In the United States Court of Appeals For the Seventh Circuit
July 20, 2010
UNITED STATES OF AMERICA, PLAINTIFF-APPELLEE,
MICHAEL MCGEE, JR., DEFENDANT-APPELLANT.
Appeal from the United States District Court for the Eastern District of Wisconsin. No. 07-CR-177-C.N. Clevert, Jr., Chief Judge.
The opinion of the court was delivered by: Easterbrook, Chief Judge.
ARGUED APRIL 16, 2010
Before EASTERBROOK, Chief Judge, FLAUM, Circuit Judge, and HIBBLER, District Judge.*fn1
Michael McGee, Jr., was elected to the Milwaukee Common Council in 2004. Almost immediately he began to demand payments from businesses that depend on liquor licenses and other permits that he could control, because the Common Council allowed each Alderman to determine licensing and zoning questions within his own district. Some of the proprietors grumbled to their friends that McGee's demands had made their businesses unprofitable. One proprietor's friend contacted the FBI, which began a criminal investigation that was assisted by a wiretap on McGee's phones. The recordings establish that McGee used his public office to wring money from constituents. He was arrested in May 2007 and charged with extortion, 18 U.S.C. §1951(a); solicitation of bribes, 18 U.S.C. §666(a)(1)(B); and structuring financial transactions to evade reporting them, 31 U.S.C. §5324(a). The jury convicted on all nine counts, and the judge sentenced McGee to 78 months' imprisonment plus $107,433 in restitution.
The evidence of guilt is strong and for the most part undisputed. The victim who paid the most was Adel "Jack" Kheirieh, who testified in detail to McGee's demands. Adel gave McGee cash, cell phones (liberally stocked with air time), and other gifts, because McGee threatened to terminate his liquor license, on which his business depended. McGee contends that the evidence on some counts is insufficient. The events underlying these counts occurred during the investigation; many of the encounters were recorded or even scripted by federal agents. For example, the transaction in Count 2-Adel's payment of $750 to McGee by money order-was recorded. McGee says that the evidence is insufficient because, although the recording proves that McGee requested and received payment, it does not prove that his intent was corrupt. Yet McGee mentioned that, to raise what he called "seed money," he had "sent" a "message" to another business by having its liquor license revoked. That's a threat to do the same to Adel unless he paid. McGee did not record the $750 as a campaign contribution and can't use that explanation for taking the money. See United States v. Allen, 10 F.3d 405, 412--13 (7th Cir. 1993). Other details, such as whether the FBI supplied this money order, do not matter. It is not necessary to traipse through the record count by count; the evidence supports all convictions.
McGee's principal argument is that the trial's first day included a narration of his guilt based on hearsay-and that's indeed what happened. An FBI agent told the jury that to obtain a warrant for a wiretap the prosecutor had to establish, to a judge's satisfaction, that the telephone was being used to commit a crime. This agent recounted what a preliminary investigation had revealed and why the United States Attorney and high-ranking officials at the Department of Justice thought it enough to support audio interception of McGee's phone calls. Then the agent explained that District Judge Adelman, who issued the warrant for the interception, agreed with this conclusion. The warrant, which recites some of this evidence (and the judge's conclusion), was introduced into evidence. Before the trial was two hours old, the essence of the prosecutor's case had been laid before the jury. And not a word of this evidence was from a witness with first-hand knowledge or subject to cross-examination. The process violated both the confrontation clause of the sixth amendment and the hearsay rule.
The prosecutor's stated rationale for exposing the jury to this damning hearsay was that it "laid a foundation" for admission of the wiretaps. Yet admissibility of evidence is a preliminary question for the judge. See Fed. R. Evid. 104(a); United States v. Martinez de Ortiz, 907 F.2d 629 (7th Cir. 1990) (en banc). There was no need to put hearsay before the jury in order to make the intercepted conversations admissible. In other cases, prosecutors have justified evidence of this kind by a supposed need to explain that in real life, unlike the world of movies and TV programs, employees of mysterious "deniable" agencies can't go around listening to other people's conversations on their own say-so. Federal agents need to persuade politically visible and responsible supervisors, then get judicial permission. It may be well and good to inform juries that wiretaps need authorization-but the means used in this trial is not the way to do it. The right way is for the prosecutor (in an opening statement) or the judge to tell the jury that judicial permission is required and was received, and that the process of listening is subject to statutory controls. There was no legitimate reason to present hearsay about the particulars of McGee's activities or the findings of the judge who issued the warrant. Evidence must be submitted through witnesses with personal knowledge, and subject to cross-examination.
Four years ago, we held in United States v. Cunningham, 462 F.3d 708 (7th Cir. 2006), that it is improper to introduce hearsay under the rationale of assuaging jurors' fears about uncontrolled snooping, and that the defendant is entitled to a new trial if an objection is made and overruled. The evidence smuggled in by a "how and why we obtained a phone-intercept order" summary is not only hearsay but also irrelevant (the validity of the order is for the judge, not the jury, to determine). In two cases in which the defense did not object, by contrast, we concluded that the introduction of hearsay was not plain error, because the evidence eventually came in properly, by live testimony and the recordings themselves. See United States v. McMahan, 495 F.3d 410, 416--18 (7th Cir. 2007), vacated on other grounds under the name Smith v. United States, 552 U.S. 1091 (2008); United States v. Noel, 581 F.3d 490, 496--99 (7th Cir. 2009).
McGee's trial occurred 22 months after our opinion in Cunningham. The prosecutor should have known that he was eliciting inadmissible testimony. The judge should have known it too, yet did nothing. And defense counsel likewise must have understood that the testimony was out of bounds-yet he did not object. It is unlikely that counsel was asleep; the hearsay rule is second nature to any trial lawyer. Perhaps he viewed the prosecutor's misstep as a godsend. Evidence of McGee's financial exactions was going to come in from the victims, who had personal knowledge, and their testimony would be bolstered by recordings from wiretaps plus hidden microphones and cameras. The main thing the hearsay did was create an issue for appeal. A lawyer who knows that the evidence is solidly against his client may see strategic value in allowing error to occur, despite the fact that the plain-error standard will make it hard to upset the verdict on appeal.
Even if we are wrong in suspecting that counsel's silence was strategic (which would imply waiver and not just forfeiture), the standard of plain-error review has not been satisfied. As the Supreme Court reiterated only a few weeks ago, the plain-error standard is hard to satisfy. See United States v. Marcus, 130 S.Ct. 2159 (2010). Defendant must establish, among other things, an adverse effect on his "substantial rights," which means serious prejudice, and on this issue the defendant bears the burden of persuasion. (This is one of several ways in which plain-error review is more confined than harmless-error review, the standard applicable when an objection is made and erroneously denied.) McGee does not say that any important part of the FBI agent's narration was left without support from admissible evidence introduced later. That would be prejudice but did not occur in this trial any more than in Noel or McMahan, and so the plain-error standard has not been met.
The only way in which this case differs from McMahan is that the wiretap orders were admitted into evidence. These orders contained Judge Adelman's finding that McGee had committed, and was continuing to commit, the sorts of acts for which he was on trial. A judicial finding may have a strong influence on jurors-stronger than it should, since the jurors may not appreciate that Judge Adelman's findings were based on a one-sided presentation by the Department of Justice. (Wiretaps must be kept secret or they won't be useful, so there cannot be an adversarial presentation before the order is issued.) But counsel was free to point out to the jury the limits of ex parte probable-cause findings. Given the overwhelming evidence at trial (all subject to cross-examination), which showed that Judge Adelman's findings were right, it is not possible to treat the jurors' knowledge of the intercept warrants as undermining McGee's substantial rights. More than that: defense counsel said at trial that he had "no objection" to the admission of the intercept orders and so has waived this topic. Not even plain-error review is possible.
Although McGee is not entitled to a new trial, we are dismayed by the prosecutor's conduct and disappointed by the district judge's failure to intervene. The extensive hearsay did not slip in by accident, in the heat of the moment; the prosecutor must have carefully planned this line of testimony. The proper way to introduce jurors to forthcoming wiretap evidence ought to be featured in the United States Attorney's Manual. The United States has not attempted to defend the propriety of the prosecutor's tactics. Waiver and the plain-error doctrine may insulate judgments from reversal, but recurrence of an episode such as this may lead to the opening of a disciplinary proceeding for the lawyers involved.
We turn to another of McGee's arguments. The district judge anticipated that the trial would be lengthy. At a pretrial conference he told counsel that he was inclined to permit both the prosecutor and defense counsel to summarize the evidence occasionally, so that the jurors could keep their bearings and maintain concentration. Principle 13G of the American Bar Association's Principles for Juries and Jury Trials (2005), recommends that judges allow such interim summaries in lengthy trials, whether civil or criminal. The Seventh Circuit American Jury Project tested seven of the ABA's proposals in trials before more than a dozen participating judges; mid-trial summaries were among the tested proposals and were used in 17 civil trials. Both the judges and counsel concluded that the summaries had helped jurors organize the evidence better, improving their attention and under-standing. None of the participants thought the summaries were hurtful, though 8% thought that they did not help much either. American Jury Project, Final Report 32--35, 63--65 (2008). The district judge, persuaded by the ABA's proposals and the Jury Project's favorable results, decided to use this approach. McGee says that doing so violated the due process clause of the fifth amendment and entailed a "structural error" that requires reversal without regard to whether any injury to the defendant ensued.
Several decisions have concluded that mid-trial summaries are permissible in civil trials. See, e.g., Consorti v. Armstrong World Industries, Inc., 72 F.3d 1003, 1008 & n.1 (2d Cir. 1995), vacated on other grounds under the name Consorti v. Owens-Corning Fiberglas Corp., 518 U.S. 1031 (1996); ACandS, Inc. v. Godwin, 340 Md. 334, 407--13, 667 A.2d 116, 152--54 (1995). But one court has held that mid-trial summaries are unconstitutional and amount to structural error in criminal trials-at least if a summary follows each witness's testimony and the judge allows the summaries to be argumentative rather than organizational. United States v. Yakobowicz, 427 F.3d 144 (2d Cir. 2005). McGee urges us to follow Yakobowicz.
Much of the reasoning in that decision responds to the fact that the judge allowed argumentative summaries after each witness's testimony. The second circuit thought that, because prosecutors usually present more witnesses than defendants, the process is bound to tip the scales in the prosecutor's favor even though the defense may use the opportunity to poke holes in the testimony and the prosecutor's case. The majority in Yakobowicz also worried that witness-by-witness summaries would force the defense to commit to a theory of the case before the prosecutor was done, undermining the defense's entitlement to wait until the prosecutor rests before announcing any theory (or deciding what evidence, if any, to offer in reply). These concerns do not arise when the district judge allows only non-argumentative summaries that are spaced days apart, so that they do not reflect the number of witnesses each side presents.
In McGee's case the district judge allowed just one opportunity to each side, after the only weekend break. (The trial concluded before the second weekend; it was shorter than counsel had estimated.) The judge thought that a refresher would help jurors regain their focus after the break. The prosecutor spoke for seven minutes; his remarks cover five pages of the transcript. He reminded the jury that the indictment had nine counts and summarized in just a few sentences per count where the evidence stood. He did not present argument, ask rhetorical questions, or propose contestable inferences. It was a simple "just the facts" recap of the sort Joe Friday would have approved. We doubt that the second circuit would see a problem with this procedure.
And if it would-well, we think that Yakobowicz over-stated the risks and understated the potential benefits. The majority in that decision seems to have been unaware that the use of mid-trial summaries has been studied in criminal as well as civil trials, and that the opinion's fears have not come to pass. The report of the Seventh Circuit American Jury Project had not been released when Yakobowicz was issued, but other reports predated that decision. For example, a pilot program in Tennessee used mid-trial summaries in both criminal and civil cases, and the participants found that the summaries helped jurors. See Neil P. Cohen & Daniel R. Cohen, Jury Reform in Tennessee, 34 U. Mem. L. Rev. 1, 31--34 (2003). Two psychologists concluded that mid-trial summaries should reduce the prosecutor's advantage in a criminal trial by allowing the defense to undermine the prose-cution's case from the outset by narratives and not just cross-examination. See Saul M. Kassin & Lawrence S. Wrightsman, The American Jury on Trial: Psychological Perspectives 136--37 (1988). They observed that the prosecutor benefits from the primacy effect-that people give extra weight to the first information they learn about a subject. Summaries during trial may help jurors understand that the first information is not necessarily the best, and if so the summaries will improve the accuracy of verdicts.
It is hard to see why mid-trial recaps should be allowed in civil trials but categorically forbidden in criminal trials, as McGee contends they should be. Support for summaries in civil trials is widespread. See, in addition to sources we've mentioned already, Federal Judicial Center, Manual for Complex Litigation §§ 12.21, 12.34 (4th ed. 2004); New York State Bar Association, Committee on Federal Courts, Improving Jury Comprehension in Complex Civil Litigation, 62 St. John's L. Rev. 549, 557--58 (1988); State Bar of Texas, Report of the Court Administration Task Force 54 (2008); B. Michael Dann, "Learning Lessons" and "Speaking Rights": Creating Educated and Democratic Juries, 68 Ind. L.J. 1229, 1255--56 (1993); Tom M. Dees, III, Juries: On the Verge of Extinction? A Discussion of Jury Reform, 54 SMU L. Rev. 1755, 1778--80 (2001); William W Schwarzer, Reforming Jury Trials, 1990 U. Chi. Legal Forum 119, 144--45; Douglas G. Smith, Structural and Functional Aspects of the Jury: Comparative Analysis and Proposals for Reform, 48 Ala. L. Rev. 441, 537 (1997). If there are skeptics, they have kept their silence.
Yakobowicz thought that criminal trials are different because juries are not supposed to reach conclusions until all of the evidence is concluded. That's true of both civil and criminal trials, however. In both civil and criminal trials jurors are exposed to persuasion from the start: lawyers get to make opening statements, and questions are asked in a way that lawyers hope will influence jurors, who inevitably form tentative opinions as they hear evidence. They must keep their minds open so that opinions can change as more evidence comes in, but this does not imply that jurors, civil or criminal, are supposed to be empty vessels until they hear the judge's instructions at the very end. The sort of objections to summaries advanced in Yakobowicz also have been essayed against allowing jurors to take notes or ask questions, but those procedures have been approved in this circuit, and elsewhere. See SEC v. Koenig, 557 F.3d 736, 741--42 (7th Cir. 2009).
The second circuit observed that argumentative questions (and for that matter argumentative objections to questions) are disallowed, but that's true of both civil and criminal trials-and the reason for keeping argument out of questions is to avoid harassing witnesses and prevent an asymmetric and time-consuming presentation. Both sides can recapitulate the evidence; summaries are not windy, unilateral harangues, as argumentative questions can be. Yakobowicz also observed that there is less discovery in criminal cases than in civil, which is true, but defendants usually know more about the prosecutor's case than the prosecutor knows about the defense case, and defendants can keep it that way if they prefer. They need not use summaries to tip their hands. It is hard to see how interim summaries could change the relative informational differences established by Fed. R. Crim. P. 16.
And we just don't see why Yakobowicz perceived a constitutional problem with mid-trial summaries. The due process clause is not a code of trial procedure. Many changes have occurred since 1791. One of the principal changes is an increase in the length of trials. In the eighteenth century multiple criminal trials were held in a single day. See James D. Rice, The Criminal Trial Before and After the Lawyers: Authority, Law, and Culture in Mary-land Jury Trials, 1681--1837, 40 Am. J. Legal Hist. 455, 463 (1996); John H. Langbein, Shaping the Eighteenth-Century Criminal Trial: A View from the Ryder Sources, 50 U. Chi. L. Rev. 1, 115--23 (1983). Today, by contrast, a single criminal trial can last multiple weeks or months. When trials are short, there's no need for mid-trial recapitulations; when trials are long, jurors' attention and memory may wane, and the opening and closing statements may be too far apart. Nothing in the constitutional text, or the original practice, implies that days or even months must pass without any opportunity for the lawyers to give the jurors their views about where the evidence stands.
Summaries equally available to both sides are no more objectionable than discovery, note-taking by juries, sending written jury instructions to the jurors, and the many changes to the rules of evidence that have accreted during the 219 years since the fifth amendment was approved. Some rules of trial procedure-juries, counsel, confrontation, and cross-examination-are in the bill of rights. Most are not. Living judges and legislatures may decide that incremental changes in trial procedure are beneficial.
Finally, we are not persuaded by the second circuit's conclusion that any misstep with respect to interim summaries is a structural error. Circuit Judge Sotomayor was willing to assume that the district judge in Yakobowicz erred by allowing argumentative summaries after each witness. But she disagreed with the majority's conclusion that such an error is "structural." See 427 F.3d at 154--58 (dissenting opinion). Judge Sotomayor's view was vindicated by the Supreme Court's decision in Marcus, which stressed that errors are "structural" only when they change the fundamental framework of the trial-when, for example, the judge is biased, the defendant lacks counsel, or a vital phrase such as "reasonable doubt" is misde-fined, so that the jurors do not understand their task. The Justices regularly declare that errors of trial management are not structural. See, e.g., Rivera v. Illinois, 129 S.Ct. 1446 (2009) (improper denial of peremptory challenge); Washington v. Recuenco, 548 U.S. 212 (2006) (improper failure to submit a sentencing factor to the jury); Arizona v. Fulminante, 499 U.S. 279, 306--07 (1991) (collecting many other examples). Allowing lawyers to be too argumentative is a problem of trial management, and it is subject to the usual doctrines of harmless error and plain error.
Yakobowicz is inapplicable to non-argumentative summaries, is mistaken in concluding that criminal trials differ categorically from civil trials with respect to mid-trial summaries (both kinds of trials permit for the exercise of wise discretion by district judges in jury management), and has been overtaken by the Supreme Court's decision in Marcus. It does not assist McGee. The district judge did not abuse his discretion in allowing one short non-argumentative summary as the trial resumed after a weekend break.
One last issue: restitution. Restitution must be awarded with respect to all crimes of which a person is convicted, but it may not be awarded with respect to other losses ("relevant conduct" in the Sentencing Guidelines' parlance) unless the defendant consents to this additional award. See 18 U.S.C. §3663A(a)(2), (3); Hughey v. United States, 495 U.S. 411 (1990); United States v. Peterson, 268 F.3d 533, 534 (7th Cir. 2001); United States v. Webber, 536 F.3d 584, 601--02 (7th Cir. 2008). McGee contends that the award of restitution is defective because it includes not only the losses attributable to the nine counts of conviction but also all other payments to McGee that were established by evidence at trial and included as relevant conduct in the calculation of "loss" under U.S.S.G. §§ 1B1.3 and 2B1.1.
Once again the United States does not defend the district court's handling of this subject. Instead it contends that McGee did not make an adequate appellate argument, because his brief does not cite §3663A(a) or contain his own estimate of the appropriate award. Neither step is necessary, however. It does not take a long argument to point out an obvious gaffe. McGee's appellate brief cites Hughey; a pointer to dispositive authority is all that is required to highlight the problem. The United States also contends that McGee consented to the additional award, or at least forfeited any objection to it, when his lawyer told the district court during sentencing that "it seems like the restitution will follow the loss amount that the court ultimately finds here." The prosecutor reads too much into this concession. Restitution usually does "follow the loss amount" for the counts of conviction. (There are exceptions, because restitution is limited to what a victim could recover in a civil suit, while "loss" includes attempted or anticipated gains, see United States v. Behrman, 235 F.3d 1049 (7th Cir. 2000); United States v. George, 403 F.3d 470, 473--74 (7th Cir. 2005), but none of the differences is relevant here.) This does not imply that restitution follows the loss calculated for relevant conduct that did not lead to a conviction. We don't think that counsel meant to say otherwise; his ambiguous statement does not sound like the consent to pay additional restitution that is essential under Hughey and the statute.
The prosecutor simplified the trial by charging only a few discrete episodes of extortion. By the calculation in the presentence report, the victims' out-of-pocket loss on the counts of conviction was $18,450. Additional countable loss under the Guidelines came from relevant conduct and attempted extortion, which counts for the purpose of determining the severity of the sentence but not for restitution. The award must be reduced-though we do not know whether $18,450 is the right figure, as the district judge did not decide whether it is appropriate. Victims remain entitled to their entire loss and, if they think that judgments will be collectable, can file civil suits in which the criminal judgment against McGee will have a preclusive effect, simplifying the civil litigation.
The judgment of conviction and sentence are affirmed, except for the award of restitution, which is vacated. The case is remanded for the entry of an award of restitution limited to victims' out-of-pocket loss on the counts of conviction.