Essays on Labor Supply and Health


This dissertation is comprised of three self-contained chapters in applied micro-econometrics. All three chapters use Danish register data to answer empirical questions within the topics of labor supply and health.

Health, Disability Insurance, and Retirement in Denmark

The aim of the first paper is twofold. First to measure the effects of pension program provisions on retirement age controlling for health based on register and SHARE health data and secondly to assess the importance of the disability insurance program for retirement controlling for health. A 1999 pension reform rolled out through 2006 brought the largest changes to Danish retirement incentives in 20 years. Early retirement for the insured was made less generous, old age pension age was reduced, and disability insurance awards became more stringent. We characterize each of these changes using an option value model for the most important routes to retirement. Controlling for health, greater stringency in awarding disability insurance has modest effects on retirement, whereas reforms to incentives in an early retirement program had large effects on retirement age, probably due to the relative magnitudes of the reforms.


How Does a Health Shock to Self or Partner Affect Economic Incentives to Retire?

The second paper exploits rich administrative data to analyze how a health shock to one-self or one’s partner affects the marginal economic incentive to retire. One important challenge when trying to get unbiased estimates of the effect of health on retirement is the endogenous character of the relationship between health and retirement. To circumvent this endogeneity, the paper exploits a 1999 reform of an early retirement program and the arguably unanticipated timing of a health shock. The main conclusions are that economic incentives have strong effects on retirement for both men and women, but that those who receive a health shock are much less responsive to economic incentives. A comparison of men’s and women’s responses finds no differences in men and women’s marginal responses to economic incentives due to a health shock to their partners–only when the health shock hits themself, then women react much stronger than men in their reduction to the economic incentive.

Can we reform Disability Insurance and increase labor supply?

The third paper investigate the effects of attempts to mitigate moral hazard in the DI program by raising the stringency of the screening process for applicants. We measure the effect on DI awards and labor supply of two DI reforms that took place in Denmark during the 1990s: (1) state audits of local awards, (2) reduction in state rebates to local authorities for DI awards to applicants aged 60-66. Benefit generosity at the individual level remained unchanged throughout the estimation period, keeping economic incentive effects stable. State audits of local county boards showed that almost no counties awarded too few DI pensions according to the state audit board, whereas many awarded too many, and when these high awarding counties were audited, they increased rejection rates the following year by on average 21%. Reduced state rebates to municipalities for DI awards for those aged 60-66 increased rejection rates by 5 percentage points for this age group compared to those aged < 60. Instrumental variables estimates show that a 10% rise in rejection rates leads to a 6.4% increase in labor supply.