Racial Disparities in Federal Sentencing: A Quantile Regression Approach PRELIMINARY AND INCOMPLETE PLEASE DO NOT CITE WITHOUT PERMISSION
نویسنده
چکیده
The Sentencing Reform Act of 1984 explicitly prohibits the use of demographic characteristics in determining prison sentence, yet researchers continue to find evidence of disparity in sentences on the basis of race, ethnicity and gender. We employ a quantile regression technique that estimates the black-white sentencing gap across the entire sentencing distribution, rather than only at the mean. Our results show that racial disparity exists for some types of offenses, however, the impact of race varies greatly dependent on the location along the sentencing distribution. For some types of offenses, disparity only exists for the lower half of the sentencing distribution. We also find that the OLS coefficient estimate for racial disparity may be misleading, suggesting disparity is much larger than what we find using quantile regression analysis. Finally, we find that much of the racial disparity in federal sentencing is due to departures from the Sentencing Guidelines from offenders providing substantial assistance to the government in the prosecution of others. JEL Codes: K14, K40
منابع مشابه
Network and Panel Quantile Effects via Distribution Regression
This paper provides a method to construct simultaneous confidence bands for quantile functions and quantile effects in nonlinear network and panel models with unobserved two-way effects, strictly exogenous covariates, and possibly discrete outcome variables. The method is based upon projection of simultaneous confidence bands for distribution functions constructed from fixed effects distributio...
متن کاملRacial Disparities in the Allocation of Wiretap Applications across Federal Judges
The degree to which discrimination causes the large racial disparities in the criminal justice system is a long-standing and difficult empirical question. This paper studies a decision in which the influence of race on prosecutors is directly observed: the process of obtaining judicial approval to conduct wiretap surveillance. The Department of Justice requires that federal prosecutors obtain i...
متن کاملFinancial market volatility and the business cycle: a stochastic volatility approach
This is an extended abstract of the paper. All the results shown are preliminary and incomplete. Do not quote without permission.
متن کاملContracting for Information (Preliminary and Incomplete)∗
We study optimal contracts between a decision maker and an expert where the decision maker can commit to make transfers to the expert contingent on his advice but cannot commit to take an action contingent on advice. We show that optimal contracts never entail full revelation. Instead, the decision maker pays the expert to reveal for some states and doesn’t compensate the expert at all in other...
متن کاملPreliminary and Incomplete Do Not Quote without Permission the Stock Market and the Fed
The paper investigates the relationship between the stock market and the monetary policy of the Federal Reserve in the last twenty years. To this purpose we construct an Index of Stock Price Misalignment in which the fundamental value of the stocks is computed on the basis of the discounted cash flow approach and we then include this index, among the regressors, into a forward looking Taylor ru...
متن کاملPRELIMINARY AND INCOMPLETE∗ A Unique Costly Contemplation Representation
We extend the costly contemplation model to preferences over sets of lotteries, assuming that the state-dependent utilities are von Neumann-Morgenstern. The contemplation costs are uniquely pinned down in a reduced form representation, where the decision-maker selects a subjective measure over expected utility functions instead of a subjective signal over a subjective state space. We show that ...
متن کامل