United States v. Henry Achiekwelu

Affirmed by published opinion. Judge MURNAGHAN wrote the opinion, in which Judge LUTTIG and Senior Judge BLACK joined. OPINION MURNAGHAN, Circuit Judge: This case is interesting and somewhat unusual, involving, as it does, the activities of someone who tried to defraud and was himself successfully defrauded by someone else. The criminal proceedings were only directed at the one whose plan produced, from his point of view, favorable results. A federal jury convicted Defendan1>-Appellant Henry Aehiekwelu on nine counts of wire fraud, in violation of 18 U.S.C.A. §§ 2, 1343 (West 1969 & Supp.1996). Aehiekwelu contends that the district court erroneously excluded one of his exhibits because the government failed to object to its admission in a timely manner. Aehiekwelu also contends that the district court committed several sentencing errors. For the reasons stated below, we affirm. I. In September 1992, Dr. Jai Gupta received a letter in Virginia on behalf of his company, EER Systems, Inc. (“EER”), from a Nigerian man who claimed to be Chief Johnny Okolie. Okolie claimed that he had received Gupta’s name from the Chamber of Commerce in Lagos, Nigeria, and the letter attempted to solicit Gupta’s participation in a business opportunity. Okolie claimed that he represented a Nigerian company, Macbos Holdings (“Macbos”), that had installed lighting at Maiduguri Airport in Nigeria for the Nigerian Ministry of Civil Aviation (the “Aviation Ministry”) in 1982. Okolie stated that the Nigerian government had not paid the $28.5 million that it owed on the lighting contract because it claimed that Macbos had overbilled the Aviation Ministry. According to Okolie, the Nigerian government had then changed its position and had agreed to pay $18.5 million now and the remaining $10 million later. Okolie stated that he wanted to deposit the $28.5 million in a bank account outside Nigeria and that he was looking for a “trustworthy” person with whom to deposit the money. Thus, Okolie stated that Macbos wanted to deposit the money in EER’s bank account and that Macbos would pay Gupta a fifteen percent commission on the $28.5 million, ie., $4,275,000. Gupta and Okolie communicated by phone and by fax throughout September and October 1992. At Okolie’s request, Gupta attempted to defraud the Nigerian government by sending Okolie an EER invoice that falsely represented that EER had performed lighting work for the Nigerian Aviation Ministry. EER had in fact done no such work. Also at Okolie’s request, Gupta travelled to Lagos, Nigeria on October 18, 1992 to finalize the arrangements for the deposit of the money into EER’s account. In Lagos, Okolie introduced Gupta to Fred Kachi, who stated that he was a deputy governor of the Central Bank of Nigeria (the “Central Bank”), and Aehiekwelu, who stated that he was a consultant to a government agency, the Nigerian Ministry of Finance (the “Finance Ministry”). At that meeting, the three Nigerians and Gupta discussed the proposed $28.5 million transfer to EER. The next day, Gupta met with Kachi and Okolie. Kachi showed Gupta a document that purported to be from the Central Bank. The document stated that the Nigerian gov- *751 eminent had authorized the Central Bank to pay $28.5 million to EER. Okolie and Kachi told Gupta, however, that EER had to pay a $712,500 “advance income tax” on the $28.5 million to the Nigerian government before the Central Bank would transfer the money. At that point, the fraud against Gupta began. Okolie and Achiekwelu later took Gupta to an office building that they claimed was an office of the Central Bank. A man who claimed to be a Central Bank official told Gupta that he had to pay the advance tax. Gupta thereafter transferred $712,500 to a Lagos bank for the tax payment. After his return to Virginia, Gupta received numerous telephone calls from Achiekwelu and Kachi. During one of those conversations, Kachi and Achiekwelu told Gupta that they had paid the Nigerian taxes with the money that Gupta had transferred. On October 27, 1992, Gupta received a fax, purportedly from the Central Bank, that confirmed that the taxes had been paid to the Nigerian government. The fax stated that the Central Bank would transfer the full $28.5 million to EER’s bank account as soon as the Bank received a “certificate of completion.” That same day, Achiekwelu and Kachi called Gupta and told him that he would have to pay $1.1 million to certain unidentified subcontractors in order to obtain the “certificate of completion.” On November 2, 1992, Gupta transferred $1.1 million to a London bank. In November 1992, Achiekwelu and Kachi told Gupta that he had to send $500,000 to Kachi’s supervisor, the purported governor of the Central Bank, in order to obtain a “release order” that would authorize the transfer of the $28.5 million to EER. Gupta sent the $500,000 on November 13, 1992. That same day, Gupta received a purported receipt from the Nigerian government for the income tax payment and a purported letter from the Central Bank that indicated that Gupta had complied with all of the necessary conditions for the transfer of the $28.5 million. By December 1992, Achiekwelu and Kachi began to tell Gupta that he would not receive the $28.5 million, or any of the money that he had already sent, if he did not send the additional money that they requested. In December 1992, Gupta made two transfers of $206,000 each to Achiekwelu’s accounts in Lagos. He also transferred $500,000 to Achiekwelu’s bank account in London after he received another letter from the Central Bank that promised to transfer the $28.5 million in return for the payment. In January 1993, following additional promises and reassurances from Achiekwelu and from a man who claimed to be an officer of a London bank, Gupta sent an additional $500,000 to an account in London. He sent an additional $450,000 later that month to an account in California pursuant to a fax signed by Achiekwelu and Kachi. Altogether, Gupta transferred almost $4.2 million to accounts designated or controlled by Achiekwelu and his associates in Nigeria, England, and California. Neither Gupta nor EER Systems ever received any of the $28.5 million. Thus, instead of receiving the $4,275,000 commission that Okolie had originally promised, Gupta paid out almost $4.2 million. On September 14,1994, a grand jury in the Eastern District of Virginia indicted Achiekwelu, Okolie, and Kachi on nine counts of wire fraud, in violation of 18 U.S.C.A. §§ 2, 1343 (West 1969 & Supp.1996). Achiekwelu was arrested in Switzerland shortly thereafter, and the United States government commenced extradition proceedings. On February 21, 1995, Achiekwelu waived extradition, and he then travelled to Virginia for trial. 1 A jury trial began on April 24, 1995. At the trial, the government presented the testimony of Richard Hart, a California businessman, pursuant to Federal Rule of Evidence 404(b). 2 Hart testified that in 1994, Achiekwelu and others defrauded him and his corn *752 pany, Xenotech, Inc., of over $400,000 through promises and representations similar to those made to Gupta. During the trial, Achiekwelu claimed that he only received $600,000 of the $4.2 million fraud proceeds. He argued that he merely acted as a conduit, that he received the money as a fee for distributing Gupta’s money, and that he did not knowingly participate in the fraud. In support of that theory, Achiekwelu presented the testimony of a retired Federal Bureau of Investigation (“FBI”) handwriting expert. The FBI expert testified that he had examined two sets of exemplars: the faxed documents that Gupta had received and a set of genuine documents that Achiekwelu provided. He further testified that, in his opinion, the same person probably had not signed the two sets of documents. Achiekwelu also introduced Defense Exhibit 3 (“DX3”). DX3 displayed a contract between Achiekwelu, Kachi, and Okolie. The DX3 contract confirmed that Achiekwelu had received $2,012,000 in his bank accounts from Gupta, and it provided that Achiekwelu would receive a five percent fee for the use of his bank accounts. The district court conditionally admitted DX3, along with all of the other exhibits submitted by both parties, at the beginning of the trial. Achiekwelu’s attorney referred to DX3 throughout the trial to support his contention that Achiekwelu did not actively participate in the fraudulent scheme. After the closing arguments, the government objected to the admission of DX3 for the first time on the ground that the exhibit lacked foundation. The trial court sustained the government’s objection and excluded DX3. On April 25, 1995, the jury convicted Achiekwelu on all nine counts of the indictment. The government’s Presentence Investigation Report (“PSR”) calculated Achiekwelu’s offense level pursuant to United States Sentencing Guidelines Manual (“U.S.S.G.”) § 2F1.1 (1994), the guideline that covers fraud and deceit. Section 2F1.1 provides for a base offense level of 6. The PSR added 13 levels for a loss of more than $2,500,000 and 2 additional levels for “more than nfinimal planning.” See U.S.S.G. § 2Fl.l(b)(l)(N), (b)(2)(A). The PSR thus calculated an offense level of 21, a criminal history category of I, and a corresponding sentencing range of 37-46 months. However, at the sentencing hearing on August 11,1995, the district court sua sponte directed the government to prepare a brief addressing whether an upward departure was warranted. After the government responded, the district court held a second sentencing hearing on August 28, 1995. The district court issued its final sentencing memorandum on October 26, 1995. The district court ultimately granted the government’s motion for a two-level enhancement pursuant to U.S.S.G. § 2Fl.l(b)(3)(A) on the ground that Achiekwelu misrepresented that he was acting on behalf of a government agency. Alternatively, the district court imposed a two-level upward departure pursuant to U.S.S.G. § 5K2.0 on the basis of the exceptional complexity of the fraudulent scheme. Thus, the district court calculated Achiekwelu’s offense level at 23, either because he misrepresented that he was acting as a government agent or because of the complexity of the fraud. The district court also departed upward to criminal history category II pursuant to U.S.S.G. § 4A1.3(e) on the basis of Achiekwelu’s similar unconvicted conduct in the fraud perpetrated on Hart. Therefore, based on an offense level of 23 and a criminal history category of II, the district court concluded that the applicable sentencing range was 51 to 63 months imprisonment. The court ultimately sentenced Achiekwelu to a 51 month term of imprisonment. The district court also noted that the sentence was independently supportable on the basis of either of the two grounds for an upward departure. Specifically, the district court noted that if it had departed only on the complexity of Aehiekwelu’s fraud, the court would have calculated Achiekwelu’s offense level at 23 and his criminal history category at I, resulting in a sentencing range of 46 to 57 months. The district court further stated that if it had departed only on the complexity of Achiekwelu’s fraud, it still “would impose a sentence on defendant of 51 months of incarceration.” *753 II. ■ Achiekwelu has first contended that the district court erred when it excluded DX3 after closing arguments because the government had failed to object to its admission in a timely manner. We give “substantial deference” to a district court’s decision to exclude evidence, and we will not reverse the district court’s decision “absent a clear abuse of discretion.” United States v. Moore, 27 F.3d 969, 974 (4th Cir.1994). Furthermore, we will find that a district court abused its discretion regarding evidentiary rulings only if the district court “acted ‘arbitrarily or irrationally.’” Id. (quoting United States v. Ham, 998 F.2d 1247, 1252 (4th Cir.1993)). Before the trial began in the instant ease, Achiekwelu’s attorney filed an exhibit list and copies of its exhibits with the district court, and he also provided copies to the government. “ The district court conditionally admitted DX3, along with all of the other exhibits submitted by both parties, at the beginning of the trial. In his opening argument, Achiekwelu’s attorney repeatedly referred to DX3. Before the defense began its case, the government objected to one of Achiekwelu’s exhibits, but it did not object to DX3. In accordance with a procedure that both parties had agreed to, Aehiekwelu’s attorney passed copies of all of its exhibits to the jury at the start of its case. The government still did not object to DX3. During his closing arguments, Achiekwelu’s attorney relied extensively on DX3 and mentioned it four times. Each time, he referred to DX3 to support his contention that Achiekwelu was merely a conduit to distribute the illegal proceeds to the other conspirators and that Achiekwelu did not actively participate in the fraudulent scheme. After the closing arguments, the government for the first time objected to the admission of DX3 on the ground that it had riot been authenticated and lacked foundation. The government explained that it had not previously objected because it thought that Achiekwelu would testify and authenticate DX3. The government did not, however, explain why it did not object before closing arguments, after the defense rested, when it knew that- Achiekwelu would not testify. Nonetheless, the district court excluded DX3 over Achiekwelu’s objection. Achiekwelu essentially concedes that he failed to authenticate DX3. However, his contention is that, regardless of whether the district court properly admitted the exhibit, it had no discretion to exclude DX3 after closing arguments because the government had failed to object to its admission in a timely manner. Achiekwelu argues that, at the latest, the government should have objected when the defense rested and the government knew that Achiekwelu would not testify and authenticate the exhibit. Since the government failed to object at that time, Achiekwelu argues that the government “waived” its objection. However, the cases and rules that Achiekwelu cites in support of his position are inapposite. He first points to Federal Rule of Criminal Procedure 51, which requires a party to “make[ ] known to the court the action which that party desires the court to take or that party’s objection to the action of the court and the grounds therefor” “at the time the ruling or order of the court is made or sought.” Fed.R.Crim.P. 51. He also points to Federal Rule of Evidence 103, which similarly provides that a party must make “a timely objection or motion to strike.” Fed. R.Evid. 103(a)(1). Thé cases that Achiekwelu cites all construe Rule 103 and hold that an appellate court will not review a district court’s admission or exclusion of evidence if the appellant failed to object below in a timely manner. See United States v. Ruffin, 40 F.3d 1296, 1298-99 (D.C.Cir.1994), cert. denied, — U.S.-, 115 S.Ct. 1716, 131 L.Ed.2d 575 (1995) (holding that the defendant waived his objection on appeal because he failed to renew his objection to testimony that the district court conditionally admitted); United States v. Benavente Gomez, 921 F.2d 378, 385-86 (1st Cir.1990) (holding that the defendant waived his objection on appeal to the admission of hearsay evidence because he did not assert his objection below until both sides had rested); United States v. Dougherty, 895 F.2d 399, 403-04 (7th Cir.1990) (holding that the defendant waived his objection on appeal to *754 evidence that the district court conditionally-admitted because he did not renew his objection at the close of the evidence); United States v. Gibbs, 739 F.2d 838, 847-50 (3d Cir.1984) (en banc) (refusing to review the defendant’s claim that the district court improperly admitted evidence because the defendant did not raise his objection below until after the government had rested). Thus, the rules and eases that Achiekwelu cites provide that a party must object timely in order to preserve the issue for appellate review. Achiekwelu does not, however, contend, as those cases suggest, that the Fourth Circuit lacks the power to review the district court’s original admission of DX3 because the government failed to object below in a timely fashion. Rather, he contends that the district court lacked any power to exclude the evidence, having admitted it, because the government failed to object to its admission in a timely fashion. The rules and eases that he has cited simply do not address that issue. They only address the power of an appellate court to review the admission or exclusion of evidence; they do not address the power of a district court to exclude evidence in the first instance after a late objection. Achiekwelu fails to cite any cases that address the issue at hand. Doubtless the government would have been well advised to have objected earlier, at the close of the defendant’s case. If the district court had excluded the evidence earlier, then Achiekwelu’s attorney could not have relied so heavily on DX3 in his closing arguments. As Achiekwelu argues, the district court’s exclusion after closing arguments might have “left a hole” in his case. Nonetheless, as noted above, we must afford “substantial deference” to the district court’s decision to exclude evidence. Moore, 27 F.3d at 974. Even when a district court admits evidence without objection, the district court has the discretion to grant a subsequent motion made after the close of the evidence to exclude the evidence. See Belmont Indus., Inc. v. Bethlehem Steel Corp., 512 F.2d 434, 437 (3d Cir.1975). Although the district court did not follow the best procedural route in the instant case, its action was not arbitrary or irrational. It, in the end, properly excluded evidence that had not been authenticated and that the jury therefore should not have seen. We therefore affirm Achiekwelu’s conviction. III. Achiekwelu next contended that the district court erred when it granted the government’s motion for a two-level enhancement pursuant to U.S.S.G. § 2Fl.l(b)(3)(A) on the ground that Achiekwelu misrepresented that he was acting on behalf of a government agency. Section 2Fl.l(b)(3)(A) provides for a two-level increase “[i]f the offense involved ... a misrepresentation that the defendant was acting on behalf of a charitable, educational, religious or political organization, or a government agency.” U.S.S.G. § 2Fl.l(b)(3)(A) (emphasis added). Achiekwelu’s contention is that the sentencing provision does not apply to those who represent that they act on behalf of a foreign government; he attempts to restrict the language to domestic governments only. However, Achiekwelu’s conduct falls within the plain language of the provision. New courts have interpreted the sentencing provision at 'issue. The courts that have interpreted it have applied it to defendants who misrepresented that they acted on behalf of a domestic federal or state government agency. See, e.g., United States v. Echevarria, 33 F.3d 175, 179-80 (2d Cir.1994) (holding that the district court correctly applied the enhancement where the defendant misrepresented that he was a “state doctor” able to approve applications for disability benefits); United States v. Hall, 996 F.2d 284, 286-87 (11th Cir.1993) (holding that the district court properly applied the enhancement where the defendants in a telemarketing fraud scheme falsely represented that the victims had to pay taxes on a prize and thus implied that they acted on behalf of the IRS); United States v. Bakhtiari, 913 F.2d 1053, 1063 (2d Cir.1990) (holding that the district court properly applied the enhancement where the defendant claimed to be a State Department official in order to facilitate the purchase of an apartment). No court has considered whether the provision also applies *755 to those who misrepresent that they act on behalf of a foreign government. We must follow the clear, unambiguous language of a particular guideline unless there is a manifestation of contrary intent. See United States v. Frazier, 53 F.3d 1105, 1111-12 (10th Cir.1995). Aehiekwelu’s misrepresentation falls within the plain, unambiguous language of section 2Fl.l(b)(3)(A). Gupta testified that Achiekwelu falsely represented that he worked for the Nigerian Finance Ministry. Thus, the offense “involved a misrepresentation that the defendant was acting on behalf of ... a government agency.” U.S.S.G. § 2Fl.l(b)(3)(A). The Nigerian Finance Ministry is “a government agency.” “Government” standing alone extends to all government agencies, domestic and foreign. The provision’s language does not limit its application to domestic government agencies. The commentary in the guidelines does not address the issue and thus does not manifest a contrary intent. Therefore, we hold that section 2Fl.l(b)(3)(A) applies to those who falsely represent that they act on behalf of either domestic or foreign government agencies. Accordingly, we hold that the district court did not err in enhancing Achiekwelu’s offense level to 23 pursuant to section 2Fl.l(b)(3)(A). IV. As an alternative to the two-level enhancement pursuant to U.S.S.G. § 2Fl.l(b)(3)(A), the district court imposed a two-level upward departure pursuant to U.S.S.G. § 5K2.0 on the basis of the exceptional complexity of the fraudulent scheme. Since the district court issued its opinion, the appropriate departure analysis has been clarified. In Koon v. United States, — U.S. -,---, 116 S.Ct. 2035, 2046-48, 135 L.Ed.2d 392 (1996), the Supreme Court held that an appellate court may review a district court’s ultimate decision to depart from the Sentencing Guidelines in an “atypical” case only for abuse of discretion. In United States v. Rybicki 96 F.3d 754 (4th Cir.1996), and United States v. Hairston, 96 F.3d 102 (4th Cir.1996), cert. denied, — U.S. -, 117 S.Ct. 956, 136 L.Ed.2d 843 (1997), the Fourth Circuit subsequently has clarified the analysis that a district court must follow in deciding whether to depart and the appropriate standards of review of departure decisions. In Rybicki 96 F.3d at 757, we first noted that each guideline provision anticipates a broad range of typical cases — a “heartland” — that represents the circumstances and consequences of ordinary crimes to which the guideline provision applies. We held that district courts “must ordinarily impose sentences within the range specified by the applicable guideline.” Id. A district court may only exercise its discretion to depart from the specified sentencing range if it determines that the circumstances and consequences of the particular case are “atypical” or “unusual” and therefore concludes that the casé does not fall within the guideline’s heartland. Id. In examining whether a case involves “atypical” or “unusual” circumstances or consequences capable of taking a ease out of the applicable guideline’s heartland, the district court must consider the Sentencing Guidelines themselves, the policy statements, and the official commentary. Id. In deciding whether to depart, the district court must follow the analysis that we set out in Rybicki and Hairston. First, the district court must determine the circumstances and consequences of the offense. See Rybicki 96 F.3d at 757. We review the district court’s resulting factual determinations only for clear error. Id. The district court then must decide whether any of the circumstances or consequences of the offense appear “atypical” enough potentially to take the case out of the applicable guideline’s heartland. Id. We do not review the district court’s identification of the potential factors. Id. Once the district court identifies the factors that potentially may remove a case from the applicable guideline’s heartland, the court must classify each factor as a factor that the Sentencing Guidelines: 1) “forbid” as a basis for departure; 2) “encourage” as a basis for departure; 3) “discourage” as a basis for departure; or 4) do not mention as a basis for departure. See Rybicki 96 F.3d at 757. We review the district *756 court’s classification de novo in the context of our ultimate review for abuse of discretion. Id. A factor that the district court classifies as “forbidden” may never provide a basis for departure. See Rybicki, 96 F.3d at 757. The district court must conduct further analysis of factors that it classifies as “encouraged,” “discouraged,” or “unmentioned.” Id. If the identified factor is an encouraged factor for departure, the district court can depart on that basis if the applicable guideline does not already take that factor into account. See Hairston, 96 F.3d at 105; Rybicki, 96 F.3d at 757-58. We review the district court’s determination of whether the guideline adequately takes the factor into account de novo to determine whether the district court abused its discretion. See Rybicki, 96 F.3d at 758. If the identified factor is a discouraged factor or an encouraged factor that the applicable guideline already takes into account, the district court can depart on that basis “‘only if the factor is present to an exceptional degree or in some other way makes the case different from the ordinary case where the factor is present.’ ” Hairston, 96 F.3d at 106 (quoting Koon, — U.S. at-, 116 S.Ct. at 2045). When the determination as to whether a factor is present to an exceptional degree merely amounts to an evaluation of a showing’s adequacy, we review the district court’s determination de novo to determine whether the district court abused its discretion. See Rybicki, 96 F.3d at 758. If the Sentencing Guidelines do not mention the identified factor at all, the district court may depart on that basis only if the “ ‘structure and theory of both relevant individual guidelines and the Guidelines taken as a whole’ ” indicate that the case falls out of the applicable guideline’s heartland. Rybicki, 96 F.3d at 758 (quoting Koon, — U.S. at-, 116 S.Ct. at 2045). However, “the Sentencing Commission expects departures based on ‘unmentioned’ factors to be ‘highly infrequent.’ ” Id. (quoting Koon, — U.S. at-, 116 S.Ct. at 2045). We review the district court’s interpretation of whether the Sentencing Guidelines’ structure and theory allow for a departure de novo to determine whether the district court abused its discretion. Id. Finally, the district court must consider whether the appropriately classified factors take the case out of the applicable guideline’s heartland and whether it should therefore depart from the guideline’s specified sentencing range. We review this ultimate departure decision for abuse of discretion. See Rybicki, 96 F.3d at 758. However, if the district court based its departure decision on a factual determination, we review that underlying determination only for clear error. Id. If the district court based its departure decision on a misinterpretation of the Sentencing Guidelines, we review the underlying ruling de novo. Id. In the instant case, the district court based its two-level upward departure on the “exceptional complexity” of the fraudulent scheme. The court noted: The exceptional complexity of defendant’s scheme to defraud Gupta is evident. Defendant and his cohorts lured their victim to Nigeria, introduced him to several others purporting to be government officials and took him to an office building that they claimed to be the Central Bank of Nigeria. They prepared false documents that appeared to be official Nigerian government certificates and arranged for all money to be sent to foreign countries where it would be more difficult for American law enforcement officials to locate. ... [T]he intricate stratagems devised by defendant and others in this case take it outside the realm of typical wire fraud schemes. Since the Koon, Rybicki, and Hairston opinions were not. available when the district court made its determination, the court did not consider whether the Sentencing Commission forbade, encouraged, discouraged, or failed to mention departure based on the complexity of a defendant’s fraud. Thus, we must consider that question in the first instance. See Hairston, 96 F.3d at 107 (considering classification question in the first instance because the district court “did not *757 have the benefit of the Koon decision when making[its] determination”). The Sentencing Guidelines do not expressly list the complexity of a defendant’s crime as a proscribed factor. Thus, the Commission did not forbid departure on that basis. Nor did the Commission expressly include complexity in the Sentencing Guidelines’ lists of encouraged or discouraged factors. Instead, the Guidelines list the complexity of a defendant’s fraud as a factor that supports an enhancement within the applicable fraud guideline. See U.S.S.G. § 2Fl.l(b)(2)(A) (providing a two-level enhancement for “more than minimal planning”). 3 In Hairston, 96 F.3d at 107, we held that when the Commission designates a factor as a basis for reduction, or in this ease, as a basis for enhancement, the designation implies that the Commission discouraged the factor as a basis for departure, or alternatively, that the Commission encouraged the factor as a basis for reduction or increase but already took it into account. We further held that, “[i]n either case, ordinarily a court should not depart based on such a factor.”. Id. Thus, the complexity of a defendant’s fraud can provide a basis for a departure only if it is “present to such an exceptional degree that it cannot be characterized as typical or ‘usual.’ ” Hairston, 96 F.3d at 108. We therefore caution that the complexity of a defendant’s fraud will rarely warrant an upward departure. However, the record reveals that the fraudulent conduct in this case was very extensive and involved extraordinary planning. Thus, since the factor is present to an exceptional degree, we hold that it may form a basis for an upward departure in extraordinary cases such as this one. See United States v. Alpert, 28 F.3d 1104, 1108-09 (11th Cir.1994) (reversing a district court’s upward departure based on the complexity of the defendant’s fraud because the fraud was not atypical, but noting that a departure may be warranted in eases where the fraudulent conduct is more extensive); United States v. Davidson, 984 F.2d 651, 655 (5th Cir.1993) (affirming the district court’s upward departure based on the “extraordinary planning and meticulous execution involved” in the defendant’s fraudulent scheme); United States v. Palinkas, 938 F.2d 456, 462-63 (4th Cir.1991) (affirming the district court’s upward departure based on the “sophistication” of the defendant’s fraudulent scheme). We next must review the district court’s ultimate decision to depart on the ground that the complexity of Achiekwelu’s fraud removed his case from the “heartland” of typical fraud cases involving “more than minimal planning.” Since the district court based its ultimate departure decision in the instant case on factual determinations, we defer to the district court’s “ ‘assessment of the many facts bearing on the outcome, informed by its vantage point and day-to-day experience in criminal sentencing’ and the comparison of the case with other Guidelines cases.” Rybicki, 96 F.3d at 758 (quoting Koon, — U.S. at---, 116 S.Ct. at 2046-47). Moreover, we agree with the- district court that the intricacy and sophistication of Achiekwelu’s scheme were substantially in excess of the typical fraud case that involves “more than minimal planning.” We therefore hold that the district court did not abuse its discretion in departing on the basis of the complexity of Achiekwelu’s fraudulent scheme. Thus, either because of the two-level enhancement pursuant to section 2Fl.l(b)(3)(A) discussed above in Part III, or because of the two-level upward departure discussed here, the district court did not err *758 in calculating Achiekwelu’s offense level at 23. V. We do not consider Achiekwelu’s final contention that the district court erred when it departed upward to criminal history category II, pursuant to U.S.S.G. § 4A1.3(e), on the basis of Achiekwelu’s similar unconvicted conduct in the fraud perpetrated on Hart. Even if we found that the district court did err in departing pursuant to section 4A1.3(e), we would still affirm Achiekwelu’s sentence. As noted above, the district court stated that its decision was independently supportable on the basis of either of the two grounds for an upward departure. If the district court had departed only on the complexity of Achiekwelu’s fraud, Achiekwelu’s offense level would have been 23 and his criminal history category would have been I, resulting in a sentencing range of 46 to 57 months. The district court expressly stated that in such a case, it still “would impose a sentence on defendant of 51 months of incarceration.” The Supreme Court has held that when an appellate court concludes that a district court departed on both valid and invalid factors, the reviewing court must remand unless it determines that the sentence imposed was reasonable and that the district court would have imposed the same sentence absent reliance on the invalid factor. See Williams v. United States, 503 U.S. 193, 203, 112 S.Ct. 1112, 1120-21, 117 L.Ed.2d 341 (1992). Since we conclude that 51 months is a reasonable sentence, and the district court clearly stated that it would have imposed the same 51 month sentence even if it had relied only on the complexity of Achiekwelu’s fraud, we would not remand for resentencing even if the district court had erred in departing to criminal history category II pursuant to section 4A1.3(e). VI. Accordingly, the judgment is AFFIRMED. Footnotes [1] Okolie and Kachi have not been apprehended. [2] Federal Rule of Evidence 404(b) provides in pertinent part: Evidence of other crimes, wrongs, or acts is not admissible to prove the character of a person in order to show action in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake, or accident.... Fed.R.Evid. 404(b). [3] The background commentary to section 2F1.1 reveals that the Commission intended the "more than minimal planning” enhancement to apply to complex cases. The commentary provides: Empirical analyses of pre-guidelines practice showed that the most important factors that determined sentence length were the amount of loss and whether the offense was an isolated crime of opportunity or was sophisticated or repeated. Accordingly, although they are imperfect, these are the primary factors upon which the guideline has been based. The extent to which an offense is planned or sophisticated is important in assessing its potential hannfulness and the dangerousness of the offender, independent of the actual harm. A complex scheme or repeated incidents of fraud are indicative of an intention and potential to do considerable harm. In preguidelines practice, this factor had a significant impact, especially in frauds involving small losses. Accordingly, the guideline specifies a 2-level enhancement when this factor is present. U.S.S.G. § 2F1.1, background (emphasis added).