United States v. Dulal Chatterji

46 F.3d 1336 UNITED STATES of America, Plaintiff-Appellee, v. Dulal CHATTERJI, Defendant-Appellant. No. 94-5379. United States Court of Appeals, Fourth Circuit. Argued Dec. 9, 1994. Decided Feb. 7, 1995. ARGUED: Mark Daryl Rasch, Arent, Fox, Kintner, Plotkin & Kahn, Washington, DC, for appellant. Maury S. Epner, Asst. U.S. Atty., Baltimore, MD, for appellee. ON BRIEF: Marilyn Tiki Dare, Arent, Fox, Kintner, Plotkin & Kahn, Washington, DC, for appellant. Lynne A. Battaglia, U.S. Atty., Baltimore, MD, for appellee. Before ERVIN, Chief Judge, and MURNAGHAN and WILKINS, Circuit Judges. Vacated and remanded by published opinion. Judge WILKINS wrote the majority opinion, in which Chief Judge ERVIN joined. Judge MURNAGHAN wrote a dissenting opinion. OPINION WILKINS, Circuit Judge: — 1 — Dulal Chatterji appeals the sentence imposed by the district court after he pled guilty to conspiring to defraud the United States, see 18 U.S.C.A. Sec. 371 (West 1966), and to obstructing proceedings before a federal agency, see 18 U.S.C.A. Sec. 1505 (West 1984). Chatterji challenges the determination of economic loss by the district court pursuant to United States Sentencing Commission, Guidelines Manual, Sec. 2F1.1(b)(1) (Oct.1987)1 and the amount of the fine imposed, see U.S.S.G. Sec. 5E4.2.2 Because we conclude that the district court improperly applied an economic loss enhancement under Sec. 2F1.1(b)(1) and failed to explain the basis for the fine imposed, we vacate Chatterji's sentence and remand. I. — 2 — Chatterji was a cofounder and 10% owner of Quad Pharmaceutical Company, Inc. (Quad), a company that manufactured generic drugs. As head of research and development, Chatterji supervised the creation and testing of various generic drugs for which Quad hoped to obtain marketing approval from the Food and Drug Administration (FDA). The two generic drugs that were the subject of his guilty plea were identified as, for purposes of this appeal, vancomycin and ritodrine hydrochloride. A. Vancomycin — 3 — In 1987, Quad submitted an abbreviated new drug application (ANDA) to the FDA in an attempt to obtain that agency's approval to market vancomycin, an injectable antibiotic. See 21 C.F.R. Secs. 314.2, 314.55 (1987). In order to gain approval, FDA guidelines required, inter alia, the submission of stability data from three different research batches of the drug. See Center for Drugs & Biologics, Food & Drug Admin., U.S. Dep't of Health & Human Servs., Guideline for Submitting Documentation for the Stability of Human Drugs and Biologics 25-26 (Feb.1987). A batch is defined as: — 4 — "a specific quantity of a drug or other material that is intended to have uniform character and quality, within specified limits, and is produced according to a single manufacturing order during the same cycle of manufacture." — 5 — Id. at 3 (quoting 21 C.F.R. Sec. 210.3(b)(2)). — 6 — The batch process for vancomycin is relatively simple. Powdered vancomycin, purchased from an FDA-approved supplier, is tested and weighed. Water is then added to make 10 liters of a vancomycin solution, which is placed into vials and freeze-dried in a process known as lyophilization. The resulting sterile powder is tested for stability, and the data is submitted to the FDA as a "batch record." — 7 — Chatterji's participation in the conspiracy as related to vancomycin consisted of the following. Chatterji first prepared 10 liters of vancomycin solution, which he then divided into two "fill sizes" of 10 ml and 5 ml labeled as 488A and 488B. Because the separate fill sizes were portions of the same batch, they were "lots" as defined by the FDA. See id. at 5.3 He then lyophilized the contents of each vial and tested the resulting powder. Further, because he had exhausted Quad's supply of powdered vancomycin in producing lots 488A and 488B, Chatterji did not have unprocessed vancomycin immediately available to create another batch. Rather than suffer the delay and expense of obtaining additional unprocessed vancomycin, Chatterji reconstituted the remaining lyophilized vancomycin from lots 488A and 488B, divided it again into two separate fill sizes labeled 495A and 495B, and lyophilized and tested them.4 Rather than skewing the test results in Quad's favor, Chatterji's use of reprocessed vancomycin had the potential to make the substance used for the test less stable and hence less likely to meet FDA standards. — 8 — Chatterji then forwarded these test results to Dilip Shah, Quad's head of regulatory affairs, who submitted Quad's ANDA for vancomycin claiming that 488A and 488B were separate batches and failing to reveal to the FDA that 495A had been made with reprocessed vancomycin. Therefore, when submitted, Quad's ANDA for vancomycin included records for three purported batches, when in fact only one acceptable batch had been produced. Although it was not required to do so, Quad later prepared and tested another batch of vancomycin and forwarded the results to the FDA. — 9 — The FDA subsequently approved Quad's marketing of vancomycin based on the records for lots 488A, 488B, and 495A, as well as the later-submitted record. Thus, the FDA's approval was based on two valid batch records. Repeated tests of vancomycin produced by Quad after FDA approval revealed that in every instance the drug met all FDA requirements for safety and effectiveness. Indeed, the Government does not dispute that Quad's vancomycin had full therapeutic value and posed no danger to consumers. Quad's gross sales of vancomycin totalled approximately $8 million. B. Ritodrine Hydrochloride — 10 — The FDA also approved Quad's application to market ritodrine hydrochloride (ritodrine), an injectable muscle relaxant. The ANDA for ritodrine submitted by Quad specified the addition of 1 mg/ml of sodium metabisulfate (bisulfate), an inert antioxidant. In its approval of the ANDA, the FDA allowed an overage of up to 1.05 mg/ml of bisulfate. Because the amount of bisulfate diminishes over time, the FDA also specified that ritodrine produced by Quad should contain no less than .7 mg/ml of bisulfate at the end of its shelf life. — 11 — Once the FDA has approved the manufacture and marketing of a generic drug according to a certain formula, a manufacturer is required to seek FDA approval before making any modification to that formula regardless of how insignificant the modification may be. See 21 C.F.R. Sec. 314.70 (1987). Soon after obtaining FDA approval for ritodrine, Chatterji directed that 1.075 mg/ml of bisulfate be used to ensure that the amount of bisulfate remaining at the end of the shelf life would comport with FDA requirements. Quad did not seek prior FDA approval for this formula change and falsely stated in a 1988 annual report to the FDA that no change in the formula for ritodrine had been made. It is not disputed that the minor formula change did not render Quad's ritodrine less effective or pose any danger to consumers who used the drug. Quad earned approximately $5.4 million in gross sales of ritodrine after the formula change. C. — 12 — The charges to which Chatterji pled guilty arose from an FDA audit of Quad's research and development department, which apparently revealed the discrepancies in the batch records for vancomycin and the change in the ritodrine formula. Whether the FDA would have allowed Quad to continue marketing vancomycin and ritodrine cannot be determined because Quad voluntarily withdrew the drugs from the market after the audit. — 13 — The plea agreement between Chatterji and the Government stipulated to the material facts, but left open the question of the appropriate economic loss calculation. The parties agreed that U.S.S.G. Sec. 2F1.1, with a base offense level of 6, applied to Chatterji's offense. The parties disagreed, however, as to whether the base offense level should be enhanced pursuant to Sec. 2F1.1(b)(1) for economic loss caused by the fraud. The district court rejected Chatterji's argument that there was no economic loss attributable to the fraud, finding instead that the effect of the regulatory fraud was that the FDA had never approved Quad's marketing of vancomycin and ritodrine and that the drugs therefore were without value to the consumers who purchased them. Based on this reasoning, the district court measured the "loss" attributable to Chatterji's conduct by Quad's gross sales of vancomycin and ritodrine--approximately $13.4 million--and accordingly increased Chatterji's base offense level by 11 levels for fraud loss in excess of $5 million. See U.S.S.G. Sec. 2F1.1(b)(1)(L). The district court made several other adjustments to reach a final adjusted offense level of 19.5 That offense level, combined with a Criminal History Category I, resulted in a guideline range of 30-37 months imprisonment. The district court sentenced Chatterji to 30 months imprisonment and imposed a $100,000 fine. II. — 14 — Chatterji first challenges the application of the loss enhancement pursuant to U.S.S.G. Sec. 2F1.1(b)(1). While the question of the amount of loss is generally one of fact subject to review only for clear error, the application of a loss enhancement to undisputed facts is a question of law which we review de novo. See United States v. Toler, 901 F.2d 399, 402 (4th Cir.1990). Loss under Sec. 2F1.1(b)(1) is the actual, probable, or intended loss to the victims of the fraud.6 See U.S.S.G. Sec. 2F1.1 comment. (n. 7). In appropriate circumstances, a defendant's gain may provide an estimate of the loss. See U.S.S.G. Sec. 2F1.1 comment. (n. 8). However, gain is only an alternative measure of some actual, probable, or intended loss; it is not a proxy for loss when there is none. See United States v. Haddock, 12 F.3d 950, 960 (10th Cir.1993). Thus, a defendant's gain "may not support an enhancement on its own if there is no actual or intended loss to the victims." Id. — 15 — The Government contends that the regulatory fraud automatically voided any FDA approval of vancomycin and ritodrine. Because unapproved drugs may not be marketed, it continues, Quad's products had no market value, and loss under Sec. 2F1.1(b)(1) properly should be measured by Quad's gain from the sale of the drug. The Government points to 21 C.F.R. Sec. 314.125(b)(7) (1987), which provides that the FDA may decline approval of an application that contains a false statement of material fact.7 However, the mere fact that the FDA may refuse to approve an application on the basis of a false statement does not mean that the FDA's approval, given for an application (or, as in the case of ritodrine, a supplement to an application) containing an undiscovered, false statement, is void ab initio. Indeed, the mere fact that the authority of the FDA to refuse to approve an application containing an untrue statement of material fact is discretionary indicates that the FDA might choose to permit marketing in any event. Moreover, the Government's position does not comport with the FDA's statutory authority, which provides that the FDA "shall, after due notice and opportunity for hearing to the applicant, withdraw approval of an application ... if the [FDA] finds ... that the ... application contains any untrue statement of material fact." 21 U.S.C.A. Sec. 355(e)(5) (West Supp.1994) (emphasis added); see also 21 C.F.R. Sec. 314.150(a)(2)(iv) (1987) (same). If we were to adopt the Government's contention that an untrue statement in the application voided the FDA's approval, Sec. 355(e)(5) would be rendered without meaning. Clearly, there would be no need to withdraw approval of a drug if, due to a materially false statement in the application, no valid approval had ever been given. We decline to adopt a construction that would render a statutory provision superfluous. See Baker v. Bethlehem Steel Corp., 24 F.3d 632, 634 (4th Cir.1994). We therefore disagree with the Government's argument that Quad's vancomycin and ritodrine were not approved by the FDA and thus had no value. — 16 — Based on the record before us, no quantifiable, actual loss can be attributed to Chatterji's conduct. Although an agency may suffer economic loss as a result of regulatory fraud (for example, costs of an investigation), the Government has not argued that Chatterji's fraud caused any such loss. Moreover, there was no loss to the consumers of vancomycin and ritodrine. Quad's products were exactly what they purported to be: vancomycin and ritodrine, approved by the FDA, manufactured in a certain strength and dosage, and producing the specified therapeutic benefits that FDA requirements were intended to ensure. We emphasize that we therefore are not presented with a product substitution case in which the product sold is something other than what it is claimed to be. It is undisputed that Quad's vancomycin performed according to FDA specifications despite the fact that records for only two batches, rather than three, were submitted to the FDA prior to its approval of vancomycin. Further, there is no dispute that the safety and therapeutic value of the ritodrine were not affected by the addition of an additional .025 mg/ml of bisulfate--an inclusion of only 2.3% more of an inert inactive ingredient than allowed under the FDA-approved formula that was intended to ensure that the drug would retain full potency over the course of its shelf life. Finally, there is no serious question that the FDA would have approved the ANDA for vancomycin had the three-batch requirement been met and would have approved the formula change for ritodrine had Quad simply requested approval of the modification.8 — 17 — In sum, this is not a situation in which a drug with fraudulently-obtained FDA approval harms consumers, fails to produce its intended effects, or is something less than it is represented to be. We have little doubt that economic loss would exist in such situations. But, when a drug possesses FDA approval, poses no threat to the health and well-being of the consumer, and meets all of the goals of FDA requirements for safety and efficiency, there can be no actual, monetary loss attributable to the regulatory fraud by which FDA approval was obtained. Economic gain to the manufacturer therefore is not the appropriate measure of loss in such a situation. — 18 — The dissent points to United States v. West, 2 F.3d 66 (4th Cir.1993), in which this court rejected the defendants' contention that the district court had misapplied Sec. 2F1.1. The district court in West determined the defendants' sentences on two bases. First, relying on U.S.S.G. Sec. 2F1.1, comment. (n. 7(b)) (1992), the district court concluded that the actual loss resulting from the defendants' fraud " ' "tend[ed] not to reflect adequately the risk of loss created by the defendant's conduct." ' "9 West, 2 F.3d at 71 (quoting presentence letter to counsel from district court). Second, the district court relied on U.S.S.G. Sec. 2F1.1, comment. (n. 8) (1992), which provides that the defendant's gain may be used as an alternative estimate of loss, and found that the amount of the brokerage fees the Government had paid to the defendants was an appropriate alternative measure of the Government's actual loss. West, 2 F.3d at 71; see U.S.S.G. Sec. 2F1.1, comment. (n. 8) (1992) ("The offender's gain from committing the fraud is an alternative estimate that ordinarily will underestimate the loss."). — 19 — The dissent's reliance on West is misplaced. Here, the district court did not base its sentence on an upward departure, and thus the question of the appropriateness of a departure is not before us.10 Consequently, we may not affirm Chatterji's sentence on this basis, as the dissent suggests. Cf. Williams v. United States, 503 U.S. 193, 200-03, 112 S.Ct. 1112, 1120, 117 L.Ed.2d 341 (1992) ("When a district court has not intended to depart from the Guidelines, a sentence is imposed 'as a result of' an incorrect application of the Guidelines when the error results in the district court selecting a sentence from the wrong guideline range."). Moreover, unlike West, there was no risk of loss that resulted from Chatterji's conduct. The dissent does not, because it cannot, point to any evidence in the record that would indicate that consumers were at risk as a result of using the products in question. Most significantly, the district court made no findings that consumers were exposed to a risk of harm. — 20 — Regarding the second basis for the sentence in West, the enhancement provided in Sec. 2F1.1(b)(1), as noted above, applies when a victim has suffered economic loss. A defendant's gain may be an appropriate estimate of loss only when there is some actual, intended, or probable loss. Because we have concluded that Chatterji's conduct occasioned no such loss, gain may not be used as an alternative basis for calculating loss. — 21 — We do not in any way denigrate the seriousness of Chatterji's offense, and we note that he will not go unpunished merely because application of the economic loss enhancement is inappropriate under these circumstances. See United States v. Smith, 951 F.2d 1164, 1169 (10th Cir.1991). An enhancement pursuant to Sec. 2F1.1(b)(1) provides for additional punishment which is meted out when the Government proves that an economic loss has occurred. See United States v. Schneider, 930 F.2d 555, 559 (7th Cir.1991). The Government has not justified the imposition of such additional punishment on this basis because it failed to prove the existence of any loss resulting from Chatterji's fraud.11 See id. III. — 22 — Chatterji also appeals the fine imposed by the district court, arguing that it improperly departed upward from the applicable fine range of $6,000-$60,000. During the sentencing hearing, the Government urged an upward departure to the maximum fine provided for Chatterji's conviction, $250,000. See 18 U.S.C.A. Sec. 3571(b)(3) (West Supp.1994). The district court concluded that a fine of $100,000 was "appropriate," stating that Chatterji's offense was "a crime of greed" and that the fine imposed would "take some of the profits out of it." — 23 — It appears that the district court applied the Guidelines Manual effective November 1, 1993 in determining Chatterji's fine and considered the fine imposed an upward departure from the fine range of $6,000-$60,000 provided in U.S.S.G. Sec. 5E1.2 (Nov.1993). However, the conclusory statements by the district court do not afford us a meaningful opportunity to determine whether the departure was based on "a factor not adequately considered by the Sentencing Commission in formulating the applicable guidelines range." United States v. Graham, 946 F.2d 19, 21 (4th Cir.1991). While the Government argued before us that the district court apparently increased the amount of the fine above the range applicable under the 1993 Guidelines Manual to provide for reimbursement to the Government of the cost of Chatterji's incarceration, this position is based on speculation for the district court articulated no such finding. And, the district court gave no indication that it was applying U.S.S.G. Sec. 5E4.2(c) (Oct.1987)12 in determining the amount of the fine. Therefore, in addition to remanding for resentencing, we remand to the district court to provide it the opportunity to make appropriate findings and articulate a more complete explanation of the reasons for the fine it imposes. See United States v. Harvey, 885 F.2d 181, 182-83 (4th Cir.1989). — 24 — VACATED AND REMANDED. MURNAGHAN, Circuit Judge, dissenting: — 25 — Section 2F1.1 of the United States Sentencing Guidelines ("U.S.S.G.") governs sentencing for offenses involving fraud or deceit, and grades the severity of all applicable offenses on the basis of the dollar value of the calculated "loss" caused by the fraud. The November 1987 version of Sec. 2F1.1, applicable to the instant case, provides for a maximum enhancement of 11 offense levels in cases in which the total fraud "loss" exceeds $5 million. — 26 — The majority, and the Appellant, contend that the district court erred in calculating the "loss" under Sec. 2F1.1 in the instant case as equalling the total dollar value of the drugs sold by Quad Pharmaceuticals as a result of the fraudulently obtained approval of certain drugs by the Federal Drug Administration ("FDA"). In particular, the majority argues that there was no actual "loss" to consumers as a result of the Appellant's misrepresentations to the FDA because the drugs received FDA approval, posed no threat to the health and well-being of the consumers, and met all of the goals of the FDA requirements for safety and efficiency. Majority Opinion at 1341.1 Underlying the majority's conclusion that there was no loss to the consumers is the contention that the drugs had some actual measurable value to the consumers. Indeed, although the Appellant, in his Brief, suggests that the actual value received by the consumers was somewhat less than the value of the drugs had they been properly approved, the majority goes further and suggests that the value received by the consumers equalled the value of the drugs had they been properly approved. Majority Opinion at 1341. Because I find no logical basis for holding that the drugs sold as a result of fraudulently obtained FDA approval had some readily measurable "actual value" to the consumers, or for holding that there was subsequently no "loss" to the consumers, I respectfully dissent. — 27 — The majority's opinion fundamentally turns on its argument that in the instant case, there was no loss inflicted upon the consumers as a result of the fraudulently obtained FDA approval of the drugs because "Quad's products were exactly what they purported to be: vancomycin and ritodrine, approved by the FDA, manufactured in a certain strength and dosage, and producing the specified therapeutic benefits that FDA requirements were intended to ensure." Majority Opinion at 1341. That conclusion, however, rests on two faulty assumptions, both of which are speculative and lack a clear basis in the record: (1) that "the safety and therapeutic value of the ritodrine were not affected by the additional .025 mg/ml of bisulfate," and (2) that "the FDA would have approved the ANDA for vancomycin had the three-batch requirement been met and would have approved the formula change for ritodrine had Quad simply requested approval of the modification." Majority Opinion at 1341. — 28 — The two assumptions underlying the majority opinion are fundamentally flawed for several reasons, and thus cast great doubt upon the majority's ultimate conclusion. First, the majority's contention that the fraudulently approved drugs were bioequivalents of properly approved drugs, and thus imposed no actual loss to the consumers, is based solely on the absence thus far, of any reports that the consumers taking those drugs have been injured or harmed. The Guidelines surely, however, could not have intended for courts to wait for such crimes of fraud to result in actual, serious harm to consumers of medicines before the perpetrator of such a fraud is subject to a sentencing enhancement for fraudulent activity. Moreover, the concept of bioequivalency has no bearing on the gravity of the fraud committed; regardless of whether the fraudulently approved drugs ultimately did or did not cause harm to the consumers, Chatterji was willing to take the risk that such harm would occur simply in order to expedite the receipt of his own profits from the sales of the drugs. He simply transferred the risk to the public whose members relied on proper FDA approval to feel comfortable in taking the drugs. The FDA regulations set out with specificity the procedures which should be followed in order to ensure the complete safety of the drugs; Chatterji's actions here were taken in complete and utter disregard of the precautionary and purposefully detailed regulations.2 — 29 — Moreover, the majority's second assumption--that there is no serious contention that the FDA would have approved the drugs had Chatterji in fact complied with the proper FDA approval regulations--simply simply is unsupported by the record. Indeed, at sentencing, FDA employee, John Harrison, expressly stated that the FDA would not have approved Quad's drugs for sales had it known that Quad had misled the FDA about the batches and formulaic makeup of the drugs.3 Thus, the majority's argument is not only speculative, but runs contrary to the evidence contained in the record. Indeed, the evidence presented at sentencing supports the conclusion that, without proper FDA approval, the drugs on the market would be lacking in saleability, and hence lacking in monetary value. — 30 — These fundamental weaknesses in the majority's logic cast great doubt upon its ultimate conclusion. More significant, in light of these weaknesses, it appears that the district court's calculation of loss as equalling gross sales by Quad was indeed proper and well supported by the Guidelines.4 — 31 — First, the commentary to the Guidelines5 specifically provides that "an offender's gain from committing the fraud" is an alternative estimate to the calculation of loss. Application Note 8, U.S.S.G. Sec. 2F1.1. In so providing, the commentary particularly states that, for the purposes of the calculation of a sentence under section 2F1.1(b)(1), the loss — 32 — need not be determined with precision. The court need only make a reasonable estimate of the loss, given the available information. This estimate, for example, may be based on the approximate number of victims and an estimate of the average loss to each victim, or on more general factors, such as the nature and duration from the fraud and the revenues generated by similar operations. The offender's gain from committing the fraud is an alternative estimate that ordinarily will underestimate the loss. — 33 — Application Note 8, U.S.S.G. Sec. 2F1.1 (emphasis added). The fact that the commentary itself notes that "revenues generated" is an appropriate measure of loss suggests that the district court's calculation in the instant case was proper. Indeed, the Guidelines commentary clearly supports the district court's decision to use Quad's gross revenues from the sales of the fraudulently approved drugs as the best estimate of victim's loss or alternatively, offender's gain, in the instant case. — 34 — Second, the Guidelines commentary to Sec. 2F1.1 provides that in calculating loss for the purposes of sentencing in a case involving a misrepresentation "concerning the quality of a consumer product," the actual loss can be measured by calculating the difference between the amount paid by the victim for the product and the "amount for which the victim could resell the product received." Application Note 7(a), U.S.S.G. Sec. 2F1.1 (emphasis added). Because it is reasonable to assume that a consumer knowing that the FDA had approved certain drugs as a result of misstatements and fraud would not be willing to purchase the drugs, it is doubtful that the fraudulently approved drugs in the instant case could have been resold at any price. Indeed, in light of the fact that the pharmaceutical industry is very tightly regulated, it was reasonable to assume that drugs not properly approved have no actual or resale value. Thus, under the assumption that the resale price of these fraudulently approved drugs would be zero, the district court's calculation of loss as amounting to gross sales was not erroneous. — 35 — Even if the consumers suffered no "loss" under the Appellant's, and the majority's, argument that the drugs sold by Quad were in fact, safe, effective bioequivalents of properly approved drugs, the alleged "zero" loss is not an appropriate measure of loss in this case because it would "tend not to reflect adequately the risk of loss created by the defendant's conduct." Indeed, this Court, in United States v. West, 2 F.3d 66 (4th Cir.1993), recently construed Sec. 2F1.1 as allowing the calculation of loss, in cases involving the making of false statements to the government, to go beyond the actual loss to the victims when the risk of loss is great. In West, the appellants had been convicted of making false statements to the government in violation of 18 U.S.C. Secs. 2, 371, and 1001. 2 F.3d at 67. In that case, the appellants had violated certain federal regulations requiring that contractors of federal construction projects secure bonds guaranteeing their performance of certain obligations; specifically, the regulations required a party wishing to qualify as an individual surety to file a Standard Form 28 Affidavit of Individual Surety, listing net worth, assets, and liability. Id. The appellants had served as matchmakers between contractors and individual sureties, and in that capacity, had filed false Form 28s. Id. at 68. The district court, in calculating the appellants' sentences under U.S.S.G. Sec. 2F1.1, had accepted the government's loss as the $2.9 million it paid in brokerage fees for bonds secured through the falsified Form 28s, and had noted that although the actual loss in the "sense of belly up contracts" was well below $2 million, the "true governing consideration is that such an actual loss approach in this case would 'tend not to reflect adequately the risk of loss created by the defendant's conduct.' " Id. at 71. At sentencing, the district court had additionally relied on Application Note 8, which allows the offender's gain to be an alternative estimate of the loss, in reaffirming its calculation of the loss as $2.9 million. This Court, in affirming the district court's sentencing calculation, stated: — 36 — [Appellants] contend that the government failed to prove either actual loss or risk of loss because it did not eliminate the possibility that, with respect to each of the fraudulently secured bonds, one of the two sureties might have had sufficient assets to cover a forfeiture. They urge that, when adequate sureties are factored into the equation, the risk of loss shrinks from $2.9 million to $429,154.... We disagree. As the district court noted, at the time [Appellants] engaged in their enterprise two individual sureties were required to secure a bond, each with a net worth equal to or exceeding the contract price. Thus, the court properly concluded that the government simply did not get what it paid for so that the amount paid out (or "the money ... unlawfully taken") fairly constitutes actual loss under Sec. 2F1.1. Nor do we find clear error in the district court's conclusion, after a careful evaluation of the evidence before it, that the $2.9 million figure adequately represents "the risk of loss created by defendant[s]' conduct." — 37 — Id. (citations omitted) (emphasis added). In light of the West decision, it is clear that this Court has been, and should continue to be, willing to go beyond hypothesized actual losses in cases in which fraud on the government creates large risks of loss to the public which are inherently difficult to measure. — 38 — Such a result is also supported by the Ninth Circuit's decision in United States v. Cambra, 933 F.2d 752 (9th Cir.1991), in which the Ninth Circuit addressed an appeal by a defendant who had pled guilty to various charges of violating the Food, Drug, and Cosmetic Act relating to the sale of anabolic steroids, which he labeled as though they had been produced by a recognized manufacturer of steroids. In calculating "loss" under section 2F1.1, the district court had found that the dollar value of the counterfeited steroids was $500,000, and sentenced the Appellant accordingly. Id. at 756. In affirming the district court's sentencing calculation, the Ninth Circuit held: — 39 — The monetary table in the fraud guideline is intended to reflect the harm to the victim and the gain to the defendant. Federal agencies may be victims of fraud in counterfeiting and misbranding drugs. There is no meaningful distinction between government as victim and individual consumer victims.... In this case, the district court found that Cambra intended to profit from his activity and that at least federal agencies were defrauded by his acts. — 40 — Id. Accordingly, the Court found that the $500,000 figure was an appropriate measure of "loss" within the meaning of Sec. 2F1.1 because it represented the retail value of the steroids. Id. In so holding, the Ninth Circuit did not insist that the district court subtract from this figure any "actual value" accruing to the consumers of the steroids. Thus, Cambra lends clear support to the district court's use of the value of the drugs as a measure of loss, even assuming that consumers did not suffer directly. — 41 — Finally, the district court was entitled to depart upwardly from the range established by U.S.S.G. Sec. 2F1.1, because the actual value of the drugs did not take into account the actual risk to society created by the scheme to defraud. Indeed, the commentary to Sec. 2F1.1 expressly provides that in cases in which the loss determined under subsection (b)(1) does not "fully capture the harmfulness and seriousness of the conduct," an upward departure may be warranted. Application Note 10, U.S.S.G. Sec. 2F1.1. As examples of such cases, the commentary lists, among others: (1) instances in which the fraud "cause[s] or risk[s] reasonably foreseeable, substantial non-monetary harm," and (2) offenses which cause "a loss of confidence in an important institution." Application Note 10, U.S.S.G. Sec. 2F1.1 (emphasis added). The fraud committed in the instant case not only created risks of potentially great harm to consumers of pharmaceutical products, but also severely diminished the confidence which consumers can justifiably have in the FDA approval process. Thus, an upward departure from the calculated guidelines range was plainly appropriate. — 42 — The majority's holding in the instant case will send two harmful messages to the public. First, the majority has in essence established that consumers cannot, and should not, justifiably rely on the FDA approval process in evaluating the safety and efficacy of the drugs that they purchase. Second, the majority has fundamentally held that courts must wait until consumers are actually harmed in some physical or economic sense before a sentencing enhancement can be deemed warranted under section 2F1.1 of the Sentencing Guidelines. The Guidelines surely could not have intended such a result. I dissent. Footnotes [1] Unless otherwise noted, all references to the sentencing guidelines are to the October 1987 Guidelines Manual [2] Chatterji was sentenced on May 12, 1994; thus, the Guidelines Manual effective November 1, 1993 applied. See 18 U.S.C.A. Sec. 3553(a)(4) (West 1985); U.S.S.G. Sec. 1B1.11(a) (Nov.1993). However, because Sec. 2F1.1 had been amended to increase the applicable offense level since the date of his offense conduct, the parties stipulated that the Guidelines Manual effective November 1, 1987 was the appropriate manual for computing Chatterji's offense level. See United States v. Morrow, 925 F.2d 779, 782-83 (4th Cir.1991); see also U.S.S.G. Sec. 1B1.11(b) (Nov.1993) (one-book rule) [3] In his brief, Chatterji's counsel erroneously asserts that FDA procedures allowed the submission of data from two separate fill sizes as two separate batches. This assertion appears to be based on the presentence report, which in turn cites a telephone interview with FDA employee John Harrison. However, this assertion is flatly contradicted by Harrison's testimony at the sentencing hearing, during which he clearly stated that Chatterji created two lots, not two batches, of vancomycin when he divided the vancomycin solution into separate fill sizes [4] Quad saved approximately $5,000 and one day of processing time as a result of Chatterji's activities [5] The district court appropriately increased Chatterji's offense level by 2 for more than minimal planning, see U.S.S.G. Sec. 2F1.1(b)(2)(A); by 3 for management or supervision of criminal conduct involving five or more participants, see U.S.S.G. Sec. 3B1.1(b); and by 2 for obstruction of justice relating to an FDA audit of Quad, see U.S.S.G. Sec. 3C1.1. It reduced Chatterji's offense level by 3 levels for acceptance of responsibility. See U.S.S.G. Sec. 3E1.1 (Nov.1993). And, the district court acquiesced in the Government's motion for a 2-level downward departure to reflect Chatterji's cooperation. See U.S.S.G. Sec. 5K1.1 [6] It appears to be undisputed that there was no probable or intended loss resulting from Chatterji's conduct. The Government has produced no evidence, and we have no reason to believe, that Chatterji intended to harm the consumers of these drugs or to market a drug that failed to do what Quad claimed it would do. In short, there is no basis upon which to find an intended or probable loss [7] The Government also relies on United States v. Cambra, 933 F.2d 752 (9th Cir.1991). Cambra pled guilty to various charges of violating the Food, Drug, and Cosmetic Act relating to the sale of anabolic steroids, which he labeled as though they had been produced by a recognized manufacturer of steroids. The Cambra court approved the use of the "dollar value" of the drugs as the measure of loss because Cambra "intended to profit from his activity." 933 F.2d at 756. Although it appears that the court held that a defendant's gain is an appropriate measure of loss in the context of a regulatory fraud without regard for loss to the victims, we cannot be sure of this because it is unclear from the opinion whether the counterfeit steroids had the same therapeutic value as approved steroids or whether they were something other than what they purported to be [8] The dissent misreads the testimony of FDA employee John Harrison at the sentencing hearing. According to the dissent, Harrison "expressly stated that the FDA would not have approved Quad's drugs for sales had it known that Quad had misled the FDA about the batches and formulaic makeup of the drugs." Dissent at 1345. To the contrary, rather than stating that the FDA would not have approved Quad's drugs, Harrison testified that had the FDA known about the false data submitted regarding the number of batches of vancomycin, it would have delayed the approval process and "ask them to make another batch." Further, with respect to the formula change for ritodrine, Harrison did not testify that the FDA would not have approved the formula change. He did testify that had the FDA been presented with a formula change request, "questions would have been asked as to why 25 percent was needed. That's a really unusual amount." However, even these tentative reservations must be viewed in light of the fact that Harrison was proceeding under the mistaken belief that 25% more bisulfate had been added to the ritodrine formula, when in fact only 2.3% more bisulfate had been added [9] It is unclear whether the district court, or this court, viewed application note 7(b) as allowing a departure from the guideline range or as permitting the risk of loss to be used to calculate loss pursuant to Sec. 2F1.1(b)(1). However, amendments to the guidelines effective November 1993 make clear that when the determination of economic loss underestimates the seriousness of the defendant's conduct--as when the risk of loss is great--a departure may be appropriate. U.S.S.G. Sec. 2F1.1, comment. (n. 7(b)) (1993) ("For example, where the defendant substantially understated his debts to obtain a loan, which he nevertheless repaid, the loss determined above (zero loss) will tend not to reflect adequately the risk of loss created by the defendant's conduct.... Where the loss determined above significantly understates or overstates the seriousness of the defendant's conduct, an upward or downward departure may be warranted.") [10] We express no opinion as to the appropriateness of consideration of an upward departure by the district court on remand [11] The Government asserted at oral argument that it was not required to prove loss because loss was not an element of the offenses to which Chatterji pled guilty. It is true that had Chatterji proceeded to trial the Government would not have been required to prove loss in order to obtain a conviction. Nevertheless, it is clearly established that the Government bears the burden of proving, by a preponderance of the evidence, facts relevant to the sentencing determination. United States v. Uwaeme, 975 F.2d 1016, 1018 (4th Cir.1992) 12 U.S.S.G. Sec. 5E4.2(c) (Oct.1987) provides that "monetary gain to the defendant" is a factor to be considered in determining the appropriate fine range. [1-1] 1 By analogy, that equates to a holding that no loss has occurred when a soldier acting as a sentry on patrol has gone to sleep when his company was resting, thus running the risk of infiltration by the enemy, because no enemy actually appeared on the sentry's watch. However, that would overlook the adverse effect on all the other members of the company who would become aware that the safety of sleep when guarded by a sentry had been substantially diminished. The increase in the risk would be much greater if the army--especially the company commander--did not punish the delinquent sentry. There would clearly be loss from the sentry's defalcation in duty even though his watch was actually not disturbed by the enemy [2-1] 2 For the majority to say that Chatterji meant no probable or intended loss is to ignore the principal congressional purposes in creating the FDA and granting it its powers. Indeed, the majority's suggestion that a drug is properly approved by the FDA until the FDA decides to withdraw such approval upon discovery of a material misrepresentation, would distort the meaning of the language relied on and diminish the purposes underlying the FDA approval process [3-1] 3 Indeed, Mr. Harrison stated that if the FDA had known that Quad had reused vancomycin in creating "batches" for testing, "I don't think that batch would have qualified." Joint Appendix at 183 [4-1] 4 Note that the district judge, in addition to imposing a sentence of imprisonment, fined the Appellant in the amount of $100,000. Because total sales of the fraudulently approved drugs amounted to more than $13,000,000, the fine imposed in no way took away all of the profits gained from Quad's drug sales [5-1] 5 Note that the commentary to the Guidelines is binding unless inconsistent with the Constitution, federal statute, or the Guidelines themselves. Stinson v. United States, --- U.S. ----, ----, 113 S.Ct. 1913, 1919, 123 L.Ed.2d 598 (1993); United States v. West, 2 F.3d 66, 71 n. 6 (4th Cir.1993)