Learning About the Bias of Others: Prosecutors and Police Before and After Body-Worn Cameras

Draft  (with Emma Harrington) (working paper)

Decision-makers often fail to correct for the bias of other actors. We model and test two behavioral mechanisms for this phenomenon: (1) the decision-maker may simply take earlier actors’ information at face value, and (2) she may have inaccurate beliefs about the extent of earlier actors’ biases. We study these mechanisms in the criminal justice system, where prosecutors rely on potentially biased police reports to make charging and sentencing decisions. Using the staggered rollout of police body-worn cameras (BWCs) in North Carolina, we evaluate how disparities change once prosecutors can see more objective information about arrests. We find that BWCs reduce incarceration disparities by 10%, little of which can be explained by a decline in arrests. To unpack the two mechanisms, we fielded an original survey of 203 prosecutors, which we linked to their half-million cases. First, we find that prosecutors who take police reports at face value increase incarceration disparities relative to others in the same unit. Yet this pattern disappears once BWCs are adopted; instead, disparities are uniformly lower. Second, we find evidence that prosecutors had inaccurate beliefs about police bias before BWCs were adopted: those with more exposure to BWCs believe that disparities can be explained more by racial bias and less by racial differences in crime.

Prosecutors, Race, and the Criminal Pipeline

• Manuscript

• Appendices

90 U. Chi. L. Rev. 1889 (2023) 

This Article presents evidence that some state prosecutors reduce racial disparities in criminal sentences. This finding challenges the prevailing view that prosecutors compound disparities. Existing literature aims to identify whether prosecutors introduce new disparities by controlling for race gaps that prosecutors inherit from earlier decision-makers in the criminal pipeline. This Article takes a more holistic approach to understanding prosecutorial discretion by interpreting prosecutors’ impacts in light of the accumulated disparities that already exist when they first open their case files.

Using North Carolina state court records from 2010 to 2019, I evaluate how the sentencing penalty for prior convictions differs by defendant race. My principal finding is that the increase in the likelihood of a prison sentence for an additional prior conviction was 23% higher for white than Black defendants with similar arrests and criminal records. While Black and white defendants without criminal records were incarcerated at similar rates, white defendants with criminal records were incarcerated at significantly higher rates. And the longer the record, the greater the divergence.

In concrete terms, racial disparities in North Carolina prison rates in 2019 would have increased by 20% had the state mandated equal treatment of defendants with similar case files. These findings should lead reformers to exercise caution when considering the urgent calls to limit or eliminate prosecutorial discretion. Constraining prosecutors or blinding them to defendant race — a proposal that jurisdictions have increasingly implemented or considered — may inadvertently increase disparities by neutralizing the offsetting effects of some prosecutors. While race-blind charging ensures that prosecutors do not introduce new bias, it also ensures that any past bias is passed through to current (and future) decisions.

Prediction Errors, Incarceration, and Violent Crime: Evidence from Linking Prosecutor Surveys to Court Records

Manuscript  with Appendices (with Emma Harrington and William Murdock III)  (forthcoming, American Economic Journal: Economic Policy)

In criminal cases, decision-makers aim to selectively incarcerate defendants with a high risk of future violent crime. If decision-makers have more accurate beliefs about this risk, can they reduce violent crime without simply incarcerating more defendants? We survey 162 prosecutors about how violent re-arrest rates vary across defendants of different ages and with different criminal records. We link prosecutors’ beliefs to their 104,039 cases, which offices assign quasi-randomly. Prosecutors’ beliefs vary widely and predict their sentencing patterns for defendants of different ages and criminal records. Prosecutors with more accurate beliefs (by one standard deviation) reduce violent crime (by 6%) without incarcerating more defendants.

Brokers of Bias in the Criminal Justice System:  Do Prosecutors Compound or Attenuate Racial Disparities Introduced by Police?

Manuscript with Appendices  (with Emma Harrington) (forthcoming, Review of Economics and Statistics)

In criminal cases, prosecutors have the discretion to adjust arresting officers’ charges — and so can offset racial disparities introduced by the police. Yet prior research suggests that prosecutors simply compound earlier racial disparities with their own biases. We investigate prosecutors’ impacts on racial disparities using discontinuities in North Carolina’s sentencing laws. For defendants with criminal histories that are slightly longer and so fall above certain thresholds, the law mandates a prison sentence. However, prosecutors can often sidestep mandatory prison for qualifying defendants by reducing their charge. We ask who benefits from these charging responses to the mandatory-prison discontinuities. Between 1995 and 2019, Black defendants were initially less likely — but ultimately became more likely — to benefit from charge reductions that avoid mandatory prison. The reversal is driven entirely by arrests typically initiated by police stops and is absent from arrests typically initiated by victims, suggesting that prosecutors have increasingly questioned disparities introduced by police.