major contribution of this thesis has been the development of methodologies and associated operational tools for analysing Auckland Region's economy-environment interactions. Moreover, these methods can be applied elsewhere. These methodological contributions are: • Methodology for the generation of regional commodity-by-industry monetary input­ output models: Typically, national industry-by-industry matrices are used to generate counterpart regional matrices. The development of methodologies for regionalising commodity-by-commodity matrices is a recent development led by the theoretical work of lackson ( 1998) and Lahr (200 1 ) . This thesis extends this pioneering work by ( 1 ) establishing a detailed methodological sequence for generating commodity-by-industry input-output matrices, and (2) moving from theory to implementation. Several of the steps within this sequence introduce novel techniques for regionalising components of the commodity-by-industry framework such as value added, final demand and in table balancing. A feature of the sequence is that it allows superior data to be incorporated. Given that the methodology adopts internationally recognised statistical c lassifications (i .e. for industries and commodities) it could easily be replicated in other nations and regions. 379 • Methodology for the generation and regionalisation of commodity-by-industry physical input-output models: In Chapter 6, an innovative methodology for generating not only national, but also regional physical commodity-by-industry input-output models. So far, physical input-output tables have been constructed by only a few statistical agencies. This thesis establishes a detailed methodological sequence for constructing national and regional physical commodity-by-industry input-output models. The methodology, l ike that developed above, allows superior data to be incorporated, ensuring that the unique characteristics of a nation or region are accounted for. The adoption of internationally recognised classification systems and datasets (e.g. harmonised system) promotes easy replication of the methodology in other nations and regions. • Cumulative effects indicator as a measure of eco-efficiency. This indicator, presented in Chapter 6, compares the total economic impact (benefit) to the total environmental impact (cost) of industries in the economy. Specifically, for a given industry it measures the Type I economic multiplier in relation to the ecological multiplier. • Methodology for measuring the dependence of economic industries on ecosystem services. This method specifically links ecosystem service values ($) to sectoral activity in the economy. This enables, by input-output analysis, the: ( 1 ) calculation of the embodied ecosystem services ($) for various sectoral outputs, (2) explicit depiction of these embodied (direct and indirect) ecosystem service inputs into each industry in the economy, using tree diagrams. This leads to an important understanding of how each industry in the economy d irectly and indirectly depends on ecosystem services. This dependence is illustrated for two service industries, namely : air transport, services to transport and storage [35] and business services [42] . Such industries, although apparently far removed from ecosystem services, do actually critically depend on ecosystem services. • Methodology for the calculation of regional ecological footprints and interdependencies using input-output analysis. This methodology, presented in Chapter 8, extends the earlier work of B icknell et al. ( 1 998) to calculate ecological footprints at a sub-national level and to show how different regions ecologically depend on each other. The method focuses on land and energy embodied in interregional trade, because it is argued that it is not only the magnitude of the ecological footprint that matters, but also the impact (or sustainable management practices) at the location of origin. This methodology was published in Ecological Economics, 2004, Volume 50, pp.49-67. • Operationalisation of a dynamic environmental-economic simulation model of an urban system. To the author's knowledge, this is one of the ftrst dynamic models to be developed for an urban region that allows the environmental consequences of economic 380 change to be investigated directly. The model incorporates several unique contributions: ( 1 ) it endogenises the cause-effect chains underpinning several key factors of production (e.g. employment, commodity use and importation)276, (2) it captures the complex monetary and physical interdependencies between industries in terms of supply and use of commodities by adopting a commodity-by-industry input­ output framework for its economic module277, and (3) it captures in physical terms the raw material inputs, and residual outputs, flowing across the Auckland Region environment-economy interface. This model was shown to be a powerful and robust tool for analysing urban sustainability issues, as illustrated by the scenario analysis. • Construction and operationalisation of a dynamic biogeochemical cycling model for the globe. This model was successful constructed for the globe (refer to Appendix B), because the original intention of building such a model for the Auckland Region proved too difficult (refer to Section 1 1 . 1 2) mainly due to lack of data. This global model, for the first time, comprehensively integrates the C, H, P, S and N cyc les. Although more detailed models have been developed for individual cycles, only a few (e.g. Mackenzie et al. ( 1 993), den Elzen et al. ( 1 995)) models have attempted to integrate the biogeochemical cycles, but not as comprehensively as the model presented in Appendix B. The complexity of the model arises from the numerous feedbacks between the biogeochemical cycles, rather than in sub-components of each cycle. It captures not only the major e lemental fluxes of each process, but also all of its by-products and their associated feedbacks. Initial runs of this global model show the model to be robust returning to a steady-state after an initial perturbation. 12.1 .3 Empirical and Knowledge Contributions This thesis has involved producing large multi-dimensional data sets that have led to improved knowledge and insight into the nature of the Auckland Region economy and its interre lationship with the biophysical environment: • Physical Input Output Model of Urban Metabolism. Previous studies have analysed and produced some data on the urban metabol ism of cities, e.g. Newcombe' s study of Hong 276 An attempt is made in Appendix L to endogenise technological change by capturing the dynamics of idea creation and formulation; although operational, only hypothetical data exists to simulate its implications. 277 Ryan ( 1 995) has also utilised input-output analysis in dynamic simulation. His model, however, adopts an industry-by-industry rather than commodity-by-industry input-output framework for analysing economic interdependencies. As industry-by-industry models assume a single homogeneous output per industry (i.e. no joint production), the number of commodities in Ryan's ( 1 995) model must equate to the number of industries. By comparison, ARDEEM may be restricted to, say, 30 industries, but can analyse hundreds, or even thousands, of commodities. 3 8 1 Kong. These studies however have been at a very aggregative level, generating few data and little information about the flows of energy and materials within the economy and across the economy-environment boundary. This thesis, for the first time, produces a detailed data-rich picture of the urban metabolism of a c ity, by using the Physical Input Output Table (PlOT) approach. In 1 997-98, the material input into the Auckland Region economy was 1 28 ,674 kt (raw materials 1 1 1 ,793 kt, imports 1 5 ,082 kt, residual inputs 1 ,799 kt). Most of this material input was destined to be residual outputs into the environment ( 1 09,789 kt), with very little recycling or re-use. The PlOT also quantified these physical inputs and outputs into each of the industries (48) in the Auckland Region economy. • A uckland Region 's Dependence on Ecosystem Services and Processes. The thesis used two separate but related analyses to quantify how the economy critically depends on ecosystem service inputs. F irstly, a preliminary quantitative picture of the main ecological processes (through input-output matrices) was generated. Secondly, the ecosystem services input ($) was quantified to be $ 1 .03 bil l ion, compared with a regional GDP of $33.2 bil l ion for 1 997-98; although it must be remembered that this included only the terrestrial ecosystem services inputs within the Auckland regional boundary. Households was the largest direct appropriator of ecosystem services at $24 1 million, fol lowed by livestock and cropping ($201 mill ion), water supply ($1 89 million), mining and quarrying ($ 1 35 million) and dairy cattle farming ($ 1 34 mil l ion). Service sectors however become more significant when indirect inputs of ecosystems are included in the analysis (refer to Chapter 7). • Ecological Footprint of the A uckland Region. It was found that Auckland Region had the largest ecological footprint (2,300,000 ha), in excess of 1 .3 times that of Canterbury, the next largest region. Auckland (2.00 ha per person) along w ith Wellington (2.40 ha per person) and Nelson ( 1 .86 ha per person) have the largest per capita footprints in New Zealand. These are the three most urban regions and this seems to be the main determinant of their low footprints. More importantly, this footprint analysis also highlights the ecological dependency of Auckland Region on other regions in New Zealand (particularly the Waikato, Northland and Otago), as well as on other countries through international trade. That is, Auckland Region is a significant net importer of embodied land ( 1 ,420,000 ha). Most of Auckland Region ' s footprint was appropriated by the manufacturing sector (48.6 percent), followed by the household sector (24.2 percent), service sector ( 1 5 .9 percent) and the agricultural sector (7.2 percent) • Eco-Efficiency of Industries in the Auckland Region Economy. Valuable information and insights about the eco-efficiency of 48 industries in the Auckland Region economy were generated by the multiplier analysis in Chapter 6. Many industries could be 382 evaluated in terms of the embodied resources (or pollutants )278 they reqUire (or generated) to produce one unit ($) of output. Some industries, for example, showed that they had large multipliers (low eco-effic iency) for energy and associated air emissions - e.g. paper and paper products [ 1 6] , non-metall ic mineral products [20] , and basic metal manufacturing [2 1 ] . Other industries had large methane emissions multipliers - e.g. the land based industries including livestock and cropping [2] , dairy cattle farming [3] and other farming [4] , as well as those associated with the downstream processing of land based products, including meat and meat products [ 1 0] and dairy product manufacturing [ 1 1 ] . 12.2 Limitations and Future Research As outlined in the relevant Chapters, several limitations could be addressed in future extensions of this research. These are d iscussed below. 12.2.1 Theoretical Analysis The theory underpinning the concept of urban sustainability is stil l not well developed and lacks integration across various discipl inary based schools of thought. As is argued in Chapter 3, a research agenda for building and maturing the New Environmental Paradigm approach to urban sustainabi lity could include: • New ecological ideas and terms. Development of specific ecological ideas uniquely applicable to c ities and urban spaces. There is a tendency to draw ideas and analogies directly from b iological ecology and apply them to c ities. At the very least, applying ideas such as carrying capacity to urban situations should be undertaken with care; • Stronger links between HEP and NEP research. Forging stronger l inks between the estab lished HEP research and the more recent NEP thinking. For example, few studies , with the exception of Huang ( 1998), attempt to explain the ecological determinants of urban phenomena (e.g. spatial zonation) that have long been observed in the HEP l iterature. The two schools of thought most often operate in complete isolation of each other; and • Institutionalisation. Institutionalising the NEP view of urban sustainabi lity and building a critical mass of research activity in this area. Institutionally the NEP-based field of 278 Other resources covered in the ecological multiplier analysis include land and water inputs. Pollutants include carbon dioxide, nitrous oxide, methane, biological oxygen demand, phosphorus, total kjeldahl nitrogen and solid wastes. 383 urban sustainabi lity is weak, with no strong international community of scholars or teaching institutions. HEP scholars from the social sciences dominate the field, while NEP scholars in urban areas are often marginalised. The issue of growth theory and how to adjust it to incorporate ecological and thermodynamic constraints is even more problematic. Most growth theory is predicated on assumptions of technological optimism and neglects ecological processes or consequences. Mackenzie et al. ( 1 993), den Elzen et al. ( 1 995), and Ayres (200 1 , 2005) inter alia are some of the few analysts to acknowledge biophysical constraints when considering growth theory. Future research agendas must critically examine the HEP assumptions of such growth theory, and propose alternative theoretical frameworks which are more appropriate for analysing urban sustainabil ity . A broad level theoretical concern is the lack of scholarly enquiry into the connections between the critical theoretical elements identified in this thesis; e.g. sustainability theory, urban theory and growth theory. Theorists tend to specialise, with l ittle or no knowledge of, or interest in, other theoretical areas. Few (if any) theoretical enquiries truly integrates these theoretical elements, but this integration is needed urgently if we are to develop a body of theory that enhances the holistic understanding of urban sustainability. 12.2.2 Static Systems Analysis The static systems analysis of Auckland Region environment-economy interactions is a strength of this thesis because it is comprehensive, and has produced high quality of data. evertheless, some areas could be developed by future research. • Chapter 5 - Economic Input-Output Model. The regionalisation process makes some assumptions that lead to imprecise calculation of the commodity-by-industry model of Auckland Region. Future research could be focus either on improving the method (thus alleviating the need to make these assumptions) or on collecting superior data to replace the assumptions. Areas needing attention are: cross hauling (only net flows are considered), self sufficiency (the regionalisation method assumes maximum self­ sufficiency in economic production), the technology assumption (the regionalisation method assumes the region has the same mix of technologies as the nation), and consumption patterns (the method assumes the same consumption patterns occur in the region as the nation). 384 • Chapter 6 - Physical Input-Output Model. The PlOT was constructed using mainly estimated data. Several consequent l imitations could be addressed by further research. These limitations include: ( 1 ) the export and import price data do not take account of import and export exclusions, coding errors and quality improvements; (2) estimated national (rather than actual regional) price data are used to convert value ($) data to physical data in the regional PlOT, but actual regional price data would better estimate the regional PlOT; and (3) the economic input-output model (from Chapter 5) was a starting point only and, as mentioned above, this could be improved. • Chapter 7 - Input-Output Model of Ecological Processes . This model could be substantially improved, as the data presented in Chapter 7 are only broadly indicative. Shortcomings and improvements include: ( 1 ) the model currently only records fluxes within the region, but future development should focus on 'cross boundary' flows; (2) all data in the model are estimated from scaled down global data. Actual data for the Auckland Region should be collected and synthesised into the model, or model led data based on knowledge of regional processes and system attributes should be used, following the type of approach used by Costanza et al. ( 1 973) for the Mississippi Deltaic P lain Region; and (3) future research should attempt to connect the flows in the ecological input-output model to the models of the economy presented in Chapters 5 and 6. • Chapter 8 - Ecological Footprint Analysis. This analysis is considered robust and needs l ittle improvement. However, future development could focus on land productivity (quality) differences in the appropriated land, and on extending the ecological footprint to include resources other than land, and other pollutants other than carbon dioxide. 12.2.3 Dynamic Systems Analysis The ARDEEM is a tool aimed not at predicting futures, but at understanding how feedbacks, non-linearities and time lags may influence key system variables, and through s imulation investigating the possible dynamic implications of change in the Auckland Region environment­ economy system. This should be borne in m ind while considering ARDEEM's primary l imitations and opportunities for development: • Price and substitution effects . The ARDEEM, l ike the L imits to Growth model, may be criticised for not considering price effects which might lead to substitution between factor inputs. If we know the price elastic ity of a commodity then changes in the commodity's supply and demand may be predicted. The ARDEEM, like other 385 simulation models, may only be used to investigate scenarios i .e. it cannot predict the future. It is however possible to test a scenario with price changes and substitutions. • Neglect of critical environmental processes. A significant weakness of the model is that it neglects the critical life supporting biogeochemical processes of the environment. These processes provide humans with resources, waste assimilation, opportunities for spiritual fulfilment, scientific learning and so on. Therefore, any environmental­ economic model which does not consider their influence is incomplete. • Number of industries and commodities. The ARDEEM is currently only a prototype covering three industries and three commodities. This is a consequence of time constraints imposed on the completion of this thesis. However, only minimal additional system dynamics model l ing is required however to extend ARDEEM to, say, 20-30 industries, and for commodities to, say, 200 or more commodities. • Spatial dynamics . Many sustainabi lity issues facing Auckland Region are localised or spatially specific and thus not suitable for simulation in ARDEEM. Consideration of these spatial dynamics is beyond the scope of this thesis; nevertheless, further research may enable ARDEEM to interface with existing static spatial models e.g. Auckland Regional Counci l 's Auckland Strategic Planning (ASP) and Auckland Regional Transport (ART) models. 3 87 List of PhD Outputs During the course of this research four papers were accepted for publication in journals ( lead author in two cases, analytical support in two cases), two papers are manuscripts in progress (one lead author, one secondary author), three papers were presented at conferences, one book chapter was accepted for publication (principal modeller), two refereed reports were published (one lead author, one analytical support), and nine unpublished reports were written for various commercial cl ients. All of these research outputs build on methodologies created in this thesis, and to a lesser extent, report its key findings. The emphasis of these outputs is not however solely on Auckland Region, but instead on the nation, other regions and specific economic industries. The full list of outputs is presented below: Papers Published Papers McDonald, G.W. & Patterson, M.G. (2004). Ecological Footprints and Interdependencies of New Zealand Regions. Ecological Economics, 50: 49-67. Jollands, .A., Golubiewski, .E. & McDonald, G.W. (2005). Impl ications of Changing Employment Patterns on Urban Ecosystem Service Requirements. The International Journal 0/ Environment, Workplace and Employment, 3-4: 3 1 0-335 . Papers Accepted/or Publication McDonald, G.W., Forgie, V .E. , & MacGregor, C. (In press). Treading Lightly: Ecofootprints of New Zealand' s Ageing ation. Ecological Economics. Jollands, N.A., Golubiewski, N.£. & McDonald, G.W. (In press). Linking Policy and Science: A Study of Metro Christchurch Ecosystem Service Appropriation. The International Journal 0/ Environment, Workplace and Employment. Forthcoming Papers - written, but not yet submitted/or publication McDonald, G.W., & Le Heron, R. (2005). Changes in the Clusters o/Comparative Advantage in the Auckland Region Economy 1986-2001 . Manuscript in progress. 388 McDonald, G. W., Le Heron, R.B. , & Patterson, M.G. (2005). Canterbury 's 'Hidden ' Economy: Assessing the Value of the Region 's Ecosystem Services. Manuscript in progress. Patterson, M.G. & McDonald, G.W. (2005). Regional Level Environmental Accounting in New Zealand: EcoLink and Other End-User Led Initiatives. Manuscript in progress. Conference Papers McDonald, G.W. & Patterson, M.G. (2000). Ecological Footprints and the Interdependencies of New Zealand Regions. International Society of Ecological Economics. Presented at People and Nature : Operationalising Ecological Economics, S - 8 July, 2000, The Australian National University, Canberra, Australia. McDonald, G.W. (2003). Treading Lightly: Recent Ecological Footprint Work in New Zealand. New Zealand Centre of Ecological Economics. Presented at Ecological Economics at the Cutting Edge, November 1 6, 2003, Auckland, New Zealand. F lemmer, c.L., Flemmer, R.c., McDonald, G. W., Archer, R.H., & Cleland, D.J. (2005). An Assessment of the Ecological Impact of the New Zealand Dairy Farming Sector. Australia New Zealand Society for Ecological Economics. Ecological Economics in Action, 1 1 - 1 3 December, Massey University, Palmerston North, New Zealand. Book Chapters Patterson, M.G., G.W. McDonald, N.E. Golubiewski, V.E. Forgie & N.A. Jollands (2006), Cl imate change Impacts on Regional Development and Sustainabil ity: An Analysis of New Zealand Regions. In M. Ruth (Ed.), Smart Growth and Climate Change: Regional Development, Infrastructure and Adaptation (pp.B2-10B) . Edward Elgar Publ ishing Ltd. Reports Published (Refereed) Reports McDonald, G.W. & Patterson, M.G. (2003). Ecological Footprints of New Zealand Regions. Environmental Reporting Technical Paper. Ministry for the Environment. Wellington, New Zealand. 1 62pp. ISBN 0-478-24085-6 . Downloadable from: http://www.mfe.govt.nzlpublications/ser/ eco-footprint -sep03/eco-footprint -sep03 . pdf 389 Patterson, M.G. & McDonald, G.W. (2004). How Clean and Green is New Zealand Tourism? Lifecycle and Future Environmental Impacts. Land Research Science Series No. 24. Lincoln: Manaaki Whenua Press, 1 4 1 pp. ISBN 0-478-09359-4. Downloadable from http://www .mwpress.co.nzJstore/viewItem.asp?idProduct=498 Unpublished Reports McDonald, G.W. & Patterson, M.G. (200 1 ). Ecological Footprint of the Waikato Region. Report for Environment Waikato. Auckland: McDermott Fairgray Group Ltd. McDonald, G.W. (2003). Canterbury Region's Hidden Economy: Assessing the Value of the Region's Ecosystem Services and Biodiversity. Report for Christchurch City Council . Palmerston North: Landcare Research Ltd. McDonald, G.W. & Patterson, M .G. (2003). Ecological Footprint and Interdependencies of Auckland City. Report for Auckland City Council . Palmerston North: Landcare Research Ltd. McDonald, G. W. (2004). Ecological Footprint of Waitakere C ity Council . Report for Waitakere City Council. Takapuna: Market Economics Ltd. McDonald, G. W. (2004). Conceptual Framework for a New Zealand F inancial and Physical Input-Output Accounting System. FoRST Research Contracts MAUX0306 and WROX0305. Takapuna: Market Economics Ltd. McDonald, G.W. (2005). A Primer on Input-Output Analysis. FoRST Research Contracts MAUX0306 and WROX0305. Takapuna: Market Economics Ltd. McDonald, G.W. (2005). Construction of a Monetary Input-Output Accounting System for New Zealand and its Regions. FoRST Research Contracts MAUX0306 and WROX0305. Takapuna: Market Economics Ltd. McDonald, G.W. (2005). Construction of a Physical Input-Output Accounting System for New Zealand and its Regions. FoRST Research Contracts MAUX0306 and WROX0305 . Takapuna: Market Economics Ltd. 390 McDonald, G .W. & MacGregor, C. (2004). Land Use Accounts Technical Report. FoRST Research Contracts MAUX0306 and WROX0305 . Takapuna: Market Economics Ltd. 39 1 References Abramovitz, M. ( 1 956). Resource and Output Trends in the United States Since 1 870. American Economic Review, 46: 5-23 Acosta, J.J. (2000). Physische Input-Output-Rechnung: Ansdt=e, Moglichkeiten und Probleme einer aktivitatsbe=ogenen StofJjlussrechnung auf nationaler und regionaler Ebene, MS Thesis. Karlsruhe, Germany. Adriaanse, A., Bringezu, S., Hammond, A., Moriguchi, Y., Rodenburg, E., Rogich, D., et al. ( 1 997). Resource Flows: The Material Basis of Industrial Economies (a joint publication of the World Resources Institute (WRI); the Wuppertal Institute; the Netherlands Ministry of Housing, Spatial Planning, and Environment; and the National Institute for Environmental Studies, Washington, DC). Aghion P. & Griffith, R. (2005). Competition and growth: reconciling theory and evidence. Cambridge, MA: MIT Press. Aghion, P. & Howitt, P. ( 1992). A Model of Growth through Creative Destruction. Econometrica 60: 323-35 l . Aghion, P. & Howitt, P. ( 1 998). Endogenous growth theory. Cambridge, MA: MIT Press. Alberti, M. ( 1 996). Measuring urban sustainability. Environmental impact assessment review, 16: 3 8 1 -424. Alfred, L.E. ( 1 995) . Urban Dynamics - The F irst F ifty Years. System Dynamics Review, 1 1 : 3 . Allaby, M . ( 1 994). The Concise Oxford Dictionary of Ecology. Oxford and New York: Oxford University Press. Allenby, B.R. ( 1 999). Industrial Ecology: Policy Framework and Implementation. New Jersey: Bell Laboratories, Lucent Technologies. AImon, C. (2000). Product-to Product Tables via Product-Technology with No Negative Flows. Economic Systems Research, 12 : 27-43 . 392 Alter, M. ( 1 982). Carl Menger and Homo Oeconomicus : Some Thoughts on Austrian Theory and Methodology. Journal of Economic Issues, 1 6: 1 49- 1 60. Anderson, c .H. ( 1 976). The Sociology of Survival: Social Problems of Growth. Homewood, I l l inois: Dorsey. Andreae, M.O. ( 1 985). The emission of sulfur to the remote atmosphere. Dordrecht, The Netherlands: Reidel. Argyris, C. ( 1 990). Overcoming Organizational Defenses. Wellesley, Massachusetts: Allyn and Bacon. Arrow, K.K. ( 1 962). The economic implications of learning by doing. Review of Economic Studies, 29: 1 55- 1 73 . Asimakopulos, A. & Weldon, J .C. ( 1 965). A synoptic view of some s imple models of growth. History of Political Economy, 1 7(4) : 6 1 9-635 . Auckland Regional Council . ( 1 999). State of the Environment Report. Auckland : Auckland Regional Council . Auckland Regional Council . (2003). Business and Economy 2003 Auckland Region. Auckland: Auckland Regional Council . Auckland Regional Council . (2004). Business and Economy 2004 Auckland Region. Auckland: Auckland Regional Council. Auckland Regional Counci l & Auckland C ity Council . ( 1 996). Preliminary Investigation of Construction and Demolition Wastes in the A uckland Region. Unpublished Report. Auckland: Auckland Regional Council. AUICK (Asian Urban Information Centre of Kobe). ( 1 994). Kobe Hosts International Symposium on Urban Metabolism. Newsletter No 1 6 Jan 1 994. Retrieved 1 4 July 2005 from http://www.auick.org/database/apc/apcO 1 6/apcO 1 604.html . Ayres, R.U. ( 1 978) . Resources, Environment, and Economics. New York: John W iley and Sons Ltd. 393 Ayres, R.U. ( 1 993a). Cowboys, Comucopians and Long-Run Sustainabi l ity. Ecological Economics, 8: 1 89-207. Ayres, R.D. ( 1 993b). Materials/Energy F lows and Balances as a Component of Environmental Statistics. In A. Franz, and C. Stahmer (Eds.), Approaches to Environmental Accounting (pp. 1 26- 1 42). Heidelberg: Physica. Ayres, R.U. ( 1 996). Industrial Metabolism and the Grand Nutrient Cycles. Fontainebleau, France: INSEAD Centre for the Management of Environmental Resources. Ayres, R.U . (2000). Commentary on the Utility of the Ecological Footprint Concept. Ecological Economics, 32: 347-349. Ayres, R.U. & Kneese, A.V. ( 1969). Production, Consumption, and Externalities. American Economic Review, 59: 282-298 . Ayres, R.U. & Nair, 1 . ( 1 984). Thermodynamics and Economics. Physics Today, November: 62-7 1 . Ayres, R. U. & Noble, S .B. ( 1 978). MaterialslEnergy Accounting and Forecasting Models. In R.U. Ayres (Ed.), Resources, Environment and Economics - Applications of the Materials/Energy Balance Principle. New York: John Wiley and Sons. Ayres, R.U., Schlesinger, W.H., & Socolow, RH. ( 1 994). Human Impacts on the Carbon and Nitrogen Cycles. In R.H. Socolow, c.J. Andrews, F.G. Berkhout & V.M. Thomas (Eds .), Industrial Ecology and Global Change (pp. 1 3 1 - 1 55) . Cambridge: Cambridge University Press. Ayres, R.U. (200 1 ). The minimum complexity of endogenous growth models: the role of physical resource flows. Fontainebleau Cedex: Center for the Management of Environmental Resources, INSEAD, Boulevard de Constance. Ayres, R.U. & van den Bergh, J.C.J.M. (2005). A theory of economic growth with material/energy resources and dematerialisation: Interaction of three growth mechanisms. Ecological Economics, 55, 96- 1 1 8 . 394 Azar C. & Dowlatabadi, H. ( 1 999). A Review of Technical Change in Assessment of Climate Policy. Annual Review of Energy and Environment, 24: 5 1 3-544. Baars, D. ( 1 997). Bridging the Gap Between Sustainability Theory and Policy: An Assessment of Regional Policy Statements. Palmerston North: Massey University. Bacc ini, P. ( 1 996). Understanding regional metabolism for a sustainable development of urban systems. Environmental Science and Pollution Research, 3(2): 1 08- 1 1 1 . Bacharach, M. ( 1 965) Estimating Nonnegative Matrices From Marginal Data. International Economic Review, 6: 294-3 1 0 . Ballance Agri-Nutrients . (2002). Ballance Kapuni Manufacturing Process Pamphlet. Kapuni: Ballance Agri-Nutrients . Balmford, A., Bruner, A., Cooper, P., Costanza, R., Farber, S . , Green, R.E. , et al. (2002). Economic Reason for Conserving Wild Nature. Science 297: 950-953 . Bannock, G., Baxter, R.E. & Davis, E. ( 1 992). The Penguin Dictionary of Economics. London: Penguin Books. Barber, M. ( 1 978). A Retrospective Markovian Model for Ecosystem Resource Flow. Ecological Modeling, 5: 1 2 5- 1 35 . Barbier, E .B . ( 1 989). Economics, Natural Resource Scarcity and Development: Conventional and Alternative Views. London: Earthscan. Barbier, E.B. ( 1 990). Alternative Approaches to Economic-Environmental Interactions. Ecological Economics, 2: 7-26. Barbier, E.N. & Markandya, A. ( 1 989). The Conditions for Achieving Environmentally Sustainable Economic Development. LEEC Paper 89-0 1 . London: London Environmental Economics Centre. Barker, T. & Kohler, J. ( 1 998). Equity and Ecotax Reform in the EU: Achieving a 1 0% Reduction in CO2 Emissions using Excise Duties. Fiscal Studies, 19: 375-402. 395 Barnett, HJ. & Morse, C. ( 1 963). Scarcity and Growth: The Economics of Natural Resource Availability. Baltimore MD and London: Johns Hopkins University Press. Bartelmus, P. ( 1 994). Environment, Growth and Development - The Concepts and Strategies 0/ Sustainability. London and New York: Routledge. Barton, L. (200 1 ). Peer-review of the EcoLink Water Accounts. Hamilton, Landcare Research Ltd. Bates, J.and Bacharach, M. ( 1 963). Input-Output Relationships: 1 954- 1 966, A Programme for Growth 3 . London: Chapman and Hall. Baumol, W. ( 1 970). Economic Dynamics: An Introduction (3rd ed.). London: The MacMillan Company. Baumol, WJ. & Oates, W.E. ( 1 979). Economics, Environmental Policy and the Quality 0/ Life. Englewood Cliffs, NJ: Prentice Hall. Baumol, W.J . & Oates, W.E. ( 1 988). The Theory of Environmental Policy (2nd ed.). Cambridge: Cambridge University Press. Baumol, WJ. , B l inder, A.S. & Scarth, W.M. ( 1 99 1 ). Economics: Principles and Policy (3rd ed.). Toronto : Harcourt Brace Jovanovich. Beckerman, W. ( 1 974). In Defence of Economic Growth. New York: J onathan Cape. Bell, M.M. (2004). An Invitation to Environmental Sociology. London: Sage Publications Inc . Bell, W. & Boat, M. ( 1 957). Urban Neighborhoods and Informal Social Relations. American Journal o/Sociology, 62: 39 1 -398. Berkes, F. & Folke, C. ( 1 994). Investing in Cultural Capital for Sustainable Use of Natural Capital. In A. Jansson, M. Hammer, C. Folke, & R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington DC: Island Press. Berry, B.J .L. ( 1 985). Islands of Renewal in Seas of Decay. In P.E. Peters on (Ed.), The New Urban Reality (pp. 69-98). Washington, DC : Brookings Institute. 396 Berry, B.J.L. & Kasarda, J.D. ( 1 977). Contemporary Urban Ecology. New York: Macmillan. Berry, B.J.L. & Rees, P.H. ( 1 969). The Factorial Ecology of Calcutta. American Journal of Sociology, 74: 445-49 1 . Betz, F. & de Azevedo, J.A.C. ( 1 976). Structural Global Models. In W.C. Churchman & R.O. Mason (Eds.), World Modelling: A Dialogue. Amsterdam: North Holland Publishing Company. B ianciardi, c., Donati, A., & Ulgiati, S. ( 1 993). On the relationship between the economic process, the Carnot cycle and the entropy law. Ecological Economics, 8: 7-1 0. B ianciardi, C., Tiezzi, E., & Ulgiati, S. ( 1 993). Complete recycling of matter in the frameworks of physics, biology and ecological economics. Ecological Economics, 8: 1 -5 . B ianciardi, C., T iezzi, E . , & Ulgiati, S . ( 1 994). Letters to the Editor. Recycling of matter: A reply. Ecological Economics, 9: 1 92- 1 93. Bianciardi, C., Tiezzi, E. , & Ulgiati, S. ( 1 996). Response. The 'recycle of matter' debate. Physical principles versus practical impossibi lity. Ecological Economics, 19: 1 95- 1 96. B icknell, K.B., Ball, RJ., Cullen, R., & Bigsby, H.R. ( 1 998). New Methodology for the Ecological Footprint with an Application to the New Zealand Economy. Ecological Economics, 27: 149- 1 60. Bigg, E.K., Chameides, W.L., & Kley, D. ( 1 985) . The transformations of sulfur and nitrogen in the remote atmosphere. In J.N. Galloway, R.J. Charlson, A.O. Andreae & H. Rodhe (Eds.). The Biogeochemical Cycling of Sulfur and Nitrogen in the Remote Atmosphere (pp.67-80). Dordrecht, The Netherlands : Reidel. B inswanger, M. ( 1 993) . From Microscopic to Macroscopic Theories: Entropic Aspects of Ecological and Economic Processes. Ecological Economics, 8: 209-234. Black, M., Guy, M., McLel lan, N. (2003). Productivity in New Zealand 1998 to 2002. Wellington: New Zealand Treasury. 397 B larney, R. & Common, M. ( 1 994). Sustainability and the Limits to Pseudo Market Valuation. In J .C.J.M. van den Bergh, & J. van der Straaten, (Eds.), Toward Sustainable Development - Concepts, Methods and Policy (pp. l 65-205). Washington DC: Island Press. B laug, M. ( 1 974). The Cambridge Revolution: Success or Failure? London: Institute of Economic Affairs. B laug, M. ( 1 985). Great Economists Since Keynes: An Introduction to the Lives and Works of One Hundred Modern Economists . Brighton: Wheatshelf-Harvester. B lumenfeld, H. ( 1 949). Theory of City Form, Past and Present. Journal of Society of Architectural Historians, 8: 7 -16 . Bolin, B . , Degens, E. T., Kempe, S . , & Kettner, P. (Eds.) . ( 1 979) The Global Carbon Cycle. SCOPE Report No. 1 3 . Chichester: Wiley. Boitzmann, L. ( 1 974). The Second Law of Thermodynamics. In B. McGuinness, (Ed.), Theoretical Physics and Philosophical Problems: Selected Writings of L. Bolt::mann. Dordrecht: D. Reidel. (Original work published 1 905) Boomsma, P. & Oosterhaven, J. ( 1 992). A Double-entry Method for the Construction of Bi­ regional Input-output Tables, Journal of Regional Science, 32: 269-84. Booth, D.E. (2004). Hooked on Growth. Lanham: Rowman and L ittlefield Publishers Inc. Borgstrom, G. ( 1 967). The Hungry Planet. New York, USA: McMillan. Borgstrom, G. ( 1 973). Harvesting the Earth. New York: Abelard - Schuman. Bostrom, A., Fischhoff, B . , & Morgan, G.M. ( 1992). Characterizing Mental Models of Hazardous Processes: A Methodology and an Application to Radon. Journal of Society, 48: 85- 1 00. Boulding, K. ( 1 966). The Economics of the Coming Spaceship Earth. In H. Jarrett (Ed.), Environmental Quality in a Growing Economy (pp.3- 1 4) . Baltimore, MD: Johns Hopkins University Press. 398 Boumans, R. , Costanza, R, Farley, J. , Wilson, M.A. , Portela, R, Rotmans, J. , Villa, F. , & Grasso, M. (2002). Modeling the dynamics of the integrated earth system and the value of global ecosystem services using the GUMBO model . Ecological Economics, 4 1 : 529-560. Bounds, M. (2004). Urban Social Theory: City, Self and Society. Victoria, Australia: Oxford University Press. Bourque, PJ. Chambers, E.J., Chiu, J.S.Y., Denman, F.1. , Dowdle, B., Gordon, G., et al. ( 1 967). The Washington State economy: An input-output study. Seattle: University of Washington Graduate School of Business Administration, Occasional Paper 22. Bowen, HJ.M. ( 1 979). Environmental Chemistry of the Elements. London: Academic Press. Boyden, S. ( 1 977). Integrated ecological studies of human settlements. Impact of Science on Society, UNESCO, 27: 1 59-69. Boyden, S., Millar, S., Newcombe, K. & O'Neill, B. ( 1 98 1 ). The Ecology of a City and its People. Canberra: Australian National University Press. Boyle, M. & Rogerson, RJ. (200 1 ). Power, Discourses and City Trajectories. In R. Paddison (Ed.), Handbook of Urban Studies (pp. 402-4 1 6). London: SAGE Publications. Breslav, M.A., Berkowitz, A.R., Nilon, C.H., & Hollweg, K .S. (2000). Cities are ecosystems! : new trend to study urban areas. Ecological Economics, 32: 337-339. Bretschger, 1. ( 1 999). Growth Theory and Sustainable Development. Cheltenham UK: Edward EIgar. BrilIouin, L. ( 1 962). Science and Information Theory (2nd ed.). New York: Academic Press Inc. Brimblecombe, P., Hammer, c., Rodhe, H., Ryaboshapko, A. , & Boutron, C.F. ( 1 989). Human influence on the sulphur cycle. In P. Brimblecombe & A.Y. Lein (Ed.), Evolution of the global biogeochemical sulphur cycle (pp. 77-1 2 1 ). SCOPE 39. Chichester, UK: John Wiley & Sons. Bringezu, S. (2000). Ressourcennutzung in Wirtschaftsraumen. StofJstromanalysen fur eine nachhaltige Raumentwicklung. Berlin, Tokyo, New York. 399 Bringezu, S. & Moriguchi, Y . (2002). Material Flow Analysis. In R. U. Ayres & L. W. Ayres (Eds.), A Handbook of Industrial Ecology (pp. 79-90). Cheltenham: Edward Elgar. Browder, J., Littlejohn, c., & Young, D. ( 1 976). The South Florida Study: South Florida, Seeking a Balance of Man and Nature. Gainesvi l le, FL: University of Florida. Brown, M.T. ( 1 98 1 ). Energy Basis for Hierarchies in Urban and Regional Systems. In WJ. Mitsch, R.W. Bosserman, & J .M. Klopatek (Eds.), Energy and Ecological Modeling: Developments in Environmental Modeling 1 (pp. 5 1 7-534). Proceedings of a symposium sponsored by the International Society for Ecological Modeling. New York: E lsevier Scientific Publ ishing Company. Brown, M.T. & Herendeen, R.A. ( 1 996). Embodied Energy Analysis and Emergy Analysis: A comparative view. Ecological Economics, 19: 2 1 9-235 . Brown, M.T. & Ulgiati, S. ( 1 998). Emergy Evaluation of the Environment: Quantitative Perspectives on Ecological Footprints. Advances in Energy Studies. Energy Flows in Ecology and Economy. Presented at VI European Week of Scientific Culture, 22-28 November 1 998. Porto Venere, Italy. Bruce, J.M. ( 1 970). lntergenerational Occupational Mobility and V isiting with Kin and Friend. Social Forces, 49: 1 1 7- 1 27. Bullard, C.W. & Hannon, B. ( 1 976). Energy Growth in the US Economy. CAC Document 1 72 . Urbana, IL : Energy Research Group, University of Il l inois . Bullard, C.W. & Herendeen, R.A. (1 975a) . Energy impact of consumption dec isions. Proceedings of IEEE, 63(3): 484-493 . BulIard, C. & Herendeen, R. ( 1 975b). The Energy Costs of Goods and Services. Energy Policy, 3 : 268-278. BulIard, C.W., Penner, P.S. & Pi lati, D .A. ( 1 978). Net energy analysis: A handbook for combining process and input-output analysis. Resources and Energy, 1 : 267-3 1 3 . 400 Buonanno, P., Carraro C., and Galeotti, M. (200 1 ) . Endogenous Induced Technical Change and the Costs of Kyoto. Milan, Italy: Nota di Lavoro 64.200 1 , Fondazione Eni Enrico Mattei (FEEM). Burch, W.R. ( 1 97 1 ). Daydreams and Nightmares: A Sociological Essay on the American Environment. New York: Harper and Row. Burch, W.R. ( 1 976). The Peregrine Falcon and the Urban Poor: Some Sociological Interrelations. In PJ. Richardson & J. McEvoy III (Eds.), Human Ecology: An Environmental Approach. North Scituate, Mass. : Duxbury. Burness, S . , Cummings, R., Morris, G. & Paik, 1. ( 1 980). Thermodynamic and economic concepts as re lated to resource-use policies. Land Economics, 56( 1 ) : 1 -9 . Burton, 1. & Kates, R.W. ( 1965). Readings in Resource Management and Conservation. Chicago: University of Chicago. Burtt, E.S. (Ed.) ( 1 999). Financial Budget Manual 1999. Lincoln: Farm Management Group, Applied Management and Computing Division, Lincoln University. Butcher, G.V. ( 1 985). Cost Benefit Handbook. Regional Income Output and Employment Multipliers: Their Uses and Estimates of Them (VolA). Well ington: Economics Division, Ministry of Agriculture and F isheries. Butcher, S .S . , Charlson, RJ., Orians, G.H., & Wolfe, G.V. ( 1 992). Global biogeochemical cycles. New York: Academic Press. Buttel, F .H. ( 1 997). Social Institutions and Environmental Change. In M. Redclift & G. Woodgate (Eds.), The International Handbook of Environmental Sociology (pp. 40-54) . Cheltenham: Edward Elgar. Buttel, F.H. & Humphrey, c.R. (2002). Sociological Theory and the Natural Environment. In R.E. Dunlap & W. Michelson (Eds.), Handbook of Environmental Sociology (pp. 33 -69). Westport, CT: Greenwood Press. Buttel, F.H., Dickens, P., Dunlap, R.E. & Gijswijt, A. (2002). Sociological Theory and the Environment: An Overview and Introduction. In R.E. Dunlap, F .H. Burtel, P. Dickens, & A. 401 Gijswijt (Eds.), Sociological Theory and the Environment: Classical Foundations, Contemporary Insights (pp. 3-32). Lanham: Rowman & Littlefield Publishers Inc. Bylund, E. ( 1 960). Theoretical conditions regarding the distribution of settlement in inner north Sweden. Geograjiska Annaler, 42: 225-3 1 . Caldwell, L.K. ( 1 990). Between Two Worlds. Cambridge: Cambridge University Press. Carson, R. ( 1 962). The Silent Spring. London: Penguin. Carter, A.J. , Peet, NJ., & Baines, J.T. ( 1 98 1 ). Direct and indirect energy requirements of the New Zealand economy. New Zealand Energy Research and Development Committee Report No.55 . Auckland: University of Auckland. Carter, L. ( 1 999). Cumulative effects assessment. In J. Petts (Ed.), Handbook of environmental impact assessment: Vol. 1 (ppA05-440). Oxford: B lackwel l Science. Caste l is, M. ( 1 977). The Urban Question: A Marxist Approach. London: Amold. Castelnuovo, E., Galeotti, M., Gambarelli, G., & Vergalli, S. Learning-by-Doing vs. Learning by Researching in a model of climate change policy analysis. Ecological Economics, 5-1, 26 1 - 276 Catton, W.R ( l 976a). Toward Prevention of Obsolescence in Sociology. Sociological Focus, 9: 89-98. Catton, W.R. ( l976b). Why the future isn' t what it used to be (and how it could be made worse than it has to be). Social Science Quarterly, 57: 276-29 1 . Catton, W.R ( 1 982). Overshoot: The Ecological Basis of Revolutionary Change. Urbana: University of Ill inois Press. Catton, W.R & Dunlap, RE. ( 1 978). Environmental Sociology: A New Paradigm. The American Sociologist, 13: 4 1 -49. Chapman, P .F. ( 1 977). Energy costs: a review of methods. In J .A.G. Thomas (Ed.), Energy Analysis. Surrey: IPC Science and Technology Press Ltd. 402 Chapman, P.F. & Roberts, F. ( 1 983). Metal resources and energy. London: Butterworths. Charison, RJ., Anderson, T.L., & McDuff, R.E. ( 1 992a). The Sulfur Cycle. In S .S. Butcher, R.J. Charlson, G.H. Orians & G.V. Wolfe (Eds.), Global Biogeochemical Cycles (pp. 285-300). London: Academic Press. Charlson, RJ., Lovelock, J.E., Andreae, M.O., & Warren, S.G. ( 1 987). Oceanic phytoplankton, atmospheric sulphur, cloud albedo, and climate. Nature, 326: 655-66 1 . Charison, R.J., Schwartz, S.E., Hales, lM., Cess, R.D., Coakley, J.A., Hansen, J.E., et al. ( l 992b). C limate forcing by anthropogenic aerosols. Science, 255: 423-429. Checkland, P. ( 1 98 1 ). Systems thinking, systems practice. New York: John Wiley and Sons. Christaller, W. ( 1 966). Central Places in Southern Germany (C.W. Baskin, Trans.). Englewood Cliffs, NJ: Prentice-Rall . (Original work published 1 933) Christensen, P.P. ( 1 989). Historical Roots for Ecological Economics - Biophysical Versus Allocative Approaches. Ecological Economics, 1 : 1 7-36. Christie, T., Brathwaite, B. & Thompson, B. ( 1 993) . Mineral Commodity Report 1 - Aluminium. New Zealand Mining, 12: 20-23 . Christie, T., Douch, C., Winfield, B., & Thompson, B. (2000). Industrial Minerals in New Zealand. New Zealand Mining, 27: 1 5-25 . Christie, T., Thompson, B . , & Brathwaite, B. (2000). Mineral Commodity Report 20 - Clays. New Zealand Mining, 27: 26-43 . Christie, T., Thompson, B . , & Brathwaite, B. (200 1 a). Mineral Commodity Report 2 1 - Limestone, marble and dolomite. New Zealand Mining, 29: 6-25 . Christie, T., Thompson, B . , & Brathwaite, B. (200 1 b). M ineral Commodity Report 22 - Aggregate. New Zealand Mining, 30: 6-26. 403 Ciriacy-Wantrup, S.V. ( 1 952). Resource Conservation: Economics and Policies. Berkeley : University of California Press. Clark, C.W. ( 1 976). Mathematical Bioeconomics. New York: Wiley. Clark, C.W., Munro, G.R. & Charles, A.T. ( 1 985). F isheries, Dynamics and Uncertainty. In H. Scott (Ed.), Progress in Natural Resource Economics. Oxford: Clarendon Press. Clark, W.C. ( 1 986). The cumulative impacts of human activities on the atmosphere. In Cumulative environment effects: A binational perspective. Proceedings of a Workshop Sponsored by the Canadian Environmental Assessment Research Council and the United States National Research Council . Hull, Quebec. Clayton, A.M.H. & Radcl iffe, NJ. ( 1 997). Sustainability: A Systems Approach. London: Earthscan Publications Limited. C lements, F .E. ( 1 9 1 6) . Plant Succession: An Analysis of the Development of Vegetation. Washington: Carnegie Institute Publishers. C lose, A. & Foran, B. ( 1 998). Canberra 's Ecological Footprint. Working Paper Series 98/ 1 2. Resource Futures Program. Canberra: CSIRO W ildlife and Ecology. Cobb, C.W. & Douglas, P.H. ( 1 928). Theory of Production. American Economic Review, 18( 1 ): 1 39-1 65 . Cocklin, C . , Harte, M. , & Lonergan, S . ( 1 989). Patterns of change in the use of energy in the New Zealand economy. Environment and Planning A, 21 : 1 1 4 1 - 1 1 56. Colby, M.E. ( 1 99 1 ) . Environmental Management in Development: The Evolution of Paradigms. Ecological Economics, 3: 1 93 -2 1 3 . Cole, A.O. (200 1 ). Air Emissions Inventory for the Waikato Region. Unpublished Research Report. Palmerston North: Landcare Research. Cole, H.S.D., Freeman, c., Jahoch, M., & Pavitt, K.L.R. (Eds.) . ( 1 973). Thinking about the Future: A Critique of the Limits to Growth. London: Sussex University Press. 404 Common, M. ( 1 988). Environmental and Resource Economics: An Introduction. London and New York: Longman. Common, M. ( 1 995). Sustainability and Policy: Limits to Economics. Cambridge: Cambridge University Press. Common, M. & Perrings, C. ( 1 992). Towards an Ecological Economics of Sustainability. Ecological Economics, 6: 7-34. Commoner, B. ( 1 972). The Closing Circle: Man, Nature and Technology. New York: Knopf. Commoner, B. ( 1 976). The Poverty 0/ Power. New York: Bantam Books. Conrad, J.M. & Clark, C.W. ( 1 987). Natural Resource Economics: Notes and Problems. Cambridge: Cambridge University Press. Contant, C.K. & L.L. Wiggins. ( 1 99 1 ). Defining and Analyzing Cumulative Environmental Impacts. Environmental Impact Assessment Review, 1 1 :297-309. Converse, A.O. ( 1 97 1 ). On the Extension of Input-Output Analysis to Account for Environmental Externalities. American Economic Review, 61 : 1 97- 1 98 . Converse, A.O. ( 1 996). Letter to the Editor. On complete recycling. Ecological Economics, 1 9: 1 93 - 1 94. Coombs, H.c. ( 1 990). The Return o/Scarcity: Strategies/or an Economic Future. New York: Cambridge University Press. Cooper, A.B. & Thomsen, C.E. ( 1 988). Nitrogen and phosphorus in streamwaters from adjacent pasture, pine, and native forest catchments. New Zealand Journal 0/ Marine and Freshwater Research, 22: 279-29 1 . Coming, P.A. (2002). Thermoeconomics: Beyond the Second Law. Journal 0/ Bioeconomics, 4: 57-88. Costanza, R. ( 1 980). Embodied Energy and Economic Valuation. Science, 210: 1 2 1 9- 1 224. 405 Costanza, R. (2000). The Dynamics of the Ecological Footprint Concept. Ecological Economics, 32: 34 1 -345 . Costanza, R, d' Arge, R , de Groot, R, Farber, S . , Grasso, M. , Hannon, B . , et al. ( 1 997). The Value of the World's Ecosystem Service and Natural Capital. Nature, 387: 253-260. Costanza, R & Hannon, B. ( 1 989). Dealing with the mixed units problem in ecosystem network analysis. In F. Wu1ff, J.G. Field, & K.H. Mann (Eds.), Network Analysis in Marine Ecology: Methods and Applications (pp. 90-1 1 5) . Springer, Berlin. Costanza, R & Neill, C . ( 1 98 1 ). The energy embodied in products of the biosphere. In W.J. Mitsch, R.W. Boserman, J.M. Klopatek, (Eds.), Energy and ecological modeling (pp . 745-755) . Amsterdam: Elsevier. Costanza, R & Neill, C. ( 1 984). Energy Initiatives, Interdependence and Value in Ecological Systems: A Linear Programming Approach. Journal o/Theoretical Biology, 1 00: 4 1 -57 . Costanza, R . , Neill, C. , Leibowitz, S.G., Fruci, J . , Bahr, L .M. & Day, J .W. ( 1 983). Ecological Models at the Mississippi Delta Plain Region: Data Collection and Presentation. Washington DC: F ish and Wildlife Service, United States Department of Interior. Costanza, R. & Patten, B.C. ( 1 995) . Defining and Predicting Sustainability. Ecological Economics, 15: 1 93 - 196. Costanza, R. & Voinov, A. (2004) . Landscape Simulation Modeling: A Spatially Explicit, Dynamic Approach. New York: Springer. Cumberland, J.H. ( 1 966). A Regional Interindustry Model for Analysis of Development Objectives. Papers, Regional Science Association, 1 7: 65-94. Cumberland, lH. & Korbach, RJ. ( 1 973) . A Regional Interindustry Environmental Model . The Regional Science Association Papers, 30: 1 6 1 - 1 75 . Czamanski, S . & Malizia, E. ( 1 969). Applicability and limitations i n the use of national input­ output tables for regional studies. Papers and Proceedings, Regional Science Association 23: 65-77. 406 d' Arge, R.c. & Kogiku, K.C. ( 1 973) . Economic Growth and the Environment. Review of Economic Studies, 59: 6 1 -77. Daily, G. ( 1 997). Nature 's Services: Societal Dependence on Natural Ecosystems. Washington DC: Island Press. Daily, G.c. & Ehrlich, P.R. ( 1 992). Population, Sustainability and the Earth's Carrying Capacity. Bioscience, 42: 76 1 -77 1 . Daly, H.E. ( 1 968). On Economics as a Life Science. Journal of Political Economics, 76: 392- 406. Daly, H.E. ( 1 973). The Steady State Economy: Toward a Political Economy of Biophysical Equil ibrium and Moral Growth. In H.E. Daly (Ed.), Toward a Steady State Economy. San Francisco: W.H. Freeman. Daly, H.E. ( 1 977). Steady-State Economics. San Francisco: W.H. Freeman. Daly, H.E. ( 1 987). The Economic Growth Debate: What Some Economists May Have Learned But Many Have Not. Journal of Environmental Economics and Management, 14: 323-336. Daly, H.E. ( 1 99 1 a) . Steady-State Economics: Second Edition with New Essays. Washington, D.C. : Island Press. Daly, H.E. ( l 99 1 b). Elements of Environmental Macroeconomics. In R. Costanza (Ed.), Ecological Economics: The Science and Management of Sustainability (pp. 32-46). New York: Columbia University Press. Daly, H.E. ( 1 992). Allocation, Distribution and Scale: Towards and Economics that is Efficient, Just and Sustainable. Ecological Economics, 6: 1 85 - 1 94. Daly, H.E. ( 1 994). Operationalizing sustainable development by investing in natural capital. In A. Jansson, M. Hammer, C. Folke, & R. Costanza (Eds.), Investing in natural capital: the ecological economics approach to sustainability. Washington D.C. : Is land Press. Daly, H.E. ( 1 995) . On Nicholas Georgescu-Roegen's contribution to economics: An obituary essay. Ecological Economics, 13 : 1 49- 1 54. 407 Daly, H. & Cobb, 1. ( 1 989). For the Common Good. Boston: Beacon Press. Dang, H.D.T. ( 1 999). Energy Data File. Wellington: M inistry of Commerce . Dasgupta, P.S . ( 1 982). The Control of Resources . Oxford: Basil B lackwell. Dasgupta, P.S. & Heal, G.M. ( 1979). Economic Theory and Exhaustible Resources. Cambridge: Cambridge University Press. Dasgupta, P .S . & Stiglitz, I .E. ( 198 1 ). Resource Depletion Under Technological Uncertainty. Econometrica, 49: 85- 1 04. Dasmen, R.F. ( 1 972). Environmental Conservation. New York: Wiley. de Groot, R.S. ( 1 987). Environmental Functions as a Unifying Concept for Ecology and Economics. Environmentalist, 7: 1 05- 1 09. de Groot, R.S . ( 1 992). Functions of Nature: Evaluation of Nature in Environmental Planning, Management and Decision Making. Netherlands: Wolfers-Noordhoff. de Groot, R.S. , Wilson, M.A. & Boumans, R.MJ. (2002). A Typology for the Classification, Description and Valuation of Ecosystem Functions, Goods and Services. Ecological Economics, 41 : 393-408. DeAngel is, D.L. & Waterhouse, J.C. ( 1 987). Equilibrium and Non Equilibrium Concepts In Ecological Models. Ecological Monographs, 57: 1 -2 1 . Deaton, M.L. & Winebrake, JJ . (2000). Dynamic Modeling of Environmental Systems. New York: Springer. Decker, E.H, El liot, S., & Smith, F.A. (2002) . Megacities and the environment. The Scientific World Journal, 2: 274-386. Deloitte & Touche Consulting Group. ( 1 997). 1 997 Top New Zealand Companies. Management, December: 68-9 1 . 408 den Elzen, M. , Beusen, A., & Rotmans, J . ( 1 995). Modelling Global Biogeochemical Cycles: An Integrated Assessment Approach. Global Dynamics and Sustainable Development Programme, GLOBO Report Series No. 7. Bilthoven: National Institute of Public Health and the Environment (RlVM) Denbigh, K.G. & Denbigh, J .S . ( 1 985). Entropy in Relation to Incomplete Knowledge. Oxford: E lsevier. Denton, R.V. ( 1 975). The energy cost of goods and servIces In the Federal Republic of Germany. Energy Policy, 3(4): 1 79-284. Department of Statistics. ( 1 98 1 ) . Census of Manufacturing. Wellington: Department of Statistics. Department of Statistics. ( 1 987). Demographic Trends 1987. Wellington: Department of Statistics. Deparment of Statistics. ( 1 989). Demographic Trends 1989. Wellington: Department of Statistics. Department of Statistics. ( 1 993). Agriculture Statistics 1993. Wellington: Department of Statistics. Department of Statistics. ( 1 994). Agriculture Statistics 1994. Wellington: Department of Statistics. DeSimone, L.D. & Popoff, F. (2000). Eco-ef iciency: the business link to sustainable development. Cambridge, MA: MIT Press Dialogue Consultants Ltd. ( 1 992). Indicators of Sustainable Development. Report Prepared for the M inistry for the Environment. Auckland: Dialogue Consultants Ltd. Dietzenbacher, E. and Lahr, M.L. (200 1 ) . Introduction. In M.L. Lahr & E. Dietzenbacher. (Eds) Input-Output Analysis: Frontiers and Extensions (pp. 1 -3 1 ). London: Palgrave. Dixit, A.K. & Stiglitz, J.E. ( 1 977). Monopol istic Competition and Optimum Product Diversity. American Economic Review, 67: 297-308. 409 Domar, E. ( 1 946). Capital expansion, rate of growth, and employment. Econometrica, 14: 1 37- 147. Dorfman, R. , Samuelson, P . , & So low, R. ( 1 95 8) . Linear Programming and Economic Analysis. New York: McGraw-Hill. Duchin, F. & Szyld, D.B. ( 1 985) . A Dynamic Input-Output w ith Assured Positive Output, Metroeconomica, 27, pp.269-282. Duncan, S. ( 1 977). Mental Maps of New York. New York, December 1 9, pp. 5 1 -62. Duniap, R.E. ( 1 997). The Evolution of Environmental Sociology : A Brief History and Assessment of the American Experience. In M. Redclift & G. Woodgate (Eds.), The International Handbook of Environmental Sociology (pp. 2 1 -39). Cheltenham: Edward Elgar. Dunlap, R.E. (2002). Paradigms, Theories, and Environmental Sociology. In R.E . Dunlap, F .H. Buttel, P. Dickens, & A. Gijswijt (Eds.), Sociological Theory and the Environment: Classical Foundations, Contemporary Insights (pp.329-3 50). Lanham: Rowman & Littlefield Publishers Inc. Dunlap, R.E., Michelson, W., & Stalker, G. (2002). Environmental Sociology: An Introduction. In R.E. Dunlap & W. Michelson (Eds.), Handbook of Environmental Sociology (pp. 1 -3 2). Westport, CT: Greenwood Press. Durkheim, E. ( 1 964). The Division of Labour in Society. New York: Free Press. (Original work published in 1 893) Eding, G.J . & Oosterhaven, J. ( 1 996). Towards a new, rectangular approach in the construction of regional input-output tables. Paper presented to the British, Irish and North American Sections of the Regional Science Association International, Edinburgh/Washington D.e. Eding, G.J. , Oosterhaven, J. , de Vet, B. , & Nijmeijer, H. ( 1 998). Constructing Regional Supply and Use Tables: Dutch Experiences. Paper presented at the 1 2th International Conference on Input-Output Techniques, New York, 1 8-22 May 1 998 . Edmeades, D. C . (Ed). (200 1 ). Fertiliser Handbook. Kapuni : Ballance Agri-Nutrients. 4 1 0 EEA (European Environment Agency) . ( 1999). Environment in the European Union at the Turn o/the Century. Copenhagen: European Environment Agency. Ehrlich, P. ( 1 968). Population Bomb. New York: Ballantine. Ehrlich, P. ( 1 994). Ecological Economics and the Carrying Capacity of the Earth. In A. Jansson, M. Hammer, F. Berkes, & R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington D.e. : Island Press. Ehrlich, P. & Ehrlich, A. ( 1 9 8 1 ). Extinction: The Causes and Consequences 0/ the Disappearance o/Species. New York: Random House. Ehrlich, P. & Ehrlich, A. ( 1 990). The Population Explosion. New York: S imon and Schuster. Ehrlich, P. & Ehrlich, A. ( 1 992). The Value of Biodiversity. Ambio, 21 : 2 1 9-226. Ehrlich, P.R., Ehrlich, A.H. & Holdren J.P. ( 1 973) . Human Ecology: Problems and Solutions. San Francisco: U.H. Freeman. Ehrlich, P.R., Ehrlich, A.H. & Holdren, J .P. ( 1 977). Ecoscience: Population, Resources, Environment. San Francisco: W.H. Freeman. Ekins, P. ( 1 994) . The Environmental Sustainabi l ity of Economic Processes: A Framework for Analysis. In J.C.J.M. van den Bergh & J. van der Straaten (Eds.), Toward Sustainable Development: Concepts, Methods and Policy (pp.25-55) . Washington DC: Island Press. Ekins, P. (2000). Economic Growth and Environmental Sustainability. London and New York: Routledge. Ekins, P. (2003) . IdentifYing Critical Natural Capital : Conclusions about Critical Natural Capital. Ecological Economics, 44(2-3): 277-292. Eltis, W. ( 1 984) . The Classical Theory 0/ Economic Growth. London, Macmillan. Emerson, MJ. and Hackmann, D.G. ( 1 97 1 ). 1969 Kansas input-output study. Report No. 33 . Manhattan, KS: Kansas State University, Department of Economics. 4 1 1 Energy Efficiency Conservation Authority (EECA). ( 1 996). Energy End-Use Database 1 994- 95. Wellington: Energy Efficiency Conservation Authority. Energy Efficiency Conservation Authority (EECA). ( 1 997). Energy Use Database: Handbook. Wellington : Energy Efficiency and Conservation Authority . Energy Efficiency Conservation Authority (EECA). (2004). Energy End-Use Database 2001- 02 . Wellington: Energy Efficiency Conservation Authority. Environment Protection Authority (EPA). ( 1 997). Auckland Air Emissions Inventory. Melbourne: Environment Protection Authority . Environment Waikato. ( 1 998). State of the Environment Report. Hamilton: Environment Waikato. Eskelinen, H & Soursa, M. ( 1 980). A note on estimating interindustry flows. Journal of Regional Science, 20: 26 1 -266. Ethier, W.J. ( 1 982). National and International Returns to Scale in the Modern Theory of International Trade. American Economic Review, 72: 389-405 . Evans, F.e. ( 1 956). Ecosystem as the basic unit i n ecology. Science, 123: 1 1 27-28. Faber, M. ( 1985). A Biophysical Approach to the Economy, Entropy, Environment and Resources. In W. van Gool & J.J.c. Bruggink (Eds.), Energy and Time in Economic and Physical Sciences. Amsterdam: North-Holland. Faber, M., Niemes, H.H., & Stephan, G. ( 1 987). Entropy, Environment and Resources. Berlin: Springer -V er lag. Faber, M., Proops, J., Ruth, M., & Michael is, P. ( 1990). Economy-environment interactions in the long-run : A neo-Austrian approach. Ecological Economics, 2 : 27-55 . Faucheux, S . & O 'Connor, M. ( 1 998). Valuation for Sustainable Development: Methods and Policy Indicators. Cheltenham, United Kingdom: Edward Elgar. 4 1 2 Faustmann, G. ( 1 968). On the determination of the value which forest land and immature stands possess for forestry. Oxford Institute Papers, 42. Femg, J.J. (200 1 ). Using Composition of Land Multiplier to Estimate Ecological Footprints Associated with Production Activity. Ecological Economics, 37(2): 1 59- 1 72. Finn, J. ( 1 976). Measure of Ecosystem Structured Function Derived from the Analysis of F lows. Journal of Theoretical Biology, 56: 363-380. Fischer, A.c. ( 1 98 1 ). Resource and environmental economics. London: Cambridge University Press. Fischer, C.S. ( 1 975). Toward a Subcultural Theory of Urbanism. American Journal of Sociology, 80: 1 3 1 9- 1 341 . Fischer, C., Jackson, R.M., Stueve, C.A., Gerson, K. & Jones, L.M. ( 1 977). Networks and Places. New York: Free Press. Fischer-Kowalski, M. (2002). Exploring the history of industrial metabolism. In R.U. Ayres & L. W. Ayres (Eds.), A Handbook of Industrial Ecology (pp. 1 6-26). Cheltenham: Edward E lgar. Fisher, A.C. ( 1 98 1 ) . Resource and Environmental Economics. Cambridge: Cambridge Univers ity Press. Fisher, F. ( 1 983). Disequilibrium Foundations of Equilibrium Economics. Cambridge: Cambridge University Press. Fiske, S .T. & Taylor, S.E. ( 1 984). Social Cognition. New York: Random House. Flanagan, W.G. ( 1 993). Contemporary Urban Sociology. Cambridge: Cambridge University Press. F lood, R.L. & Carson, E.R. ( 1 988). Dealing with Complexity: An Introduction to the Theory and Application of Systems Science. New York: Plenum Press. Foley, V. ( 1 973) . An Origin of the Tableau Oeconomique. History of Political Economy, 5(2): 1 2 1 - 1 30 . 4 1 3 Folke, C . ( 1 99 1 ). Socio-Economic Dependence on the Life-Supporting Environment. In C. Folke & 1. Kaberger (Eds.), Linking the Natural Environment and the Economy: Essays from the Eco Eco Group. Dordrecht: Kluwer Academic. Folke, c., Hammer, N., & Jansson, AM. ( 1 989). Life-Support Value of Ecosystems: A Case Study of the Baltic Sea Region. Ecological Economics, 3: 1 23 - 137 . Folke, C . , lansson, A., Larsson, J . , & Costanza, R . ( 1 997). Ecosystem Appropriation by Cities. Ambio, 26: 1 67 - 1 72. Folke, c., Larsson, l., & Sweitzer, l . ( 1 994) . Renewable resource appropriation by cities. Paper presented at "Down To Earth: Practical Applications of Ecological Economics," Third International Meeting of the International Society for Ecological Economics, San Jose, Costa Rica (24-28 October 1 994). Fonterra Co-operative Group. (n.d.). Retrieved from http://www.fonterra.comJdefault.jsp Food and Agriculture Organisation of the United Nations (n.d.) Retrieved from http ://apps.fao.org Ford, A. ( 1 999). Modeling the Environment. An Introduction to System Dynamics Models of Environmental Systems. Washington, DC: Island Press. Forrester, l .W. ( 1 96 1 ). Industrial Dynamics. Portland, Oregon: Productivity Press. Forrester, l .W. ( 1 968). Principles of Systems. Walthan, MA: Pegasus Communications. Forrester, l .W. ( 1 969). Urban Dynamics. Cambridge: MIT Press. Forrester, J .W. ( 1 97 1 ). World dynamics . Massachusetts : Wright-Al len Press, Inc. Foster,J. & Metcalfe, J.S. (Eds.). (200 1 ). Frontiers of evolutionary economics: competition, self­ organization, and innovation policy. Cheltenham, UK; Northampton, MA: Edward Elgar Publishing. 4 1 4 Frank, 1 . & Babunovic, M . ( 1 984). An Investment Model o f Natural Resource Markets. Economica, 51: 83-96. Freney, J.R., Ivanov, M.W., & Rodhe, H. ( 1 983). The sulfur cycle. In B. Bolin & C.B. Cook (Eds.), The major biogeochemical cycles and their interactions. SCOPE 2 l . New York: John Wiley & Sons. Frey, W.H. & Zimmer, Z. (200 1 ) . Defining the City. In R. Paddison (Ed.), Handbook of Urban Studies (pp. 1 4-35). London: SAGE Publications. Galloway, 1. N. ( 1 985) . The deposition of sulfur and nitrogen from the remote atmosphere: background paper. Chapter 8. In J .N. Galloway, R.J. Charison, A.O. Andreae & H. Rodhe (Eds.). The Biogeochemical Cycling ofSulfur and Nitrogen in the Remote Atmosphere (pp. 1 43- 72). Boston: Reidel. Gans, H. ( 1 962). The Urban Villagers. New York: Free Press. Gans, H. ( 1 982). The Urban Villagers: Group and Class in the Life of Italian Americans. New York: Free Press. Garrels, R.M., Mackenzie, F.T. & Hunt, C. ( 1 975) . Chemical Cycles and the Global Environment, Assessing Human Influences (3rd ed.). Los Altos, CA: W. Kaufmann, Inc . Gasson, B. (2002). Towards Ecologically Sustainable Cities: A Conceptual Framework and a Case Study. Paper presented at the IERM Congress 2002 held in Midrand, South Africa, 1 8th - 2 1 st November. Retrieved 1 4 July 2005 from www. ierm.org.zaJlibrary/congress2002/towards%20eco%20sustainable%20cities.pdf Geddes, P. ( 1 885). An Analysis of the Principles of Economics. Proceedings of the Royal Society of Edinburgh. London: Williams and Norgate. Georgescu-Roegen, N. ( 1 9 7 1 ). The Entropy Law and the Economic Process. Cambridge, Massachusetts: Harvard University Press. Georgescu-Roegen, N. ( 1 976). Energy and Economy Myths: Institutional and Analytical Economic Essays. New York: Pergamon Press. 4 1 5 Georgescu-Roegen, N. ( 1 977a). The Steady State and Ecological Salvation. Bioscience, 13: 268. Georgescu-Roegen, N. (1 977b). Bioeconomics : a new look at the nature of economic activity. In Junker, L. (Ed.), The Political Economy of Food and Energy (pp. 1 05-134). Ann Arbor, MI: The University of Michigan Press. Georgescu-Roegen, N. ( 1 977c). Inequality, limits and growth from a bioeconomic viewpoint. Review of Social Economy, 35:36 1-375 . Georgescu-Roegen, N. ( 1 979a). Comments on the Papers by Daly and Stiglitz. In V.K. Smith (Ed.), Scarcity and Growth Reconsidered. RFF, Baltimore and London: Johns Hopkins Press. Georgescu-Roegen, N. ( 1 979b). Energy analysis and economic valuation. Southern Economic Journal, 45: 1 023-1 058. Georgescu-Roegen, N. ( 1 984). Feasible Recipes versus V iable Technologies. Atlantic Economic Journal, 12: 20-3 1 . Giampietro, M. & Pimentel, D. ( 1 99 1 ) . Energy Analysis Models to Study the Biophysical L imits for Human Exploitation of Natural Processes. In C. Rossi & E. Tiezzi (Eds.), Ecological Physical Chemistry (pp. 1 39- 1 84). Amsterdam: Elsevier. Gi l l iland, M.W. ( 1975). Energy Analysis and Public Policy. Science, 189: 1 05 1 - 1 056. Gi ll i land, M.W. ( 1 977). Energy Analysis: A Tool for Evaluating the Impact of End Use Management Strategies on Economic Growth. In R.A. Fazzolare & C.B. Smith (Eds.), Energy Management: Proceedings of the International Conference (pp.363-6 1 9) . New York: Pergamon Press. G irardet, H. ( 1992). The Gaia Atlas of Cities. London: Gaia Books. Girardet, H. ( 1 996). The Gaia A tlas of Cities: New Directions for Sustainable Urban Living. London: Gaia Books Limited. Glansdorff, P. & Prigogine, I . ( 1 97 1 ) . Thermodynamic Theory of Structure, Stability and Fluctuations . New York: Wiley . 4 1 6 Gleick, P. H. (Ed.). ( 1 993). Water in Crisis: A Guide to the World's Freshwater Resources. Oxford: Oxford University Press. Goeller, H. & Weinberg, A. ( 1 976). The Age of Substitutabi l ity. Science, 191 : 1 90 1 - 1 909. Goh, K.L. & Fairgray, D. ( l 999a). International Tourism Forecasts to New Zealand 1999-2005. Auckland: McDermott Fairgray Group Ltd. Goh, K.L. & Fairgray, D. ( 1 999b). New Zealand Domestic Tourism Forecasts . Auckland: McDermott Fairgray Group Ltd. Goldsmith, E., Allen, R., Allaby, M. , Davoll, J., & Lawrence, S. ( 1972). Blueprint for Survival. Harmondsworth: Penquin. Goodland, R. & Daly, H.E. ( 1 993) . Why Northern Income Growth is not the Solution to Southern Poverty. Ecological Economics, 8: 85- 1 0 l . Goodstein, E. ( 1 995). Economics and the Environment. Englewood Cliffs, New Jersey: Prentice Hall. Gordon, H.S. ( 1 954). Economic theory of common property resources. Journal of Political Economy, 62: 1 24- 1 42. Gottdiener, M . ( 1 994). The New Urban Sociology. New York: McGraw-Hil l . Gottdiener, M. & Feagin, J . ( 1 988). The Paradigm Shift in Urban Ecology. Urban Affairs Quarterly, 24: 1 63- 1 87. Gowdy, J .M. ( 1 984). Marx and Resource Scarcity: An Institutional Approach. Journal of Economic Issues, 18: 1 45- 1 56 . Graedel, T.E. ( 1 994). Industrial Ecology: Definition and Implementation. In R.H. Socolow, J.A. Clinton, F.G. Berkhout, & V.M. Thomas (Eds.), Industrial Ecology and Global Change (pp. 23-42). Cambridge, UK: Cambridge University Press. 4 1 7 Graede1, T.E. & Allenby, B.R. ( 1 995). Industrial Ecology. Englewood Cliffs, NJ: Prentice­ Hall. Graham, P. (2004, December 1 3) . Pulp mill sets production record. New Zealand Herald. Retrieved August 4, 2005, from http://www.nzherald.co.nzlindex.cfm ?c _ id=3 &Obj ectID=9002 862. Gravgard, O. ( 1 998). Physical Input-output Tables for Denmark, 1 990. Extract of a forthcoming report on physical input-output tables and emissions accounts for Denmark, 1 990. Copenhagen: Statistics Denmark. Greer, S. ( 1 962). The Emerging City. New York: Free Press. Gril iches, Z. ( 1 992). The Search for R&D Spillovers. Scandinavian Journal of Economics, 94: 29-48. Grossman, G.M. & Helpman, E. ( 1 99 1 ) . Innovation and Growth in the Global Economy. Cambridge, MA: MIT Press. Grossman, G.M. & Helpman, E. ( 1 994). Endogenous Innovation in the Theory of Growth. Journal of Economic Perspectives, 8: 23-44. Grubb, M., Chapuis, T., & Duong, M.H. ( 1 995). The Economics of Changing Course. Implications of Adaptability and Inertia for Optimal Climate Policy. Energy Policy, 23: 4 1 7- 43 1 . Hall, C.A., Cleveland, C.J. & Kaufmann, R. ( 1 986). Energy and resource quality: the ecology of the economic process. New York: Wiley. Hamilton, K. ( 1 994). Green Adj ustments to GDP. Resources Policy, 20(3): 1 55 - 1 68. Hannon, B.M. ( 1973a). An Energy Standard of Value. Annals of the American Academy of Political and Social Science, 410: 1 39-1 53 . Hannon, B.M. ( l 973b). The Structure of Ecosystems. Journal of Theoretical Biology, 41 : S3 S- 546. 4 1 8 Hannon, B.M. ( 1 979). Total Energy Costs in Ecosystems. Journal of Theoretical Biology, 80: 27 1 -293 . Hannon, B.M. ( 1 982). Analysis of the Energy Cost of Economic Activities: 1 963 to 2000. Energy Systems and Policy Journal, 6(3) : 249-278. Hannon, B. ( 1 99 1 ). Accounting in Ecological Systems. In R. Costanza (Ed.), Ecological Economics: The Science and Management of Sustainability (pp. 234-252). New York: Columbia University Press. Hannon, B. , B lazeck, T., Kennedy, D., & Il lyes, R. ( 1984). A comparison of energy intensities: 1 963, 1 967 and 1 972. Resources and Energy, 5: 83- 1 02. Hannon, B . and Ruth, M. ( 1994). Dynamic Modeling. New York: Springer-Verlag. Hannon, B . & Ruth, M. ( 1 997). Modeling Dynamic Biological Systems. New York: Springer. Hannon, B. & Ruth, M. (200 1 ) . Dynamic Modeling. New York: Springer. Hansen, J., Johnson, D., Lacis, A., Lebedeff, S . , Lee, P., Rind, D., et al. ( 1 98 1 ). Climate impact of increasing atmospheric carbon dioxide. Science, 213: 957-966. Hardin, G. ( 1 968). Tragedy of the Commons. Science, 1 62 : 1 243-1 248. Harris, C.D. & Ullman, E.L. ( 1 945). The Nature of Cities. Annals of the American Academy of Political and Social Science, 242: 7- 1 7 . Harrison, M . ( 1975). Entropy Concepts i n Physics. In L . Kubat & J . Zeman (Eds.), Entropy and Information in Science and Philosophy. Amsterdam: Elsevier. Harrod, R.F. ( 1 939). An Essay on Dynamic Theory. Economic Journal, 49: 1 4-33 . Harrod, R.F . ( 1 948). Fundamental Dynamic Theorems. In Towards a Dynamic Economics: Some Recent Developments of Economic Theory and their Application to Policy, Lecture 3 (pp .63- 1 00). London: Macmi l lan. 4 1 9 Hartwick, J.M. ( 1 977). Intergenerational Equity and the Investing of Rents from Exhaustible Resources. American Economic Review, 67(5): 972-974. Hartwick, J .M. ( 1 978). Investing Returns From Depleting Renewable Resource Stocks and Intergenerational Equity. Economic Letters, 1: 85-88. Hartwick, J.M. and Olewiler, N.D. ( 1 986). The Economics of Natural Resource Use. New York: Harper and Row. Harvey, D. ( 1 978). The urban process under capitalism. International Journal of Urban and Regional Research, 2( 1 ) : 1 0 1 -3 1 . Harvey, D . ( 1 985) . The Urbanisation of Capital. Oxford: B lackwell. Hawley, A. & Duncan, O.D. ( 1 957) . Social Area Analysis: A Critical Appraisal. Signs: Journal of Women in Culture and Society, 4: 274-290. Heal, G. & Barrow, M. ( 1980). The Relationship Between Interest Rates and Metal Price Movements. Review of Economic Studies, 47: 1 6 1 - 1 8 1 . Helpman, E . (2004). The Mystery of Economic Growth. Cambridge, Massachusetts: The Belknap Press of Harvard University Press. Hendtlass, C.A., Peet, NJ . & Baines, J .T. ( 1 988). Energy Analysis of Goods and Services in New Zealand. Auckland: New Zealand Energy Research and Development Committee, Univers ity of Auckland. Herendeen, R.A. ( 1 972). The Energy Costs of Goods and Services Report No. 69. Center for Advance Computation, Urbana, I l l inois: University of I llinois. Herendeen, R.A. ( 1 98 1 ). Energy Intensities in Ecological and Economic Systems. Journal of Theoretical Biology, 91 : 607-620. Herendeen, R.A. ( 1 998). Embodied Energy, Embodied Everything .. . Now What? Advances in Energy Studies. Energy Flows in Ecology and Economy. Presented at VI European Week of Scientific Culture, 22-28 November 1 998 . Porto Venere, Italy. 420 Herfindahl, O. & Kneese, A.V. ( 1 965) . The Quality of the Environment. Baltimore: Johns Hopkins University Press. Herfindahl, O. & Kneese, A.V. ( 1 974). Economic Theory of Natural Resources. Columbus, Ohio: Charles E. Meri l l . Hewings, O.J.O. & Jensen, R.e. ( 1 986). Regional, interregional and multiregional input-output analysis. In P. Nijkamp (Ed.), Handbook of regional and urban economics: Vol. 1 . Amsterdam: North Holland. Hicks, J .R. ( 1 946). Value and Capital (2nd ed.). Oxford: Oxford University Press. Hicks, J .R. ( 1 985). Methods of Dynamic Economics. New York: Oxford University Press. Hinterberger F. and Stiller H. ( 1 998) Energy and Material F lows. In Proceedings of the 1st International Workshop Advances in Energy Studies. Energy F lows in Ecology and Economy 275-286/642. Roma, Italy : MUSIS. Hirata, H. & Ulanowicz, R.E. ( 1 985). Information Theoretical Analysis of the Aggregation and Hierarchical Structure of Ecological Networks . Journal of Theoretical Biology, 166: 32 1 -34 1 . Hirsch, W.Z. ( 1 959). Interindustry re lations of a metropolitan area. Review of Economics and Statistics, 41 : 360-369. Hite, J .e. & Laurent, E.A. ( 1 97 1 ). Empirical study of economic-ecologic l inkages in a coastal area. Water Resources Research, 7(5) : 1 070- 1 078 . Hite, J .C. & Laurent, E .A. ( 1 972). Environmental Planning: An Economic Analysis. New York: Praeger. Hodgson, G.M. ( 1 993). Economics and Evolution: Bringing Life Back Into Economics. Cambridge: Polity Press. Hogan, O. & Wil liamson, B. ( 1 999). New Zealand is Different: Chemical Milestones in New Zealand History. Christchurch : C lerestory Press. 42 1 Holdren, J.P. & Ehrlich, P.R. ( 1 974). Human Population and the G lobal Environment. American Scientist, 62: 282-292. Holdren, J.P., Ehrlich, P.R., & Daily, G.e. ( 1 993). Physical and B iological Sustainabil ity. Population and Environment, 18(3): 23 1 -255 . Hollander, S. ( 1 973). The Economics of Adam Smith. Toronto: University of Toronto Press. Hollander, S. ( 1 985). The Economics of John Stuart Mill. Toronto: University of Toronto Press. Holling, e .S . ( 1 973). Resilience and Stability of Ecological Systems. Annual Review of Ecological Systems, 4: 1 -23 . Holling, C.S. ( 1 986). Resi lience of Terrestrial Ecosystems: Local Surprise and Global Change. In Clark, W.e. & Munn, R.E. (Eds.), Sustainable Development of the Biosphere. Cambridge : Cambridge University Press. HoIling, e.S . ( 1 995). Sustainability: The Cross Scale Dimension. In M. Munasinghe & W. Shearer (Eds.), Defining and Measuring Sustainability: The Biogeophysical Foundations. Washington DC : W orId Bank. Holmen, K. ( 1 992). The Global Carbon Cycle. In S .S . Butcher, R.J. Charlson, G.H. Orians, & G.Y. Wolfe (Eds.), Global Biogeochemical Cycles (pp. 239-262). London: Academic Press. Hotel l ing, H. ( 1 93 1 ). The Economics of Exhaustible Resources. Journal of Political Economy, 39: 1 37-1 75 . Howard, E . ( 1 965). Garden Cities of To-morrow. Cambridge, MA: MIT Press. HoweIl, J.I. ( 1 973). Hard Living on Clay Street. Garden C ity , NY: Anchor. Hoyt, H. ( 1 939). The Structure and Growth of Residential Neighborhoods in American Cities. Washington, DC: Federal Housing Administration. Hua�g, S-L. ( 1 998). Spatial Hierarchy of Urban Energetic System. In Advances in Energy Studz s: Energy flows in ecology and economy (pp.499-5 1 4) . Rome: Museum of Science and Scie tific Information. 422 Hubbard, L.J. & Brown W.A.N. ( 1 9 8 1 ) . Multipliers Jrom Regional Non-Survey Input- Output Tables Jor New Zealand. Research Report No. 1 1 7 . Canterbury: Agricultural Economics Research Unit, Lincoln College. Hudson, J. ( 1 969). A location theory for rural settlement. Annals oJ the Association oJ Am erican Geographers, 59: 3 65 -3 8 l . Hueting, R. ( 1 980). Ne w Scarcity and Economic Growth: More Welfare Through Less Production? Amsterdam: orth-Hol land Publishing Company. Hueting, R., Bosch, P. & de Boer, B. ( 1 992). Methodology Jor the Calculation oJ Sustainable National Income. Statistical Essay M44. Voorburg/Heerlen: Netherlands Central Bureau of Statistics. Hughes, M . K. ( 1 974). The urban ecosystem. Biologist, 21(3): 1 1 7- 1 2 7. Humphrey, C.R. & Buttel, F. ( 1 982). Environment, Energy and Society. Belmont: Wadsworth. Husar, R.B . ( 1 994). Ecosystem and the biosphere: metaphors for human-induced material flows. In R.U. Ayres & U. Si monis (Eds .), Industrial Metabolism: Restructuring Jor Sustainable Development. Tokyo: United ations University Press. IFIAS (International Federation of Institutes for Advanced Study). ( 1 974). Energy Analysis Workshop on Methodology and Conventions. 25-30 August 1 974. Stockholm: IFIAS. Irwin, J . ( 1 977) . Scenes. Beverly Hil ls, CA: Sage. Isard, W. ( 1 95 1 ) . Interregional and regional input-output analys is: a model of a space economy. Review oJ Econom ics and Statistics, 33: 3 1 9-3 2 8 . Isard, W. ( 1 968). Some Notes o n the Linkage of Ecological and Economic Systems. Papers, Regional Science Association, 22: 8 5 -96. Isard, W. ( 1 972). Ecologic and Economic Analysis Jor Regional Development. New York: Free Press. 423 Isard, W. ( 1 975) . Introduction to Regional Science. New Jersey : Prentice Hall . Isard, W. and Langford, T . W. ( 1 969). Impact of Vietnam War expenditures on the Philadelphia economy: Some initial experiments with the inverse of the Philadelphia input-output table. Papers and Proceedings, Regional Science Association, 23: 2 1 7-65. Isard, W. and Langford, T. W. ( 1 97 1 ). Regional input-output study: Recollections, reflections, and diverse notes on the Philadelphia experience. Cambridge: MIT Press. Isard, W., Langford, T.W. & Romanoff, E. ( 1966-68) . Philadelphia region input-output study (Vols 1 -3). Philadelphia: Regional Science Research Institute Working Paper. ISO (International Standards Organisation). ( 1 997). Environmental Management - Life Cycle Asessment - Principles and Framework, ISO 1 4040. IUCNfUNEP/WWF. ( 1 980). World Conservation Strategy: Living Resource Conservation for Sustainable Development. Gland, Switzerland: IUCN, UNEP & WWF. IUCNIUNEP/WWF. ( 1 99 1 ). Caring for the Earth: A Strategy for Sustainable Living. Gland, Switzerland: IUCN, UNEP & WWF. Jackson, R. W. ( 1 998). Regionalizing National Commodity-by-Industry Accounts. Economic Systems Research, 1 0(3): 223-238 . Jacobs, J . ( 1 961) . The Death and Life of Great American Cities. New York: Random House. Jacobs, J. ( 1 970). The Economy of Cities. New York: V intage. Jacobs, J. ( 1 984). Cities and the Wealth of Nations. New York: Random House. Jacobs, M. ( 1 99 1 ). The Green Economy: Environment, Sustainable Development and the Politics o/the Future. London: P luto Press. Jaffe, D.A. ( 1992). The Nitrogen Cycle. In S .S . Butcher, RJ. Charlson, G.H. Orians & G.V. Wolfe (Eds.), Global Biogeochemical Cycles (pp. 263-282). London: Academic Press. Jaffe, W. ( 1 976). Menger, Jevons and Walras Dehomogenized. Economic Inquiry, 14: 5 1 1 -524. 424 Jahnke, R.A. ( 1 992). The Sulfur Cycle. In S .S . Butcher, RJ. Charison, G.H. Orians & G.Y. Wolfe (Eds.), Global Biogeochemical Cycles (pp. 3 0 1 -3 1 4) . London: Academic Press. Jansson, A. & Jansson, B. ( 1 994). Ecosystem Properties as a Basis for Sustainability. In A. Jansson, M. Hammer, C. Folke, & R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington D.C. : Island Press. Jansson, A. & Zucchetto, J. ( 1 978) . Man, Nature and Energy F low on the Island of Gotland. Ambio, 7(4): 1 40-1 49. Jensen, R.C. ( 1 980). The Concept of Accuracy in Input-Output. International Regional Science Review 5: 1 39-54. Jensen, R.C. ( 1 990). Construction and Use of Regional Input-Output Models: Progress and Prospects. International Regional Science Review, 13: 9-25 . Jensen, R.c . , Mandeville, T.D. & Karunaratne, N .D . ( 1 979). Regional Economic Planning. London: Croom Helm Jensen, R.c. & West, G.R. ( 1 980). The Effect of Relative Coeffic ient Size on Input-Output Multipliers. Environment and Planning A, 12 : 659-670. Jensen, R.C. & West, G.R. ( 1 988). An operational hybrid or partial survey system for the construction of regional and interregional input-output tables: The 'GRIT ' system. Paper presented at the Conference on the Construction and Use of Input-Output Models, Alpine Lake Resort, WV. Jensen, R.C., West, G.R., & Hewings, G.J.D. ( 1 988). The study of regional economic structure using input-output tables. Regional Studies, 22: 209-220. Jevons, W.S . ( 1 909). The Coal Question: An Inquiry Concerning the Progress of the Nation and the Probable Exhaustion of Our Coal-Mines . London: MacMillan. (Original work published 1 865) Johnson, A. H., & T. G. S iccama. ( 1 983) . Acid deposition and forest decl ine. Environment, Science & Technology, 1 7:294A-305A 425 Johnson, M.H., & Bennett, J.T. ( 1 98 1 ). Regional Environmental and Economic Impact Evaluation: An Input-Output Approach. Regional Science and Urban Economics, 1 1 : 2 1 5-230. Johnston, R.J. ( 1 976). Residential Area Characteristics. In D.T. Herbert & R.J. Johnston (Eds.), Social Areas in Cities: Vol. 1, Spatial Processes and Form (pp. 1 93-235). New York: Wiley. Jollands, N., Ruth, M., Golubiewski, N., and Bemier, C. (2005). Climate' s Long-term Impact on New Zealand Infrastructure (CLINZI) - A Case Study of Hamilton C ity, New Zealand . . Submitted to Journal of Environmental Management. Jollands, N.A., Golubiewski, N.E. , & McDonald, G.W. (2005). Implications of Changing Employment Patterns on Urban Ecosystem Service Requirements. The International Journal of Environment, Work place and Employment, 1 : 3 1 0-335 . Iollands, N.A., Golubiewski, N.E., & McDonald, G.W. (In press b). Linking Policy and Science: A Study of Metro Christchurch Ecosystem Service Appropriation. The International Journal of Environment, Work place and Employment. Jones, c.l . ( 1 998). Introduction to Economic Growth. New York: W.W. Norton & Company, Inc. Jones, C.1. (2002). Introduction to Economic Growth (2nd ed.). New York: W.W. Norton & Company, Inc. Jones, C.T. ( 1 994). Accounting for Technical Progress in Aggregate Energy Demand. Energy Economics, 16: 245-52. Jongeneel, R.A. ( 1 992). The Neo-Classical and Steady State Approaches. In J.J. Krabbe & WJ.M. Heijman (Eds.), National Income and Nature: Externalities, Growth and the Steady State. Dordrecht: K luwer Academic. Jorgenson, D.W. & Griliches, Z. ( 1967). The Explanation of Productivity Change. The Review of Economic Studies, 34(3): 249-83. 426 Jorgenson, D.W. & Yip, E. (200 1 ) . Whatever happened to Productivity Growth? In E. Dean, M.J. Harper, & C. Hulten (Eds.), New Developments in Productivity Analysis (pp. 205-246). Chicago: University of Chicago Press. Judge, A.A. & Ledgard, S.F. (2002). Non-point source emissions from agriculture in 26 catchments in the Waikato region. Hamilton: AgResearch, Ruakura Research Centre. Judson, O.P. ( 1 994). The Rise of the Individual-based Model in Ecology. Trends in Ecology and Evolution, 9: 9- 1 4. Kahn, H. ( 1 979). World Economic Development. Ontario: Hudson Institute. Kahn, H., Brown, W., & Martel, L. ( 1 976). The Next 200 Years - A Scenario For America and the World. New York: William-Morrow. Kahn, J.R. (2005). The Economic Approach to Environmental and Natural Resources (3rd ed). Australia: Thomson Southwestern. Kahn, R.F. ( 1 93 1 ) . The Relation of Home Investment to Employment. Economic Journal, June. Kaldor, N. ( 1 957). A model of economic growth. Economic Journal, December: 59 1 -624. Kaldor, N. ( 1 96 1) . Capital Accumulation and economic growth. In F.A. Lutz & D.C. Hague (Eds), The Theory of Capital (pp. 1 77-222). London: Macmillan. Kamien, M.I . & Schwartz, N.J. ( 1 978). Optimal Exhaustible Resource Depletion with Endogenous Technical Change. Review of Economic Studies, 45: 1 79- 1 96 . Kamien, M.I. & Schwartz, N.J . ( 1 982). The Role of Common Property Resources in Optimal Planning Models with Exhaustible Resources. In V.K. Smith & J.V. Krutilla (Eds.), Explorations in Natural Resources Economics. Baltimore: Johns Hopkins University Press. Kapp, K. ( 1 970). Environmental Disruption and Social Costs: A Challenge to Economics. Kyklos, 23: 833-48 . 427 Karr, J .R. ( 1 992). Ecological Integrity: Protecting Earth's Life-Support Systems. In R. Costanza, B .G. Norton, & B .O. Haskel l (Eds.), Ecosystem Health: New Goals for Environmental Management. Washington D.e . : Island Press. Katterl, A. & Kratena, K. ( 1 990). Reale Input-Output Tabelle und okologischer Kreislauf Heidelberg: Physica. Kaufmann, R.K. ( 1 995). The Economic Multiplier of Environmental Life Support: can Capital Substitute for a Degraded Environment. Ecological economics, 1 2, 67-79. Kauffman, S . ( 1 993). The Origins of Order: Self-Organisation and Selection in Evolution. New York: Oxford University Press. Kay, J.J. ( 1 99 1 ). A Non-Equilibrium Thermodynamic Framework for Discussing Ecosystem Integrity. Environmental Management. 15: 483-495. Kay, J.J. & Schneider, E. ( 1 994) . Embracing Complexity: The Challenge of the Ecosystem Approach. Alternatives, 20(3) : 32-38 . Kemp, M.e . & Long, N.V. ( 1 984) . Essays in the Economics of Exhaustible Resources. Amsterdam: North-Holland. Kerr, G.N. & Sharp, B.M.H. ( 1987). Valuing the Environment. Economic Theory and Applications . Centre for Resource Management. Studies in Resource Management No.2, Canterbury: L incoln Col lege. Kerr, G.N., Sharp, B.M.H., & Gough, J .D. ( 1986). Economic Benefits of Mt. Cook National Park. Christchurch: Centre for Resource Management, University of Canterbury and Lincoln College . Keynes, J .M. ( 1 936). The General Theory of Employment Interest and Money. New York: Harcourt and Company. Khalil , E. ( 1 990). The Entropy Law and Exhaustion of Natural Resources: Is Nicholas Georgescu-Roegen's Paradigm Defensible? Ecological Economics, 2: 1 63- 1 78 . 428 Khali l, E.1. ( 1 994). Letters to the Editor. Recycling of matter: Further. Ecological Economics, 9: 1 93- 1 94. Khalil, E.1. ( 1 995) . Ecological Economics, Neoclassical Economics and the TechnologicallInstitutional Regime of Production. British Review of Economics, 1 7: 41 -70. Khalil, E.1. ( 1 997) . Production and Environmental Resources: A Prelude to an Evolutionary Framework. Southern Economic Journal, 63: 929-946. Kim, D.H. ( 1 993) . The Link Between Individual and Organizational Learning. Sloan Management Review, 35: 14. Kim, D.H. & Senge, P.M. ( 1 994). Putting systems thinking into practice. System Dynamics Review, 10: 277-290. King, A.W. ( 1 993). Consideration of Scale and Hierarchy. In S. Woodley, J . Kay, & G. Francis (Eds.), Ecological Integrity and the Management of Ecosystems (pp. 1 9-45) . F lorida: St.Lucie Press. Kitching, R.1. ( 1 983). System Ecology: An Introduction to Ecological Modelling. Saint Lucia: Univers ity of Queensland Press. Klaassen, G.A.J. & Opschoor, J .B. ( 1 991) . Economics of sustainability or the sustainability of economics: different paradigms. Ecological Economics, 4: 93- 1 1 5 . Kleniewski, . (2002). Cities, Change, and Conflict: A Political Economy of Urban Life. Belmont, CA: Wadworth Thomson Learning. Kneese, A.V., Ayres, R.U. & d' Arge, R.e. ( 1 970). Economics and the environment - a materials balance approach. Washington, DC: Resources for the Future. Kneese, A.V. & Bower, B.T. ( 1 979). Environmental Quality and Residuals Management. Baltimore: Johns Hopkins University Press. Konijn, PJ.A. & Steenge, A.E. ( 1 995). Compilation of input-output data from the national accounts. Economic Systems Research, 7: 3 1 -45 . 429 Kop Jansen, P. & ten Raa, T. ( 1 990). The choice of model in the construction of input-output coefficients matrices. International Economic Review, 31 : 2 1 3-227 . Krelle, W . ( 1984). Economic Growth with Exhaustible Resources and Environmental Protection. Z. Ges. Staatswiss, 140: 399-429. Krutilla, J.Y. ( 1 967). Conservation Reconsidered. American Economic Review, 45: 777-786. Kuenne, R. ( 1 963). The Theory of General Equilibrium. Princeton: Princeton University Press. Kuhn, T. ( 1 962). The Structure of Scientific Revolutions. Chicago: Chicago University Press. Kula, E. ( 1 994). Economics of Natural Resources, the Environment and Policies. London: Chapman and Hall. Kulhavy, R.W. & Stock, W.A. ( 1 996). How Cognitive Maps Are Learned and Remembered. The Annals o/the Association of American Geographers 86 (March) : 1 23 -45 . Kummel, R. ( 1 994). Letters to the Editor. Recycl ing of matter: Energy, entropy - economy, ecology. Ecological Economics, 9: 1 94- 1 96. Kuschel, G . & Kingsland, S . ( 1 998). Total Air Emissions Inventory for New Zealand. NIW A Cl ient Report No. AK98064. Auckland: NIW A. Kuschel, G. & Petersen, J. ( 1 999) . Natural Emissions Inventory for the Waikato Region. NIW A C lient Report No. AK98 1 54. Auckland: NIW A. Lahr, M. (200 1 ) . Reconciling Domestication Techniques, the Notion of Re-exports and Some Comments on Regional Accounting. Economic Systems Research 13(2) : 1 65 - 1 79. Landcare Research (2003). Land Resources Inventory. Palmerston North: Landcare Research. Lashof, D.A. & Ahuja, D.R. ( 1 990). Relative Contributions of Greenhouse Gas Emissions to Global Warming. Nature, 344: 529-53 1 Leach, G. ( 1 975). Net Energy Analysis - Is It Any Use? Energy Policy, 3. 4 3 0 Lecomber, R . ( 1 97 5 ) . Economic Growth Vers us the Environment. London: Macmillan. Lecomber, R. ( 1 979). The Econom ics of Natural Resources. London and Basingstoke: Macmi llan. Le Corbusier. ( 1 929). The City of To-morrow and its Planning. London: J. Rodker. (Original work published 1 924) Le Corbusier. ( 1 967). The Radiant City. New York: Or ion Press. (Original work pub lished 1 93 5 ) Lee, K. ( 1 989). Social Philosophy and Ecological Scarcity. London and New York: Routledge. L e febvre, H. ( 1 99 1 ). The Production of Space (D. icholson-Sm ith, Trans .) . Oxford: B lackwell . (Original work published 1 974) Le Heron, R. & Mc Donald, G.W. (2005 ). Auckland Region 's Drivers of Structural Change. A uckland Un iversity. Manuscript in preparation. Lenzen, M. (2004). Measuring our Progress: Canberra 's Journey to Sustainability. Report on consultancy work carried out for the Chief Min ister's Department of the Australian Capital Territory. Canberra: University of Sydney Leontief, W. ( 1 9 70). Environmental Repercussions and the Economic Structure: An Input­ Output Approach. Review of Economics and Statistics, 52: 262-27 1 . Leontief, W . ( 1 9 8 5 ) . Input-Output Analysis. Reprinted from the Encyclopedia of Materials Sc ience and Engineering. In W. Leontief (Ed.), Input- Output Economics (2nd ed). New York: Oxford University Press. Leontief, W., & Duchin, F. ( 1 9 86). The Future Impact of A utomation on Workers . New York: Oxford University Press. Leopold, L. ( 1 974). Water: A Primer. San Franc isco: W.H. Freeman & Co. Lerman, A., MacKenzie, F.T. & Garrels, R.M. ( 1 975). Modeling of geochemical cycles: Phosphorus as an example. Geological Society of A merica Memoir, 142: 205-2 1 8 43 1 Leathwick, J., Wilson, G., Rutiedge, D. , Wardle, P., Morgan, F. , Johnston, K., McLeod M. & Kirkpatrick. R. (2003). Land Environments of New Zealand. Auckland: David Bateman Ltd. Levin, S.A. ( 1 998). Ecosystems and the Biosphere as Complex Adaptive Systems. Ecosystems, 1 : 43 1 -436. Levin, S.A., Barrett, S., Aniyar, S . , Baumol, W., Bliss, C., Bolin, B., et al. ( 1998). Resilience in natural and socioeconomic systems. Environment and Development Economics, 3(2) : 222-234. Lifset, R. & Graedel, T.E. (2002). Industrial Ecology: Goals and Definitions. In R.U. Ayres & L.W. Ayres (Eds.), A Handbook of Industrial Ecology (pp. 3-1 5). Cheltenham: Edward Elgar. Likens, G. & Borman, F. . ( 1974). Linkages between Terrestrial and Aquatic Ecosystems. Bioscience, 24: 447-456 Lindeman, R.L. ( 1 942). The trophic-dynamic aspect of ecology. Ecology, 23(4): 399-4 1 8. Lister, N-M. ( 1 998). A Systems Approach to Biodiversity Conservation Planning. Environmental Monitoring and Assessment, 49: 1 23-1 5 5 . Loeschel, A . (2002). Technological Change in Economic Models of Environmental Policy: A Survey. Ecological Economics, 43: 1 05- 1 26. Lofland, L. ( 1 973). A World of Strangers. New York: Basic Books. Logan, J . & Molotch, H. ( 1987). Urban Fortunes: A Political Economy of Place. Berkeley : University of California Press. Loh, J. (Ed.) . (2000). Living Planet Report 2000. Gland, Switzerland: WWF-World Fund For Nature. Lonergan, S.C. & Cocklin, C. ( 1 985). The Use of Input-Output Analysis in Environmental Planning. Journal of Environmental Management, 20: 1 29- 1 47. Losch, A. ( 1 954). Location and Space Economy (W.H. Woglom, Trans.) . Cambridge, MA: MIT Press. (Original work published 1 943) 432 Lotka, AJ. ( 1 922). Natural selection as a physical principle. Proceedings of the National Academy of Sciences, 8: 1 47- 1 55 . Lotka, AJ . ( 1 925). Physical Biology. Williams and Wilkins, Baltimore, MD. Lotka, AJ. ( 1 945). The law of evolution as a maximum principle. Human Biology, 1 7: 1 67 . Lovelock, J.E. ( 1 979). Gaia: A New Look at Life on Earth. New York: Oxford University Press. Lucas, R.E. ( 1 988). On the mechanics of economic development. Journal of Monetary Economics, 22: 3-42. Lynch, K. ( 1 960). The Image of the City. Cambridge, MA: MIT Press. MacArthur, R.H. ( 1 95 5). Fluctuations of animal populations and a measure of community stability. Ecology, 36: 533-536. Macionis, J.J. & Parrillo, V.N. ( 1 998). Cities and Urban Life. Upper Saddle River, NJ: Prentice Hal l . Mackenzie, F .T. & Mackenzie, J .A. ( 1 995) Our Changing Planet: An Introduction to Earth System Science and Global Environmental Change. New Jersey: Prentice Hall . Mackenzie, F .T., Ver, L.M., Sabine, c., Lane, M. , & Lerman, A. ( 1 993). C, N, P, S Global Biogeochemical Cycles and Modeling of Global Change. In R. Wollast, F.T. Mackenzie & L. Chou (Eds.), Interactions of C, N, P and S Biogeochemical Cycles and Global Change (pp. 1 - 62) . Berlin: Springer-Verlag. Maddox, J. ( 1 972). The Doomsday Syndrome. New York: McGraw-Hill Madsen, B. & Jensen-Butler, C. ( 1 998). Commodity Balance and Interregional Trade: Make and Use Approaches to Interregional Modelling. Paper presented at the 1 2th International Conference on Input-Output Techniques, New York, 1 8-22 May 1 998. Madsen, B . & Jensen-Butler, C . ( 1 999). Make and Use Approaches to Regional and Interregional Accounts and Models. Economic Systems Research, 1 1 : 277-299. 433 Maler, K.G. ( 1 974). Environmental Economics: A Theoretical Inquiry. Baltimore: Johns Hopkins University Press. /---, Malthus, 1.R. ( 1 964). An Essay on the Principle of Population. London: J.M. Dent and Sons. (Original work published 1 798) MandeviIIe, T.D. ( 1 975). Linking APMAA to Representative Regional Input-Output Models. Aggregative Programming Model for Australian Agriculture Research Report No. 8. Armidale: Department of Agricultural Economics, University of New England. Manne, A.S. & Richels, RG. ( 1 992). Buying Greenhouse Insurance: The Economic Costs of CO2 Emission Limits. Cambridge, MA: MIT Press. Mansson, B.A. ( 1 994). Letters to the Editor. Recycling of matter: A response. Ecological Economics, 9: 1 9 1 - 1 92. MarcotuIIio, P.l . & Boyle, G (Eds.). (2003). Defining an Ecosystem Approach to Urban Management and Policy Development. UNU/IAS Report. Tokyo: United Nations University Institute of Advanced Studies. Retrieved 1 4 July 2005 from www. ias.unu.edulbinariesIUNUIAS_UrbanReport l .pdf. MarshaIl, A. ( 1 949). Principles of Economics: An Introductory Volume (8th ed.). London: MacmiIIan. (Original work published 1 890) Martinez-Al ier, 1. ( 1 987). Ecological Economics. Oxford: Basil B lackwel l Ltd. Marx, K. ( 1 909). Das Kapital. (S. Moore, E. Aveling and E Untermann, Trans.). Chicago: Kerr. (Original work publ ished 1 867) Marx P. ( 1 984). A slowdown in labour productivity growth rates in New Zealand in the 1 970s. In B. Easton (Ed), 1 984 studies in the labour market. Wellington: NZIER. Matthews, E. , Amann, C. , Bringezu, S., F ischer-Kowalski, M. , Hlittler, W., Kleijn, K, et al. (2000). Weight of Nations: Material outflows from industrial economies. Washington DC: World Resources Institute. 434 May, R.M. ( 1 972). Will a Large Complex System be Stable? Nature, 238: 41 3-4 1 4 . May, R.M. ( 1 974). Biological Populations with Non-Overlapping Generations: Stable Points, Stable Cycles and Chaos. Science, 186: 645-647. Mayer, W. & Pleeter, S . ( 1 975) . A theoretical justification for the use of location quotients. Regional Science and Urban Economics, 5: 343-3 5 5 . McCann, P . (200 1 ). Urban and Regional Economics. New York: Oxford University Press Inc. McDermott Fairgray Group Ltd and Massey University. ( 1 999). EcoLink Database. Takapuna: McDermott Fairgray Group Ltd. McDermott Fairgray Group Ltd. ( 1 998). Unpublished Report on Residential Water Supply and Use in Auckland Region. Takapuna: McDerrnott Fairgray Group Ltd. McDermott, P. ( 1 998). Tourism. In M. P ickford & A. Bi l lard (Eds.), The Structure and Dynamics of New Zealand Industries (pp.323-356) . Palmerston North: Dunmore Press. McDonald, G.W. ( 1 997). Integrated Economic and Ecological Accounts for the Manawatu­ Wanganui and Wellington Regions. Unpublished master's thesis, Massey Univers ity, Palmerston North, New Zealand. McDonald, G.W. ( 1 999a). EcoLink Economic Accounts Technical Report. Takapuna: McDermott Fairgray Group Ltd. Ch5 McDonald, G.W. ( 1 999b). EcoLink Land Accounts Technical Report. Takapuna: McDermott Fairgray Group Ltd & Massey University. Ch 6 McDonald, G.W., Forgie, V .E. & MacGregor, C. (in press). Treading Lightly: Ecofootprints of New Zealand's Ageing Nation. Ecological Economics. McDonald, G.W., Le Heron, K., & Patterson, M .G. ( 1 999). EcoLink Water Accounts Technical Report. Takapuna: McDerrnott Fairgray Group Ltd. McDonald, G.W., Le Heron, R., & Patterson, M.G. (2005) . Assessing the Value of Regional Ecosystem Services: Canterbury 's 'Hidden ' Economy. Manuscript in preparation. 435 McDonald, G.W., & Le Heron , R. (2005). Changes i n the Clusters of Comparative Advantage in the A uckland Region Economy 1986-2001 . Manuscript in preparation. McDonald, G.W. & MacGregor, C. (2004). Land Accounts Technical Report. Takapuna: Market Economics Ltd. McDonald, G.W., MacGregor, C. & Patterson, M.G. (2004). Energy Accounts Technical Report. Takapuna: Market Economics Ltd. McDonald, G.W. & Patterson, M.G. ( 1 994). Manawatu-Wanganui Regional Council: Environmental Input-Output Models. Palmerston North: Department of Resource and Environmental Planning, Massey University. McDonald, G.W. & Patters on, M.G. ( 1 995a). Wellington Regional Council: 1992- 1993 Commodity-by-Industry Environmental Input-Output Model. Palmerston North : Department of Resource and Environmental Planning, Massey University. McDonald, G.W. & Patterson, M.G. ( 1 995b). Wellington Regional Council: 1992-1 993 Industry-by-Industry Environmental Input-Output Model. Palmerston North: Department of Resource and Environmental Planning, Massey University. McDonald, G.W. & Patterson, M.G. ( 1 995c). Wellington Regional Council: 1992- 1993 Industry-by-Industry Multipliers. Palmerston North: Department of Resource and Environmental Planning, Massey University. McDonald, G. W. & Patterson, M. ( 1 998). Ecolink Water Accounts: Technical Report. Takapuna, Auckland: McDermott Fairgray Group. McDonald, G.W. & Patterson, M.G. ( 1 999). EcoLink Database. Takapuna: McDermott Fairgray Group Ltd & Massey University. McDonald, G.W. & Patterson, M.G. (2002). Our Debt to Nature: Ecological Footprints o/New Zealand and its Regions. Palmerston North: Massey University 436 McDonald, G.W. & Patterson, M.G. (2003a). Canterbury Region 's 'Hidden ' Economy: Assessing the Value of the Region 's Ecosystem Services and Biodiversity. Palmerston North: Landcare Research Ltd. McDonald, G.W. & Patterson, M.G. (2003 b). Ecological Footprints of New Zealand and its Regions. Wellington: New Zealand Ministry for the Environment. ISBN 0-478-24085-6. McDonald, G.W. & Patterson, M.G. (2003c). Ecological Footprints of New Zealand Regions. Environmental Reporting Technical Paper. Wellington, New Zealand: Ministry for the Environment. McDonald, G.W. & Patterson, M.G. (2004). Ecological Footprints and Interdependencies of New Zealand Regions. Ecological Economics, 50: 49-67. McDonald, G.W. & Patterson, M.G. (2005) . A System Dynamics Model of Global Biogeochemical Cycling. Manuscript in preparation. McElroy, M.B. & E lkins, J .W., Wofsy, S .c. , & Yung, Y.L. ( 1 976). Sources and sinks for atmospheric N20. Reviews of Geophysics and Space Physics, 14(2) : 1 43-50. McElroy, M.B. , Wofsy, S .c. , & Tung, Y.L. ( 1 977). The nitrogen cycle: perturbations due to man and their impact on atmospheric N20 and 03 . Philosophical Transactions of the Royal Society of London, B277: 1 59- 1 8 1 . McMenamin, D.G. & Haring, J.V. ( 1 974). An appraisal of non-survey techniques for estimating regional input-output models. Journal of Regional Science, 14: 1 9 1 -205. McShane, O. ( 1 998). Land Use Control Under the Resource Management Act. A Report to the Minister for the Environment. Kaiwaka: McShane Venture Management Ltd. Meadows, D.H., Meadows, D.L. , & Randers, J. ( 1 992). Beyond the Limits: Global Collapse or a Sustainable Future. London: Earthscan Publications Limited. Meadows, D.H., Meadows, J., Randers, J . & Behrens, W.W. ( 1 972). The Limits to Growth. New York: Universe Books. 437 Meadows, D., Randers, J . & Meadows D. (2004). Limits to Growth: The 30 Year Update. Vermont: Chelsea Green Publishing Company. Meat & Wool Economic Service of New Zealand. ( 1 999). Annual Review of the New Zealand Sheep and Beef Industry 1998-99. Wellington: Meat & Wool Economic Service of New Zealand. Menger, C. ( 1 950). Principles of Economics. (J. Dingwall, & B.F. Hoselitz (Eds.), Trans.) . Glencoe, Illinois: Free Press. (Original work published 1 87 1 ) Michelson, W . & Van Vl iet, W. (2002). Theory and the Sociological Study of the Bui lt Environment. In R.E. Dunlap & W. Michelson (Eds.), Handbook of Environmental Sociology (pp. 70-95). Westport, CT: Greenwood Press. Miernyk, W. ( 1 968). Long range forecasting with a regional input-output table. Western Economic Journal, 6: 1 65 - 1 76. Miernyk, W. ( 1 969). Comment of Czamanski and Mal izia. Papers and Proceedings, Regional Science Association, 23: 8 1 -82. Miernyk, W. ( 1 976). Comments on recent developments III regional input-output analysis. International Regional Science Review, 1: 2, 47-55 . Miernyk, W.H., Bonner, E.R. , Chapman, J.H., & Shellhammer, K.L. ( 1 967). Impact of the space program on local economy: An input-output analysis. Morgantown, WV: West Virginia University Foundation. Miernyk, W.H., Shellhammer, K.L., Brown, D.M., Coccari, R.L., Gallagher, C.J. and Wineman, W.H. ( 1 970). Simulating regional economic development with an input-output model. West Virginia: Regional Research Institute, West Virginia University. Miles, S. & Miles, M. (2004). Consuming Cities. New York: Palgrave Macmillan. Milgram, S. 1 972. A Physical Map of New York C ity. American Scientist, 60: 1 94-200. M ill, J .S. ( 1 962). Principles of Political Economy With Some of Their Application to Social Philosophy (5th ed.). London: Parker, Son & Bourn. (Original work published 1 848) 438 Mil ler, G.T. ( 1 995). Environmental Science: Sustaining the Earth (5th ed.). New York: Wadsworth Publishing Company. Mil ler, G.T. ( 1 996). Living in the Environment: Principles, Connections, and Solutions. New York: Wadsworth Publishing Company Miller, R.E. & Blair, P.D. ( 1 98 5) . Input-output analysis: Foundations and extensions. Englewood Cliffs, NJ: Prentice-Hal l . Ministry for the Environment. (n.d.a). Retrieved from http://www.environment.govt.nziindicators/waste/landfil l/ Ministry for the Environment. (n.d.b). Retrieved from http://www.environment.govt.nziindicators/waste/recycle/ Ministry for the Environment. ( 1 997a). National Waste Data Report. Well ington: Ministry for the Environment. Ministry for the Environment. ( 1 997b). State of the Environment Report. Wellington: Ministry for the Environment. M inistry of Agriculture & Forestry. ( 1 997). Impacts of Dairy Conversions in the Taupo District. Well ington: Ministry of Agriculture and Forestry. Ministry of Agriculture & Forestry. ( 1 998). Farm Monitoring Report North Region. Hamilton: M inistry of Agriculture & Forestry. Ministry of Agriculture & Forestry. ( 1 999). A National Exotic Forest Description as at 1 April 1998. Well ington: Ministry of Agriculture and Forestry. M inistry of Agriculture & Forestry. (2000). A National Exotic Forest Description as at 1 April 1999. Wellington: Ministry of Agriculture & Forestry. Ministry of Economic Development. ( 1 998). Crown Mineral Production Statistics. Wellington: Ministry of Economic Development. 439 Mishan, EJ. ( 1 977). The Economic Growth Debate: An Assessment. London: George Allen & Unwin. Mitchell, R.e. & Carson, R.T. ( 1 984). Willingness to Pay for National Freshwater Quality Improvements. Draft Report Prepared for U.S. Environmental Protection Agency, Washington, DC, Resources for the Future. Moffatt, L (2000). Ecological Footprints and Sustainable Development. Ecological Economics, 32: 3 59-362. Moore, C.L ( 1 98 1 ). An Input-Output Model of Northland 's Economy: With Application to Forestry. Unpublished PhD Thesis. Auckland: Auckland University. More, T.A., Averi ll, J . , & Stevens, T.H. ( 1 996). Values and Economics in Environmental Management: A Perspective and Critique. Journal of Environmental Management 48: 397- 407. Morrill, R. ( 1 963). The development of spatial distributions of towns in Sweden. Annals of the Association of American Geographers, 53: 1 - 1 4. Morrison, D.E. ( 1 976). Growth, Environment, Equity and Scarcity. Social Science Quarterly, 5 7: 292-306. Morrison, W.!. & Smith, P. ( 1 974). Nonsurvey Input-Output Techniques at the Small Area Level : An Evaluation. Journal of Regional Science, 14: 1 - 1 4 . Mumford, L . ( 196 1 ) . The City in History. New York: Harcourt, Brace and World. Munasinghe, M. & McNeely, J. ( 1 995). Key Concepts and Terminology of Sustainable Development. In M. Munasinghe & W. Shearer (Eds.), Defining and Measuring Sustainability: The Biogeophysical Foundations. Washington D.C. : World Bank. Murray, J.W. ( 1 992). The Global Carbon Cycle. In S .S . Butcher, RJ. Charlson, G.H. Orians & G.V. Wolfe (Eds.), Global Biogeochemical Cycles (pp. 239-262). London: Academic Press. National Business Review. ( 1 994, July 8). Auckland's drought threatens $800 million drain on economy. National Business Review. 440 Nebbia, G. ( 1 999). Contabilita Monetaria e Contabilita Ambientale. Lectio doctoralis, Laurea honoris causa in Economia e Commercio, Universita di Bari, 30 January. Nelson, R.R. & Winter, S. ( 1 982). An Evolutionary Theory of Economic Change. Cambridge MA: Harvard University Press. Nentjes, A. & Wiersema, A. ( 1 992). Economic Growth, Technology and the Environment. In J .J . Krabbe & WJ.M. Heijman (Eds.), National Income and Nature: Externalities, Growth and the Steady State. Dordrecht: Kluwer Academic. Neumayer, E. (2003) . Weak versus Strong Sustainability (2nd ed.). Cheltenham: Edward Elgar. New Zealand Mining. ( 1 998). Mineral Production Summary 1996 Calendar Year, 23. New Zealand Vegetable and Potato Growers ' Federation Inc. (n.d.). Retrieved from http://www.vegetables.co.nz New Zealand Wool Group (n.d.) Retrieved from http://www.woolgroup.co.nzlmarket_infolhistorical_info.htmI Newcombe, K. ( 1 975a) . Energy use in Hong Kong, Part I: An Overview. Urban Ecology, 1 : 87- 1 1 3 . Newcombe, K . ( 1 975b). Energy use i n Hong Kong, Part IT: Sector end-use analysis. Urban Ecology, 1 : 285-309. Newcombe, K. ( 1 976). Energy use in Hong Kong, Part III: Spatial and temporal patterns. Urban Ecology, 2: 1 39- 1 72. Newcombe, K. ( 1 977a). Apparent consumption and socio-economic distribution of nutrients in an urban settlement: Hong Kong. Ecology of Food and Nutrition, 6: 9-22. Newcombe, K. ( 1 977b). Nutrient flow m a major urban settlement: Hong Kong. Human Ecology, 5(3) : 1 79-208. 441 Newcombe, K . , Kalma, J .D., & Aston, A.R. ( 1 978). Metabolism of a c ity: the case of Hong Kong. Ambio, 7( 1 ) : 1 . Newman, P .W.G. ( 1996) . Greening the city : The ecological and human dimensions of the city can be part of town planning. Alternatives 22(2): 1 0- 1 7 . Newman, P . W.G. ( 1 999). Sustainability and cities: extending the metabol ism model. Landscape and Urban Planning, 44: 2 1 9-226. Newton, P. W. (200 1 ). Human Settlements Theme Report: Australian State of the Environment Report 200 1 . Canberra: CSIRO Building, Construction and Engineering. Newman, P., B irre1, B., Ho1mes, D., Mathers, c. , Newton, P . , Oakley, G., et al. ( 1 996). Human settlements. In Australia: State of the Environment Report. Canberra: Department of Environment, Sport and Territories. Newman, P. & Kenworthy, J. ( 1 999). Sustainability and Cities: Overcoming Automobile Dependence. Washington, DC : Island Press. Nijkamp, P. ( 1 977) . Theory and Application of Environmental Economics. Amsterdam: North­ Holland. Nijkamp, P. & Pepping, G. ( 1 998). A Meta-analytical Evaluation of Sustainable City Initiatives. Urban Studies, 35(9): 148 1 - 1 500. Nordhaus, W.D. ( 1 969). An Economic Theory of Technological Change. American Economic Review,59(2) : 1 8-28 . Nordhaus, W.D. ( 1 990). Green Economics: Count Before You Leap. The Economist, July 7: 57 . Nordhaus, W.D. ( 1 992). The 'DICE'Model: Background and Structure of a Dynamic Integrated Climate-Economy Model of the Economics of Global Warming. Cowles Foundation D iscussion Paper No. 1 009. New Haven, Conn. : Cowles Foundation for Research in Economics. Nordhaus, W.D. ( 1 994). Managing the Global Commons. The Economics of Climate Change. Cambridge MA: MIT Press. 442 Nordhaus, W.D. ( 1 999). Requiem for Kyoto: An Economic Analysis of the Kyoto Protocol. Energy Journal, Kyoto Special Issue, 93- 1 30. Nordhaus, W & Tobin, J . ( 1 972). Is Growth Obsolete? In Economic Growth. National Bureau of Economic Research General Series #96E. New York: Columbia University Press. Norgaard, R.B. ( 1 986). Thermodynamic and Economic Concepts as Related to Resource-Use Policies: A Synthesis. Land Economics, 62: 325-327. Norgaard, R.B . ( 1 989). The Case for Methodological Pluralism. Ecological Economics, 1 ( 1 ) : 37-57. Norgaard, R.B. ( 1 990). Economic Indicators of Resource Scarcity: A Critical Essay. Journal of Environmental Economics and Management, 19( 1 ) : 1 9-25 . North, D. ( 1 98 1 ) . Structure and Change in Economic History. New York: Norton. Norton, B.G. ( 1 986). The preservation of species: the value of biological diversity. Princeton, NJ. : Princeton University Press. Norton, B.G. ( 1 995). Evaluating Ecosystem States: Two Competing Paradigms. Ecological Economics, 14: 1 1 3 - 1 27. Noss, R.F. ( 1 992). Issues of Scale in Conservation Biology. In P.L. Fiedler & S.K. Jain (Eds.), Conservation Biology: The Theory and Practice of Nature Conservation Preservation and Management. London: Chapman & Hall. O'Connor, M. ( 1 99 1 ). Entropy, Structure, and Organisational Change. Ecological Economics, 3 : 95-1 22. O'Connor, M. ( 1 993) . Entropic irreversibil ity and uncontrolled technological change In economy and environment. Evolutionary Economics, 3: 1 -3 1 . O'Connor, R. & Henry, E.W. ( 1 975). Input-Output Analysis and it Application. London: Charles Griffith. 443 Odum, E.P. ( 1 969). The Strategy of Ecosystem Development. Science, 164: 262-270. Odum, E.P. ( 1 97 1 ). Fundamentals of Ecology (3rd ed.). Philadelphia: W.B. Saunders. Odum, E.P. ( 1 983). Basic Ecology. Philadelphia: Saunders College Publishing. Odum, E.P. ( 1 994). Ecological and General Systems: An Introduction to Systems Ecology. Colorado: University Press of Colorado. Odum, H.T. ( 1 957) . Trophic Structures and Productivity of Si lver Springs, Florida. Ecological Monographs, 2 7: 5 5-1 1 2 . Odum, H.T. ( 1 97 1 ) . Environment, Power, and Society. New York: Wiley-Interscience. Odum, H. T. ( 1 975). Energy quality and carrying capacity of the Earth (Response at Prize Ceremony, Institute La Vie, Paris). Tropical Ecology, 1 6( 1 ): 14. Odum, H.T. ( 1983). Systems Ecology: An Introduction. New York: Wiley and Sons. Odum, H.T. ( 1988). Self-organization, transformity and information. Science, 242: 1 1 32- 1 1 39 . Odum, H.T. ( 1 99 1 ) . Emergy and biogeochemical cycles. In C . Rossi & E. Tiezzi (Eds.), Ecological Physical Chemistry (pp. 25-56). Proceedings of an International Workshop, 8- 1 2 November 1 990. Amsterdam: Elsevier Science Publishers. Odum, H. T. ( 1 996). Environmental Accounting: Emergy and Environmental Decision Making. New York: Wi ley. Odum, H.T. & Brown, M.T. ( 1 975) . Carrying capacity for man and nature in South Florida : energy models for recommending energy, water and land use for long range economic vitality in South Florida. Gainesville: University of F lorida. OECD (Organisation for Economic Cooperation and Development). ( 1 998). Eco�efficiency. Paris: OECD. O'Neill , R.V., DeAngelis, D.L., Waide, J.B., & AlIen, I.F.H. ( 1 986). A hierarchical concept of ecosystems. Princeton: Princeton University Press. 444 O'Neill , R.V. , Johnson, A.R., & King, A.W. ( 1 989). A Hierarchical Framework for the Analysis of Scale. Landscape Ecology, 3(3) : 1 93 -205 . Oosterhaven, J. ( 1 984). A Family of Square and Rectangular Interregional Input-Output Tables and Models. Regional Science and Urban Economics, 14: 565-82. O'Riordan, T. ( 1 993). Interpreting the Precautionary Principle. CSERGE Working Paper PA 93-0 1 . Norwich: University of East Anglia. Orr, A. ( 1 989). Productivity trends and cycles in New Zealand: a Sectoral and Cyclical Analysis 1961-1987. Research Monograph 48. Wellington: NZIER Orum, A.M. & Chen, X. (2003). The World of Cities: Places in Comparative and Historical Perspective. MaIden, MA: B lackwel l Publishing. Oser, J. & Brue, S .L. ( 1 988) . The Evolution of Economic Thought (4th ed.). New York: Harcourt Brace Jovanovich. Pacione, M. (200 1 ). Urban Geography: A Global Perspective. London: Routledge. Packaging Council of New Zealand http://www . packaging.org. nzJpackaging_ index.html Inc. (n.d.). Retrieved from Paddison, R. (200 1 ) . Studying C ities. In R. Paddison (Ed.), Handbook of Urban Studies (pp. 1 - 9). London: SAGE Publications. Paddock, P. & Paddock, W. ( 1 967). Famine - 1975. Boston: Little & Brown. Paelinck, J. & Waelbroeck, J. ( 1 963) . Etude empiri su l' evolution des coefficients input-output. Economie Appliquee, 16: 8 1 - 1 1 1 . Paglin, M. ( 1 96 1 ). Malthus and Lauderdale: The Anti-Ricardian Tradition. New York: M. Kelley. 445 Panico, C. (2003) . Old and new growth theories: what role for aggregate demand? In: N. Salvadori (Ed.), Old and New Growth Theories: An Assessment (pp .53-66). Cheltenham, UK: Edward Elgar. Park, R.E. ( 1 967). The City : Suggestions for the Investigation of Human Behavior in the Urban Environment. In Park, R.E. and Burgess, E.W. (Eds.), The City (pp. 1 -46). Chicago: University of Chicago Press. (Original work published 1 9 1 6) Park, R.E. & Burgess, E. W. (Eds). ( 1 967). The City. Chicago: University of Chicago Press. Park, R.E., Burgess, E.W., & McKenzie, R. D. ( 1 925). The City. Chicago: University of Chicago Press. Patten, B.c. ( 1 98 1 ) . Environs: the supemiches of ecosystems. American Zoologist, 2 1 : 845- 852. Patten, B.C. ( 1 982). Environs: relativistic elementary particles or ecology. American Naturalist, 1 19: 1 79-2 19. Patten, B .C. , Bosserman, C.R.W., F inn, J.T. & Cale, W.G. ( 1 976). Propagation and Cause in Ecosystems. In B. Patten (Ed.), Systems Analysis and Simulation in Ecology, Vo!. 4 (pp. 457- 479). New York: Academic Press. Patten, B.C. & Jorgensen, S. ( 1 995). Complex Ecology: The Part-Whole Relationship in Ecosystems. Englewood C liffs, NJ: Prentice-Hall . Patterson, M.G. ( 1 980). Wood Energy Resources in New Zealand. Unpublished master's thesis. University of Canterbury and Lincoln College, Christchurch, New Zealand. Patterson, M.G. ( 1 998). Commensuration and Theories of Value in Ecological Economics. Ecological Economics, 25: 1 05 - 1 25 . Patterson, M.G. (2002a). Ecological Production Based Pricing of Biosphere Processes. Ecological Economics, 41(3) : 457-478. Patters on, M.G. (2002b). Headline Indicators for Tracking Progress to Sustainability in New Zealand. Wellington: Ministry for the Environment. 446 Patterson, M.G. (2004). Construction of EECA Energy End-Use Database 2002: Technical Report. Palmerston North: Riverdale Associates Ltd Patterson, M.G. (2005) . Global Biogeochemical Cycling Model . Manuscript in preparation. Patterson, M.G. & Co le, A.O. ( 1 999a). Assessing the Value of New Zealand 's Biodiversity. Occasional Paper Number 1 . Palmerston orth: School of Resource and Environmental Planning, Massey University. Patterson, M.G. & Cole, A.O. ( 1 999b). Estimation of Ecosystem Services in the Waikato Region. Environment Waikato Internal Series 1 999/02. Hamilton: Environment Waikato. Patterson, M.G. & McDonald, G.W. ( 1 996). Regional Level Environmental Accounting Systems in New Zealand, Using Input-Output Methodologies. Institute of Environmental Studies (Ed.), Tracking Progress: Linking Environment and Economy through Indicators and Accounting Systems Conference Papers. Australian Academy of Science Fenner Conference on the Environment. Sydney: Institute of Environmental Studies, The University of New South Wales, 30 September to 3 October 1 996. Patterson, M.G. & McDonald, G.W. (2004). How Clean and Green is New Zealand Tourism? Lifecycle and Future Environmental Impacts. Landcare Research Science Series No. 24. L incoln, Canterbury : Manaaki Whenua Press. Patterson, M.G., McDonald, G.W., Golubiewski, N.E. , Forgie, V.E. & lollands, N.A. (In press). C limate change impacts on regional development and sustainabi lity: an analysis of New Zealand regions. In M. Ruth (Ed.), Smart Growth and Climate Change: Regional Development, Infrastructure and Adaptation. Edward Elgar Publishing Ltd . . PCE (Parliamentary Commissioner for the Environment). ( 1 998). The Cities and Their People. New Zealand's Urban Environment. Well ington: Office of the Parliamentary Commissioner for the Environment. Pearce, D. & Turner, K. ( 1 990). Economics of Natural Resources and the Environment. London: Harvester Wheatshelf. 447 Pearce, D., Markandya, A., & Barbier, E.B. ( 1 989). Blueprint for a Green Economy. London: Earthscan. Pearce, D., Markandya, A., & Barbier, E.B. ( 1 990). Sustainable Development. London: Earthscan. Pearce, D.W. & Atkinson, G. ( 1 993) . Capital Theory and the Measurement of Weak Sustainability. Ecological Economics, 8: 1 03 - 1 08. Pearce, D. W. & Barbier, E.B. (2000). Blueprint for a Sustainable Economy. London: Earthscan Publications Ltd. Pedersen, P.O. ( 1 967). Modeller for Befokningsstrucktur og Bejlolkingssudvikling I Storbymorader Specielt med Henblik pa Storkobenhavn. Copenhagen, Denmark: Stat Urban Planning Institute. Peet, N.J. ( 1 986). Energy requirements of output of the New Zealand economy, 1 976-77. Energy, 1 1(7): 659-670. Pennington, N. & Hastie, R. ( 1 99 1 ). A Cognitive Theory of Juror Decision Making: The Story Model. Cardo=o Law Review, 1 3 : 5 19-557. Perrings, C . ( 1987) . Economy and environment: a theoretical essay on the interdependence of economic and environmental systems. Cambridge: Cambridge University Press. Perrings, e. ( 1 99 1 ). Ecological Sustainability and Environmental Control. Australia National University: Centre for Resource and Environmental Studies. Perrings, C. ( 1 994). Biotic Diversity, Sustainable Development and Natural Capital. In A. Jansson, M. Hammer, e. Folke, & R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington D.e . : Island Press. Perrings, e . ( 1995a). Biodiversity Loss: Economic and Ecological Issues. Cambridge: Cambridge University Press . Perrings, e. ( 1 995b). The Economic Value of B iodiversity. In V.B. Heywood (Ed.), Global Biodiversity Assessment (pp 823-9 1 4). Cambridge: Cambridge University Press. 448 Pezzey, J. ( 1 992). Sustainability: An Interdisciplinary Guide. Environmental Values, 1 : 3 2 1 - 362 . Pezzey, J.C.V. ( 1 997). Sustainabil ity constraints versus 'optimality' versus intertemporal concern, and axioms versus data. Land Economics, 73(4): 448-467. Pezzoli, K. ( 1 997a). Sustainable Development: A Transdisc iplinary Overview of the Literature. Journal of Environment Planning and Management, 40(5): 549-574. Pezzol i, K. ( 1 997b). Sustainable Development Literature: A Transdisciplinary Bibliography. Journal of Environment Planning and Management, 40(5): 5 75-60 1 . Piispala, J . ( 1 998). Regional Input-Output Tables Based on Supply and Use Framework: The Finnish Case. Paper presented at Structures and Prospects of Nordic Regional Economics, Savonlinna, 4-7 June 1 998 . Piispala, J. (2000) . On Regionalising Input/Output Tables - Experiences from Compiling Regional Supply and Use Tables in Finland. Paper presented at the 1 3th International Conference on Input-Output Techniques at the University of Macerata, Italy, 2 1 -25 August 2000. Pindyck, R.S. ( 1 978). The Optimal Exploration and Production of Non-Renewable Resources. Journal of Political Economy, 86: 84 1 -86 1 . Phelps, E.S. ( 1 966). Models o f Technical Progress and the Golden Rule of Research. Review of Economic Studies, 33: 1 3 3-45 . Popenoe, D . & Michelson, W. (2002). Macro-Environments and People: Cities, Suburbs, and Metropolitan Areas. In R.E. Dunlap & W. Michelson (Eds.), Handbook of Environmental Sociology (pp. 1 37- 1 66). Westport, CT: Greenwood Press. Poultry Industry Association of New Zealand Inc. (n.d.). Retrieved from http://www.pianz.org.nzl Pribram, K. ( 1 983). A History of Economic Reasoning. Baltimore: Johns Hopkins University Press. 449 Prigogine, I. ( 1 967). Thermodynamics and Irreversible Processes. New York: Wiley. Prigogine, I . ( 1 973). The Statistical Interpretation of Non-Equi librium Entropy. Acta Phys - Austriaca Suppl, 1 0: 40 1 -450. Prigogine, I . , Nicolis, G., & Babloyantz, A. ( 1 972). Thermodynamics of Evolution. Physics Today, 25: 23-38 . Prigogine, I . & Stengers, I . ( 1 984). Order Out of Chaos . London: Heinemann. Pugh-Roberts Associates. ( 1986). Professional Dynamo Reference Manual. Cambridge : Pugh­ Roberts Associates. Quotable Value New Zealand. ( 1998). Unpublished Land Area Data Disaggregated by 99 Use Codes for all Regions in New Zealand. Wellington: Quotable Value New Zealand. Radermacher, W. & Stahmer, C. ( 1 996). Material and Energy Flow Analysis in Germany: Accounting Framework, Information Systems, Application. Paper presented at the Special IARlW Conference, International Symposium on Integrated Environmental and Economic Accounting in Theory and Practice, Tokyo, March 5-6. Ramsey, F.P. ( 1 928). A Mathematical Theory of Saving. Economic Journal, 38( 1 52) : 543-559. Randal1, H. ( 1 987). Resource Economics: An Economic Approach to Natural Resource and Environmental Policy (2nd ed.). New York: John Wiley & Sons. Raphael, D.D. ( 1 985). Adam Smith. London: Oxford University Press. Rees, J. ( 1 990). Natural Resources: Allocation, Economics and Policy (2nd ed.). London and New York: Methuen. Rees, W.E. ( 1 992). Ecological Footprints and Appropriated Carrying Capacity: What Urban Economics Leaves Out. Environment and Urbanization, 4(2) : 1 2 1 - 1 30 . Rees, W .E . (2000) . Eco-footprint Analysis: Merits and Brickbats. Ecological Economics, 32: 3 7 1 -374. 450 Reid, R. ( 1 995). Sustainable Development: An Introductory Guide. London: Earthscan Publications. Resource Management Act. ( 1 99 1 ) . Wellington: Government Printer. Revelle, R. ( 1 982). Carbon dioxide and world climate. Scientific American, 247(2) :35-43 . . Ricardo, . ( 1 973). The Principles of Political Economy and Taxation. London: J.M. Dent and "sG-I:I.s. ·Original work published in 1 8 1 7) Richardson, H.W. ( 1 972). Input-Output and Regional Economics. London: Weidenfeld and Nicho1son Ltd. Richardson, H.W. ( 1 973). Regional Growth Theory. London: Macmillan Press. Richardson, L. (Ed.) (2000). Bateman New Zealand Encyclopedia (5th ed). Auckland: David Bateman Ltd. Richardson, G.P. & Pugh, A.L. ( 1 98 1 ). Introduction to System Dynamics Modelling with DYNAMO. London: MIT Press. Ricklefs, R.E. ( 1 990). Ecology (3rd ed.) . New York: W.H. Freeman & Co. Rima, I.H. ( 1 986). Development of Economic Analysis (4th ed.). Homewood, Il l inois: Richard D. Unwin. Ritzer, G. ( 1 975). Sociology: A Multiple Paradigm Science. Boston : Al lyn and Bacon. Robinson, T.J .C. ( 1 989). Economic Theories of Exhaustible Resources. London and New York: Routledge. Robinson, W.A. (200 1 ). Modeling Dynamic Climate Systems. New York: Springer. Rogers, E. ( 1 995). The Diffussion of Innovation, 4th Ed. New York: Free Press. 45 1 Romer, P.M. ( 1 986). Increasing returns and long run growth. Journal of Political Economy, 94: 1 002- l O37. Romer, P.M. ( 1 990). Endogenous Technological Change. Journal of Political Economy, 98: S7 1 --S 1 02. Roseland, M. ( 1 992). Towards sustainable communities. Ottawa: Canadian National Round Table on the Environment and the Economy. Rosenberg, N. ( 1 982). Inside the Black Box: Technology and Economics. Cambridge : Cambridge University Press. Rotty, M .R., & Masters, D.e. ( 1 985). Carbon Dioxide from Fossil Fuel Combustion: Trends, Resources, and Technological Implications. Atmospheric Carbon D ioxide and the Global Carbon Cycle. DOE/ER-0239. Washington, DC : US Department of Energy. Roughgarden, J. ( 1 979). Theory of Population Genetics and Evolutionary Biology: An Introduction. New York: Macmillan. Round, J.1. ( 1 983) Non-survey techniques: A critical review of the theory and evidence. International Regional Science Review 8: 1 89-2 1 2 . Ruth, M . ( 1 993). Integrating Economics, Ecology and Thermodynamics. Boston and London: K luwer Academic . Ruth, M. & C leveland, e . ( 1 994). Modelling the dynamics of resource depletion, substitution, recyling and technical change in extractive industries. Paper presented to the ISEE Conference in Costa Rica. Ruth, M. (2002). Optimal Resource Extraction. In R.U. Ayres & L.W. Ayres (Eds.), A Handbook of Industrial Ecology. Cheltenham and Northampton MA: Edward Elgar. Ruth, M. , Bernier, e. Jollands, N. and Golubiewski, N. (2005). Adaptation of urban water supply infrastructure to impacts from climate change and socioeconomic changes : the case of Hamilton, New Zealand. Submitted to Journal of Environmental Management. 452 Ruth, M. & Hannon, B. ( 1 997). Modeling Dynamic Economic Systems. New York, Berlin, Heidelberg: Springer-Verlag. Ryan, G.J. ( 1 995). Dynamic physical analysis of long term economy-environment options. Unpublished doctoral dissertation, University of Canterbury, Christchurch. Salvadori, N. (Ed) (2003). Old and New Growth Theories: An Assessment. Cheltenham, UK: Edward Elgar. Samuelson, P. & Nordhaus, W. ( 1 985). Economics ( 1 2th ed.). New York: McGraw-Hill . Samuelson, P.A. & Nordhaus, W.D. ( 1 989). Economics ( 1 3th Ed). New York: McGraw-Hil l . Sassen, S. ( 1 99 1 ). The Global City: New York, Tokyo, and London. Princeton: Princeton University Press. Saunders, P. ( 1 98 1 ). Social Theory and the Urban Theory. London : Hutchinson. Saunders, P . (200 1 ) . Urban Ecology. In R. Paddison (Ed.), Handbook of Urban Studies (pp. 36- 5 1 ). London: SAGE Publications. Sawyer, C .H. & Mil ler, R.E. ( 1 983). Experiments in regionalisation of a national i nput-output table. Environment and Planning A, 15: 1 5 0 1 - 1 520. Schaffer, W.A. & Chu, K. ( 1 969). Nonsurvey techniques for constructing regional interindustry models. Papers and Proceedings, Regional Science Association 23:83- 1 0 1 . Schaffer, W.A. & Chu, K. ( 1 97 1 ). Simulating regional interindustry models/or Western States. Paper presented at the Pacific Regional Science Conference, Honolulu, Hawaii, August 26-29. Papers of the F irst Pacific Regional Science Conference, 1 969. Tokyo: University of Tokyo Press. Schank, R.D. & Abelson R. ( 1 977). Scripts, Plans, Goals, and Understanding. Hillsdale, New Jersey: Erlbaum. Schlesinger, W.H. ( 1 99 1 ) . Biogeochemistry: An Analysis of Global Change. San Diego: Academic Press. 453 Schmidt-Bleek, F. ( 1 994a). MIPsbook or The Fossil Makers - Factor la and more. Wuppertal: Wuppertal Institute for Climate, Environment and Energy. Schmidt-Bleek, F.B. ( 1994b). Where We Stand Now: Actions Toward Reaching a Dematerali::ed Economy. Declaration of the First Meeting of the Factor 1 0 Club held in Carnoules, France, September 1 994. Schmidt-Bleek, F.B. ( l 994c). Wieviel Umwelt braucht der Mensch? MIPS, Das Mass fur okologisches Wirtchaften (How Much Environment for Human Needs?). Berlin, Basle: Brikhauser Verlag. Schmidt-Bleek, F.B. ( 1 997). MIPS and Factor 10 for a Sustainable and Profitable Economy. Wuppertal : Wuppertal Institute Schnaiberg, A. ( 1 972). Environmental Sociology and the Division of Labor. Mimeograph: Department of Sociology, Northwestern University. Schnaiberg, A. ( 1 975). Social Syntheses of the Societal-Environmental Dialectic : The Role of Distributional Impacts. Social Science Quarterly, 50: 5-20. Schrodinger, E. ( 1 944). What is Life? Cambridge: Cambridge University Press. Schroeder, W. III, & Strongman, I. ( 1 974). Adapting Urban Dynamics to Lowel!. In Reading in Urban Dynamics. Walthan, MA: Pegasus Communications. Schumacher, E.F. ( 1 973 ) . Small is Beautiful: Economics as Though People Mattered. New York: Harper & Row. Schumpeter, 1.A. ( 1 942). Capitalism, Socialism, and Democracy. New York: Harper and Brothers. Schumpeter, 1.A. ( 1 954). History of Economic Analysis. New York: Oxford University Press. Schwartz, P. ( 1 972). The New Political Economy of Js. Mill. London: Allen and Unwin. 454 Scott, A.D. ( 1 9 55) . Natural Resources: The Economics of Conservation. Toronto: University of Toronto Press. Screpanti, E. & Zamagni, S. ( 1 99 3 ) . An Outline of the History of Economic Thought. Oxford: Clarendon Press. Shampine, L., & Gear, C. ( 1 979). A User's View of Solving Stiff Ordinary Different Equations. SIAM Review, January, 2 1 : l . Shannon, C . E. & Weaver, W. ( 1 949). The Mathematical Theory of Communication. Urbana: Univers ity of Il l inois Press. Shell, K. (Ed.) ( 1 967). Essays on the Theory of Optimal Economic Growth. Cambridge, MA : MIT Press. Shevky, E. & Bell, W. ( 1 955). Social Area Analys is . Stanford, C A : Stanford University Press. Shevky, E. & Williams, M. ( 1 949). The Social Areas of Los A ngeles. Berkeley, Univers ity of California Press. Shiklomanov, LA. ( 1 993 ) . World fresh water resources. In P.H. G le ick (Ed.), Water in crisis: A Gu ide to the World's Fresh Water Resources (pp. 1 3-24). New York: Oxford Univers ity Press. Siddiqi, Y.M. & Salem, M. ( 1 995). Regionalisation of commodity-by-industry input-output accounts: the Canadian Case. Statistics Canada. Siebert, H. ( 1 98 1 ). Economics of the Environment. Massachusetts: Lexington Books. Siebert, H. ( 1 982). Nature as a Life-Support System: Renewable Resources and Environmental Disruption. ZeitschriJt fur Nationalokonomie, 42: 1 3 3 - 1 42 . S iebert, H. ( 1 987). Economics of the Environment (2nd ed.). Berlin: Springer. S iebert, H. ( 1 995). Economics of the Environment: Theory and Policy. Berlin: Springer-Verlag. S immel, G. ( 1 964). The Metropolis and Mental Life . In K. Wolff (Ed.), The Sociology of Georg Simmel (pp. 409-424). New York: Free Press. (Original work published 1 905) 455 Simmie, J . (200 1 ). Innovative Cities. London: Spon Press Taylor & Francis Group. Simon, J.L. & Kahn, H. ( 1 984). The Resourceful Earth. Oxford: Basil B lackwell . Simon, J .L. ( 1 977). The Economics of Population Growth. Princeton, New Jersey: Princeton University Press. S imon, J .L. ( 1 9 8 1 ). The Ultimate Resource. Princeton, New Jersey: Princeton University Press. Skirrow, G. ( 1 975) The dissolved gases - carbon dioxide. In J.P. Riley & G. Skirrow (Eds), Chemical Oceanography: Vol. 2 (2nd ed.) (pp. l - l92). London, New York, San Francisco: Academic Press. S lesser, M. ( 1 973). Energy Analysis in Policy Making. New Scientist, November: 323-330. Smil, V. ( 1 997). Cycles of Life: Civili=ation and the Biosphere. New York: Scientific American L ibrary. Smil, V (2000) Phosphorus in the Environment. Annual Review of Energy and the Environment, 25: 53-88 Smil, V. (2002) . Global biogeochemical cycles. In R.U. Ayres & L.W. Ayres (Eds.), A Handbook of Industrial Ecology (pp.249-259). Cheltenham: Edward Elgar Publishing Ltd. Smith, A. ( 1 953) . An Inquiry into the Nature and Causes of the Wealth of Nations. London: J.M. Dent & Sons. (Original work published in 1 776) Smith, P. & Morrison, W.I. ( 1 974). Simulating the urban economy: experiments with input­ output techniques. London: P ion L imited. Smith V.K. ( 1 98 1 ) . The empirical relevance of HoteIIing's model for natural resources. Resources and Energy, 3: 1 05-1 1 7 . Society of Environmental Chemistry and Toxicology (SECAT). ( 1 993). A Technical Framework for Lifecycle Assessments. Society of Environmental Chemistry and Toxicology. 456 Socolow, R.H. ( 1 994). S ix perspectives from industrial ecology. In RH. Socolow, c.J. Andrews, F.G. Berkhout, & V .M. Thomas (Eds.), Industrial Ecology and Global Change (pp. 3- 1 6) . Cambridge, OK: Cambridge University Press. Soderlund, R & Svensson, RH. ( 1 976). The Global Nitrogen Cycle. In Nitrogen Phosphorus & Sulfur: Global Cycles (pp.23 -74). 22 (SCOPE Report 7), Ecology Bul letin, Stockholm. Soleri, P . ( 1 969). Arcology: The City in the Image of Man. Cambridge, MA: MIT Press. Solow, RM. ( 1 956). A contribution to the theory of economic growth. Quarterly Journal of Economics, 70: 65-94. Solow, R.M. ( 1 957) . Technical Change and the Aggregate Production Function. Review of Economics and Statistics, 39: 3 1 2-320. Solow, R.M. ( 1 974). Intergenerational Equity and Exhaustible Resources. Review of Economic Studies, Symposium: 29-46. So low, R.M. ( 1 986). On the Intergenerational Allocation of Natural Resources. Scandinavian Journal of Economics, 88( 1 ) : 1 4 1 - 1 49. Solow, R.M. ( 1 988). Growth Theory. An exposition. New York: Oxford University Press. Solow R.M. (2000). Growth Theory. An exposition. (2nd ed.). New York: Oxford University Press. Sowell , T. ( 1 985). Marxism: Philosophy and Economics. New York: William Morrow. Spaling, H. ( 1 994). Evaluation of Methods for Cumulative Effects Assessment. In R.E. Munn (Ed.), Looking Ahead: The Inclusion of Long-Term 'Global ' Futures in Cumulative Environmental Assessments (pp. 1 99-2 1 4) . Environmental Monograph No. 1 l . Toronto, Ontario: Institute for Environmental Studies, University of Toronto. Spence, M. ( 1976). Product Selection, Fixed Costs, and Monopolistic Competition. Review of Economic Studies, 43, (June) : 2 1 7-35 . Spencer, H . ( 1 862). First Principles. London: Wil liams and Norpale 457 Spiegel, H.W. ( 1 99 1 ). The Growth of Economic Thought (4th ed.) . North Carolina: Duke University Press, Durham. St. Louis, L.V. ( 1 989). Empirical Tests of Some Semi-Survey Update Procedures Applied to Rectangular Input-Output Tables. Journal of Regional Science, 29(3) : 373-385. Stahmer, C . ( 1988). Umwelt-Satellitensystem zu den Volkswirtschaftlichen Gesamtrechnungen. Allgemeines Statistisches Archiv, 72. Stahmer, C. ( 1 993). Umweltbezogene Erweiterungen der VolkswirtschaftIichen Gesamtrechnung: D ie Konzeption der Vereinten Nationen Mit Input- Output-Anwendungen. In H. Schnabl (ed.), Okointegrative Gesamtrechnung: Ansdt:e, Probleme, Prognosen (pp. 1 1 -62). Berlin and New York: Gruyter. Stahmer, c. , Kuhn, M., & Braun, N. ( 1 996) . Physical Input-output Tables: German Experiences. London Group Meeting on Environmental Accounting, Stockholm, Sweden, May 29-3 1 , 1 996. Stahmer,C. , Kuhn, M., & Braun, N. ( 1 997). Physische Input-Output Tabellen 1 990. Schriftenreihe Beitriige zu den UmweIt okonomischen Gesamtrechnungen. Statisches Bundesamt (ed.), 1 , Wiesbaden. Stahmer, c. , Kuhn, M., & Braun, N. ( 1 998). Physical Input-output Tables for Germany, 1990. Eurostat Working Papers, 211 998/B/l . Brussels: European Commission. Statistics New Zealand. ( 1 99 1 ) . Inter-Industry Study of the New Zealand Economy 1987. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 994). Demographic Trends 199�. Wellington: Statistics New Zealand. Statistics New Zealand. (1 996a). Agriculture Statistics 1996. Wellington: Statistics New Zealand. Statistics New Zealand. (1 996b). Business Directory 1996. Wellington: Statistics New Zealand . 458 Statistics New Zealand. ( l 996c). Census of Population and Dwellings. Wellington: Statistics New Zealand. Statistics New Zealand. ( l 996d). Digital Statistical Boundaries of New Zealand. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 996e ). GDP Series for 25 Sectors from INFOS. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 996 f) . Key Statistics June 1996. Well ington: Statistics New Zealand. Statistics New Zealand. ( 1 998a). Agriculture Statistics 1996: Wellington: Statistics New Zealand. Statistics New Zealand. ( l 99 8b). A Regional Profile - A uckland. Wellington: Statistics New Zealand. Statistics New Zealand. ( l 998c). Business Directory 1998. Wellington: Statistics New Zealand. Statistics New Zealand. ( l 998d). GDP Series for 25 Sectors from INFOS. Well ington: Statistics New Zealand. Statistics New Zealand. ( 1 998e). Harmonised System Statistics 1 998. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 998f). Key Statistics June 1998. Wellington: Statistics ew Zealand. Statistics New Zealand. ( l 998g). Labour Market Statistics 1 998. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 998h). New Zealand Of icial Yearbook 1998: W e l lington: Statistics New Zealand. Statistics New Zealand. ( 1 998i). 1994-95 Inter-Industry Study of the New Zealand Economy. Wellington: Statistics New Zealand. 459 Statistics New Zealand. (1 998j) . Of icial New Zealand Year Book. Wellington : Statistics New Zealand. Statistics New Zealand. (l998k) . Sub-National Population Estimates as at 30 June 1 998. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 999a). Agriculture Statistics. Wellington: Statistics New Zealand. Statistics New Zealand. ( 1 999b). A Regional Profile Auckland. Wellington: Statistics New Zealand. Statistics New Zealand. (1 999c). 1998 Business Directory. Wellington: Statistics New Zealand. Statistics New Zealand. (1 999d). Official New Zealand Year Book. Retrieved from http://www.stats.govt.nzlproducts-and-serv ices/adverts/nz-off-year-book-99.htm Statistics New Zealand. (2000). New Zealand Official Yearbook. Wellington: Statistics New Zealand. Statistics New Zealand. (2000a). Demographic Trends. Wellington: Statistics New Zealand. Statistics New Zealand. (200 1 a) . Agriculture Statistics 1999. Wellington: Statistics New Zealand. Statistics New Zealand. (200 I b). Business Directory 2001 . Wellington: Statistics New Zealand. Statistics New Zealand. (200I c) . Census of Population and Dwellings . Wellington: Statistics New Zealand. Statistics New Zealand. (200 1 d) . GDP Series for 25 Sectors from INFOS. Wellington: Statistics New Zealand. Statistics New Zealand. (200 1 e). Harmonised System Statistics 2001 . Wellington: Statistics New Zealand. Statistics New Zealand. (200 !£). Key Statistics June 2001. Wellington: Statistics New Zealand. 460 Statistics New Zealand. (200 1 g) . New Zealand System of National Accounts. Inter-Industry Study 1996: Interim Release of Tables. Christchurch: Statistics New Zealand. Statistics New Zealand. (2002). Regional Input-Output Study: User Needs Review. Christchurch: Statistics New Zealand. Statistics New Zealand. (2003a). Regional Input-Output Study. Christchurch: Statistics New Zealand. Statistics New Zealand. (2003b). Regional Input-Output Study: Data Evaluation. Christchurch: Statistics New Zealand. Statistics New Zealand. (2003c). Regional Input-Output Study: Methodological Review. Christchurch: Statistics New Zealand. Statistics New Zealand. (2003d). Business Directory 1987-2003. Wellington : Statistics New Zealand. Statistics New Zealand. (2004) . Sub-national Population Projections. Well ington: New Zealand. Statistics New Zealand. (2005). Labour Force Projections . Wellington : New Zealand. Statistics New Zealand Infos Series DFM-DF ASFRNZ, DFM-RASFRNZ, DMM-RASDRSNZ, DMM-RASDRSNZ, FLTA.SBEA3, FLTA. SMAA, FLTA.SMBA, FLTA.SBDA, FLTA.SPFA, FLTA.SPGA, MAGQ.SAB, NRGA.SCR7ZM, NRGA.SGP3AM, PRPH.SAACD, PRPH.SAACE, PRPH.SAACF, SEPA.SAFRZM, SEPA.SATTD, SEPA.SYZ, SEPQ.SABBA, SEPQ.SABBC, SEPQ.SABBD, SEPQ.SABBE, SEPQ.SABBF, SNC (n.d.) Retrieved from http://www.stats.govt.nzJdefault.htm Stedman, D. H. & Shetter, R. E. ( 1 983). The global budget of atmospheric nitrogen species. Advances in Environmental Science and Technology, 12: 4 1 1 -54 Steenge, A.E. ( 1 977). Economic-ecologic analysis: A note on Isard' s Approach. Journal of Regional Science, 1 7: 97-1 05 . 46 1 Sterman, J.D. ( 1 992) . Teach ing Takes Off: Flight Simulators for Management Education. OR/MS Today, October: 1 0-44. Sterman, J .D. ( 1 984). Appropriate Summary for Evaluating the Historical Fit of System Dynamic Models. Dynamica, 1 0: 5 1 -66. Sterman, J.D. ( 1 991 ) . A Sceptics Guide to Computer Models: Managing a Nation. The Microcomputer Software Catalog. Oxford: Westview Press. Sterman, J .D. ( 1 994). Learning in and about complex systems. System Dynamics Review, 10 : 29 1 -330. Steurer, A. ( 1 996). Material F low Accounting and Analysis - Where to go at a European level. In Statistics Sweden (Ed.), Third meeting of the London Group on Natural Resource and Environmental Accounting Proceedings Volume (pp. 2 1 7-22 1 ). Stockholm: Statistics Sweden. Steurer, A. & Schiitz, H. (2000). Economy-wide Material Flow Accounts and Balances with derived Resource Use Indicators: A Methodological Guide. Draft for Public Review. Germany: Eurostat. Stevens, B.H. and Lahr, M.L. ( 1 993) . Sectoral aggregation error in regional input-output models: a simulation study. RSRI Discuss ion Paper No. 1 32, Highstown, NJ: Regional Science Institute. Stevens, B.H. , Treyz, G.I . , & Lahr, M.L. ( 1 989). On the comparative accuracy of RPC estimating techniques. In R.E. Miller, K.E. Polenske, & A.Z. Rose (Eds.), Frontiers of Input­ Output Analysis (pp. 245-257) . New York: Oxford University Press. Stevens, B.H., Treyz, G.1 . , Ehrlich, DJ. , & Bower, J .R. ( 1 983). A New Technique for the Construction of Non-Survey Regional Input-Output Models. International Regional Science Review, 8(3) : 27 1 -286. Stiglitz, J. ( 1 974). Growth with Exhaustible Natural Resources: Efficient and Optimal Growth Paths. Review of Economic Studies, Symposium: 1 23- 1 3 7. Stiglitz, J.E. ( 1 979). A Neoclassical Analysis of the Economics of Natural Resources. In V.K. Smith (Ed.), Scarcity and Growth Reconsidered. London: Johns Hopkins University Press. 462 Stone, R. ( 1 96 1 ). Input-output and National Accounts. Paris : Organization for European Economic Cooperation. Stone, R. ( 1 966). Mathematics in the social sciences, and other essays. Cambridge Massachusetts: M.LT. Press. Stone, R. and Brown, A. ( 1 962). A Programme for Growth, Part 1: A Computable Model of Economic Growth. Andover: Chapman and Hall. Strassert, G. (2000). The German Throughput Economy: Lessons from the First Physical Input­ Output Table (PlOT) for Germany. In J .c. Dragan, E.K. Seifert, G. Strassert, M.C. Demetrescu. & Constantin (eds.), Cybernetics, Ecology and Bioeconomics (pp. 3 1 4-33 0). International Joint Conference of the Cybernetics Academy, Stefan Odobleja and the European Association for Bioeconomic Studies (EABS), Palma de Mallorca, November 7- 1 0, 1 998 . Milan: Nagard,. Strassert, G. (2002). Physical Input-Output Accounting. In Ayres, R.U. and Ayres, L.W. (Eds. ), A Handbook of Industrial Ecology (pp. 1 02- 1 1 3). Cheltenham: Edward Elgar. Street, A. ( 1 998). Auckland Waste Analysis 1997. Unpublished Report. L incoln: AgFirst Consultants Environmental Ltd. Street, D. & Associates. ( 1 978). Handbook of Contemporary Urban Life . San Francisco: Jossey Bass. Stumm, W. ( 1 973). The Acceleration of the Hydrogeochemical Cycling of Phosphorus. Water Research, 7: l 3 1 - 1 4 1 . Suttles, G. ( 1 968). The Social Order of the Slum. Chicago: University o f Chicago Press. Suttles, G. ( 1 972). The Social Construction of Communities. Chicago: University of Chicago Press. Swan, T.W. ( 1 956). Economic Growth and Capital Accumulation. Economic Record, 32: 3 34- 36 1 . 463 Tansley, A .G. ( 1 935). The use and abuse of vegetational concepts and terms. Ecology, 16(3): 284-307 . ten Raa, T . , Chakraborty, D. , & Small, J .A . ( 1984) . An Alternative Approach of Negatives in Input-Output Analysis. Review of Economics and Statistics, 66: 88-97. Tennet, R.M. (Ed.) ( 1 993). Science Data book. Essex: Oilver & Boyd. Thiel, H. ( 1 967). Economics and Information Theory. Amsterdam: North-Holland. Thill, J .c. & Sui, D.Z. ( 1 993). Mental Maps and Fuzziness m Space Preferences. The Professional Geographer, 45(August) : 264-276. ThirlwalI, A.P. (2002). The Nature of Economic Growth: An Alternative Framework for Understanding the Performance of Nations. Cheltenham, UK: Edward Elgar. Thorn, R. ( 1 975) . Structural Stability and Morphogenesis. Reading MA: Benjamin. Thompson, A.M. & Cicerone, RJ. ( 1 986). Possible perturbations to atmosphere CO, CH4, and OH. Journal of Geophysical Research, 92: 1 0853-1 0864. Thorns, D.e. (2002). The Transformation of Cities: Urban Theory and Urban Life. New York: Pal grave Macrnillan. Tiessen, H. ( 1 995). Phosphorus in the Global Environment: Transfers, Cycles and Management. Chichester: John Wiley. Tisdell, C. ( 1 990). National Resources, Growth and Development: Economics, Ecology and Resource Scarcity. New York: Praeger. TjalIingii, S.P. ( 1 993). Ecopolis: Strategies for Ecologically Sound Urban Development. Leiden: Backhuys Publishers. T6nnies, F. ( 1 963). Community and Society [Gemeinschaft and Gesellschaft]. New York: Harper & Row. (Original work published 1 887) 464 Treyz, G.!. & Petragl ia, L. (200 1 ). Consumption equations for a multiregional forecasting and pol icy analys is model. In M . L . Lahr & R.W. Miller (Eds.), Regional Science Perspectives in Economic A nalysis: A FestschriJt in Memory of Benjamin H. Stevens (pp.28 7-300). Amsterdam: Elsevier Sc ience. Treyz, G.!. & Stevens, B.H. ( 1 985). The TFS Regional Mode lling Methodology. Regional Studies, 19: 547-562. Turner, R.K. (Ed.). ( 1 9 88). Sustainable Environmental Economics and Management: Principles and Practice. London: Be lhaven Press. Turner, R.K. (Ed. ). ( 1 993 ). Sustainable Environmental Economics and Management: Principles and Practice (2nd ed.). London: Belhaven Press. Turner, R. K., Doktor, P. & Adger, N. ( 1 994). Sea-Level Rise and Coastal Wetlands in the U.K.: Mitigation Strategies for Sustainable Management. In A. Jansson, M. Hammer, C . Fo lke, & R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability. Washington D.C. : Island Press. U lanowicz, R.E. ( 1 99 1 ). Contributory Values of Ecosystem Resources. In R. Costanza (Ed.), Ecological Economics. The Science and Management of Sustainab ility (pp. 2 5 3 -268). ew York: Columbia Univers ity Press. UNCED. ( 1 992). Rio Earth Summit Conference on Environment and Development. New York: UNCED. UNFCCC (United Nations Framework Convention on C limate Change). ( 1 997). Kyoto Protocol to the United Nations Framework Convention on Climate Change. United Nations Framework Convention on C limate Change, Kyoto, Japan, 1 to 1 1 December 1 997. United Nations. ( 1 968). A System of National A ccounts. Studies i n Methods . Series F, No. 2. Paris: United Nations. United Nations. ( 1 993a). Integrated Environmental and Economic A ccounting. Handbook of National Accounting: Studies in Methods. Series F, No. 6 l . New York: United Nations. United Nations. ( 1 993b). System of National Acco unts. New York: Un ited Nations. 465 United Nations. ( 1 999).Handbook of Input-Output Table Compilation and Analysis. Handbook of National Accounting: Studies in Methods. Series F, No. 74. New York: United Nations. United Nations. (2002). Report of the World Summit on Sustainable Development. United Nations Conference, Johannesburg, South Africa, 26 August to 4 September 2002. United Nations. (2003 ). Integrated Environmental and Economic Accounting 2003 (Final Draft)· United Nations Statistical Commission. (2002). System of Integrated Environmental and Economic Accounting (Draft SEEA 2002) . Draft report of the London Group on Environmental Accounting. V.S. Department of Agriculture. ( 1 974). Wood Handbook: Wood as an engineering material. Agriculture Handbook No. 72. V.S. Government Printing Service. Uzawa, H. ( 1 965). Optimal technical change III an aggregate model of economic growth. International Economic Review, 6: 1 8-3 1 Van den Bergh, J.C.J.M. & Verbruggen, H. ( 1 999). Spatial Sustainability, Trade and Indicators: an Evaluation of the 'Ecological Footprint ' . Ecological Economics, 29( 1 ) : 6 1 -72. van Gigch, J.P. ( 1 978). Applied General System Theory. New York: Harper & Row. van Vuuren, D.P. & Smeets, E .M.W. (2000). Ecological Footprints of Benin, Bhutan, Costa Rica and the Netherlands. Ecological Economics, 34( 1 ) : 1 1 5- 1 3 0. Vensim® Reference Manual ( 1999). Belmont: Ventana Systems Inc. Verbrugge, L.M. & Taylor, R.B. ( 1 980). Consequences of Population Density and S ize. Urban Affairs Quarterly, 16: 1 35- 1 60. Verdoorn, P.J. ( 1 980). Verdoorn's Law in Retrospect: A Comment. Economic Journal, 90 (June 1 980): 382-385 . 466 Verdoorn, PJ. ( 1 993) . On the factors determining the growth of labour productivity. Italian Economic Papers, 2: 59-68. Originally published: Verdoom, PJ . ( 1 949). Fattori che regolano 10 sviluppo della produttivita del lavoro. L 'Industria, 1 : 3 - 1 0. Verfail l ie, H.A. & Bidwell, R. (2000). Measuring Eco-ef iciency - A Guide to Reporting Company Performance. Geneva: World Business Council for Sustainable Development. Victor, P.A. ( 1 972a). Economics of Pollution. London: Macmi llan. Victor, P.A. ( 1 972b). Pollution: economy and environment. London: George Al Ien and Unwin Ltd. Victor, P.A. ( 1 99 1 ). Indicators of Sustainable Development: Some Lessons from Capital Theory. Ecological Economics, 4: 1 9 1 -2 1 3 . Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., & Matson, P.A. ( 1 986). Human Appropriation of the Products of Photosynthesis. Bioscience, 36: 368-373 . Vitousek, P.M., & Howarth, R . ( 1 99 1 ) . Nitrogen l imitation on land and in the sea: how can it occur? Biogeochemistry, 13: 87- 1 1 5 . Vollebergh, H.RJ. and Kemfert, C . (2005). The role o f technological change for a sustainable development. Ecological Economics, 54, 1 33 - 147 . Von Bertalanffy, L. ( 1 968). General Systems Theory: Foundations, Development, Applications. New York: George Braziller. Wackernagel, M. ( 1 99 1 ). Using 'appropriated carrying capacity ' as an indicator: measuring the sustainability of a community. Report for the UBC Task Force on Healthy and Sustainable Communities. Vancouver: UBC School of Community and Regional Planning. Wackernagel, M., Onisto, L., Bel lo, P., Callejas L inares, A. , Lopez Falfan, I .S . , Mendez Garcia, J., et al. ( 1 999). National Natural Capital Accounting with the Ecological Footprint Concept. Ecological Economics, 29: 3 75-390. Wackernagel, M. & Rees, W.E. ( 1 996). Our Ecological Footprint: Reducing Human Impact on the Earth. Philadelphia: New Society Publishers. 467 Wackernagel , M. , Schulz, N.B., Deumling, D., Linares, A.c., Jenkins, M., Kapos, V., et al. (2002). Tracking the ecological overshoot of the human economy. Proceedings of the National Academy of Sciences, 99( 1 4) : 9266-927 1 . Wackernagel, M. & Silverstein, J . (2000). Big Things First: Focusing on the Scale Imperative with the Ecological Footprint. Ecological Economics, 32: 39 1 -394. Walker, B.H. ( 1992). Biodiversity and Ecological Redundancy. Conservation Biology, 6: 1 9- 23 . Wallerstein, I . (1 979). The Capitalist World-Economy. New York: Cambridge University Press. Walras, L. ( 1 954). Elements of Pure Economics . (W. Jaffe (Ed.), Trans.). Homewood, Illinois: Richard D. Irwin. (Original work published 1 874) Walton, J. ( 1 98 1 ). The New Urban Sociology. International Social Science Journal, 33: 3 74- 90. Waring, R.H., Rogers, J.J., & Swank, W.T. ( 1 98 1 ). Water relations and hydrological cycles . In D.E. Reichle (Ed.), Dynamic Properties of Forest Ecosystems (pp. 205-264). Cambridge: Cambridge University. Warner, R. S. ( 1 997). A Paradigm Is Not a Theory: Reply to Lechner. American Journal of Sociology, 1 03: 1 92- 1 98. Warren-Rhodes, K. & Koenig, A. (200 1 ). Escalating trends in the urban metabolism of Hong Kong: 1 97 1 - 1 997 . Ambio, 30(7): 429-43 8. Watson, A. J. & Liss, P.S. ( 1 998). Marine biological controls on c limate via the carbon and sulphur geochemical cycles. Philosophical Transactions of the Royal Society of London, 353: 4 1 -5 1 . Watson, R.T., Rodhe, H., Oeschger, H. , & S iegenthaler, U . ( 1 990). Greenhouse gases and aerosols. In J.T. Houghton, G.J. Jenkins, & J.J. Emphraums (Eds.), Climate Change: the IPCC Scientific Assessment. Intergovernmental Panel on Climate Change (IPCC). Cambridge : Cambridge University Press. 468 Watt, K.E. ( 1 973) . Principles of Environmental Science. New York: McGraw Hil l . WBCSD (World Business Counci l for Sustainable Development). (2000). Eco-efficiency: creating more value with less impact. Retrieved 30 August 2005 from http://www.wbcsd.orgIDocRoot/02w8IK1 4V8E3HMIiFYue/eco _efficiency _ creating_more _ val ue .pdf WCED (World Commission on Environment and Development). ( 1 987). Our Common Future. New York: Oxford University Press. Weber, M. ( 1 966). The City. New York: Free Press. Wedepohl, K.H. (Ed.) ( 1 969). Handbook ofGeochemistry (Vol . 1 ) . New York: Springer-Verlag. Weinstein, M.C. & Zeckhauser, R.J . ( 1 975) . The Optimal Consumption of Depletable Natural Resources. Quarterly Journal of Economics, 89: 3 7 1 -392. Weizsacker, E. , Lovins, A.B. & Lovins, L.H. ( 1 997). Factor Four Doubling Wealth - Halving Resource Use: The New Report to the Club of Rome. London: Earthscan. West, G.R. ( 1990). Regional trade estimation: A hybrid approach. International Regional Science Review 13: 1 03 - 1 1 8 . West, G.R., Wilkinson, J.T. & Jensen, R.C. ( 1 980) . Generation of Regional Input-Output Tables for the Northern Territory GRIT If. Report to the Northern Territory Department of the Chief Minister. St. Lucia, Queensland: Department of Economics, University of Queensland. Westman, W.E. ( 1990). Managing for Biodiversity : Unresolved Science and Pol icy Questions. Bioscience, 40: 26-33 . Wetzel, K.R. ( 1 995). Sizing the Earth: Recognition of Economic Carrying Capacity. Ecological Economics, 12: 1 3-2 1 . Weyant lP. & Olavson T . ( 1 999). Issues in modeling induced technological change in energy, environmental, and c limate policy. Environmental Modeling and Assessment, 4: 67-86. 469 Whittaker, R.H. 1 975. Communities and ecosystems (2nd ed.). New York: MacMillan Publishing Co. Inc. Whyte, W.F. ( 1 943). Street Corner Society: The Social Structure of an Italian Slum. Chicago: University of Chicago Press. Wicken, J. ( 1 988). Thermodynamics, Evolution and Emergence. In B .H. Weber, D.J. Depew, & J.D. Smith (Eds.), Entropy, Information and Evolution. Cambridge MA: MlT Press. Wiens, J .A. ( 1 984). On Understanding a Non-Equi librium World: Myth and Reality in Community Patterns and Processes. In D.R. Strong Jr., D. S imberloff, L.G. Abele, & A .B. Thistle et al. (Eds.), Ecological Communities: Conceptual Issues and Evidence (pp. 439-457). Princeton, NJ: Princeton University Press. W il liams, P.J. Le B. (1 975) . B iological and chemical aspects of dissolved organic material in sea-water. In I.P. Riley and G. Skirrow (Eds.), Chemical Oceanography; Vol. 2 (2nd ed.) (pp. 3 0 1 -363). London: Academic Press. Wilson, E.O. ( 1 988). The Current State of Biological Diversity. In E.O. Wilson (Ed.), Biodiversity. Washington: National Academy Press. Wirth, L. ( 1 93 8). Urbanism as a Way of Life. American Journal o/Sociology, 44: 1 -24. Wirth-Nesher, H. (200 1 ). Impartial Maps: Reading and Writing C ities. In R. Paddison (Ed.), Handbook of Urban Studies (pp. 52-66). London: SAGE Publications. Wollast, R., Mackenzie, F.T., & Chou, L. (Eds.) ( 1 993). Interactions of C, N, P and S Biogeochemical Cycles and Global Change. Berlin: Springer-Verlag. Wolman, A . ( 1 965). The metabol ism of cities. Scientific American, 213(3): 1 79- 1 90. Wolstenholme, E . ( 1990). System Enquiry: A System Dynamics Approach. Chichester: John Wiley & Sons Ltd. Works Consultancy Services Ltd. ( 1 996). Transit New Zealand National Traffic Database. Contents and Operation 0/ Database. Transit New Zealand Research Report No.53 . Well ington: Transit New Zealand. 470 Wright, D.J. ( 1 975). The Natural Resource Requirements of Commodities. Applied Economics, 7: 3 1 -39. Wright, F .L. ( 1 958). The Living City. New York: Horizon. Young, A. ( 1 993). Invention and Bounded Learning by Doing. Journal of Political Economy, 101(3): 443-72. Young, A. ( 1 995). The Tyranny of Numbers: Confronting the Statistical Realities of the East Asian Growth Experience . Quarterly Journal of Economics, 1 10: 54 1 -80. Young, M.D. ( 1 992). Sustainable Investment and Resource Use: Equity, Environmental Integrity and Economic Ef iciency. Carnforth: Parthenon. Zanders, J. (2002). Comparison of EcoLink point source water pollutants with instream water pollutants for the Waikato Region. Pa1merston North: Landcare Research Ltd. Zaret, T.M. ( 1 982). The Stabi litylDiversity Controversy: A Test of Hypotheses . Ecology, 63: 72 1-73 1 . Zucchetto, J. & J ansson, A . ( 1 979). Integrated regional energy analysis for the island of Gotland, Sweden. Environment and Planning A, 1 1 : 91 9-942. ApPENDICES Appendix A Input-Output Analysis: History, Mathematics and Assumptions A.I Brief History of Input-Output Modelling 475 The origins of input-output model ling may be traced back to the Physiocrats of the 1 8th Century. Fran + 208°2 + N2 3 C land plant respiration C7soH,S((,o7SON'SS4P ", 782.502 --+ 750C02 + 750H20 + 15 N03- + 4 sot ... PO/" 4 C tropospheric oXidation of CO 2CO + O2 ..... 2C02 5 C release of CO2 HCO/- ... H+ ....... CO2 ,,, H20 6 C absorption of CO2 CO2 ,,, H20 -- HC03" + W 7 C gross land production 750C02 + 750H20 + 15N03" + 4S0t + PO/, -- C?SOH'SOCP150N'SS4P + 782.502 8 C production of CO C'OOH2S00'40N,5SP ", 70 5°2 '" H· -- 100CO + 125H20 -+ 1 6N03 + 804 + HP04 9 C oXidation of CH4 CH" ... 1.502 - CO + 2HzO 10 C sorption of CO 800eO ... 500H20 + 32N03 ... 4804 -+ P04 ---> C800HlOCI(P800Nn54P -+ 40802 1 1 C production of CH4 C250Hsoc,o,ooNsoPJS '" 125H20 ---> 150CH4 '" 100C02 ... 50NH3 -+ 3P04 -+ 804 -+ 4 502 12 C land humus formation (p) 2C7SOH,soo0750N,sP84 -+ 35302 -+ 18NO� --> 1 5CaooH,ol):,oaooS4P + 300C02 + 750H20 ... 0 5P04 -+ 2 504 13 C land consumption C750H,soo0750N,sS4P ", 1 35N03 + 8P04 ---> 3C2SOHsooO'OONSOPJ8 ... 504 + 441 5°2 14 C gross manne productIOn 100 HC03• ... 4NO)' ", HPO/ -+ 50/ -+ 25H20 ---> C'OOH'50090N4SP ", 132 502 -+ H+ 15 C manne plant reSpiration C,ooH,soOsoN48P -+ 132 5°2 -+ H" ---> 100HC03' + 4N03' -+- HP04 3. -+- 80/ -+- 25H20 16 C productton of CaCO) C�ooH�soOeoN�8P '" 157 502 -+ 100Ca]<. '" H· ...... 1 00CaC03 -+ 4NO; .,. HPO/' ''' SO/- -+ 75H20 17 C manne humus formation (p) C,ooH,soOsoN48P -+ 9°2 ...... C'OOH'200S0N�PS + 15H20 '" NO, 18 C manne consumption C,OOH'SO080N48P .,. 50H20 -+ 12N03 - C,ooH2SOO'40N,SSP -+ 1302 19 C land humus formation (c&d) 4C:�soHsooO,ooN50P3S + 8240:z ...... C900H'oo:P600N32S4P -+ 200C02 -+ 1 68N03 -+ SOOH20 -+ 1 1 P04 20 C marine humus formation (c&d) C,ooH2SOO'40N,SSP .,. 22°2 ...... C'OOH'200SON3PS + 13NO) '" 6SH20 21 C coal formation C800H'00J0eooNnS4P -+ 280/ -+ 6C'JOH" oO'oN2S -+ 20CO '" 170H20 '" iON, -+ 177°2 ", P04 22 C transfer of land humus C800H,oooOsooNnS4P -+ 7HP04 -+ 4804 .,. 2002 ...... 8C,ooH,200soN3P8 -+ 8N03 + 20HzO + 7H+ 23 C oXidation of marine humus C'OOH'200SON3P8 -+ 104H20 -+ 6002"" 100HC03' -+ 108Hz -+ HPO/' ''' H28 + 3NH3 24 C kerogen formation 7C,ooH1200soN3PS -+ 280H20 -+ 2180/' -+ 413°2 ...... C700H,4O(I0'7S0526P7N,4 . 3 SN2 25 C formatIOn of limestone CaCO, -+ CaCO, 26 C weathenng of limestone CaCO, .,. H+ -+ HCO, -+ Ca2< 27 C Igneous rock formation C7ooH,400 1750S2SPIN14 '" 1 83°2 " 25HPO/ -+ C'OOH'400 '0I)oPn82QN2 -+ 8S02 '" 6N2 .,. 600e02 -+ 25H< 28 N reducl!On of N20 2NzO -+ 2N2 ... 02 29 N oXidatIOn of N2 2N2 '" 02 - 2NzO 30 N manne dentrlflcatlon 2NO� + 12H ... ...... N2 '" 6H20 31 N NOx formatIOn by lightning 2N2 '" 3°2 ..... 2NO -+ 2N02 32 N biological fixation by phytoplankton 400C02 -+ 300H20 -+ 4804 -+ 4HP04 ... NzO -+ 7N, ..... 4C,ooH,soOaoN4SP -+ 406 502 ... 4W 33 N release of N2 & N20 4C,ooH,soOaoN4SP -+ 4H+ .,. 406 5°2 "" 400C02 .,. 300HzO + 4804 ", 4HP04 + N;p + 7N, 34 N microbial production of N2 & N20 C800HlI»P500N�2S4P -+ 635 502 ..... 800G02 -+ 248H20 + 1 3N2 .,. 3Np .,. PO/, -+ 450, .,. 504W 35 N uptaKe of N2 & N20 aoocoz '" 248HzO -+ 3N.p .,. 13N2'" PO/, -+ 4S0� '" 504H· - C&;()H,oooO&;()N:uS�P + 635 50, 36 N photochemical oXidation of NzO Np .,. 02 -- NO -+ N02 37 N wet depOSItion of NOx NO, s + 0 7502 ..... N03 38 N aCid rain NO,s -+ 0 7502 -- NO, 39 N microbial production of NH3 CSOOH'OO"J0600N3284P .,. 364HzO -- 480C02 -+ 320CH4 ... 32NH3 -+ 172H2 + 4H,S + PO, 40 N uptake of NH3 by sOil 480C02 -+ 320CH, .,. 32NH, .. 172H2 -+ 4H28 -+ P04 ---> C&lQH'OOO0600N3254P .. 364HzO 41 N uptake of NH3 by ocean NH3 (gas) ...... NH3 (dissolved) 42 N N runoff NO)- (SOil) -+ NO)' (ocean) 43 N sedlmenta.ry N rock formation N03' -+ 4H'" ..... NH4 < .,. 1 502 44 N weathering of sedimentary N rock NH,+ -+ 1.5°2 ...... N03 -+ 4H· 45 5 release of H25 C'OOH'500SON,SP -+ 28H20 -+ 48CH4 ... 52G02 -+ H2S .,. 4NH3 -+ P043. 46 5 ocean spray of SO, SO/" - 802 .,. 02 47 5 uptake of S02 by ocean S02 + 02 - SO/ 48 8 uptake ofS02 by soil S02 -+ 02 _ 50/- 49 8 ocean spray to land SO/" ...... SO/- 50 S S runoff SO/--+ 80/ 51 8 5 sedimentation SO/ ...... 80/ 52 8 sedimentary 5 rock formatIOn 5°4 """ S2- + 202 53 8 uplift & weathering of sedimentary S rock 5 2. + 2°2 """ 80,2. 54 S 1-\2S dlSSolutlon HzS ... 4H20 -- 80/ "\' 2H+ -\" 4H2 55 P weathering of sedimentary P rock Cas(P04h(OH) (sediment) ....... 5Ca 2< + 3P043- .,. OH 56 P 5011 to ocean P sedlments 5Ca2+ -+ 3PO/' + OH---> Cas{PO,h(OH) 57 P P runoff PO/- ... H+ ....... HPO/ 58 P evaporallon of soluble P 3HPO/- -+ 5Caz+ -+ 30H- -+ Cas(PO,):rt0H) -+ 2H20 '" H+ 59 P depOSItion of soluble P Cas(PO,h(OH) + 2H20 -+ H+ ..... 3HPO/ -+ 5Caz+ .,. 30H- 60 P depOSition of Insoluble P Cas(P04h(OH) (atmosphere) """ Cas(PO.b(OH) (sedlments) 61 P P parnculate flux C'OOH,500soN,8P + 10802 + 3H+ -+ 100C02 + 76H20 + 4N03 -+ SO/- -+ HPO/ 62 P formation of atmosphenc P dust particles 5Ca + OH + 3PO, ...... Cas(P04hOH 63 P dissolution of atmospheriC P dust particles Cas(PO_hOH --> 5Ca '" OH -+ 3P04 64 P P sedimentation 6 HPO/ + 1 Oea2+ + 80H- ..... 2Ca5(PO�h(OH) -+ 6Hp 65 P sedimentary P rock formation 66 H tranSpiration 67 H evaporabon from ocean 68 H preclprtation to ocean 69 H preclpltatlon to land 70 H evaporation from land 71 H uptake of Hp 72 H photoQ)odallon of H2 73 H river diSCharge Ca",(P04h{OH) (manne sedlments) -> Ca.,.{P04h{0H) (uplifted sedlments) H20 (liqUid) -+ H20 (gas) H20 (liqUid) -- H20 (gas) H20 (gas) ...... H20 (liquid) H20 (gas) - H20 (llqula) H20 (liqUid) -- Hp (gas) H20 (liqUid) -- HzO (liqUid) H2 -+ 0 50z - H20 H20 (liqUid) --H20 (liqUid) MarKer Element Mass Pg 40 1885 0.0082 90 0033 1 .4820 64 9949 70.1082 142 5052 0 1 100 1 5460 0 1900 0.5612 45 0966 0 961 1 40.7666 1 1 9990 1 0999 23 5986 0 3800 0 4000 'Cl 2700 0 0075 0.4000 20 9983 0 5000 0 3500 0 3500 0 0006 0.0300 0 0150 0 0270 0.0100 0.0400 0 0400 0 2425 0 1847 0 0050 0 0150 0 0130 0 0840 0 0540 0 .1396 0 0320 0.0150 0.0050 0 0550 0.0440 0.0960 0 1 160 0.0040 0 1 329 0 0550 0 0550 0 0470 C 4606 0.0214 0 0187 0.0180 0 0003 0 0005 0 0008 0.0420 0 0031 0.0021 0.0019 0.0214 3,000.0587 43,000.8411 39,000.7629 1 1 ,004 6122 4,000.0782 3,0000587 3.9864 3,955.4374 Table B.1 Biosphere Inputs Into Processes (Continued) BlogeochemlcaJ Process 1 C oXidation of land humus 2 C volcanic action 3 C land plant respiratIOn 4 C tropospheric oXidation of CO 5 C release of C� 6 C absorption of CO2 7 C gross land production 8 C production of CO 9 C oXidation of CH� 1 0 C sorption of CO 11 C production of CH. 1 2 C land humus formation (p) 1 3 C land consumptIOn 14 C gross manne production 15 C manne plant reSpiration le C production of Ca CO; 17 C marme humus formation (p) , 8 C manne consumption 19 C land humus formation (c&d) 20 C manne humus formation (c&d) 21 C coal formation 22 C transfer of land humus 23 C OXidation of manne humus 24 C Kerogen formation 25 C formation of I)mestone 26 C weathenng of limestone 27 C Igneous rock formation 28 N reduction of N]O 29 N OXidation of N2 30 N manne dentnficatlon 3 1 N NOx formation by lightning 32 N biological fixation by Phytoplanl5 land proQuctmn 8 C production of CO 9 C oXidation of CH4 10 C sorption of CO i i C prcctl.lc\lon of CH4 12 C land humus formation (p) 1 3 e land consumpllOn 14 C gross marine production 15 C marine plant reSpiration 16 C production ofCaC03 17 C marine humus formation (p) 18 C marine consumption 19 e land hjJmus formatIOn {c&d) 20 C marine humus formaMn (C&d) 21 C coal formation 22 C transfer of land humus 23 C ClXldatVJI1 of marme humus 24 C kerogen formation 25 C formation of limestone 26 C weathering of limestone 27 C Igneous rock formaMn 28 N reduction of N20 29 N oXidation of N2 30 N manne dentnflcatlon 31 N NOx formation by lightning 32 N bIOlogical fixation by phytoplankton 33 N release of N2 & N20 34 N microbial production of N2 & N20 35 N uptake of N2 & Np 36 N photocherl1lcal oXidation of N�O 37 N wet depOSition of NOx 38 N aCId ram 39 N microbial production of NH3 40 N uptake of NH3 by soil 41 N uptake of NH3 by ocean 42 N N runoff 43 N sedimentary N rock formation 44 N weathenng of sedimentary N rock 45 5 release of H2S 46 S ocean spray of S04 47 S uptake of 502 by ocean 48 S uptake of S02 by SO!) 49 S ocean spray to land 50 S S runoff 51 S S sedimentation 52 S sedimentary S rock formation 53 S uplift & weathenng of sedimentary S rock 54 S H2S dissolution 55 P weathenng of sedimentary P rock 56 P SOil to ocean P sedlments 57 P P rI,moff 58 P evaporation of soluble P 59 P deposrtJOn of soluble P 60 P deposition of Insoluble P 61 P P partlculate flux 62 P formation of atmospheriC P dust particles 63 P dissolutiOn of atmospheriC P dust particles 64 P P sedimentation 65 P sedimentary P rock formatiOn 66 H transplfatlon 67 H evaporation from ocean 68 H preCipitation to ocean 69 H preclpttatlon to land 70 H evaporation from land 71 H uptake of H20 72 H phOtooxidatJon of Ht 73 H nver diSCharge 3rd Reactant 15N03 H· 32N03 18N03 8P04 HP043 H· 100Ca2+ 12N03· 4S04 60°2 2180/- 25HP043- 32NH3 OH" 30H H> 3H· 3P04 80H· Ongm Reservoir SOIl Ocean NEe SOil SOil 5011 Ocean NEe Ocean NEe Ocean NEe Ocean NEe Ocean NEe 0, Ocean NEe Ocean NEe Ocean NEe 0, Atmosphere N:P Atmosphere NEC SOil Ocean NEe Atmosphere NEe Ocean NEe 5011 Ocean NEe 4th Reactant Mass Chemical Species Pg 14 7�57 4S0t 0 0001 0 0392 4504 2 7941 0 08 1 1 3 2579 5042. 0 0101 3 6706 H· 0 2354 0 0160 200, 335694 0 1200 413°2 0.0002 0 0666 4HP04 2 3217 0.0544 13N2 0 0657 172H2 0.0034 0.0002 0 0000 0 0041 0.0095 0 0014 5th Reactant Ongm Reservoir Mass Chemical Species Pg Sail 6 0790 pot SOil 0 0076 PO, Ocean NEe 3 2607 25Hp Ocean NEe 0 0009 0, 0 0266 0, 0.7860 Ocean NEe Atmosphere N2 0 1 500 PO/" Atmosphere NEC 0 0418 4H2S Origin ReservOIr Mass Pg So<' 1 5025 5011 0 0019 Ocean Hp 15 2883 0 0079 5011 0 0391 Atmosphere NEe 0 0164 Table B .1 Biosphere Inputs Into Processes (Continued) Slogeochem!ca! Process 1 C OXidation or land humus 2 C volcanic action 3 C land plant reSpiration 4 C tropospheric oxidation of CO 5 C release of CO� 6 C absorption of CO2 7 C gross land production 8 C production of CO 9 C oXidation of CH4 10 C sorption of CO 11 C production of CH .. 12 C land humus formatlQn (P) 13 C land consumption 14 C gross manne production 15 C manne plant respiratIOn 16 C production of CaCOJ 17 C manna humus formation (p) 1 8 C manne consumption 19 C land humus formation (c&d) 20 C manne humus formation (C&d) 21 C coal formation 22 C transfer of land humus 23 C oXidation of manne humus 24 C kerogen formation 25 C formation of limestone 26 C weathering of limestone 27 C Igneous rock formation 28 N reduction of N�O 29 N Olo;\datlon 01 N, 30 N manne dentnflcatlon 31 N NOx formation by lightning 32 N biological fixatIOn by phytoplankton 33 N release of N2 & N20 34 N microbial production of N2 & N20 35 N uptake of N2 & N20 36 N photochemical OXidation of N20 37 N wet depOSition of NOx 38 N aCid rain 39 N microbial productIOn of NH� 40 N uptake of NHJ by soil 41 N uptake of NH3 by ocean 42 N N runoff 43 N sedimentary N rock formatJon 44 N weathenng of sedimentary N rock 45 8 release of H28 46 8 ocean spray of 804 47 S uptake of 802 by ocean 48 S uptake of 802 by SOil 49 S ocean spray to land 50 S 8 runoff 51 8 8 sedimentation 52 S sedimentary S rock formatJon 53 8 uplift & weathenng of sedimentary S rock 54 S H:>8 dissolutIOn 55 P weathering of sedimentary P rock 56 P SOil to ocean P sedlments 57 P P runoff 58 I? e .... a?Oratlon of soluele P 59 P depos!tlon of soluble P 60 P depOSition of Insoluble P 61 P P partlculate flux 62 P formation of atmosphenc P dust particles 63 P diSsolution of atmosphenc P dust particles 64 P P sedimentation 65 P sedimentary P rock formation 66 H transpiration 67 H e .... aporatlOn from ocean 68 H precIpitation to ocean 69 H preCIpitation to land 70 H evaporatlOn from land 71 H uptake of H20 72 H photooxldanon of H2 73 H rIVer discharge 6\1'1 Reactant Chemical Origin ReseIVO!r Species 7N, Atmosphere N2 4804 SOil PO, Sod 7th Reactant Mass Chemical On91n ReservOir Species Pg 0 0350 0 1583 504H· 8011 0 01 1 4 Mass Pg 0 2092 Total Reactant Flux Mass Pg 195 2518 0 1458 478 9071 5 4306 335.6683 365 1747 758 2696 0.5712 8.2437 0 6699 1 5296 145 6658 3 4167 237 3453 69 8590 1 0 8069 59 7190 1 3909 1 .6683 1 0530 0 0164 0 9375 1 1 3 6583 2 3328 2 9166 2 9460 0 0034 0 0471 0 0236 0 1 312 0 0271 4 2866 4 2866 22 2648 16 9529 0 0136 0 0664 0.0575 5 1315 3 2988 0 1698 0 1417 0.0707 0 0236 5 5860 0 1 318 0 2876 0 3475 0 0120 0 3982 0.1648 0 1648 0.1408 1 5249 0 1 157 0 . 1011 0 0558 0 0017 0 0029 0 0043 8.4209 0 0168 0.01 14 0 01 1 4 0 1 157 26,810 0482 384.277 3578 348,530 6268 98,342 8045 35,746 7310 26,810 0482 35 6241 35,347.7974 489 490 Table B .2 Biosphere Outputs From Processes Venslm Processes 1 C oXidation of land humus 2 C volcanic action 3 C land plant respiratIOn 4 C tropospheric OXidation of CO 5 C release of CO2 6 C absorption of CO2 7 C gross land pmductlon 6 C production of CO 9 C oXldatton of CH4 10 C sorption of CO 1 1 C production of CH4 12 C land humus formation (p) 13 C land consumption 14 C gross manne production 15 C marine plant reSpiration 16 C productIOn of CaC03 17 C marine humus formation (p) 18 C marine consumption 19 C land humus formation (c&d) 20 C marine humus formabon (c&d) 21 C coal formation 22 C transfer of land humus 23 C OXidation of manne humus 24 C kerogen formation 25 C formation of limestone 26 C weathering of limestone 27 C Igneous rock formatlOn 28 N reductJon of N20 29 N OXidation of N2 30 N marine dentnficatlOn 31 N NOx formation by lightning 32 N biological fixatiOn by phytoplankton 33 N release of N2 and N20 34 N microbial productIOn of N2 and N20 35 N uptake of N2 and N20 36 N pnotochemlcal OXidation of N20 37 N wet depOSitIOn of NOx 38 N aCid ram 39 N microbial production of NH3 40 N uptake of NH; by SOil 41 N uptaKe of NH3 by ocean 42 N N runoff 43 N sedimentary N rock fOlmaton 44 N weathering of sedimentary N rock 45 5 release of H2S 46 5 ocean spray of S04 47 S uptake of S02 by ocean 48 5 uptake of S02 by SOil 49 S ocean spray to land 50 5 S runoff 51 S S sedimentation 52 S sedimentary S rock formabon Chemical Equation CeooH'CC(10600NnS4P '+ 80802 ..... 800C02 '+ 500H20 '+ 32N03 '+ PO/ '+ 4S0/ C'OOH'.40P'OO(P32S2llN2 '+ 3402 -+ 100C02 '+ 700HP '+ 32PO/ '+ 20502 '+ N2 C750H,sOP7SON,SS.P '+ 782 502 ....... 750C02 " 750H,0 '+ 15 NO,: '+ 4 sol '+ Po/" 2CO '+ O2 --> 2C02 HC032" '+ H· ...... CO2 '' H20 CO2 '' H20 -- HC03" '+ H' 750C02,," 750H'20 '+ 15NO;' '+ 450/' "" PO..," � ClsoH,�0750N'5S4P " 782 502 C,OOH2S00'40N,6SP " 70 50, '+ H+ - 100eo .. 125H20 " 16N03 " S04 " HP04 CH, '+ 1 502 ...... CO '+ 2H20 800CO "" 500HP .. 32N03 '+ 4S04 '+ P04 ..... C800H'CC(10500N32S4P '+ 40802 Cz:,oH:ooO,ooN50P�S " 125H20 � 150CH. '+ 100C02 '+ SONH3 '" 3P04 '+ 804 " 4 502 2CrsoH,soo07SON,5PS. '+ 35302 '+ 18N03 ...... 1 5C800H,000 600S.P ... 300C02 '+ 750HzO ... 0 5P04 .. 2 504 C750H,sO:P750N,SS.P .. 135N03 '+ 8PO. � 3C2SOHsooO,ooNsoP35 '+ SO, '+ 441 502 100 HCOl '+ 4N03 .. HPO/ '+ SO/ '+ 25H20 -+ C'OOH'500SON4SP '+ 132 50. '+ H+ C,ooH,so06(lN.SP '+ 132.502 '" H+ - 100HC03 ... 4N03 '+ HPO/" '+ 50/" '+ 25H20 C'OOH'SO060N,SP '+ 157.502 '+ 100Ca z+ " H+ ..... 100CaCOJ '+ 4N03' '+ HPO/" " 80/" '+ 75HzO C,OCH'SO080N.SP '+ 902 --; C'OOH'2006(lN3PS '+ 15H20 '+ N03 C'OOH'SO080N.SP '+ 50H20 '+ 1 2N03 ..... C,ooH2SOO'40N,SSP '+ 1302 4C2SOH�)p'OONSOP3S '+ 82402 ..... CeooH'OC(l0SOON32S.P " 200C02 '+ 168N03 " 500HzO " 1 1PO. C'OOHZSOO'40N'6SP '+ 2202 ..... C'OOH'20080N3PS '" 13N03 " 65H20 C800H,000 600NnS.P '+ 250/ ...... 6C,30H, 100,oN2S '+ 20CO '+ 170H20 '+ 10N2 '+ 17702 " P04 C800H'CC(l05OQN32S4P '+ 7HP04 '+ 4504 " 2002 --> 8C,ooH,2QOsoNPS " 8N03 '+ 20H20 " 7H' C'OOH'200SON3PS '+ 104H20 '+ 6002 --; 100HCO; '+ 108H. '+ HPO/" '+ H2S '+ 3NH3 7C,00H'2Q06QN3PS '+ 280H20 '+ 21S0/ " 41301 "", CrooH'4oo017SOS2SP7N14 '+ 3 5N2 CaCO) � CaC03 CaC03 '+ H+ --; HCO," '+ Ca2'+ C7ooH'400 1750S2sPrN14 '+ 18302 " 25HPO/" --; C'OOH'400 'OOOPnS2QN2 '+ 8502 " 6N2 '+ 600C02 " 25H· 2N20 ..... 2Nz '+ O2 2N2 '+ O2 --> 2N20 2NO; '+ 12H· ...... N2 '+ 6H20 2N2 ... 302 --+ 2NO '+ 2N02 400C02 '" 300H20 .. 4S0 . .. 4HPO. '+ N20 '+ 7N2 ...... 4C,ooH':'::P&lN.5P ... 406 50',? '+ 4H· 4C,ooH,so06QN4SP '+ 4H+ '+ 406 502 ...... 400C02 .. 300HP .. 4504 .. 4HPO . .. N20 .. 7Nz CSOOH1000 sooNn5.P " 635.502 - 800C02 .. 24SH20 .. 13N. " 3NzO " PO/" '+ 450. '" S04H+ 800C02 '+ 248H20 '+ 3N20 '+ 13N2 '+ PO/" '+ 450 . .. 504H· ..... CeooH,oooOsooNnS.P '+ 635 502 N.O .. O2 -. NO .. NO. NO'5 '+ 0 7502 - NO) NO,5 '+ 0 7502 ..... NO) CSOOH,coo0600NnS,P '+ 364H20 -+ 480C02 '+ 320CH. '+ 32NH3 '" 172H2 '+ 4H',?,S '+ PO. 480C02 " 320CH. '+ 32NH3 '+ 172H2 .. 4H2S '+ PO. � C800H,000 600N32S4P '+ 364HzO NH3 (gas) -- NH� (dissolved) NOl" (SOil) --+ N03 (ocean) NO)" '+ 4H' --- NH4' -+ 1 502 NH/ '+ 1 5� ..... NO] '+ 4H+ C'OOH'500SON4SP .. 28HzO ..... 48CH4 ... 52C02 .. H� '+ 4NH3 '+ PO/" S042" ... , S02 '+ O2 S02 '+ O2 ..... S042" S02 '+ O2 ..... 50; SO/" - SO/· sot ..... SO/" SO/ - SO/" 504 ..... S 2 " '+ 20. 53 S uplift and weathering of sedimentary S roc� 5 2 . ... 202 ..... 504 2 54 S H2S dlssolutlon H2S ... 4H20 -> 50/ ... 2H� ... 4H2 55 P weathenng of sedimentary P rock Cas(P04h(OH) (sediment) ..... 5Ca 2 . ... 3PO/" + OH" 56 P 5011 to ocean P sedlments 5Ca2" ... 3PO/ ... OH" --+ Cas(P04h(OH) 57 P P runoff pol ... H" --+ HPO/ 58 P evaporation of soluble P 3HPO/ ... Sea 2 . ... 3QH" ..... Cas(PO.h(OH) ... 2H20 + H' 59 P deposItion of soluble P Cas(P04h(OH) '" 2H20 '" H* --+ 3HPO/'" Sea2> ... 30H" 60 P deposltJon of Insoluble P Cas(PO,h(OH) {atmosphere} ..... Cas{P04h(OH) (sedlments) 61 ? P particulate flux C,ooH,soOsoN4SP '" lQB02 ... 3H'" ..... 1QOC02 ;' 76H20 '" 4N03' ", sot -I- HPO/" 62 P formatIon of atmosphenc P dust particles Sea '" OH ... 3P04 ..... Cas(PO.hOH 63 P dissolution of atmospheric P dust particles Cas{P04hOH ..... 5Ca '+ OH '+ 31'04 64 P P sedimentation 65 P sedimentary P rock formation 66 H transpiration 67 H evaporation from ocean 68 H p�eC\Pltatlon to ocean 69 H preclPltatlon to land 70 H evaporation from land 71 H uptake of HP 72 H photooxldatlon of H2 73 H river discharge 6 HPO.z- .. l OCal' .. 80H" ....... 2Ca5(P04h{OH) -+' 6HP Ca5(P04h(OH) (marine sedlments) ..... Cas(P04ls(OH} (uplifted sedlments) H20 (liqUid) ....... H20 (gas) H20 (liqUid) ..... Hp (gas) H::,o (gas) ..... H20 (hquld) H20 (gas) ..... H20 (liqUid) H20 (liqUid) ..... H20 (gas) H20 (liqUid) --+ H20 (liqUid) H, '+ 0.502 --+ H20 H20 (liqUid) ..... H20 (liquId) Marker Element Mass Pg 40.1885 0.0082 90.0033 1 4820 64 9949 70 7082 142.5052 0 1 100 1 .5460 0 1 900 0 5612 45 0966 0 9611 40 7666 1 1 9990 1 0999 23.5986 0 3800 0 4000 0 2700 0 0075 0 4000 20 9983 0 5000 0 3500 0 3500 0 0006 0 0300 0.0150 0 0270 0.0100 0 0400 0.0400 0 2425 0 1847 0.0050 0 0150 0 0130 0 0840 0 0540 0 1 396 0 0320 0 0150 0 0050 0 0550 0 0440 0.0960 0 1 160 0 0040 0 1329 0 0550 0.0550 0.0470 0 4606 0 0214 0.0187 0.0180 0.0003 0.0005 0 0008 0 0420 0.0031 0 0021 0 0019 0.0214 3000.0587 43000.8411 39000 7629 1 1004 6122 4000 0782 3000.0587 3.9864 3955.4374 Table B.2 Biosphere Outputs From Processes (Continued) Venslrn Processes 1 C OXidatIOn of land humus 2 C volcanic acbon 3 C land plant reSpiration 4 C tropospheriC oXidation of CO 5 C release o f CO2 6 C absorption of CO2 7 C gross land production 8 C production of CO 9 C oXidation of CH. 10 C sorption of CO '\ i C productlon of CH. 12 C land humus formatIOn (p) 13 C land consumptIOn 14 C gross manne production 15 C manne plant resplrabon 16 C production of CaCOs 1 7 C manne humus formatIOn (pj 1 8 C manne consumptIOn 19 C land humus formaMn (C&d) 20 C manna humus formation (c&dl 21 C coal formation 22 C transfer of land humus 23 C oXidation of manne humus 24 C Kerogen formation 25 C formatIOn of limestone 26 e weathering of limestone 27 C Igneous rock. formation 28 N reduction of N10 29 N oXidation of N� 30 N manne dentnficallon 31 N NOx formation by IIghtnm9 32 N biological frxatlCn by pnytoplankton 33 N release of N2 and Np 34 N microbial production of N, and N20 35 N uptake of N2 and N20 36 N photochemical oXidation of N20 37 N wet depositIOn of NOx 38 N aCid ram 39 N microbial producllon of NH3 40 N uptake of NH3 by sOil 41 N uptake of NH, by ocean 42 N N runoff 43 N sedimentary N rock formaMn 44 N weathenng of sedimentary N rock 45 S release of H2S 46 S ocean spray of 804 47 5 uptake of 502 by ocean 48 5 uptake of 501 by sOil 49 5 ocean spray to land 50 S 5 runoff 51 S S sedimentation 52 S sedimentary S rock formatJon 1st Product Chemical Species 800CO. 100CO, 750C02 2C02 DestmatJOn Reservoir Atmosphere CO. Atmosphere CO2 Atmosphere CO2 Atmosphere CO2 CO2 Atmosphere CO, HC03 Ocean NEC C750H,soo07soN,sS4P land plants 1 00CO Atmosphere NEC CO Atmosphere NEC 1 5C�Hl0000600N325.P Land humus 3C2SOH5Cl:l0,OON;oP35 land consumers/decomposers C,ooH,soO&QN4SP Manne plants iOOHCOJ Ocean NEC 100CaC03 Ocean NEC C'OOH'200&QN3PS Manne humus C,OOH2SOO,4(lN,s5P Manne consumersJdecomposers C800Hl0c006OQN12S.P Land humus C'OOH':l(PWN3PS Manne humus 6C'30Hl l00,oN2S Fossil fuels 8C'00H'2{,o�ONJPS Manne humus 100HCO� Ocean NEC C700H,.oo0,7S052SP7N., Fossil fuels Sedimentary rock HC03 Ocean NEC C'ooH,.ooO'OOOP32S20N2 Igneous rocK 2Nz Atmosphere N, Atmosphere N20 Nz Atmosphere N2 2NO Atmosphere NEe 4C'ooH'soOaoN.SP Manne plants 400C02 Atmosphere CO2 800C02 Atmosphere CO2 C1lOOH'oco01(){lN325.P Land humus NO Atmosphere NEG NO, Ocean NEC NO, Soil 480C02 Atmosphere CO2 CwoH'OOJO�oNnS.P Land humus NH3 (diSSolved) NO,- (oceanj NH4+ N03 48CH. SO, 50/ SO/" 50/ 50,2. 50/ Si Ocean NEC Ocean NEC Sedimentary rocK Ocean NEC Atmosphere NEC Atmosphere NEC Ocean NEC Sod Ocean NEC Manne sedlments 53 5 uplift and weathenng ef sedlment2lry 5 rocK 50/ Sedimentary rocK SOil 54 S H25 dissolution 55 P weathering of sedimentary P rock 56 P sOil to ocean P sedlments 57 P P runoff 58 P evaporation of soluble P 59 P depOSitIOn of soluble P 60 P depOSItion of insoluble P 61 P P partlculate flux 50/' 5Ca2• Cas(P04h(OH) HPO/ Cas(PO.h(DH) 3HPD42 Ca5(P04i,(OH) 100C02 62 P formation of atmosphenc P dust particles Cas(PO.)PH 63 P dissolutIOn of atmosphenc P dust particles Sea 64 P P sedimentation 65 P sedimentary P rock formatJon 66 H transpiration 67 H evaporation from ocean 68 H precIpitation to ocean 69 H precIpitation to land 70 H evaporation from land 71 H uptaKe of Hp 72 H photooxldanon of H2 73 H nver discharge 2Ca5(PO.h(OH) Cas(P04h(OH) H,o H,o H,o H,O H,o H,o H,o H,o Ocean NEe Sorl Manne sedlments Ocean NEC Atmosphere NEC Ocean NEC Manne sedlments Atmosphere CO2 Atmosphere NEC 5011 Manne sedlments Sedimentary roCK Atmosphere H20 Atmosphere Hp Ocean H,O Fresh water H;"o Atmosphere Hp Land plants Atmosphere H20 Ocean H20 2nd Product Mass Chemical Destmatlon Mass Species Pg 147 2686 500H20 o 0300 700H20 329 8123 750H20 5 4306 238 1 705 Hp 359 2402 H" 362 1 1 85 782 5°2 0 2565 125H20 3 6055 2HzO 0 41 1 6 40802 0 4497 100C02 78 1908 3DDeD2 1 8989 SO. 93 3889 132 5°2 60 9623 4N03 9 1 664 4NO;' 53 1 906 15Hp 1 2592 1302 o 6935 2ooCD2 0 6085 13N03 0 0089 20CO 0 9015 8NO, 106 6842 108Hz 2 3270 3 5N2 2 9166 1 7781 Caz. 0 0014 8502 0 0300 02 0 0236 0 0270 6H20 0 0107 2NDz 1 9642 406 502 3 1420 300H20 1 9 0454 248HzO 8.5769 635 5°2 0 0054 NO. 0 0664 0 0575 3 9590 320CH, 2 5088 364H,O 0 1698 0 14 1 7 0 0193 1 5°2 0 0221 4H+ 1 32 1 1 52C02 0 0879 O2 0 2876 0.3475 0 0120 0 3982 0 1648 0 0550 202 0 1408 1 3800 2H+ 0 0462 3PO/ 0. 10 1 1 Q 0558 0.0016 2H20 0 0015 5Ca2+ 0 0043 5 9676 76H20 0 0168 0 0045 OH 0 0103 6H20 0 1 157 26,810.0482 384,277.3578 348,530 6268 98,342.8045 35,746.7310 26,810 0482 35.6241 35,347 7974 ReservOir Pg Fresh water Hp 37 6788 Fresh water Hp 0 0861 Fresh water Hp 135.0124 0 0000 Ocean Hp 97 4978 Ocean NEC 5 9345 02 396 151 1 Ocean H,O 0 2062 Atmosphere Hp 4 6382 02 0 2582 Atmosphere CO2 Atmosphere CO2 0 8226 33 0508 5011 0 0103 02 143 9222 Ocean NEe 2 4781 Ocean NEC 0 2272 Ocean Hp 5 3100 O2 0 1 316 Atmosphere CO2 Q 2931 Ocean NEC Q 1812 Atmosphere NEe 0 0004 Ocean NEC Q 0207 Atmosphere NEe 3 8068 Atmosphere N2 0.0058 Ocean NEC Atmosphere NEe 0, Ocean H20 Atmosphere NEe 0, Ocean H20 Fresh water H20 0, Almosphere NEe Atmosphere CH. Fresh water Hp 0, Ocean NEe Atmosphere CO2 0, 0, Ocean NEC Soil Atmosphere H20 Ocean NEC Ocean Hp 0 0000 1 1679 0 0000 0 0171 0 0000 o � 042 0 0164 2 321 7 0 9647 2 4169 8 3760 0 0082 0 0000 0 0000 0 9620 0 7901 0 0000 0 0000 0 0514 0 0014 3 9265 0 0439 0 0000 0.0000 0 0000 0 0000 0 0000 0 1098 0 0000 0.0290 0 0656 0 0000 0 0000 0 0001 0 001 1 0 0000 1 8566 0 0000 0 0004 0 001 1 0 0000 0 0000 0.0000 0 0000 0.0000 0 0000 0 0000 0.0000 0.0000 49 1 492 Table B.2 Biosphere Outputs From Processes (Continued) Vensim Processes 1 C oxidation of land humus 2 C volcanic action 3 C land plant respiration 4 C troposphenc oxidauon of CO 5 C release of Co, 6 C absorption of Co, 1 C gross land production 8 C productIon ot CO 9 C oxidation of CH, 10 C sorption of CO 1 1 C production of CH, 12 C land humus formation (p) 13 C land consumption 14 C gross manne productIon 15 C marine plant reSpiration 16 C produCtlCn of CaCe, 17 C marine humus formation (p) 18 C marine consumption 19 C land humus formation (C/J.d) 20 C manne humus formation (c&d) 21 C coal lormatlon 22 C transfer of land humus 23 C oxidatIOn of marine humus 24 C kerogen formation 25 C formation of limestone 26 C weathenng ot limestone 27 C igneous rock formation 28 N reductIOn of N,o 29 N oxidation of N2 30 N manna dentnfication 31 N NOx formation by lightning 32 N biological fIXation by phytoplankton 33 N release ot N2 and N20 34 N microbial production of N2 and N20 35 N uptake of N2 and NzO 36 N photochemical OXidation of N20 37 N wet depOSition of NOx 38 N acid rain 39 N microbial production of NHl 40 N uptake of NHl by 5011 41 N uptake of NHs by ocean 42 N N runoff 43 N sedimentary N rock formation 44 N weathering of sedimentary N rock 45 5 release of HzS 46 5 ocean spray of 50_ 47 5 uptake of 50t by ocean 48 5 uptake of 50: by soil 49 5 ocean spray to land 50 5 5 runoff 51 5 5 sedimentation 52 5 sedimentary 5 rock formation 53 5 uplift and weathering of sedimentary 5 rock 54 5 HzS dissolution 55 P weathering of sedimentary P rock 56 P soil to ocean P se. formation of Ocean NEC - N l imestone / ..... �/ fl rt Carbon Cycle Influence Diagram gmp rt / / - Atmosph ere NEe - E photooxidat ion of H2 ..--D evaporati on from land el rt p recipitation to land et pi rt Land P l ants - G o U h20� F reshwate r H20 - P ---� ..... �river d ischarge Figure B.2 Hydrological Cycle Influence Diagram transp i ration � o t rt Ocean H20 - Q 505 preci p itation t o o cean 6 p o rt Atmosphere NEC - E l )( 1 1""11. "'-========;J formation 01 atmospher ic P dust pa rticles dissolution 01 atmospheric P dust ______ P" ;'� J 1apdp rt P runoff p r rt dapdp rt o sops� _-------.......... soi l to o cean P ---- sediments Soi l - I ,£2. weathering of sedimenta ry P rock o deposition 01 inso luab le P 6 dip rt � evaporation 01 so luab le P Sed imentary Rocks -I .. )( �1arine Sediment - t.1 ... j( depositio n 01 soluable P 6 dsp rt Marine P lants - K P pa rticulate flux 6 " ppf rt i Ocean NEC - N R sedimentary P rock 10rmation P sed imentation � Figure 8.3 Phosphorus Cycle Influence Diagram rr==================lIAtmosphere NEe - Erl ..... II==========;! uptake of S021Y SOi l o us02s rt Soi l - I up l ift and ""eathering o".';� ' •• uwssr rt .-----'---"----, " uptake of S 02 "�. us020 rt S runoff d ocean spray to land sr rt Sedimentary Rocl (Pg H) (1) Land Humus (Pg N) Fresh Water (Pg H) (2) Soil (Pg N) (k) (I) F igure B.6 Baseline Analysis of the Steady State Cond itions of Critical Atmosphere Stocks i n the GBCM, 2001 -2051 (Continued) 5 1 2 Glob.11 Tenestr i.11 Biosphel'e P Stocks, 2001-2051 Global Tenestl i.11 Biosphere S Stocks. 2001-2051 400 100 300 75 200 50 100 25 0 0 2001 201 1 2021 2031 2041 2051 2001 2011 2021 2031 2041 2051 Time (Ye .. ) Time (ye.,) 50 i l (Pg P) 50il 5 (Pg 5) (m) (n) F igure B.6 Baseline Analysis of the Steady State Conditions of Critical Atmosphere Stocks in the GBCM, 2001 -2051 (Continued) B.6 Limitations and Caveats of the GB CM There are several l imitations and caveats of the GBCM including: • Scenario analysis through anthropogenic perturbation. Scenarios need to be formulated to simulate these implications of human perturbation of the biogeochemical cyc les. This could include: ( 1 ) depletion of a non-renewable natural resource stocks, (2) unsustainable harvests of renewable resource stocks ( i.e. harvest plus natural decrease rates > regeneration rates), (3) exogenous injection of residuals into stocks, (4) one-off stochastic events, and (5) any combination of above. • Temporal dimension - stiff system. The model is a stiff system i.e. the biogeochemical processes operate over diverse time frames, with residence times ranging from nanoseconds to geological time. The consequences of change on the slower parts of the cycle are therefore difficult to establish. • Spatial dynamics. The effects of many of the ecological processes identified in the GCBM occur across distinct local areas. To capture these effects accurately a spatial dimension would need to be added to the GCBM. This would allow impacts to be assigned to, say, biomes or bioregions. It is however noted that spatial dynamic modelling is still very much in its infancy; refer to Costanza and Voinov (2004) for further details. B.7 Finite Difference Equations for the GBCM 5 1 3 In this Section a full set of finite difference equations is presented for the GBCM. These equations adopt the conventions outlined above in Section B .4 of this Appendix. Table B.8 provides details of the arrayed flows, and sets of processes, used in describing the non-marker e lement flows in the GBCM. 5 1 4 Table B.8 Non-marker Arrayed Flows and Sets of Processes Stock Atmosphere Stocks Atmosphere C02 Atmosphere CH4 Atmosphere N2 Atmosphere N20 Atmosphere NEC Terrestrial Biosphere Stocks Land Cons & Decomps Land Plants Land Humus Soil Marine Biosphere Stocks Marine Cons & Decomps Marine Plants Marine Humus Marine Sediments Ocean NEC Set! Process Name Set Elements/Processes ac02rp igneous rock formation, land humus formation (c&d), land humus formation (p), microbial production of N2 and N20, microbial production of NH3, P particulate flux, production of CH4, release of H2S, release of N2 and N20, oxidation of land humus, volcanic action, land plant respiration, tropospheric oxidation of CO, release of C02 ac02dp ach4rp ach4dp an2rp an2dp an20rp an20dp anecrp biological fixation by phytoplankton, uptake of N2 and N20, uptake of NH3 by soil , absorption of C02 ,gross land production microbial production of NH3, production of CH4, release of H2S oxidation of CH4, uptake of NH3 by soil coal formation, igneous rock formation, kerogen formation, microbial production of N2 and N20, release of N2 and N20, volcanic action biological fixation by phytoplankton, NOx formation by lightning, uptake of N2 and N20 microbial production of N2 and N20, oxidation of N2, release of N2 and N20 biological fixation by phytoplankton, photochemical oxidation of N20, reduction of N20, uptake of N2 and N20 oxidation of CH4, NOx formation by lightning, photochemical oxidation of N20, H2S dissolution, coal formation, igneous rock formation, microbial production of NH3, oxidation of marine humus, production of CH4, release of H2S sub, volcanic action, production of CO, ocean spray of S04, evaporation of soluable P , formation of atmospheric P dust particles anecdp deposition of soluable P, uptake of NH3 by soil, tropospheric oxidation of CO, sorption of CO, wet deposition of NOx, acid rain, uptake of NH3 by ocean, uptake of S02 by ocean, uptake of S02 by soil, deposition of insoluable P, dissolution of atmospheric P dust particles Icdrp Icddp Iprp Ipdp Ihrp Ihdp srp sdp mcdrp mcddp mprp mpdp mhrp mhdp msrp msdp onecrp onecdp land consumption land humus formation (c&d), production of CH4 gross land production, uptake of H20 land consumption, land humus formation (p), land plant respiration, transpiration land humus formation (c&d) , land humus formation (p), sorption of CO, uptake of N2 and N20, uptake of NH3 by soil coal formation, microbial production of N2 and N20, microbial production of NH3, oxidation of land humus, transfer of land humus coal formation, dissolution of atmospheric P dust particles, land consumption, land humus formation (c&d), land humus formation (p), land plant respiration, microbial production of N2 and N20, microbial production of NH3, oxidation of land humus, production of CH4, volcanic action, weathering of sedimentary P rock, acid rain, uptake of S02 by soil, ocean spray to land, uplift and weathering of sedimentary S rock coal formation, formation of atmospheric P dust particles, gross land production, land consumption, land humus formation (p), P runoff, soil to ocean P sediments, sorption of CO, uptake of N2 and N20, uptake of NH3 by soil, weathering of l imestone, N runoff, S runoff marine consumption marine humus formation (c&d), production of CO biological fixation by phytoplankton, gross marine production, gross marine production marine consumption, marine humus formation (p), marine plant respiration, P particulate flux, production of CaC03, release of H2S, release of N2 and N20 marine humus formation (c&d) , marine humus formation (p), transfer of land humus kerogen formation, oxidation of marine humus deposition of insoluable P, P sedimentation, S sedimentation, soil to ocean P sediments sedimentary P rock formation, sedimentary S rock formation absorption of C02, deposition of soluable P, H2S dissolution, marine humus formation (c&d), marine humus formation (p), marine plant respiration, oxidation of marine humus, P particulate flux, production of CaC03, production of CO, release of H2S, release of N2 and N20, transfer of land humus, weathering of l imestone, wet deposition of NOx, uptake of NH3 by ocean, N runoff, weathering of sedimentary N rock, uptake of S02 by ocean, S runoff, P runoff S sedimentation, biological fixation by phytoplankton, evaporation of soluable P, gross marine production, igneous rock formation, kerogen formation, marine consumption, marine dentrification, P particulate flux, P sedimentation, release of C02, sedimentary N rock formation, transfer of land humus, formation of l imestone, ocean spray of S04, ocean spray to land 5 1 5 Table B.8 Non-marker Arrayed Flows and Sets of Processes (Continued) Stock Hydrosphere Stocks Atmosphere H20 Freshwater H20 Ocean H20 Lithosphere Stocks Sedimentary Rocks Fossil Fuels Igneous Rocks Miscellaneous Stocks Oxygen Set! Process Name ah20rp ah20dp fh20rp fh20dp oh20rp Set Elements/Processes evaporation of soluable P, transpiration, evaporation from ocean, evaporation from land, photooxidation of H2, oxidation of CH4 deposition of soluable P, H2S dissolution, precipitation to land, precipitation to ocean coal formation, land humus formation (c&d) , land humus formation (p), land plant respi ration, microbial production of N2 and N20, oxidation of land humus, uptake of NH3 by soil, volcanic action, precipitation to land gross land production, microbial production of NH3, production of CH4, sorption of CO, uptake of N2 and N20, evaporation from land, uptake of H20, river discharge marine dentrification, marine humus formation (c&d), marine humus formation (p), marine plant respiration, P particulate flux, P sedimentation, production of CaC03, production of CO, release of C02, release of N2 and N20, transfer of land humus, precipitation to ocean, river discharge oh20dp absorption of C02, biological fixation by phytoplankton, gross marine production, kerogen formation, marine consumption, oxidation of marine humus, release of H2S, evaporation from ocean srrp srdp ffrp ffdp irrp irdp 02rp formation of limestone, sedimentary N rock formation, sedimentary P rock formation weathering of l imestone, weathering of sedimentary N rock, weathering of sedimentary coal formation, kerogen formation igneous rock formation igneous rock formation volcanic action biological fixation by phytoplankton, coal formation, gross land production, gross marine production, land consumption, marine consumption, ocean spray of S04, production of CH4, sedimentary N rock formation , sedimentary S rock formation, sorption of CO, uptake of N2 and N20, reduction of N20 02dp tropospheric oxidation of CO, acid rain, igneous rock formation, kerogen formation, land humus formation (c&d), land humus formation (p), land plant respiration, marine humus formation (c&d), marine humus formation (p), marine plant respiration, microbial production of N2 and N20, oxidation of land humus, oxidation of marine humus, P particulate flux, photooxidation of H2, production of CaC03, production of CO, release of N2 and N20, transfer of land humus, uplift and weathering of sedimentary S rock, uptake of 502 by ocean, uptake of S02 by soil , volcanic action, weathering of sedimentary N rock, wet deposition of NOx, oxidation of CH4, oxidation of N2, NOx formation by lightning, photochemical oxidation of N20 B.7.1 Atmosphere Stocks Atmosphere C02e(t + dt) where: Initial Atmosphere C02e atmos C02 from oPe absorption ofC02e = Atmosphere C02e(t) + (atmos C02 from oPe + land plant respiratione + oxidation of land hum USe + release of C02e + tropospheric oxidation of COe + volcanic actione - absorption ofC02e - atmos C02 for OPe - gross land productione) x dt = initial atmos C02e for the 1 998 base year m n L ( atmos C02 nmrsmaco2rp,e X L ( {atmos C02 nmrsm aco2rp=1 e=l processesaco2rp,e} )) = Atmosphere C02e(t) x aco2 rte 5 1 6 atmos C02 for OPe gross land productione Atmosphere CH4eCt + dt) where: Initial Atmosphere CH4e atmos CH4 from OPe atmos CH.:! for OPe oxidation ofCH4e Atmosphere N2eU + dt) where: Initial Atmosphere N2e atmos N2from OPe atmos N2 for OPe NOx formation by lightninge oxidation ofN2e Atmosphere N20e(t + dt) m L ( atmos C02 nmdsmaco2dp,e X aco2dp=1 nmdsm processesaco2dp,e } )) = Atmosphere C02e(t) x gip rte n L ( {atmos C02 e=1 = Atmosphere CH4e(t) + (atmos CH4 from OPe + production of CH.:!e - atmos CH4 for OPe - oxidation ofCH4e) x dt = initial atmos CH4e for the 1 998 base year m n L ( atmos CH4 nmrSmach-lrp,e X L( {atmos CH4 nmrsm ach4rp=1 e=1 procesSeSach-lrp, e} )) m n L ( atmos CH4 nmdsmach-ldp,e X ach4dp=1 L ( {atmos CH4 e=1 nmdsm procesSeSach-ldp,e } )) = Atmosphere CH4eCt) x och4 rte = Atmosphere N2e(t) + (atmos N2 from OPe + marine dentrificatione + reduction of N20e - atmos N2 for OPe - NOx formation by lightninge - oxidation ofN2e) x dt = initial atmos N2e for the 1 998 base year m L ( atmos N2 nmrsman2rp,e an2rp=1 processesan2rp,e } )) m L ( atmos N2 nmdsman2dp,e an2dp=1 processesan2dp,e } )) = Atmosphere N2e(t) x noxfl rte = Atmosphere N2eCt) x on2 rte n X L ( {atmos N2 e=1 n X L( {atmos N2 e=1 nmrsm nmdsm = Atmosphere N20eCt) + (atmos N20 from OPe + microbial production of N2 and N20e + oxidation of N2e + release of N2 and N20e - atmos N20 for OPe - biological fixation by phytoplanktone - photochemical oxidation of N20e - reduction ofN20e - uptake ofN2 and N20e) x dt where: Initial Atmosphere N20e atmos N20 from oPe atmos N20 for OPe = initial atmos N20e for the 1 998 base year m n 5 1 7 I ( atmos N20 nmrsman2orp,e X Ie {atmos N20 nmrsm an2orp=! e=! processesan2orp,e} » m I ( atmos N20 nmdsman2odp,e X an2odp=! nmdsm processesan2odp, e } » n I ( {atmos N20 e=! biological fixation by phytoplanktone = Atmosphere N20e(t) X biP rte photochemical oxidation of N20e = Atmosphere N20e(t) x pon20 rte reduction ofN20e uptake of N2 and N20e Atmosphere NECe(t + dt) where: Initial Atmosphere NECe atmos nec from OPe acid raine atmos nec for OPe deposition ofinsoluable Pe deposition of soluable Pe = Atmosphere N20e(t) x rn20 rte = Atmosphere N20e(t) x un2n20 rte = Atmosphere NECe(t) + (atmos nec from OPe + evaporation of soluable Pe + formation of atmospheric P dust particlese + microbial production of NH3e + NOx formation by lightninge + ocean spray of S04e + oxidation of CH.:!e + photochemical oxidation of N20e + production of COe + release of H2Se - uptake of S02 by oceane - uptake of S02 by soile - acid raine - atmos nec for OPe - deposition of insoluable Pe - deposition of soluable Pe - dissolution of atmospheric P dust particlese - H2S dissolutione - photooxidation of H2e - sorption of COe - uptake of NH3 by oceane - uptake of NH3 by soile - wet deposition of NOxe - tropospheric oxidation of co e) x dt = initial atmos nece for the 1998 base year m I ( atmos nec nmrsmanecrp,e x aMecrp=1 processesanecrp,e} ) = Atmosphere NECe(t) x ar rte n I ( {atmos nec nmrsm e=! m n I ( atmos nec nmdsmanecdp,e x I ( {atmos nec nmdsm anecdp=l e=! processesanecdp,e} )) = Atmosphere NECe(t) x dip rte = Atmosphere NECe(t) x dsp rte dissolution of atmospheric P dust particlese = Atmosphere NECe(t) x dapdp rte 5 1 8 H2S dissolutione = Atmosphere NECe(t) x h2sd rte photooxidation of H2e = Atmosphere NECe(t) x ph2 rte sorption ofCOe = Atmosphere NECe(t) x sco rte tropospheric oxidation ofCOe = Atmosphere NECe(t) x toco rte uptake ofNH3 by oceane = Atmosphere NECe(t) x unh30 rte uptake ofNH3 by soile = Atmosphere NECe(t) x unh3s rte uptake ofS02 by oceane = Atmosphere NECe(t) x uso20 rte uptake ofS02 by soile = Atmosphere NECe(t) x uso2s rte wet deposition ofNOxe = Atmosphere NECe(t) x wdnox rte B.7.2 Terrestrial Stocks Land Consumers and DecomposerseCt + dt) = Land Consumers and Decomposerse(t) + (land cons & decomps from OPe + land consumptione - land cons & decompsfor oPe - land humus formation (c&d)e - production of CH4e) x dt where: Initial Land Consumers and Decomposerse = initial land cons & decompse for the 1 998 base year land cons & decomps from OPe m n L ( land cons & decamps nmrsmlcdrp,e x L ( { land cons lcdrp=1 e=1 & decamps nmdsm pracesseslcdrp,e })) m n land cons & decomps for OPe = L ( land cons & decomps nmdsmlcddp,e x L ( {land cons lcddp=! e=1 & decomps nmdsm processeslcddp,e } )) land humus formation (c&d)e = Land Consumers and Decomposerse(t) x llifcd rte production ofCH4e = Land Consumers and DecomposerseCt) x pch4 rte Land PlantseCt + dt) = Land Plantse(t) + (gross land productione + land plants from OPe + uptake of H20e - land consumptione - land humus formation (P)e - land plant respiratione - land plants for OPe - transpiratione) x dt where: Initial Land Plantse = initial land plantse for the 1 998 base year land plants from OPe land consumptione land humus formation (P)e land plants for OPe land plant respiratione transpiratione Land Humuse(t + dt) where: Initial Land Humuse land humus from OPe coal formatione land humus for OPe m n 5 1 9 L ( land plants nmrsmlprp. e x I ( { land plants nmrsm processes/prp.e } » = Land Plante(t) x le rte = Land PlanteCt) x lhfp rte e=! m n = I ( land plants nmdsmlpdp.e x I ( {land plants nmdsm Ipdp=! e=! processeslpdp.e} » = Land PlanteCt) x lpr rte = Land Plante(t) x t rte = Land Humuse(t) + (land humus formation (c&d)e + land humus formation (P)e + land humus from OPe + sorption of COe + uptake of N2 and N20e + uptake of NH3 by soile - coal formatione - land humus for OPe - microbial production of N2 and N20e - microbial production of NH3e - oxidation of land humuse - transfer of land humuse) x dt = initial land hum USe for the 1998 base year m n = I ( land humus nmrsmlhrp. e x I ( {land humus nmrsm Ihrp=! e=l processesfhrp.e } » = Land HumuseCt) x cf rte m n = I ( land humus nmdsmlhdp.e x I ( { land humus nmdsm fhdp=1 e=1 processes Ihdp. e} » microbial production ofN2 and N20e = Land HumuseCt) x mpn2n20 rte microbial production ofNH3e = Land HumuseCt) x mpnh3 rte oxidation of land humuse = Land HumuseCt) x olh rte transfer of land hum USe = Land Humuse(t) x tlh rte Soile(t + dt) = Soile(t) + (acid raine + dissolution of atmospheric P dust particlese + ocean spray to lande + soil from OPe + uplift and weathering of sedimentary S rocke + uptake of S02 by soile + weathering of sedimentary P rocke - formation of atmospheric 520 where: Initial Soile soil from OPe P dust particlese - N runoff. - P runoff. - S runof.fe - soil for OPe - soil to ocean P sedimentse) x dt = initial soile for the 1 998 base year m n = L ( soil nmrsmsrp,e x L( {soil nmrsm processessrp,e })) srp=1 e=1 formation of atmospheric P dust particlese = Soile(t) x fapdp rte N runoff. P runo!fe S runoff- soilfor OPe soil to ocean P sedimentse B.7.3 Marine Stocks = Soile(t) x nr rte = Soile(t) x pr rte = Soile(t) x sr rte m n = L ( soil nmdsmsdp,e x L ( {soil nmdsm proceSSessdp,e } )) sdp=l e=l = Soile(t) x sops rte Marine Consumers and Decomposerse(t + dt) = Marine Consumers and Decomposerse(t) + (marine cons & decamps from OPe + marine consumptione - marine cons & decamps for OPe - marine humus formation (c&d)e -production of co e) x dt where: Initial Marine Consumers and Decomposerse = initial marine cons & decompse for the 1 998 base year m marine cons & decamps from OPe = L ( marine cons & decamps nmrsmmcdrp, e x mcdrp=l n L ( {marine cons & decomps nmrsm processesmcdrp,e })) e=1 m n marine cons & decomps for OPe = L ( marine cons & decamps nmdsmmcddp,e x L ( {marine mcddp=l e=l cons & decamps nmdsm processesmcddp,e })) marine humus formation (c&d)e = Marine Consumers and Decomposerse(t) x mhfcd rte production ofCOe Marine Plantse(t + dt) = Marine Consumers and Decomposerse(t) x pco rte = Marine PlantseCt) + (biological fixation by phytoplanktone + gross marine productione + marine plants from OPe - marine consumptione - marine humus formation (P)e - marine plant where: Initial Marine Plantse marine plants from OPe 521 respiratione - marine plants for OPe - P particulate fluxe - production of CaC03e - release of H2Se - release of N2 and N20e) x dt = initial marine plantse for the 1 998 base year m L ( marine plants nmrsmmprp,e x mprp�l nmrsm processesmprp,e } )) n I ( {marine plants e�l marine consumptione = Marine Plantse(t) x mc rte marine humus formation (P)e = Marine Plantse(t) x mhfp rte marine plant respiratione = Marine PlantseCt) x mpr rte marine plants for OPe P particulate fluxe production ofCaC03e release of H2Se release of N2 and N20e Marine HumuseU + dt) where: Initial Marine Humuse marine humus from OPe kerogen formatione marine humus for OPe oxidation of marine humuse Marine Sedimente(t + dt) m I ( marine plants nmdsmrnpdp,e x mpdp�l nmdsm processesmpdp,e } » = Marine Plantse(t) x ppfrte = Marine Plantse(t) x pcac03 rte = Marine PlantseCt) x rh2s rte = Marine PlantseCt) x rn2n20 rte n I ( {marine plants e�l = Marine HumuseCt) + (marine humus formation (c&d)e + marine humus formation (p)e + marine humus from OPe + transfer of land hum USe - kerogen formatione - marine humus for OPe - oxidation of marine humuse) x dt = initial marine humuse for the 1 998 base year m I ( marine humus nmrsmrnhrp,e x mhrp�l nmrsm processesmhrp,e}» = Marine Humuse(t) x !if rte n I ( {marine humus e�l m n I ( marine humus nmdsmrnhdp,e x I ( {marine humus mhdp�l e�l nmdsm processeSmhdp,e } » = Marine HumuseU) x omh rte = Marine Sedimente(t) + (deposition of insoluable Pe + marine sediments from OPe + P sedimentatione + S sedimentatione + 522 where: Initial Marine Sedimente marine sediments from oPe marine sediments for OPe soil to ocean P sedimentse - sedimentary P rock formatione - marine sediments for OPe - sedimentary S rock formatione) x dt == initial marine sedimente for the 1 998 base year == m L e marine sediments nmrsmmsrp,e x msrp=1 sediments nmrsm processesmsrp, e } )) m L ( marine sediments nmdsmmsdp,e x msdp=1 n Le {marine e=1 n L ( {marine e=] sediments nmdsm processesmsdp,e } )) sedimentary P rock formatione == Marine Sedimente(t) x sprf rte sedimentary S rock formatione == Marine Sedimente(t) x ssrf rte Ocean NECe(t + dt) where: Initial Ocean NECe ocean nec from OPe evaporation ofsoluable Pe formation of limestonee gross marine productione marine dentrificatione ocean nec for OPe == Ocean NECe(t) + (absorption of C02e + deposition of soluable Pe + H2S dissolutione + marine plant respiratione + N runoff. + ocean nee from OPe + oxidation of marine humuse + P particulate fluxe + P runoff. + production of CaC03e + S runoff. + uptake of NH3 by oceane + uptake of S02 by oceane + weathering of limestonee + weathering of sedimentary N rocke + wet deposition of NOxe - ocean spray of S04e - evaporation of soluable Pe - formation of limestonee - gross marine productione - marine dentrificatione - ocean nec for OPe - ocean spray to lande - P sedimentatione - release of C02e - S sedimentatione - sedimentary N rock formatione) x dt == initial ocean nece for the 1 998 base year m L ( ocean nec nmrsmonecrp,e x onecrp=1 processesonecrp, e } )) == Ocean NECe(t) x esp rte == Ocean NECe(t) x fl rte == Ocean NECe(t) x gmp rte == Ocean NECe(t) x md rte n L ( {ocean nec nmrsm e=] m n L ( ocean nec nmdsmonecdp,e x L ( {ocean nec nmdsm onecdp=1 e=] processesonecdp,e } )) ocean spray of S04e ocean spray to lande P sedimentatione release ofC02e S sedimentatione = Ocean NECe(t) x oss04 rte = Ocean NECe(t) x osl rte = Ocean NECe(t) x ps rte = Ocean NECe(t) x rc02 rte = Ocean NECeCt) x ss rte 523 sedimentary N rock formatione = Ocean NECe(t) x snrf rte B.7.4 Hydrosphere Stocks Atmosphere H20e(t + dt) where: Initial Atmosphere H20e atmos H20 from OPe atmos H20 for OPe precipitation to lande precipitation to oceane Freshwater H20eCt + dt) where: Initial Freshwater H20e freshwater H20 from OPe evaporation from lande freshwater H20 for OPe = Atmosphere H20e(t) + (atmos H20 from OPe + evaporation from lande + evaporation from oceane + photooxidation of H2e + transpiratione - atmos H20 for OPe - precipitation to lande - precipitation to oceane) x dt = initial atmos H20e for the 1 998 base year m n I ( atmos H20 nmrsmah2orp, e X L ( {atmos H20 nmrsm ah2orp=1 e=1 processesah2orp,e} )) m L ( atmos H20 nmdsmah2odp,e x ah2odp=1 nmdsm processesah2odp,e } )) = Atmosphere H20e(t) x pi rte = Atmosphere H20e(t) x po rte n L ( {atmos H20 e=l = Freshwater H20eCt) + (freshwater H20 from OPe + precipitation to lande - evaporation from lande - freshwater H20 for OPe - river dischargee - uptake of H20e) x dt = initial freshwater H20e for the 1 998 base year m n I (freshwater H20 nmrSmjh2orp,e x L ( {freshwater jh2orp=1 e=l H20 nmrsm processesjh2orp,e} )) = Freshwater H20eCt) x el rte m n I (freshwater H20 nmdsmjh2odp,e x L ( {freshwater jh2odp=1 e=1 H20 nmdsm processesjh2odp, e } )) 524 river dischargee uptake of H20e Ocean H20e(t + dt) where: Initial Ocean H20e ocean H20 from OPe evaporation from oceane ocean H20 for OPe B.7.S Lithosphere Stocks Sedimentary Rockse(t + dt) where: Initial Sedimentary Rockse sedimentary rocks from OPe sedimentary rocks for OPe = Freshwater H20e(t) x rd rte = Freshwater H20eCt) x uh20 rte = Ocean H20e(t) + (ocean H20 from OPe + precipitation to oceane + river dischargee - evaporation from oceane - ocean H20 for OPe) X dt = initial ocean H20e for the 1 998 base year m n I ( ocean H20 nmrsmoh2orp,e x I ( {ocean H20 nmrsm oh2orp=1 e=l processesoh2orp,e } )) = Ocean H20e(t) x eo rte m I ( ocean H20 nmdsmoh2odp,e x oh2odp=1 nmdsm processesoh2odp,e } )) n I ( {ocean H20 e=1 = Sedimentary Rockse(t) + (formation of limestonee + sedimentary N rock formatione + sedimentary P rock formatione + sedimentary rocks from OPe + sedimentary S rock formatione - sedimentary rocks for OPe - uplift and weathering of sedimentary S rocke - weathering of limestonee - weathering of sedimentary N rocke - weathering of sedimentary P rocke) x dt = initial sedimentary rockse for the 1 998 base year m n I ( sedimentary rocks nmrsmsrrp, e x I ( {sedimentary srrp=1 e=1 rocks nmrsm processessrrp, e })) m n I ( sedimentary rocks nmdsmsrdp,e x I ( {sedimentary srdp=1 e=1 rocks nmdsm processessrdp, e })) uplift and weathering of sedimentary S rocke = Sedimentary RockseCt) x uwssr rte weathering of limestonee = Sedimentary Rockse(t) x wl rte weathering of sedimentary N rocke = Sedimentary Rockse(t) x wsnr rte 525 weathering of sedimentary P rocke = Sedimentary Rockse(t) x wspr rte Fossil FuelseU + dt) where: Initial Fossil Fuelse fossil fuels from oPe fossil fuels for OPe igneous rock formatione Igneous RockeCt + dt) where: Initial Igneous Rocke igneous rocks from OPe igneous rocks for OPe volcanic actione B.7.6 Oxygen Stock Oxygene(t + dt) where: Initial Oxygene 02from oPe 02for oPe = Fossil Fuelse(t) + (coal formatione + fossil fuels from OPe + kerogen formatione - fossil fuels for OPe - igneous rock formatione) x dt = initial fossil fuelse for the 1 998 base year m n I (fossil fuels nmrsmjfrp,e x I ( {fossil fuels nmrsm jfrp=1 e=1 processesjfrp,e } » m n = I (fossil fuels nmdsmjfddp,e x I ( {fossil fuels nmdsm jfdp=! e=1 processesjfdp,e } ») = Fossil Fuelse(t) x irfrte = Igneous Rocke(t) + (igneous rock formation" + igneous rocks from OPe - igneous rocks for OPe - volcanic actione) x dt = initial igneous rocke for the 1 998 base year m n = I ( igneous rocks nmrsmtrrp,e x I ( { igneous rocks nmrsm Irrp=1 processesmp,e} » m I ( igneous rocks nmdsmlrdp,e x irdp=1 nmdsm processes1rdp, e}») = Igneous RockeCt) x va rte n I ( { igneous rocks e=1 = OxygeneCt) + (02from OPe - 02for oPe) x dt = initial 02e for the 1 998 base year m n = I ( 02 nmrsmo2rp,e X I ( { 02 nmrsm processeso2rp,e} » o2rp�1 e=1 m n = I ( 02 nmdsmo2dp,e x L ( {02 nmdsm processeso2dp,e } » o2dp=1 e=1 Appendix C Mathematical Description of the Interregional Trade Flows Optimisation used in the Ecological Footprint Analysis 527 This Appendix describes the optimisation approach used to determine the interregional trade flows between the study region and its trading partners . The optimisation problem is portrayed diagrammatically in Figure C . l and described mathematically below. Figure C.1 From R1 to F rom R2 to F rom Rn_1 to From Rn to Rn R 1 S1 · · · Sm S1 " Sm + 1 Rn R 1 S1 " , Sm " . S 1 " , Sm q C D I tJ Matrix X Rn R 1 S1 " , Sm S 1 " , Sm + � � + i�J L...J ,---, 1 + : L....j I � Travel time (hrs) Vector z :: :: :: :: :: + Structure of the Interregional Trade F lows Optim isation Problem Exports/ Imports ($) +�RnSm I Scalar v I The capital letters R and S are used respectively to denote regions and industries i .e . , R]S] denotes Industry 1 in Region 1 . The letter n represents the number of regions, while the letter m represents the number of industries. 528 Matrix X [n x m x 2, (n x n - n) x m]. This matrix describes the flow of trade between regions. A negative sign (-) denotes exports, while a positive s ign (+) denotes imports. Column 1 , for example, describes the trade flows between Region 1 , RI, and Region 2, R2• Vector y [n x m x 2, 1 ] . This column vector describes imports to, and exports from, each region by industry. The element in row 1 , for example, represents the sum of all Industry 1 exports originating from Region 1 , -LRjS j . The elements in this vector are used as binding constraints in the optimisation. Vector = [ 1 , (n x n - n) x m]. This row vector denotes freight haulage times between regions per dollar of trade flow. Freight haulage times are calculated using an origin-destination matrix. Scalar v ( 1 , 1 ) . This scalar is the sum of row vector z. It represents the total freight travel time needed to move goods and services between all permutations of regions and industries. Minimisation of this scalar is the objective function. The optimisation is solved as follows: Min: z w v subject to: X w = y W >= 0 where: w = column vector [en x n - n) x m, 1 ] describing the flow ($) of m industry commodities between n regions to be solved for. In the analysis undertaken in Chapter 8 of this thesis, there are 5 ,520 possible flows between industries that are quantitatively determined by solving for w. 529 Appendix D Non-Survey Regionalisation Methodologies This Appendix provides a brief outline of the most commonly used non-survey regionalisation methodologies. The intention is to provide only background knowledge rather than a comprehensive evaluation of non-survey methodologies290• For further information refer to Richardson ( 1 972), Smith and Morrison ( 1 974), Jensen et al. ( 1 979), Sawyer and Miller ( 1 983), Butcher ( 1 985), Stevens et al. ( 1 989), and Dietzenbacher and Lahr (200 1 ). The development of regional input-output methodologies began with Hirsch ( 1959), and produced studies of the economies of inter alia Boulder (Miemyk et al. 1 967), West Virginia (Miernyk et al. 1 970), Philadelphia (Isard and Langford, 1 969; Isard, Langford, and Romanoff 1 966-68; Isard and Langford, 1 97 1), Kansas (Emerson and Hackman, 1 97 1 ), and Washington (Bourque et al. 1 967). These survey based studies were the product of coordinated research groups. They established best practice procedures that have subsequently been adopted by many countries in producing survey tables. By the mid 1 970s high costs and time delays associated with bui lding survey based tables had led input-output analysts to seek alternative construction methods. The resulting non-survey approaches typically attempted to reduce national technical coefficients to the regional equivalents by purely mathematical means. Analysts such as Schaffer and Chu ( 1 969), Smith and Morrison ( 1 974), Eskelinen and Suosra ( 1 980) and Sawyer and Miller ( 1 983) conducted notable reviews of the leading non-survey methodologies. Their fmdings concluded that in general non-survey based approaches tend to overestimate inter-industry flows and underestimate exports, and by implication, overestimate multipliers. Although most analysts acknowledge that non-survey methods carmot be expected to represent adequately the complex interrelationships in a regional economy, as captured by a survey based approach, virtual consensus has emerged that supplementing non-survey based tables with superior survey data greatly improves the analytical accuracy of the regional table derived. The major non-survey techniques are discussed below. 290 It should be noted that the techniques reviewed have for the most part only been assessed in the context of regionalising inter-industry matrices, rather than commodity-by-industry matrices as are used in this thesis. 530 D . I Coefficient Reduction Methodologies The location quotient, commodity balance and constrained matrix methodologies attempt to mechanically reduce national input-output coefficients to the regional level . The first two approaches operate on the premise that regional and national coefficients differ only be the magnitude of the regional import coefficient. The impl icit assumption of maximum regional self-sufficiency means that these approaches will generally overestimate regional input-output coefficients. By comparison, the constrained matrix approach ruthlessly employs mathematical iteration to scale down national level intermediate demand transactions until input-output sub­ totals approximate regional estimates. D.1 . 1 Location Quotient Methodologies The Simple Location Quotient Method Schaffer and Chu ( 1 969) assessed the performance of five non-survey techniques in estimating a regional input-output table for the 1 958 State of Washington economy. The performance of each technique was assessed in terms of how the table generated compared with a survey based table for the same year. They found that the simple location quotient (SLQ) technique generally provided the best estimates of regional inter-industry structure. Similar studies conducted by Smith and Morrison ( 1 974), Eskelinen and Suorsa ( 1 980) and Sawyer and Mil ler ( 1 983) all corroborated Schaffer and Chu's ( 1 969) findings, namely that the SLQ technique produced the best estimates of the techniques reviewed, although coefficients were l ikely to be overestimated. The SLQ method is arguably the most widely employed data reduction method. Smith and Morrison ( 1 974) attribute this to its simplicity, modest data demands, and time and cost effectiveness. Essentially, this method compares the relative importance of regional industries, usually in terms of output or employment, to their relative importance nationally. Mathematically, the SLQ for industry j is calculated as fol lows: (D. 1 ) 53 1 where P represents output or employment and the superscript r the region. The regional input­ output coefficients for each industry j may be determined by multiplying the national input­ output coefficients for that industry j by the corresponding SLQj. If the SLQ; is greater than or equal to 1 the regional coefficient is set to its national equivalent, otherwise the regional coefficient is set to the product of the multiplication with any shortfall assigned to imports. When an industry is absent from the region, but present nationally, the national coefficient is assigned in full to imports. Mayer and Pleeter ( 1 975) offer theoretical justifications for the use of location quotients. Other Location Quotient Methodologies Several location quotient derivatives have targeted, with only limited success, the major shortcomings of the SLQ approach. These include the Purchases Only Location Quotient, the Cross-Industry Location Quotient, and the Logarithmic Cross-Industry Location Quotient. For a complete description of these approaches refer to Richardson ( 1972, p. 1 1 8-1 22). D.1.2 The Commodity Balance Approaches The Supply-Demand Pool (SDP) method is most frequently util ised commodity-balance approach. For each industry the SDP method subtracts total regional requirements from total regional outputs to obtain a net commodity surplus or deficit. If a surplus results, regional supply is sufficient to satisfY regional demand, and national coefficients may be substituted for regional coefficients. If a deficit occurs, regional supply is insufficient to satisfY regional demand, and imports from other regions will be required. In this case, regional coefficients are estimated as follows: (D.2) where r represents the regional coefficient in industry j row i, a the national coeffic ient for industry j row i, fil total national inputs for industry j, and PI' total regional inputs of industry j. Comparisons between the SDP approach, other adaptation approaches, and actual regional coeffic ients have produced mixed results (see for example McMenamin and Haring, 1 974, Morrison and Smith, 1974; Sawyer and Miller, 1 983) . 532 D.1 .3 Constrained Matrix Techniques A key characteristic of constrained matrix techniques is their iterative nature. The most commonly utilised constrained matrix technique is the RAS technique (early application were undertaken by Stone and Brown ( 1 962), Czamanski and Malizia (1 969), and explicated by Bacharach ( 1 965), modified by inter alia Morrison and Smith ( 1 974), McMenamin and Haring ( 1 974) and more recently by D ietzenbacher and Lahr (200 1 » . Interestingly, St Louis ( 1 989) adapted the standard US approach for use with commodity-by-industry tables291 . Tables D . 1 to D.6 provide a hypothetical example of the application of the US technique. Table D . I provides estimates of primary inputs, final demand and total input and output for each industry. Using these estimates it is possible to calculate, for each industry, intermediate input/output by subtracting primary inputs/final demand from total input/output e .g. total intermediate input for the primary sector would be calculated as 964 - 586 = 378 . Table 0 . 1 Hypothetical Transactions Table (Target Year) Primary Manufacturing Services Intermed iate Final Demand Gross Output Demand Primary 647 31 8 964 Manufacturing 1 , 575 1 ,893 3,468 Services 1 ,248 3,380 4,628 Intermed iate Demand 378 1 ,826 1 ,265 3,470 5,590 9,060 Primary I nput 586 1 ,641 3 ,363 5,590 1 ,3 14 6,904 Gross Input 964 3,468 4,628 9,060 6 ,904 1 5 ,964 Using a regional transactions table, it is then possible to calculate row ratios for intermediate demand by dividing total intermediate demand for each row industry in the target year by the sum of the total intermediate demand for each row industry in the base year e.g. for primary industry the ratio would be 647 -7- 65 1 = 0.99368 (Table D .2). Each row element in Table D.2 is then multiplied by the corresponding row ratio e .g. 1 7 1 x 0.99368 = 1 70 (0 d.p.), 464 x 0 .99368 = 461 (0 d.p.), and 1 6 x 0.99368 = 1 6 (0 d.p.). 291 St Louis ( 1 989, p.3 84) found that the modified commodity-by-industry RAS "compared very favourably with the [inter-industry] Leontief MS ". Table 0.2 Hypothetical Transactions Table (Base Year) Primary Manufacturing Services Intermediate Demand Primary Manufacturing Services In termediate Intermediate Ratio 1 7 1 464 1 6 1 1 3 860 442 128 454 809 41 1 1 ,778 1 , 267 Demand Demand - Target Yr (a 1 ) (b ) 651 647 1 ,41 4 1 ,575 1 , 391 1 ,248 3,456 3,470 (bla 1 ) 533 0.99368 1 . 1 1 363 0.89757 The resulting values are transferred to Table 0.3 and column ratios are then computed by dividing estimated total intermediate demand for each column industry in the target year by the total intermediate demand for each column industry e.g. for primary industry the ratio would be 3 78 -:- 4 1 0 = 0.92308. Each column element in Table D.3 is then multiplied by the corresponding column ratio and transferred to Table D.4 e.g. 1 70 x 0.92308 = 1 57 (0 d.p.), 1 25 x 0.92308 = 1 1 6 (0 d.p.), and 1 1 5 x 0.92308 = 1 06 (0 d.p.) . Table 0.3 Hypothetical Transactions Table (Target Year) _ 1st Iteration Primary Manufacturing Services Intermediate Demand - Target Yr Primary 1 70 461 16 647 Manufacturing 1 25 957 492 1 ,575 Services 1 1 5 407 726 1 ,248 Total (C 1) 4 10 1 ,826 1 ,234 3,470 I nter Demand - Target Yr (d) 378 1 ,826 1 ,265 Ratio (dlc 1 ) 0 .92308 1 .00019 1 .02526 Table 0.4 Hypothetical Transactions Table (Target Year) - 2nd Iteration Primary Manufacturing Services Total (a 2) Primary 1 57 461 1 6 Manufacturing 1 1 6 957 505 Services 1 06 407 744 I nter Demand - Target Yr 378 1 ,826 1 ,265 634 1 ,578 1 ,258 3,470 Intermed iate Ratio Demand - Target Yr (b ) (bla 2) 647 1 ,575 1 ,248 3,470 1 .01 981 0.998 1 1 0.99238 In a s imilar manner Tables 0.5 and 0.6 are then derived. As successive iterations are performed the ratio values approach unity. After several iterations the results obtained are sufficient for practical purposes. 534 Table 0 .5 Hypothetical Transactions Table (Target Year) - 3rd Iteration Primary Manufacturing Services Intermediate Demand - Target Yr Primary 1 60 471 1 6 647 Manufacturing 1 1 5 956 504 1 ,575 Services 1 05 404 739 1 ,248 Total (C2) 380 1 ,830 1 ,259 3,470 Inter Demand - Target Yr (d) 378 1 ,826 1 ,265 Ratio (d/C2) 0.99455 0.99769 1 .00501 Table 0 .6 Hypothetical Transactions Table (Target Year) - Final Iteration Primary Manufacturing Services Total (a 3) Primary 1 59 469 16 Manufacturing 1 1 5 953 506 Services 1 05 403 742 Inter Demand - Target Yr 378 1 ,826 1 ,265 645 1 ,575 1 ,250 3,470 Intermed iate Ratio Demand - Target Yr (b ) (b/a 3) 647 1 ,575 1 ,248 3,470 1 .00291 1 .00020 0.99825 Overall, the RAS method depends crucially on the choice of, in the case of an inter-industry table, the pre-assigned A matrix on which the iterations are performed. Concerns with the RAS technique have included its uniform substitution effect and the so-called ' ripple effect' (see Bates and Bacharach (1 963)) whereby one erroneous estimate in the RAS generates errors throughout the rest of the table. D.2 Other Approaches D.2 .1 Regional Weights Using employment, value added, or output data this technique computes regional weights, these weights in turn are used to assemble a h ighly disaggregated national transaction table into a more aggregate regional table. In comparison with transactions tables derived from survey based data this approach has not produced favourable results and is considered to be in the tentative stage of development. 535 D.2.2 Representative Regional Coefficients Mandeville ( 1 975) advocates the use of representative regional coefficients taken from the survey coefficients developed for regions with s imilar economic compositions. This method has two major drawbacks. F irstly, an analyst may experience extreme difficulty in finding representative regional coefficients from survey data. Secondly, careful consideration must be given to using representative regional coefficients as subtle differences can easily exist between differing regions. Appendix E Industry Definitions and Concordances Table E.1 Industry Defin itions Industry Description Other horticulture 2 Apple and pear growing 3 Kiwifruit growing 4 Other fruit growing 5 Mixed livestock and cropping 6 Sheep and beef cattle farming 7 Dairy cattle farming 8 Other farming 9 Services to agriculture, hunting and trapping 1 0 Forestry 1 1 Services to forestry 1 2 Logging 1 3 Fishing 14 Coal mining 15 Services to mining 16 Other mining and quarrying 1 7 Oil & gas extraction 1 8 Oil & gas exploration 1 9 Meat processing 20 Poultry processing 2 1 Bacon, ham and smallgood manufacturing 22 Dairy product manufacturing 23 Fruit and vegetable, oil and fat, cereal and fiour manufacturing 24 Bakery, sugar and confectionery manufacturing 25 Seafood processing 26 Other food manufacturing 27 Soft drink, cordial and syrup manufacturing 28 Beer, wine, spirit and tobacco manufacturing 29 Textile manufacturing 30 Clothing manufacturing 31 Footwear manufacturing 32 Other leather product manufacturing 33 Log sawmilling and timber dressing 34 Other wood product manufacturing 35 Paper and paper product manufacturing 36 Printing and services to printing 37 Publishing and recorded media manufacturing 38 Petroleum refining 39 Petroleum & coal product manufacturing nec 40 Fertiliser manufacturing 41 Other industrial chemical manufacturing 42 Medicinal, detergent and cosmetic manufacturing 43 Other chemical product manufacturing 44 Rubber manufacturing 45 Plastic product manufacturing 46 Glass and glass product and ceramic manufacturing 47 Other non-metallic mineral product manufacturing 48 Basic metal manufacturing 49 Structural, sheet and fabricated metal product manufacturing 50 Motor vehicle and part manufacturing 51 Ship and Boat Building 52 Other transport equipment manufacturing 53 Photographic and scientific equipment manufacturing 54 Electronic equipment and appliance manufacturing 55 Agricultural machinery manufacturing 56 Other industrial machinery and equipment manufacturing 57 Prefabricated building manufacturing 58 Furniture manufacturing 59 Other manufacturing 60 Electricity 61 Gas supply 62 Water supply ANZSIC Code A01 1 1 0-A01 1 30 A01 1 50 A01 1 70 A01 140, A01 1 60 , A01 1 91-A0 1 1 99 A01 2 1 0-A01220, A01 591 -A01 592 A01 230-A01 250 A01 300 A01410-A01530, A01 593-A01699 A02120-A02200 A03010 A03030 A03020 A041 1 0-A04200 B 1 1 01 0-B1 1 020 8 15 140-81 5200 B131 1 0-814200 B1 2000 B151 10-81 5 1 20 C21 1 1 0 C21 1 20 C21 1 30 C21 2 1 0-C2 1 290 C21 300-C2 1 520 C21610-C21720 C21 730 C21 740-C21790 C21 8 1 0 C21 820-C21 900 C22 1 1 0-C22390 C22400 C22500 C2261 1-C22620 C23 1 1 0-C231 30 C2321 0-C23290 C2331 0-C23390 C24 1 1 0-C241 30 C24210-C24300 C251 00 C25200 C25310 C25320 C25430, C25450-C25460 C2541 0-C25420, C25440, C25470-C25490 C25510-C25590 C25610-C25660 C26100-C26290 C26310-C26400 C27 1 1 0-C27330 C27 41 0-C27690 C28 1 1 0-C281 90 C28210-C28220 C28230-C28290 C2831 0-C28390 C28410-C28590 C28610 C28620-C28690 C291 1 0-C291 90 C2921 0-C29290 C2941 0-C29490 036 100 036200 03701 0 537 538 Table E.1 Industry Defin itions (Continued) Industry Description 63 Residential building construction (incl owner builders) 64 Non-residential building construction 65 Non-building construction 66 Site preparation services 67 Building structure services 68 Plumbing services 69 I nstallation trade services 70 Building completion services 71 Other construction services 72 Wholesale trade 73 Retail trade 74 Accommodation 75 Bars, clubs, cafes and restaurants 76 Road Freight transport 77 Road passenger transport 78 Water and rail transport 79 Air transport, services to transport and storage 80 Communication services 81 Finance 82 Life insurance 83 Superannuation fund operation 84 Health insurance 85 General insurance 86 Services to finance and insurance 87 Residential property operators 88 Commercial property operators 89 Real estate agents 90 Ownership of owner-occupied dwellings 91 I nvestors in other property 92 Vehicle and equipment h ire 93 Scientific research 94 Technical services 95 Computer services 96 Legal services 97 Accounting services 98 Advertising and marketing services 99 Business administrative and management services 1 00 Employment, security and investigative services 1 0 1 Pest control and cleaning services 1 02 Other business services 1 03 Central government administration 1 04 Defence 1 05 Public order and safety services 1 06 Local government administration services and civil defence 1 07 Pre-school education 1 08 Primary and secondary education 1 09 Post school education 1 1 0 Other education 1 1 1 Hospitals and nursing homes 1 1 2 Med ical, dental and other health services 1 1 3 Veterinary services 1 14 Child care services 1 1 5 Accommodation for the aged 1 1 6 Other community care services 1 1 7 Motion picture, radio and TV services 1 1 8 Libraries, museums and the arts 1 1 9 Horse and dog racing 120 Lotteries, casinos and other gambling 1 2 1 Other sport and recreational services 1 22 Personal and other community services 1 23 Waste disposal, sewerage and drainage services ANZSIC Code E41 1 1 0-E41 1 20 E41 1 30 E41 2 1 0-E41 220 E421 00 E4221 0-E42240 E42310 E42320-E42340 E4241 0-E42450 E4251 0-E42590 F45 1 1 0-F47990 G5 1 1 0 1 -G53290 H571 00-H571 09 H57200-H57400 161 1 00 161 2 1 0- 161230 , 1661 1 0-1661 90 162000-163030, 16621 0-166290 164010-165090, 166300-167090 J71 1 1 0-J7 1 200 K73 100-K73400 K741 1 0 K74 1 20 K7421 0 K74220 K751 10-K75200 L771 1 0-L771 1 9 L77 1 20-L77 1 29 L77200 N/A L77300-L77309 L7741 0-L77430 L781 00 L7821 0-L78290 L7831 0-L78340 L7841 0 L78420 L7851 0-L78530 L78540-L78550 L7861 0-L78640 L78650-L 78660 L78670-L78690 M8 1 1 1 0 , M81 300 M82000 M81 200, Q9631 0-Q96330 M81 1 30 N841 00 N8421 0-N84240 N8431 0-N84320 N84400 0861 1 0-086 1 30 08621 0-086390 086400 0871 00 08721 0 087220-087290 P91 1 1 0-P91220 P92 1 00-P92590 P93 1 1 0-P931 1 2 P9321 0-P93290 P931 20-P931 9� P93300 Q95 1 1 0-Q96290 , Q97000 D37020, Q96340 Table E.2 Industry Defin itions Concordance 123 Industry Description 1 Other horticulture 2 Apple and pear growing 3 Kiwifruit growing 4 Other fruit growing 5 Mixed livestock and croppmg 6 Sheep and beef cattle farming 7 Dairy cattle farming 8 Other farming 9 Services to agnculture, hunting and trapping 10 Forestry 11 Services to forestry 12 Logging 13 Fishing 14 Coal mining 15 Services to mining 16 Other mining and quarrying 17 011 & gas extractIon 18 Oil & gas exploration 19 Meat processing 20 Poultry processing 21 Bacon, ham and smailgoad manufactunng 22 Dall"y product manufacturing 23 Fruit and vegetable, 011 and fat, cereal and flour manufactunng 24 Bakery, sugar and confectionery manufactunng 25 Seafood processing 26 Other food manufacturing 27 Soft drink. cordial and syrup manufactunng 28 Beer, wine, Sptnt and tobacco manufactunng 29 Textile manufacturing 30 Clothing manufactunng 31 Footwear manufactunng 32 Other leather product manufacturing 33 Log sawmllling and timber dreSSing 34 Other wood product manufactunng 35 Paper and paper product manufactunng 36 Pnnting and services to pnntlng 37 Publishing and recorded media manufactunng 38 Petroleum refimng 39 Petroleum & coal product manufactunng nec 40 Fertiliser manufactunng 41 Other Industnal chemical manufacturing 48 Industry Descnption 1 Horticulture and frUIt growing 1 Horticulture and fruit growing 1 Horticulture and fruit growing 1 Horticulture and fruit growing 2 Livestock and cropping farming 2 livestock and cropping farming 3 Dairy cattle farming 4 Other farming 5 Services to agriculture, hunting ana trapping 6 Forestry and logging 6 Forestry and logging 6 Forestry and logging 7 Fishing 8 Mining and quarrying 8 Mming and quarrying 8 Mining and quarrymg 9 OIl and. gas exploration and. extraction 9 Oll and gas exploration and extraction 10 Meat and meat product manufactunng 10 Meat and meat product manufactunng 10 Meat and meat product manufactunng 11 Dairy product manufactunng 12 Other food manufactunng 12 Other food manufacturing 12 Other food manufacturing 12 Other food manufactunng 13 Beverage, malt and tobacco manufactunng 13 Beverage, ma�t and tobacco manufactunng 14 Textile and apparel manufactunng 14 Textile and apparel manufactunng 14 Textile and apparel manufactunng 14 Textile and apparel manufactunng 15 Wood product manufactunng 15 Wood product manufacturing 16 Paper and paper product manufactunng 17 Pnntlng , publishing and recorded media 35 Industry Description 1 Horticulture and fruit growing 1 Horticulture and fruit growing 1 Horticulture and fruit growing 1 Horticulture and frUIt growing 2 Livestock and cropping farming 2 Livestock and cropping farming 3 Dairy cattle farming 4 Other farming 5 Services to agriculture, hunting and trappmg 6 Forestry and logging 6 Forestry and logging 6 Forestry and Jogging 7 Fishing 8 Mining, quarrying, 011 and gas exploration and extraction 8 Mining, quarrying, 011 and gas exploratIon and extraction 8 Mining, quarrying, 011 and gas exploration and extraction 8 Mming, quarrymg, 01\ and gas exploration and extraction 8 Mining, quarrymg, 011 and gas exploratIon and extraction 9 Meat and meat product manufactunng 9 Meat and meat product manufactunng 9 Meat and meat product manufactunng 10 Dairy product manufactunng 1 1 Other fOOd, beverage, malt and tobacco manufactunng 11 Other fOOd, beverage, malt and tobacco manufactunng 11 Other foad, beverage, malt and tobacco manufactunng 11 Other food, beverage, malt and tobacco manufactunng 1 1 Other food, beverage, malt and tobacco manufactunng 11 Other food, beverage, malt arid tobacco manufactunng 12 Textile and apparel manufacturing 12 Textile and apparel manufactunng 12 Textile and apparel manufacturing 12 Textile and apparel manufactunng 13 Wood, paper and paper product manufactunng 13 Wood, paper and paper product manufactunng 13 Woad, paper and paper product manufactunng 14 Pnntlng, publishing and recorded media 17 Printing , publ1shing and recorded 14 Pnnting, publishing and recorded media media 18 Petroleum and Industnal chemical 15 Petroleum, rubber, plastic and other manufactunng chemical product manufactunng 18 Petroleum and industnal chemical 15 Petroleum, rubber, plastiC and other manufactunng chemJcal product manufacturing 18 Petroleum and industnal chemical 15 Petroleum, rubber, plastiC and other manufactunng chemical product manufactunng 18 Petroleum and Industrial chemical 15 Petroleum, rubber, plastic and other manufacturing chemical product manufactunng 23 Industry Descnption 1 Agriculture 1 Agriculture 1 Agriculture 1 Agnculture 1 Agriculture 1 Agriculture 1 Agriculture 1 Agriculture 1 Agnculture 3 Forestry and logging 3 Forestry and logging 3 Forestry and Jogging 2 Fishing 539 4 Mining and quarrying, oil and gas exploratIon and extraction 4 Mining and quarrying, 011 and gas exploration and extraction 4 MIning and quarrymg, 011 and gas exploration and extraction 4 Mining and quarry'lng, oil and gas exploration and extractJOn 4 Mining and quarrying, oil and gas exploratIOn and extraction 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufacturing 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufactunng 5 Food, beverages and tobacco manufacturing 5 Food., beverages and. tobacco manufacturing 6 Textile and apparel manufactunng 6 Textile and apparel manufactunng 6 Textile and apparel manufactunng 6 Textile and apparel manufactunng 7 Wood product manufacturing 7 Wood product manufacturing S Paper, paper products, printing , publishing and recorded media manufacturing 8 Paper, paper products, pnnting, publishing and recorded media manufactunng 8 Paper, paper products, printing, publishing and recorded media manufactunng 9 Petroleum, Chemical, rubber and plastic product manufactunng 9 Petroleum, chemical, rubber and plastiC product manufactunng 9 Petroleum, chemical, rubber and plastic product manufacturing 9 Petroleum, chemical, rubber and plastic product manufacturing 540 Table E.2 Industry Defin itions Concordance (Continued) 1 23 Industry Descnptlon 42 Medicinal, detergent and cosmetic manufactunng 43 Other chemical product manufacturing 44 Rubber manufacturing 45 Plastic product manufacturing 46 Glass and glass product and ceramic manufacturing 47 Other non-metallic mineral product manufactunng 48 Basic metal manufacturing 49 Structural, sheet and fabncated metal product manufacturing 50 Motor vehicle and part manufacturing 51 Ship and Boat BUilding 52 Other transport equipment manufacturing 53 Photographic and sCientific eqUipment manufacturing 54 Electronic equipment and appliance manufacturing 5S Agncultural machInery manufactunng 56 Other Industrial machinery and eqUipment manufactunng 57 Prefabricated bUilding manufacturing 58 Furniture manufacturing 59 Other manufactunng 60 Electricity 61 Gas supply 62 Water supply 63 Residential bUilding construction (incl owner bUilders) 64 NonMresidentlal building construction 65 NonMbullding construction 66 Site preparation services 67 Building structure services 68 Plumbing services 69 Installation trade seNices 70 Building completion services 71 Other construction services 72 Wholesale trade 73 Retail trade 74 Accommodation 75 Bars, clubs, cafes and restaurants 76 Road Freight transport 77 Road passenger transport 78 Water and rail transport 79 Air transport, services to transport and storage 80 Communication services 81 Finance 82 Ufe insurance 48 Industry Descnption 19 Rubber, plastiC and other chemical product manufacturing 19 Rubber, plastic and other chemical product manufactunng 19 Rubber, plastiC and other chemical product manufactunng 19 Rubber, plastiC and other chemical product manufactunng 20 Non-metallic mineral product manufactunng 20 Non-metallic mineral product manufactunng 21 BasIc metal manufactunng 22 Stnuctural, sheet, and fabncated metal product manufactunng 23 Transport equipment manufactunng 23 Transport equipment manufacturing 23 Transport equipment manufactunng 24 MaChinery and equipment manufacturing 24 Machinery and equipment manufacturing 24 Machinery and equIpment manufactunng 24 Machinery and equipment manufactunng 25 Furniture and other manufacturing 25 Fumlture and other manufacturing 25 Fumiture and other manufactunng 26 Electncity generation and supply 27 Gas supply 28 Water supply 29 Construction 29 ConstructIOn 29 Construction 29 Construction 29 Construction 29 Construction 29 Construction 29 Construction 29 Construction 30 Wholesale trade 31 Retail trade 32 Accommodation, restaurants and bars 32 AccommodatJon, restaurants and bars 33 Road transport 33 Road transport 34 Water and rail transport 35 Air transport, services to transport and storage 36 Communication services 37 Finance 38 Insurance 35 Industry Descnption 15 Petroleum, rubber, plastiC and other chemical product manufacturing 15 Petroleum, rubber, plastic and other chemical product manufacturing 15 Petroleum, rubber, plastiC and other chemical product manufactunng 15 Petroleum, rubber, plastic and other chemical product manufacturing 16 NonMmetallic mmeral product manufacturing 16 Non-metallic mineral product manufacturing 17 Basic metal manufacturing 18 Other manufactunng 18 Other manufactunng 18 Other manufactunng 18 Other manufacturing 1 8 Other manufacturing 18 Other manufactunng is Other manufacturing 18 Other manufactunng 18 Other manufactunng 18 Other manufactunng 18 Other manufactunng 19 Electncity generatIon and supply 20 Gas supply 21 Water supply 22 Construction 22 Construction 22 Construction 22 Construction 22 Construction 22 ConstructIOn 22 ConstructIon 22 Construction 22 ConstructIon 23 Wholesale and retail trade, accomodation, restaurants and bars 23 Wholesale and retail trade, accomodatlon, restaurants and bars 23 Wholesale and retail trade, accomodation, restaurants and bars 23 Wholesale and retai l trade, accomodation, restaurants and bars 24 Transport and storage 24 Transport and storage 24 Transport and storage 24 Transport and storage 25 Communication services 26 Finance and insurance 26 Finance and insurance 23 Industry Description 9 Petroleum, chemical. rubber and plastic product manufactunng 9 Petroleum, chemical, rubber and plastic product manufactunng 9 Petroleum, chemical. rubber and plastic product manufactunng 9 Petroleum, chemical, rubber and plastiC product manufacturing 10 Non-matellic mineral product manufacturing 10 Non-mateUic mineral prOduct manufacturing 1 1 BaSIC metal manufacturing 12 Structural, sheet, fabncated metal, transport, machinery and equipment product manufacturing 12 Structural, sheet. fabricated metal, transport, machinery and equipment product manufactunng 12 Structural, sheet. fabncated metal, transport, machinery and equipment product manufacturing 12 Structural, sheet. fabricated metal, transport, machinery and equipment product manufactunng 12 Stnuctural, sheet, fabncated metal, transport, machinery and equipment product manufacturing 12 Structural, sheet. fabncated metal, transport, machinery and equipment product manufactunng 1 2 Structural, Sheet, fabricated metal, transport, machinery and equipment product manufactunng 12 Structura l, sheet, fabricated metal, transport, machinery and equipment product manufacturing 1 3 Other Manufacturing 1 3 Other Manufactunng 13 Other Manufacturing 1 4 Electnclty, gas and water 14 Electnclty, gas and water 1 4 Electncity, gas and water 1 5 Construction 15 Construction 1 5 Construction 15 Construction 15 Construction 15 Construction 15 Construction 1 5 Construction 15 Construction 1 6 Wholesale and retail trade, accommodation, restaurants and bars 16 Wholesale and retail trade, accommodation, restaurants and bars 16 Wholesale and retail trade, accommodation, restaurants and bars 1 6 Wholesale and retail trade, accommodation, restaurants and bars 1 7 Transport and storage 1 7 Transport and storage 17 Transport and storage 1 7 Transport and storage 1 8 CommUnication services 1 9 Finance, insurance, real estate and business servIces 19 Finance, insurance, real estate and bUSiness services Table E.2 Industry Defin itions Concordance (Continued) 1 23 Industry Description 48 Industry Description 83 Superannuation fund operation 38 Insurance 84 Health insurance 38 Insurance 8S General insurance 38 Insurance 86 Services to finance and Insurance 39 Services to finance and investment 87 Residential property operators 40 Real estate 88 Commercial property operators 40 Real estate 89 Real estate agents 90 OwnerShip of owner·occupled dwellings 91 Investors In other property 92 Vehicle and equipment hire 93 SCIentific research 94 Technical services 95 Computer services 96 Legal services 97 Accounting services 98 Advertlsmg and marketing services 99 Busmess administrative and management services 100 Employment, secunty and Investigative services 101 Pest control and cleaning services 1 02 Other business services 103 Central government administration 104 0efence 1 05 Public order and safety services 1 06 Local govemment administration services and CIVil defence 1 07 Pre-school education 1 08 Pnmary and secondary education 1 09 Post school education 1 10 Other education 1 1 1 Hospitals and nursmg homes 1 1 2 Medical, dental and other health services 1 13 Vetennary services 1 14 Child care services 1 1 5 Accommodation for the aged 1 1 6 Other community care services 1 1 7 Motion picture, radiO and TV services 1 1 8 Libraries, museums and the arts 1 19 Horse and dog racing 1 20 Lotteries, casinos and other gambling 1 2 1 Other sport and recreational services 122 Personal and other community services 1 23 Waste disposal, sewerage and drainage services 40 Real estate 41 Ownership of owner-occupied dwellings 42 Business services 42 Busmess services 42 Bustr'less servlces 42 Business services 42 Business services 42 Business services 42 Business services 42 Business services 42 Business services 42 Busmess services 42 Business services 42 Business services 43 Central govemment administration, defence, public order and safety services 43 Central government administration, defence, publiC order and safety services 43 Central government administration, defence, public order and safety services 44 Local government administration services and Civil defence 45 Education 45 Education 45 Education 45 Education 46 Health and community services 46 Health and community services 46 Health and community services 46 Health and community services 46 Health and community services 46 Health and community services 47 Cultural and recreational services 47 Cultural and recreational services 47 Cultural and recreatIonal servIces 47 Cultural and recreational services 47 Cultural and recreational servIces 48 Personal and other community servIces 48 Personal and other community services 35 Industry Description 26 Finance and insurance 26 Finance and insurance 26 Finance and insurance 26 Fmance and insurance 27 Real estate 27 Real estate 27 Real estate 28 Ownership of owner-occupied dwellings 29 BUSiness services Including equipment hire 29 Busmess servIces includmg eqUIpment hire 29 BUSiness services Includmg equipment hire 29 Busmess servIces including equipment hIre 29 Busmess services Including equipment hire 29 BUSiness services including eqUipment hIre 29 Busmess services Including equipment hire 29 BUSiness services Includmg equipment hire 29 Business servIces including equipment hire 29 BUSiness services Including eqUipment hire 29 BUSiness servIces IncludIng equIpment hire 29 BUSiness services Including equipment hire 30 Central government administratIon. defence, publ ic order and safety services 30 Central government administration, defence, publiC order and safety services 30 Central govemment administration, defence, publiC order and safety services 31 Local government administration services and CIVJl defence 32 Education 32 Education 32 Education 32 Education 33 Health and communIty services 33 Health and community services 33 Health and community services 33 Health and community services 33 Health and community services 33 Health and community services 34 Cultural and recreational seNlces 34 Cultural and recreational services 34 Cultural and recreational services 34 Cultural and recreational services 34 Cultural and recreational services 35 Personal and other community services 35 Personal and other communIty services 54 1 23 Industry Descnption 19 Finance, insurance, real estate and bustness servIces 1 9 Finance, insurance, real estate and busmess services 19 Finance, insurance, real estate and business services 1 9 Finance, insurance, real estate and business services 1 9 Finance, insurance, real estate and bUSiness services 19 Finance, insurance, real estate and business services 1 9 Finance, Insurance. real estate and business services 20 Ownership of owner-occupied dwellings 19 Finance, insurance, real estate and bUSiness services 1 9 Finance, jnsurance, real estate and busmess services 19 FInance , insurance, real estate and bUSiness services 1 9 Finance, insurance, real estate and busmess services 19 Finance, Insurance, real estate and bUSiness services 19 Finance, Insurance, real estate and busmess services 1 9 Fmance, insurance, real estate and business services 19 Finance, Insurance, real estate and bUSiness services 19 Fmance, insurance, real estate and business services 19 Finance, Insurance, real estate and busmess services 19 Flnance, insurance, real estate and bUSiness services 19 Finance, Insurance, real estate and bUSiness services 22 Central government administration, defence, public order and safety services 22 Central government administration, defence, publiC order and safety services 22 Central government administration, defence, publiC order and safety services 23 Local government admimstration servIces and Civil defence 21 Community, SOCial and personal srves 21 CommunIty, SOCial and personal srvcs 21 Community, SOCial and personal srves 21 Community, social and personal srvcs 2 1 Community, social and personal srvcs 21 Community, SOCial and personal srvcs 2 1 Community, social and personal srves 21 Community, social and personal srves 21 Community, social and personal srvcs 2 1 Community, social and personal srves 21 Community, SOCial and personal srvcs 2 1 Community, social and personal srves 21 Community, SOCial and personal srves 21 Community, social and personal srves 21 Community, SOCial and personal srvcs 2 1 Community, SOCial and personal srves 2 1 Community, SOCial and personal srves 542 Table E.3 Commodity Definitions Commodity Description 1 Vegetables 2 Pipfruit 3 Kiwifruit 4 Other fruit and nuts 5 Oil seeds 6 Plants, flowers, seeds 7 Raw vegetable materials 8 Sheep 9 Cattle 10 Wool 1 1 Grain and other crops 12 Beverage and spice crops 13 Unmanufactured tobacco 14 Raw milk 15 Pigs 16 Poultry 1 7 Deer 18 Other livestock 19 Other animal products 20 Agriculture services 21 Forestry and logging 22 Natural gums 23 Standing timber 24 Other forestry products 25 Services to forestry 26 Fish 27 Crustaceans 28 Fishing services 29 Fishing quota leases 30 Coal 31 Metal ores 32 Building stone 33 Gypsum and limestone 34 Sand, pebbles, gravel, clay 35 Chemical and fertilizer minerals 36 Salt 37 Precious metals and stones 38 Services incidental to mining 39 Capitalised exploration 40 Crude petroleum and natural gas 41 Capitalised oil and gas exploration 42 Meat and meat products 43 Poultry products 44 Bacon, ham and smallgood products 45 Hides and skins 46 Processed milk and cream 47 Yoghurt, buttermilk, icecream 48 Other dairy products 49 Prepared fish 50 Prepared vegetables 51 Fruit juices 52 Prepared fruit and nuts 53 Oils and fats 54 Grain products 55 Starches 56 Animal feedings 57 Bakery products 58 Sugar 59 Confectionery 60 Macaroni and noodles 61 Other food products 62 Spirits, wines, beer, tobacco 63 Soft drinks, bottled water 64 Natural textiles 65 Cotton textiles 66 Man-made fibres and textiles 67 Yarn and thread 68 Woven fabrics 69 Other textiles 70 Carpets ANZSCC Code 012 013.1 5 0 13 . 18 . 16 0 13. 1 1 -013 . 14 . 0 13 . 16. 0 13 . 17, 01 3 . 18 . 1 1-0 13 . 18. 15 , 0 13. 18 . 17-01 3 . 18 .26, 01 3.20-013.40 014 015 019 021 . 1 5 021 . 1 1 , 021 . 12 .01 , 021 . 1 2.02 029 . 12 0 1 1 016 , 018 0 17 No corresponding ANZSCC 021 . 13 021 . 1 4 021 . 16 . 13 021 . 16. 1 1 , 021 . 16 . 12, 02 1 . 16 . 14-021 . 1 6.90 029 . 1 1 , 029 . 14-029.90 881 . 1 0-881 .30 031 032 No corresponding ANZSCC 039 881 .40 041 , 049 042 882 892.90 1 1 0 1 4 15 1 1 52 1 53, 1 54 161 162 163 883 No corresponding ANZSCC 120 No corresponding ANZSCC 21 1 . 1 1 , 2 1 1 . 1 2 , 2 1 1 . 1 5-21 1 . 1 8, 21 1 .24, 2 1 1 .25, 21 1 .27 03, 2 1 1 .28-2 1 1 .3 1 , 2 1 1 .34- 2 1 1 .37 , 21 1 .41 , 21 1 .43, 2 1 1 .44 21 1 .21 -21 1 .23, 2 1 1 .42 2 1 1 . 1 3, 2 1 1 . 14 , 2 1 1 .26, 2 1 1 .27.01 , 2 1 1 .27.02, 2 1 1 .33, 2 1 1 38 029 . 13 221 229 . 13 , 229 .14 , 229 . 1 8 229.1 1 , 229 . 12 , 229 . 1 5-229 . 17 , 229.21-229.23 21 1 .32, 212 213 2 14 2 15 216-2 1 8 231 232 233 234 235 236 237 239 241-243, 250 244 261 .01 261 .90 262, 355 263, 264 265-268 271 , 279 272 Table E.3 Commodity Defin itions (Continued) Commodity Description 71 Twine, rope, netting 72 Tanned skins and leather 73 Knitted fabrics 74 Clothing 75 Handbags and articles of leather 76 Footwear 77 Wood 78 Panels and boards 79 Veneer sheets and plywood 80 Builders joinery 81 Wood containers 82 Other wood products 83 Pulp, paper and paperboard 84 Non metal wastes and scraps 85 Books and stationery 86 Prepared printing plates 87 Newspapers and journals 88 Petroleum products 89 Industrial chemicals 90 Other chemical products 91 Plastics 92 Rubber 93 Rubber tyres 94 Paints 95 Pharmaceutical products 96 Soap and perfumes 97 Fertilisers 98 Pesticides 99 Glass and glass products 1 00 Cement, lime and plaster 101 Articles of concrete and stone 102 Other mineral products 1 03 Metal wastes 1 04 Iron and steel 1 05 Other metals 1 06 Structural metal products 1 07 Tanks, reservoirs and containers 1 08 Steam generators 1 09 Other fabricated metal products 1 1 0 Engines 1 1 1 Motor vehicles and parts 1 1 2 Coachwork 1 1 3 Ships 1 14 Pleasure and sporting boats 1 15 Other transport equipment 1 1 6 Aircraft and parts 1 1 7 General industrial machinery 1 1 8 Machinery for textile production 1 19 Agricultural and forestry equipment 1 20 Machinery for mining 1 21 Machinery for food production 1 22 DomestiC appliances 1 23 Office equipment 1 24 Computers and parts 1 25 Electric equipment 1 26 Audio and video records and tapes 1 27 Watches and clocks 1 28 Medical equipment 1 29 Photographic and scientific eqUipment 1 30 Furniture 131 Jewellery 1 32 Musical instruments 1 33 Sports goods 1 34 Games and toys 1 35 Prefabricated buildings 1 36 Other manufactured articles 1 37 Services incidental to manufacturing 1 38 Electricity 1 39 Services incidental to electricity distribution 1 40 Water ANZSCC Code 273 283-291 281 282 292 297 31 1 -313 314 3 1 5 3 1 6 3 17 3 1 9 321 392 322-326 327 328 331 -335 336-345 354 , 391 347, 363-369, 372 348, 362 361 351 352 353 346.01 -346.08 346.09 371 , 373 374 375, 376 379 393 4 1 1 4 13-416 412, 421 422 423 429 431 491 492 493 494 495, 499 496 432-439, 442, 443 446 441 444 445 448 449-451 452 461 -474 475 484 481 482, 482 381 382 383 384 385, 386 387 389, 447 884, 885 1 71 887 173-180 543 544 Table E.3 Commod ity Defin itions (Continued) Commodity Description 141 Pre-erection work 1 42 Residential building construction 1 43 Non-residential building construction 1 44 Civil engineering 1 45 Prefabricated constructions 1 46 Other installation work 1 47 Plumbing 148 Electrical installation work 1 49 Fencing 1 50 Building completion work 151 Land and land improvements 1 52 Wholesale trade 1 53 Retail trade 1 54 Repair services to machinery and equipment 1 55 Accommodation 1 56 Meal services 1 57 Takeaways 1 58 Beverage services 1 59 Road passenger transport 1 60 Road freight transport 1 61 Supporting services for road transport 1 62 Sea, water and rail services 1 63 Air transport, other transport and storage services 1 64 Commun ication services 1 65 Finance 1 66 Life insurance 1 67 Superannuation services 1 68 Health insurance 1 69 General insurance 1 70 Services to finance and insurance 1 71 Leased commercial property services 1 72 Leased residential property services 1 73 Other real estate services 1 74 Owner-occupied dwellings 1 75 Equipment h ire 1 76 Computer software and services 1 77 Legal services 1 78 Accounting services 1 79 Taxation services 180 Architectural and engineering services 18 1 Advertising and marketing 1 82 Management consultancy 1 83 Research and development 1 84 Placement and supply of personnel 1 85 Investigation and security services 1 86 Cleaning 1 87 Photographic services 1 88 Other business services 1 89 Central government administration services 190 Public order, safety and defence 1 91 Local government administration services 1 92 Preschool education 1 93 Primary education 1 94 Secondary education 1 95 Higher education 196 Other education services 1 97 Hospital and nursing care 198 Medical, dental and other health services 1 99 Vet services 200 Accommodation for the aged 201 Other social services 202 Child care services 203 Motion picture, radio, TV services 204 News agency services 205 Libraries, museums, art 206 Sport and recreation services 207 Other personal and other community services 208 Sewerage services 209 Waste disposal 210 Direct purchases abroad by residents ANZSCC Code 5 1 1 51 2.01-5 1 2.03, 521 .01 51 2.04, 5 1 2.05, 521 .02 5 13 , 518 , 522 514 515 5 16 . 12 , 5 16 . 1 3 5 16. 1 1 , 5 16 . 14 , 51 6.90 516 . 15 , 5 16 . 16 51 7 532-539 621 , 622 631 , 632 61 1 .03, 6 1 2.91-612.99, 633, 886 641 642 . 10.01 , 642.90 642. 10.02 643 7 12. 1 1 -712 .20 71 2.30-71 2.40 744 71 1 , 72 1 , 722, 743, 745 713. 10, 731-742, 746-749 751 -754 8 1 1 , 8 13 812 . 1 1 8 12 . 12 8 12 .21 81 2.29 8 14 821 .20 821 . 1 0 822 821 . 1 0 831 841-849, 892 . 1 1 861 862 863 867 864, 871 , 892. 1 2 865, 866 851-853 872 873 874 875 876-879 91 1 . 1 1 , 9 1 1 .20-91 2.50, 91 2.70, 912.90-91 3.40 912 .60, 9 1 2.61 , 91 2.80 9 1 1 . 12 , 9 1 1 . 1 3 921 . 1 0 921 .90 922 923 924-929 931 . 1 1-93 1 . 1 3 931 .21-931 .99 932 933 . 1 1 933 . 19 , 933.29 933.21 961 962 963 964 832, 940.40-959.90, 970-980 940 . 10 940.20-940.30 No corresponding ANZSCC Table E.4 Commodity Definitions Concordance 210 Commodity Description 1 Vegetables 2 Pipfruit 3 Kiwifrult 4 Other fruit and nuts 5 Oil seeds 6 Plants, flowers, seeds 7 Raw vegetable materials 8 Sheep 9 Cattle 10 Wool 1 1 Grain and other crops 12 Beverage and spice crops 13 Unmanufactured tobacco 1 4 Raw milk 15 Pigs 16 Poultry 1 7 Deer 18 Other livestock 19 Other anlma! products 20 Agriculture services 21 Forestry and logging 22 Natural gums 23 Standing timber 24 Other forestry products 25 Services to forestry 26 Fish 27 Crustaceans 28 Fishing services 29 Fishing quota leases 30 Coal 31 Metal ores 32 Building stone 33 Gypsum and l imestone 34 Sand, pebbles, gravel, clay 35 Chemical and fertilizer minerals 36 Salt 37 Precious metals and stones 38 Services InCidental to mining 39 Capitalised exploration 40 Crude petroleum and natural gas 41 Capitalised 011 and gas explrtn 42 Meat and meat products 43 Poultry products 44 Bacon, ham and smallgood products 45 Hides and skins 46 Processed milk and cream 47 Yoghurt, buttermilk, icecream 48 Other dairy products 49 Prepared fish 50 Prepared vegetables 51 Fruit jUices 52 Prepared fruit and nuts 53 Oils and fats 54 Gra\n products 55 Starches 56 Animal feedings 57 Bakery products 58 Sugar 59 ConfecliOnery 60 Macaroni and noodles 61 Other food products 62 Spirits, wines, beer, tobacco 63 Soft drinks, bottled water 64 Natural textiles 65 Cotton textiles 66 Man-made fibres and textiles 67 Yarn and thread 68 Woven fabrics 69 Other textiles 70 Carpets 64 Commodity Description 1 Horticulture and fruit 1 Horticulture and frUit 1 Horticulture and fruit 1 Horticulture and frUit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 2 Sheep 3 Cattle 4 Wool 5 Grain and other crops 5 Grain and other crops 5 Grain and other crops 6 Raw milk 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other Ijvestock and animal prdcts 8 Agriculture services 9 Forestry and logging 9 Forestry and logging 9 Forestry and logging 9 Forestry and logging 9 Forestry and logging 10 Fishing and fish products 10 Fishing and fish products 10 Fishing and fish products 1 0 Fishing and fish products 1 1 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 11 Mining and quarrying 1 1 Mining and quarrying 1 1 Mining and quarrying 12 Oil and gas 12 0,1 and gas 13 Meat products 13 Meat products 13 Meat products 13 Meat products 14 Dairy products 14 Dairy products 14 Dairy products 1 5 Other food 15 Other food 15 Other food 15 Other food 15 Other food 1 5 Other food 15 Other food 15 Other food 1 5 Other food 15 Other food 1 5 Other food 1 5 Other food 15 Other food 16 Beverages and tobacco 16 Beverages and tobacco 1 7 Textiles 17 Textiles 17 Textiles 1 7 Textiles 17 Textiles 1 7 Textiles 17 Textiles 48 Commodity Description 1 Horticulture and fruit 1 Horticulture and frUit 1 Horticulture and fruit 1 Horticulture and frUit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 2 Sheep 3 Cattle 4 Wool 5 Grain and other crops 5 Grain and other crops 5 Grain and other crops 6 Raw milk 7 Other livestock and animal prdcts 7 Other livestock and animal prdets 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 9 Fishing and fish products 9 Fishing and fish products 9 Fishing and fish products 9 Fishing and fish products 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 MIning and quarrying 10 MIning and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 1 1 011 and gas 11 Oil and gas 12 Meat products 12 Meat products 12 Meat products 12 Meat products 13 Dairy products 1 3 Dairy products 1 3 Dairy products 14 Other food 14 Other food 14 Other food 14 Other food 1 4 Other food 14 Other food 14 Other food 1 4 Other food 14 Other food 1 4 Other food 1 4 Other food 14 Other food 1 4 Other food 15 Beverages and tobacco 1 5 Beverages and tobacco 16 Textiles 16 Textiles 16 Textiles 16 Textiles 16 Textiles 16 Textiles 16 Textiles 35 Commodity Description 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 1 Horticulture and fruit 2 Sheep 3 Cattle 4 Wool 5 Grain and other crops 5 Grain and other crops 5 Grain and other crops 6 Raw milk 545 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 7 Other livestock and animal prdcts 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 8 Forestry and logging 9 Fishing and fish products 9 Fishing and fish products 9 Fishing and fish products 9 Fishing and fish products 10 Mining and quarrying 1 0 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 10 Mining and quarrying 1 0 Mining and quarrying 10 Mining and quarrying 1 1 0,1 and gas 1 1 Oil and gas 12 Meat products 1 2 Meat products 12 Meat products 12 Meat products 13 Dairy products 1 3 Dairy products 1 3 Dairy products 14 Other food 14 Other food 1 4 Other food 1 4 Other food 1 4 Other food 1 4 Other food 14 Other food 1 4 Other food 14 Other food 1 4 Other food 1 4 Other food 1 4 Other food 1 4 Other food 1 5 Beverages and tobacco 1 5 Beverages and tobacco 16 Textiles, clothing and footwear 16 Textiles, clothing and footwear 1 6 Textiles, clothing and footwear 1 6 Textiles, clothing and footwear 16 Textiles, clothing and footwear 16 Textiles, clothing and footwear 16 Textiles, clothing and footwear 546 Table E.4 Commodity Defin itions Concordance (Cont inued) 2 10 Commodity Descnption 71 Twine, rope, netting 72 Tanned skms and leather 73 Knitted fabnes 74 Clothing 75 Handbags and articles of leather 76 Footwear 77 Wood 78 Panels and boards 79 Veneer sheets and plywood 80 Builders Joinery 81 Wood containers 82 Other wood products 83 Pulp, paper and paperboard 84 Non metal wastes and scraps 85 Books and stationery 86 Prepared pnntlng plates 87 Newspapers and joumals 88 Petroleum products 89 Industnal chemicals 90 Other chemical products 91 Plastics 92 Rubber 93 Rubber tyres 94 Paints 95 Pharmaceutical products 96 Soap and perfumes 9? Fertilisers 98 PestiCides 99 Glass and glass products 1 00 Cement, lime and plaster 10 1 Articles of concrete and stone 1 02 Other mineral products 1 03 Metal wastes 1 04 Iron and steel 105 Other metals 106 Structural metal products 1 07 Tanks, reservoirs and containers 1 08 Steam generators 109 Other fabncated metal products 1 1 0 Engines 1 1 1 Motor vehicles and parts 1 1 2 Coachwork 1 1 3 Ships 1 1 4 Pleasure and sporting boats 1 1 5 Other transport equipment 1 1 6 Aircraft and parts 1 1 7 General indUstnal machmery 1 1 8 Machinery for textlie production 1 1 9 Agricultural and forestry equipment 1 20 Machmery for mining 12 1 Machinery for foOd production 1 22 Domestic appliances 1 23 Office equipment 1 24 Computers and parts 125 ElectriC equipment 126 Audio and Video records and 1 27 Watches and clocks 128 Medical equipment 129 Photographic & sCJentific eqpmnt 1 30 Furniture 1 3 1 Jewellery 1 32 MUSical Instruments 1 33 Sports goods 134 Games and toys 1 35 Prefabricated bUildings 1 36 Other manufactured articles 1 37 Services incidental to mnfctrng 1 38 Electncity 1 39 Services incidental to electriCity distnbution 140 Water 64 Commodity Descnptlon 17 Textiles 17 Textiles 18 Clothing and footwear 18 Clothing and footwear 18 Clothing and footwear 1 8 Clothing and footwear 1 9 Wood products 19 Wood products 19 Wood products 19 Wood products 19 Wood products 19 Wood products 20 Paper products 20 Paper products 21 Pnntlng, pblshng & recorded media 21 Printing, pblshng & recorded media 21 Pnntlng, pblshng & recorded media 22 Petroleum products 23 industrial chemicals 24 Rubber, plastiC & othr ch em prdcts 24 Rubber, plastic & othr chem prdcts 24 Rubber, plastic & othr chem prdcts 24 Rubber, plastic & othr chem prdcts 24 Rubber, plastiC & othr chem prdcts 24 Rubber, plastiC & othr chem prdcts 24 Rubber, plastic & othr chem prdcts 25 Non metalliC mineral products 25 Non metalliC mineral products 25 Non metallic m ineral products 25 Non metalliC mmeral products 25 Non metalliC mmeral products 25 Non metallic mineral products 26 BaSIC metals 26 Basic metals 26 Basic metals 27 Strct�, sheet & fbrctd metal prdcts 27 Strctrl, sheet & fbrctd metal prdcts 27 Strct�, sheet & fbrctd metal prdcts 27 Strctrl, sheet & fbrctd metal prdcts 28 Motor vhcls & other trnsprt eqpmnt 28 Motor vhcls & otner trnsprt eqpmnt 28 Motor vhcls & other trnsprt eqpmnt 28 Motor vhcls & other trnsprt eqpmnt 28 Motor vhcls & other trnsprt eqpmnt 28 Motor vhcls & other tmsprt eqpmnt 28 Motor vhcis & other trnsprt eqpmnt 29 Industnal machmery 29 Industrial machinery 29 Industrial machinery 29 Industna! machinery 29 Industrial machinery 30 Electronic equipment and appliances 30 Electronic equipment and appliances 30 Electronic equipment and appliances 30 Electronic equipment and appliances 30 ElectrOnic equipment and appliances 30 Electronic equipment and appliances 31 Photographic & sCientific eqpmnt 31 Photographic & sCJentific eqpmnt 32 Fumiture 33 Other manufactures 33 Other manufactures 33 Other manufactures 33 Other manufactures 33 Other manufactures 33 Other manufactures 34 Services incidental to manufacturing 35 Electricity 35 ElectriCity 36 Water supply 48 Commodity Description 16 Textiles 16 Textiles 17 Clothing and footwear 1 7 Clothing and footwear 17 Clothing and footwear 17 Clothing and footwear 1 8 Wood products 1 8 Woad products 18 Wood products 18 Wood products 1 8 Wood products 1 8 Wood products 19 Paper products 19 Paper products 20 Printing, pblshng & recorded media 20 Pnnting, pblshng & recorded media 20 Printing, pblshng & recorded media 21 Petroleum products 22 Chemical, rubber and plastiC products 22 Chemical, rubber and plastiC products 22 Chemical, rubber and plastiC products 22 Chemical, rubber and plastic products 22 Chemical, rubber and plastiC products 22 Chemical, rubber and plastiC products 22 Chemical, rubber and plastiC products 22 Chemical. rubber and plastiC products 23 Non metallic mineral products 23 Non metallic mineral products 23 Non metalliC mineral products 23 Non metalliC mineral products 23 Non metallic mineral products 23 Non metalliC m ineral products 24 Basic metals 24 Basic metals 24 BaSIC metals 25 Strctrl, sheet & fbrctd metal prdcts 25 Strctn, sheet & fbrctd metal prdcts 25 Stretrl , sheet & fbrctd metal prdcts 25 Strctr1 , sheet & fbrctd metal prdcts 26 Motor vhcls & other tmsprt eqpmnt 26 Motor vhcls & other tmsprt eqpmnt 26 Motor vhcls & other tmsprt eqpmnt 26 Motor vhcis & other trnsprt eqpmnt 26 Motor vhcls & other tmsprt eqpmnt 26 Motor vhcls & other tmsprt eqpmnt 26 Motor vhcls & other tmsprt eqpmnt 27 Industrial machinery 27 lndustnal machinery 27 Industnal machinery 27 Industnal machinery 27 Industnal machinery 28 Elctmc, phtgnphc, scntfc eqp & pplnes 28 Elctrnc, phtgrphc, sentfc eqp & pplnes 28 Elctrnc, phtgrphc, scntfc eqp & pplnes 28 Elctmc, phtgrphc, scntfc eqp & pplnes 28 Elctmc, phtgnphc, scntfc eqp & pplncs 28 Elctmc, phtgrphc, scntfc eqp & pplnes 28 Elctrnc, phtgnphC, scntfc eqp & pplnes 28 Elctrnc, phtgrphc, scntfc eqp & pplncs 29 Furniture 30 Other manufactures 30 Other manufactures 30 Other manufactures 30 Other manufactures 30 Other manufactures 30 Other manufactures 30 Other manufactures 31 Electricity 31 Electricity 32 Water supply 35 Commodity Description 16 Textiles, clothing and footwear 16 Textiles, clothing and footNear 16 Textiles, clothing and footwear 16 Textiles, clothing and foot'Near 16 Textiles, clothing and footNear 16 Textiles, clothing and footwear 17 Wood and paper products 17 Wood and paper products 17 Wood and paper products 17 Wood and paper products 17 Wood and paper products 1 7 Wood and paper prOducts 17 Wood and paper prOducts 17 Wood and paper prOducts 17 Wood and paper products 1 7 Wood and paper products 1 7 Wood and paper products 1 8 Ptrlm, chem, rubber & plastiC prdcts 1 8 Ptr1m, chem, rubber & plastiC prdcts 18 Ptr1m, chem, rubber & plastiC prdcts 18 Ptrlm, chem, rubber & plastiC prdcts 18 Ptr1m, chem, rubber & plastiC prdcts 18 Ptr1m, chem, rubber & plastiC prdcts 18 Ptnm, chem, rubber & plastiC prdcts 18 Ptr1m, chem, rubber & plastiC prdcts 18 Ptnm, chem, rubber & plastiC prdcts 19 Non metalliC mineral products 19 Non metalliC mineral products 19 Non metallic mineral products 19 Non metalliC mineral products 1 9 Non metallic mineral products 1 9 Non metallic mmeral products 20 Basic metals 20 BaSIC metals 20 Basic metals 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mcnnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 21 Fbrctd metal prdcts, mchnry & eqpmnt 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 22 Other manufactured products 23 Electricity 23 Electncl/y 24 Water supply 547 Table E.4 Commodity Definitions Concordance (Continued) 210 Commodity Oescnptlon 141 Pre-erection work 142 Residential building cnstrctn 143 Non-residential bUilding cnstrctn 144 Civil engineering 145 Prefabricated constructions 146 Other installation work 147 Plumbing 148 Electncal installation work 149 Fencing 150 Building completion work. 151 Land and land Improvements 152 Wholesale trade 153 Retail trade 154 Repair srvcs to machinery & eqpmnt 155 Accommodation 156 Meal services 157 Takeaways 158 Beverage services 159 Road passenger transport 160 Road freight transport 161 Supporting srvcs for road transport 162 Sea, water and rail services 163 Air transport, other transport and storage serviCes 164 Commumcatlon sef\llces 165 Finance 166 Lite msurance 167 Superannuation services 168 Health Insurance 169 General msurance 170 Services to finance and Insurance 17 1 Leased commerCIal property SNCS 172 Leased residential property services 173 Other real estate services 174 Owner-occupied dwellings 175 Equipment hire 176 Computer software and services 1 77 Legal services 1 78 Accounting services 179 Taxation services 180 Architectural and engineering srvcs 181 Advertlsmg and marketing 182 Management consultancy 183 Research and development 184 Placement and supply of personnel 185 I nvestIgation and secunty serv,ces 1 86 Cleaning 1 87 Photographic services 188 Other bUSiness services 189 Central govemment administration services 190 Public order, safety and defence 191 Local government administration services 192 Preschool education 193 Pnmary education 194 Secondary education 195 Higher educatIon 196 Other education services 197 Hosplta! and r'lurslng care 198 Medical, dental & other health srves 199 Vet services 200 Accommodation for the aged 20� Other SOCial services 202 Child care services 203 Motion picture, radiO, TV services 204 News agency services 205 Libraries, museums, art 206 Sport and recreation services 207 Other prsnl & other community srvcs 208 Sewerage services 209 Waste disposal 210 Direct purchases abroad by reSidents 64 Commodity Descnption 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 37 Construction 38 Wholesale trade 39 Retail trade 40 RepaIr srvcs to machinery & eqpmnt 41 Accmmdtn, rstmt & bar srvcs 41 Accmmdtn, rstmt & bar srvcs 41 Accmmdtn, rstmt & bar SIVCS 41 Accmmdtn, rstmt & bar srves 42 Road transport services 42 Road transport servIces 42 Road transport seMces 43 Water and rail transport services 44 Air transport, other transport and storage services 45 Communication services 46 Finance services 47 Insurance services 47 Insurance services 47 Insurance services 47 Insurance services 48 Services to finance and insurance 49 Real estate servIces 49 Real estate services 49 Real estate services 50 Owner-occupied dwellings 51 Equipment hire services 52 Computer services 53 Legal and accounting services 53 Lega( and accounting services 53 Legal and accounting services 54 Archltectura! and englneenng services 55 AdvertiSing and mat1