Knowledge that Transforms

Using annual longitudinal data, I show that all children in families with teen childbearing are on a downward trajectory several years before pregnancy begins. Compared to students on similar trajectories from families without teenage childbearing, siblings of teen mothers have lower test scores, higher high school dropout, and higher juvenile justice system exposure following the birth. The change in test score outcomes occurs after the baby is born, indicating that the child's arrival affects performance, rather than some unobserved occurrence leading to both teen pregnancy and poor outcomes. The test scores for teen mothers drop in the year of pregnancy.

This paper proposes a new measure of the disincentive cost of unemployment insurance (UI): the ratio of the behavioral cost (BC) to the mechanical cost (MC) of a UI reform. This measure represents the labor supply distortion relative to the additional (mechanical) transfer from the UI reform. We show the BC/MC ratio naturally arises from a model of optimal UI and can be readily computed and compared across different types of reforms and labor market contexts. We summarize the evidence regarding the BC/MC ratio for existing studies and relate it to typical measures of employment effects of UI.

We discuss an underutilized dataset to examine the causes of migration. The United Nations High Commissioner for Refugees publishes annual binational asylum applications and the resulting decisions. Asylum is granted to protect individuals from persecution. They are a small part of overall migration patterns: one-tenth of overall migration flows into OECD countries. The European Union receives the largest share of asylum applicants and has a low acceptance rate, but the rate increases when source countries have positive deviations from historic trends. Countries outside the EU and OECD receive almost all of the applications from neighbors with a contiguous land border.

Jurisdictions across the United States have adopted “ban the box” (BTB) policies preventing employers from conducting criminal background checks until late in the job application process. Their primary goal is to increase employment for those with criminal records. If individuals with criminal records view these policies as improving their labor market opportunities, they might move to BTB-adopting places in search of employment. In this paper, we consider BTB's effects on the demographic composition of labor markets and the likelihood that residents report recently moving from other labor markets. We find no evidence that BTB affects migration.

We introduce a framework of matching with flexible allotments that can be used to model firms with cross-division hiring restrictions. Our framework also allows us to nest some prior models of matching with distributional constraints. Building upon our recent work on observable substitutability, we show that multi-division choice functions with flexible allotments enable stable and strategy-proof matching; these results illustrate the value of clearly mapping when stable and strategy-proof matching is possible.

Is coachability a usable concept for startups? Using a dataset on high-growth startups in a prominent university incubator, we show that firm characteristics do predict which firms are likely to follow advice but that these “coachable” firms are not more likely to achieve positive outcomes. The reason is that advice may be rejected either because the firm is stubborn and uncoachable, or because the firm possesses capabilities related to the advice which is hard for outsiders to observe.

This paper provides a snapshot of undergraduate research at top 30 liberal arts institutions and top 30 national universities (based on U.S. News and World Report rankings, 2017). This description provides a broader perspective than exists in the literature as we identify departmental motivations for providing undergraduate research opportunities despite the lack of mandates. We also document the extent to which a variety of forms of undergraduate research occur (programmatically and by student numbers). Also included are descriptions of innovative approaches to promoting undergraduate research in light of constraints and methods for creating a culture conducive to undergraduate research.

Successful platforms attract not just many users, but also those of the right kind. 'The right kind of user' is one who can either be directly monetized or who differentially attracts other valuable users. Bonacich centrality on the network of user sorting with direct value of monetization captures this feedback loop and thus characterizes the value of user characteristics. We use this value to determine optimal steady-state platform design and reliable means for platforms to reach such a steady state. We apply these results respectively to explain the dynamic growth strategy of social networks and urban development policies of cities.

Many school and college admission systems use centralized mechanisms to allocate seats based on applicant preferences and school priorities. When tie-breaking uses non-randomly assigned criteria like distance or a test score, applicants with the same preferences and priorities are not directly comparable. The non-lottery setting does generate a kind of local random assignment that opens the door to regression discontinuity designs. This paper introduces a hybrid RD/propensity score empirical strategy that exploits quasi-experiments embedded in serial dictatorship, a mechanism widely used for college and selective K-12 school admissions. We use our approach to estimate achievement effects of Chicago's exam schools.

Aggregate real wages exhibit less procyclicality than most macroeconomic models predict. We use 35 years of Current Population Survey data to confirm that the puzzling behavior of wages largely owes to changes in the composition of the employed over the business cycle. This composition effect relates to changes in both the number and the relative wage levels of those entering and exiting. The changing gap in wages of entrants and exiters is especially important for the unemployed. A large part of this wage gap is due to differences in average Mincer residuals between entrants and exiters.