Household spending is the driving force behind economic booms and recessions. Understanding the determinants of consumption is therefore of utmost importance. This proposal will look at the effect of labor market risks on household consumption and saving. Consumers who perceive to live in risky times may behave cautiously and defer spending; they may also respond to risk by working longer hours. The aggregate and welfare effects of risk may therefore be substantial.
More specifically, the proposal clarifies the distinction between the various kinds of risks that workers face in the labor market. It distinguishes between productivity risk and unemployment risk. This distinction is often not well understood. Productivity risk is individual-specific uncertainty that exists regardless of the employer's characteristics. For example, health shocks affect individual productivity and hence wages in all kinds of jobs. Unemployment risk captures instead the uncertainty about having a job. This includes the possibility of firm bankruptcy, the risk of being laid off and that of not receiving offers while unemployed. Moreover, if different firms pay different wages, unemployment risk will be magnified by firm heterogeneity, because unemployed workers may reject offers from bad firms and stay unemployed longer. The proposal also aims at quantifying the extent of productivity risk and unemployment risk using individual panel data. Correct identification of the two risks depends crucially on the assumptions one makes about the economic environment confronted by workers. Simple models may lead to wrong inference concerning the quantitative importance of the two risks. For example, suppose one makes the standard assumptions that job offers are received with certainty and all firms offer the same wage. In this case there will be no unemployment risk as such because unemployment would be entirely voluntary. To make the distinction between productivity risk and unemployment risk meaningful the author proposes a model with saving and endogenous labor market participation and intra-firm mobility. He also allows for government insurance programs such as unemployment and disability insurance, and the Food Stamp programs, so that behavioral responses to government insurance are built in his model through changes in participation and changes in savings. The exercise we propose may have important implications. When formulating public policy it is imperative to have an idea of what risks matter. With a quantitative assessment of the risks facing agents it is possible to use economic theory to guide the construction of optimal social insurance arrangements.
Broader impacts: Once the various sources of risk have been identified, one can simulate individual saving behavior and measure the welfare costs of the various risks. These simulations give an indication of the extent of precautionary behavior (both precautionary saving and precautionary labor supply). They also raise a number of questions that can be addressed: How much would individuals be willing to pay to avoid the various risks? How much of the precautionary response is due to unemployment risk and how much to productivity risk. The investigator plans to evaluate the welfare benefits of existing government insurance programs, as well as those of proposed reforms and of alternative programs.
