United States v. Royal, 7 F. Supp. 2d 96 (D. Mass. 1998)

KEETON, District Judge.

The matter before the court is Defendant Jerome Royal's Motion for New Trial (Docket No. 230, filed December 24, 1997), with supporting submissions (Docket Nos. 231, filed December 21, 1997), opposition by the government (Docket No. 233, filed January 22, 1998), and additional documents before the court, including the transcript of two depositions. The first is the deposition of The Honorable Rya W. Zobel (Docket No. 236), who served as Liaison Judge for implementation of this court's Amended Jury Plan of 1989 during a period commencing before and extending beyond the time of selection of the panel for service in the trial of defendant Royal. The second is the transcript of the deposition of Judith Cook (Docket No. 238), who as Deputy Clerk was assigned as Jury Clerk, responsible for administration of the Plan during a period commencing before and extending beyond the time of selection of this jury.

The asserted ground for new trial is that defendant Royal "was denied his Sixth Amendment right under the United States Constitution and his statutory right under the Jury Selection and Service Act, 28 U.S.C. § 1861 et seq., to a jury selected at random from a fair cross section of the community." Docket No. 230.

Royal was convicted of one count of conspiracy to commit mail fraud and eight counts of mail fraud after a six-day jury trial in February, 1995. On October 13, 1996, this court sentenced Royal to 27 months imprisonment and 36 months supervised release, and ordered that he pay restitution of $30,000.00. The Court of Appeals affirmed as to issues unrelated to jury selection but reversed this court's denial of Royal's motion relating to inspection of records of the Clerk of the United States District Court for the District of Massachusetts and remanded with instructions to:

*98 allow Royal access to the contents of records or papers used by the jury commission or clerk in connection with the jury selection process ... in order to support a motion to strike the jury venire.

United States v. Royal, 100 F.3d 1019, 1026-27 (1st Cir. 1996).

Following receipt of the Mandate, this court set an initial hearing to schedule steps to be taken by the Clerk to comply with the Mandate, and at that hearing made an order initiating the process and providing for periodic status reports and hearings to resolve any disputes that might arise about implementation. The court also allowed defendant's motions for appointment and compensation of experts to assist defense counsel, and on several occasions allowed defendant an extension of time for filing a motion for new trial with a supporting submission particularizing the grounds of the motion and proffering evidence and expert opinion as well as legal argument in support. Eventually the court denied requests for additional extensions of time, and defendant's Motion for New Trial and supporting submissions were filed on December 24, 1997. The matter is now ripe for consideration on the motion and supporting and opposing submissions.

The manner in which the opposing parties have chosen to present their respective contentions has led to a failure of the parties' opposing memoranda to focus precisely upon what this court sees as a threshold issue with respect to which clarity is essential to a full understanding of the merits of this controversy. It is important not only that the court decide this controversy on the merits but also that the court's reasons for its decision be clear and understandable. Not only the parties but as well many others have a legitimate interest either in the outcome of this specific case or in the ongoing evaluation of Plans of this court and other federal and state courts for developing methods of jury selection that achieve in practice the constitutionally and statutorily declared aims.

An essential element of this court's Amended Jury Plan in effect at times material to the motion now pending is the meaning of the phrase "Master Jury Wheel." That phrase came into usage long before electronic data processing was possible. A large number of slips of paper or light cardboard, say 1,000, with a name of a prospective juror on each, were placed through an opening into a kind of circular basket, called a "wheel," the opening was closed, the wheel was spun to scramble the slips thoroughly, a blindfolded Jury Clerk reopened the circular basket, reached in and drew a designated number of slips, say 50, one by one, and that Deputy or another then recorded onto a list the names in the order drawn. Thus, the selection of which 50 among the 1,000 were drawn and the order in which they were drawn were random. With advances in electronic technology came some obvious, and some less obvious, possibilities for improving the whole process. As courts have adopted improvements, they have often continued to use the phrase Master Jury Wheel as a convenient label for a device that is not literally a wheel and probably contains no wheel anywhere within it. To communicate with each other about whether a modern Master Jury Wheel is functioning properly, we need to understand exactly what it is that we are labeling as our Master Jury Wheel for a particular set of uses in a particular court.

For the reasons explained here, I find that the "Synopsis of the Amended Jury Plan" as presented in Defendant Jerome Royal's Memorandum in Support of His Motion for New Trial is not a fair and accurate description of the Plan's provisions with respect to what constitutes the Master Jury Wheel for this court's Plan.

Defendant's synopsis states that, for the purpose of conducting an inspection of documents authorized to be inspected under the Mandate of the Court of Appeals and this court's orders for compliance with the Mandate, defendant's counsel created an "Inspection Team."

The defense "Inspection Team" consisted of two paralegals (Chad Steele and David Pacheco) and two legal interns (Meegan Thibaudeau and Michael Goodwin).

*99 During approximately three months commencing on March 17, 1997 and extending until on or about June 11, 1997, the "Inspection Team" reviewed the contents of 23 boxes of documents. Among these documents were computer-generated sheets listing prospective jurors selected by use of the computer software program created for the purpose of random selection from the large database used during a designated period of four years (or less, if so ordered for good cause). The "Inspection Team" reported that handwritten notes appeared on some of these computer-generated sheets. Also in the boxes were original juror qualification forms (completed by persons among those whose names appeared on the computer-generated sheets), correspondence, non-deliverable summonses, and folders that served to segregate one category of documents from another. Defendant characterizes these documents as "in the aggregate compris[ing] the Master Jury Wheel for the Eastern Division" of the District of Massachusetts. Defendant Jerome Royal's Memorandum in Support of His Motion for New Trial (Docket No. 231 at 4), citing as support for this characterization "Inspection Team Affidavits," Attachment 2 to Docket No. 231 at ¶¶ 5-17.

I find that the characterization of all these documents together as being the Master Jury Wheel is not consistent with the terms of the Amended Jury Plan of September 6, 1989, which is in the record as Attachment 1 to Docket No. 231. Part 7 of that document bears the caption, THE MASTER JURY WHEEL. The text of part 7 is as follows:

(a) The Master Jury Wheel for each division shall consist of the names and addresses of all persons randomly selected from the local resident lists in accordance with Section 6 of this Plan. The physical form of record on which names from the Master Wheel are kept may include labels or such electronic devices as punch cards, magnetic tapes or disc files.

(B) Initially, the Clerk shall place in each Master Jury Wheel the number of names that are perceived to be needed in order to provide qualified jurors for the Court, but this number shall always be at least 1,000 names. The Clerk shall empty and refill each Master Jury Wheel every four years in conformance with this Plan or at more frequent intervals as deemed necessary or expedient by the Clerk under the supervision of the Chief Judge. The Chief Judge, or in the absence of the Chief Judge, the Jury Judge, may order additional names to be placed in each Master Jury Wheel at other times as needed.

Part 6 of "this Plan" is as follows:

6. SELECTING NAMES BY MACHINE METHOD

(a) The Clerk finds that electronic data processing methods can be advantageously used for selecting and copying names from the local resident lists. Therefore, a properly programmed electronic data processing system may, at the Clerk's option and after consultation with the Chief Judge, be used to select master wheel names from the local resident lists, provided that the required proportions of names for each county are maintained.

(b) The Clerk shall request the Jury Commissioner to utilize the procedures outlined in Regulation No. 7, entitled Specifications of Random Selection Methods and Procedures, and attached hereto, for the random selection of the names to be placed in the Master Jury Wheel for each division so that each county shall be represented in proportion to the number of names on its resident lists.

Regulation No. 7 does not appear in the record submitted by the parties, but the court has obtained a copy of that Regulation from the files of the Clerk.

Regulation No. 7 sets forth specifications for the generation of random numbers and random selection procedures to be used by the Jury Clerk. ¶ 7.1. Each year before August 1, the Jury Clerk generates a computer file of random numbers. ¶ 7.2. The random numbers are uniformly distributed between zero and one, and each number is a decimal fraction consisting of a decimal point and seven digits to the right of the decimal point. ¶ 7.3. The numbers are arranged in blocks of ten thousand numbers per block. Id. A computer printout of the blocks is available for public inspection. Id.

*100 A Fortran computer program is used to generate the random number file. ¶ 7.7. The program uses the multiplicative congruential method, also known as the power residue method, to generate the random numbers. ¶ 7.8. A chi-square statistic is then used to check the integrity of the random numbers, and any block for which the chi-squared statistic exceeds the acceptable limit is discarded. ¶ 7.9.

The Jury Clerk generates such a list of random numbers for each city or town in the district before August 1st of each year. ¶ 7.12. Each random number corresponds to a particular resident appearing on the local resident list of the city or town. Id. The master juror list is the aggregate of prospective jurors from all cities and towns. ¶ 7.13.

Another program selects at random from a local resident list without replacement (meaning once a resident is selected, even if that resident is not used as a juror, his or her name is not returned to be chosen again). Town by town, each resident is drawn by the computer program. ¶ 7.15. As each resident is drawn, that resident's assigned random number is tested. Id. If the probability of drawing that person's ballot at that time is greater than that person's random number, that person's name is rejected and not included in the jury wheel list. If the probability of drawing that person's ballot is less than that person's assigned random number, that person's name is selected for inclusion in the jury wheel list. Id. This process is designed to ensure that representative numbers of persons are randomly selected from all Massachusetts cities and towns in the Eastern District.

The defense team has not challenged the random selection process itself. Rather, the defense team challenges the composition of the list for the Master Jury Wheel allegedly caused by the disproportionately low response to summonses by black persons, or caused by functions conducted by the Jury Clerks (such as granting excusals requested by persons who respond to a summons).

The government contends that an evaluation of the Master Jury Wheel and its functioning cannot properly be made on the basis of a one-year analysis. Government's Response (Docket No. 233 at 4 n. 2). The government also argues that even if defendant's proposal to do the analysis on the basis of only one year (1994) is accepted by the court, defendant still fails to meet its burden under United States v. Benmuhar, 658 F.2d 14, 19 (1st Cir.), cert. denied, 457 U.S. 1117, 102 S. Ct. 2927, 73 L. Ed. 2d 1328 (1982).

Defendant proposes 21,900 as the number of jury summonses mailed out and as the number of juror names in the Master Jury Wheel, and argues that "a random sampling of the 21,900 jurors summoned should yield 1,064 black jurors" (4.86% of 21,900). Defendant makes this argument even though acknowledging elsewhere, first, that only 415 of the "yield" (the completed or partially completed questionnaires returned) were identifiable as having come from "black jurors;" second, that more than that number of the persons who received questionnaires were Blacks; and, third, still more questionnaires were never in fact delivered even though deposited in the United States mail by a Jury Clerk, with addresses as shown in the list used for the Master Jury Wheel. Also, the defense argues that the court should find that 1,064 (all of 4.86% of 21,900) were "Blacks ... eligible for jury service" in the face of reasonable certainty that even if that many of the 21,900 were in fact black persons not all of them would be "eligible for jury service." I find that these government contentions have merit, and are factors that weigh significantly against the defense motion, even if not alone decisive.

The defense "inspection team" met with William Ruane, Chief Deputy Clerk, and Shawna Connolly, Jury Administrator, serving in their respective positions as officials of the United States District Court for the District of Massachusetts. In addition to asking questions of Mr. Ruane and Ms. Connolly about the jury selection process, the team *101 examined 23 boxes of materials related to 1994 jury selection.

The defense inspection team refers to the 23 boxes as comprising the Master Jury Wheel for 1994. See, e.g., Affidavit of Chad Steele at ¶ 5; Affidavit of Michael Goodwin at ¶ 5. For reasons explained above in Part II of this Opinion, I find that the definition of the Master Jury Wheel is significantly narrower than what the defense team proposes. In any event, the inspection team inspected the contents of the 23 boxes.

Essentially, the team counted the names on the computer-generated Master Jury Wheel sheets and determined that a total of 21,900 potential jurors were summoned in 1994. By also examining individual juror qualification forms, and supporting materials (such as requests for exclusion, doctors' notes, and excuse letters), the inspection team attempted (with partial success) to collect data regarding each juror's number; zip code; occupation; ethnicity; whether the juror had been qualified, disqualified, excused or exempt; and the reason for disqualification, exclusion or exemption.

Although juror qualification forms contain a box "for official use only" where a Jury Clerk could designate whether a particular juror was qualified, disqualified, excused, exempt, excluded, or "other," these boxes were often, if not uniformly, unmarked on the juror qualification forms. Instead, the Jury Clerks used a notation system on the computer-generated Master Jury Wheel sheets to indicate the status of a particular potential juror.

The inspection team examined each juror qualification form that contained some indication that the person was black. Through this examination, the inspection team reported the data they were able to collect regarding the numbers of black persons who

returned qualification forms (415)

qualified to serve on a jury (257)

appeared for orientation (167)

were labeled "no shows" by the Jury Clerk (24)

did not appear because a panel was canceled as not needed (66)

were disqualified, exempted, or excused (158)

See Inspection Team Affidavits at ¶ 17.

The defense team argues that of the 21,900 persons summoned, 1,064 (4.86%) of them should have been black. The defense team then subtracts from the 1,064 expected number of black persons 4.86% of the total summonses that were undeliverable, 4.86% of the total juror qualification forms not returned, and 4.86% of the total persons who declined to designate race. This resulted in a calculation of 838 "expected" black persons who were "randomly selected" and who "responded." See Docket No. 231 at 11. The defense team argues that "[w]hen a random selection of jurors produces less than one-half the number of black jurors that should have been generated randomly, common sense dictates that there is no dispute Blacks are underrepresented in the jury pool." Id.

The inspection team then calculates that all but 162 of the 415 black persons were disqualified, excused, or excluded. The team reports that ten black prospective jurors were excused without reasons given by the Jury Clerks. The defense team focuses on this apparent oversight by the Jury Clerks as "def[ying] the fair cross-section requirement of the Jury Act", and as an abuse of discretion. Docket No. 31 at 12, 24.

Based on the inspection team effort and the affidavit of defendant's expert witness, Gordon F. Sutton, the defense team argues that use of the resident lists from which the Master Jury Wheel names are drawn is problematic and underrepresents black persons because the resident lists are created in the same manner as the annual census, which is known to undercount black persons significantly.

Federal statutory provisions ordinarily require use of voter lists as a source of potential jurors. A special federal statute, however, makes a specific exception for jury selection in the Commonwealth of Massachusetts. See 28 U.S.C. § 1863(b) (3). The District of Massachusetts is permitted *102 to use resident lists, as provided for in Mass. Gen. L. ch. 234A, rather than voter lists. Resident lists are broader than voter lists because resident lists prepared by each city and town in Massachusetts are required to include all residents, regardless of voter registration status. See Mass. Gen. L. ch. 234A, § 10.

Undeniably, census-taking practices may undercount black persons. Defendant's expert witness, Mr. Sutton, criticizes the process for creating resident lists on the ground that "there is no corrective measure applied to the List to compensate for the undercounting of Blacks." Sutton Affidavit at ¶ 16.

Even if that criticism is credited, however, it does not support either the conclusions proffered or any definable remedy based upon them. This is so because in order to obtain randomly selected jurors, Jury Clerks and administrators need a list identifying each individual resident so that resident may be summoned and qualified. Knowing that a resident list undercounts black persons in a particular area, and even scaling up the number of estimated black persons in that area, would not benefit the jury selection process because it would not provide additional identifiable persons to whom summonses should be sent.

Mr. Sutton states that he is currently working with the Jury Commissioner of the Commonwealth of Massachusetts investigating alternative means of providing more complete resident lists. Sutton Affidavit at ¶ 19. The United States District Court for the District of Massachusetts lauds and encourages that effort. And if and when the Commonwealth adopts an alternative, more inclusive, basis for Commonwealth jury selection, the court will of course consider its use, if it is within the statutory authorizations and constraints under which this court must operate. That time has not yet come.

Failed summonses are those summonses that are either undeliverable, or not returned. The defense team argues that the Jury Plan fails to account for and address failed summonses. Mr. Sutton notes that failed summonses are more likely to occur in those zip code areas that have significant black populations. For example, 60.84% of summonses fail in zip code areas where 20% or more of the population is black, whereas only 14.87% of the summonses fail in zip code areas where less than 20% of the population is black. Sutton Affidavit at table 5.

The defense team has not provided the court with any supportable theory or hypothesis to explain, measure, and validate various potential reasons why summonses fail more often in predominantly black areas. See Docket No. 31 at 19-20; Sutton Affidavit at ¶¶ 22-35. Mr. Sutton does provide two suggestions as preventive or remedial measures (that the summonses be mailed in plain envelopes and that follow-up letters be mailed several weeks after a summons). Sutton Affidavit at ¶ 34. And, implicitly at least, these suggestions point to possible inferences as to why summonses fail. First, a summons may be undeliverable when a person moves and provides no forwarding address. If persons in some identifiable population groups tend to move more often, summonses of those persons will be more likely to fail. Second, persons who have had negative experiences with the criminal justice system, or the federal government, may be less likely to return summonses.

A universal practice of sending a follow-up letter might improve the ultimate response percentage, but any prediction as to whether the improvement would be greater for black persons than others, or in any event would be sufficient to outweigh the costs, is speculative.

The defense team, with supporting references to Mr. Sutton's Affidavit, further complains that the district court has failed to enforce § 1864(b) of the Jury Act requiring the district court to order persons who fail to appear for jury service to appear and show cause for that failure. See Docket No. 31 at 23.

Lax enforcement, however, cannot account for the disparity in the numbers of white and black persons that respond to summonses. *103 Even if increased resources were committed to enforcement (and the defense team does not suggest methods or means for increased enforcement), improvement in responses would be likely across all races and geographic areas. No basis appears in the record now before the court for expecting a differential increase that would significantly affect the percentage of black persons appearing for jury service. Perhaps improved response could be effected from identifiable zip code areas. We do not know, however, what effects race, low income, and other factors have, and whether more effective enforcement would increase potential juror responses differentially.

The defense team is critical of the one-step juror summoning and qualification process. The Jury Act allows this process. 28 U.S.C. § 1878(b). The defense team correctly notes that a challenge resting solely on the one-step characteristic cannot succeed. They seek to overcome this barrier by charging that Jury Clerks abuse discretion by "systematically" excluding black jurors. With one exception, however, they fail to identify any exercise of discretion other than use of the one-step process to support the charge of "systematically" excluding black persons. That exception is that they argue that excusing ten black persons with no assigned reason is an abuse of discretion. But to show a differential effect, the team would need to probe into excuses of non-blacks without assigned reasons. And in any event the number ten is too small to support a broad inference of systematic discrimination. Thus, the team still has not provided any reasoned basis for challenging the one-step process. They have presented no evidence as to why the one-step process might be less likely to produce potential black jurors or more likely to exclude black persons, or how any two-step process could be made systematically to work better in this respect.

The government does not challenge, and the court will accept as true, the fact that the Jury Clerk did excuse ten black jurors for no apparent reason. That is, for ten black jurors who were excused, the Jury Clerk listed no reason for their being excused.

Failure to provide the reasons for excusing ten out of 104 black jurors does not amount to a material abuse of discretion. One deficiency in the defense argument is a failure to show that the Jury Clerk failed to list a reason for excusing white jurors a significantly lower percentage of the time.

An obvious fallacy in the defense team's data analysis concerns their flawed definition of Master Jury Wheel, as explained in Part II above. The Master Jury Wheel includes not just those persons who respond to the summons, but the entire group of persons to whom summonses are mailed. Thus, the defense team's use of the number of black persons as 415 for calculating disparities is erroneous. The number of black persons to whom summonses were mailed is indisputably larger than the number of black persons who responded to the summonses. Because the Jury Clerks (and we, acting with the benefit of hindsight) do not know the race of a person until that person responds, if even then, it is impossible for the Jury Clerks to tell exactly the number of black persons in the Master Jury Wheel.

The government proposes "adding back" 304 black persons to the Master Jury Wheel based on its calculation using data from Table 3 of Mr. Sutton's Affidavit. This suggestion is as flawed as the argument to which it responds. Adding back of any number of persons expected to exist, but not individually identifiable, is spurious.

The purpose of calculating any disparity is to quantify the discrepancy between the percentage of black persons that the plan was expected to produce and the percentage actually produced. No precedent has been cited for using some percentage other than the actual percentage of responding persons who can be identified as being black as the percentage to be compared with a target *104 percentage. Specifically, the number properly used in the disparity calculations is not the number of black persons in the Master Jury Wheel (a number that can only be estimated), but the number of black persons actually responding to summonses.

The defense team did make an effort to calculate the number of black persons expected out of the 21,900 prospective jurors summoned. In this effort, the defense team assumes that since 4.86% of the population of the Eastern Division is black, 4.86% of the non-deliverable summonses were for black persons, 4.86% of the unreturned summonses were from black persons, and 4.86% of the jurors who declined to designate race were black. See Docket No. 31 at 10. But, the defense team's own analysis belies its argument.

Since failed summonses (non-deliverable summonses and unreturned summonses) are more likely to occur in zip code areas with greater than 20% black population, one reasonably infers that greater than 4.86% of the unreturned and undelivered summonses were from black persons. This may partially account for why only 415 black persons responded, and not the 1,024 expected from the defense team's calculations. Furthermore, without having some basis for saying whether black and white persons are equally likely to fail to designate race on the juror qualification form, one cannot validate the assumption that 4.86% of those persons who did not designate race were black.

The defense team also states that the "universe of potential black jurors in the Eastern Division is roughly 170,000" based on the approximate black population over eighteen years of age in the Eastern Division. See Docket No. 31 at 19. As the government points out, however, not all of these black residents of the Eastern Division are eligible to serve as jurors. Thus, the numbers by the defense team in both its absolute and comparative disparity calculations are overstated. This overstatement results in a greater disparity than would be shown if the percentage of those black persons responding to summonses were compared to the percentage of those eligible for service in the Eastern Division. But, again, we are faced with the problem of a lack of data on how many among the 4 .86% of the black population of the Eastern Division are eligible to serve. In these circumstances, I find that it is appropriate to use the best available data on the percentage of the population that is black in making comparisons.

The Supreme Court has provided a framework for determining whether a defendant has made a prima facie case showing violation of the fair cross-section requirement:

(1) that the group alleged to be excluded is a "distinctive" group in the community, (2) that the representation of this group in venires from which juries are selected is not fair and reasonable in relation to the number of such persons in the community; and (3) that this underrepresentation is due to systematic exclusion of the group in the jury-selection process.

Duren v. Missouri, 439 U.S. 357, 364, 99 S. Ct. 664, 58 L. Ed. 2d 579 (1979).

No dispute exists in this case that black persons represent a distinctive group. See Docket No. 233 at 4 n. 1. The analysis next turns to whether the representation of black persons is "fair and reasonable." Courts have typically used one of two measures of disparity for this purpose: absolute disparity or comparative disparity.

Absolute disparity measures the shortfall between the percentage of eligible black persons and the percentage of black persons serving. See Pion, 25 F.3d at 23 n. 5 (absolute disparity compares the percentage of eligible black persons with the percentage of black persons on the Master Jury Wheel). But see United States v. Yazzie, 660 F.2d 422, 426 n. 3 (10th Cir.1981) (defining absolute disparity as a comparison of the percentage of eligible black persons with the percentage of black persons on the venire), cert. denied, 455 U.S. 923, 102 S. Ct. 1282, 71 L. Ed. 2d 464 (1982). Comparative disparity, on the other hand, is calculated by dividing the shortfall by the percentage of black persons in the eligible population. See Pion, 25 F.3d at 23 n. 5. In other words, comparative disparity "measures the diminished likelihood that members of an underrepresented group, *105 when compared to the population as a whole, will be called for jury service." Ramseur v. Beyer, 983 F.2d 1215, 1231 (3d Cir.1992), cert. denied, 508 U.S. 947, 113 S. Ct. 2433, 124 L. Ed. 2d 653 (1993).

The defense inspection team states that black persons "comprise only 1.89% of the prospective jurors on the Master Jury Wheel." Docket No. 31 at 10. Although they do not explicitly say so, this number is calculated by dividing 415 (the number of black persons who received summonses and responded) by 21,900 (the number of jurors on the list for the Master Jury Wheel). This calculation yields an absolute disparity of 2.97%, and a comparative disparity of 60.9%. These calculations understate the percentage of black persons because they compare (i) black persons who responded to (ii) all those (of whatever race) summoned, and not instead to all those actually responding after having been summoned.

The Court of Appeals for the First Circuit observed this difficulty in United States v. Pion, where the spread between the percentage of eligible Hispanics and the percentage of Hispanics on the list for the Master Jury Wheel had not been established, and could not be discerned, because it was not disclosed on the appellate record. 25 F.3d 18, 23 (1st Cir.1994). This case presents the same difficulty, as the defense team points out, because the race and ethnicity of the persons on the list for the Master Jury Wheel is not known until they respond by sending in their jury qualification forms.

In Duren, the Supreme Court noted that the 14.5% of women at the final venire stage was not "reasonably representative" of the community (54% of which was female). 439 U.S. at 365, 99 S. Ct. 664. The Duren analysis calculated the percentage of the total persons summoned for service that were women (26.7%). Id. at 367, 99 S. Ct. 664. The opinion does not disclose how this was possible. Perhaps the sex of each person was known from voter registration lists, or perhaps each person's sex could be inferred from his or her name. In any case, when evaluating the race and ethnicity of persons on the list for the Master Jury Wheel, we do not have such a luxury.

Given the lack of information about race of persons on the list for the Master Jury Wheel before they respond, we might consider instead, comparing the number of black persons responding (415) to the total number of persons responding (18,451), this alternative calculation yields a result that 2.25% of the jurors receiving summonses and responding were black. The shortfall by this comparison is then 2.61% (4.86-2.25). Thus, the absolute disparity by this method of calculation would be 2.61% and the comparative disparity would be .54% (2.61/4.86).

Regardless of whether this court considers the defense team's proposed disparity numbers, or the numbers obtained by the different analysis suggested above, defendant Royal's Motion for a New Trial (Docket No. 230) must be rejected. The defense team argues that the absolute disparity approach adopted by the First Circuit does not bind this court. That argument fails for two reasons.

First. As the defense team recognizes, in Pion the defendant had not timely raised the issue of jury composition. Thus the court's absolute disparity analysis was not essential to disposition. Of course partisans may, as the defense team chooses to do, refer to the analysis in such an opinion as mere orbiter dicta. See Docket No. 231 at 15 n. 10. A district court, however, must be respectful of reasoned dicta in higher court opinions. In any event, the Court of Appeals for the First Circuit had already adopted absolute disparity and rejected comparative disparity in United States v. Hafen, 726 F.2d 21, 24 (1st Cir.), cert. denied, 466 U.S. 962, 104 S. Ct. 2179, 80 L. Ed. 2d 561 (1984).

Second. The defense team, even though pointing to Chief Judge Torruella's discontent with the use of absolute disparity analysis, as an important factor, still constructs a form of intricate argument in the course of which they suggest that this court should not disregard that analysis. As noted above, however, the defense team also argues elsewhere that references to disparity analysis were orbiter dicta, apparently intimating that it could properly be disregarded. Inconsistent arguments in the alternative are permissible *106 advocacy, but often, as in this instance, unpersuasive in both prongs.

Although the defense team did not provide relevant authority on this point, if this jurisdiction were to adopt the comparative disparity analysis, a strong possibility exists that this and higher courts would conclude that the defense team's 60.9% comparative disparity or the 54% comparative disparity calculated through the alternative analysis would not rise to the level of a constitutional violation. This is so because most courts that consider comparative disparity also consider absolute disparity as well, and here the absolute disparity (calculated at 2.61% or 2.97%) is not egregious. See, e.g., Ramseur, 983 F.2d at 1231 (treating 14.1% actual and 39.3% comparative disparity as insufficient); United States v. McAnderson, 914 F.2d 934, 941 (7th Cir.1990) (treating 8% actual and 40% comparative disparity as not establishing a constitutional violation), cert. denied sub nom. Cranshaw v. United States, 513 U.S. 953, 115 S. Ct. 372, 130 L. Ed. 2d 323 (1994); Yazzie, 660 F.2d at 427 (treating 4.29% absolute disparity and 46.3% comparative disparity as insufficient). But see United States v. Rogers, 73 F.3d 774, 776 (8th Cir.1996) (treating .579% actual and 30.96% comparative disparity as sufficient), cert. denied, 517 U.S. 1239, 116 S. Ct. 1889, 135 L. Ed. 2d 183 (1996). On the other hand, many circuits have rejected comparative analysis altogether. See, e.g., United States v. Sanchez-Lopez, 879 F.2d 541, 547 (9th Cir.1989); United States v. Maskeny, 609 F.2d 183, 190 (5th Cir.), cert. denied, 447 U.S. 921, 100 S. Ct. 3010 (1980); United States v. Rioux, 97 F.3d 648, 655 (2d Cir.1996); United States v. Grisham, 63 F.3d 1074, 1078-79 (11th Cir.1995), cert. denied, 516 U.S. 1084, 116 S. Ct. 798, 133 L. Ed. 2d 746 (1996).

If outside precedents differ from those of the Court of Appeals for this Circuit, I must follow the guidance of this Circuit's Court of Appeals. If the defense team wishes to change the law of the Circuit, they must do so in the appellate court, though of course I understand their need to present the argument here as a prerequisite to arguing it on appeal.

I find that the precedents in this jurisdiction direct this court to apply absolute disparity analysis, as at least a factor if not alone decisive of a discretionary choice. See Pion, 25 F.3d at 23; Hafen, 726 F.2d at 24. Applying the absolute disparity analysis, along with other factors bearing on the decision, I find that the defense team's contention fails. An absolute disparity of either 2.61% (as calculated by the court) or 2.97% (as calculated by the defense team) does not constitute a "gross discrepancy" of such dimension as to require the conclusion that the venire from which Jerome Royal's jury was chosen was not "reasonably representative." Duren, 439 U.S. at 365-66, 99 S. Ct. 664; see also Rioux, 97 F.3d at 657 (finding 1.58% absolute disparity insufficient to constitute Sixth Amendment violation); United States v. Esquivel, 88 F.3d 722, 726 (9th Cir.1996) (finding 4.9% absolute disparity insufficient to constitute Sixth Amendment violation), cert. denied, ___ U.S. ___, 117 S. Ct. 442, 136 L. Ed. 2d 339 (1996); Grisham, 63 F.3d at 1078 (finding less than 10% absolute disparity insufficient to constitute Sixth Amendment violation).

As the defense team has failed to satisfy the second prong of the Duren prima facie case, there is no need to examine more critically the defense team contention that they have shown some degree of systematic exclusion.

In addition, the defense team arguments fail to show how a court could come to a reasoned conclusion that a new trial would be a remedy tailored to any constitutional, statutory, or regulatory violation occurring in the design, implementation, and administration of the Jury Plan of the United States District Court for the District of Massachusetts.

For the foregoing reasons, it is hereby ORDERED:

(1) All pending motions other than Jerome Royal's Motion for a New Trial are DISMISSED as moot.

(2) The Clerk is directed to enter forthwith on a separate document a Final Order as follows:

*107 For the reasons stated in the Opinion of June 12, 1998, it is Ordered:

Jerome Royal's Motion for a New Trial (Docket No. 230) is DENIED.