The Impact of Self-Assessed Health on Labour Supply

Abstract

Health has long been recognised as an influence on labour supply. To date there has been little New Zealand research on this topic. This paper uses census area unit level data from the 1996 Census to estimate the effects of changes in the prevalence of self-assessed disability and health problems on the labour force participation race. The best results were obtained from questions asking about the effect of health on ability to carry out common everyday activities, and having a long-term disability. Despite the use of aggregate data the goodness of fit of the models was low (approximately 0.50). Coefficients on non-health variables were robust to changes in the specification of the health variable. The elasticity of labour force participation rates with respect to ill health varied between -0.02 and 0.02. The use of grouped data is less than ideal, and the impact of different corrections for this data structure is explored. The weighted least squares methods used in this paper have been argued to be inefficient if group sizes vary widely, and this is an area for future research with the current dataset. Future research directions with New Zealand survey data are suggested.