Academic Summary for Final Report

Focus of Project

Faced with aging populations and the looming insolvency of social security, governments around the world are grappling with the issue of how to raise employment. One avenue is retirement reform, which is geared toward keeping older workers employed longer. Given the severity of the issue, we should, however, ask how to boost employment at all ages. In order to make any policy suggestions we must first understand the driving forces behind labor supply behavior, and in particular how individuals respond to changes in government programs. Our focus has been on developing heterogeneous-agent life cycle models to study the effect of various government policies – notably retirement systems and the government provision of family services – on individual and household labor supply.

Implementation

Our original proposal included the study of three types of government programs: (1) retirement programs and their effect on the employment of older workers, (2) the government provision of family services and its effect on household time allocation, fertility and the gender wage gap, and (3) the financing of higher education and its effect on youth employment. Parts (1) and (2) unfolded very much as planned. This has already yielded two high quality journal publications. The first-stage of the project spurred a new collaboration with researchers from the Frisch Centre in Oslo. Our expertise is on structural modeling, while theirs is on reduced form analysis. This felt like a good opportunity for us to benefit from their deeper knowledge and good access to data, while introducing them to life cycle modeling. This collaboration resulted in a working paper, which was recently submitted to a journal. In light of this new avenue of research, stage-three of our original project is still in the relatively early stages. We have explored the relevant data and are currently developing the model. In our research, we first develop a model and then use that framework to gain insights. Our work has a distinct quantitative focus. This type of research is computationally intensive and time consuming, yet highly policy relevant.

Important Research Findings

In the first part of the project, we developed a life cycle model of labor supply and health to study the role of old-age pensions, disability insurance and health insurance, in accounting for the large differences in employment rates of older individuals across OECD countries. Our model features heterogeneous agents who make decisions regarding employment and the applications for old-age pension/disability insurance benefits. The granting of disability insurance benefits (conditional on applying) is imperfectly correlated with health. Agents can partially insure against health shocks by investing in health.

Older workers face very different incentives for continued employment across OECD countries. We find that generous social insurance programs create incentives for early retirement in France, Spain, Sweden, the Netherlands and Denmark. With the exception of France, this is to a substantial degree driven by generous disability insurance schemes. Conversely, the model predicts high employment rates of older workers for Australia, Canada, the U.K. and Germany. The existence of public health insurance depresses labor supply in all of the studied countries. Despite this dampening effect, the model predicted employment rates for older workers in Canada, the U.K. and Germany are somewhat too high relative to the data. On the contrary, the model predicts somewhat too low employment of older workers in Spain, Sweden, Denmark and the Netherlands. Given our findings, the puzzle is not, why do Europeans work so much less than Americans, but why, given the incentives built into the social insurance programs in place, do some Europeans work as much as they do. Moreover, viewed through the same lens, it is equally puzzling why people in Canada and the U.K. do not work more than they do.

This project was published in the Review of Economic Dynamics, which is one of the preeminent field journals in macroeconomics.

In the second stage of the project, we study the effect of family policies on female employment, fertility and the gender wage gap. We develop a life cycle model of heterogeneous households featuring endogenous labor supply, human capital accumulation, fertility and home production. The model is calibrated to Swedish data and institutions. Yet, the insights are general.

Our results suggest that human capital accumulation is important in accounting for the widening of the gender wage gap following children. We find that, in aggregate, childcare subsidies promote maternal employment and fertility, although the effects are heterogeneous across couples. A reduction in the childcare subsidy leads some women to work more and have fewer children, while others continue to have several children but stay at home longer. We find that labor supply and fertility decisions are very much linked. The prevalence of part-time work, and the resulting widening of the gender wage gap following children, has been linked to high female hours of home work. This motivated us to study the labor supply implications of introducing a subsidy on the market input in home production. We find that such a subsidy does lead to more market work and less home work, but the increase in market work stems primarily from the retirement margin. Consequently, it does not lead to a narrowing of the gender wage gap. Moreover, while part-time work contributes to the widening of the gender wage gap, in the absence of generous part-time work options, female employment would be lower and the gender wage gap would widen even more.

This paper is forthcoming in the Scandinavian Journal of Economics, which is a high quality general interest journal.

Currently, we are working on an extension of this analysis to a cross-country setting in order to study whether the differences in government policies can account for the cross-country differences in female employment and fertility rates. The coding is complete and the data moments are in place. We are currently calibrating the model.

Our work on this project caught the attention of researcher at the Frisch Centre. Together we delve into questions pertaining to inequality in health and longevity. In many countries, solvency of social security in the future will require either that people work longer or that benefits are cut/taxes are raised. Population aging is largely driven by improvements in longevity. However, there is substantial heterogeneity in the life expectancy, and also health, of older people. More educated individuals enjoy better health, and can expect to live longer, than their less educated counterparts. Yet, in most countries these heterogeneous workers face homogenous social security rules.

We develop a dynamic, structural life cycle model of heterogeneous agents who face health and mortality risk. We use the model to study the labor supply and welfare effects of alternative pension reforms. In particular, we contrast different ways of making pension schemes robust to improvements in longevity. We consider the following policy reform measures: (1) increasing the early access age to old-age retirement, (2) raising income taxes, (3) lowering old- age retirement benefits and (4) lowering old-age retirement and disability benefits. We find that, of the considered policies, proportionally lowering old-age retirement and disability benefits results in the highest average welfare for all education categories. It is also the most successful at boosting employment.

Three Important Takeaways from Our Project

(1) The previous literature has considered it a puzzle why Europeans work so much less than Americans. We find that older workers face very different incentives for continued employment across OECD countries. In fact, in light of the generous social insurance in place in many European countries, it is surprising that Europeans work as much as they do.

(2) We find that modeling the full life cycle, and endogenizing fertility and male employment, are important when studying female employment. These aspects have been largely ignored in this literature.

(3) When studying retirement reform, it is important to include a disability insurance channel, as changes to old-age retirement can have the unintended consequence of increasing disability benefit claiming. With a few exceptions, the literature on retirement has evolved separately from the literature on disability.

New Research Questions

As noted previously, our work in the first stage of the project on the effect of social insurance on retirement entry lead us to collaborate with researchers at the Frisch Centre. That paper deals with questions pertaining to solvency of social security and specifically the differential effects of alternative policy reforms on individuals who are heterogeneous in productivity, health and life expectancy.

International Dimensions of Project and Dissemination of Research
We have presented our work at several international conferences, workshops and seminars:

Society for Economic Dynamics conferences in Seoul and Toulouse, Arizona State University Reunion Conference, Mannheim Workshop on Quantitative Macro, Goethe University, Sabanci University, IAE/UAB Barcelona, DIW Berlin, NORMAC in Iceland and Smögen, Nordic Workshop on Tax Policy, SOFI, Greater Stockholm Macro Group, Swedish National Conference, and Bank of Lithuania.

This allowed us to get feedback from leading researchers at other institutions. Again, our work on this project caught the attention of researchers at the Frisch Centre and led to a project on retirement reform and longevity. Wallenius was invited to the Finnish and Swedish Ministries of Finance and to various policy institutions, allowing our work to become known also in policy circles. Additionally, Wallenius presented her research to the SSE Corporate Partners.