On the drivers of fiscal consolidation attempts : a Bayesian approach : a thesis presented in partial fulfilment of the requirements for the degree of Master of Business Studies in Economics at Massey University
In this paper we investigate which factors determine successful fiscal consolidation attempts. The literature on fiscal consolidation suggests a variety of fiscal, macroeconomic and political factors that are likely to influence the success rate of fiscal adjustment attempts. We focus on 31 of these aspects, as independent variables, and analyse their influence on a binary dependent success variable in a Bayesian model averaging framework for generalised linear models. The applied dataset includes 18 OECD countries and comprises the time period from 1975 to 2009. Bayesian model averaging estimates the generalised linear models with the highest posterior probabilities. By applying this method we can test for model uncertainty. As the posterior probabilities for the best models are, however, quite low we emphasise the posterior probabilities of the individual variables being included in the model. We find that the three variables that control for government social spending, the monetary stance and the formation of the government are highly likely to be included in the best model. Accordingly, a decrease in government social spending, an expansionary monetary policy and a unified government are supportive of successful budget consolidation. These results are mostly in line with findings in the previous research. Finally, several robustness checks confirm that the three variables have a high explanatory power for the dependent success variable.