Health effects on labour market exits and entries

Abstract

This paper analyses the role of health on exits out of and entries into employment using data from the first twelve waves of the British Household Panel Survey (1991–2002). We use discrete-time duration models to estimate the effect of health on the hazard of becoming non-employed and on the hazard of becoming employed. The results show that general health, measured by a variable that captures health limitations and by a constructed latent health index, affects entries into and exits out of employment; the effects being higher for men than for women. The results are robust to different definitions of employment, and to the exclusion of older workers from the analysis.

Introduction

In many developed countries there is increasing concern about the fiscal implications of an ageing population. This has prompted discussion about potential changes in financial incentives geared towards discouraging early retirement and/or postponing the retirement age. In the UK, for example, the age of eligibility for the State Pension is set to increase from the current age of 65 years for men and 60 years for women to 68 years for both men and women by 2046. Financial incentives alone, however, may not be sufficient if individuals chose to cease working due to shocks or deteriorations in health. This is particularly relevant for younger age groups, as they will not be affected by supply side policies targeting individuals approaching retirement.

The majority of the evidence on health and labour outcomes (Bound et al., 1999, Au et al., 2005, Disney et al., 2006, Hagan et al., 2006, Rice et al., 2006, Zucchelli et al., 2007) has focused on the role played by health in retirement transitions. These studies show that decreases in health status have explanatory power for retirement decisions. In addition, there is limited evidence that the onset of a health shock decreases the probability of employment for younger individuals, who mainly transit into inactivity (García-Gómez and López-Nicolás, 2006, García-Gómez, 2008). This suggests a dual role for policies aimed at retaining workers within the labour force with, perhaps, separate incentives geared towards encouraging older and younger workers to continue to supply labour. In this respect, understanding the role of health, and in particular shocks to health, in the labour supply decisions of individuals at all ages is of great importance.

In order to design integrated policies that avoid the possible adverse employment effects from disability benefits, further research is needed to understand the importance of health when individuals are deciding either to enter or re-enter the labour market. The limited evidence suggests that individuals with impaired health have longer unemployment spells (Stewart, 2001) and a higher probability of transiting from unemployment to economic inactivity (Boheim and Taylor, 2000). However, the general lack of evidence on the relative importance of health as a determinant of transitions into and out of employment restricts the design of policies aimed at integrating individuals with health problems into the labour market, be it through strengthening policies to aid working individuals who suffer health deteriorations remain in the work force or to help encourage individuals in poor health to enter or re-enter the labour market.

This paper contributes to the existing literature by analysing the effects of a deterioration to health has on entries into and exits from employment. This is the first attempt, to the best of our knowledge, to analyse the relative role that health plays as a determinant of these separate transitions. We use data from the British Household Panel Survey (BHPS) to analyse employment transitions using discrete-time hazard models. Moreover, we focus on the working-age population, rather than a more narrowly defined sample of individuals approaching retirement.

Our analysis explores the effect of changes in both psychological and physical measures of health. Mental health is a leading cause of disability and sickness leave across OECD countries being the second largest category of occupational ill-health after muscular–skeletal problems (Weiler, 2006). In the UK, mental health problems account for 40% of Incapacity Benefit claims and 23% of new claimants of Disability Living Allowance (Lelliott et al., 2008). It is, therefore, important to understand whether and to what extent mental ill-health influences entries into and exit from the labour market and how this might differ from other dimensions of health. However, evidence suggesting that disability benefit receipt in many OECD countries has been increasing in the absence of an overall increase in reported mental health problems among the working-age population (OECD, 2008) would suggest better knowledge of all dimensions of health on labour market transitions is required.

Our empirical findings show that individuals' health is an important determinant of employment transitions, and that the effects are greater for men than for women. However, there are some differences depending on the measure of health used. General health, measured by health limitations and a constructed latent health index increasing in ill-health, has a significant positive effect on exits from employment and a negative effect on entries into employment. Furthermore, a measure of psychological well-being appears to influence positively the hazard of becoming non-employed for the stock sample of workers, but it has also a positive effect on the hazard of becoming employed for the flow sample of non-employed women. The effects are robust to the exclusion of older workers from the analysis.