Estimating the causal effect of income on health: Evidence from post-reunification Germany

In this paper we investigate if there was a causal effect of changes in current and 'permanent' income on the health of East Germans in the years following reunification. Reunification was completely unanticipated and therefore can be seen as a providing some exogenous variation, which resulted in a substantial increase in average household incomes for East Germans. Our data source is the German Socio-Economic Panel (GSOEP) between 1991 and 2002, and we fit fixed-effects estimators to our ordinal health measures. Whilst the exogeneity of reunification allows us to establish the causality between income and health, the fixed-effects methodology additionally enables us to control for individual unobservable heterogeneity such as parental background and general attitudes to health. We also provide new evidence on how major life-events impact on health, and we pay close attention to the issue of panel attrition, given that there might be endogenous exits from the panel if the unhealthy are more likely to drop out of the sample. Using cross-sectional variations in income and health we find evidence of a significant positive effect of current income on health. However, after controlling for heterogeneity and using a new decomposition of the fixed-effects estimates, we find no evidence that increased income leads to improved health. This is the case with respect to current income and a measure of 'permanent' income and two alternative definitions of health