Profiling long-term unemployment utilising the logit model : a New Zealand case study : a thesis presented in partial fulfilment of the requirements for the degree of Master of Applied Economics at Massey University

Abstract

This study attempts to fit the logit model to a random sample of data compiled by the New Zealand Employment Service on individuals who have completed unemployment spells, over the period 1988-1997. The objective is to estimate the probability that an individual job seeker, with a certain set of personal attributes, will become long-term unemployed. The regression results are consistent with a priori expectations. However, the predictive power of the model is low, lending support to conclusions from other empirical studies that have used other approaches to modelling long-term unemployment in New Zealand. That is, the current set of personal attributes on which data arc collected in New Zealand are inadequate for modelling long-term unemployment.