Become a Patron! 



excerpted from: Sherod Thaxton, Disentangling Disparity: Exploring Racially Disparate Effect and Treatment in Capital Charging, 45 American Journal of Criminal Law 95 (Spring, 2018) (328 Footnotes) (Full Document)


Sherod ThaxtonEmpirically oriented legal scholars and social scientists have developed a voluminous literature documenting racial disparities in sentencing at both the state and federal levels. With very few exceptions, these studies demonstrate the persistence of racial disparities across time, place, and offense type, even after accounting for a wide range of nonracial factors purported to influence sentencing. Of course, judges' and jurors' sentencing decisions come at the tail end of the adjudicative process, and earlier discretionary choices by legal actors--primarily prosecutors--also influence final outcomes. As a consequence, there has been increased emphasis on, and scrutiny of, prosecutorial decision-making because prosecutors are generally less constrained by the law--and their choices are less visible to the public--than judges and juries. The adjudicative process begins with the charging decision, and not only does research suggest that racial disparities are strongest at this stage, but also that racial disparities are not rectified during sentencing. Furthermore, studies that focus exclusively on sentencing and ignore earlier discretionary choices, which are vulnerable to racially discriminatory practices, tend to mask racial disparities.

Any abuse of discretion by prosecutors creates serious cause for concern, but unjustified racial disparities in charging decisions in the capital punishment context is especially alarming. “One of the enduring arguments in Supreme Court death penalty jurisprudence is that the death penalty is ‘qualitatively different’ from all other punishments in ways that require extraordinary procedural protection against error.” And the omnipresent influence of impermissible racial considerations on the administration of capital punishment has figured prominently in the Court's decisions. In fact, the case credited with “launching one hundred years of federalism” involved an African American defendant who, inter alia, challenged the legality of his death sentence (from state court) under the Due Process and Equal Protection Clauses of the U.S. Constitution because of overt racism at the pretrial, trial, and appellate stages. The vast majority of statistically sophisticated studies examining capital charging have discovered that race still exerts an impact: all else equal, African American defendants are more likely to be charged with the death penalty than Caucasian defendants and defendants of any race charged with killing Caucasian victims are significantly more likely to face a capital charge than defendants charged with killing non-Caucasian victims.

Despite near consensus in the scholarly literature about the persistence of racial disparities in the criminal justice system, judges, attorneys, legislators, and the general public continue to debate whether these racial disparities in criminal justice outcomes are primarily a function of differential criminal culpability (disparate effect) or discriminatory legal decision-making (disparate treatment). Defendants have raised claims of racially discriminatory capital charging practices, referred to as selective prosecution, in federal court, as violative of their rights to equal protection under the law as guaranteed by the Fifth and Fourteenth Amendments, and have relied on statistical evidence to support their assertions. The Supreme Court has uniformly rejected these claims, underscoring a very troubling fact about the Court's selective prosecution jurisprudence: the Court has not ruled in favor of a defendant raising a selective prosecution claim based on racial discrimination in over 130 years. In 1886, in Yick Wo v. Hopkins, the Court ruled for the first time that racially biased enforcement of a facially neutral law violated the Equal Protection Clause of the Fourteenth Amendment. Yick Wo was also the last time a race-based selective prosecution claim was successfully argued before the Court. This fact is especially mystifying given the weight of social scientific evidence of unjustified systemic racial disparities not only in prosecutorial charging decisions, but in virtually all aspects of criminal justice legal decision-making that has emerged in the aftermath of Yick Wo.

The ineffectiveness of this body of research in racial discrimination litigation in the criminal context can be primarily attributed to the Court's anti-discrimination jurisprudence which has simultaneously failed to specify the type of statistical evidence necessary to support an inference of discrimination and increasingly suggested that, in order to be successful with statistical evidence, claimants must show how systemic racial bias leads to racial disparities, and not merely that there was an opportunity for discriminatory decision-making. In other words, the Court has emphasized that claimants must do a better job of demonstrating the manner in which racial discrimination influences decision-making, while at the same time neglecting to provide guidance to judges and litigants as to what kinds of circumstantial evidence would be demonstrative of a constitutional violation. The purpose of this article is to not only answer the Court's clarion call--that is, carefully demonstrating how racially differential treatment produces a racially disparate outcome--but also bring attention to some peculiarities of the current doctrine that have made it unduly burdensome on claimants to prevail in selective prosecution actions for well over a century.

The primary contribution of this article pertains to the operationalization of systemic discriminatory treatment, subjecting prosecutorial decision-making in the capital charging process to a more granular analysis that is directly responsive to several of the Supreme Court's prior concerns about the use of statistical analyses of capital charging-and-sentencing behavior to provide evidence of racially disparate treatment. The statistical models described in this Article provide a template for the investigation of discriminatory charging dynamics in capital and non-capital cases. Concretely, my analytical approach carefully separates an observed racial disparity in capital charging into two components. The first component pertains to differences in the distribution of aggravating and mitigating evidence across Caucasian-victim and African American-victim cases and is analogous to a disparate effect (as defined in this Article). The second component captures the differences in the returns on that aggravating and mitigating evidence; in other words, differences in prosecutors' behavioral responses to that evidence. This latter component is a measure of discriminatory treatment. Under this analytical framework, the discriminatory treatment component does not purport to directly capture racial animus on the part of the decision-maker, although such effects may be highly probative of such animus and support an inference that it exists. This article is the first to apply the analytical approach to capital charging decisions, and decision-making in the capital punishment process, more generally.

Prior research has yet to sufficiently disentangle the sources of racial differences in capital charging at a descriptive level, even though such differences have been explored by legal scholars and social scientists for more than 70 years. This shortcoming may partly stem from the fact that scholars have been primarily concerned with measuring overall differences in criminal justice outcomes between racial groups that remain after taking into account a wide range of legally relevant variables. Under this approach, racial discrimination is said to exist because no other valid explanation accounts for the observed differences. These studies have been helpful in highlighting the fact that purely legal justifications fail to explain why, in the aggregate, members of certain groups are routinely subject to harsher punishments than others, net of the actual social harm they cause. Yet judges, lawyers, legislators, and scholars still lack an understanding of how prosecutors differentially assess legally relevant (and legally suspect) factors across different racial groups--that is, how prosecutors' evaluations of seemingly objective criteria may shift based upon race. Put differently, this existing scholarship has failed to inform the legal community about the potential ways in which race modifies the impact of legally relevant (and legally suspect) factors on legal behavior, independent of the distribution of these characteristics across the various racial groups. My analytical approach provides a more nuanced understanding of racially disparate treatment, which is especially necessary in light of the Supreme Court's sparse case law that has failed to articulate a uniform and workable evidentiary standard for statistical evidence of discrimination.

Part I explores the differing conceptualizations of race-based discrimination present in the U.S. Supreme Court's constitutional and statutory anti-discrimination jurisprudence, describing and evaluating the rationales for these differing conceptions--both theoretical and practical--as well as the critiques of those rationales. The Court has blurred the line between these seemingly opposing notions of discrimination, raising important questions about the appropriateness of various types of evidence in particular contexts, the standards governing its applications, and the Court's competency in assessing such evidence.

Part II describes both the Court's embrace and skepticism of statistical evidence of intentional discrimination over the past forty years. The Court's sharply divided opinions and uneven approach to statistical evidence has failed to provide workable standards for lower courts to apply.

Part III discusses and assesses the empirical social scientific literature on capital charging dynamics over the past quarter-century. This literature has almost unequivocally identified racial disparities in charging decisions based on the victim's race and the combination of the defendant's and victim's race, but the analytical frameworks utilized in these studies have impaired the ability of analysts to ask and answer questions that now appear to be of central interest to courts and legislators--namely, how are racial differences in outcomes connected to racial differences in process? I explain, both mathematically and in plain English, how prior studies have measured racial discrimination, their specific findings, and why their methodologies prevent addressing more fundamental questions that often lie at the heart of courts' inquiries.

Part IV presents a set of statistical tools--again, both mathematically and in plain English--capable of disentangling disparate effect from disparate treatment in capital charging. After discussing the statistical model, I describe an originally compiled dataset of capital charging decisions from Georgia over an eight-year period to which I apply the aforementioned analytical approach.

Part V explains the results of the statistical analyses. My findings make it unequivocally clear that race still very much matters for capital charging decisions. I find that 60%-80% of the race-of-victim gap in capital charging behavior in Georgia is attributable to disparate treatment. In addition to the overall disparate treatment effect, I demonstrate how much the racially differential treatment of specific case characteristics contributes to the race-of-victim gap in capital charging. This aspect of my analysis demonstrates how unjustified racial differences in process directly contribute to racial differences in outcomes, and is thereby directly responsive to several Supreme Court Justices' heightened evidentiary standard for statistical evidence of discrimination.

Part VI discusses the legal implications of my findings for discriminatory prosecution claims and examines the durability of the results in the presence of potential uncertainty about the underlying statistical model and measurement of key variables.

. . .

In his historic dissenting opinion in Glossip v. Gross, Justice Breyer remarked that the “arbitrary imposition of punishment is the antithesis of the rule of law. [ ... ] How then can we reconcile the death penalty with the demands of a Constitution that first and foremost insists upon a rule of law?” He described the vast social scientific literature over the past 40 years documenting the unconstitutional administration of the death penalty, including “numerous studies [that] have concluded that individuals accused of murdering white victims, as opposed to black or other minority victims, are more likely to receive the death penalty.” According to the Justice, the “circumstances and the evidence of the death penalty's application have radically changed” since the Court upheld the constitutionality of the death penalty forty years earlier in Gregg v. Georgia. He “believe[s] that it is now time to reopen the question” of the constitutionality of the administration of the death penalty and invited “full briefing that would allow [the Court] to scrutinize [the empirical scholarship on the administration of the death penalty] with more care.”

This Article accepted Justice Breyer's invitation and set forth a framework that more carefully parses race-of-victim differences in capital charging than prior studies into the part explained by actual differences in the defendant's level of culpability and the part explained by prosecutors' racially discriminatory treatment of these cases. The model is directly responsive to the Court's critique of much of the existing statistical evidence of racial discrimination--its inability to explicitly connect racial differences in process to racial differences in outcomes. The approach I adopt quantifies the extent of prosecutorial “shifting standards” in capital charging according to the victim's race and establishes the foundation for an articulation of a more powerful and appropriately nuanced story about the role of race on prosecutorial decision-making.

Accompanying my methodological contribution is an important substantive one: race still matters a lot in capital charging decisions in Georgia. And there is good reason to believe that similar results would be obtained in other jurisdictions based on the similarity of empirical findings across studies, as well as the fact that many states modeled their own death penalty statutes after Georgia's (which, itself, was based on the American Law Institute's Model Penal Code). I discover that 60%-80% of the race-of-victim gap in capital charging results from disparate treatment. More importantly, I show that many of the case characteristics relevant to defendant culpability (i.e., aggravation and mitigation evidence) have radically different effects on the likelihood that the prosecutor seeks the death penalty depending on the victim's race in the case. In other words, I demonstrate how race influences the process of prosecutorial decision-making that leads to racially disparate outcomes.

Of course, any statistical approach to measuring discrimination will only be as reliable as the assumptions of the underlying statistical model are reasonable. So in addition to presenting a novel framework for examining racial discrimination in capital charging, I also describe and implement various diagnostic tools to examine the sensitivity of my results. These tools, like the statistical model to which they are applied, are also responsive to the Court's general concerns about the reliability of statistical evidence. The diagnostic tools, along with underlying framework, constitute a template for the investigation of discriminatory dynamics in the capital context, and therefore are critically important to how judges, lawyers, legislators, and legal scholars think about the constitutional constraints on prosecutorial decision-making and the courts' role in ensuring the rule of law remains operative. As Justice Brennan eloquently explained in McCleskey, the “diminished willingness to render [capital punishment] when blacks are victims, reflects a devaluation of the lives of black persons. [ ... ] Race is a consideration whose influence is expressly constitutionally proscribed ... and evidence that race may play even a modest role in levying [capital punishment] should be enough to characterize that [punishment] as [unconstitutional].

Assistant Professor of Law, UCLA School of Law.