Issues

Sentencing has become the most important part of a criminal case. Over the past century, criminal trials have given way almost entirely to pleas. Once a case is charged, it almost always ends up at sentencing. And notably, judges learn little sentencing-relevant information about the case or the defendant prior to sentencing and have significant discretion in sentencing decisions. Thus, sentencing is the primary opportunity for the defense to affect the outcome of the case by presenting mitigation: reasons why the nature of the offense or characteristics of the defendant warrant a lower sentence. It is surprising, then, that relatively little scholarship in criminal law focuses on mitigation at sentencing. Fundamental questions have not been explored: Do the Sentencing Guidelines—which largely limit the relevance of mitigating evidence—make mitigation unimportant? Does the extent or type of mitigation offered have any relationship with the sentence imposed?

This Article fills that gap by examining a previously unexplored data set: sentencing memoranda filed by defense attorneys in federal felony cases. By systematically parsing categories of mitigating evidence and quantitatively coding the evidence, I show that mitigation is a central predictor of sentencing outcomes and that judges approach mitigation in a modern way: rather than adhering to the strict, offense-centric structure that has dominated sentencing since the advent of the Sentencing Guidelines in the 1980s, judges individualize sentences in ways that consider the personal characteristics of each defendant, beyond what the Guidelines anticipate. And particular types of mitigation, such as science-based arguments about mental and physical health, appear especially persuasive.

The results have significant implications for criminal justice policy: while my data show that mitigation is critical to judges’ sentencing decisions, both the Guidelines and procedural rules minimize mitigation, failing to encourage both defense attorneys and prosecutors to investigate and consider it. I suggest reforms to make sentencing more equitable, such as requiring the investigation and presentation of mitigation to constitute effective assistance of counsel, easing the barriers to obtaining relevant information on mental and physical health mitigation, and encouraging prosecutors to consider mitigation in charging decisions and sentencing recommendations.

This Article empirically tests the conventional wisdom that a permissive constitutional standard bearing on pretextual traffic stops—such as the one announced by the Supreme Court in Whren v. United States—contributes to racial disparities in traffic stops. To gain empirical traction on this question, we look to state constitutional law. In particular, we consider a natural experiment afforded by changes in the State of Washington’s rules regarding traffic stops. Following Whren, the Washington Supreme Court first took a more restrictive stance than the U.S. Supreme Court, prohibiting pretextual stops by police officers, but later reversed course and instituted a laxer standard, effectively equivalent to Whren’s.

We investigate the effect of this retreat to a Whren-like standard on the degree of racial disparities in traffic stops in Washington. For that purpose, we use a dataset of over 7 million traffic stops and employ a range of empirical techniques—including the estimation of difference-in-difference and triple-differences specifications—that are designed to isolate the effect of the change in Washington constitutional law and account for both observable and unobservable factors that may also impact racial disparities in traffic-stop rates. In particular, we employ a novel methodological approach designed to separate the effect of the change in constitutional standards in Washington from the effect of Washington’s contemporaneous legalization of recreational marijuana.

Across our deep dive into these matters, we fail to find evidence that supports the conventional wisdom that a Whren-like standard intensifies racial bias in officers’ decisions to initiate stops. On the contrary, our results suggest that constitutional standards, at best, have little to no impact on the gap between the stop rates of non-white and white drivers (or between Black and white drivers).

Racial disparities in traffic stops are an undeniable problem. And to best address this problem, we need to understand which legal tools do and do not work in regulating officer behavior. We suggest our findings may be due to certain inherent weaknesses in the way in which the relevant constitutional standards are enforced—i.e., via the exclusionary rule. To the extent an officer’s decision to initiate a traffic stop is heavily driven by factors other than the remote possibility that any evidence obtained during a pretextual stop will be suppressed, it is unlikely that a constitutional-rights-based approach will meaningfully reduce racial disparities in traffic-stop rates. Instead, we propose several extraconstitutional approaches to this critical problem, including the use of administrative disciplinary systems that evaluate an officer’s aggregate pattern of behavior, not their behavior in individual cases—i.e., approaches designed to bolster a deterrent channel—along with the development of technologies that rely less on officer discretion in the first place.