Quatloos! > Investment Fraud > HYIP & Bank Debentures > HYIP > U.S. v Polichemi

United States v. Polichemi

219 F.3d 698 (7th Cir. 07/05/2000)

U.S. Court of Appeals, Seventh Circuit

Nos. 96-3866, 96-3867, 96-3868 & 96-3869

July 05, 2000

UNITED STATES OF AMERICA, PLAINTIFF-APPELLEE

v.

JOSEPH POLICHEMI, ET AL., DEFENDANTS-APPELLANTS

Before Flaum, Rovner, and Diane P. Wood, Circuit Judges.

The opinion of the court was delivered by: Diane P. Wood, Circuit Judge.

On Petition for Rehearing

This case arose out of the government's prosecution of an elaborate financial scam, perpetrated in large part by the four defendants whose cases we consider on the government's petition for rehearing. In our original opinion, United States v. Polichemi, 201 F.3d 858 (7th Cir. 2000), we concluded that the convictions of Joseph Polichemi, Lyle "Pete" Neal, Oscar William Olson, and Charles Padilla on various counts of wire fraud, money laundering, conspiracy, and perjury, had to be overturned. Our reason was that the trial court erroneously failed to grant the defendants' motion to dismiss a certain juror for cause. This in turn forced the defendants to use a peremptory challenge to eliminate her from the jury; they ran out of peremptory challenges; and the court refused to give them any more. Six days after our decision was handed down, however, the Supreme Court decided United States v. Martinez-Salazar, 120 S.Ct. 774 (2000), which rejected the theory on which we had relied. In due course, the government filed a petition for rehearing and the defendants responded. The panel hereby grants the government's petition, vacates Parts I-III of its earlier opinion, and substitutes the following opinion in their place. (We note, however, that the petition for rehearing does not affect case No. 96-3870, the appeal of Larry P. Oesterman. We affirmed Oesterman's sentence in Part IV of our original opinion, which remains undisturbed.) Finding no reversible error other than the juror problem that is governed by Martinez-Salazar, we affirm all four convictions and sentences with two qualifications noted below.

I.

We review briefly the general facts relating to the financial frauds at issue here; to the extent that additional details are relevant to particular claims, we mention them there. Polichemi, Neal, Olson, Padilla, and others were the creators of a scheme to market so-called "prime bank instruments" to unsuspecting victim-investors. In fact, the instruments were phony from top to bottom. Nonetheless, the defendants talked a good game, and they described their "prime bank instruments" as multi-million dollar letters of credit that had been issued by the top 50 to 100 banks in the world. They offered these instruments to investors, telling them that they could purchase the paper at a discount and then resell it to other institutions at face value. The difference in price represented the profit to be earned. The trades, sadly, were fictional; there was no market for the trading of these (or any other) letters of credit, and nothing capable of generating profits ever occurred. None of the "investors" earned a cent, but the defendants for a time were living off the fat of the land. Polichemi, for example, wound up in a $6.2 million home in Florida as a result of the scheme; Olson bought a $4.4 million villa nearby; and Neal's luxury of choice was speed boats (he bought eight).

Between 1991 and 1994, the "prime bank instruments" yielded more than $15 million for the defendants. Their largest patsy was the Chicago Housing Authority, which (with the cooperation of the dishonest former director of employee benefits, one John Lauer, who also landed in prison for his part in the arrangement) turned over more than $13 million of its pension funds to the defendants. In all, some 30 investors were victimized. The defendants passed the monies they received through various bank accounts, used some to pay off prior investors and old debts, and spent the rest on themselves.

Each person had a particular role to play. Polichemi was the president of "Copol," a company that supposedly traded in the "prime bank instruments." He held himself out to be one of a handful of people in the world with a license to trade that kind of financial instrument. Neal was president of Konex Holding and Konex Marketing, companies that marketed Copol's "product" through a network of salespeople. Olson served as an attorney for both Copol and Polichemi, in addition to being a participant in many of the deals. Last, Padilla was Copol's "stateside banker." He served as a reference for the other defendants and reassured potential investors that Copol was a sound and successful company.

II.

In this section of the opinion, we discuss the many issues relating to the trial and convictions of the four defendants. Some points pertain to all of them, while others are individual in nature.

A. Juror Disqualification

We begin with the point that we are overruling as a result of Martinez-Salazar. During the jury selection process, prospective juror Lorena Nape came up as a potential member of the final jury. As a result of questioning, the parties learned that she was a 15-year employee of the U.S. Attorney's Office for the Northern District of Illinois--precisely the same office from which the prosecuting attorneys came. Based on her affiliation with the prosecutor's office, the defendants moved to strike her for cause. Even though Nape, in response to questions, stated that she could be fair and impartial, the defendants took the position that at a minimum she was excludable on the ground of implied bias. The district court denied their motion, and they then used one of their peremptory challenges to remove her from the jury.

Two different questions are presented by this scenario: the first is whether Nape should have been excluded for cause, and the second is whether an error in that ruling deprives the defendants of any rule-based or constitutional right, if the jury that actually sat was an impartial one. We first address the question whether the district court erred in refusing to strike Nape, because if its ruling was correct, we would have no need to address the Martinez-Salazar issue.

In spite of the government's arguments to the contrary in its petition for rehearing, we continue to be of the view that the implied bias doctrine applies in the particular circumstances of this case, and that it required the disqualification of prospective juror Nape. If she had sat on the final jury, this appeal would be quite different, and we note that Martinez-Salazar rejected the contention that "a defendant is obliged to use a peremptory challenge to cure the judge's error." 120 S.Ct. at 777. But here, Nape did not sit, and we suspect that prudent defense counsel will continue to use peremptory challenges to protect their clients against potentially biased jurors, rather than gambling everything on their ability to show bias after-the-fact and to obtain a reversal of a conviction on this basis.

The concept of implied bias is well-established in the law. Many of the rules that require excusing a juror for cause are based on implied bias, rather than actual bias. For example, a court must excuse a juror for cause if the juror is related to one of the parties in the case, or if the juror has even a tiny financial interest in the case. See, e.g., United States v. Annigoni, 96 F.3d 1132, 1138 (9th Cir. 1996); Getter v. Wal-Mart Stores, 66 F.3d 1119, 1122 (10th Cir. 1995). Such a juror may well be objective in fact, but the relationship is so close that the law errs on the side of caution.

In its decision in United States v. Haynes, 398 F.2d 980, 984 (2d Cir. 1968), the Second Circuit traced the implied bias doctrine back to Chief Justice John Marshall's opinion in United States v. Burr, 25 Fed. Cas. 49 (No. 14692g) (C.C. Va. 1807), one of several opinions in the prosecution of Aaron Burr. There the Chief Justice addressed the ways in which the law strives to assure an impartial jury:

Why is it that the most distant relative of a party cannot serve upon his jury? Certainly the single circumstance of relationship, taken in itself, unconnected with its consequences, would furnish no objection. The real reason of the rule is, that the law suspects the relative of partiality; suspects his mind to be under a bias, which will prevent his fairly hearing and fairly deciding on the testimony which may be offered to him. The end to be obtained is an impartial jury; to secure this end, a man is prohibited from serving on it whose connexion with a party is such as to induce a suspicion of partiality. 22 Fed. Cas. at 50. The Second Circuit later reviewed different grounds on which jurors were excusable for presumptive bias under the common law: kinship, interest, former jury service in the same cause, or because the prospective juror was a master, servant, counselor, steward, or of the same society or corporation. United States v. Haynes, 398 F.2d at 984.

We agree with the United States that government employment alone is not, and should not be, enough to trigger the rule under which an employee is disqualified from serving as a juror in a case involving her employer. But one need not adopt such a broad rule to find a problem in this case. Here, Nape was a long-time employee of the very U.S. Attorney's Office that was conducting the prosecution.

The Supreme Court had no such problem before it in United States v. Wood, 299 U.S. 123 (1936), or in Dennis v. United States, 339 U.S. 162 (1950), on which the government relies. Wood rejected the sweeping proposition that no government employee of any kind, and no recipient of government largesse such as a pension, could sit in any criminal case. Dennis raised a similarly broad challenge to all government employees as jurors, because the defendant there had been charged with failing to comply with a subpoena issued by the House Committee on Un-American Activities. The defendant's theory, which was rejected by the Court, was that the loyalty oath all government employees were required to sign automatically disqualified them from jury service. Finally, this case is unlike Smith v. Phillips, 455 U.S. 209 (1982), also cited by the government, in which a juror submitted a job application to the prosecuting attorney's office during a murder trial. The Court there rejected the defendant's effort to obtain habeas corpus relief on imputed bias grounds. But, of course, that juror was not an employee of the office; had no actual or perceived access to confidential information within the office; and had done little more than demonstrate an interest in the office.

A 15-year employee inside the prosecutor's office is in a materially different position. We note as well that nothing in 28 U.S.C. sec. 1866(c) mandates the sitting of biased jurors; that statute simply forbids disqualification of individuals or classes of individuals on other grounds. Although the case is not before us, we freely acknowledge that a proceeding brought by the Securities and Exchange Commission in Chicago, or by the Department of Housing and Urban Development (to the extent it was not using U.S. Attorney's office personnel) would not present the problem we have here.

We reaffirm our conclusion, therefore, that the district court should have excused Nape when the defendants moved to strike her for cause. Nevertheless, just as in United States v. Patterson, Nos. 97-3132 et al., 2000 WL 706020 (7th Cir., June 1, 2000), we find that this was not the kind of error that calls into question the impartiality of the jury ultimately selected. Id. at *2. It is not like the situation this court faced in United States v. Underwood, 122 F.3d 389 (7th Cir. 1997), where the entire process of jury selection was infected with ambiguity. As we recognized in Patterson, "[i]n any given situation there remains the possibility that a blunder affects a right that is substantial in the sense of Kotteakos v. United States, 328 U.S. 750 (1946): that it 'had substantial and injurious effect or influence in determining the jury's verdict.'" 2000 WL 706020 at *5. This, however, is a far cry from the automatic reversal rule of Swain v. Alabama, 380 U.S. 202 (1965), which Martinez-Salazar overruled. It is, in effect, a way in which someone challenging the jury selection process might show that an error (or set of errors) were not harmless.

Polichemi points out in his brief responding to the government's petition for rehearing that there is a factual difference between Martinez-Salazar's case and his own that might have been important, which Justice Souter highlighted in his concurring opinion. Martinez-Salazar and his co-defendant exhausted all of their peremptory challenges, but they failed to request any more (as they were entitled to do under Fed. R. Crim. P. 24(b)), and at the close of jury selection Martinez-Salazar's lawyer affirmatively told the district judge that he had no objections to the final list of jurors to be seated. Martinez-Salazar, 120 S.Ct. at 778. Polichemi and his co-defendants, in contrast, preserved their objections to the final jury and did request additional peremptories. But in the end, Martinez-Salazar's handling of his jury selection process was not what caused the majority of the Court to rule as it did. First, the Court noted that Martinez-Salazar received all the peremptory challenges to which he was entitled under Rule 24(b). Second, the Court stressed the fact that the ultimate jury was impartial. Third, just as in our case, there was no allegation that "the trial court deliberately misapplied the law in order to force the defendants to use a peremptory challenge to correct the court's error," or that the district court's ruling "result[ed] in the seating of any juror who should have been dismissed for cause." 120 S.Ct. at 782.

Perhaps a majority of the Supreme Court will some day accept the distinction between curative and non-curative uses of peremptory challenges offered by Justice Souter, but in our view it did not do so in Martinez-Salazar. Just as in that case, the defendants here opted to use one of their peremptories to strike a juror who should have been eliminated for cause. That was a choice they were free to make, but the fact that it had its basis in an error by the district court does not amount to a violation under either Rule 24(b) or the Due Process Clause. We thus reject this argument for reversal, made by all four defendants, and move on to the remainder of their challenges to their convictions.

B. Sufficiency of the Evidence: Single Scheme

The jury convicted Polichemi, Neal, and Padilla of engaging in a single wire fraud scheme, in violation of 18 U.S.C. sec. 1343, and it convicted Polichemi and Neal of participating in a single money laundering conspiracy, in violation of 18 U.S.C. sec.sec. 1956 and 371. (It acquitted Olson on these charges.) The defendants argue that their convictions should be overturned because there was a prejudicial variance between the indictment and the proof offered at trial. The indictment charged a single scheme or conspiracy, but, they claim, the evidence showed multiple schemes. See Kotteakos, 328 U.S. 750.

The defendants concede that the government produced enough evidence to allow the jury to find that there was a single scheme or conspiracy with respect to the CHA, and that separately it produced enough evidence to show a scheme or conspiracy with respect to the so-called Truckstop and Deanthorpe transactions (referring to two of the overt acts charged). They urge, however, that those transactions were distinct from one another, and that each was distinct from other deals (i.e. the Sovran, Glavinovitch, Reynolds, and Medema transactions). The same people did not participate in each "subscheme," for example. Neal did not even meet Polichemi and Padilla until 1992. The results of the earlier transactions were not used in later schemes or mentioned to later investors, and the transactions differed in their factual details.

We have held more than once that "[a] conspiracy variance claim amounts to a challenge to the sufficiency of the evidence supporting the jury's finding that each defendant was a member of the same conspiracy." United States v. Townsend, 924 F.2d 1385, 1389 (7th Cir. 1991) (emphasis added); see also, e.g., United States v. Magana, 118 F.3d 1173, 1185-86 (7th Cir. 1997); United States v. Whitt, 211 F.3d 1022, 1027 (7th Cir. 2000). Here, there was ample evidence to support the jury's finding that the defendants joined together with the common design and purpose to defraud investors through the sale of false financial instruments and then to launder the proceeds of the fraud. This common scheme extended from the 1991-92 transactions through the CHA deal. The jury could have found that the following common elements, all of which the government amply proved, demonstrated the existence of a single conspiracy: (1) in each case, there was an attempt to sell "prime bank instruments" through Polichemi and his company, Copol; (2) the defendants played set roles in each transaction--Polichemi the trader, Padilla the banker, and Neal the marketer (starting in 1992); (3) each time, the defendants laundered the proceeds quickly and frequently, through bank accounts in Europe and the United States. The fact that each defendant's relative role may have varied somewhat from transaction to transaction is not enough to undermine the jury's conclusion.

C. Sufficiency of the Evidence: Money Laundering

At issue here are convictions under three different money laundering provisions: 18 U.S.C. sec. 1956(a)(1)(A)(i) (knowingly conducting a financial transaction involving the proceeds of specified unlawful activity "with the intent to promote the carrying on of specified unlawful activity"); 18 U.S.C. sec. 1956(a)(2)(B)(i) (transporting or transferring funds from within the United States to a place outside the country, knowing that the funds represent the proceeds of unlawful activity and knowing that the transportation is designed to conceal source, ownership, etc.); and 18 U.S.C. sec. 1957 (engaging in monetary transactions in property derived from specified unlawful activity). In reviewing a sufficiency of the evidence challenge, this court views the evidence in the light most favorable to the government and will reverse only if there is no evidence from which the jury could find guilt beyond a reasonable doubt. E.g., United States v. Brown, 71 F.3d 1352, 1354 (7th Cir. 1995); United States v. Jocic, 207 F.3d 889, 892 (7th Cir. 2000).

Before addressing the particular claims here, we note that the government has conceded that Count 18 of the indictment, under which both Polichemi and Neal were convicted, and which dealt with a wire transfer of $850,000, was legally insufficient. Count 18 charged that an April 26, 1993 wire transfer in that amount from an account at Oak Trust Bank to an account held by one of the participants in a New Jersey bank was part of the fraud scheme. The government concedes in its brief, page 68 n.10, that this transfer did not involve "proceeds" of the scheme, as this court has defined the term for purposes of 18 U.S.C. sec. 1956. See United States v. Mankarious, 151 F.3d 694, 705 (7th Cir. 1998). After looking at the record, we accept this concession and vacate Polichemi's and Neal's convictions on Count 18. This has no effect on their sentences, however, because the total amount of the money laundered during the conspiracy still exceeds $10 million even without this $850,000.

Polichemi and Neal were convicted under Counts 17 and 19 of money laundering in violation of sec. 1956(a)(1)(A)(i). In order to prove that violation, the government had to show (among other things) that they conducted the financial transaction knowing that it involved the proceeds of "specified unlawful activity" and "with the intent to promote the carrying on of specified unlawful activity." The evidence supporting these counts shows that Polichemi and Neal transferred CHA funds (obtained as a result of specified unlawful activity--the fraud) to Ray Starkey, a potential investor with whom they were working on a deal, and to the CHA's John Lauer, as payment to him for "future administrative fees" (i.e. to make sure that the unlawful activity could continue). A reasonable jury could conclude, as this one did, that these transfers met the statutory requirements.

The same two defendants, Polichemi and Neal, were also convicted of money laundering under Counts 20-22, which charged violations of sec. 1956(a)(2)(B)(i). As noted above, that statute prohibits moving illegal funds out of the country, knowing that the transportation will help disguise the money. The evidence showed that Polichemi and Neal transferred CHA funds from the United States to a Copol bank account in Switzerland, and then from the Swiss bank to Olson and Lauer's accounts back in the United States. The government also introduced evidence that transactions are difficult to investigate once funds are moved through foreign bank accounts, and that Polichemi and Neal lied to Lauer about the whereabouts of the CHA funds (once again showing that there is no honor among thieves). Even though the transfers were made in the Copol name, a reasonable jury could have concluded that the two schemers intended to disguise their source.

Although both Polichemi and Olson were convicted of money laundering under sec. 1957 (Counts 23-26), only Olson is appealing his conviction on those counts. The evidence against him showed that there were two payments of $25,000 each by Copol into Olson's bank account, and one $25,000 payment by Olson to Edward Bergmann (a business associate to whom Olson owed money). Olson (the lawyer, recall) argues that the Copol payments were simply for legal services, and that the government failed to prove that Olson knew the monies were the proceeds from the illegal scheme.

Olson makes a superficially persuasive argument when he points out that the jury acquitted him on all charges pertaining to the scheme to defraud and conspiracy to launder money, and it convicted him only on three substantive counts of money laundering. Nevertheless, even if these were inconsistent verdicts, that is not enough for us to conclude that the jury lacked sufficient evidence to support the counts on which it did convict. See United States v. Powell, 469 U.S. 57, 64 (1984); United States v. Sims, 144 F.3d 1082, 1084 (7th Cir. 1998). While the evidence may not have been overwhelming against Olson on the sec. 1957 charges, there was enough to meet the generous standard of review that applies here. Olson was a signatory on the Swiss bank account where the CHA proceeds were held. At the time the transfers took place, Polichemi and Olson had been working together on the fraud scheme for years. The government notes in particular the evidence pertaining to the attempted, but unsuccessful, Sovran Bank swindle in 1991, in which Olson, Polichemi, and Padilla were involved. The government also points to other evidence of Olson's active fraud--evidence that the jury evidently did not find persuasive enough to convict on some charges, but which it was entitled to consider for the sec. 1957 charges. We reject Olson's invitation to second-guess the jury on this point.

D. Olson: Adequacy of Money Laundering Jury Instructions

Olson is the only defendant who has attacked the district court's instructions on the money laundering counts under which he was convicted (Counts 24, 25, and 26, all of which charged violations of sec. 1957). Olson alleges that those instructions erroneously told the jury that the government did not need to show that Olson knew the funds transferred to him were criminally derived. He also complains that the court should not have given an "ostrich" instruction with respect to him.

As a preliminary matter, we note an odd dispute over the standard of review that applies here. The government argues that it should be for abuse of discretion, citing United States v. Neville, 82 F.3d 750, 759 (7th Cir. 1996), while Olson has suggested the more deferential standard of plain error applies, because he failed to object to the instructions at trial. See United States v. Olano, 507 U.S. 725, 732 (1993).

Olson is probably right, but he has lost little as a practical matter, because we see no problem with the instructions under either standard of review. The jury was told that the government must show that "the defendant knew that the money involved in the financial transaction represented proceeds from some form, though not necessarily which form, of activity which constitutes a felony offense under state or federal law." While there may have been smoother ways of phrasing this, we think it was enough to convey the key idea to the jurors that knowledge that the funds were criminally derived was an element of the offense. Similarly, given the fact that Olson defended himself by claiming lack of guilty knowledge, and the fact that the government had rebutted this defense with evidence of Olson's long-standing participation in Copol and had suggested that at best he was deliberately indifferent, the district court appropriately gave an ostrich instruction. See United States v. Graffia, 120 F.3d 706, 713 (7th Cir. 1997).

E. Olson: Admission of Evidence--Bergmann and Merchants' National Bank Transactions

Olson is also the only person raising this point. In 1990, Edward Bergmann gave Olson $100,000 to invest through a trading account that Olson had established at Merchants' National Bank. Olson's proposed investments for Bergmann were similar to the "prime bank instruments" that formed the basis of the indictment. Shortly thereafter, Merchants' Bank closed Olson's account, when it found itself unable to complete several transactions he had orchestrated. In 1993, Olson repaid Bergmann his investment money using CHA funds and a check drawn on a Copol bank account. That repayment formed the basis for Count 23 of the indictment, a money laundering charge on which Olson was acquitted. Olson later (in 1994) paid $25,000 to Bergmann; this payment gave rise to Count 26, on which Olson was convicted.

The government wanted to admit evidence about the 1990 Merchants' Bank/Bergmann episode under Fed. R. Evid. 404(b), as relevant to Olson's knowledge and intent concerning the investment fraud. The district court ruled that it could admit evidence of Olson's dealings with Bergmann only insofar as it related to the 1994 transaction charged in the indictment. Nevertheless, at trial Bergmann testified for the government both about his 1990 investment with Olson and the 1993 and 1994 repayments. Olson was permitted to testify that the Bergmann investment was associated with an account he had at the bank for purposes of trading capital market instruments, and that the bank decided to discontinue trading with that account, but the court did not permit him to answer when asked if the bank attempted to conduct any trades.

These kinds of decisions about where to draw the line in the admission or exclusion of evidence are matters for the district court's discretion. See United States v. Poole, 207 F.3d 893, 897 (7th Cir. 2000). Here, the government also argues that Olson waived his right to challenge the exclusion of the Merchants' Bank evidence, because he had filed a motion in limine trying to exclude everything (which the court had denied). We are reluctant to rest our decision on waiver, because not exploring something at all is quite a different matter from exploring it in part. Nevertheless, we see no abuse of discretion in the court's rulings. It was entitled to find that any probative value of the additional evidence about Merchants' Bank that Olson wanted to introduce would be outweighed by its prejudicial effect, and moreover would be confusing to the jury.

F. Neal: Admission of Evidence on Prior Fraud Conviction

Over defense objections, the district court allowed the government to introduce evidence of Neal's 1980 guilty plea in a Virginia federal court to charges of investor fraud and aiding and abetting in the preparation of false tax returns pertaining to fraudulent investments. It is not clear from the transcript whether the court found the evidence admissible under Fed. R. Evid. 404(b) or directly admissible as "intricately related" to the acts charged in the indictment. See United States v. Spaeni, 60 F.3d 313, 316 (7th Cir. 1995).

In the end, the theory does not matter, because we find no abuse of discretion in the district court's decision either way. The court did not allow evidence of the conviction itself to be introduced; it allowed only evidence of the underlying conduct. It explained that the Virginia fraud was intricately related to the charged conduct because Neal had lied to Lauer (of the CHA) about the nature of the earlier conviction and Lauer relied on these lies in deciding to invest the CHA funds. The defendants also misrepresented Neal's past "successes" in the materials they gave prospective investors. The evidence is also weak support for Neal's intent to defraud, see Rule 404(b), even though the Virginia conduct occurred more than 10 years prior to the charged conduct and thus (without the intricate relation to the later activity) is at the outer edges of the requirement for Rule 404(b) that the prior crime be close enough in time to be relevant. See, e.g., United States v. Kreiser, 15 F.3d 635, 640 (7th Cir. 1994) (finding seven year time gap to be close enough to allow admission under Rule 404(b)); see also United States v. Wimberly, 60 F.3d 281, 285 (7th Cir. 1995) (finding 13 year time gap close enough where prior evidence was almost identical to current charged crime).

G. Padilla: Sufficiency of Evidence for Perjury Conviction

Padilla was convicted of three counts of perjury arising from statements he made to SEC investigators during a 1994 deposition. "To sustain a perjury conviction, the government must prove that, while under oath, the defendant knowingly made a statement that was both false and material." United States v. Gulley, 992 F.2d 108, 112 (7th Cir. 1993). Padilla makes the familiar argument that his statements were literally true, even if misleading, and that they were in any event not material. See Bronston v. United States, 409 U.S. 352 (1973) (recognizing that it is not a violation of the perjury statute to give an answer that is literally true, even if it is non-responsive or misleading); Kungys v. United States, 485 U.S. 759, 770 (1988) (defining materiality as something having "a natural tendency to influence, or capable of influencing, the decision of the decision-making body to which it was addressed").

There was sufficient evidence to support the jury's conclusions about truth and materiality on each of the three counts. For example, on Count 27 Padilla was asked why he left Nationsbank, and he replied that it was because he did not want to move to Richmond, Virginia, after Nationsbank merged with C&S Sovran. That was a lie. Sovran fired Padilla when it discovered his involvement with a fraudulent discounted letter of credit transaction. He also lied about where he went after he left Nationsbank, telling the SEC investigator that it was to Starbank, when it really was to Copol. Furthermore, these lies were material to the SEC, which was trying to find out who had worked for Copol and whether Padilla had engaged in any of the discounted letter of credit transactions.

Without recounting the evidence here for Counts 28 and 29, it is enough to say that we have reviewed the record and are satisfied that the jury's verdict was supported for these two counts as well. Were we to accept Padilla's view, the SEC investigators would literally have needed to know ahead of time the answer to each question they asked in order to be specific enough. That is not the law, nor is his somewhat odd idea that if his answers were not material to questions designed to elicit material information, that element has not been shown.

H. Polichemi: Juror Coercion

After the jury reported that it had reached a verdict, the district court polled the jurors to ensure that the verdict was unanimous. During the polling, juror Thomasina Jackson said that she did "not totally" agree with the verdict. The court immediately reminded the jurors that the verdict had to be unanimous and it instructed them to return for further deliberation. Not long thereafter, the jury returned again with a unanimous verdict. When polled again, juror Jackson answered "yes" when asked if the jury's final verdict was her own verdict.

Polichemi argues that the district court should have conducted a separate inquiry of Jackson to make sure she had not been coerced into the verdict. As evidence that she might not have freely assented to the verdict, he points to the facts that the jury delivered its first verdict at 4:30 p.m. on a Friday afternoon, that the jurors were probably eager to finish up before the weekend, that one juror had a vacation scheduled to begin five days later, and that the trial had lasted weeks longer than the jurors had originally been advised. Polichemi did not raise these objections at trial, however, and thus we would overturn the district court's decision not to conduct a separate interrogation only if plain error were present. None of the factors to which Polichemi points, however, suggest that the result of the trial was fundamentally unfair or that the judge should have disregarded Jackson's affirmation that the final verdict was indeed her own.

I. Polichemi: Mistrial Based on Exhibit Sent to Jury

During its examination of fraud victim Sylvia Field, the government introduced Exhibit 77, which was a letter to Olson from Field in which Field demanded the return of the money she had invested with Copol. At the request of defense counsel, the district court ruled that the following two sentences had to be redacted from the letter: "The Securities and Exchange Commission agent visited me and said Copol has been served with a suit against it for fraud. I was horrified by their revelations." Unfortunately, when the exhibit books were prepared for the jury's use, an unredacted copy of that letter was used and thus, for a time, the jury had the complete letter with it in the jury room.

Two days after the jury began deliberating, the prosecutors discovered the error. They immediately notified defense counsel and the court, and defense counsel moved for a mistrial. The court denied the motion, observing that the two sentences were "trivial in relation to the evidence in this case." At the same time, the court ordered that the unredacted copies should be removed and replaced with the redacted copies, which was done. We see no reversible error in this course of events. The district court was in the best position to assess whether any prejudice was likely from that slip, and as the court pointed out, there was ample evidence properly before the jury that revealed the existence of the SEC's fraud investigation. Additionally, there is no evidence indicating that the use of the wrong version of Exhibit 77 was anything but inadvertent.

J. Polichemi: Double Jeopardy

The last conviction issue any of these defendants raises is Polichemi's complaint that the prosecution against him was barred under the Double Jeopardy Clause of the Constitution by two earlier actions: (1) a 1994 civil action that the SEC brought against the defendants in federal court in Chicago, and (2) a 1996 civil action that the CHA brought against the defendants in a Florida state court. This is a frivolous argument that requires little discussion. The Double Jeopardy Clause is not implicated at all, because both prior actions were civil proceedings. See Hudson v. United States, 522 U.S. 93, 99 (1997) (the clause protects only against imposition of multiple criminal punishments for the same offense). The SEC received only injunctive relief, and so by no stretch of the imagination could its proceedings be thought to resemble a criminal prosecution, and the CHA is not even a federal prosecutorial agency. As far as the Double Jeopardy Clause is concerned, the government was entirely free to prosecute Polichemi for his crimes.

III.

Individually, the four defendants have also challenged the sentences they received on varying grounds. Those sentences were as follows:

(1) Polichemi received a term of 210 months on each of counts 17 through 22, a term of 60 months on each of counts 1 through 16, and a term of 120 months on each of counts 23 and 24, all to be served concurrently, along with a term of supervised release of two years and a $10,000,000 restitution order;

(2) Neal received a term of 235 months on each of counts 18, 19, 21, and 22, and a term of 60 months on each of counts 1-13, 16, 31, 32, and 33, all to run concurrently with one another, a two-year supervised release term, and the same $10,000,000 restitution order;

(3) Olson was sentenced to concurrent 120 month terms on counts 24 and 25, and also a consecutive one month sentence on count 26, a two-year term of supervised release, $10,000,000 in restitution, and an order by the court directing the Clerk of the Court to send a copy of the judgment of conviction to the Illinois Attorney Registration and Disciplinary Committee and to the Executive Committee of the Northern District of Illinois; and

(4) Padilla was sentenced to a term of 78 months, which consisted of concurrent 60 month terms on each of counts 1-13, 18 months each on counts 27, 28, and 29, which were concurrent with one another but consecutive to the 60-month terms, two years' supervised release, and $10,000,000 in restitution.

We take their sentencing claims in the order in which we have listed them.

A. Polichemi

Polichemi raises three arguments in an effort to reduce his sentence: (1) the court should have given him the downward departure for a minor or minimal participant provided by U.S.S.G. sec. 3B1.2; (2) the court erred in its calculation of the total amount of money lost because of the scheme; and (3) the court should have found him eligible for the safety valve provided in U.S.S.G. sec. 5C1.2. We find no merit in any of these points.

Polichemi initially argued that because the district court found that the total amount attributable to the money laundering conspiracy was between $10 and $20 million, the court should have imposed a nine-level increase over the base rather than ten. See U.S.S.G. sec. 2S1.1(b). But, as the government points out, that is what the court did. In his reply brief, he has also tried to argue that the evidence was insufficient to support the $10-20 million figure, but (a) he has waived that point, and (b) the evidence supports that range quite handily.

As for his claim that he should have received the minor participant adjustment offered by U.S.S.G. sec. 3B1.2, we find no clear error in the district court's handling of the matter. It is in fact quite impossible to think of Polichemi, the president of Copol and purported master trader of prime bank instruments, as "substantially less culpable than the average participant" in the scheme. See United States v. Stephenson, 53 F.3d 836, 850 (7th Cir. 1995). Finally, the safety valve provision on which he is trying to rely, sec. 5C1.2, applies only to convictions under Title 21 (drug offenses). This indictment charged no such thing, which is the end of the matter.

B. Neal

Neal argues only that the district court should not have increased his sentence under U.S.S.G. sec. 3B1.1(b), which calls for an increase of three levels for someone who was a "manager or supervisor" of a criminal activity that involved five or more persons. Our review is once again for clear error, and we find none. To qualify for the increase, the district court had to find that Neal had control over at least one participant in the criminal activity. United States v. Fones, 51 F.3d 663, 668 (7th Cir. 1995). Although the court did not find this fact explicitly, its discussion of Neal's relationship to Edward Russey and Larry Oesterman indicates that it found the necessary supervision. Neal was president of Konex Marketing, and Russey and Oesterman were salesmen for the company. As such, Neal was their boss, not their equal.

C. Olson

Four sentencing issues appear in Olson's separate brief: (1) the relevant conduct determination that more than $10 million was involved, when Olson himself was convicted only on three counts involving a total of $75,000; (2) the two-level enhancement he received under U.S.S.G. sec. 3C1.1 for obstruction of justice (by committing perjury); (3) the decision to increase his offense level by two for his leadership role, under U.S.S.G. sec. 3B1.1; and (4) the decision to hold him responsible for the full $10 million in restitution.

Olson plainly considers it unfair that he should be held responsible, under the relevant conduct provisions, for money laundering and fraud covered in the counts on which he was acquitted. But that fact did not prevent the district court from making its own assessment of the proper sentence, see United States v. Watts, 519 U.S. 148, 157 (1997), and taking into account for sentencing purposes the conduct that underlay the other charges. The district court recognized this after it had initially denied the nine-level increase, when it found that Olson "had a hand in" the more than $10 million that was laundered through the various transactions.

Under the provisions that relate to money laundering, relevant conduct includes all acts or omissions, including jointly undertaken criminal endeavors, "that were part of the same course of conduct or common scheme or plan as the offense of conviction." U.S.S.G. sec. 1B1.3(a)(2). This is the part of the relevant conduct guideline that applies to money laundering, because money laundering is an offense for which sec. 3D1.2(d) requires grouping of multiple counts; indeed, sec. 3D1.2(d) specifically refers to sec.sec. 2S1.1 and 2S1.2, which are the guidelines for money laundering offenses under 18 U.S.C. sec.sec. 1956 and 1957. (Olson is incorrect when he asserts that sec. 1B1.3(a)(1)(B) governs here, under which we would restrict our consideration to activities that occurred "during the commission of the offense of conviction," or in preparation for that offense or to avoid detection or responsibility for that offense. Compare United States v. Gabel, 85 F.3d 1217 (7th Cir. 1996).)

The question boils down to whether the district court found that Olson participated in the CHA transaction underlying his money laundering conviction, and if so, whether that finding was supported by a preponderance of the evidence. Our review once again is for clear error only. The district court certainly made the necessary finding, and our review of the record and the court's comments indicates that there was an adequate evidentiary basis for it.

Olson's next argument is that he should not have received two extra levels under the aggravating role enhancement provided in sec. 3B1.1 for an "organizer, leader, manager, or supervisor." To qualify, the defendant must have been the organizer, leader, etc., of one or more of the participants. U.S.S.G. sec. 3B1.1, application note 2. The district court found that Olson had control over at least one participant in the criminal activity. See id. application note 4. In response to defense counsel's question about whom exactly Olson was supposed to have controlled, the court named Padilla, Lauer, Oesterman, and Russey. The judge also commented that "[w]ithout Mr. Olson, I think Mr. Polichemi, with all due deference, wouldn't be able to write a letter."

Like many of the other sentencing issues we have examined, this one involves factual findings that we review for clear error. Even if we took the position that a lawyer typically does not control his client, and thus that the judge may have been mistaken to think that Olson was pulling the strings for Polichemi, and even if Olson, as he argues, never even met or corresponded with Russey, Oesterman, or Lauer, as the government argued in the alternative, the two-level increase Olson actually received (instead of the four-level increase that would have been possible under sec. 3B1.1(a)) can also be upheld on the basis that Olson "used a special skill, in a manner that significantly facilitated the commission or concealment of the offense." U.S.S.G. sec. 3B1.3. Although the district court did not address this specifically in its findings, we think the record is quite clear that Olson did use his skills as a lawyer to facilitate the commission of the offense. See id. application note 3. We therefore find no reversible error with respect to the two-level enhancement.

The district court increased Olson's offense level by two, relying on U.S.S.G. sec. 3C1.1, upon finding that Olson perjured himself when he "testified to the validity of the prime bank instruments." As Padilla argued with respect to his conviction on perjury, Olson asserts here that his statements were neither false nor material. He claims that he testified only that he believed the prime bank instruments to be valid, not that they were valid. But this statement takes his testimony out of context. Olson testified at length about the mechanics of such instruments, and overall made it clear that he was portraying them as legitimate investment vehicles. Furthermore, this perjury was material to the offense of conviction--laundering the proceeds of an illegal investment scheme. We find no clear error in the district court's decision to adjust Olson's sentence accordingly.

Last, Olson complains that the district court erred in holding him jointly and severally liable for the $10 million restitution payment designed to go to the victims of this group. The government concedes that restitution cannot be ordered that relates to conduct for which the defendant has been acquitted. See United States v. Kane, 944 F.2d 1406, 1415 (7th Cir. 1991). Because the $10 million order necessarily had this effect, we must remand the order of restitution that runs against Olson so that the district court can recalculate the amount he owes. In so doing, the court must follow the provisions of the Mandatory Victim Restitution Act of 1996, 18 U.S.C. sec. 3663A, because Olson was convicted of an offense of property as described in sec. 3663A(c)(1)(A)(ii). Furthermore, the order may not take into account Olson's own economic circumstances. See 18 U.S.C. sec. 3664(f)(1)(A). On the other hand, the burden is on the government to prove by a preponderance of the evidence the amount of the loss sustained "as a result of the offense." To the extent that Olson's financial resources and needs are pertinent to other issues, such as the rate of payment he must make, the burden will be on him to show those facts. See id. sec. 3664(e). See also United States v. Walton, Nos. 99-2638, 99-2640, 2000 WL 767891 *7 (7th Cir. June 14, 2000).

D. Padilla

Padilla argues that the district court should not have sentenced him to consecutive terms on his fraud and perjury convictions, when the perjury conduct was used to support an upward adjustment for obstruction of justice in the offense level and the court should have grouped the perjury conviction with the fraud conviction. See U.S.S.G. sec. 3D1.5 and illustration #3. The government responds that Padilla did not raise this objection during sentencing and thus our review can only be for plain error.

Padilla did not address the waiver point in his reply brief, and so we agree that plain error is the governing standard. In any event, the district court did not err. The wire fraud counts under which Padilla was convicted carry a five-year statutory maximum. See 18 U.S.C. sec. 1343. Padilla's sentencing range as computed by the guidelines, however, was 78 to 97 months. In circumstances in which the guidelines range exceeds the highest statutory maximum in a multiple-count conviction, the guidelines instruct that "the sentence imposed on one or more of the other counts shall run consecutively, but only to the extent necessary to produce a combined sentence equal to the total punishment. In all other respects, sentences on all counts shall run concurrently, except to the extent otherwise required by law." U.S.S.G. sec. 5G1.2(d). That is exactly what the district court did. It accepted the 60-month maximum on the wire fraud counts, and imposed an 18-month consecutive sentence on the other counts, which was just enough to reach the bottom of the guideline range that applied to Padilla. This is just what sec. 5G1.2(d) orders, and it could thus hardly be called error.

IV.

To summarize, we re-confirm our holding in Part IV of United States v. Polichemi, 201 F.3d 858 (7th Cir. 2000), in which we affirmed the sentence imposed upon Larry Oesterman. We affirm the convictions of defendants Polichemi, Neal, Olson, and Padilla, with the exception of the convictions against Polichemi and Neal under Count 18, which are hereby vacated. We affirm the sentences imposed on Polichemi, Neal, and Padilla in their entirety, and we affirm the sentence imposed upon Olson with the exception of the order of restitution. As to that, we order a limited remand for the purpose of the district court's recalculation of the amount of restitution he must pay.

______________________________________________

United States v. Polichemi, 201 F.3d 858 (7th Cir. 01/13/2000)

U.S. Court of Appeals, Seventh Circuit

Nos. 96-3866, 96-3867, 96-3868, 96-3869, 96-3870

January 13, 2000

UNITED STATES OF AMERICA, PLAINTIFF-APPELLEE,

v.

JOSEPH POLICHEMI, ET AL., DEFENDANTS-APPELLANTS.

Appeals from the United States District Court for the Northern District of Illinois, Eastern Division. No. 94 CR 555--William T. Hart, Judge.

Before Flaum, Rovner, and Diane P. Wood, Circuit Judges.

The opinion of the court was delivered by: Diane P. Wood, Circuit Judge.

Argued January 4, 1999

Decided January 13, 2000*fn1

Not everyone has a perfect understanding of the complex workings of modern financial markets, and unfortunately, sometimes unscrupulous individuals manage to exploit that fact for a period of time. This case, at its outset, involved seven such individuals, who allegedly engaged in a breathtakingly ambitious phony investment scheme through which they bilked nearly 30 investors out of more than $15 million. Their largest patsy was the Chicago Housing Authority, which lost more than a third of its pension fund, some $13 million, in the scheme. The CHA's former director of employee benefits, John Lauer, landed in prison for fraud and related offenses as a result of these dealings. See United States v. Lauer, 148 F.3d 766 (7th Cir. 1998).

The present appeal comes to us from the convictions after a jury trial of four of the defendants, Joseph Polichemi, Lyle "Pete" Neal, Oscar William Olson, and Charles Padilla, on assorted counts of wire fraud, money laundering, conspiracy, and perjury. A fifth, Larry Oesterman, who pleaded guilty, argues that his sentence is too harsh. Although taken together the four who went to trial have raised nearly a score of issues for our consideration, we conclude that the trial court's handling of the defendants' effort to strike a juror for cause was an error that fundamentally tainted the fairness of the trial. Under this circuit's decision in United States v. Underwood, 122 F.3d 389 (7th Cir. 1997), we have no choice but to reverse and remand for a new trial. This error was irrelevant to Oesterman, because of his guilty plea. Finding no error in the district court's sentencing decision, we affirm Oesterman's sentence.

I.

We need not delve into the details of the scheme, given the fact that they are not necessary for our analysis of the jury selection problem. Briefly, however, they are as follows. Polichemi, Neal, Olson, Padilla, and Oesterman, along with others not relevant here, devised a system under which they marketed so-called "prime bank instruments" to investors (i.e. victims). They described these "prime bank instruments" as multi-million-dollar letters of credit issued by the top 50 or 100 banks in the world. The defendants told their victims that they could purchase these instruments at a discount and then resell them to other institutions at face value; the difference in price represented the profits that would go to the defendants and their "investors." This was nothing more than a song and dance: the trades were fictional; there was no market for the trading of letters of credit; and nothing capable of generating profits ever occurred. Somehow, notwithstanding the implausibility of the "prime bank instruments" to one familiar with normal business practice for letters of credit, they managed to persuade their victims to give them money to finance the purchase of the phantom discounted instruments. While this did not earn a cent for any of the investors, it definitely changed the defendants' own lifestyles. Polichemi, for example, was living in his sister's 2-bedroom condominium in Florida when things began, but he ended up in a $6.2 million home. Olson's and Neal's stories were similar.

Between 1991 and 1994, the defendants collected more than $15 million in this way. As noted above, their largest source was the CHA, which turned over more than $13 million of its pension funds to the defendants, thanks in large part to the unfaithful Lauer. They passed the monies they received through various bank accounts (often Swiss), used some of the money to pay off prior investors and old debts, and spent the rest on themselves. Each person had his own role to play. Polichemi was the president of "Copol," a company that purportedly traded in the prime bank instruments. He held himself out to be one of a handful of people in the world with a license to trade these "securities." Neal was president of Konex Holding and Konex Marketing, companies that marketed Copol's product through a network of salespeople. Olson was an attorney for both Copol and Polichemi, in addition to being a participant in many of the deals at issue. Padilla was Copol's "stateside banker." He served as a reference for the other defendants and provided reassurance of Copol's soundness and success to potential investors. Oesterman was one of Neal's salespeople and a director at Konex marketing. These five were allegedly joined by Lauer, Edward Russey (another salesperson for Neal who pleaded guilty and testified pursuant to a plea agreement), and John DeVincens, a Konex attorney who was later acquitted on all charges by the jury.

II.

In the end, the final benefit the defendants reaped from their scheme was an indictment from the grand jury in the Northern District of Illinois. Count 1 of the indictment charged the entire group with a scheme to defraud in violation of 18 U.S.C. sec.sec. 1343 and 2. Counts 2 through 15 charged the individual defendants with engaging in wire transmissions in furtherance of the fraud. Count 16 accused Polichemi, Neal, Olson, and DeVincens of conspiring to launder money in violation of 18 U.S.C. sec. 371, and Counts 17-26 charged that individual defendants had committed specific money laundering offenses in violation of 18 U.S.C. sec.sec. 1956(a)(1)(A)(i), 1956(a)(2)(B)(i), 1957, and 2. Finally, Counts 27-33 involved perjury charges against Padilla (27-29), DeVincens (30), and Neal (31-33), in violation of 18 U.S.C. sec. 1621(1). As noted above, DeVincens was acquitted; Russey pleaded guilty and is not involved in this appeal; Oesterman pleaded guilty and raises only sentencing challenges on appeal; and Polichemi, Neal, Olson, and Padilla (to whom we refer as the Polichemi defendants for convenience) were each convicted on some or all of the charges brought against them and appeal both their convictions and sentences.

III.

At this court's direction, the Polichemi defendants filed one consolidated brief in which they addressed issues common to all four, and separate supplemental briefs on their individual issues. Oesterman, of course, filed his own brief. Because we find it dispositive of the appeals of the Polichemi defendants, we address first their complaint about the jury selection process. It centers on the district court's handling of their request to strike potential jurors Lorena Nape, John Buck, and David Maines for cause. Under the system the trial court adopted, the defendants were to have 10 peremptory challenges collectively, plus one additional challenge for each two jurors selected as alternatives. After sixteen potential jurors had been qualified, the judge would call for simultaneously submitted peremptory challenges and then empanel those who were not struck.

Jury selection began on May 15, 1996. A number of jurors were excused for cause during the morning session. Among the prospective jurors called after the lunch break were Nape, Buck, and Maines. Both Buck and Maines stated, in response to questions posed during the voir dire, that they would tend to credit testimony from a law enforcement officer more than testimony of a lay witness. Maines noted that his wife's side of the family included several police officers, but he indicated that he did not believe this would affect his ability to be fair and impartial. On the other hand, he said that he tended "to have a less scrutinizing point of view when it comes to government officials and police officers." Buck said that he would start out believing the testimony of a law enforcement agent more than the testimony of a lay witness. Faced with these damaging statements, the prosecutor elicited statements from both men that they could, in the final analysis, reach a verdict solely based on the evidence they heard in the courtroom. On that record, the court denied the defense motions to excuse Maines and Buck for cause, and the defendants used two of their remaining peremptory challenges to strike them from the panel.

Nape's situation was somewhat different. At the time of the trial, Nape was a 15-year employee of the U.S. Attorney's Office for the Northern District of Illinois, based in Chicago--precisely the same office from which the prosecuting attorneys came. At oral argument and in a letter filed after argument, the AUSAs (in response to questions posed by this court) indicated that the official who signed the indictment did not have supervisory authority over Nape, because she was a secretary in the Civil Division. Nonetheless, it also appears that the sharing of work from division to division within the U.S. Attorney's Office led to varying levels of communication between the employees of the Civil Division and of the Criminal Division. See Tr. 176-77 (when asked whether her work had involved any assignments from the Criminal Division, Nape responded "maybe in the asset forfeiture area, of course, that seems to be slowing down or going mostly criminal now."); Tr. 179 (Nape admitted to recognizing the names of the prosecuting attorneys in the case, and being aware that they worked in her office.). See generally Attorney General, Memorandum on Coordination of Parallel Criminal, Civil, and Administrative Proceedings (dated July 28, 1997), available at (calling for information-sharing among divisions during the investigation and prosecution of cases, in order for the government to allocate its resources efficiently and come to effective, comprehensive settlements).

Based on her affiliation with the prosecutor's office, the defendants moved to strike Nape for cause as well. Even though, in response to questions, Nape stated that she could be fair and impartial, the defendants took the position that at a minimum she was excludable for cause on the ground of implied bias. As before, the court denied the motion, and the defendants had to use a peremptory challenge to remove her from the jury.

The Polichemi defendants argue that the district court erred in all three of its rulings rejecting their challenges for cause, with respect to prospective jurors Buck, Maines, and Nape. These errors forced them to use peremptory challenges to eliminate the objectionable panel members, which in turn, they argue, violated their Fifth Amendment due process rights by impairing the intelligent exercise of their peremptory challenges. They rely on this court's decision in United States v. Underwood, 122 F.3d 389 (7th Cir. 1997), cert. denied sub nom. United States v. Messino, 118 S.Ct. 2341 (1988), to show that this error is a structural one requiring automatic reversal.

Our analysis of these claims will follow that basic order. First, we consider whether the district court erred in any of the three rulings on the challenges for cause. If not, then we would have no occasion to proceed further with this ground of the appeals. Second, assuming at least one of those decisions was in error (given the fact that defendants here exhausted all their peremptory challenges), we must decide whether the automatic reversal rule of Underwood applies in these circumstances, or if this was a lesser error subject to harmless error analysis, as in our more recent decision in United States v. Osigbade, 195 F.3d 900 (7th Cir. 1999). If the automatic reversal rule applies here, then there is nothing left to do with the Polichemi defendants but to reverse the convictions and remand for further proceedings; if not, then we would proceed to a harmless error analysis.

Before proceeding down this path, we acknowledge that the circuits have been split on the question whether the automatic reversal rule of Swain v. Alabama, 380 U.S. 202, 218 (1965), still applies to a situation in which trial court error with respect to challenges for cause forces a defendant to use up her peremptory challenges, thus effectively reducing the number of such challenges below that are permitted in Fed. R. Crim. P. 24(b). Courts following the automatic reversal rule include (in addition to our own, as reflected in Underwood), the Fifth Circuit, in United States v. Hall, 152 F.3d 381, 408 (5th Cir. 1998), quoting United States v. Nell, 526 F.2d 1223, 1229 (5th Cir. 1976), the Ninth Circuit, in United States v. Martinez-Salazar, 146 F.3d 653 (9th Cir. 1998), cert. granted, 119 S.Ct. 2365 (1999), the First Circuit, in United States v. Cambara, 902 F.2d 144, 147 (1st Cir. 1990), the Third Circuit, in United States v. Ruuska, 883 F.2d 262, 268 (3d Cir. 1989), the Fourth Circuit, in United States v. Ricks, 776 F.2d 455, 461 (4th Cir. 1985), and the Sixth Circuit, in United States v. Hill, 738 F.2d 152, 153-54 (6th Cir. 1984). The Eighth Circuit and the Tenth Circuit take the opposite approach, under which they regard errors in denying challenges for cause as reversible only if the eventual jury that sat was biased--in other words, only if the error was not harmless. See United States v. Sithithongtham, 192 F.3d 1119, 1123 (8th Cir. 1999); Getter v. Wal-Mart Stores, 66 F.3d 1119, 1122 (10th Cir. 1995); see also Ross v. Oklahoma, 487 U.S. 81, 88 (1988). As we explain in more detail below, this court has taken a middle ground, under which some such errors trigger the automatic reversal rule and others do not. The Supreme Court will be speaking soon on the subject, in Martinez-Salazar, but unless or until we learn that we have erred, we must follow our own circuit's precedents.

We can dispose of the challenges to prospective jurors Buck and Maines relatively easily. In both those cases, the individuals made some remarks during voir dire that indicated they would harbor an actual bias in favor of witnesses from the law enforcement community, but upon further questioning, they both promised to evaluate the evidence fairly and to decide the case based on what was presented to them. We review a trial court's decision whether to dismiss a juror for cause deferentially, recognizing that the judge on the scene is in the best position to evaluate the juror's ability to serve. United States v. Beasley, 48 F.3d 262, 266 (7th Cir. 1995); United States v. Casey, 835 F.2d 148, 151-52 (7th Cir. 1987). We cannot say on this record that the district court clearly erred when it concluded, after full questioning, that the defendants had failed to justify striking prospective jurors Buck and Maines for cause. Thus, we need not address the question whether the defendants' choice to use peremptory challenges for them infringed their due process rights. Peremptory challenges are most often used to eliminate from the jury those whom the defendant has no other way to remove. Since the district court committed no reversible error in these two rulings, no further comment is necessary.

The situation with prospective juror Nape is significantly different. As a long-time employee of the same U.S. Attorney's Office that was handling the prosecution, Nape could not avoid a public association with the lawyers for one side of the case. While she assured the trial judge that she believed she could be fair and impartial, such assurances cannot be the last word in these circumstances. Considering the deferential standard of review, we do not quarrel with the district court's conclusion that Nape did not consciously harbor ill will toward the defendants, or good will toward the prosecution. But the court erred in stopping its inquiries there. In addition to the kind of overt, actual bias that the court was looking for, a prospective juror might be excludable for implied bias.

While the term "implied bias" is not one that is used often, the concept is nonetheless well established in the law. Indeed, the risk of implied bias lies behind many of the rules that require excusing a juror for cause. A court must excuse a juror for cause if the juror is related to one of the parties in the case, or if the juror has a financial interest in the case. See, e.g., United States v. Annigoni, 96 F.3d 1132, 1138 (9th Cir. 1996); Getter, 61 F.3d at 1122. It is possible, of course, that a particular juror may be quite capable of maintaining her objectivity, even if her nephew is a lawyer in the case, but the relationship is so close that the law errs on the side of caution. Indeed, much the same rationale underlies the statute that requires disqualification of judges when the risk of at least the appearance of partiality or the risk of an unconscious tendency to favor one side is particularly great. See 28 U.S.C. sec. 455(b) (requiring disqualification when, for example, the judge or an immediate member of her family has a financial interest in the case, or when a close relative of the judge is a party or lawyer in the case). In many, if not most, such cases, the judge will in fact be unbiased and capable of rendering a fair decision, but the need to prevent even an appearance of partiality leads to a type of implied bias rule.

In its decision in United States v. Haynes, 398 F.2d 980, 984 (2d Cir. 1968), the Second Circuit traced the implied bias doctrine back to Chief Justice John Marshall's opinion in United States v. Burr, 25 Fed. Cas. 49 (No. 14692g) (C.C. Va. 1807), one of a series of opinions in the famous prosecution of Aaron Burr. In Burr the Chief Justice addressed the ways in which the law strives to assure an impartial jury:

Why is it that the most distant relative of a party cannot serve upon his jury? Certainly the single circumstance of relationship, taken in itself, unconnected with its consequences, would furnish no objection. The real reason of the rule is, that the law suspects the relative of partiality; suspects his mind to be under a bias, which will prevent his fairly hearing and fairly deciding on the testimony which may be offered to him. The end to be obtained is an impartial jury; to secure this end, a man is prohibited from serving on it whose connexion with a party is such as to induce a suspicion of partiality. The relationship may be remote; the person may never have seen the party; he may declare that he feels no prejudice in the case; and yet the law cautiously incapacitates him from serving on the jury because it suspects prejudice, because in general persons in a similar situation would feel prejudice.

25 Fed. Cas. at 50. From that time to the present, the federal courts have included among the reasons supporting a challenge for cause the kinds of presumptive sources of bias to which the Chief Justice referred. In Haynes, the Second Circuit wrote that

[a]t common law, jurors were challengeable on principle for bias for partiality due to kinship, interest, former jury service in the same cause, or because the prospective juror was a master, servant, counselor, steward, or of the same society or corporation. . . . Not only have these common law grounds for causal challenge retained their vitality, . . . but to them have been added others from which prejudice or bias may be implied.

398 F.2d at 984 (internal citations omitted). See also United States v. Torres, 128 F.3d 38, 43 (2d Cir. 1997) (recognizing three possible grounds for challenges for cause: those based on actual bias, those based on implied bias, and those based on "inferable" bias); United States v. Nell, 526 F.2d at 1229 (5th Cir. 1976); United States v. Dellinger, 472 F.2d 340, 367-69 (7th Cir. 1972) (noting that prospective jurors may be unacceptably biased for a variety of reasons, and, because they themselves are often unaware of these actual biases, it is essential to explore their backgrounds and attitudes in order to uncover them).

Here, prospective juror Nape was a "servant"--that is, employee--of the particular office representing the United States, that of the U.S. Attorney for the Northern District of Illinois. We offer no comment on the application of the implied bias concept to links more remote than the one she had, such as affiliation with a different agency of the United States in Chicago, or affiliation with a different part of the Department of Justice. In this case, unfortunately, the fit was perfect. Although the government has argued that she should be disqualified only if she was under the actual supervision of the officer signing the indictment, we think that fine-tunes matters too far. The U.S. Attorney has authority over the entire office and can manage its personnel as he sees fit. The court therefore erred when it denied defendants' motion to dismiss Nape for cause.

Given this error, what should be its consequence under our decisions in Underwood and Osigbade? In Underwood, the district court gave a confusing explanation of the system it was planning to use for jury selection. Under a misapprehension about the actual system, the defendants used their peremptory challenges on a number of prospective jurors whom they would not have chosen had they understood the court's plan. This court, noting that the specific right to peremptory challenges is found in Fed. R. Crim. P. 24(b), not the Constitution, held that "the 'right' to peremptory challenges is denied or impaired only if the defendant does not receive that which [statutory] law provides." 122 F.3d at 392 (internal quotations and citations omitted). Recognizing that not all restrictions on the right to peremptory challenge constitutes the denial or impairment of the right, the court held that harmless error analysis was nonetheless inappropriate when a fundamental denial or impairment had occurred.

In Underwood, the fact that the entire process for exercising peremptory challenges had been misunderstood supported a finding of a denial or impairment fundamental enough to justify application of the Swain automatic reversal rule. This was true even though there was no showing that the jury that actually sat was biased or could not be impartial. On the other hand, United States v. Osigbade presented a case of a clerical mistake that led to one juror's being excused and another being substituted in her place. 195 F.3d at 901-02. Using the "struck jury" selection system, both sides had questioned the entire venire, completed their challenges for cause, and then had submitted a list of peremptory challenges to the court. The court reviewed the peremptory challenges and then compiled the list of jurors, including alternates, for trial. After it entertained Batson challenges, it read off a list of the jurors who had been selected. At that point the parties alerted the court to the fact that one juror on that list had already been excused for cause. The court called the next name on the list. While waiting for that individual to return to the courtroom, the parties discovered that one prospective juror, Rhoda Richardson, had inadvertently been deleted from the list, because the court erroneously thought the government had struck her. Efforts to find her were unsuccessful, and the court eventually replaced her with one of the alternates.

These facts, we found, did not add up to such a basic impairment of the right to use peremptory challenges that the automatic reversal rule of Swain and, by then, Underwood, should be applied. As in Underwood, we recognized that some jury mistakes justify automatic reversal and others do not. Id. at 904. In Osigbade, automatic reversal was not required for several reasons. First, the defendant did not run out of peremptory challenges. Second, there was no evidence that the mistaken dismissal of the one juror resulted in anyone's being seated to whom the defendant would have objected (even with a peremptory challenge). Third, as the government argued in Osigbade, the essence of the defendant's complaint was that Richardson herself did not sit on the jury. Defendants have no legally cognizable right to have any particular juror participate in their case. See United States v. Duff, 76 F.3d 122, 125 (7th Cir. 1996).

While we understand that the distinctions here are fine, we have concluded that the district court's error in failing to excuse prospective juror Nape for cause goes to the fundamental integrity of the jury, and thus falls on the Underwood side of the line. Osigbade did not purport to overrule Underwood, nor could it have done so under this circuit's rules without undergoing circulation to the full court under the procedures established in Circuit Rule 40(e). Instead, it sensibly distinguished fundamental errors that undermine the integrity of the jury selection system from mere clerical errors that do not result in the seating of a single individual to which the defendant would have objected. (We know this to be the case, because the court used one of the previously submitted alternates in Osigbade after the problem became clear.) Here, not only did the court fail to consider Nape's implied bias, but its error resulted in the subversion of the defendants' ability to exercise their peremptory challenges effectively. See Osigbade, 195 F.3d at 904. Unlike in Osigbade, the defendants ran out of peremptory challenges; unlike in Osigbade, the attorneys specifically told the court that the error resulted in the seating of jurors to whom they would have objected. These are important differences. When, under these circumstances, the court commits the legal error of failing to apply the principle of implied bias in its administration of challenges for cause, the structure of the jury selection process itself is compromised and the Underwood rule applies.

We therefore conclude that the convictions of defendants Polichemi, Neal, Olson, and Padilla must be reversed and the case must be remanded for a new trial.

IV.

The only remaining issue we need consider, in light of our disposition of those four appeals, is Oesterman's challenge to his sentence. Oesterman pleaded guilty to one count of wire fraud, without any plea agreement. Although he was personally responsible for only $450,000, which he had obtained from victim investors, the district court relied on the relevant conduct provisions of the Sentencing Guidelines to hold him accountable for the entire $10-20 million loss. It found that he had joined the conspiracy and increased his offense level by 15, following U.S.S.G. sec. 2F1.1(b).

We review the district court's assessment of relevant conduct for clear error, including its determination of the questions whether the defendant participated in jointly undertaken criminal activity and whether the actions of the others were reasonably foreseeable to him. See United States v. Edwards, 115 F.3d 1322, 1325 (7th Cir. 1997); see also U.S.S.G. sec. 1B1.3 (1)(B). The record showed that Oesterman was a salesman of fraudulent securities for Konex. Neal solicited the CHA investment, and that investment was foreseeable to Oesterman according to evidence showing that Oesterman and Lauer (the CHA insider) were fellow participants in a Konex Summit Meeting. Oesterman also used Lauer as a reference (playing the role of a "satisfied investor") to facilitate his sales. This evidence justifies the district court's finding that Oesterman participated in the scheme as a whole and that its activities were reasonably foreseeable to him. In the absence of clear error with respect to the relevant conduct calculation, there is no reason to disturb Oesterman's sentence.

V.

For the reasons stated, we therefore Reverse the convictions of Joseph Polichemi, Lyle E. Neal, Oscar W. Olson, and Charles Padilla and Remand for further proceedings consistent with this opinion. We Affirm the sentence imposed on Larry P. Oesterman.

Opinion Footnotes

*fn1 This opinion was initially released in typescript.