Working Papers
-
Information and the Persistence of the Gender Wage Gap: Early Evidence from California's Salary History Ban
Under Review, with D. McNichols
Media Coverage:
Aiming to reduce the gender wage gap, several states and cities have recently adopted legislation that prohibits employers from asking about previously earned salaries. The advocates of these salary history bans (SHBs) have suggested pay history perpetuates past discrimination. We study the early net impact of the first state-wide SHBs. Using both difference-in-difference and synthetic control approaches, we find the gender earnings ratio increased by 1 percent in states with SHBs. We find these population wide increases are driven by an increase of the gender earnings ratio for households with all children over 5 years old, by workers over 35, and are principally driven by those who have recently switched jobs.
-
Do State Tobacco 21 Laws Work?
Under Review, with C. Bryan, D. McNichols and J. Sabia
Tobacco 21 (T-21) laws prohibit the sale of tobacco products to individuals under age 21. This study is the first to comprehensively examine the impacts of statewide T-21 laws on youth tobacco consumption, including spillovers to minor teens. Using data from the 2009-2019 Behavioral Risk Factor Surveillance Survey (BRFSS) and a difference-in-differences approach, we find that the enactment of a statewide T-21 law was associated with a 2.5 to 3.9 percentage-point decline in smoking participation among 18-to-20-year-olds. A causal interpretation of our estimates is supported by event-study analyses and falsification tests for young adults ages 21 and older. Next, using data from the 2009-2019 State Youth Risky Behavior Surveys (YRBS), we find that statewide T-21 laws reduced tobacco cigarette and electronic cigarette (e-cigarette) consumption among 18-year-old high school students. However, descriptive evidence suggests that the negative impact of T-21 laws on e-cigarette use among 18-year-olds may be partially blunted by an increase in borrowing e-cigarettes from others. Finally, we find that T-21 laws generate important spillovers including (i) a reduction in tobacco cigarette use among 16-to-17-year-olds, a group that relies heavily on the “social market” — including 18-year-old peers — to access tobacco, and (ii) reductions in both marijuana use and frequency of alcohol consumption among older teenagers.
-
Vertical Integration and Production Inefficiency in the Presence of a Gross Receipts Tax
Under Review, with K. Miller and C. Weber
We quantify the effects of a gross receipts tax (GRT) on vertical integration for the first time. We use data from the Washington state recreational cannabis industry, which has numerous advantages including a clean natural experiment: a 25% GRT imposed on cannabis firms was subsequently replaced by an excise tax at retail. We find the short-run elasticity of vertical integration with respect to the intermediate good net- of-tax rate is -0.15 and the long-run elasticity is more than twice as large. We find these incentives lead to large output losses – production increases by 23 percent when the GRT is eliminated.
-
Getting into the Weeds of Tax Invariance
Under Review, with K. Houghton, K. Miller and C. Weber
Media Coverage:
We provide the first general empirical test of tax invariance (TIV). When a 25 per- cent tax remitted by manufacturers was eliminated in Washington state and the retail cannabis excise tax was simultaneously increased from 25 to 37 percent—a shift intended to be revenue-neutral—TIV did not hold. Manufacturers kept two-thirds of their tax savings instead of passing all their savings through to retail firms via lower prices as predicted by TIV. One-third of the retail tax increase was passed on to consumers via higher retail prices—TIV would have predicted constant or even declining tax-inclusive retail prices.
-
Social Insurance Benfit Generosity and Moral Hazard: Evidence from Workers' Compensation
Joint with T. Nguyen and G. Waddell
In this paper, we investigate the effect of benefit generosity on claim duration and temporary benefits paid among temporary disability claims for workers' compensation. While previous studies have focused on natural experiments created by one-time large changes in minimum or maximum weekly benefits, we exploit variation around a kink in benefit generosity inherent in all workers' compensation systems in the United States. Using administrative data on the universe of injured workers in Oregon, we also find that more-generous benefits leads to longer injuries, but with implied elasticities that are smaller than the average elasticity from previous difference-in-difference studies. Our preferred estimates suggest that a 10-percent increase in benefit generosity leads to a 2- to 4-percent increase in injury duration. We derive similar duration-benefit elasticities when studying changes in benefits paid at the kink. We also introduce the first evidence that more-generous benefits encourage subsequent claim filing.
School Year Length and Student Performance: Quasi-Experimental Evidence
This paper investigates the impact of instructional days on student performance. Because school year length is endogenously determined, I estimate the causal impact of school year length through two quasi-experiments that exploit difference-in-differenceerent sources of variation in instructional days. The first identifes school year lengthĂs effect through weather-related cancellations in Colorado and Maryland. Weather-related cancellations are made up at the end of school years, allowing relatively large floatuctuations in instructional days within school districts prior to test administration. Because school cancellations are not recorded for past school years, this data limitation is overcome by using two-sample indirect least squares. The second identification strategy takes advantage of state-mandated changes in test-date administration in Minnesota, which moved 5 times in 5 years. The results are similar for either source of instructional day variation: more instructional time prior to test administration increases student performance. The effects are consistent across various thresholds of performance and grade levels.