Arguably the most aggressive affirmative action program ever implemented in the United States was a series of court-ordered racial hiring quotas imposed on municipal police departments.My best estimate of the effect of court-ordered affirmative action on workforce composition is a 14 percentage point gain in the fraction African American among newly hired officers.Evidence on police performance is mixed.Despite substantial black-white test score differences on police department entrance examinations, city crime rates appear unaffected by litigation.However, litigation lowers slightly both arrests per crime and the fraction black among serious arrestees.

Arguably the most aggressive affirmative action program ever implemented in the United States was a series of court-ordered racial hiring quotas imposed on municipal police departments. My best estimate of the effect of court-ordered affirmative action on work-force composition is a 14-percentage-point gain in the fraction African American among newly hired officers. Evidence on police performance is mixed. Despite substantial black-white test score differences on police department entrance examinations, city crime rates appear unaffected by litigation. However, litigation lowers slightly both arrests per crime and the fraction black among serious arrestees.

Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime: Comment by Justin McCrary. Published in volume 92, issue 4, pages 1236-1243 of American Economic Review, September 2002

Forward-looking behavior on the part of the monetary authority makes it difficult to estimate the effect of monetary policy interventions on output. We present instrumental variables estimates of the impact of interest rates on quarterly real output for several European countries, using German interest rates as the instrument. These estimates confirm a strong forward-looking bias in least squares estimates that persists even conditional on standard controls for the history of the system. Due to the potential for correlation of output shocks across countries, we interpret our estimates as lower bounds for the effect of monetary policy on real output.

This paper uses age-at-school-entry policies to identify the effect of female education on fertility and infant health. We focus on sharp contrasts in schooling, fertility, and infant health between women born just before and after the school entry date. School entry policies affect female education and the quality of a woman's mate and have generally small, but possibly heterogeneous, effects on fertility and infant health. We argue that school entry policies manipulate primarily the education of young women at risk of dropping out of school.

Frölich (2004) compares the finite sample properties of reweighting and matching estimators of average treatment effects and concludes that reweighting performs far worse than even the simplest matching estimator. We argue that this conclusion is unjustified. Neither approach dominates the other uniformly across data-generating processes (DGPs). Expanding on Frölich's analysis, this paper analyzes empirical as well as hypothetical DGPs and also examines the effect of misspecification. We conclude that reweighting is competitive with the most effective matching estimators when overlap is good, but that matching may be more effective when overlap is sufficiently poor.

We show current market practices relating to odd-lot quotes create a large “inside” market where better prices routinely exist relative to the National Best Bid or Offer. We show that odd-lot quotes play a price discovery role, and these quotes provide valuable information to traders with access to proprietary data feeds. Using a XGBoost machine learning algorithm that uses odd-lot data to predict future prices, we demonstrate a simple and profitable trading strategy. We argue the SEC’s proposed round-lot redefinition reduces—but does not eliminate—the high incidence of superior odd-lot quotes within the NBBO.

In the single-IV model, researchers commonly rely on t-ratio-based inference, even though the literature has quantified its potentially severe large-sample distortions. Building on Stock and Yogo (2005), we introduce the tF critical value function, leading to a standard error adjustment that is a smooth function of the first-stage F-statistic. For one-quarter of specifications in 61 AER papers, corrected standard errors are at least 49 and 136 percent larger than conventional 2SLS standard errors at the 5 percent and 1 percent significance levels, respectively. tF confidence intervals have shorter expected length than those of Anderson and Rubin (1949), whenever both are bounded.