825 F.2d 1104
23 Fed. R. Evid. Serv. 212
UNITED STATES of America, Plaintiff-Appellee,
David BENTLEY, Richard Degen, Allen Yung, and Walter Josten,
Nos. 86-1858, 86-1873, 86-1907 and 86-2626.
United States Court of Appeals,
Argued May 20, 1987.
Decided June 24, 1987.
Rehearing Denied Aug. 3, 1987.
Martin S. Agran, Agran & Agran, Ltd., Terence F. MacCarthy, Federal Defender Program, David P. Schippers, Davis P. Schippers & Assoc., Chtd., Chicago, Ill., for defendants-appellants.
Pierre Talbert, Asst. U.S. Atty., Anton Valukas, U.S. Atty., Chicago, Ill., for plaintiff-appellee.
Before FLAUM, EASTERBROOK, and MANION, Circuit Judges.
EASTERBROOK, Circuit Judge.
Universal Precious Metals, Inc., was a bucket shop. It "sold" metals, principally copper and silver, for future delivery. It used the mails and phones extensively. Salesmen represented that Universal had the metals in vaults or that the firm would hedge the customers' positions through futures contracts. But Universal had no significant inventories of the metals. When it purchased futures contracts it was more likely to be short (i.e., promising to deliver metals it did not have, which magnified the risk of customers who wanted to buy the metals) than to be long. To protect its customers completely by hedging, Universal had to take a 100% net long position. (Because the business may have other assets to secure customers' positions, the industry standard is 90% net long.) When a firm is properly hedged and the price of the metal rises, the value of the futures contract rises too; the firm may pay this profit to its customer or stand for delivery on the contract, obtaining the metal for the customer's account. If the price of the metal falls, Universal loses on the futures contract, but this matches the customer's loss, and the firm (which writes down its debts to customers) is no worse off. If the firm is short and the market falls, it gains while reducing its debts to customers; when it is short and the market rises, customers' portfolios show "profits" but Universal has no way to pay its customers (unless it has other sources of wealth), having just lost in the market as much money as the customers gained. In other words, by going short, Universal exposed its customers to the risk of calamitous loss with no corresponding prospect of gain for their accounts. Sooner or later, the loss was bound to occur.
Universal had a net long position less than half of the time, and it never covered more than 32% of the customers' positions in silver or 41% in copper. Universal also did not return customers' money when there were paper profits. People who wanted to sell their positions were termed "problem accounts" and turned over to especially high-powered salesmen. When this did not work, the firm promised to send money but often did not; salesmen then claimed to be baffled by the failure of the mails. Universal raked in almost $15 million during its short life (April 1981 to December 1982) and returned about $3 million to customers.
The scheme could not last. Customers frustrated by inability to withdraw their money tend to complain--first within the firm, then to the press and the government. The FBI obtained a search warrant for Universal's premises in September 1982 and carted away most of its records. This did not prevent the firm from carrying on for another ten weeks. The inevitable criminal prosecution did not commence until 1985, in Chicago rather than Ft. Lauderdale, Universal's headquarters. Stranger still, the case was prosecuted under the mail and wire fraud statutes, 18 U.S.C. Secs. 1341 and 1343, rather than the commodities and securities statutes designed for frauds of the type. Counsel for the government could not explain at oral argument the venue or the selection of the charges; as the defendants have not complained about either, we do not pursue the matter.
The indictment charged seven out of the more than 25 people who worked at Universal. Jack Rose, the chief of the sales staff, and one other participant pleaded guilty and testified for the prosecution at trial. A third defendant is a fugitive. This left four for trial. David Bentley, who made Universal's investments, was convicted on 22 counts and sentenced to 12 years' imprisonment to be followed by five years' probation; he was ordered to pay $75,000 as restitution. Richard Degen, Walter Josten, and Allen Yung were salesmen from Universal's boiler room. Yung, convicted on 13 counts, received two years' imprisonment, five years' probation, and an order to pay $30,000 of restitution. Josten, convicted on a single count, was sentenced to three months' imprisonment and required to pay, as restitution, one-third of his earnings while at Universal. Degen was convicted on 13 counts and sentenced to five years' probation on condition that he spend 90 days of that time in prison, perform 1,000 hours of community service, and make restitution of $50,000.
All four maintain that the evidence is insufficient to show that they participated in a "scheme" to defraud. Bentley says he had no idea that Rose and his staff were telling customers that their positions would be covered; the salesmen say that they had no idea that Bentley was not covering the customers' positions. The jury was entitled to disagree. It could conclude that Bentley saw sales literature that touted the security of the firm's investment and hedging practices. Bentley also had been in the metals business before, indeed had run a bucket shop before. After a court issued an injunction against Bentley's fraudulent business practices at First Guaranty Metals, Bentley changed his name (it had been David Smith) and went back into business at Universal. After the collapse of Universal, Bentley went to the well a third time, leading to a citation for contempt of court. The district court kept from the jury most of the history of Bentley's encounters with the law but allowed the jury to learn that Bentley had experience in the metals business. The jury also learned that when the FBI conducted the search in September 1982, it found in Bentley's office many newspaper articles describing bucket shops in metals. The jury could infer that Bentley knew that a business like Universal attracts money on promises of security that he was dishonoring.
The salesmen have a better claim of ignorance, but we must now take all the evidence and permissible inferences in the light most favorable to the prosecution. The evidence permitted the jury to conclude that the salesmen were in on the scheme. (The court gave an ostrich instruction, see United States v. Ramsey, 785 F.2d 184, 188-91 (7th Cir.1986).) The salesmen told tales they knew were false--about their background (exaggerating their credentials and experience), about the success of their other customers (making investments in metals sound profitable for everyone, although they knew that such investments are risky even when the positions are hedged and that most of Universal's customers lost money), about the standing and expertise of the firm (saying that it had representation on the floor of the commodities exchanges and telling customers that the salesmen personally had talked to the phantom representative). No legitimate metals business tags an investor who wants to draw on his account as a "problem account" to be turned over to a different salesman, yet Universal did this. Yung and Degen were among the specialists in "problem" customers. No legitimate commodities business offers to disregard the losses in an investor's account if the investor will send more money rather than asking for his balance to be returned; yet Universal did this too, in an effort to ensure that cash flowed in one direction only. Rose told the sales staff that its function was to separate the public from its money. The jury was entitled to infer that salesmen who listened to Rose and then told lies, "negated" losses on condition of further investment, and identified ordinary investment transactions as "problems", knew enough about the character of Universal's business to be participants in its scheme to defraud. The evidence here was smashing compared to that in United States v. Pearlstein, 576 F.2d 531 (3d Cir.1978), on which the salesmen rely.
There is also enough evidence to permit the jury to infer that there was a single scheme, that the salesmen acted with the necessary intent, and that their activities facilitated the success of the scheme. The participation of the defendants in a common endeavor made joinder appropriate under Fed.R.Crim.P. 8. United States v. Wozniak, 781 F.2d 95 (7th Cir.1985). The defenses were not mutually inconsistent, so severance was unnecessary. See United States v. Buljubasic, 808 F.2d 1260, 1263 (7th Cir.1987). And although the defendants complain about some of the district court's evidentiary decisions, none was erroneous or need be discussed in detail. Only two require comment.
The evidence about Bentley's involvement at First Guaranty Metals showed that he understood the operation of a metals merchant and the importance of adequately hedging customers' positions. It was not admitted to show propensity, the district court considered its potential to support such a forbidden inference, and the court's decision therefore did not offend Fed.R.Evid. 404(b). See United States v. Beasley, 809 F.2d 1273 (7th Cir.1987). The district court excluded, as unduly prejudicial under Fed.R.Evid. 403, evidence of Bentley's escapades after Universal's collapse; this was all he had a right to expect.
The court admitted charts showing Universal's net position in silver and copper futures (both absolute and as a percentage of customers' open positions) every 20 calendar days from May 1981 through September 1982. The defendants grouse that this chart did not show whether Universal was properly hedged on other days, but charts need not be encyclopedic to be admissible. A district judge may admit charts that summarize other evidence. Fed.R.Evid. 1006; United States v. Dana, 457 F.2d 205, 207-08 (7th Cir.1972). A decision on this subject may be disturbed only if the court abuses its discretion. It did not. In a complex trial, such as this one, charts help the jury understand the evidence. The defendants were free to show, if they could, that Universal was properly hedged at other times. A sample of one trading day every 15 over a period of 16 months is large enough to be statistically valid, and this sample showed that more than half of the time Universal had a net short position. This was "prejudicial" to the defendants only because of its probative force. Damning evidence is not inadmissible on that account. The position of defendants' "expert" that Universal might have been hedged on other days--and would have made money if the market had fallen while Universal had short positions--has nothing to do with the admissibility of the charts. It is also lunacy. The probability that a firm short, or holding less than half of the necessary long position, on every day in a large sample, might be properly hedged on all the rest of the days is about one divided by the number of atoms in the solar system.* Someone who says, in essence: "I don't believe in statistics, and you have to look at the position every day because it could be different" has identified himself as a crackpot. A judge need not allow such testimony to go to the jury; trials are complicated enough without the "assistance" of cranks. See United States v. Davis, 772 F.2d 1339 (7th Cir.1985). The district judge was generous to the defendants in allowing them to present this "expert's" attack on the charts. (Evidence that a 1-in-15 sample was too small to produce a statistically significant result, or that the charts had been compiled improperly, would have been valuable, but the defendants offered nothing along these lines.) The "expert's" observation that the firm makes money if the market drops while it is short is true but irrelevant--it simply underlines the prosecution's point that being short could help the defendants while exposing the customers to uncompensated risk. If the market rose, Universal would not have the money to pay its debts to the customers on their purchases, which had become more valuable on paper.
The last matter requiring attention is the search in September 1982. The warrant authorizing the search set out 21 categories of documents that collectively covered every business document likely to be in Universal's files. It excluded only personal documents. The 13 agents executing the warrant (including a legal adviser) examined and seized almost every business document in sight--and followed up by cleaning out the medicine chests and carting away a locked filing cabinet. They searched one room not mentioned in the warrant. Bentley moved to suppress all of the evidence, contending that the warrant was unconstitutionally "general". See Stanford v. Texas, 379 U.S. 476, 85 S.Ct. 506, 13 L.Ed.2d 431 (1965); cf. Andresen v. Maryland, 427 U.S. 463, 478-84, 96 S.Ct. 2737, 2747-50, 49 L.Ed.2d 627 (1976). The district court held that Bentley has standing to contest the search and set for a hearing the questions whether the warrant was too broad and whether the agents exceeded its scope in conducting the search. At the conclusion of the hearing the judge denied the motion with a brief oral statement "that [the warrant] is sufficiently specific" and that the search of the additional room was lawful.
It is far from clear that Bentley has standing to challenge the search. Bentley did not have a privacy interest in the whole of Universal's files, which were maintained by other people. But the government has raised this issue only in a footnote, which is insufficient to present it for decision. The government also has not argued in this court that the warrant was good enough to avoid suppression under United States v. Leon, 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984), even if not constitutionally adequate. Compare United States v. Buck, 813 F.2d 588, 592-93 (2d Cir.1987) (Leon principle applies), with United States v. Fuccillo, 808 F.2d 173, 177-78 (1st Cir.1987) (Leon inapplicable unless warrant is as detailed as possible). We therefore deal exclusively with the two questions that concerned the district judge.
Although Bentley believes the warrant was executed improperly, he does not say that the agents indiscriminately seized personal documents. The government did not use the contents of the medicine chests against him, and the locked cabinet was taken whole only because it appeared to contain business records covered by the warrant, and no one present would open it for the agents. The search of one office not listed in the warrant came about because the officers realized that the warrant had the wrong number (Room 203 instead of Room 201); the right number was readily ascertainable, and the officers did just what they should in searching the right office rather than the one with the number mistakenly given in the warrant. Cf. Maryland v. Garrison, --- U.S. ----, 107 S.Ct. 1013, 94 L.Ed.2d 72 (1987). So only the particularity of the warrant matters.
This is the rare case in which even a warrant stating "Take every piece of paper related to the business" would have been sufficient. Universal was fraudulent through and through. Every transaction was potential evidence of that fraud. In order to show whether Universal hedged properly, the prosecutor had to determine the customers' open positions and Universal's net position in the futures market at any moment. Bentley does not suggest what more limited description would have allowed the FBI to pick up all potential evidence of wrongdoing yet protected any legitimate privacy and property interests Universal may have had. When the whole business is a fraud, the warrant properly may permit the seizure of everything the agents find. See United States v. Kail, 804 F.2d 441, 444-45 (8th Cir.1986); United States v. Weinstein, 762 F.2d 1522, 1532 n. 4 (11th Cir.1985); United States v. Offices Known as 50 State Distributing Co., 708 F.2d 1371, 1374-75 (9th Cir.1983); National City Trading Corp. v. United States, 635 F.2d 1020, 1026 (2d Cir.1980); United States v. Brien, 617 F.2d 299, 306 (1st Cir.1980). The Constitution requires that the warrant "particularly describ[e]" the things to be sought and seized, but when there is probable cause to seize every business paper on the premises, a warrant saying "seize every business paper" particularly describes the things to be searched for and seized.
This does not mean that warrants may use open-ended descriptions. The description must be as particular as the circumstances reasonably permit. So if the fraud infects only one part of the business, the warrant must be so limited--but within that portion of the business "all records" may be the most accurate and detailed description possible. E.g., United States v. Scherer, 523 F.2d 371, 376 (7th Cir.1975) (upholding warrant to search business premises for "business records relating to the purchase and sale of firearms"). See also, e.g., Rickert v. Sweeney, 813 F.2d 907, 909 (8th Cir.1987); Voss v. Bergsgaard, 774 F.2d 402, 404-06 (10th Cir.1985); United States v. Cardwell, 680 F.2d 75, 78 (9th Cir.1982); United States v. Abrams, 615 F.2d 541, 545 (1st Cir.1980). How detailed the warrant must be follows directly from the nature of the items there is probable cause to seize; detail is necessary only to the extent the judicial officer must limit the search and seizure to those items. When there is probable cause to seize all, the warrant may be broad because it is unnecessary to distinguish things that may be taken from those that must be left undisturbed. A generic description adequately defines the officers' authority. When the probable cause covers fewer documents in a system of files, the warrant must be more confined and tell the officers how to separate the documents to be seized from others. See Marron v. United States, 275 U.S. 192, 196, 48 S.Ct. 74, 76, 72 L.Ed. 231 (1927); United States v. Roche, 614 F.2d 6, 7 (1st Cir.1980). But cf. Andresen (sustaining a warrant that did not tell the officers, except by implication, how to separate innocent documents from those to be seized). The fourth amendment bans exploratory rummaging (which diminishes privacy) and excessive seizures (which interfere with property). A warrant authorizing the seizure of truckloads of documents, all on probable cause, offends neither rule. "Take every business document in Room 201" may be perfectly detailed, in the circumstances. A criminal venture does not get extra protection by being large, generating extra paper. Because the agents had probable cause to believe that all of Universal's business records were evidence of crime, the warrant could authorize their capture.
Universal was in business for about 300 trading days, and the chart covered 26 of them. Suppose Universal was properly hedged for 274 of these days and inadequately hedged on the other 26. The probability that a random sample of the 300 days would turn up only the 26 improperly hedged days is 26/300 X 25/299 X ... X 1/275. An alternative way to put the question is: suppose you take 26 draws from a sample of 300, and these draws are 0, 0, 0, 0, 0, 0, 0, -33, -31, -20, 0, -1, 0, -16, -14, 32, 32, 0, 3, 31, 0, -4, 0, -6, -7, -8. (This summary of the evidence concerning silver represents net short positions as negative numbers, and all numbers are futures contracts as percentages of the customers' open positions.) What is the probability that the median over the whole 300-day period is 90? You needn't work through the calculations to see that the answer either way has a lot of zeros after the decimal point before the first significant digit