1  Introduction

This paper describes in detail the estimation of preference functions for hours of work and income for four subgroups of the New Zealand population, from which the expected hours of labour supply can be derived. The groups are the following: couples with and without children, single men, single women, and sole parents. Each of these groups is relatively homogenous and we specify one separate utility function for each group. The four groups together add up to a sample representing the New Zealand working-age population.

No individual structural labour supply models have been estimated for New Zealand. Chiao and Walker (1992) estimated a discrete choice model using four earnings levels instead of a choice model based on different levels of labour supply and Maloney (2000) estimated a reduced form labour supply and participation equation based on average labour supply and participation of groups of individuals with similar characteristics. The lack of information on earnings in the data he used complicated his analysis.

The model in this paper allows for the presence of fixed costs associated with working and for observed and unobserved heterogeneity in preferences for labour supply and income. A similar specification has been estimated for Australia (Kalb, 2002) allowing a comparison with the results from New Zealand. In addition, the estimation of similarly specified models for different groups allows us to compare the effect of characteristics on labour supply across the different demographic groups.

The emphasis of the basic framework is on the separation of income into different categories and on a correct representation of net income at all levels of gross income, taking taxes and benefit withdrawal rates into account. This results in a highly nonlinear and non-convex budget set which differs for each individual. Estimation of a continuous labour supply model for two persons, using this budget constraint, is complicated and computationally intensive, which is one of the reasons to discretise labour supply for all groups.[1] This simplification with regard to hours of work allows us to take the full details of the benefit and tax system into account. Following Van Soest (1995), we use a multinomial logit specification in the discrete choice model, which allows us to choose a relatively large number of labour supply points for both adults in the household.

The models estimated in this paper can be used to simulate the behavioural effects of policy changes, as long as these policy changes are of a financial nature, affecting net household income levels, such as for example, a change in the withdrawal rate of a benefit payment, a change in the level of payment or a change in eligibility rules. For a policy simulation, the model is usually calibrated to make sure that the prediction in the starting situation is equal to the observed situation.

Section 2 briefly discusses the economic model. Section 3 describes the data. Section 4 contains the econometric details. The results from the models for the different groups are discussed in Section 5. First the estimated parameters are discussed and then predicted labour supply using the estimated parameters is presented. Finally, in Section 6 some conclusions are presented.