| Two of the goals of the criminal justice system are deterring criminal activity and ensuring that individuals have a fair trial. In order to achieve these goals, the government uses policies which affect the incentives of individuals on both sides of the system. This dissertation evaluates three such policies.;The first chapter, "I'd rather be Hanged for a Sheep than a Lamb: The Unintended Consequences of 'Three-Strikes' Laws," evaluates the impact of longer sentences on the distribution of crimes committed by repeat offenders. To discourage repeat offenders, many states passed "Three-Strikes" laws, which impose enhanced penalties for multiple felony convictions. Assuming that more serious crimes have higher expected payoffs, the flattening of the penalty gradient implies that repeat offenders will commit more serious crimes when they do engage in criminal activity. Using data from California's criminal records for 1993-1995, I find that repeat offenders became more likely to commit serious crimes after Three-Strikes was implemented.;The second chapter "Does the Certainty of Arrest Reduce Domestic Violence? Theory and Evidence from Mandatory Arrest laws," explores the impact of policies that mandate arrest when a domestic violence incident is reported. These laws were justified by a randomized experiment which found that arrests reduced future violence. Using the FBI homicide data from 1980-2000, I provide evidence that mandatory arrest laws increased intimate partner homicides. I provide theoretical and empirical evidence that this increase in homicides is due to decreased reporting.;The third chapter, "An Analysis of the Performance of Federal Indigent Defense Counsel," measures performance differences between the two types of indigent defense attorneys in the federal system. Exploiting the random assignment of cases between the types of attorneys, an analysis of cases from 1997-2001 from 51 districts indicates that public defenders secure lower conviction rates and sentence lengths than CJA attorneys. An analysis of data from three districts finds that attorney experience, wages, law school quality and average caseload account for over half of the overall difference in performance. This performance difference disproportionately affects minorities and as such may constitute a civil rights violation under Title VI of the Civil Rights Act. |
