In this paper, four separate basic labour supply models for couples, single men, single women and sole parents are estimated for New Zealand. The preference parameters for labour supply and income and the parameters for fixed costs include observed heterogeneity in the form of the number and age of children in the income unit, age and education of the head and partner (if present), and the place of residence of the income unit. It was found that adding unobserved heterogeneity did not change the estimated values of the other parameters and the unobserved heterogeneity parameters were all very small and insignificant.
The results are similar for all demographic groups. The basic results seem sensible (and similar to results found in other countries, such as Australia and to earlier more aggregate results for New Zealand), with the preference for labour supply highest for people who are in their thirties with a high education level, although education levels (including those of the partner) seem somewhat more important for women than for men. The preference for labour supply is lower for women with children, in particular when the children are young, whereas no effect is found for married men. Finally, the predicted distribution over the different labour supply hours using the point estimates of the parameters is similar to the actual distribution. This leads us to conclude that the four models in this paper seem a good starting point for further experimentation with alternative specifications and extensions in future research and as the basis for policy simulations. However, the model for sole parents may need to be refined further.
The available data from 1991 to 2001 allowed us to include some variables in the preference parameters, which do not usually form part of labour supply models. The most interesting of these was the eligibility for New Zealand Superannuation indicator. In addition to this variable, a time trend, a dummy variable for being 60 years or older and an unemployment rate were included to take out possible other effects that might pollute the eligibility parameter. For all groups, a significant effect is found for the Superannuation indicator, even after controlling for the other possible influences. The delayed eligibility for this state provided superannuation scheme has increased labour supply for all groups. Finally, an indicator for living with one’s parents was included to explain labour supply preferences and found to have the expected sign for sole parents (indicating that living with one’s parents may be a childcare strategy for sole parents), although the effect was not significant.
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.