OPINION

BOYCE F. MARTIN, Jr., Chief Judge.

On December 8, 1998, a federal grand jury returned a second superseding indictment charging defendant John Bass with the intentional firearm killing of two individuals. Shortly thereafter, the United States filed its notice of intent to seek the death penalty on those charges. Bass moved to dismiss the death penalty notice and, in the alternative, requested discovery pertaining to the United States's capital charging practices. The district court granted Bass's discovery request and, after the United States refused to comply with the order, dismissed the death penalty notice. We now affirm the district court's discovery order, and remand to allow the United States to submit the requested materials for an in camera review.

I.

According to a Department of Justice report, "The Federal Death Penalty System: A Statistical Survey" (September 12, 2000), all death-eligible charges brought by the United States since 1995 are subjected to the Department's death penalty decision-making procedures. Under the protocol, the individual United States Attorneys offices retain discretion in only three areas: whether to bring federal charges or defer to state prosecutions, whether to charge defendants with a capital-eligible offense, and whether to enter into a plea agreement. Otherwise, the sole power to authorize seeking the death penalty lies with the Attorney General. Once the Attorney General authorizes seeking the death penalty, the United States must file a notice of its intent to do so. See 18 U.S.C. § 3593(a). Each time the United States charges a defendant with a death-eligible crime, it must submit specific forms, including a recommendation on whether to seek the death penalty, a "Death Penalty Evaluation Form," and a memorandum outlining the theory of liability, the facts and evidence, including any evidence relating to any aggravating or mitigating factors, the defendant's background and criminal history, the basis for federal prosecution and other relevant information. See U.S. Attys. Man. § 9-10.040.

Bass requested from the Michigan United States Attorney's office all such materials relating to his prosecution, all policies or manuals used in the Eastern District of Michigan to determine whether to charge defendants federally, and a list of all death-eligible defendants in that district since January 1, 1995, including each defendant's race, and the ultimate disposition of each case. Bass also requested all materials submitted to the Attorney General for death-eligible prosecutions between January 1, 1995 and September 1, 2000, as well as captions and case numbers of such cases, a description of the offense charged, and the ultimate disposition of the case. Finally, Bass requested all standards, policies, practices, or criteria employed by the Department of Justice to guard against the influence of race in the death penalty protocol, any correspondence between the Department of Justice and the United States Attorneys regarding such policies or requesting identification of death-eligible defendants, and a list of all nonnegligent homicide cases throughout the United States since January 1, 1995, in which one or more offenders were arrested and charged and in which the facts would have rendered the offender eligible for the death penalty. As evidence in support of his discovery motion, Bass introduced, among other studies, the Department of Justice's Survey. Bass also introduced public comments regarding the Survey made on the day of its release by then-Attorney General Janet Reno and then-Deputy Attorney General Eric Holder, as well as comments by the current Attorney General, John Ashcroft. The United States opposed Bass's motion on the grounds that the requested information was protected by both the work-product and deliberative process privileges, that Bass had failed to make the evidentiary showing necessary to obtain further discovery, and that the requested materials were either non-existent or already in Bass's possession.

On October 24, 2000, following a hearing on Bass's motion, the district court found that he had presented sufficient evidence of racial bias in the death-penalty decision process to justify further discovery. The district court, noting that the United States did not offer any of the allegedly privileged materials for in camera review, further found that any privileges that may have attached to the materials were outweighed by the constitutional interests implicated by Bass's allegations and the death penalty context. Finally, it cited 18 U.S.C. § 3593(f), "Special precaution to ensure against discrimination," which requires a jury to determine that its individual members would have imposed a death sentence regardless of the defendant's race as a prerequisite to imposing such a sentence under Section 3593(e). The district court noted that for Section 3593(f) to have its intended effect of ensuring that a defendant's race plays no role in his death sentence, discovery of the sort requested by Bass must be allowed.

The United States refused to comply with the discovery order. On January 10, 2001, the district court sanctioned the United States by dismissing its notice of intent to seek the death penalty. The United States timely appealed.

II.

As an initial matter, we agree with the United States that we have jurisdiction to review the district court's pre-trial discovery order because that court's dismissal of the death penalty notice constitutes a final, appealable order under 18 U.S.C. § 3731. Under Section 3731, "an appeal by the United States shall lie to a court of appeals from a decision, judgment, or order of a district court dismissing an indictment or information. . . ." We have previously allowed the United States to appeal, under Section 3731, pre-trial discovery rulings. See, e.g., United States v. Presser, 844 F.2d 1275, 1280 (6th Cir. 1988). In Presser, we exercised jurisdiction over the United States's appeal from an order granting a defendant's discovery request because the district court indicated that if the United States failed to comply, it would suppress the relevant evidence, which would likely result in dismissal of the indictment. Here, the United States's failure to comply with the district court's discovery order resulted in dismissal of the death penalty notice — in effect, a partial dismissal of the charge. Accordingly, we find that, as in Presser, we have jurisdiction to hear the United States's appeal from the district court's pre-trial discovery order, and will now proceed to the merits of its argument.

III.

"It is well established that the scope of discovery is within the sound discretion of the trial court." United States v. One Tract of Real Property, 95 F.3d 422, 427 (6th Cir. 1996) (citations and internal punctuation omitted). We thus review a district court's discovery order in a criminal case for an abuse of discretion. See United States v. Kincaide, 145 F.3d 771, 780 (6th Cir. 1998). Under this standard, "the relevant inquiry is not how the reviewing judges would have ruled if they had been considering the case in the first place, but rather, whether any reasonable person could agree with the district court." Morales v. American Honda Motor Co., Inc., 151 F.3d 500, 511 (6th Cir. 1998).

The United States argues that Bass's evidence in support of his discovery request, including the Department of Justice's Survey and the Department officials' statements, did not constitute sufficient evidence of selective prosecution to warrant discovery. Accordingly, it contends that the district court abused its discretion in ordering the United States to produce the relevant documents. Bass does not dispute that the evidence he presented to the district court was insufficient to constitute a prima facie case of selective prosecution. He does, however, argue that the evidence was sufficient to warrant further investigation through discovery.

To make out a claim of selective prosecution, a defendant must show both a discriminatory effect and a discriminatory purpose or intent. See United States v. Armstrong, 517 U.S. 456, 465, 116 S.Ct. 1480, 134 L.Ed.2d 687 (1996). In Armstrong, the Supreme Court discussed the threshold showing a criminal defendant must make in order to obtain discovery on a selective prosecution claim. See id. at 463, 116 S.Ct. 1480. The Supreme Court noted that "in the absence of clear evidence to the contrary, courts presume that [prosecutors] have properly discharged their official duties." Armstrong, 517 U.S. at 464, 116 S.Ct. 1480 (citations omitted). Accordingly, "the showing necessary to obtain discovery [in a selective prosecution case] should itself be a significant barrier to the litigation of insubstantial claims." Id. Armstrong's plain language requires only that a defendant must present "some evidence tending to show the existence of the discriminatory effect element." Id. at 469, 116 S.Ct. 1480 (citations and internal punctuations omitted). Nonetheless, we have read Armstrong to require some evidence of the discriminatory intent element as well. See United States v. Jones, 159 F.3d 969, 978 (6th Cir. 1998). To establish discriminatory effect, a defendant "must show that similarly situated individuals of a different race were not prosecuted." Armstrong, 517 U.S. at 465, 116 S.Ct. 1480. As an example of "some evidence" showing a "discriminatory effect on blacks as compared to similarly situated whites," Armstrong cited a statistic showing that blacks were "at least 1.7 times as likely as whites" to have a state's disenfranchisement law applied to them. Id. at 467, 116 S.Ct. 1480 (citing Hunter v. Underwood, 471 U.S. 222, 105 S.Ct. 1916, 85 L.Ed.2d 222 (1985)). To establish discriminatory intent, a defendant must show that the prosecutorial policy "was motivated by racial animus." Jones, 159 F.3d at 976-77.

A. Bass's Evidence

Through the Department of Justice's Survey and other statistical evidence, Bass presented the following evidence tending to show that selective prosecution taints the death penalty protocol. First, the Survey showed a significant difference between the percentage of white and black prisoners in the general federal prison population (white: fifty-seven percent; black: thirty-eight percent) and those charged by the United States with death-eligible crimes (white: twenty percent; black: forty-eight percent). Of the seventeen defendants charged with a death-eligible crime in the Eastern District of Michigan, none were white and fourteen were black (the other three were Hispanic).

Second, the Survey showed that the United States entered into a plea bargain with forty-eight percent of the white defendants against whom it sought the death penalty, compared with twenty-five percent of similarly situated black defendants. The United States entered into plea agreements with twenty-eight percent of Hispanics, and twenty-five percent of other non-white defendants.

Third, the Survey showed that two of the three death-eligible offenses charged most frequently against whites and blacks were the same, but that the percentages by race of those charged with each crime were vastly different. Sixteen percent of death-eligible whites were charged with firearms murder, compared with thirty-two percent of death-eligible blacks. Fifteen percent of death-eligible whites were charged with racketeering murder, compared with twenty-two percent of death-eligible blacks. The Survey noted that firearms murder, racketeering murder, and continuing criminal enterprise murder (the three charges brought most frequently against death-eligible blacks) "can be charged in a wide array of circumstances, and [are] therefore more likely to be available as a charging option in a given case than more narrowly defined offenses such as kidnaping-related murder." However, death-eligible whites were most often charged with murder within a federal jurisdiction (twenty-one percent of all death-eligible whites).

Bass also introduced other statistics indicating that blacks are no more likely to commit violent federal offenses than whites. For instance, the United States Sentencing Commission's statistics for 1999 (the most recent statistics currently available) show that twenty-eight percent of people sentenced for federal murder were white, while eighteen percent were black. See 1999 Sourcebook of Federal Sentencing Statistics. In fact, there were only four federal offense categories where whites comprised twenty percent or less of the total defendants sentenced. The Commission's 1999 sentencing statistics reflect three of them: manslaughter (whites: seventeen percent; blacks: eleven percent), sexual abuse (whites: eighteen percent; blacks: seven percent), and immigration (whites: four percent; blacks: four percent). The Survey reflects the fourth: death-eligible defendants (whites: twenty percent; blacks: forty-eight percent). In contrast, the only federal offense reflected in the 1999 sentencing statistics where blacks represented forty-eight percent or more of the total defendants sentenced was robbery (blacks: forty-eight percent; whites: forty-one percent). In the few non-death-eligible offense categories in which blacks actually constituted a higher percentage of total offenders sentenced than whites, none reflected a statistical racial disparity comparable to the disparity reflected by the Survey for death-eligible charges.

In addition to the statistical evidence, Bass introduced public comments made on the Survey's release date by then-Attorney General Reno and then-Deputy Attorney General Holder who expressed concern over the significant racial disparities uncovered by the Survey. For instance, Holder commented:

I can't help but be both personally and professionally disturbed by the numbers that we discuss today. To be sure, many factors have led to the disproportionate representation of racial and ethnic minorities throughout the federal death penalty process. Nevertheless, no one reading this report can help but be disturbed, troubled, by this disparity.

In response to another question, Holder tacitly recognized that the Survey's results implicate the very concerns forming the basis of Bass's selective prosecution claim: "I'm particularly struck by the facts that African-Americans and Hispanics are over-represented in those cases presented for consideration of the death penalty, and those cases where the defendant is actually sentenced to death." (emphasis added). Reno also expressed concern over the Survey's results, even while acknowledging the various non-racial factors that could affect them: "So in some respects I'm not surprised [by the racial disparities], but I continue to be sorely troubled." While cautioning that intentional racial bias could not fairly be inferred simply as a result of the numbers, Reno emphatically endorsed future studies to determine whether the disparities were shaped, in part, by racial animus: "More information is needed to better understand the many factors that affect how homicide cases make their way into the federal system and, once in the federal system, why they follow different paths. An even broader analysis must therefore be undertaken to determine if bias does in fact play any role in the federal death penalty system" (emphasis added). Therefore, the top Department of Justice officials have taken the position that, although the Survey's results do not conclusively show intentional racial bias, neither do they conclusively show the lack of bias. Rather, in Reno's and Holder's view, the results demonstrate a clear racial disparity and raise questions warranting further study to determine whether that disparity is caused by intentional racial discrimination.

B. Discriminatory Effect

Bass's evidence shows the same type of statistical disparity the Supreme Court previously approved of in both Hunter and Armstrong as "indisputable evidence" of a law's discriminatory effect. For example, the evidence shows that although whites make up the majority of all federal prisoners, they are only one-fifth of those charged by the United States with death-eligible offenses. The United States charges blacks with a death-eligible offense more than twice as often as it charges whites. Cf. Armstrong, 517 U.S. at 467, 116 S.Ct. 1480 (citing evidence in Hunter as example of type of proof needed to obtain discovery in selective prosecution claim); Hunter, 471 U.S. at 227, 105 S.Ct. 1916 (citing single statistic showing blacks at least 1.7 times as likely as whites to have challenged state law applied to them). In addition, the United States charges blacks with racketeering murder one-and-a-half times as often as it charges whites, and with firearms murder (Bass's charge) more than twice as often as it charges blacks. Among death penalty defendants, the United States enters plea bargains with whites almost twice as often as it does with blacks. Under the "1.7 times" standard approved of in Armstrong, then, the statistics presented by Bass constitute sufficient evidence of a discriminatory effect to warrant further discovery as a matter of law.

The United States concedes that the Survey shows a statistical disparity at the charging stage, but argues that Bass's evidence does not satisfy the "similarly situated" requirement because Bass has failed to identify white defendants who could have been charged with death-eligible crimes but were not. We find, however, that with the plea bargaining statistics, Bass has identified a pool of similarly situated defendants — those whose crimes shared sufficient aggravating factors that the United States chose to pursue the death penalty against each of them. Of those defendants, the United States enters plea bargains with one in two whites; it enters plea bargains with one in four blacks. Therefore, the United States's assertion that Bass has failed to show a discriminatory effect on similarly situated whites and blacks is simply wrong.

Of course non-discriminatory reasons may explain such glaring discrepancies, but Bass need not address any of them at this pre-discovery stage. The current death penalty protocol leaves only three areas where the United States Attorneys can exercise discretion (and, of course, it is the manner in which that discretion is exercised that forms the basis of Bass's claim): bringing federal charges, bringing death-eligible charges, and plea bargaining. In the two areas addressed by the Survey — bringing death-eligible charges and plea bargaining — the racial disparities are clear. Viewing the totality of Bass's evidence, the district court did not abuse its discretion in finding that the statistical disparities are, at the least, some evidence tending to show the death penalty protocol's discriminatory effect warranting discovery.

C. Discriminatory Intent

With regard to the discriminatory intent element, we again find that Bass presented some evidence tending to show that the United States considers the defendant's race when determining whether to charge him or her with a death-eligible offense. The racial disparities identified by Bass in the death penalty charging phase do not occur in any non-death-eligible federal offenses. Therefore, they suggest that a defendant's race does play a role during the death penalty protocol. The Department of Justice's officials' comments bear this out. As Bass states in his brief, "[t]he precise point made by the Attorneys General and the former Deputy Attorney General is that they are deeply troubled because race may be systemically biasing federal capital charging, and that they need more information to know for sure" (emphasis added). If the Department of Justice's official position is that these statistics, standing alone, show sufficient evidence of the possibility of racial animus to warrant further study, we cannot fairly deny Bass the same opportunity to investigate when he has introduced not only the Survey, but several other statistics showing that the grave racial disparities identified by the Survey are unique to the death penalty protocol.

The United States attempts to preclude us from drawing any inference of intentional race discrimination from Bass's statistics by arguing that McCleskey v. Kemp, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987), prohibits such inferences. McCleskey held that statistics showing the discriminatory effect of a state's death penalty procedure do not, without more, constitute proof of a discriminatory intent. See id. at 294-95, 107 S.Ct. 1756. In McCleskey, however, the Supreme Court, sitting in federal habeas review, was addressing whether the defendant had carried his burden of proof on the merits of his selective prosecution claim. In contrast, we must determine only whether Bass has shown "some evidence tending to show the existence of . . . discriminatory intent" sufficient to warrant discovery. Jones, 159 F.3d at 978 ("Obviously, a defendant need not prove his case in order to justify discovery on an issue."). McCleskey will certainly preclude Bass's selective prosecution claim if, at the end of discovery, he fails to show any additional evidence that the United States intentionally discriminates against blacks through the death penalty protocol. It does not, however, pose any bar to Bass at this preliminary stage. Here as well, the United States has failed to show that the district court abused its discretion in ordering discovery on the grounds that the stark discriminatory effect of the federal death penalty protocol, coupled with the Department of Justice's official statements recognizing the possibility of intentional discrimination in light of the protocol's discriminatory effect, presents some evidence tending to show that race in fact plays a role in the United States's decision-making process.

IV.

We can dispose of both parties' remaining arguments relatively easily. The United States argues that the district court abused its discretion by citing 18 U.S.C. § 3593(f) as an alternative ground for its discovery order. The United States is correct. That statute, by its plain language, requires only that each jury member swear that he or she voted to impose the death penalty without regard to improper considerations, including the defendant's race. Jury members are simply not responsible for also asserting that the defendant was not prosecuted because of his or her race. Questions regarding the prosecution's motivation are more properly addressed in the type of selective prosecution claim Bass presents. Section 3593(f) does nothing to grant him a right to either obtain discovery or present evidence to a jury regarding alleged discrimination in the prosecution, and the district court abused its discretion in holding otherwise. Nonetheless, because we affirm the district court's discovery order on other grounds, our decision to reverse the district court on this issue does not effect the outcome of this appeal.

Second, the United States argues that the requested items are either not relevant, non-existent, or already in Bass's possession. The district court indicated that it would hear from the United States as to the unavailability or irrelevancy of particular documents, but the United States chose instead to refuse to comply with the entire order. That refusal also prohibited the district court from reviewing the requested documents to determine whether the United States' claimed privileges applied to any of them. Therefore, we find the record insufficiently developed to allow us to assess the merits of the United States's arguments relating to the content of the requested documents. Because of this, and because we think the district court should have the opportunity to review each requested item's relevancy and privileged status in the first instance, we remand to the district court with instructions to allow the United States to produce the documents for an in camera review. If the United States again fails to comply, the district court remains free to impose whatever sanction it deems appropriate under the circumstances.

V.

For the foregoing reasons, we AFFIRM the district court's discovery order, and REMAND for further proceedings consistent with this opinion.