Knowledge that Transforms

We use newly linked tax records to show that the large responses of UK company owner-managers to personal taxes are due to intertemporal income shifting and not to reductions in real business activity. Around half of this shifting is short-term and helps prevent volatile incomes being taxed more heavily under progressive personal taxes. The remainder reflects systemic profit retention over long periods to take advantage of lower tax rates, including preferential treatment of capital gains. We find no evidence that this tax-induced retention increases business investment. It does, however, substantially reduce the tax revenue raised from high income business owners.

This paper exploits location choices of all generalist physicians who graduated in Brazil between 2001 and 2013 to study policies aimed at increasing the supply of physicians in underserved areas. We set up and estimate a supply and demand model for physicians. We estimate physicians' locational preferences using a random coefficients discrete choice model. The demand has private establishments competing for physicians with private and public facilities around the country. Policy counterfactuals indicate that quotas in medical schools for students born in underserved areas and the opening of vacancies in medical schools in deprived areas are more cost-effective than financial incentives.

This paper examines methods of inference concerning quantile treatment effects (QTEs) in randomized experiments with matched-pairs designs (MPDs). The standard multiplier bootstrap inference fails to capture the negative dependence of observations within each pair, and thus, is conservative. The analytical inference involves estimating multiple functional quantities that requires several tuning parameters. In this paper, we propose two bootstrap methods that can consistently approximate the limit distribution of the original QTE estimator and lessen the burden of tuning parameter choice. In particular, the inverse propensity score weighted multiplier bootstrap can be implemented without knowledge of pair identities.

Little is known about where hotspots of gentrification emerge within a city and the role that some types of businesses play in the process. We develop a method to detect the sectors whose presence heralds the process of gentrification in a neighborhood. We show that these sectors, mostly found in cultural and creative industries, help to anticipate neighborhood change and that their predictive power complements that of traditional gentrification determinants. We also examine mechanisms related to amenities, worker characteristics, and signaling that are consistent with these results. The analysis illustrates the importance of businesses in the sociodemographic dynamics of neighborhoods.

We analyze dynamic incentives in pension systems created by the use of a small set of final years of earnings to compute benefits. Using social security records and household surveys from Uruguay, we show that self-employed workers and some employees of small firms respond to these incentives by increasing reported earnings in the benefit calculation window. We find evidence that suggests that these responses are explained by changes in earnings reporting and not in total earnings or labor supply. Back-of-the-envelope calculations indicate that this behavior increases the cost of pensions by about 0.2% of the GDP.

We introduce two empirical strategies harnessing the randomness in school assignment mechanisms to measure school value-added. The first estimator controls for the probability of school assignment, treating take-up as ignorable. We test this assumption using randomness in assignments. The second approach uses assignments as instrumental variables (IVs) for low-dimensional models of value-added and forms empirical Bayes posteriors from these IV estimates. Both strategies solve the underidentification challenge arising from school undersubscription. Models controlling for assignment risk and lagged achievement in Denver and New York City yield reliable value-added estimates. Estimates from models with lower-quality achievement controls are improved by IV.

A key parameter in the analysis of wage inequality is the elasticity of substitution between skilled and unskilled labor. We show that the empirical literature is consistent with both publication and attenuation bias in the estimated inverse elasticities. Publication bias, which exaggerates the mean reported inverse elasticity, dominates and results in corrected inverse elasticities closer to zero than the typically published estimates. The implied mean elasticity is 4, with a lower bound of 2. Elasticities are smaller for developing countries. To derive these results, we use nonlinear tests for publication bias and model averaging techniques that account for model uncertainty.

We evaluate the effect of vehicle recalls on vehicle transactions in the second-hand market. Using a rich data set of Dutch vehicle registrations, we exploit the quasiexperimental variation in recalls across nearly identical cars. We find strong heterogeneities across market segments: transactions increased for cars with lower listed price or with defects, and decreased for those with higher price or no defects. Based on our theoretical model, this suggests that recalls increase sorting in low-end markets, yet exacerbate adverse selection in high-end markets. Our results shed light on the effect of information arrival in markets subject to uncertainty and information asymmetries.

This paper evaluates the impact of a sudden and unexpected nationwide alcohol sales ban in South Africa. We find that this policy causally reduced injury-induced mortality in the country by at least 14%. We argue that this estimate constitutes a lower bound on the true impact of alcohol on injury-induced mortality. We also document a sharp drop in violent crimes, indicating a tight link between alcohol and aggressive behavior in society. Our results underscore the severe harm that alcohol can cause and point toward a role for policy measures that target the heaviest drinkers in society.

Selection forces often confound the effects of policy changes. In the immigration enforcement context, we tackle this challenge tracking arrested immigrants along the deportation pipeline, isolating local and federal efforts. 80% of counties exhibit strategic substitutabilities in responding to federal enforcement, while the federal level is very effective at directing its efforts toward cooperative counties. We estimate that changes in the profile of immigration cases, and not weakened federal efforts, drove the reduction in deportations following a 2011 shift in federal priorities. Reducing immigration-court discretion and removing their dependence from the executive would have a significant impact on deportations.