The purpose of disability insurance (DI) is to protect people who develop functional impairments that limit their ability to work. In this project, we evaluate the effectiveness of DI benefit programs in delivering this protection by following people’s health and financial wellbeing after the take-up of disability insurance benefits. Since the mid-1990s, there have been incisive reforms to reduce the generosity of the DI systems in many countries. They mainly lowered DI generosity along two dimensions: Stronger screening mechanisms with stricter eligibility rules aiming at reducing the number of DI claimants and lower replacement rates aiming at decreasing the amount of DI payments. A key question is whether these generosity reductions have an impact on health and well-being.
This paper takes advantage of internationally harmonized panel data and the differences across DI programs in Europe and the United States, as well as their changes over time. For this purpose we harmonize data from three different surveys for the years 2004 to 2015: The Survey of Health Ageing and Retirement in Europe (SHARE), The English Longitudinal Study of Ageing (ELSA) and the US Health and Retirement Study (HRS).
The estimation of the causal effect of DI benefit receipt on physical and mental health as well as for psychological and financial well-being is challenging because of two underlying econometric problems. First, there is reverse causality due to the fact that DI benefits may not only change health and well-being but DI benefit uptake is also determined by health status. Second, even with modern microdata at hand, there are unobserved variables that influence both DI uptake and health. This creates a selectivity problem since the initial health status of those who receive DI benefits and of those who do not may not be observed. We several econometric approaches, specifically instrumental variable estimation and fixed effects estimation, to account for the potential endogeneity of DI enrollment and sample selectivity.
We find that self-reported health stabilizes after DI benefit receipt. Mental health improves more for DI benefit recipients than non-recipients relative to the beginning of DI benefit receipt. This effect is stronger in countries with more generous DI systems. The effects on objective health measures are positive but largely insignificant.
The project was funded by the US Social Security Administration. A report has been submitted in September 2017. Currently the paper is revised and submitted to conferences.