Essays in Labor and Applied Microeconomics

Ong, Pinchuan

doi:https://doi.org/10.21985/n2-51zb-1197

Work

Essays in Labor and Applied Microeconomics

Public

Download PDF

Download All Files (.zip)

This thesis explores questions in labor economics and applied microeconomics, with particular focus on issues that have implications for public policy. The first essay estimates the Frisch elasticity, sometimes known as the wage elasticity of labor supply in response to anticipated wage changes. Despite its importance in macroeconomic and public finance models, its estimation requires a setting that is difficult to find; as a result, we have little quasi-experimental evidence on its magnitude. In the essay, I explain why child support---tax-like payments from noncustodial parents towards custodial parents in cases of divorce and nonmarital births---satisfies the two key features lacking in almost all settings that other researchers have looked at in the past. Specifically, we require that individuals anticipate their future effective wage in advance, satisfied because it is common knowledge that child support payments end on emancipation of the youngest eligible child, and exogeneity, satisfied if we believe that the ages of these children (who live away from the payers) do not directly affect the labor supply decision. Exploiting these two features, I estimate the Frisch elasticity in an event study design using individual-level panel data from four countries. Empirically, I find that the observed child support rate that fathers face drops to nearly zero upon emancipation of the children; correspondingly, these fathers increase their work hours and earnings at this time. Based on these results, I estimate Frisch elasticities of 0.7 to 0.8 on the intensive margin and 0.1 to 0.2 on the extensive margin. The second essay, coauthored with Janjala Chirakijja and Seema Jayachandran, shows that lower heating prices reduce mortality in winter months, driven mainly by cardiovascular and respiratory causes. To obtain a causal estimate, we combine the considerable geographic variation across counties in whether an area relies on natural gas or electricity for home heating with temporal variation in the national prices of natural gas and electricity. The effect we find is meaningfully large---US natural gas prices dropped 42 percent in the late 2000s and this averted 11,000 winter deaths per year---and does not just represent short-run hastening of mortality. Because the effect seems to be larger for low-income households, our results have implications for heating-related policies like the federal Low Income Home Energy Assistance Program (LIHEAP). In addition, our findings highlight a health benefit of increases in the U.S. energy supply, and hence have direct implications for energy and environmental policies. The third essay, coauthored with Jason Ward, explores the extent to which seasonal unemployment affects drug use. The unemployment rate varies across months of the year, generally for reasons unrelated to the current macroeconomic conditions and outside the control of individuals. This seasonality in the unemployment rate also varies across industries and occupations, motivating a difference-in-differences strategy. We find that an increase in seasonal unemployment leads to lower legal substance use among working individuals. For illicit drug use, we find increased use of marijuana only among women with income above \$20,000. Because seasonal unemployment is both highly predictable and temporary, it allows us to disentangle the mechanisms involved. We find some evidence that the income effect is larger than the psychological distress caused by unemployment. Our results also suggest that the workplace might facilitate legal substance use, and that the increase in marijuana use among women with income above \$20,000 is due to an increase in leisure time.

Creator