Working Papers

Patient vs. Provider Incentives in Long Term Care (with Martin B. Hackmann)

Abstract: We study how patient and provider incentives affect the mode and cost of long term care. Our analysis of 1 million nursing home stays yields four main insights. First, Medicaid-covered residents prolong their stays instead of transitioning to community-based care due to limited cost-sharing. Second, nursing homes shorten Medicaid stays when capacity binds to admit more profitable residents who pay out-of-pocket. Third, longer stays for marginal Medicaid beneficiaries do not improve health outcomes on average, pointing to annual overspending of $0.9bn. Fourth, transitioning from fee-for-service to episode-based provider reimbursement is more effective than increasing resident cost-sharing in shortening Medicaid stays.

Time Trends Matter: The Case of Medical Cannabis Laws and Opioid Overdose Mortality

Abstract: Mortality due to opioid overdoses has been growing rapidly in the U.S., with some states experiencing much steeper increases than others. Legalizing medical cannabis could reduce opioid-related mortality if potential opioid users substitute towards cannabis as a safer alternative. I show, however, that a substantial reduction in opioid-related mortality associated with the implementation of medical cannabis laws can be explained by selection bias. States that legalized medical cannabis exhibit lower pre-existing mortality trends. Accordingly, the mitigating effect of medical cannabis laws on opioid-related mortality vanishes when I include state-specific time trends in state-year-level difference-in-differences regressions.

Health Shocks, Human Capital, and Labor Market Outcomes (with Francisco Parro)

Abstract: Health, human capital, and labor market outcomes are linked through complex connections that are not fully understood. We explore these links by estimating a flexible yet tractable dynamic model of human capital accumulation in the presence of health shocks using administrative data from Chile. We find that (i) human capital mitigates the negative labor market effects of health events, (ii) these alleviating effects operate through channels involving occupational choice, the frequency of exposure to health events, and access to health care, and (iii) the effect of health shocks on labor market outcomes is heterogeneous across industries and types of diagnoses.

Media and policy coverage: Upjohn Institute

Targeting Policies: Multiple Testing and Distributional Treatment Effects (with Steven F. Lehrer and Kyungchul Song) [Online Appendix]

Abstract: Economic theory suggests that individual responses to welfare reform depend on pre-treatment labor supply and other characteristics. Resulting changes in earnings may be positive or negative, leading to heterogenous treatment effects. In this paper, we extend the literature on treatment effect heterogeneity by introducing six nonparametric tests that are implemented using bootstrap testing of functional inequalities. The proposed tests treat treatment effect heterogeneity as a multiple testing problem and make corrections for the family-wise error rate. To facilitate comparisons to the existing literature we reexamine the extent of heterogeneity in labor supply responses to the Jobs First welfare experiment across both quantiles of outcome distribution for the full sample as well as individual subgroups. Our results shed new light on who truly benefits from welfare reform and demonstrate the importance of correcting for multiple testing.

Minimum Wages and Healthy Diet (with Kathryn L. Clark and Ryan C. Thomas)

Abstract: A healthy diet is often unaffordable for low-income individuals, so income-lifting policies may play an important role in not only alleviating poverty but also in improving nutrition. We investigate if higher minimum wages can contribute to an improved diet by increasing consumption of fruits and vegetables. Exploiting recent minimum wage increases in the U.S. and using individual-level data from the Behavioral Risk Factor Surveillance System we identify the causal effect of minimum wage changes on fruit and vegetable intake among low-wage individuals in a triple-differences framework. Our results indicate that higher minimum wages contribute positively but moderately to improved nutrition.

Labor Market Effects of Medical Innovation: The Case of Breast and Prostate Cancer (with Sung-Hee Jeon) [email me for draft]

Abstract: Innovations in cancer treatment have lowered mortality, but little is know about their economic benefits. In this paper, we assess the effect of improved treatment options over the last three decades on the labor market outcomes of breast and prostate cancer patients. We combine administrative tax return and cancer registry data from Canada with measures of medical innovation (approved drugs, academic publications, and patents) to estimate triple-differences regressions of employment and annual earnings. Our results show that the reductions in these labor market outcomes among cancer patients are partially offset by improved treatment options. Specifically, the decline in employment due to a cancer diagnosis is cut by 50 to 75 percent by the medical innovation that has occurred during the last three decades.

Wage Premiums, Shirking Deterrence, Gift Exchange and Employee Quality: Firm Evidence (with Constanca Esteves-Sorenson and Ernesto Freitas) [email me for draft]

Abstract: Why may firms pay efficiency wages: above-market clearing wages which are not contingent on output? We explore the contribution to productivity of three main mechanisms underpinning efficiency wages (shirking, gift-exchange, and selection) by exploring a novel personnel data set from two Portuguese hotel chains as well as particular features of the Portuguese labor market. We find evidence of the three mechanisms. In the first firm, higher wage premiums at the time of hiring decrease shirking, consistent with selection. Higher yearly wage premiums also reduce shirking, consistent with the shirking and gift-exchange models. However, tenure induces an increase in shirking which is consistent with the shirking model.