Abstract

The paper analyses the impact of retirement on the provision of informal care using a fuzzy regression discontinuity design. Demographic change implies consequences concerning the care for the elderly. Rising life expectancy increases the number of individuals in need for care. At the same time women’s labor market participation is increasing at the extensive and intensive margin. As care is predominantly carried out by women this reduces the supply of informal care. Women face time conflicts by having to decide between their labor supply and care provision.

So far the literature has primarily been focused on the impact of informal care provision on labor supply. The main methodological issue is to circumvent simultaneous causality. This paper closes a research gap by focusing on the choice to provide informal care within the context of retirement decisions when time resources can be re-allocated. I exploit age thresholds in German retirement legislation for identification under simultaneous causality. A fuzzy regression discontinuity design is applied using thresholds at ages 60, 63 and 65 as individual instruments for the retirement decision. I am then able to estimate the causal effect of retirement on the provision of informal care.

SOEP panel data from the years 2001-2013 not only offer a variety of socio-economic variables on individuals and households. The SOEP also contains information on the time use of adults. I can therefore track the individual retirement decisions and time they spend on informal care. The sample is restricted to people providing informal care to a person within the household as information on external care is too noisy.

I find that the share of women providing informal care among all women retiring at the age of 60 increases significantly by around 9 percentage points. In this group women provide 0.5 hours more of care per day due to retirement. For women retiring from a full time position these effects increase to 15 percentage points and around 0.8 hours per day. The estimates are robust to the inclusion of fixed effects. In line with previous research results show no effect of retirement on the care provision of men. The findings are consistent with time conflicts between the supply of labor and informal care in the household among elderly citizens. This points to a need for reform in care and retirement schemes to reconcile labor and care activities and to close the care gap.