Chairman Feingold, Senator Thurmond, Honorable members of the Subcommittee, thank you for inviting me to testify before you this morning. I have been asked to speak about race and the federal death penalty generally, and in particular about a recent report on this topic from the Department of Justice.

I. Summary

On June 6, 2001 the Department of Justice released a report entitled The Federal Death Penalty System: Supplementary Data, Analysis and Revised Protocols for Capital Case Review (the "Ashcroft Report"). This report follows a detailed study of the federal death penalty system that was released in September of last year, by former Attorney General Janet Reno. That earlier study found stark racial and geographic disparities in federal capital prosecutions. The most important was that the vast majority of federal capital defendants since 1988 have been African American or Hispanic.

Announcing the release of the new report, Attorney General John Ashcroft said in testimony before the House Judiciary Committee: "Our conclusion is, as the Reno study concluded, that there is no evidence of racial bias in the administration of the federal death penalty." In fact, the June 6 report provides no basis to conclude that the administration of the federal death penalty is free of racial discrimination. What’s more, former Attorney General Reno did not reach that conclusion. On the contrary, she expressed deep concern about the racial patterns revealed in the DOJ’s self examination, and she called for more detailed study by academic experts from outside the Department.

Attorney General Ashcroft seems to have concluded that an external, professional study of race and the federal death penalty is unnecessary. That is a serious mistake. Based on the evidence reported last September, there was every reason to be worried that racial discrimination might play a role in the use of the federal death penalty. The new report does nothing to change that.

II. The Reno Study

On September 12, 2000 former Attorney General Janet Reno released a study entitled Survey of the Federal Death Penalty System (1988-2000) (the "Reno Study").1 Among others, the Reno Study reported the following findings:

   • The Department of Justice sought the death penalty against 206 defendants from 1988 through 2000. Of these 75% (155) were minorities (and 51% (105) were African Americans); only 25% (51) were white. Reno Study pp. 23-24, Table ("T") -245, T-266.

   • Of 19 defendants under a federal sentence of death as of July 20, 2000, 79% (15) were minorities and 68% (13) were African American; only 21% (4) were white. Reno Study p. 36. By contrast, as the study points out, 55% of state death row inmates across the country at the end of 1998 were white, and 63% of the 505 inmates executed in the United States from 1988 through 1999 were white. Id. at p. 36 n.28.1

   • These racial disparities are generated primarily at the early stages of federal capital cases. Thus, among the 235 defendants for whom United States Attorneys recommended seeking the death penalty from 1988 through July, 2000, 77% (180) were minorities (and 51% (120) were African American), while only 23% (55) were white. Reno Study, T-5, T-6. For the death penalty "protocol" period separately (1995-2000) the pattern is nearly the same: United States Attorneys recommended capital charges for 183 defendants, of whom 26% (48) were white and 74% (135) were minorities, including 44% (81) African Americans. Id. p.12. In fact, the major problem seems to occur in the initial selection of cases for federal prosecution on capital-eligible charges. Of 682 such cases across the country from 1995 through July, 2000, 80% (548) involved minority defendants (and 48% (324) involved African American defendants), while only 20% (134) involved white defendants. Id. at T-6.

The DOJ report also contains many other troubling items. For example, since 1995 only 49 of the 94 United States Attorney offices have recommended any capital prosecutions (Reno Study, p. 12); 21 districts did not even file charges in a single capital-eligible case. Id. At the other end of the spectrum, the Eastern District of Virginia sought the death penalty against 21 of 66 defendants in potentially capital cases filed from 1995 through 2000. Id. at T-41. At the end of the judicial process, 12 of the 19 men on federal death row as of July, 2000, were sentenced in the South, including 6 from Texas and 4 from Virginia. Id. at T-307 - T-309. There were also pronounced disparities by race of victim. For example, United States Attorneys were twice as likely to recommend capital punishment for black defendants charged with killing white victims (22 out of 55, or 40%) as for black defendants charged with killing black victims (46 out of 227, or 20%). Id. T-67.

When the Reno Study was released, Attorney General Reno and Deputy Attorney General Holder conducted a press conference. The Attorney General summarized the key findings of the study, expressed her concern about them, and described some of the study’s limitations, including the absence of information on why the defendant was "arrested and prosecuted by federal authorities rather than state authorities," and why "the U.S. attorney submitt[ed] the case for review rather than enter a plea bargain." She added that:

"More information is needed to better understand ... how homicides make their way into the federal system, and once in the federal system, why they follow different paths. An even broader analysis must be undertaken to determine if bias does, in fact, play any role in the federal death penalty system."

She called for studies by experts outside the Department. Later, in response to a question, Attorney General Reno amplified this point: "[W]e want to continue to do everything we can to expose any bias if it exists. But at this point, we are troubled by the figures, but we have not found the bias."

Deputy Attorney General Holder was equally explicit:

"I am a career law-enforcement officer...I have approved the death penalty in several cases. But I can’t help but be both personally and professionally disturbed by the numbers that we discuss today.... [N]o one reading this report can help but be disturbed, troubled by this disparity. We have to be honest with ourselves. Ours is still a race-conscious society, and yet people are afraid to talk about race."

The present Attorney General, John D. Ashcroft, in response to written questions submitted to him as part of his confirmation hearing before the U.S. Senate Judiciary Committee, echoed the sentiments of Ms. Reno and Mr. Holder . For example, Senator Russel D. Feingold asked: "Are you troubled by the fact that about 75% of those against whom the Department of Justice seeks the death penalty are people of color or ethnic minorities, even though far less than 75% of the people who commit federal capital crimes are people of color and ethnic minorities?" and Attorney General Designate Ashcroft answered: "Yes, it troubles me deeply." Asked to comment on Ms. Reno’s statement that further studies are needed "to determine if bias does in fact play a role in the federal death penalty system," Mr. Ashcroft said: "I fully agree that the Department of Justice should do everything necessary to eliminate any racial bias from the federal death penalty system, including undertaking all reasonable and appropriate research necessary to understand the nature of the problem." Attorney General Ashcroft also stated that "federal law should be applied uniformly across the country," and promised to help ensure that, if confirmed.2

III. The National Institute of Justice

On January 10 of this year the National Institute of Justice convened a meeting of practitioners, researchers and government representatives, to discuss the federal death penalty. The main purpose of the meeting was to discuss how best to proceed to conduct the study that Attorney General Reno requested. I attended that meeting.

The Ashcroft Report (p.12) says that the discussion at that meeting "indicated" that such a study "could not be expected to yield definitive answers concerning the reason for disparities in federal death penalty cases." This description is puzzling. The researchers at the meeting did not talk in these terms, which have a peculiar lawyerly ring.3 (What is a "definitive" answer? Do we have "definitive" evidence that nicotine is addictive?) In fact, the clear consensus at the meeting was that a thorough and highly informative study of the federal death penalty could be done, given the resources and the will.

The Ashcroft Report correctly states that the researchers present saw this study as a "multi-year" project. The general estimate was two years. The main requirement that was discussed, in addition to funding, was cooperation from the Department of Justice. As I recall, the representatives of the Department who were present were strongly opposed to the notion of providing information on federal capital charging for such a study, regardless of any guarantees of confidentiality that were discussed by the researchers and that are available by statute under the authority of the NIJ. Excellent studies of capital charging and sentencing have been completed in several states, using data from a wide range of state agencies. With DOJ cooperation, the same could be done for the federal system.

The NIJ representatives present on January 10 said that they were committed to going ahead with this study, and promised to keep the participants at the meeting informed of their plans. Since then, I have heard nothing from the NIJ on the matter. In his testimony before the House Judiciary Committee, Attorney General Ashcroft said that he had already concluded "that there is no racial bias in the way we are administering the death penalty in the federal system." I believe this conclusion is unsupported, as I will explain. Given that conclusion, Mr. Ashcroft seems to have decided that there is no need to proceed with the study that Ms. Reno requested in order "to determine if bias does, in fact, play any role in the federal death penalty system." Instead, he announced that:

"[I]n order to assure public confidence and guarantee that our future efforts in the enforcement of the federal death penalty are consistent with the high standards of fairness that are required in charging, trying and sentencing those accused of federal death-eligible murders, I am directing today that the National Institute of Justice initiate a study of how death penalty cases are brought into the federal system." (Emphasis added.)

To summarize: Former Attorney General Reno requested a study by outside experts to determine whether there is racial bias in the system; a meeting was convened by the NIJ, the study was discussed, and we were assured that it would take place; no action was taken on the proposed study; despite the absence of the planned study, Attorney General Ashcroft concluded that there is no racial discrimination; he then proposed a similar sounding study, on a problem that he has already stated does not exist, for the explicit purpose of generating "public confidence". I know no independent researcher who would agree to conduct a study under these circumstances.

IV. The Ashcroft Report

The most striking thing about the Ashcroft Report is how little new material it contains. The new information in this report consists of two things:

   (1) Information on 291 additional potentially capital federal cases that were not included in the Reno Study. Unlike the Reno Study, the Ashcroft Report provides few details on these new cases. It seems, however, that the great majority of these new cases are comparatively low severity crimes that were initially omitted because the defendants had pled guilty to non-capital offenses before capital charges were ever filed. Ashcroft Report, n. 10.

   (2) The DOJ’s own explanations for the racial disparities in federal capital prosecutions, in general and in four selected federal districts.

These items add nothing of substance to the Reno Study. As a result, the Ashcroft Report does not support any new conclusions about the administration of the federal death penalty.

1. Initial Federal Charging and Later Stages of the Prosecution

Federal prosecutors occupy an unusual position in our system. For state prosecutors, charging in most homicide cases is automatic. If someone has been killed and there is good evidence against a known suspect, they almost always file charges (although not necessarily first degree murder). But nearly all federal crimes can also be prosecuted locally, so the Department of Justice can pick and choose a small number of cases and leave the rest to state authorities. The most conspicuous racial disparities in the use of the Federal death penalty are generated at this initial step. The pool of potentially capital cases that are selected for federal prosecution consists overwhelmingly of minority defendants, and nothing that happens later does much to change that stark disparity. That is why Attorney General Reno asked for additional information on "how homicides make their way into the federal system."

The Ashcroft Report purports to study this issue without looking at the much larger universe of cases in which federal capital charges could have been filed, but were not. It cannot be done. Whatever this report may be, it is not the sort of factual research that any scholar would ever rely on.

Imagine a company that is charged with gender discrimination for hiring a workforce that is 90% male. What if they responded by talking only about the men they did hire: "Mr. Smith had ten years of experience, so that’s why we hired him, not because he’s a man; Mr. Jones did an outstanding job for his previous employer;" and so forth. The immediate reaction would be: "That’s no good. You have to tell me about the female applicants that you didn’t hire. For all we know they were just as qualified." But that is just what the Department of Justice did not do. For all we know there were many white defendants with cases just as suitable for Federal prosecution as the minority defendants who were charged, or more so.

The Ashcroft Report emphasizes that among cases that are charged as federal capital crimes, the death penalty is sought more frequently for white defendants than for minority defendants. In his testimony on June 6, Attorney General Ashcroft relied on this finding repeatedly as evidence of lack of discrimination. This finding is not new – the same pattern was reported last September – and it does not show lack of discrimination.

What if the same company we discussed before said: "Look, we pay our female employees just as much as our male employees. Clearly we don’t discriminate by gender." Nobody would believe it. We’d answer: "Wait a second. Maybe that shows that you don’t discriminate in pay; but you were charged with discrimination in hiring." And yet this is the substance of the argument on this point in the Ashcroft Report and in the Attorney General’s testimony.

It is, of course perfectly possible that the DOJ does discriminate by race in the initial intake decision on who to prosecute on capital charges, and then does not further discriminate among those who are chosen for federal prosecution. In fact, it would not be surprising. The Ashcroft Report focuses on the professionalism of Assistant United States Attorneys, the lawyers who make the legal decisions once a case has been taken on. But the initial decision to undertake a federal investigation is often made by law enforcement agents rather than prosecutors, by the FBI or the DEA rather than the United States Attorneys. Perhaps these two sets of DOJ employees have different patterns of behavior.

In general, it is impossible to conclude anything about discrimination from the proportions of cases that are treated in a particular manner without detailed information on those cases. The imaginary firm I mentioned could discriminate against its female employees in compensation, even if they are paid more on average than the men, if the few women it hires are far more skilled and experienced than most of the men. This report provides essentially no information about the characteristics of the cases that were prosecuted federally. As a result, we cannot know why DOJ lawyers asked for the death penalty in some but not others.

The new data that are included in the Ashcroft Report illustrate how little can be learned from aggregate numbers like these. The Reno Study reported that among those charged with federal capital offenses from 1995 through 2000, the death penalty was sought for 38% of the white defendants, 25% of the black defendants, and 20% of the Hispanic defendants. Reno Study, p. 7. In the Ashcroft Report, the corresponding percentages are 27% for whites, 17% for blacks, and 9% for Hispanics. Do these new figures – which show that Hispanics are only 1/3 as likely as whites to face the death penalty – provide new evidence of absence of discrimination against Hispanics, or even of discrimination in their favor? Not at all.

As I mentioned, it appears that most of the 291 new cases that are added in the Ashcroft Report are comparatively low seriousness cases in which the defendant was allowed to plead guilty to non-capital charges. We also know that 53% of all the 291 new cases involve Hispanic defendants. See pie chart attached to Ashcroft Report. As result, the proportion of Hispanic federal defendants in capital eligible cases increased from 29% in the Reno Study (p.6) to37% in the Ashcroft Report (n.10), mostly by adding low aggravation cases. Inevitably, the proportion of death charges went down.

Does this decrease in the percentage of death charges for Hispanics mean there has been a decrease in discrimination against them? Consider a police department that is charged with racial profiling because 70% of the tickets it issued were given to minorities. What if they said: "But wait. There’s another batch of cases where we just gave warnings, and those drivers were 90% minorities. So, overall, among all the drivers we stopped, minorities were less likely than whites to get tickets rather than warnings." Would anyone take this defense seriously? Unfortunately, the Ashcroft Report’s use of numbers is no more convincing.

2. Explanations for the Racial Disparities in Federal Charging

The Ashcroft Report does offer some explanations for the racial disparities in federal death cases, but they are unsupported by data. The main one is that federal prosecutors target crimes associated with drugs, and that in the districts where they do so most actively "organized drug trafficking is largely carried out by gangs whose membership is drawn from minority groups." Ashcroft Report, p.3. No evidence is offered for this sweeping assertion.

The report goes into some detail about federal capital prosecutions in the Eastern District of Virginia. Ashcroft Report, pp.16 -18. This is a natural choice. Overall, 26 of the 206 federal cases in which the death penalty was requested from 1988 through July 2000 were from this one district, 13% of the national total. Reno Study, T-203 and T-207. All of these 26 death penalty defendants were African American.

Most of the potentially capital federal cases in the Eastern District of Virginia are homicides in the course of drug trafficking. The Ashcroft Report explains why there are no whites among the 34 federal defendants charged with capital murder for drug-related killings in that district:

"[T]he members of the drug gangs that engage in large-scale trafficking in the Eastern District of Virginia are not white." Ashcroft Report, p.17.

How does the Department of Justice know that all major drug traffickers in that entire district – from Arlington to Norfolk to Richmond – are minorities? The report does not say. Are we supposed to accept this extraordinary statement on faith?

Worse, this explanation has a depressingly familiar ring. Police departments that are charged with racial profiling sometimes respond: "It’s not discrimination. We're stopping and searching mostly black and Hispanic drivers because we're looking for major drug traffickers, and they’re all black and Hispanic." Is something similar going on here? Are Federal law enforcement agencies, the FBI and the Drug Enforcement Administration, searching for African American and Hispanic drug dealers because they think they know that the worst drug traffickers are all black or Latin American? Are the racial disparities in Federal capital prosecutions a manifestation of race-specific drug investigations? We don’t know, and this report does nothing to allay our fears.

*Samuel R. Gross is Thomas & Mabel Professor of Law at the University of Michigan Law School. He is an expert on criminal procedure, evidence, and the use of social science in legal proceedings. He has written extensively about the death penalty, including a book on racial discrimination in capital punishment (Death and Discrimination, Northeastern University Press, 1989, with Robert Mauro), and about the use of expert evidence in litigation.

1 See http://www.usdoj.gov/dag/pubdoc/dpsurvey.html/.

1 That disproportion has since become even more extreme, as a few new defendants have joined federal death row, and a few others have been removed by the courts, or, in the case of Mr. McVeigh, by execution. As of today, 18 of 20 of Federal capital defendants – 90% – are minorities.

2 http://www.senate.gov/~leahy/press/200101/ashcroft.html

3 The report lists no author(s), so it is unclear who provided the information on which this description is based.