United States v. William F. Helbling

OPINION OF THE COURT RENDELL, Circuit Judge. William F. Helbling appeals from his conviction and sentence. A jury found that Helbling embezzled funds from a profit sharing plan covered by the Employee Retirement Income Security Act (“ERISA”) to pay the operating expenses of three failing companies he owned, and engaged two lawyers to help him by creating false documents indicating that the withdrawals had been part of a lawful Employee Stock Ownership Plan (“ESOP”) conversion. Helbling’s appeal raises numerous issues relating to the timeliness of his indictment, the sufficiency of the evidence presented at trial, and the calculation of his sentence. 1 We will affirm his conviction and sentence in all aspects. An understanding of the facts of the case is a necessary foundation for a discussion of the issues he raises. I. On December 18, 1996, a federal grand jury returned a thirty-five count indictment against Helbling. The indictment included: (1) one count of conspiracy to embezzle employee pension plan funds and falsify ERISA documents (18 U.S.C. § 871); (2) four counts of embezzlement of employee pension plan funds from an ERISA covered plan (18 U.S.C. § 664); (3) eighteen counts of falsifying documents required by ERISA (18 U.S.C. § 1027); (4) six counts of wire fraud (18 U.S.C. § 1343); and (6) six counts of mail fraud (18 U.S.C. § 1341). The mail fraud counts were dismissed during trial. 2 The jury convicted Helbling of twenty-seven of the remaining twenty-nine counts. Before trial, Helbling filed a motion to dismiss the indictment on the basis that the indictment was not timely. On July 22, 1996, Helbling had signed an agreement waiving his statute of limitations defense. However, Helbling argued to the District Court that the waiver was invalid because he had been coerced into signing it by fraud and misconduct. Helbling also argued that the government had failed to investigate allegations he had made about third parties to a degree Helbling says he believed the waiver agreement required. The waiver agreement specifically allowed Helbling “to present for investigation” his allegations which included claims that a number of individuals purposely injured his companies. After an evidentiary hearing, the District Court denied the motion in part by finding that the government had fulfilled its part of the bargain. The Court found that on October 15, 1996, government agents met with Helbling and accepted from him documents he believed supported his claims. Before ceasing their investigative activities, the agents reviewed the documents and spoke with another agent who had previously investigated related complaints. *232 Trial commenced on May 18, 1998. At trial, the government offered proof that Helbling illegally withdrew money from the profit sharing plan covered by ERISA, used the funds to pay the operating costs of three companies he owned, and had two lawyers help him withdraw the money and legitimize the withdrawals by creating backdated documents to reflect that the plan had been lawfully converted into an ESOP. Helbling did not contest many of the background facts presented, at trial including his control of the three companies, his administration of the plan, or the financial transactions themselves. Hel-bling instead argued that the government failed to establish that he had acted with the requisite criminal intent, and that the government witnesses were lying. In his pro se brief, Helbling explains that he acted on the advice of counsel who told him that he could withdraw funds from the plan and document the ESOP conversion later as long as he had secured the consent of the company’s board of directors. To prove its case, the United States presented numerous documents and several witnesses. The witnesses included Hel-bling’s alleged co-conspirators, the two lawyers, Gerald S. Susman and Stephen Sokolic, who testified to the false ESOP conversion, Laura Scurko, who testified about the financial transactions and explained that ERISA covered the plan, Barry Penn, Susan Kramer and Donald Mayle, who testified to the forgery of two important documents, and John Grikis and Barry Katz who managed the plan investments at NatWest Bank and Oppenheimer & Co. Several of the plan participants also testified. The witnesses explained that Helbling was the president, chief executive officer, and sole shareholder of three companies, Micro-Technology Co. (and its subsidiary Micro-Products Engineering Co.), Scranton Electronics, Inc., and Yardley Group, Inc., which Helbling ran as one company, and the administrator of Micro-Products Engineering Company Profit Sharing Retirement Plan. The plan was funded exclusively by Micro-Products. Ed Wisniewski, a plan participant and long-time employee, testified that the plan was established by a previous owner in 1965 to provide retirement income as an incentive to salaried employees to remain with the company. Laura Scurko, an attorney who was appointed trustee of the plan in a civil suit brought by the plan participants, testified as a lay witness and explained that the plan was covered by ERISA. She pointed out that the plan documents stated that the plan was amended and restated in 1976 to comply with the Employee Retirement Income Security Act of 1974. As of March 1991, the plan had assets of approximately $625,000 and covered ten salaried employees. The plan’s assets were held by Nat-West Bank which also acted as trustee to the plan. Helbling’s companies were manufacturing companies that relied heavily upon military contracts. In May 1990, at Hel-bling’s direction, the companies filed for Chapter 11 reorganization. In February 1992, the Bankruptcy court converted the Chapter 11 bankruptcy proceedings into a Chapter 7 liquidation because the companies failed to provide the Court with the required monthly filings. The companies subsequently ceased business. While in bankruptcy, the companies continued to struggle because they lacked sufficient cash flow and were unable to procure new military contracts. In February 1991, Helbling directed John Grikis, then a trust officer at Nat-West, to transfer the plan’s assets to Oppenheimer & Co. Helbling’s companies had been contacted by Barry Katz, an Oppenheimer fund manager, in late 1990. The funds were transferred on March 13, 1991 after Grikis received a letter sent by Katz indicating that Oppenheimer would assume responsibility for the funds. The day after the transfer to Oppenheimer, Helbling moved $125,000 to an account, at Farmers & Mechanics Bank and used it to pay operating expenses. In July, after conver *233 sations with Katz during which Helbling explained that he wanted to remove more money, Helbling converted the account to a margin account and borrowed $350,000 against the plan’s assets. Katz approved the conversion after receiving a fax supposedly signed by Barry Penn, a lawyer, indicating that Micro-Product’s plan permitted the company to borrow against the plan’s assets. The money was again used to pay operating expenses. At trial, Penn and his secretary, testified that the letter was a forgery. They explained that Hel-bling, who had been Penn’s client, twice asked Penn to write the opinion letter, first in person and later by fax, but that Penn refused because he was not familiar with pension law or the profit sharing plan. Subsequently, Helbling instructed Katz to close the account and sent Katz a letter, dated August 29, 1991, which confirmed that the plan had been converted into an ESOP. Helbling subsequently withdrew money from the account in the form of a check for $55,300 on September 6, 1991, and a wire transfer for $29,700 on September 16, 1991. Helbling withdrew a final amount of $3,500 on January 3,1992. Attorneys Gerald Susman and Stephen Sokolic were the government’s two key witnesses. Both testified pursuant to plea agreements. Susman testified first. He explained that he first spoke with Hel-bling in the early part of 1990 after Hel-bling had been referred to him by Arnold Kaminer, who provided health insurance for Helbling’s companies, to help Helbling establish a trust to hold his life insurance. During the summer of 1990, Susman completed some estate planning work for Hel-bling. Kaminer confirmed that he referred Helbling to Susman for estate planning, and also testified that he mentioned the possibility of an ESOP in a conversation with Helbling in February of 1991. Susman testified that Helbling called him again in March, and July, of 1991. During the conversation in March, Hel-bling asked Susman if he knew of any sources of financing because his companies were having financial trouble. During the conversation in July, Helbling asked Sus-man about withdrawing money from a profit sharing plan. Sometime around August, 20, Helbling called Susman again. This time, according to Susman, Helbling told him that he had made the large March and July withdrawals from the Micro-Products plan. Susman testified that he “was flabbergasted” when Helbling told him in August of the large withdrawals. Susman had thought that Helbling had been asking about removing money from his own personal stake in a plan when he answered Helbling’s questions in July. Susman immediately told Helbling that he could not remove funds from the plan unless he had an ESOP and explained that to have an ESOP, Helbling needed a valid resolution of the board of directors that predated the distribution, and an appraisal of the stock. Susman testified further that he explained that since the companies were in bankruptcy, the stock was worthless. When Hel-bling asked if Susman would prepare the required resolution, Susman refused. When Helbling asked if they could still go ahead with the ESOP, Susman said that he would look into the matter and get back to him. Helbling and Susman met the next day, and Helbling provided Susman with several documents including a copy of the plan. Susman subsequently sought advice from Stephen Sokolic, an attorney with more experience with ERISA, to find out whether they could document what had been done as an ESOP. Susman testified that after talking with Sokolic, he felt that the documentation could legitimately be completed. Subsequently, Susman reviewed Helbling’s August 29, 1991 letter to Katz that said that the ESOP had been converted into an ESOP under Susman’s direction. The letter also authorized the liquidation of the account. Both Susman and Sokolic testified that they reviewed the documents Helbling *234 gave them and met with Helbling on September 11, 1991. During that meeting, Susman and Sokolic explained to Helbling that the withdrawals were prohibited transactions under ERISA. However, after the meeting, Susman and Sok-olic prepared a retainer agreement which Helbling later signed. The retainer agreement stated explicitly that the loan Helbling took from the plan was prohibited by ERISA, and that there could be a subsequent Department of Labor investigation and possibly fines because the transactions might violate Department of Labor regulations. However, on cross-examination both Susman and Sokolic testified that while they told Helbling the withdrawals had been prohibited transactions under ERISA, they did not tell Helbling that the withdrawals or the creation of the backdated ESOP could be the basis for criminal prosecution. Susman and Sokolic testified that on October 14, 1991 they met again with Hel-bling who was accompanied by Donald Mayle, an employee of Helbling’s companies. In preparation for this meeting, Susman and Sokolic drafted most of the documents needed to achieve the ESOP conversion. At the meeting, Helbling signed the employee stock ownership plan and the ESOP trust agreement. Both documents were backdated to July 25, 1991. He also signed the stock purchase agreement, which purported to transfer 12.5% of the stock to the plan for $500,000, which was backdated to March 1, 1991. 3 Mayle witnessed Helbling’s signatures. A promissory note for $85,000, representing the additional funds withdrawn from the Oppenheimer account in September, was signed by Helbling in November, and backdated to September 8, 1991. The conversion was made dependent upon a later appraisal of the companies validating the stock price. Both Susman and Sokolic testified that they were originally unwilling to backdate the ESOP plan document and the trust agreement, but gave in to Helbling’s demand during the meeting. Sokolic, however, testified that he backdated the stock purchase agreement and the stock certificates to coincide with Helbling’s withdrawals before the meeting. In response to questions from Helbling’s bankruptcy attorneys in December of 1991, Susman and Sokolic drafted an opinion letter with Hel-bling’s input that Susman and Sokolic testified was misleading and false. Ronald Santora, Helbling’s bankruptcy attorney, subsequently presented the letter to the Bankruptcy Court. When asked why he agreed to create the documents, Susman testified that “he wanted the fee.” On cross-examination, Susman testified that he did not intend to steal from the plan participants and did not conspire with Helbling to steal money. Susman testified that after he first met with Sokolic he thought the ESOP conversion could be completed legally, and that he first understood that he had committed a crime in November of 1991. He did, however, admit that earlier, in August, he had committed a crime because he had participated in Helbling’s illegal activities by reviewing a draft of Helbling’s August 29 letter to Katz which confirmed that the plan had been converted into an ESOP and authorized the liquidation of the Oppenheimer account. Sokolic corroborated much of Susman’s testimony, particularly the meetings with Helbling, the retainer agreement, and the backdating of documents. Sokolic testified that he realized that he had committed a crime in “hindsight,” explaining that he originally believed that he was serving his client to the best of his ability by documenting the ESOP, and that it was not until November and December of 1991 that he first began to reconsider the legality of his acts. Sokolic, however, also conceded repeatedly that his actions were illegal because he created false or misleading documents that were required by ERISA. *235 Sokolic explained his initial view by saying that he had understood that Helbling intended to create the ESOP in March of 1991, as indicated by the board of directors’ consent dated March 8, 1991, and that he had felt he was simply documenting that transaction. Sokolic said that he first saw the consent sometime between the September and October meetings with Helbling. At trial, Donald Mayle, who along with Helbling supposedly signed the consent, testified that his signature was forged. Susman testified that Helbling brought the consent to the October 14 meeting. Susman also testified that Helbling had asked him to create a board of directors’ consent during the first August conversation. The government also presented witnesses who testified that Helbling gave them specific directions on how to refer to the infusions of cash in reports to the Bankruptcy Court and in certain letters. Donald Mayle testified that Helbling directed him to record the money as receipts and inventory in reports filed with the Bankruptcy Court. Thomas Taylor, former sales manager and later vice-president of sales at Scranton Electronics, testified that Helbling directed him to sign two letters intended to help Helbling’s companies obtain a Certificate of Competence from the Small Business Administration stating that Helbling had personally injected $350,000 into the companies. Obtaining a certificate would have made it easier for Helbling’s companies to procure military contracts. Throughout the period, the plan participants requested information from Hel-bling about the plan. After the money was moved from NatWest, NatWest stopped providing the plan participants with monthly statements. Helbling first told the participants about the conversion in October. After two meetings during which plan participants protested the conversion, they filed a lawsuit. The record from the sentencing hearing reflects that Helbling had sent the participants letters blaming them for the companies’ difficulties and threatening them with unemployment, sought to file retaliatory and harassing civil suits against them, falsely informed the state of New Jersey that they were responsible for approximately $240,000 in back taxes due to the ESOP conversion, and refused to comply with the order entered at the end of the civil proceedings. As a consequence of the civil lawsuit, Lauren Scurko was appointed trustee. In her role as trustee, Scurko recovered less than $24,000 of the original amount. At the end of the government’s case, Helbling offered no witnesses on his behalf. Helbling moved to dismiss the conspiracy count arguing that the government failed to establish that either Susman or Sokolic entered into an agreement with Helbling. The District Court denied the motion. After closing arguments were heard, Helbling moved for a new trial based upon alleged misconduct of the prosecutor for making impermissible statements in the opening and closing arguments, and for reconsideration of his statute of limitations argument. The District Court also denied both motions. Helbling was sentenced on January 25, 1999. The District Court increased Hel-bling’s offense level under S 3B1.1 of the Sentencing Guidelines for Helbling’s role in the offense and under § 3C1.1 for obstruction of justice based on Helbling’s efforts to pressure Susman, Sokolic and Mayle to lie during the investigation. The District Court also departed upwards under § 5K2.3 for extreme psychological damage to the victims of Helbling’s embezzlement. Helbling now appeals the District Court’s sentencing decisions. He also seeks a new trial based on statements made in the prosecutor’s opening and closing statements, the insufficiency of the evidence presented at trial, allegations of prosecutorial misconduct, and the District Court’s decision on the validity of the stat *236 ute of limitations waiver. We will address all of these issues beginning with the statute of limitations waiver. II. Helbling argues that the District Court erred by upholding the validity of his waiver of the statute of limitations. His primary contention is that the waiver must be invalidated because the government failed to fulfill its part of the bargain, which required the government to allow Helbling “to present for investigation” charges he had leveled against individuals he felt sabotaged his companies. 4 He urges that the government was obligated to complete a full and thorough investigation of his allegations. He alleges that the District Court erred in rejecting this contention because it failed to inquire into what Hel-bling reasonably understood the waiver to mean. We will affirm the District Court’s decision because, even if Helbling is correct to assert that we must compare the government’s actions to what Helbling “reasonably understood” it had promised to do, 5 we cannot conclude, on the basis of the language of the waiver or the record in front of us, that Helbling reasonably understood the waiver to require an investigation any broader than he actually received. The District Court addressed this specific contention, along with others, in a written opinion, after holding an evi-dentiary hearing into the circumstances surrounding the signing of the waiver agreement. Finding the government’s witnesses credible, the District Court addressed each of Helbling’s allegations as to why the waiver was fraudulently coerced and rejected each of them. 6 Im *237 portantly for this discussion, the District Court found that Helbling knew and understood the implications of waiving the statute of limitations, was not unduly-pressured into signing the waiver since he engaged in extensive negotiations regarding the waiver provision, and was repeatedly advised to retain, or request, counsel. Addressing whether the government fulfilled its bargain, the District Court noted that the provision itself did not clearly describe the extent of investigation required, and stated that Helbling had not introduced any evidence showing that a particular meaning had been agreed upon. The District Court proceeded on the basis that the agreement imposed a duty on the government to investigate and found that the government had done enough to satisfy its obligation and, therefore, the waiver was valid. We apply the clearly erroneous standard of review to the District Court’s finding of facts and review the District Court’s legal conclusion that the government fulfilled the terms of the agreement under the plenary standard of review. See United States v. Moscahlaidis, 868 F.2d 1357, 1360 (3d Cir.1989) (reviewing a district' court’s determination that the government did not breach a plea agreement). The difficulty with Helbling’s argument on appeal is that his assertion — that the government promised to engage in a full scale investigation — -cannot be squared with the record evidence, or the language of the waiver itself. Helbling argues in effect that he reasonably understood that the government would complete a full scale investigation because he would not have agreed to the waiver without such a promise. However, even if we were to step beyond the language of the provision which only requires the government to allow him to “present for investigation” — a promise the government clearly fulfilled— we cannot find support for his assertion because at the pre-trial hearing a government agent testified that she told Helbling that the government would not agree to more than a promise to evaluate Helbling’s evidence. Helbling signed the waiver agreement on July 22, 1996. According to testimony presented at the hearing, and which the District Court found credible, SA Donald Wadsworth had spoken to Helbling on July 19, and Helbling advised him that he wanted Wadsworth to perform an expansive investigation and include Helbling’s allegations against others in his review. When Helbling, thereafter, told Department of Labor investigator Daphne Rich that Wadsworth had promised him that if he signed the waiver, the government would investigate his charges, Rich said that the government would only evaluate the evidence he gave to Wadsworth and Rich. Thereafter, that day Helbling signed the waiver in the presence of two other FBI agents. Therefore, we conclude that Helbling’s belief that the waiver required the government to fully investigate his claims was unreasonable. Furthermore, we note that the government did indeed do more than permit Hel-bling to “present” his allegations. Reviewing the evidence, the District Court found that, before the government ceased its investigation into Helbling’s allegations, government agents met with Helbling and discussed his allegations, reviewed the documents Helbling gave them, and culled out those which the government already had, those that were new and those that were related to a prior investigation. The agents also spoke to Robert Connolly, the Chief of the Antitrust Division of the Department of Justice in Philadelphia, who had already investigated allegations related to many of Helbling’s complaints of sabotage in a protracted investigation from 1989-1992, and reviewed the documents with him. ■ Accordingly, we find the waiver to have been valid. *238 III. Next, Heibling challenges the sufficiency of the evidence supporting all the counts of his conviction. Hethling argues that the government failed: (1) to establish an agreement necessary for a conspiracy; (2) to establish a scheme to defraud required by the wire fraud convictions; or (3) to establish that ERISA covered the Micro-Products plan, as required by the embezzlement and false documents convictions. 7 Our standard of review is highly deferential. "We determine whether there is substantial evidence that, when viewed in the light most favorable to the government, would allow a rational trier of fact to convict." Government of the Virgin Islands v. Charles, 72 F.3d 401, 410 (3d Cir.1995). We find that substantial evidence did support the verdicts and we will affirm all the convictions on all counts. Helbling first argues that the government failed to establish that he conspired with either Susman or Sokolic. 8 To establish a conspiracy, the government must show: (1) a unity of purpose between two or more persons; (2) an intent to achieve a common goal; and (3) an agreement to work together. See United States v. Carr, 25 F.3d 1194, 1201 (3d Cir.1994). Circumstantial evidence may be used to establish a conspiracy although if the government relies entirely on circumstantial evidence "the inferences drawn must have a logical and convincing connection to the facts established." Id. (quoting United States v. Casper, 956 F.2d 416, 422 (3d Cir.1992)) (internal quotations omitted). The indictment charges Helbling with conspiracy (1) to defraud the plan and participants, or (2) to put false information in documents required by ERISA. We must affirm the jury verdict if the government adduced sufficient evidence of either object of the conspiracy. See id. at 1201-02 (citing Griffin v. United States, 502 U.S. 46, 112 S.Ct. 466, 116 L.Ed.2d 371 (1991)). Heibling argues that the government failed to show that he entered into an agreement with either Susman or Sokolic because both testified that they did not intend to commit unlawful acts and that they came to understand the illegality of their actions only by hindsight. However, their testimony was replete with statements from which the jury could reasonably have inferred that Sokolic and Susman had the necessary intent and knowledge regardless of their statements to the contrary. For example, both testified that they acted illegally, and that they agreed to complete work for Hel-hung after explaining to him that the withdrawals were prohibited transactions. Contesting his wire fraud convictions, Helbling argues that the government failed to establish that Helbling participated in a scheme with the specific intent to defraud. See 18 U.S.C. § 1343; United States v. Veksler, 62 F.3d 544, 551-52 (3d Cir.1995). Helbling argues that the government failed to prove this element of the crime because he relied upon the expertise of Susman and Sokolic, and because both testified that they did not intend to defraud the participants. Clearly, sufficient evidence existed to establish that Susman and Sokolic had the requisite intent. However, even if *239 Helbling’s contention about Susman’s and Sokolic’s intent had merit, their intent is irrelevant to Helbling’s conviction since a wire fraud conviction is not dependent upon the existence of an agreement. See United States v. Nelson, 54 F.3d 1540, 1546 (10th Cir.1995). Susman and Sokolic testified that Helbling directed them to construct the ESOP and backdate the documents. Testimony also supports the conclusion that Helbling forged Barry Penn’s letter to Oppenheimer before engaging the services of Susman or Sokolic, and that he alone forged the consent of the directors sometime after engaging their help. These acts are clearly probative of Hel-bling’s intent. Finally, Helbling contends that the government failed to establish that the plan was subject to ERISA. Counts 2 through 5 of the indictment charged Helbling with embezzlement of plan assets in an “employee benefit plan subject to any provision of title 1 of [ERISA]” in violation of 18 U.S.C. § 664 and counts 6 through 23 charged Helbling with violating documents required by title I of ERISA to be published, kept as records or certified to the administrator of the plan. See United States v. Furst, 886 F.2d 558, 565 (3d Cir.1989) (explaining that a conviction under § 664 requires a showing that the plan is covered by ERISA); see also United States v. Coyle, 63 F.3d 1239, 1245 (3d Cir.1995) (explaining that a conviction under 18 U.S.C. § 1027 requires a showing that (1) the defendant made a false statement; (2) knowing it to be false; (3) in a document required by ERISA). A plan is subject to ERISA if it fits under 29 U.S.C. § 1003(a), and is not exempted by the 29 U.S.C. § 1003(b). Relevant definitions are provided in 29 U.S.C. § 1002. Helbling contends that the government failed to introduce sufficient evidence from which the jury could conclude that the plan was covered by ERISA, and argues that the plan was not covered by ERISA as a matter of law. 9 Generally, whether a plan is covered by ERISA is an issue of fact to be decided by the jury, which we review under the substantial evidence standard, see Furst, 886 F.2d at 565, though we exercise plenary review when our review requires the interpretation and application of legal precepts. See United States v. Martorano, 767 F.2d 63, 65 (3d Cir.1985). At trial, the District Court precluded the testimony of two experts offered by the government who would have testified about the nature of the plan because the government failed to comply fully with discovery requirements. 10 The government therefore relied upon the lay testimony of Laura Scurko, the court appointed plan trustee, Susman and Sokolic. Scurko testified that the plan was covered by ERISA, that the plan was restructured to be an ERISA plan, and that the plan itself suggested that it was covered by ERISA. Susman and Sokolic predicated their testimony, as well as their guilty pleas, on the basis that the plan was covered by ERISA. The government also introduced testimony *240 of the plan participants indicating that the plan had the characteristics of an ERISA covered plan, including testimony that the plan was established to provide retirement income to workers who did not receive piece rate incentives, after they left the company. Reviewing the factual evidence adduced at trial and the legal arguments raised in response by the government, we conclude that sufficient factual evidence and legal support existed for the conclusion that the plan was covered by ERISA. The government convinces us as an initial matter, with reference to the facts discussed above, that the plan fell within the general ERISA definitions, particularly 29 U.S.C. § 1002(2)(A). See In re New Valley Corp., 89 F.3d 143, 148-49 (3d Cir.1996) (explaining that a court must begin first look to see whether the plan fits under the general provisions of ERISA and then consider the exceptions). Section 1002(2)(A) extends ERISA to cover plans established or maintained by an employer to the extent that it provides retirement income regardless of the method used to determine contributions, or calculate and distribute benefits. See 29 U.S.C. § 1002(2)(A). The government also explains that the plan fits within the definition of a defined contribution plan, see 29 U.S.C. § 1002(34), and correctly rebuts Helbling’s arguments as to why the plan was exempt from ERISA coverage in whole or in part. Finally, the government points out that even if the plan was exempt from certain provisions of ERISA, for example, by having the characteristics of a “top hat” plan, 18 U.S.C. § 1027 and 18 U.S.C. § 664 would still apply because they do not require that the plan in question be subject to all of ERISA’s various provisions. Rather, these criminal provisions require that the plan be a plan “subject to any provision of title 1,” 18 U.S.C. § 664, or that the fraudulent statement be contained in a document required by title 1. See 18 U.S.C. § 1027. rv. Helbling next argues that the prosecutor, made prejudicial comments in her opening and closing statements that require the grant of a new trial. Helbling argues that her opening statement included impermissible characterizations and impermissible vouching, and that in her closing and rebuttal arguments, the prosecutor again acted impermissibly by characterizing Helbling and his actions, implying that Helbling committed improprieties in the past, and by describing the evidence that the letter supposedly signed by Barry Penn was forged as “uncontroverted.” Without parsing the prosecutor’s comments, which we have paraphrased in the margin, 11 we note that, although the gov- *241 eminent disputes many of Helbling’s contentions of misconduct, the United States acknowledged in its brief that the prosecutor’s opening statement “overstepped the parameters of an opening statement” in the manner in which it characterized Hel-bling. See Appellee’s Br. at 38. We quite agree that the prosecutor’s remarks were over the line, and arguably even out of line. The government will not be sanctioned in this instance, however, because our case law punishes the government by granting a new trial in such a situation only if the defendant was prejudiced by the remarks in question. Here, we conclude that a new trial is not warranted because Helbling was not in fact prejudiced. We reach this conclusion even though we view the prosecutor’s remarks to have been inappropriate, and we urge that the United States Attorney for New Jersey remind his assistants of the limits of appropriate advocacy. “Prosecutorial conduct does not always warrant the granting of a mistrial.” United States v. Zehrbach, 47 F.3d 1252, 1264 (3d Cir.1995) (en banc). “An appellate court should not exercise its ‘supervisory power to reverse a conviction ... when the error to which it is addressed is harmless since, by definition, the conviction would have been obtained notwithstanding the asserted error.’ ” Id. (quoting United States v. Hasting, 461 U.S. 499, 506, 103 S.Ct. 1974, 76 L.Ed.2d 96 (1983)). The harmless error standard requires us to consider the record as a whole. The standard of review depends upon whether the error was constitutional or non-constitutional. If the error is non-constitutional, we will affirm “when it is highly probable that the error did not contribute to the judgment.” Id. (quoting Government of Virgin Islands v. Tato, 529 F.2d 278, 284 (3d Cir.1976)). If the error is constitutional, we will affirm if we find that the error is harmless beyond a reasonable doubt. See United States v. Molina-Guevara, 96 F.3d 698, 703 (3d Cir. 1996) (citing Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967)). Even under the most stringent harmless error analysis, we cannot conclude that Helbling was prejudiced. “In determining prejudice, we consider the scope of the objectionable comments and their relationship to the entire proceeding, the ameliorative effect of any curative instructions given, and the strength of the evidence supporting the defendant’s conviction.” Zehrbach, 47 F.3d at 1265. First, the evidence presented by the United States was overwhelming. See id. at 1267 (finding the evidence at trial substantial). As discussed above, the government adduced substantial direct evidence of Helbling’s activities, and substantial circumstantial evidence of his motivation and intent. Second, the District Court did give curative instructions. It repeatedly instructed the jury not to consider counsel’s arguments as evidence, doing so before the prosecutor’s opening, after the prosecutor’s opening, and before her closing argument, and appeared to do so in a way that conveyed the Court’s displeasure with counsel’s tactics. See id. at 1267 (finding characterizations neutralized in part by instructions not to consider the opening evidence); United States v. Somers, 496 F.2d 723, 738 (3d Cir.1974). Furthermore, the Court told the jury to disregard the vouching in the prosecutor’s opening statement and to disregard the references to “arrogant” and “despicable” in rebuttal. Last *242 ly, we note that defense counsel asked the Court not to give a curative instruction to neutralize reference to “uncontroverted” testimony in a side bar conference. Helbling seeks to distinguish this case by arguing that he was prejudiced because the government’s comments poisoned the jury's mind regarding his intent, a crucial element in each count of his indictment. Helbling places considerable reliance on United States v. Mastrangelo, 172 F.3d 288 (3d Cir.1999), in which we found the defendant to be prejudiced by the comments of a prosecutor in closing and rebuttal arguments which inflated a limited stipulation that the defendant “had the chemical background to know the ingredients and equipment necessary to make methamphetamine” into one stating that the defendant knew how to make methamphetamine. See Mastrangelo, 172 F.3d at 295. We find Mastrangelo to be easily distinguishable from our case. First, and most importantly, in Mastrangelo, the prosecutor misstated a stipulation to make it encompass a fact central to the government’s charges and not otherwise supported by any evidence the government presented. See id. at 298. Here, the government introduced overwhelming circumstantial evidence of Helbling’s intent. Second, in Mastrangelo, we explained that the district court’s curative instruction may have served to exacerbate the problem because the court reaffirmed the prosecutor’s incorrect view of the stipulation. See id. at 296. Here, we view the curative instructions as having neutralized the possible prejudice. We note that Helbling’s argument is also weakened by United States v. Retos, 25 F.3d 1220 (3d Cir.1994), where we found that a prosecutor’s reference in opening argument to the defendant’s “crooked law practice” was not misconduct requiring a new trial since the comment was clearly related to the charges the government had to prove and was supported by the evidence presented at trial. See Retos, 25 F.3d at 1226. But see Somers, 496 F.2d at 737-38 (‘Whether or not proofs were ultimately adduced warranting such characterizations is irrelevant. Such characterizations add nothing to the legitimate education of the jury which is not afforded by the proper presentation of the facts to be proved.”). Here, although the prosecutor’s comments may have been a pointed assertion of Helbling’s guilt, the characterizations were related to the charges contained in the indictment which the evidence presented later did in fact establish. Accordingly, we find prejudice to be lacking. V. Helbling also appeals the District Court’s determination of his sentence. Helbling contends that the District Court erred by increasing his offense level four-levels under U.S.S.G. § 3Bl.l(a), “Aggravating Role,” by increasing his offense level two-levels for obstruction of justice, U.S.S.G. § 3C1.1, and by departing upwards two-levels for psychological harm. See U.S.S.G. § 5K2.3. A. § SBl.l(a), Aggravating Role Helbling argues that the District Court erred in applying a four-level enhancement to his offense level pursuant to § 3Bl.l(a) of the Sentencing Guidelines for “Aggravating Role.” Section 3Bl.l(a) states in full: “If the defendant was an organizer or leader of a criminal activity that involved five or more participants or was otherwise extensive, increase by 4 levels.” U.S.S.G. § 3Bl.l(a). Helbling finds fault with the District Court’s view regarding both aspects of this Guideline, namely that he was an organizer or leader, and that the activity was “otherwise extensive.” Both Hel-bling and the government agree that Hel-bling’s criminal activity did not involve five “participants” as defined by the commentary to the Guideline. We review a District Court’s factual determinations underlying the application of the sentencing guidelines for *243 clear error. See United States v. Ortiz, 878 F.2d 125, 126-27 (3d Cir.1989). Although we give due deference to the District Court’s application of the sentencing guidelines to those facts, as required by 18 U.S.C. § 3742(e), we exercise plenary review over legal questions involving the proper interpretation and application of the sentencing guidelines. See United States v. Katora, 981 F.2d 1398, 1402 (3d Cir.1992). We will address the two aspects of the relevant Guideline section in turn. 1. Leader or Organizer Section 3Bl.l(a) applies only if a sentencing court finds that “the defendant was an organizer or leader of a criminal activity.” Application Note 4 explains that a court should consider a number of factors when determining if the defendant was an organizer or leader, rather than a manager or supervisor. In distinguishing a leadership and organizational role from one of mere management or supervision, titles such as “kingpin” or “boss” are not controlling. Factors the court should consider include the exercise of decision making authority, the nature of participation in the commission of the offense, the recruitment of accomplices, the claimed right to a larger share of the fruits of the crime, the degree of participation in planning or organizing the offense, the nature and scope of the illegal activity, and the degree of control and authority exercised over others. U.S.S.G. § 3B1.1, Application Note 3; United States v. Ortiz, 878 F.2d 125, 127 (3d Cir.1989). We have explained that to be considered an organizer or leader, “the defendant must have exercised some degree of control over others involved in the commission of the offense.” United States v. Phillips, 959 F.2d 1187, 1191 (3d Cir. 1992) (quoting United States v. Fuller, 897 F.2d 1217,1220 (1st Cir.1990)). The District Court made specific factual findings, which are not clearly erroneous, in determining that Helbling acted as an organizer or leader. The Court found that: Helbling recruited all of the participants mentioned above and his attorneys, Sok-olic and Susman, to help him in his criminal activity. He alone made the decision to conceive and implement his conversion scheme. The benefits from his scheme went either to his personal use or into shoring up companies which he was the sole owner. Helbling played the key role in structuring the operation to defraud the retirement plan and convert its assets. There is no evidence that anyone else exercised control over the criminal scheme, nor exercised any decision-making authority. On appeal, Helbling raises two arguments. Helbling argues first that the evidence failed to show that Helbling recruited Sus-man and Sokolic with a criminal purpose or controlled their activities. Helbling also argues, citing United States v. Katora, 981 F.2d 1398 (3d Cir.1992), that he cannot be a leader or organizer because the other participants — Susman and Sokolic — were equally responsible for the criminal activity. Given the guideline provision and the record evidence, we find the District Court’s factual findings to be proper and its legal conclusion that Helbling was an organizer and leader unassailable. In so doing, we necessarily reject Helbling’s arguments which are essentially fact based. Contrary to his urging, the District Court could properly find that he had a criminal purpose in recruiting Susman, that he controlled their activities in orchestrating the scheme through them thereafter, and that he was more culpable than they were. 12 On *244 these facts, Katora does not dictate a different result. 13 Finally, we note that evidence that certain individuals provided expertise or planning does not necessarily counter evidence that their actions were controlled by another. 2. “Otherwise Extensive” The second part of § 3Bl.l(a)’s analysis requires the sentencing court to determine whether the “criminal activity ... involved five or more participants or was otherwise extensive.” U.S.S.G. § 3Bl.l(a). Since the government acknowledges that Hel-bling, Susman and Sokolic are the only “participants,” this case requires us to determine when a criminal activity is “otherwise extensive.” See United States v. Col-letti, 984 F.2d 1339, 1346 (3d Cir.1992) (holding that the leader or organizer may be included when determining the number of participants); see also U.S.S.G. § 3B1.1, Application Note 1 (defining “participant”). This seemingly simple phrase, which also appears in § 3Bl.l(b), has spawned much discussion, and some disagreement, in the opinions of several courts of appeals. The courts that have considered the question agree that the “otherwise extensive” language makes § 3Bl.l(a) applicable when criminal activity involves the equivalent of five participants, but disagree as to whether, if fewer than five participants are involved, the determination of equivalence must focus upon a headcount of the individuals involved, or may also rely upon other indices of extensiveness such as the magnitude of the harm, the complexity of the planning, or the number of victims. Compare United States v. Carrozzella, 105 F.3d 796, 802-03 (2d Cir.1997) (focusing analysis on the number of individuals involved), with United States v. Dietz, 950 F.2d 50, 53-54 (1st Cir.1991) (permitting consideration of a broad range of indices). In Carrozzella, the Court of Appeals for the Second Circuit explained that the analysis of “otherwise extensive” must focus initially upon the number of participants, and the knowing or unknowing persons, involved in the criminal activity, and then a subsequent determination must be made as to whether the roles and involvement of all those persons constitute the functional equivalent of five “participants” as defined by the Application Notes. See Carrozzella, 105 F.3d at 802-04. This contrasts with the rulings of several other courts of appeals that permit a broader inquiry that examines other characteristics of the criminal activity as well. See United States v. Brockman, 183 F.3d 891, 900 (8th Cir.1999), cert. denied,— U.S.-, 120 S.Ct. 800, 145 L.Ed.2d 674 (2000) (permitting consideration of the number of persons involved and the amount of loss); United States v. Yarnell, 129 F.3d 1127, 1139 (10th Cir.1997) (following Dietz)-, United States v. Tai, 41 F.3d 1170, 1175 (7th Cir.1994) (permitting courts to “to examine factors in addition to a headcount to justify a finding that a given criminal activity is extensive”). For example, the Court of Appeals for the First Circuit, in United States v. Dietz, explained that “the extensiveness of a criminal activity is not necessarily a function of the precise number of persons, criminally culpable, or otherwise, engaged in the activity” and upheld a four-level increase based upon a general conception of extensiveness which included a finding that the activity spanned twelve years, crossed seven states and victimized a number of governmental agencies without an exact determination of the number of individuals involved. See Dietz, 950 F.2d at 53-54. In United States v. Rose, *245 20 F.3d 367, 374 (9th.Cir.1994), the Court of Appeals for the Ninth Circuit explained that “[w]hether criminal activity is ‘otherwise extensive’ depends on such factors as (i) the number of knowing participants and unwitting outsiders; (ii) the number of victims; and (iii) the amount of money fraudulently obtained or laundered.” Id. at 374 (internal citations omitted). In United States v. Bennett, 161 F.3d 171 (3d Cir.1998), the only opinion in which we have previously considered the “otherwise extensive” prong of § 3Bl.l(a), we affirmed the application of § 3Bl.l(a) to enhance the offense level of a defendant convicted of running a large and complex “Ponzi” scheme which the district court found involved two participants and at least thirteen non-participants who assisted Bennett. See id. at 194. Many of these individuals served Bennett by withholding information from investors and legitimizing his activities by preparing reports based upon false information he had provided. See id. Because the evidence of extensiveness was so clear, due to the involvement of at least thirteen others, we did not discuss whether other indices of “extensiveness” could be appropriately considered under § 3B1.1. 14 We take the opportunity to decide this issue here. We will subscribe to the analysis of the Court of Appeals for the Second Circuit as described in Carrozzella. We believe that the focus upon the number and roles of the individuals knowingly, and unknowingly, involved best comports with the text of § 3B1.1, its application notes and commentary, as well as the overall structure of the Sentencing Guidelines. We note that while the Guideline itself provides no clear guidance as to the meaning of “otherwise extensive,” the application notes and the commentary strongly indicate that the focus should be upon the number of persons involved. Although we disagree with the District Court’s decision to count one “non-participant,” we agree with the District Court, which essentially applied the Carrozzella test, that the combination of Helbling, Susman and Sokolic — the participants — together with the countable non-participants, made Hel-bling’s criminal activity “otherwise extensive” for the purposes of § 3B1.1. Application Note 3 specifically includes an example of “otherwise extensive” that is instructive. “In assessing whether an organization is ‘otherwise extensive,’ all persons involved during the course of the entire offense are to be considered. Thus, a fraud that involved only three participants but used the unknowing services of many outsiders could be considered extensive.” U.S.S.G. § 3B1.1, Application Note 3 (emphasis added). The background commentary is consistent with a numerical focus: This section provides a range of adjustments to increase the offense level based upon the size of the criminal organization (ie., the number of participants in the offense) and the degree to which the defendant was responsible for committing the offense.... The Commission’s intent is that this adjustment should increase with both the size of the organization and the degree of the defendant’s responsibility. U.S.S.G. § 3B1.1, Background (emphasis added). The background commentary also explains why the size of the criminal activity is important to the purpose of the provision. It notes that the Sentencing Commission drafted the Guideline’s separate provisions to respond to criminal organizations whose size and structure increases the significance of the defendant’s role. *246 Compare U.S.S.G. § SBl.l(a)-(b), with, § 3Bl.l(c). “In relatively small criminal enterprises that are not otherwise to be considered as extensive in scope or in planning or preparation, the distinction between organization and leadership, and that of management or supervision, is of less significance than in larger enterprises that tend to have dearly delineated divisions of responsibility.” U.S.S.G. § 8B1.1, Background (emphasis added). If the Commission intended the courts to utilize a broader analysis to determine if a criminal activity was “extensive” it could easily have said so. Instead, the notes clearly indicate the Commission meant the courts to consider the role of the defendant and the size of the criminal organization as its way of determining the relative responsibility of the defendant, so as to mete a greater sentence to those who likely received a greater division of the illegal profit, posed a greater danger to the public, and are more likely to commit more crimes in the future. See U.S.S.G. § SB1.1, Background (“This adjustment is included primarily because of concerns about relative responsibility. However, it is also likely that persons who exercise a supervisory or managerial role in the commission of an offense tend to profit more from it and present a greater danger to the public and/or are more likely to recidi-vate.”). Furthermore, limiting our construction of “extensiveness” to a head counting analysis reduces the potential for double counting certain aspects of criminal activity that are considered elsewhere in the scheme of the guidelines, thus helping to maintain the distinct character of various guideline sections. See Carrozzella, 105 F.3d at 802-03. For example, as the court in Car-rozzella explained, in a fraud conviction the base offense level can be increased based on the amount of loss, the extent of planning, and the number of victims. See U.S.S.G. § 2F1.1; Carrozzella, 105 F.3d at 802. Further adjustments may be made to account for the vulnerability of the victims, see U.S.S.G. § 3A1.1, the defendant’s role, see U.S.S.G. § 3B1.2, and abuse of a position of trust. 15 See U.S.S.G. § 3A1.3. If we were to adopt a broad reading of the phrase “otherwise extensive,” so that some of these possible adjustments could be considered in that inquiry as well, a defendant would possibly receive two sentence enhancements for the same attributes of the crime. We avoid this result by focusing upon the number of individuals involved. We disagree with the government that our opinion in United States v. Wong, 3 F.3d 667, 668 (3d Cir.1993), undercuts this rationale. In Wong, we held that where the guidelines happen to direct the sentencing court to increase the defendant’s sentence under two separate sections^ — which respond to different evils— based on different aspects of the same conduct, the district court does not err by “double counting” those aspects in determining, the sentence. See Wong, 3 F.3d at 668; see also United States v. Johnstone, 107 F.3d 200, 211-13 (3d Cir.1997) (rejecting a double counting argument in accordance with Wong); United States v. Maurello, 76 F.3d 1304, 1315-16 (3d Cir.1996) (same). Thus in Wong, we found no error in the district court’s decision to increase the defendant’s offense level for both more than minimal planning, see U.S.S.G. § 2Bl.l(b)(5), and aggravating role under *247 U.S.S.G. § SBl.l(c) — the degree of planning and defendant’s role were different attributes of the same conduct. See Wong, 3 F.3d at 668, 669 n. 4. Here, the guideline focuses on the attribute or aspect of the numerosity of the group that the defendant led or organized. The language of two or more sections may well require consideration of different attributes of the same conduct. This is not an untoward result, but, rather an intended one. 16 Having adopted the test that requires us to focus upon the number and participation of the individuals involved in the criminal activity, we must address how the district courts are to apply it to the situations facing them. Although Application Note 3 supports our mode of analysis, the guidelines are silent as to how the sentencing court actually is to decide which non-participants should be considered, and what combination of participants and countable non-participants is the “equivalent” of five participants. We agree with the court in Carrozzella that not every individual tangentially involved in the criminal activity can fairly be considered in the analysis. See Carrozzella, 105 F.3d at 803. The purpose of the provision would rarely be achieved by counting the unknowing services of some actors in a criminal scenario, a taxicab driver or bank teller, for instance. The court in Carrozzella developed a three step inquiry to help determine which individuals should be counted. 17 See id. at 803-04. This test distinguishes non-participants who should be considered from those who should not be considered on the basis of the defendant’s intent in involving them in the criminal activity and the nature of their role in the offense. 18 Under this test, a *248 sentencing court must first separate out the “participants” as defined by Application Note 1 from other individuals, nonparticipants, who were involved in the criminal activity. “A ‘participant’ is a person who is criminally responsible for the commission of the offense, but need not have been convicted.” U.S.S.G. § 3B1.1, Application Note 1. The defendant may be considered as one of the participants. See Colletti, 984 F.2d at 1346. The court must next determine whether the defendant used each non-participants’ services with specific criminal intent. See Carrozzella, 105 F.3d at 804. Third, the court must determine the extent to which the services of each individual, non-participant, were peculiar and necessary to the criminal scheme. See id. at 804. Utilizing this test, the Court of Appeals for the Second Circuit has upheld the consideration of individuals used by the defendant to legitimize, facilitate or hide the criminal activity. See United States v. Napoli, 179 F.3d 1, 15 (2d Cir.1999) (applying Carrozzella and upholding the finding that the criminal activity was otherwise extensive on the basis of casino personnel who unwittingly helped the defendant launder money); United States v. Nolan, 136 F.3d 265, 273 (2d Cir.1998) (approving the consideration of accountants, bank officers, and lawyers used to perpetuate embezzlement from an ERISA covered pension plan including the filing of incomplete documents). We view this test as helpful in defining which non-participants should be counted, and subscribe to it. After deciding which individuals may be counted, the court must then consider whether the sum of the participants and countable non-participants is the “functional equivalent” of five participants. We note that, at a minimum, a criminal scheme must involve more than one participant in order to be found otherwise extensive; there can be no less than the defendant and one participant the defendant led or organized. See U.S.S.G. § 3B1.1, Application Note 2 (“To qualify for an adjustment under this section, the defendant must have been the organizer, leader, manager, or supervisor of one or more other participants.”). However, a court’s determination as to the functional equivalence does not lend itself to a mere numerical analysis, but will necessarily be guided by the sentencing court’s discretion. In deciding whether the total is the equivalent to five participants, the sentencing court may consider other factors, including the nature of the criminal scheme, to evaluate the relative value of the various countable non-participants. We will apply the test we have crafted to the case at hand. We note that Helbling urged us to adopt the Second Circuit’s test and does not contest the District Court’s determination that Helbling, Susman and Sokolic should be counted as participants. Therefore, the analysis rests upon the validity of the court’s decision to count six non-participants — John Grikis, Arnold Kaminer, Bruce Katz, Donald Mayle, Ronald Santora and Thomas Taylor — and the conclusion that the combination of these three participants and these six non-participant individuals is the equivalent of five participants. Helbling disputes the consideration of each individual. He argues that Katz and Grikis should not be considered because both merely transferred funds at Helbling’s instruction. Helbling argues that Mayle, Santora and Taylor should not be considered because they all acted as part of a bankruptcy fraud which was not peculiar and necessary to the criminal activities that supported Helbling’s conviction, and, finally, Helbling argues that Kaminer should not be considered because the letter Kaminer *249 wrote at Helbling’s behest, falsely saying that they had a conversation about an ESOP conversion, was never sent. We agree with the District Court’s conclusion in all but Kaminer’s case, and find that the District Court properly concluded that the combination of participants and countable non-participants makes Helbling’s scheme “otherwise extensive.” In arguing that Katz and Grikis should not be considered, Helbling unduly narrows the scope of our inquiry. While our test, which considers both the intent with which the defendant directed the individual’s involvement and the extent to which the services of the individual were peculiar and necessary to the criminal scheme, instructs the sentencing court to disregard individuals who are essentially functionaries, the activities of Katz and Grikis are distinguishable from those of other banking employees who, for example, might have simply moved the money for Helbling. See generally Napoli, 179 F.3d at 15 (applying Carrozzella and upholding the finding that the criminal activity was otherwise extensive on the basis of casino personnel who unwittingly helped the defendant launder money). Helbling used the services of Grikis and Katz to convert an account held by NatWest as a trustee owing fiduciary duties, into an account that from which he could withdraw funds, borrow against and which he could eventually liquidate. We also believe that the District Court was correct to count the activities of Mayle, Santora and Thomas. Each individual helped Helbling hide his criminal activities. See Bennett, 161 F.3d at 194 (counting non-participants who helped hide and legitimize the criminal activity). At the direction of Helbling, Mayle attested to Helbling’s signatures on the ESOP conversion documents and hid the influx of $350,000 from the Bankruptcy Court. Similarly, Thomas, at Helbling’s direction, submitted false documents to the Small Business Administration describing the $350,000 as a personal cash infusion from Helbling. Lastly, at Helbling’s direction, Santora, Helbling’s bankruptcy lawyer, submitted false documents given to him by Helbling to the bankruptcy court. We reject Helbling’s argument that Mayle, Santora, and Thomas ought not be considered because their activities were related to crimes for which Helbling was not indicted, and, therefore, their activities were not peculiar and necessary to Hel-bling’s criminal behavior. 19 These activities aided Helbling’s embezzlement activities by concealing Helbling’s withdrawals from the Bankruptcy Court, from the Small Business Administration, and from the plan participants. We agree, however, that the District Court erred in counting Kaminer. The District Court found that Helbling directed Kaminer to write a letter in December 1991 stating falsely that he had a conversation with Helbling about an ESOP conversion in February 1991. While we disagree with the Court’s factual conclusion, that is not the focal point of our inquiry. 20 As far as can be determined *250 from the record, Helbling never used the letter, or even placed it in a situation where it might be used, and, therefore, we cannot say that Kaminer’s conduct, innocuous in itself, was peculiar and necessary to Helbling’s criminal activities. We will therefore affirm the District Court’s decision to enhance Helbling’s sentence under § 3B1.1 on the basis that he was the organizer or leader of a criminal activity that was otherwise extensive. Helbling’s criminal activity involved three criminally responsible participants and five other non-participant individuals whose involvement Helbling directed and whose actions were peculiar and necessary to the furtherance of Helbling’s efforts. Given the nature of Helbling’s criminal conduct, the combination of these participants and non-participants was properly held by the District Court to be the functional equivalent of five participants. B. Obstruction of Justice Helbling contends that the District Court did not find that Helbling had the requisite willful intent to obstruct justice as is required in order to increase Helbling’s offense level under § 3C1.1 of the Guidelines. 21 See U.S.S.G. § 3C1.1; United States v. Belletiere, 971 F.2d 961, 965 (3d Cir.1992) (“[T]he government bears the burden of proving by a preponderance of the evidence that the defendant willfully obstructed or impeded, or willfully attempted to instruct or impede, the administration of justice.”). We disagree. The District Court applied the two-level enhancement on the basis of two acts by Helbling which it found were “clear-cut attempts to influence” testimony and which it concluded were covered under Application Note 4. Application Note 4 lists as one type of applicable behavior, “(a) threatening, intimidating, or otherwise unlawfully influencing a co-defendant, witness, or juror, directly or indirectly, or attempting to do so.” U.S.S.G. § 3C1.1, Application Note 4(a). The first act relied upon by the District Court was a letter written by Helbling to his two attorneys, Susman and Sokolic, in May of 1994. In the letter Helbling threatened to file malpractice actions against the lawyers and file complaints with the bar association if they refused to stand behind the documents they created. Helbling, however, contends that the letter evidences his frustration rather than an attempt to intimidate. The second act was a conversation between Helbling and Donald Mayle in July of 1996 during which Helbling pressured Mayle to say his signature on the board of directors’ consent was genuine. 22 Helbling argues that the discussion of Mayle’s signature was innocuous and that the District Court failed to put it in context. We disagree on both scores. We review the District Court’s factual findings under the clearly erroneous standard. See United States v. Boggi, 74 F.3d 470, 478 (3d Cir.1996) (citing United States v. Cu-sumano, 943 F.2d 305, 315 (3d Cir.1991)). While we recognize that the Helbling’s letter varies in tone and could be subject to differing interpretations, the letter de *251 manded that the lawyers stand behind documents found by the jury to be false, and threatened adverse consequences if they did not do so. With respect to the conversations with Mayle, the inference drawn from the conversations is that Helbling desired Mayle to say that he signed the board of directors’ consent and was urging Mayle to lie. We conclude that the District Court’s finding that these acts represented attempts to influence testimony was not clearly erroneous. C. Extreme Psychological Injury to the Victims Helbling also contends that the District Court abused its discretion when it departed upward two-levels from the applicable guideline level on the basis of extreme psychological damage to the vie-, tims because there was insufficient evidence of psychological harm to justify the departure. See U.S.S.G. § 5K2.3 (Extreme Psychological Injury to the Victims) (Policy Statement); United States v. Neadle, 72 F.3d 1104, 1111-12 (3d Cir.1995); United States v. Astorri, 923 F.2d 1052, 1058 (3d Cir.1991). Section 5K2.3 permits a sentencing court to sentence above the guideline sentence level “[i]f a victim or victims suffered psychological injury much more serious than that normally resulting from commission of the offense ...” The section continues to explain that: Normally, psychological injury would be sufficiently severe to warrant application of this adjustment only when there is a substantial impairment of the intellectual, psychological, emotional, or behavioral functioning of a victim, when the impairment is likely to be of an extended or continuous duration, and when the impairment manifests itself by physical or psychological symptoms or by changes in behavior patterns. The court should consider the extent to which such harm was likely, given the nature of the defendant’s conduct. U.S.S.G. § 5K2.3. In Astorri, we upheld a departure for extreme psychological harm when the District Court found, both through testimony and its own observation, that the victims’ health suffered after they lost their life savings. See Astorri, 923 F.2d at 1059. We review a district court’s decision to depart from the guidelines under an abuse of discretion standard. See United States v. Jacobs, 167 F.3d 792, 798 (3d Cir.1999) (citing United States v. Baird, 109 F.3d 856, 862 (3d Cir.1997)). We have noted that “[i]f there is any place in sentencing guidelines analysis where a fact-finder is to be given considerable deference, it is here where the district court is to be called upon to assess the psychological impact upon victims.” See Astorri, 923 F.2d at 1058. However, in Jacobs, a case addressing the psychological effects of an aggravated assault, we explained that a finding that the victim’s psychological injury was “much more serious than that normally resulting from the commission of the crime ... is a prerequisite for a departure under § 5K2.3.” Jacobs, 167 F.3d at 799. Relying principally on Astorri, a fraud case, the District Court found that a two level departure was warranted under § 5K2.3 because of the effects of Hel-bling’s criminal activities and harassment. Helbling argues that the District Court’s factual findings were insufficient for the departure. The District Court found: [T]he age of Helbling’s victims did not facilitate the commission of his crimes, it did heighten their impact.... The Government has convincingly detailed the emotional and psychological costs of surviving Helbling’s crimes and resisting his harassment. These costs include the humiliation of being forced to seek work at an advanced age and rely on help from family members, the trauma that comes with losing one’s savings, and the psychological damage resulting from resisting slurs, threats, frivolous lawsuits, and pressure from the tax authorities. The evidence fully justifies this two level upward departure. We believe that the District Court’s findings are clearly supported by the record *252 and that the record itself supports the determination that Helbling caused psychological injury “much more serious” than would normally result from his type of fraudulent activity. In fact, as the District Court recognized, the record, in the form of the Presentence Report and the testimony of some of the victims at trial, contains the same type of evidence of individual loss and resulting medical complications we found sufficient in Astorri. Accordingly, we find that the District Court did not abuse its discretion in departing upward under § 5K2.3. VI. • For the reasons above, we will affirm the judgment of conviction and sentence. * and to present for investigation by the FBI and the Department of Labor those allegations which I have set forth in my letter to the U.S. Attorneys Office dated July 13, 1996, specifically paragraph # 3. Footnotes [1] We have jurisdiction over Helbling’s appeal over his conviction under 28 U.S.C. § 1291 and his sentence under 18 U.S.C. § 3742. We gave Helbling leave to file a supplemental pro se brief and a reply brief and leave to the government to respond. We have considered both his counseled and pro se submissions. We have denied Helbling's motions to file further supplemental briefs and appendices. [2] The District Court granted Helbling’s motion to exclude the documents charged in the indictment. The United States filed a superseding indictment on May 20, 1998. [3] The ESOP conversion was limited to $500,-000 to avoid the need for filing with the SEC. [4] The relevant portion of the agreement, with the added rider, stated: In order to allow ample time for me to confer with representatives of the U.S. Attorney's Office,* I hereby knowingly, voluntarily and expressly waive the right of defense provided by any statute of limitations with regard to the above criminal statutes of the United States Code based upon the failure of the U.S. Attorney's Office to obtain an indictment for violations of those during the period from July 22, 1996 through December 31, 1996. [5] The parties do not direct us to any controlling authority where we have actually considered whether the government had breached a statute of limitations waiver in a criminal case. Helbling argues that we should use the same standard here as we use to determine whether a plea agreement has been complied with. See, e.g., United States v. Nolan-Cooper, 155 F.3d 221, 236 (3d Cir.1998) (considering “whether the government’s conduct is inconsistent with what was reasonably understood by the defendant when entering the plea of guilty.”); see also United States v. Levine, 658 F.2d 113, 124, 124 n. 17 (3d Cir.1981) (noting that waivers of statutes of limitations have been viewed as being similar to guilty pleas). We decide neither that this standard is necessarily applicable to waivers, nor that defendants who sign waiver agreements should always receive the same protections as those who sign plea agreements. Rather, even assuming the standard does apply, Helbling cannot succeed. [6] Considering Helbling's allegations that he was coerced into signing the waiver by threats and by misrepresentations, and that he was not represented during the negotiations, the District Court determined that the government was ready to indict Helbling when the waiver was signed so it did not coerce him into signing the waiver by falsely threatening immediate indictment, and that Helbling realized, or should have realized, that the waiver would provide the government more time to investigate, as well as provide him more time to prepare, and, therefore, the government did not misrepresent the effect of the waiver. In his pro se brief, Helbling alleges that the prosecutor conspired with Susman’s lawyer to coerce Helbling to sign the waiver, and states that Department of Labor investigator Daphne Rich testified that Susman and Sokolic had agreed to cooperate with the government before Helbling signed his statute of limitations waiver. However, Susman’s lawyer’s testified at the hearing, and the District Court rejected Helbling’s argument that he was coerced by him. The District Court also found that Susman’s plea agreement came months after Helbling signed the waiver. The Court’s factual findings are not clearly erroneous, and we agree with its legal conclusions. Therefore, to the extent that Helbling asserts these arguments as separate grounds to invalidate the waiver, we reject them. We find no evidence in the record to support Helbling’s other allegations. [7] We note at the outset of this discussion that Helbling also alleges in his pro se brief that the government withheld exculpatory evidence and suborned false testimony in return for plea agreements. Since these arguments were not raised to the District Court, we review for plain error. To the extent that Helbling's argument rests on United States v. Singleton, 144 F.3d 1343 (10th Cir.1998), rev'd en banc, 165 F.3d 1297 (10th Cir.), cert. denied, - U.S. -, 119 S.Ct. 2371, 144 L.Ed.2d 775 (1999), we note that this Court has now rejected the holding of the first Singleton decision in United States v. Hunte, 193 F.3d 173, 174 (3d Cir.1999), decided after briefing. Furthermore, Heibling had the opportunity to cross-examine the witnesses at trial, and we find no factual support for his other allegations. [8] Helbling raised this issue at trial as a Rule 29 motion to dismiss. The District Court denied the motion. [9] Helbling’s counseled brief argues that the government failed to submit sufficient evidence. Helbling raised and developed his second theory in his subsequent pro se brief. The government points out that he did not raise his second argument in front of the District Court. However, we have explained that since the alleged error directly relates to his criminal responsibility, such error would be plain error. See United States v. Cusumano, 943 F.2d 305, 309 (3d Cir.1991) (considering whether a plan was covered by ERISA for the first time on appeal). [10] The government intended to have Debra Golding and Kevin Long testify, and presented a letter to Helbling outlining the general areas about which they would testify. Hel-bling moved to preclude their testimony, asserting that the government’s letter failed to comply with Federal Rule of Criminal Procedure 16(a)(1)(e) which requires the government to “describe the witnesses’ opinions, the bases and the reasons for those opinions, and the witnesses' qualifications.” Rejecting the government’s contention that Golding and Long were lay witnesses not covered by 16(a)(1)(e), the District Court granted Hel-bling’s motion. [11] Early in opening arguments, the prosecutor told the jury that the defendant had "[u]gly values: Dishonesty, disloyalty, deceit, and above all, greed. To be sure, ladies and gentlemen, the Defendant, William Helbling is a thief. He lied to these men, he stole their retirement moneys boldly and brazenly.” The prosecutor also called Helbling a looter while pointing a finger in his face, and said, while discussing the companies’ financial problems, that "the one source [of money] was what he called his 'secret fund,’ his slush find, his private kitty. And he stole the money, he stole the retirement money.” The prosecutor also called Susman and Sokolic “unscrupulous,” and explained that they assisted Hel-bling in “covering up his fraud, his looting of moneys.” After a defense counsel objection, the prosecutor completed her opening by stating that: "We believe that you will find the Defendant guilty as charged.” The District Court directed the jury to disregard the comment. During the prosecutor's opening, the Court sustained a number of defense counsel objections and brought counsel to side bar where the judge admonished the prosecutor. At the close of the prosecutor’s opening, the judge reminded the juiy, as he had already explained during his preliminary instructions, that the opening statement was not to be considered as evidence. Before closing arguments began, the District Court again reminded the jurors, as part of the jury instructions, that they must not consider the opening and closing arguments as evidence. In the early portion of her closing argument, the prosecutor referred to “a greedy person’s plan,” the "Defendant’s plun *241 der and thievery," and the Defendant’s "track record of lying, cheating and deceit." The prosecutor also called evidence that the letter from Barry Penn to Oppenheimer was forged "uncontroverted.” In rebuttal, the prosecutor again referred to the same evidence as "uncontroverted," and further characterized the defendant as "arrogant” and "despicable” which the court instructed the jury to disregard. Upon defense counsel objection to the second reference to uncontroverted evidence, the District Court told the prosecutor to avoid such references. Later at side bar, the Court asked defense counsel what counsel wanted the court to do to neutralize the statement. Counsel asked the Court to "leave it, please.” [12] Helbling’s reliance on the terms of Sus-man’s and Sokolic’s plea agreements as proof of their equal culpability is misplaced. The Guideline clearly contemplates that a defendant may be the leader or organizer of another criminally responsible individual; the amount of loss stipulated to in a plea agree *244 ment is surely not the only factor that bears on the relationship between those individuals. [13] Katora is simply not applicable to this case. In Katora, we explained that a defendant cannot be assessed an offense level enhancement under S 3B1.1 where the district court determined that the two participants were equally culpable and did not organize, lead, manage or supervise a third participant. See Katora, 981 F.2d at 1405. In this case, however, the District Court determined that Helbling alone organized and led the criminal activity which involved two other participants. [14] Since we did not engage in a broader analysis, we place no significance upon citations in Bennett to United States v. D’Andrea, 107 F.3d 949, 957 (1st Cir.1997), and United States v. Dietz, 950 F.2d at 53, decisions that permit the district court to consider the "width, breadth, scope, complexity, and duration of the scheme.” See Bennett, 161 F.3d at 194 n. 14. These citations were placed in a footnote attached to the last part of the analysis and are no more than dicta. [15] Similarly, following a broad reading of “otherwise extensive” would permit a court to consider the amount of drugs sold in both the base offense level and as an element leading to the determination that the activity was extensive. See United States v. Rodriguez, 981 F.2d 1199, 1200, 1200 n. 3 (11th Cir.1993) (per curiam) (finding no problem with considering drug quantities as evidence of extensiveness). Not surprisingly, courts following a broad test for extensiveness repeatedly encounter, though they have rejected them, arguments complaining of double counting relating to the finding of extensiveness in fraud cases. See, e.g., Brockman, 183 F.3d at 900 n. 8 (rejecting argument based upon double counting by way of § 2F1.1 and § 3B1.1.). [16] Application Note 4, of § 1B1.1, was amended effective November 1, 1993, to state that "[a]bsent an instruction to the contrary, the adjustments from different guideline sections are applied cumulatively (added together). For example, the adjustments from § 2F 1.1 (b)(2) (more than minimal planning) and § 3B1.1 (Aggravating Role) are applied cumulatively.” U.S.S.G. § 1B1.1, Application Note 4 (amended November 1, 1993, Amendment 497). We believe that this amendment, which followed a few months after our decision in Wong, reinforces the decision in Wong and our reasoning here. We read the Application Note to explain that a defendant cannot attack a sentence by arguing that the district court impermissibly considered evidence of more than minimal planning when both adjusting under § 2F 1.1 (b)(2) and when engaged in finding the defendant to be a leader, organizer, manager or supervisor as specifically directed by § 3B1.1, Application Note 4. Indeed, prior to the amendment, the Sixth Circuit Court of Appeals, in United States v. Romano, 970 F.2d 164 (6th Cir.1992), found that because it viewed consideration of more than minimal planning as being required by § 3Bl.l's application notes an enhancement under both provisions was not permitted. See Romano, 970 F.2d at 167. In United States v. Cobleigh, 75 F.3d 242 (6th Cir.1996), the court recognized that Romano had been effectively overruled by the amendment. See Cobleigh, 75 F.3d at 251. Nothing about the amendment leads us to believe that it meant to implicate the breadth of analysis required or permitted under the second part of § 3Bl.l(a) — the determination of the size of the criminal activity. [17] The Second Circuit wrote in full: [W]e believe that the following must be determined by the sentencing court: (i) the number of knowing participants; (ii) the number of unknowing participants whose activities were organized or led by the defendant with specific criminal intent; (iii) the extent to which the services of the unknowing participants were peculiar and necessary to the criminal scheme. Carrozzella, 105 F.3d at 803-04. [18] We note that the Introductory Commentary to Part B, which contains § 3B1.1, was amended effective November 1, 1990 to state that: “The determination of a defendant's role in the offense is to be made on the basis of all conduct within the scope of § lB1.3(a)(l-4), and not solely on the basis of elements and acts cited in the count of conviction'.” Prior to the amendment, we held that § IB 1.3 did not apply to the analysis under 3B1.1. See United States v. Murillo, 933 F.2d 195, 199 (3d Cir.1991); see also United States v. Pollen, 978 F.2d 78, 89-90(3d Cir.1992). In these cases, we focused upon the language of the provision — the offense — to conclude that consideration of who could be considered a participant must be limited to those involved in the offense charged and all conduct in furtherance of the offence of con *248 viction. See Murillo, 933 F.2d at 199. In both cases, we noted that our decisions did not address the effect of the amendment on our analysis. See id. at 198 n. 1; see also Pollen, 978 F.2d at 89 n. 24. We do not decide today whether the scope of § IB 1.3 might affect our analysis. [19] In the proceedings in front of the District Court and again on appeal, Helbling considers Mayle, Santora and Thomas together because they provided services relating to bankruptcy fraud. See Appellant’s Br. at 46. Our review of the trial testimony indicates that the filings signed by Taylor were not part of the bankruptcy case. Instead, they were submitted to the United States Small Business Administration. However, we consider the three individuals together because Helbling's arguments why each person should not be considered are identical. [20] The District Court found that the letter was false because the conversation never took place. We find this finding clearly erroneous because Kaminer testified at trial that the February 1991 conversation did occur and that the letter merely confirmed that fact. As we find that Kaminer should not be counted because the letter was never used in furtherance of Helbling’s activities, we need not determine whether submitting a truthful letter in an effort to support a false conclusion is the type of involvement that should be counted under § 3B1.1. [21] Section 3C1.1 of the guidelines states: If (A) the defendant willfully obstructed or impeded, or attempted to obstruct or impede, the administration of justice during the course of the investigation, prosecution, or sentencing of the instant offense of conviction, and (B) the obstructive conduct related to (i) the defendant’s offense of conviction and any relevant conduct; or (ii) closely related offense, increase the offense level by 2 levels. U.S.S.G. § 3C1.1. [22] Two conversations between Helbling and Mayle are reproduced in the record. The first conversation took place on July 19, 1996. The second conversation took place on July 20, 1996. The District Court's opinion does not make it clear whether it is relying upon both conversations or just one. The opinion quotes from the July 19, 1996 conversation. Helbling suggests that part of the District Court’s error was failing to consider both conversations. We believe, however, that both conversations indicate the same desire to influence Mayle. The issue is therefore moot.