Self-insurance and social insurance over the life-cycle

Lead Research Organisation: University of Cambridge
Department Name: Economics


Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Description The project analysed how the different risks that individuals face over their lifetimes affect behaviour, and how government provided insurance can mitigate those risks and change individual behaviour. The focus was on incentives to save and to work across the life-cycle: labour supply and saving when young is affected, for example, by insurance against bad health and insurance against unemployment in later life; similarly insurance against bad outcomes early in life may reduce effort and affects the accumulation of skills and saving which would have been of benefit later in life. These dynamic effects were analysed by computer simulations of life-cycle behaviour with realistic characterisation of the government's social insurance programs.
This led to the following findings about social and self-insurance:
1) On social insurance:
a. The welfare value of means-tested programs such as Food Stamps, which can insure persistent shocks, is substantially greater than the value of unemployment insurance which provides (partial) insurance against unemployment shocks, which are transitory
b. Disability insurance is characterised by high rejections rates of disabled applicants; acceptances of healthy applicants is less widespread. Relaxing the strictness of the assessment is beneficial despite the incentive costs because of lower rejection rates among the severely work limited.

3) On self-insurance and the effects of risk:
a. Savings rates across all age groups rise in recessions. This rise happens because of greater uncertainty in recessions not because of an inability to access credit markets.
b. Divorce leads to large falls in living standards for women, but not for men. Benefits provide insurance for those households that were already poor prior to divorce.
Exploitation Route To academics, the framework for thinking about uncertainty and social insurance can be used to simulate policy reforms. The code alongside the papers provides the means to do this and for the framework to be developed.
To non-academics, the key lessons are in diverse applications: eg. the macro effects of micro behaviour under uncertainty is a different way of thinking about the current crisis. The joint anaylsis of different social policies in one framework highlights the interaction between different policies, such as unemployment benefit and disability insurance. This is a key lesson from the simulations of policy reform. The work on the implications of divorce can be used for reform of the basis of financial settlement on divorce.
Sectors Communities and Social Services/Policy,Government, Democracy and Justice

Description Each of the main applications of my research from the fellowship have had impact in the wider community. I describe here the impact of each of these applications. Central banks often focus on aggregate numbers and ignore the heterogeneity among individuals. My research using micro data in a macro context has shown how different individuals respond differently to the same aggregate shocks. In the most recent recession, the key debate that I have been involved in with central banks has been whether the lack of consumption is due to a fall in the supply of credit or due to a lack of demand for credit. The impact of this work has been on the way policy makers and researchers in Central Banks think about aggregate risk and on the importance that central banks place on the use of micro data to understand macro phenomenon. This is an ongoing agenda that I am pushing. The central message of work on social insurance is that the value of insurance depends on the persistence of shocks because more persistent shocks are where self-insurance through saving is less feasible. This means that income support is much less valuable than unemployment insurance, for example. This has helped changed the prior that unemployment insurance was a key social insurance programme. It has also focused attention on the interaction between different social insurance programs. Similarly, my work on disability insurance has focused attention on the trade-off in disability insurance between improving incentives and the insurance value. Much of the policy debate has been focused on the false claimants of disability insurance and how to mitigate these incentive costs. My research balanced this focus on the incentive costs against the value of the insurance. In the current climate, it is not a message that has been easy to sell, but I have argued that this is about how we should think of the problem. My research on the economic consequences of divorce has shown the inequalities between men and women that result from divorce. It has also shown the ways that individuals mitigate the costs of divorce. This has fed into the discussion of what the basis should be of financial settlements on divorce, and also into the discussion of whether there should be clear formula for the division of financial assets on divorce in order to protect women.
Sector Communities and Social Services/Policy,Education,Government, Democracy and Justice
Impact Types Economic