United States v. Richard Goldberg

BOUDIN, Circuit Judge. Richard Goldberg was convicted of two counts of conspiracy to defraud the Internal Revenue Service, 18 U.S.C. § 371, and eight counts of aiding and assisting the filing of false income tax returns, 26 U.S.C. § 7206(2). Goldberg’s appeal is now before us. We begin by describing the factual background and proceedings in the district court. In the years prior to his indictment in 1995, Goldberg was involved in several businesses in and around Boston. His ventures included a billboard company, Logan Communications, and a partial interest in a “Park ‘N Fly” lot located in East Boston near Logan Airport. Goldberg also owned and operated Liverpool Lumber, Inc., which Goldberg used as a management company for various of his other enterprises. In-or around 1988, Goldberg became aware that the Commonwealth of Massachusetts planned to také all or part of the East Boston Park ‘N Fly lot by eminent domain as part of its Third Harbor Tunnel project. The planned taking not only threatened Goldberg’s profitable parking business, but also his billboard company, since many of its signs were located on the parking lot’s land. Goldberg began an intense lobbying effort against the proposal in 1988, eventually spending over $1 million of his and his partners’ money to oppose the tunnel plans. *772 Two of those hired to oppose the project— community activist Robert A. Scopa and consultant Vernon Clark — were named as co-conspirators in the two separate conspiracies for which Goldberg was ultimately convicted. Taking the evidence most favorable to the verdict, the facts pertaining to the two different conspiracies were as follows. Scopa Conspiracy. From 1990 to 1995, Goldberg employed Scopa to help organize the East Boston community against the tunnel project and to perform other services. But Goldberg never paid Scopa in Scopa’s own name. Instead, Goldberg had his Liverpool Lumber company issue paychecks to three successive “straw” employees, none of whom worked for Goldberg and all of whom agreed to hand the money over to Scopa. To reflect the “wages” of the straw employees, Goldberg directed his bookkeeper at Liverpool Lumber to prepare various W-2, W-3, and W-4 reporting statements, which were then filed with the IRS. These documents falsely described wage payments to straws who had performed no work for Liverpool Lumber. The straws, in turn, falsely included the phantom wages from Liverpool on their own individual returns. Reporting the money on the straws’ returns instead of Scopa’s resulted in a loss of about $150 to the Internal Revenue Service. The government claimed at trial that the scheme was devised so that Scopa would seem to be unemployed and thus could continue to collect monthly benefits under a disability insurance policy. Evidence also indicated that Scopa sought to hide the payments in order to preserve his status as an “independent” activist in the East Boston community and to prevent an extramarital affair from being discovered by his wife. The district court later found that Scopa, but not Goldberg, was motivated by all of these objectives. Clark Conspiracy. In the course of opposing the Third Harbor Tunnel project, Goldberg also retained Vernon Clark, a lobbyist in Washington, D.C., who performed various services to this end. Goldberg’s companies owed Clark a substantial sum of money in 1991 for work performed in opposition to the tunnel project. Rather than pay the bill directly, the two men agreed with others to a more complicated method for Goldberg to discharge his debt to Clark. At the time, Clark was having a secret affair with a woman named Patricia McNally. The pair occasionally spent time in a Maine beach house of which McNally was part owner. Clark sought to fund an expansion of the beach house without his wife’s knowledge. Goldberg agreed to pay the money he owed to Clark to a landscaping company owned by John Lango, McNally’s brother-in-law, who would in turn construct the beach house expansion. Goldberg arranged for the preparation of two separate $10,000 invoices to Park ‘N Fly from Lango, dated October 15, 1991 and January 1, 1992, respectively. The invoices were ostensibly for landscaping services, although Lango performed no work for any of Goldberg’s companies. At Goldberg’s direction, the invoices were paid by Park ‘N Fly. Lango testified at trial that the payments were structured in two installments so as to reduce his taxes on the transaction. The triangular flow of money and services involved the plainly foreseeable preparation and filing of several false tax documents. In due course, Park ‘N Fly sent forms 1099-MISC, one for each $10,000 payment, to the IRS and to Lango. The forms falsely listed the payments as non-employee compensation to Lango. Lango in turn reported the payments as income on his own income tax returns in 1991 and 1992. Clark did not report the money. The foreseeable tax loss to the IRS based on this scheme was about $3,000. A federal grand jury indicted Goldberg on April 6, 1995 for offenses relating to the above activities. The indictment charged Goldberg with two counts of conspiring to defraud the United States government, 18 U.S.C. § 371, several counts of aiding and assisting the filing of false income tax returns, 26 U.S.C. § 7206(2), and several counts of mail fraud based on his alleged efforts to conceal his employment of Scopa from the latter’s disability insurer. 18 U.S.C. § 1341. *773 After moving unsuccessfully to dismiss the indictment, Goldberg waived his right to a trial by jury. Goldberg’s trial before the district judge took eight days, and on September 6,1995, the court announced its findings. The court found Goldberg guilty of conspiring to defraud the government and of aiding and assisting in the preparation of false tax returns, but acquitted him on the mail fraud charges on the ground that his motive to help defraud the insurer had not been proved beyond a reasonable doubt. At Goldberg’s sentencing in December 1995, the district court made guideline calculations (described below) but then departed downward two levels and sentenced Goldberg at the bottom of the range. The result was a ten-month sentence — five months to be served in prison and five in community confinement — as well as three years of supervised release and a $20,000 fine. Goldberg now appeals, challenging his convictions and sentence. The most important and difficult issues on appeal relate to Goldberg’s conviction for conspiracy under 18 U.S.C. § 371 to defraud the IRS. This type of conspiracy is known as a Klein conspiracy, taking its name from an earlier case involving a complex scheme designed to escape taxes. United States v. Klein, 247 F.2d 908 (2d Cir.1957). Goldberg argues that the district court misunderstood the crime’s “purpose” element and that the evidence did not support a conviction. The defraud clause of Section 371 criminalizes any' conspiracy “to defraud the United States, or any agency thereof in any manner or for any purpose.” 18 U.S.C. § 371. Such conspiracies to defraud are not limited to those aiming to deprive the government of money or property, but include conspiracy to interfere with government functions. See, e.g., United States v. Tarvers, 833 F.2d 1068, 1075 (1st Cir.1987). The crime with which Goldberg was charged, therefore, was that he conspired to interfere with the proper functioning of the IRS, through the filing of false tax documents. It is commonly said that in such a conspiracy the fraud has to be a purpose or object of the conspiracy, and not merely a foreseeable consequence of the conspiratorial scheme. Dennis v. United States, 384 U.S. 855, 861, 86 S.Ct. 1840, 1844, 16 L.Ed.2d 973 (1966); 1 Sand et al., Modem Federal Jury Instructions § 19.02 (1990). Consider, for example, the case of a band of bank robbers. All know that the agreed-upon robbery will generate “income” that none of the robbers will report. Yet it would be straining to describe interference with the IRS as a purpose or object of the conspiracy. E.g., United States v. Vogt, 910 F.2d 1184, 1202 (4th Cir.1990). This requirement of purpose accords generally with conspiracy doctrine, United States v. Alvarez, 610 F.2d 1250, 1256 (5th Cir.1980), but it is especially important under the defraud clause of section 371. There are not many financial crimes without some implications for false reporting in someone’s tax filings, if not for tax liability itself. If section 371 embraced every foreseeable consequence of a conspiracy, many joint financial crimes having no other federal nexus — and perhaps many non-criminal acts as well — would automatically become federal conspiracies to defraud the IRS. The “purpose” requirement, however, is easier to state than to apply. The laundering of drug money, for example, normally involves the deliberate concealment of the money’s origin. The primary purpose is almost always to avoid detection of the underlying crime; but can a jury also find an implied secondary objective to conceal income from the IRS? We have held, on specific facts, that a jury could draw such an inference and also find a violation of section 371. E.g., United States v. Cambara, 902 F.2d 144, 147 (1st Cir.1990); Tarvers, 833 F.2d at 1075-76. Such cases are the source of Goldberg’s first argument on this issue. He argues, inventively, that the conspirators either must have as their primary purpose the aim of frustrating the IRS or must be agreeing to undertake the conduct in question to. conceal some other crime. An example of the first alternative (primary purpose) is Klein itself where a web of shell companies and deceptive arrangements was devised to evade taxes; the second alternative (concealment of *774 crime) captures the money laundering precedents. This view of section 371 might explain a number of eases and create a barrier against overreaching by prosecutors. But it makes no doctrinal sense. A conspiracy can have multiple objects, Ingram v. United States, 360 U.S. 672, 679-80, 79 S.Ct. 1314, 1320, 3 L.Ed.2d 1503 (1959), and any agreed-upon object can be a purpose of the conspiracy and used to define its character. The central problem, which ought not be shirked, is to distinguish rationally between cases where interfering with government functions is a purpose and those where it is merely a foreseeable effect of joint action taken for other reasons. This effort poses subtle problems in discriminating “purpose” from “knowledge” and in separating the objects of a conspiracy from its more remote consequences. Volumes could be written on these subjects but — for cases like ours — a more compact solution is at hand: where the conspirators have effectively agreed to falsify IRS documents to misstate or misattribute income, we think that (depending upon the circumstances) the factfinder may infer a purpose to defraud the government by interfering with IRS functions in the sense endorsed by the Supreme Court in Dennis. It may well be that the conspirators in this case had no subjective desire — -primary or secondary — to throw sand in the wheels of the IRS, let alone a subjective aim to reduce tax liability. Goldberg’s argument on this point, with one qualification as to the Clark conspiracy, may be plausible. But fifing a number of false tax documents misattribut-ing income can interfere so clearly and proximately with IRS functions — or at least a factfinder could (but need not) so find — that we see no sharp distinction under section 371 between a purpose to file such documents and a purpose to interfere. In permitting a factfinder to equate the two purposes, we leave untouched the general precept, namely, that mere collateral effects of jointly agreed-to activity, even if generally foreseeable, are not mechanically to be treated as an object of the conspiracy. This would be a different case if, without fifing false tax documents, Goldberg had agreed with his partners to pay Jones under the table, knowing that Jones had no intention of reporting the money to the IRS. If the difference is in degree, then here the degree matters. This brings us to the evidentiary question raised by Goldberg which we rephrase to accord with our just-stated view of the law: does the evidence in this case show that Goldberg and at least one other conspirator shared a purpose to interfere with IRS functions by the fifing of false income reports with the IRS? This question must be asked and answered separately as to each conspiracy, as Goldberg was convicted of two separate conspiracies under section 371 and each conviction involves a separate assessment. In each conspiracy, the illicit purpose that gives rise to section 371 violation must be shared by two or more conspirators. Although the government’s brief stresses the evidence pertaining to Goldberg’s own role and knowledge, a conspiracy to defraud requires at least two who share that aim. Innocent third parties may be the unwitting instruments of a conspiracy. But when it comes to characterizing the purposes or objects of the conspiracy, it is those that are shared by at least two co-conspirators that make up the illegal agreement between them. United States v. Krasovich, 819 F.2d 253, 255 (9th Cir.1987). Here, the district court found that a purpose of the conspirators, in each conspiracy, was to interfere with the IRS. As we have said, such a purpose can be inferred, depending upon the facts, where the very acts agreed to by the conspirators included the fifing of false income-related tax documents. This purpose can fairly be imputed to Goldberg who arranged for the creation of several or more false tax documents in each scheme. The duration and complexity of the schemes, and Goldberg’s own sophistication, add to the inference. There is no evidence that Goldberg discussed the fifing of false tax documents with other conspirators. Yet we think that such conduct was an integral and self-evident part of each conspiracy, permitting the inference *775 that other eo-eonspirators shared in that purpose. In the case of the Seopa conspiracy, false W-2s were given to the straws, who were participants in the scheme, over an extended period. Scopa himself signed a tax return with his wife, who was one of the straws, that incorporated a false W-2. As to the Clark conspiracy, Lango received the false form 1099s, and he in turn reported the false figures to the IRS. Indeed, Lango asked that the amount be divided so that it could be reported in two different years, testifying later that Clark had made the suggestion. This indicates a tax motive but, in addition, shows that both men knew that the filing of false tax documents was an integral part of the scheme, and both ■shared in this purpose with Goldberg. In sum, the evidence supports the trial court’s findings of a common purpose to interfere with IRS functions. In addition to “the danger [of injustice] inherent in a criminal conspiracy charge,” Dennis, 384 U.S. at 860, 86 S.Ct. at 1843, the defraud clause of section 371 has a special capacity for abuse because of the vagueness of the concept of interfering with a proper government function. For that reason, we have examined with special care both the concept and the evidence in this case. But having done so, we conclude that the conduct and purpose of the defendants, although markedly less sinister than in Klein, could properly be found to fall within the outer bounds of section 371. Goldberg next challenges the admission at trial of two out-of-court conversations between Lango and Clark, in which they discussed the false landscaping invoices and the solicitation of Goldberg’s participation in the scheme. These statements were admitted, over Goldberg’s objection at trial, pursuant to Fed.R.Evid. 801(d)(2)(E), which provides that “a statement by a co-conspirator of a party during the course and in furtherance of the conspiracy”' is not considered hearsay. Goldberg does not dispute that Lango and Clark made the challenged statements during and in furtherance of the conspiracy, but he argues that the statements were not admissible against him because they were made before he joined. He relies heavily on our opinion in United States v. Petrozziello, 548 F.2d 20 (1st Cir.1977), where we said that “if it is more likely than not that the declarant and the defendant were members of a conspiracy when the hearsay statement was made, and that the statement was in furtherance of the conspiracy, the hearsay is admissible.” Id. at 23. Although this language has been cited with approval in a few later cases, e.g., United States v. McCarthy, 961 F.2d 972, 976-77 (1st Cir.1992), it conflicts with United States v. Baines, 812 F.2d 41 (1st Cir.1987). Baines expressed the traditional notion that — insofar as hearsay is concerned — a late-joining conspirator takes the conspiracy as he finds it: “a conspiracy is like a train,” and “when a party steps aboard, he is part of the crew, and assumes conspirator’s responsibility for the existing freight....” Id. at 42; accord United States v. Saccoccia, 58 F.3d 754, 778 (1st Cir.1995). Frankly, the underlying co-conspirator exception to the hearsay rule makes little sense as a matter of evidence policy. No special guarantee of reliability attends such statements, save to the extent that they resemble declarations against interest. The exception derives from agency law, an analogy that is useful in some contexts but (as the Advisory Committee noted) is “at best a fiction” here. The most that can be said is that the co-conspirator exception to hearsay is of long standing and makes a difficult-to-detect crime easier to prove. United States v. Gil, 604 F.2d 546, 549 (7th Cir.1979). If starting afresh, one might argue that the narrow Petrozziello version of the exception should be preferred, if only because it accords better with the companion rule imposing substantive liability for other crimes committed during the conspiracy; a co-conspirator is held liable for foreseeable acts of others done in furtherance of the conspiracy but only if committed during the defendant’s period of membership. United States v. O’Campo, 973 F.2d 1015, 1021 (1st Cir.1992). Symmetry is at least convenient. But we are not- starting afresh. The broader Baines test describes the traditional approach, United States v. United States *776 Gypsum Co., 333 U.S. 364, 393, 68 S.Ct. 525, 541, 92 L.Ed. 746 (1948), presumptively-adopted by the Federal Rules of Evidence. It is followed in most circuits. See 2 Saltz-burg, et al., Federal Rules of Evidence Manual 219-22 (5th ed. 1990) (collecting cases). Most important, it is the test in most of our own recent cases, including Saccoccia, decided only 19 months ago. 1 This panel is arguably not free, but is in any event not inclined, to depart from Saccoccia. Goldberg’s next claim on appeal is based on his motion filed prior to trial asking the district court to dismiss the indictment on the ground of selective prosecution. The district court denied the motion without holding a full evidentiary hearing. Goldberg claims that he alleged facts sufficient to require a hearing on his complaint, and he now asks this court to remand the case so that he may have such a hearing. The government is allowed “broad discretion” in deciding whom to prosecute, Wayte v. United States, 470 U.S. 598, 607, 105 S.Ct. 1524, 1530, 84 L.Ed.2d 547 (1985), but there are some limitations. It is said that the government may not base its decision to prosecute on an “unjustifiable standard,” including the defendant’s exercise of protected statutory and constitutional rights. Wayte, 470 U.S. at 608, 105 S.Ct. at 1531. Goldberg bases his selective prosecution claim on the theory that he was targeted by the government in response to his vigorous — and constitutionally protected — lobbying activities in opposition to the Third Harbor Tunnel project. In seeking an evidentiary hearing, a defendant need only allege “some facts (a) tending to show that he has been selectively prosecuted and (b) raising a reasonable doubt about the propriety of the prosecution’s purpose.” United States v. Saade, 652 F.2d 1126, 1135 (1st Cir.1981). But the court may refuse to grant a hearing if the government puts forward adequate “countervailing reasons” to refute the charge, id., and if the court is persuaded that the hearing will not be fruitful. Review on appeal is for abuse of discretion. United States v. Gary, 74 F.3d 304, 313 (1st Cir.1996). Here, Goldberg alleged that one of the prosecutors on the case made a comment to Goldberg’s counsel during a preindietment meeting to the effect that Goldberg “should not have won” his fight with Frederick Sal-vueci, Massachusetts’ secretary of transportation during the tunnel planning stage. Goldberg also claimed that the initials “D.D.” on a prosecution file reflect the complicity in the investigation of David Davis, executive director of MassPort. Finally, he pointed to the fact that several of his coconspirators in the two schemes, including Clark and Lango, were never indicted. The government filed several affidavits to rebut the claim. It denied that the prosecutor in question made the alleged statement to Goldberg’s attorney. It explained that the initials “D.D.” on the file that raised Goldberg’s suspicion in fact referred not to David Davis but to Denise Doherty, an FBI agent assigned to the case. In another district court paper, the government described the origins of its investigation into Goldberg’s activities and gave examples of other recent prosecutions for mail fraud and Klein conspiracies. The district court ultimately denied a hearing, saying that Goldberg’s claims were “close to eonelusory.” We have reviewed the complete filings of both sides and the district court’s explanation. What is involved is a judgment call — tempered on appeal by the deferential standard of review — as to the force and specificity of the allegations, the strength of the response, and the likelihood that a hearing would be helpful. United States v. Lopez, 71 F.3d 954, 963-64 (1st Cir.1995). Here, the district court did not abuse its discretion. The claim of selectivity was quite weak; the government largely explained its choice mainly to pursue Goldberg and Seopa. And the rather modest evidence of wrongful mo *777 tive also melted away, leaving only a single dispute. As to this, four prosecutors denied under oath that the alleged remark had been made. But it is in any event too thin a reed to require an evidentiary hearing, given the lack of surrounding evidence to support a selective prosecution claim. Even less need be said about Goldberg’s later new trial motion, whose summary denial is also cited as error. In substance Goldberg complained that the government did not follow its own internal rules for tax prosecutions or reveal to him information about this decision. The government’s procedures do not create substantive rights, United States v. Michaud, 860 F.2d 495, 499 (1st Cir.1988), and there is no substantial basis for believing that the government withheld Brady material, Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), let alone that its actions were prejudicial. Goldberg’s last claim of error concerns his sentencing. In fixing the sentence, the district court enhanced Goldberg’s base offense level of 10, by four levels — two levels for his role in the offense, U.S.S.G. § 3Bl.l(c), and two more levels for obstruction of justice, id. § 3Cl.l^-to arrive at an adjusted offense level of 14. (All citations are to the November 1995 edition of the guidelines). However, the court departed downward two levels to 12 because it thought Goldberg’s conduct was outside the “heartland” contemplated by the Klein conspiracy sentencing guideline. Id. § 2T1.9. The district judge stated that although Goldberg’s conduct “as a matter of law constitutes a Klein conspiracy, as a matter of sentencing law, it seems to me inappropriate to apply the Klein conspiracy guidelines.” He chose to depart downward two levels because he thought the guideline section for aiding and assisting tax fraud, § 2T1.4, with a base offense level of 8 (when the tax loss is $3,001 to $5,000), was more reflective of Goldberg’s conduct than the Klein conspiracy guideline, § 2T1.9, which has a base offense level of 10. 2 The government, sensibly in our view, has chosen not to pursue an appeal from the downward departure. But Goldberg, as is his right, challenges , the district court’s decision to impose a two-level enhancement for his managerial or supervisory role in the Scopa and Clark conspiracies. The applicable guideline calls for an increase if “the defendant was an organizer, leader, manager, or supervisor in any criminal activity.” U.S.S.G. § 3Bl.l(c). In such a case, a two-level increase applies to joint criminal activity that involved fewer than five participants and was not otherwise extensive. Id. Goldberg says that the only person he managed or supervised was his bookkeeper, Arlene Meucci. Meucci, he argues, does not count under the guideline because she was not a culpable participant. See United States v. Morillo, 8 F.3d 864, 872 & n. 13 (1st Cir.1993); U.S.S.G. § 3B1.1, comment n. 1. However, at the sentencing hearing, the district court found that Goldberg “had a management role” in connection with false payroll arid tax documentation directed to the straw employees in the Scopa conspiracy. We review a district court’s fact-finding at sentencing under a clearly erroneous standard. United States v. Thompson, 32 F.3d 1, 4 (1st Cir.1994). On the record before us, ample evidence shows that Goldberg superintended the straws’ receipt of false tax documents. Goldberg says that Scopa and Clark were the true leaders of the two conspiracies. But a defendant need not be at the top of a criminal scheme to be a manager or supervisor. United States v. Savoie, 985 F.2d 612, 616 (1st Cir.1993). Here, Goldberg’s role was sufficient for the enhancement even if we assume that Scopa conceived of the payroll scheme and may have exercised primary supervision over the straws. Affirmed. Footnotes [1] See Saccoccia, 58 F.3d at 778; O’Campo, 973 F.2d at 1023 n. 5; United States v. Fields, 871 F.2d 188, 194 (1st Cir.1989); United States v. Murphy, 852 F.2d 1, 8 (1st Cir.1988); United States v. Angiulo, 847 F.2d 956, 969 (1st Cir.1988); United States v. Reynolds, 828 F.2d 46, 47-48 (1st Cir.1987); United States v. Cintolo, 818 F.2d 980, 997 (1st Cir.1987). [2] The ten-month sentence — five months to be served in prison and five in community confinement — was the minimum end of the resulting guideline range of 12. U.S.S.G. ch. 5, pt. A (10-16 months at offense level 12).