Randall Carroll pled guilty to one count of conspiracy in violation of 18 U.S.C. § 371 in connection with an odometer-tampering scheme. He was sentenced to 37 months of imprisonment, three years of supervised release, 300 hours of community service and a special assessment of $50. Carroll appeals from his sentence, alleging that the district court erred in calculating the amount of the loss under the sentencing guidelines and in enhancing his sentence for his role as the leader of a criminal conspiracy. For the following reasons, we affirm.

Randall Carroll owned and operated an automobile dealership in Cleveland, Tennessee. He purchased late-model automobiles at auctions and paid mechanics to roll back the odometers. To cover up his actions, he forged the signatures of previous owners and fraudulently obtained replacement titles showing the reduced mileage. He then sold the automobiles to unsuspecting buyers who thought they were purchasing automobiles with much less mileage.

A federal grand jury indicted Carroll on one count of conspiracy in violation of 18 U.S.C. § 371, 13 counts of odometer tampering in violation of 15 U.S.C. §§ 1984 & 1990c,1 and three counts of interstate transportation of counterfeited securities, i.e., motor vehicle titles, in violation of 18 U.S.C. § 2314. Pursuant to a plea agreement, Carroll pled guilty to the conspiracy charge, and the remaining counts were dismissed at the time of sentencing.

The probation office prepared a presentence investigation report (PSR). The PSR concluded that the base offense level was six. United States Sentencing Commission, Sentencing Guidelines, § 2F1.1(a). The PSR then recommended adding 10 levels for the amount of the loss, which it calculated to be $753,000. Id. at § 2F1.1(b)(1)(K). In calculating the amount of the loss, the PSR relied on a report by the Chief of the Odometer Fraud Staff of the National Highway Traffic Safety Administration (NHTSA). In that report, NHTSA used issues of the Galves Auto Price List to determine that, for intermediate-size used automobiles, the value of an automobile increases $85 for every 1,000 miles rolled back. Here, Carroll had rolled the odometers back an average of approximately 47,000 miles per automobile. Multiplying this by $85, the average loss per automobile was $3,995. The PSR conservatively estimated the loss at only $3,000 per automobile. There was evidence that Carroll had rolled back the odometers of 440 automobiles; however, only 251 could be documented. Thus, the total loss was calculated as $753,000.

The PSR also used other measures of the loss to confirm the reasonableness of its calculation. The PSR relied on a study of wholesale prices conducted by NHTSA and the Pennsylvania Attorney General Bureau of Consumer Protection. That study found that the average loss was between six and ten cents per mile rolled back. When multiplied by the average rollback of 47,000 miles, the loss would be between $2,820 and $4,700 per automobile. In addition, that study found that many consumers would never knowingly buy an automobile with a rolled-back odometer, regardless of the price, and that the loss in many cases would be the full amount paid for the automobile. Finally, the PSR relied on studies which showed that the owners of these automobiles incurred additional maintenance expenses, lost time due to breakdowns, and paid increased taxes, finance charges, and insurance.

The PSR then recommended a two-level enhancement because the offense involved more than minimal planning or a scheme to defraud more than one victim. U.S.S.G. § 2F1.1(b)(2). The PSR also recommended a four-level enhancement for Carroll's role as a leader in a criminal activity involving five or more participants. Id. at § 3B1.1(a). Finally, the PSR recommended a three-level reduction for acceptance of responsibility. Id. at § 3E1.1(a). This brought the final offense level to 19.

Carroll raised two objections to the PSR. First, he objected to the PSR's calculation of the amount of the loss. Second, he objected to the PSR's conclusion that the scheme involved five or more participants.

At the sentencing hearing, the district court heard evidence on these issues. The government submitted evidence similar, if not identical, to the evidence relied on in the PSR. It also submitted evidence that many of Carroll's victims had suffered consequential losses as a result of their purchases, such as increased maintenance and repair costs, and that the automobiles were virtually unmarketable.

Carroll argued that the district court should use figures created from the N.A.D.A. Official Used Car Guide. Carroll's counsel compared the prices for which Carroll had sold the automobiles to the prices the automobiles would have fetched according to the N.A.D.A. guide had Carroll not rolled back their odometers. Counsel was only able to do this for 98 of the 251 automobiles, and came up with an average loss of approximately $1,534. This was approximately 20 percent of the sale price of these automobiles. Extrapolating to the remainder of the automobiles, Carroll calculated the total loss as $348,454, which would only allow an eight-level enhancement. See U.S.S.G. § 2F1.1(b)(1)(I).

Carroll also called Gary Rymer, the used car manager at a local Honda dealership. Rymer testified from his unquestioned expertise in selling used cars that he did not believe that rolling back an automobile's odometer by 47,000 miles would cause it to increase in value by $3,000. He also estimated that 95% of banks and credit unions use the N.A.D.A. guide in valuing used cars. On cross-examination, Rymer admitted that he only used the N.A.D.A. guide in about ten percent of his transactions. He also admitted that he tried to avoid purchasing automobiles if their true mileage was not known, and that he could only sell such cars on the wholesale market at a discount.

The district court rejected Carroll's proposed method of valuing the loss and accepted the government's method. The court concluded that the government's estimate was, if anything, conservative.

Also at the hearing, the government called investigator Charles Bradley of the Tennessee Highway Patrol. He testified that he interviewed Dwight Ryan, who helped Carroll roll back the odometers.2 Ryan told him that Keith Sewell and Babe Lawson had also helped Carroll roll back odometers. Ryan had previously testified before the grand jury that Barron Carroll, the defendant's son, told Ryan that he knew that he was selling automobiles with rolled-back odometers for his father. Bradley also testified that he had confirmed some of these facts with Sewell.

Bradley also testified that Sherry Cantrell had told him that Carroll had paid her to forge affidavits to get replacement titles. Cantrell herself testified to that in a trial in the Northern District of Mississippi in 1994. Bradley also testified that Lonnie Cross, Cantrell's step-father, told him that he had helped Cantrell to obtain the replacement titles.

In response, Carroll introduced the affidavit of his attorney, which stated that Cross had denied to Carroll's attorney that he knew that Carroll was rolling back odometers. Carroll offered no further testimony. The district court agreed with the PSR, and found that Randall Carroll, Ryan, Sewell, Lawson, Barron Carroll, Cantrell, and Cross all participated, for a total of seven participants. It also found that Carroll was the leader of the conspiracy, and gave him a four-level enhancement pursuant to § 3B1.1(a).

Carroll now appeals, claiming that the district court erred both in its calculation of the loss and in applying the enhancement for a leadership role in a conspiracy involving five or more individuals.

Carroll first challenges the district court's calculation of the amount of the loss. A district court's initial definition of loss is a question of law to be reviewed de novo. United States v. Kohlbach, 38 F.3d 832, 841 (6th Cir.1994). Once the definition of loss has been established, the district court's factual determinations are reviewed for clear error. Id. Here, both parties agree that Carroll is challenging the district court's factual determinations, and that the district court's findings should only be reversed if they are clearly erroneous. Under the clear error standard, the district court's findings should only be overturned if the circuit court is satisfied that "the court's evaluation of the loss was not only inexact, but was outside the universe of acceptable computations." Id. (quoting United States v. Tardiff, 969 F.2d 1283, 1288 (1st Cir.1992)).

Where, as here, the fraud involves the misrepresentation of the quality of the consumer product, the loss is the difference between the amount paid by the victim and the amount for which the victim could resell the product received. U.S.S.G. § 2F1.1, comment. (n. 7(a)). However, the district court does not have to determine the loss with precision. Id. at § 2F1.1, comment. (n. 8). Rather, it must make a "reasonable estimate of the loss, given the available information." Id.

Carroll argues first that the district court erred by calculating the gross loss instead of the net loss to the victims. The guidelines do require the district court to determine the net loss. See United States v. Lavoie, 19 F.3d 1102, 1105 (6th Cir.1994); United States v. Sloman, 909 F.2d 176, 182 (6th Cir.1990). However, the district court properly calculated the net loss; it calculated how much rolling back the odometers changed the value of the cars.

Next, Carroll argues that the district court erred by using the government's method of calculating the loss value instead of his method. Carroll offers several reasons why his method is superior. First, Carroll claims that the N.A.D.A. guide is superior to the Galves guide as a matter of law and points to a Fifth Circuit case that used the N.A.D.A. guide in calculating the amount of the loss. See United States v. Whitlow, 979 F.2d 1008, 1012 (5th Cir.1992). However, Whitlow stands for the proposition that the use of the N.A.D.A. guide to determine the loss is not clearly erroneous; it does not hold that the N.A.D.A. guide is the required guide, or even the preferred one. Id. In addition, the Fifth Circuit in Whitlow approved the use of the N.A.D.A. guide to conclude that the cars had lost 40 percent of their value. Id. If a 40 percent loss ratio were applied in this case, the average loss per automobile would still be approximately $3,000.

Carroll, relying on Rymer's testimony, also claims that his method is superior because the government's method overstates the amount of the loss. However, the government presented evidence that its method understates the amount of the loss, because it failed to take into consideration that the owners of these automobiles often cannot resell them at any price, have increased maintenance costs, higher taxes, higher finance costs, and higher insurance. In light of this evidence, the district court did not err in determining that the government's method provided a reasonable estimate of the amount of the loss.

In addition, there was evidence that Carroll's method understates the amount of the loss. Carroll's method assumes that the victims could sell the automobiles for what they would have been worth had Carroll not rolled back the odometers. However, Rymer testified on cross-examination that one cannot get full value for an automobile with a rolled-back odometer; in fact, he testified that it was hard to get any value. Thus, the victims could not sell the automobiles for the price listed in the N.A.D.A. guide. Moreover, in calculating the net loss, the amount paid must be compared to the amount for which the victim could resell the product, not to the amount the product would be worth had the fraud never occurred. U.S.S.G. § 2F1.1, comment. (n. 7(a)). Because Carroll's method looks at the worth of the product but for the fraud, its comparison is faulty.

Carroll also argues that the government's method is inaccurate because it relies on averages and does not take into account individual differences in the prices of automobiles. However, the guidelines specifically sanction the use of averages: "This estimate [of the loss], for example, may be based on the approximate number of victims and an estimate of the average loss to each victim." U.S.S.G. § 2F1.1, comment. (n. 8). In addition, Carroll's own method relies on averages to calculate the total loss. Carroll could only find sufficient records to calculate the precise loss for 98 of the automobiles. He then extrapolates from the average percentage loss to determine the total loss.

Furthermore, even if the government's method overstates the loss on some of the automobiles, that does not mean that the district court erred in using that method. The government's method is an average; by definition, an average will overstate some losses and understate others. As stated above, the government presented a plethora of evidence that its method understates many of the losses.

In summary, the district court did not err in calculating the amount of the loss. The district court was faced with two different methods, with testimony supporting each method. The method the district court chose was a reasonable way of estimating the loss in light of the evidence before it. The result is not outside the universe of acceptable computations and is not clearly erroneous.

Carroll also challenges the enhancement he received for his role in his offense under U.S.S.G. § 3B1.1. We review the district court's factual findings for clear error. United States v. Alexander, 59 F.3d 36, 38 (6th Cir.1995). However, the district court's legal conclusions are reviewed de novo. United States v. Williams, 962 F.2d 1218, 1221 (6th Cir.), cert. denied, 506 U.S. 892 (1992).

First, Carroll argues that the district court erred by applying both the enhancement for more than minimal planning under U.S.S.G. § 2F1.1(b)(2)(A) and the enhancement for role in the offense under § 3B1.1(a). As an initial matter, Carroll failed to raise this argument below. Thus, he has waived it unless he shows good cause for not raising it below. United States v. French, 974 F.2d 687, 697 (6th Cir.1992), cert. denied, 506 U.S. 1066 (1993). Carroll makes no such showing and the argument is waived.

Even if Carroll had not waived this argument, he would not prevail. This Court has held that a district court may not apply both the enhancement for more than minimal planning under § 2F1.1(b)(2)(A), and the enhancement for a leadership role in the offense under § 3B1.1(a). United States v. Chichy, 1 F.3d 1501, 1505-07 (6th Cir.), cert. denied, 114 S. Ct. 620 (1993).3 However, Section 2F1.1(b)(2) provides for an enhancement of two levels if the offense involved either (A) more than minimal planning or (B) a scheme to defraud more than one victim. U.S.S.G. § 2F1.1(b)(2). While the district court may not apply both the enhancements for more than minimal planning and for a leadership role in the offense, it may apply both the enhancement for a scheme to defraud more than one victim and the enhancement for a leadership role in the offense. United States v. King, 55 F.3d 1193, 1196-97 (6th Cir.1995); United States v. Aideyan, 11 F.3d 74, 76 (6th Cir.1993). Here, there is no dispute that Carroll's scheme defrauded more than one victim, and the district court properly enhanced the sentence on that basis. Therefore, the district court did not err in applying both §§ 2F1.1(b)(2)(B) & 3B1.1(a).

Next, Carroll argues that he should not receive an enhancement for his leadership role because the conspiracy did not include five or more participants. U.S.S.G. § 3B1.1(a) provides: "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." Carroll does not dispute that he was the leader of the conspiracy; he instead claims that there were only two participants: himself and Ryan. The district court found that there were seven participants, Carroll, Ryan, Sewell, Lawson, Barron Carroll, Cantrell, and Cross.

First, Carroll argues that he should not be counted as a participant. This argument is without merit. Section 3B1.1(a) provides for the enhancement if the "criminal activity ... involved five or more participants." It does not require five or more participants in addition to the defendant. Furthermore, the guidelines define a participant as "a person who is criminally responsible for the commission of the offense, but need not have been convicted." Id. at comment. (n. 1). Carroll meets this definition. Moreover, the ordinary use of the word participant includes the leader or organizer of the activity.

In addition, every circuit which has considered this question has concluded that the defendant counts as a participant within the meaning of § 3B1.1. See United States v. Colletti, 984 F.2d 1339, 1346 (3rd Cir.1992); United States v. Marshall, 976 F.2d 658, 658-59 (11th Cir.1992) (per curiam), cert. denied, 507 U.S. 1009 (1993); United States v. Schweihs, 971 F.2d 1302, 1318 (7th Cir.1992); United States v. Atkinson, 966 F.2d 1270, 1276 n. 8 (9th Cir.1992), cert. denied, 507 U.S. 1004 (1993); United States v. Harry, 960 F.2d 51, 53 (8th Cir.1992); United States v. Fells, 920 F.2d 1179, 1182 (4th Cir.1990), cert. denied, 501 U.S. 1219 (1991); United States v. Reid, 911 F.2d 1456, 1464 (10th Cir.1990), cert. denied, 498 U.S. 1097 (1991); United States v. Barbontin, 907 F.2d 1494, 1498 (5th Cir.1990); United States v. Preakos, 907 F.2d 7, 10 (1st Cir.1990). In addition, this circuit has come to the same conclusion in two unpublished opinions. United States v. Genoa, 47 F.3d 1171 (6th Cir.1994), cert. denied, 115 S. Ct. 2629 (1995); United States v. Newell, 977 F.2d 583 (6th Cir.1992), cert. denied, 113 S. Ct. 1657 (1993). Thus, the district court did not err in counting Carroll as a participant.

Next, Carroll attacks the testimony that the government presented at the sentencing hearing as to the involvement of the other participants. Carroll complains that the government did not produce any of these people as witnesses in court and relied on hearsay testimony. However, the rules of evidence and hearsay do not apply to sentencing hearings. Fed.R.Evid. 1101(d)(3). Instead, hearsay evidence may be admitted if the Court is satisfied that the evidence has sufficient indicia of reliability to support its probable accuracy. U.S.S.G. § 6A1.3(a); United States v. Silverman, 889 F.2d 1531, 1535 (6th Cir.1989). Here, the testimony of Officer Thomas was corroborated by the grand jury testimony of Ryan and the trial testimony of Cantrell, as well as by a conversation Thomas had with Sewell. This gives the testimony sufficient indicia of reliability to be admitted.

Similarly, Carroll argues that the alleged participants had an incentive to lie to Thomas because the government would prosecute them if they did not cooperate. As Carroll so aptly puts it in his brief, however, the issue is credibility. Judging the credibility of evidence is for the district court; this court should not reverse, absent clear error. United States v. Brannon, 7 F.3d 516, 520 (6th Cir.1993). There is no such error here.

In addition, Carroll produced scant evidence to rebut the government's case. He only submitted a hearsay affidavit challenging Cross's involvement. It is within the province of the district court not to credit that affidavit. Moreover, even if the district court were to believe the affidavit submitted by Carroll and not count Cross as a participant, that would still leave six participants counting Carroll.

In summary, the question of who was a participant is a question of fact for the district court, and the district court did not commit clear error in concluding that there were seven participants.

After reviewing the sentence imposed by the district court, we find no error. Accordingly, we AFFIRM the sentence in all respects.