OPINION OF THE COURT
MANSMANN, Circuit Judge.Michael E. Katora and Daniel A. Squire appeal from judgments in a criminal case after a jury convicted them of wire fraud and other related interstate offenses. We are called upon to decide whether subsection 3Bl.l(c) of the United States Sentencing Guidelines, which increases a person’s offense level for his role as an organizer, can apply when the only participants in an offense have shared equal responsibility. We are also called upon to determine whether the district court followed the correct legal standard in calculating the amount of loss resulting from the fraud. Because we conclude that subsection 3Bl.l(c) does not apply unless a defendant has organized, led, managed or supervised another participant, (i.e., another legally culpable person), we will remand the case for resentencing.
After full consideration, we have concluded that the other arguments advanced by Katora and Squire are without merit.1 We will therefore confine our discussion to the applicability of subsection 3Bl.l(e) and to the calculation of loss.
*1400I.
In essence, Katora and Squire telephoned their own phone service thousands of times so as to create the illusion that the service had large, bona fide accounts receivable, and then they sold those illusory accounts.
First, Katora and Squire, as officers of a corporation called Jobs 900, contracted with MCI to obtain several 900-prefix telephone numbers. The 900-prefix telephone service would allow Jobs 900, as a subscriber, to provide information to callers for a fee. MCI would collect the fee from the caller, subtract service charges, then forward the balance to the subscriber (Jobs 900).
In the same office as Jobs 900, Katora and Squire connected automatic dialing equipment to four telephone lines. Renaissance Marketing, a company that Squire controlled, subscribed to regular MCI long-distance service on these four lines. Although Renaissance Marketing and Jobs 900 shared an office, Squire gave a different billing address for Renaissance Marketing.
Meanwhile, Katora and Squire negotiated a contract with James Janota of Commercial Factors. A factor is a firm that buys accounts receivable at a discount.2 As part of Janota’s due diligence, he visited the Jobs 900 office where he observed equipment, a receptionist, and other employees. Janota, Katora, Squire and Kato-ra’s attorney then reviewed the contract, line by line.
In the contract, Katora and Squire gave personal guarantees and warranted that:
16. Each account offered for sale to FACTOR is an accurate statement of a bona fide sale, delivery and acceptance of merchandise or performance of service by CLIENT to customer.
17. CLIENT does not own, control or exercise dominion over, in any way whatsoever, the business of any account/customer to be factored by CLIENT to FACTOR.
R. at 866-69. The contract also set a $250,-000 limit, subject to increase at Commercial Factors’ discretion. After reviewing the contract, Katora’s attorney warned him in a letter that paragraph 16 “could pose some problems for you.” He also addressed Katora’s personal guarantee and whether a bankruptcy creditor could reach Katora’s personal income, which consists of tax-free disability pay.
On October 2, 1990, Katora and Squire began using the automatic dialer to call one of the 900-prefix numbers. MCI would bill Renaissance Marketing for the calls. Except for MCI’s service charges, the amount due from Renaissance Marketing (Squire’s company) would become an account receivable for Jobs 900 (Katora’s and Squire’s company). Each call was the equivalent of an I.O.U., written on MCI stationery, from Renaissance Marketing to Jobs 900.
In the first seven days, Renaissance Marketing’s bill totalled over $200,000, and correspondingly, Jobs 900 had accounts receivable for almost the same amount (after MCI’s service charges). Katora and Squire then raised rates on some of the other 900-prefix numbers. They eventually made some 350,000 calls, worth about $3.6 million. Most of the $3.6 million would appear to be bona fide accounts receivable for Jobs 900; MCI’s service charges on the calls totalled less than $130,000.
On October 22, Katora and Squire gave Janota an MCI report showing receivables due from some of Renaissance Marketing’s calls to Jobs 900. They did not tell Janota that they controlled Renaissance Marketing. Nor did they tell Janota that they had made the calls themselves.
Katora did tell Janota that MCI had approved the factoring arrangement in writ*1401ing, even though MCI had not. At first, Katora claimed that MCI’s signature on the approval did not come through on a fax. Finally, Katora produced a faxed acknowl-edgement from MCI that it had received the approval agreement for purposes of review. Janota faxed the acknowledgment to his supervisor, who authorized him to send a $123,390 check by Federal Express. FBI agents picked up the check at the Federal Express Office and delivered it to Jobs 900.
During a jury trial, Katora and Squire asserted that they made the 350,000 phone calls to test MCI’s ability to generate billing information. They also asserted that they intended to pay MCI with the check from Commercial Factors and to pay Commercial Factors with Katora’s own funds, if necessary. Despite these assertions, the jury convicted each defendant of five counts of wire fraud in violation of 18 U.S.C. § 1343, and four counts of related acts in violation of 18 U.S.C. § 2314 (interstate travel in connection with, and transportation of, a fraudulently obtained check) and 18 U.S.C. § 371 (conspiracy).
In calculating the offense level for both Katora and Squire, the district court began with a base level of 6 pursuant to subsection 2Fl.l(a) of the United States Sentencing Guidelines. The district court increased the base offense level by 2 because the offense required more than minimal planning, see U.S.S.G. § 2F1.1(b)(2)(A), and again by 2 for their roles as organizers, see U.S.S.G. § 3Bl.l(e).
In support of its conclusion that both were organizers, the district court found that “Katora both conceptualized and created Jobs 900 and shared much if not most of the final decision making authority that led to the implementation of the fraud....” (R. at 2298.) The district court also concluded that “Squire shared most of the final decision making authority that lead to the creation and implementation of the fraud.... His role in organizing and staffing [Renaissance Marketing], along with his numerous directives to MCI and Commercial Factors representatives, highlight his role in the offense as an organizer, leader and manager.” (R. at 2229.)
With respect to the amount of loss, the presentence reports had set the amount of loss at $3,615,838, the total bill generated by Renaissance Marketing’s calls. The district court rejected that calculation, recognizing that the $3.6 million figure overstated the amount of loss. (R. at 2306.) Instead, the district court found that the actual or intended loss was $373,600, an amount slightly less than the sum of MCI’s loss ($126,190) plus the factor’s self-imposed limit ($250,000). The court thus increased the offense level by 9 under subsection 2Fl.l(b)(l)(J) for an offense resulting in loss between $350,000 and $500,000.
The court thus set each defendant’s offense level at 19. The resulting sentence range for each was 30-37 months. Consistent with its view that neither defendant was more culpable than the other, the district court sentenced each defendant to the identical term of imprisonment, 34 months.
Katora and Squire offer two challenges to their sentences. First, they argue that the district court should not have applied subsection 3Bl.l(c) of the sentencing guidelines to them. Second, they argue that the district court calculated loss (pursuant to U.S.S.G. § 2Fl.l(b)) under the wrong legal standard. We address each of these challenges in turn.
II.
We have jurisdiction of these appeals from judgments in a criminal case. 18 U.S.C. § 3742(a); 28 U.S.C. § 1291. The district court had subject matter jurisdiction of the prosecution pursuant to 18 U.S.C. § 3231.
We “exercise plenary review over legal questions about the meaning of the sentencing guidelines, but apply the deferential clearly erroneous standard to factual determinations underlying their application.” United States v. Inigo, 925 F.2d 641, 658 (3d Cir.1991).
*1402III.
A.
Part 3B1 of the guidelines directs a sentencing court to adjust a defendant’s offense level based on his role in the offense. See U.S.S.G. Part 3B1, Introductory Commentary. Section 3B1.2 directs the court to decrease the offense level for a minor role. Pursuant to section 3B1.1, the section involved here, the court may increase the offense level for an aggravating role. It provides:
§ 3B1.1 Aggravating Role
Based on the defendant’s role in the offense, increase the offense level as follows:
(a) 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.
(b) If the defendant was a manager or supervisor (but not an organizer or leader) and the criminal activity involved five or more participants or was otherwise extensive, increase by 3 levels.
(c) If the defendant was an organizer, leader, manager or supervisor in any criminal activity other than described in (a) or (b), increase by 2 levels.
U.S.S.G. § 3B1.1.
Having concluded that each of the two defendants was an organizer, leader and manager, the district court applied subsection 3Bl.l(c) to enhance their offense levels. The court based its conclusion on findings that each shared in the “final decision making authority” in the creation and implementation of the fraud, and that Squire had staffed the office with outsiders. The court also found that the fraud involved fewer than five participants and was not otherwise extensive; accordingly, the court did not apply subsection 3Bl.l(a) or (b).
Katora and Squire challenge the district court’s interpretation and application of section 3B1.1. We therefore exercise plenary review. See Inigo, 925 F.2d at 658.
B.
The Introductory Commentary to Part 3B1 of the guidelines informs that section 3B1.1 or 3B1.2 (or neither) may apply when an offense is committed by more than one “participant,” which means more than one “person who is criminally responsible for the commission of the offense, but need not have been convicted.” See U.S.S.G. Part 3B, Introductory Commentary; U.S.S.G. § 3B1.1, Application Note 1 (defining “participant”). Accordingly, we have ruled that sections 3B1.1 and 3B1.2 do not apply to solitary offenders. See United States v. Badaracco, 954 F.2d 928, 934 (3d Cir.1992); United States v. Bierley, 922 F.2d 1061, 1065 (3d Cir.1990). For example, in Badar-acco, we held that the sentencing court should not have enhanced the offense level under subsection 3Bl.l(c) if the offender had acted alone; in Bierley, we held that a solo offender who purchased child pornography from a postal inspector could not receive a downward adjustment under section 3B1.2.
The government argues that section 3B1.1 applies here because more than one participant committed the offense. The application of section 3B1.1, however, requires more than the existence of multiple participants. To apply section 3B1.1, a district court must find that the defendant exercised control over at least one other person. United States v. Fuentes, 954 F.2d 151, 153 (3d Cir.), cert. denied, — U.S. —, 112 S.Ct. 2950, 119 L.Ed.2d 573 (1992). In Fuentes, for example, a defendant who managed a crack house, but not people, could not receive an enhancement under subsection 3Bl.l(b). Fuentes, 954 F.2d at 153-54.
The district court’s findings that Katora and Squire shared responsibility for creating and carrying out the fraud do not indicate that either Katora or Squire organized the other. Rather, the district court’s findings indicate that Katora and Squire were “organizers” only in the sense that they were “planners” of the offense. Just as section 3B1.1 cannot enhance the sentence of a solo offender, see, e.g., Badaracco, 954 F.2d at 934, neither can it *1403enhance the sentences of a duo when they bear equal responsibility for “organizing” their own commission of a crime. We note that to the extent the district court’s findings indicate “organization” in the sense of planning, each defendant received a 2-level increase under 2F1.1(b)(2)(A) for more than minimal planning.
Because the district court found that neither Katora nor Squire directed the other, application of section 3B1.1 would only be appropriate if Katora and Squire had organized a third person. Here, the government did not identify — in its briefs or at oral argument — any third (culpable) participant whom Katora and Squire might have organized. Rather, the government has argued that the defendants’ sentences are subject to enhancement under section 3B1.1 because the defendants organized the non-culpable office staff, who were outsiders to the fraud.
Although the language of section 3B1.1 is not entirely clear, our precedent, when informed by the guidelines’ policy of reducing disparity in sentences, see S.Rep. No. 225, 98th Cong., 2d Sess. 39, reprinted in 1984 U.S.C.C.A.N. 3182, 3183, 3222, dictates that management of a non-culpable party does not warrant application of section 3B1.1.
As a starting point, we note that it is impossible to reconcile the section’s threshold requirement of more than one participant — a requirement imposed by the Introductory Commentary and by precedent— with a definition of management that encompasses non-participants. Such a definition would create an unprincipled disparity between, on the one hand, the sentence of a solo defendant who manages an outsider but who cannot receive the enhancement because of the threshold requirement, and, on the other hand, the sentence of a defendant who manages an outsider but who happens to have a co-participant.
In Badaracco, we held that a solo offender who had involved non-culpable developers in bank fraud could not receive an enhancement under section 3Bl.l(c). 954 F.2d at 934-35. Here, as in Badaracco, we are faced with outsiders to the crime who are essentially victims; one employee worked for eight weeks but never received his wage (R. at 665), the other received a check that bounced (R. at 745). It would contradict the guidelines’ policy of reducing sentencing disparity to exempt a solo offender like Badaracco who managed outsiders but to penalize Katora and Squire for the same conduct simply because they reach the threshold requirement of “more than one participant.”3 If the section were meant to apply to the management of outsiders, then the threshold requirement of “more than one participant” would not exist. The existence of a second, “equally culpable” participant does not provide a principled distinction between the two outcomes.4
*1404Our holding in Fuentes also informs our holding here, for in Fuentes we held, even in a crime that involved more than one participant, that the management of property could not trigger application of subsection 3Bl.l(c). 954 F.2d at 153-54. It would surely create disparate sentencing treatment to exempt the superintendent of a crack house while penalizing the supervisor of a non-culpable employee.5
Similarly, based on Fuentes, the government’s alternate theory—that the use of different corporate forms warrants application of subsection 3Bl.l(c)—must fail. If “management” does not apply to real property, Fuentes, 954 F.2d at 154, then it cannot apply to intangible corporate entities.
The language of section 3B1.1 and of its commentary supports our holding here. The title and first clause of section 3B1.1 indicate that the defendant’s “role in the offense” is the central criterion for application6; size, or “extensiveness,” of the activity is secondary.7 The application notes only authorize use of non-culpable “outsiders” to calculate “extensiveness,” not to determine “role.”8 To be sure, once a *1405court has determined that a defendant’s role among other participants dictates application of 3B1.1, the severity of the enhancement increases as a function of both relative culpability and extensiveness of the crime. See U.S.S.G. § 3B1.1, Background; compare § 3Bl.l(a)-(b) (increasing severity for extensive activity) with § 3Bl.l(c) (enhancing sentence for leadership role in non-extensive offense); compare § 3Bl.l(a) (imposing a severe enhancement for leadership) with § 3Bl.l(b) (imposing an intermediate enhancement for management). There is no indication in the commentary or in the section itself, however, that outsiders may be used to determine a defendant’s role in the offense, especially given that outsiders can never be used to determine a solo offender’s “role in the offense.”
Thus, a common sense reading of sections 3B1.1 and 3B1.2, together with the commentary and caselaw, demonstrates that these two sections are meant to adjust for a defendant’s responsibility in criminal activity relative to other participants. Moreover, the harm of manipulating outsiders could not possibly be the concern of a provision that would not apply to a solo offender who brings about that harm.9 Indeed, in both Fuentes and Bierley we recognized that “adjustments authorized [by §§ 3B1.1 and 3B1.2] for role in the offense are directed to the relative culpability of participants in group conduct.” Fuentes, 954 F.2d at 153; Bierley, 922 F.2d at 1065; see also United States v. Phillips, 959 F.2d 1187, 1191 (3d Cir.1992) (for § 3B1.1 to apply, defendant must have exercised some control over others involved in commission of offense); United States v. DeCicco, 899 F.2d 1531, 1535 (7th Cir.1990) (“§ 3B1.1 to apply only to situations where the offender organizes or leads criminally responsible individuals”); United States v. Carroll, 893 F.2d 1502, 1509 (6th Cir.1990) (§ 3B1.1 requires at least two culpable individuals so that leadership, however minimal, can be claimed); Badaracco, 954 F.2d at 934 (citing DeCicco and Carroll).10
In summary, we conclude that the application of sections 3B1.1 and 3B1.2 has two prerequisites: multiple participants and some differentiation in their relative culpabilities. Once these prerequisites are found to exist, a court will apply subsection 3Bl.l(a) or (b) to enhance the more culpable participants’ offense levels so long as there were five or more participants or so long as the criminal activity was “otherwise extensive.” Alternatively, the court may apply subsection 3Bl.l(c) if there were fewer than five participants and the criminal activity was not otherwise extensive.
Here, because the only two participants were equally culpable and, furthermore, because they did not organize, lead, manage or supervise a third participant, neither section 3B1.1 nor section 3B1.2 could apply. Therefore, we find that the district court committed an error of law in applying subsection 3Bl.l(c) to the defendants.
IV.
Subsection 2Fl.l(b) directs a court to increase a defendant’s offense level in *1406proportion to the amount of loss caused by a fraudulent offense. U.S.S.G. § 2Fl.l(b)(l)(A)-(S) (table). The application notes indicate, in various places, that “Loss” means the value of property taken, damaged, or destroyed; that the loss need not be determined with precision; that a court should consider whether actual loss overstates or understates the seriousness of the defendant’s conduct; and that the offender’s gain will usually underestimate the loss. See U.S.S.G. § 2B1.1, Application Note 2; § 2F1.1 Application Notes 7-8; § 2X1.1, Application Note 4. See generally United States v. Kopp, 951 F.2d 521, 526-536 (3d Cir.1991) (discussing calculation of loss in fraud cases).
In Kopp, we held:
[F]raud “loss” is, in the first instance, the amount of money the victim has actually lost (estimated at the time of sentencing), not the potential loss as measured at the time of the crime. However, the “loss” should be revised upward to the loss that the defendant intended to inflict, if that amount is higher than actual loss.
951 F.2d at 536.
The district court found the loss to be $373,600 — slightly less than the sum of the $250,000 factoring limit plus MCI’s total unpaid service charges ($126,190.21). There was sufficient evidence for the district court to determine that Katora and Squire inflicted a loss of at least $123,000 on MCI, and there was sufficient evidence that Katora and Squire intended to inflict a loss of at least $250,000 on Commercial Factors. See United States v. Kikumura, 918 F.2d 1084, 1099 (3d Cir.1990) (preponderance standard). The MCI affidavit sufficiently proves MCI’s loss of $126,190.21. An intended loss of at least $250,000 can be reasonably inferred from the following evidence. Katora and Squire generated over three million dollars in false accounts receivable. (R. at 631.)11 Prior to that, they had attempted to negotiate a factoring arrangement of up to two million dollars. (R. at 860-66.) They settled for an agreement that limited the amount to be factored at $250,000, but the agreement also left open the possibility of an increase in the $250,-000 limit at Commercial Factors’ discretion. (R. at 860-66.) The only explanation Kato-ra and Squire have offered for generating the calls is the incredible story that they were “testing” MCI. Thus the only logical inference is that Katora and Squire intended to exchange the receivables for cash from Commercial Factors, at least to the contractual limit of $250,000, and perhaps beyond that amount if Commercial Factors raised the limit. This inference is supported by Squire’s own statement to Janota after their last meeting: “We will see you next week with additional business.” (R. at 894.) Given the above facts and reasonable inferences, we cannot say it was clearly erroneous for the district court to find that Katora and Squire intended to inflict a loss of $250,000 on Commercial Factors.12
Katora and Squire premise a number of their challenges to this finding on their assertion that they intended to repay MCI and Commercial Factors. The district court, however, was not bound to believe them and obviously did not. Indeed, the government presented evidence that the money advanced from Commercial Factors was not going to MCI, but would be used to pay salaries, rent, insurance, and other bills. (R. at 2191-92 (government’s argument at sentencing).) Thus, we cannot accept their argument that MCI’s loss must be attributed to the FBI (for shutting down Jobs 900) rather than to themselves (for intent not to pay).
Similarly, we cannot accept their argument that their personal guarantees to repay Commercial Factors were somehow *1407more reliable than their contractual warranties that they would sell bona fide receivables from unrelated customers. This situation thus differs from Kopp, 951 F.2d at 521, in which we required a district court to consider the value of a mortgage on real property used to fraudulently obtain a loan. Here, in contrast to an ascertainable market value on real property at issue in Kopp, the personal guarantees did not have a market value given Squire’s inability to pay and Katora’s inquiries into the protections of bankruptcy.
Finally, although Katora and Squire are correct in arguing that the $123,000 check represents a measurable loss that Commercial Factors would have actually suffered, there was evidence that Katora and Squire intended to return to Commercial Factors the next week for “more business” (R. at 894), that is, further intended loss. As we remarked above, the district court reasonably could have concluded that Katora and Squire intended to inflict losses on Commercial Factors up to the $250,000 limit. See Kopp, 951 F.2d at 536 (instructing courts to use higher of actual or intended loss).
In summary, the district court did not err as a matter of law or fact in finding that MCI and Commercial Factors were victims of a fraud and in calculating their “loss” under subsection 2F1.1(b)(1) to be the sum of the actual loss suffered by MCI and the loss intended by Katora and Squire for Commercial Factors.
Y.
For the foregoing reasons, we will remand the case to the district court for resentencing without a 2-level enhancement pursuant to subsection 3Bl.l(e) of the United States Sentencing Guidelines. With respect to all other issues, we will affirm the judgments of the district court.
. Katora and Squire contended that there was insufficient evidence to support a conviction; that there was an improper variance between the indictment and proof presented at trial; that the district court erred when it refused to allow evidence of MCI’s factoring program; that the district court erroneously allowed a witness without personal knowledge to testify; that the district court erred in denying Katora’s motion for a new trial; and that the judge incorrectly charged the jury regarding intent to repay in fraud cases. Katora also challenged the judge’s order that he pay restitution. In addition, Squire challenged the district court’s denial of his motion to join in Katora’s motion for a new trial.
. For example, after shipping $100,000 worth of widgets, a widget manufacturer might take the $100,000 in accounts receivable to a factor. The factor would buy the right to collect the $100,-000. The amount the factor would pay its client (the manufacturer) depends on the time-value of money, on the likelihood that the client’s customers will pay for the widgets, and on the factor's own profit margin. Thus the factor might pay $50,000 to the manufacturer for the right to collect $100,000 later. If the factor were to receive the full $100,000 from the manufacturer's customers, the factor might then also pay an additional sum to its client, the manufacturer.
. The dissent essentially relegates Badaracco to a footnote and, while arguing about disparity, downplays the simple disparity created by penalizing these defendants but not Badaracco.
. In contrast to our view of disparity, which is premised on the disparity between punishing these defendants but not a solo offender, the dissent's disparity argument is premised on the view that the management of other criminal participants is no worse than the management of non-culpable outsiders. Although the latter is reprehensible, it is certainly logical to distinguish the two. Arguably, the management of participants presents a greater danger to the public because, unlike a non-culpable party, even a minimally informed participant can take individual action to accomplish a crime. But regardless of whether one type of management is worse, our view of disparity is premised on § 3Bl.l's inability to punish identical conduct (management of outsiders), whereas the dissent’s view is premised in part on the section’s inability to punish two different, albeit similar, forms of conduct (management of participants versus management of outsiders).
Thus the dissent's example of "disparity” of the sentences of two defendants, defendant A who managed an unindicted participant and defendant B who managed a non-culpable party, fails because — although A would receive the increase and B would not — their predicate behavior differs. (The dissent also ignores that Ba-daracco would prohibit applying § 3Bl.l(c) to defendant B in any event.)
The rest of the dissent’s concern for "disparity" stems from the unfairness that results from imposing no greater sentence on the "organizer” of outsiders than on the non-organizer. However well-founded the dissent’s concern may be, we find the dissent’s solution — that of extending the application of § 3Bl.l(c) to those deemed *1404deserving of additional punishment—to be inconsistent with Badaracco in which we declined to apply § 3Bl.l(c) to the solo organizer of outsiders, who presumably deserved greater punishment than a comparable "non-organizer."
. The distinction made by the dissent between Fuentes and the case at bar is based on the obvious difference between manipulation of things (which are always "innocent”) and manipulation of innocent people. The central fallacy of this distinction is that it makes no more sense to evaluate an offender’s responsibility in relation to a innocent non-participant than it does to evaluate an offender's responsibility in relation to “innocent" tangibles or intangibles. In either case the participant will be deemed “more” responsible because, by definition, neither things nor non-participants are ever culpable.
The logical extension of the a distinction between things and people would lead to some rather disparate results. For example, offenders who rent a car to transport stolen goods across the state line would not be penalized, but the same offenders would if they direct a cab driver instead; offenders who conduct mail fraud from their home would not be penalized, but the same offenders would be penalized if they hired an innocent teenager to lick stamps and carry mail bags.
Indeed, because the employees manipulated by Katora and Squire were themselves victims of the fraud, the dissent's approach may lead to even more bizarre results: bank robbers "supervising” tellers who put cash in bags and extortionists "organizing” their victims into compliance would both be entitled to enhancement under the dissent's theory.
Furthermore, the dissent merely dismisses as dicta the underpinnings of our reasoning in Fuentes that "[w]hen the terms 'manage' and 'supervise' are applied to the management or suppression of other persons, they serve well the purpose of assigning relative fault among criminal coadventurers." Id. at 153 (emphasis added). In no sense can a non-participant be considered a coadventurer, and, as demonstrated in the examples above, it is as pointless to assess an offender’s relative fault by reference to the innocent as it is to assess it by reference to things. Either assessment inevitably leads to assessment of greater relative fault.
. The dissent does not discuss the significance of the title and first clause of § 3B1.1.
. The Background to the section confirms that the adjustment for role in the offense "is included primarily because of concerns about relative responsibility.” U.S.S.G. § 3B1.1 Background. Although the dissent opines that the Background suffers from “a general lack of clarity,” Typescript at 20, we take the word "primary" at face value. Thus, notwithstanding other concerns expressed in the Background, it clearly sets a priority on relative responsibility.
Contrary to the dissent’s reading, the second paragraph of the Background does not retreat from the importance of relative responsibility. Rather it merely de-emphasizes the necessity for a strict hierarchy in smaller criminal enterprises.
Our threshold requirement flows from the obvious importance of relative responsibility coupled with the threshold requirement of “more than one participant.” If the guideline were meant to address equally the "multi-facet-ed” concerns listed by the dissent, then it would presumably apply to solo offenders. The dissent concedes that it does not.
. The dissent’s reliance on the use of "others” in Application Note 3, which lists a defendant's "control of others” as a factor to determine leadership, is misplaced. The word "others” is ambiguous, meaning either "other participants” or "other people."
Moreover, by ignoring the first sentence of Application Note 3, the dissent ignores the probability that “control of others” is only a factor to be considered “[i]n distinguishing a leadership or organizational role from one of mere management or supervision,” not a factor to be *1405considered in defining who is an organizer or manager or to whom § 3B1.1 applies initially. Similarly, by ignoring the third sentence of Application Note 3, the dissent ignores the indication that the section applies to "a criminal association or conspiracy,” groups in which outsiders are not counted.
. The dissent relies on the obvious harmfulness of involving innocent outsiders to argue that the guidelines’ policy of punishing harmful conduct counsels an inclusive reading of section 3B1.1. To the contrary, the dissent’s reliance on “harm" simply begs the question, “To which particular harm is section 3B1.1 addressed?” The threshold requirement of more than one participant— imposed not by this opinion but by the guidelines and precedent — would suggest that the specific harm addressed is the role among participants in the offense.
. The dissent eschews what amounts to reasonable interpretation of § 3B1.1 — that § 3B1.1 increases the sentence for the more culpable participants in a criminal conspiracy in proportion to the participant’s role and the extensiveness of the crime. Instead, the dissent chooses a nebulous "multifaceted” approach that — despite a lengthy treatment — finds little or no real support in the text of the section or its commentary, offers no concrete criteria for determining when a non-culpable party has been "organized, led, managed, or supervised,” and ultimately gives the government an automatic two-level enhancement whenever it can find a warm body in the vicinity of a crime.
. Katora and Squire objected to the $3.6 million figure only insofar as it was used in the presentence report to calculate loss. R. at 2217, 2256.
. Indeed, in his response to the presentence report Katora primarily argued that MCI was not a victim and did not suffer a loss. (R. 2252-60.) Although Katora argued that Commercial Factors' loss was less than $250,000, Katora appeared to acknowledge that "the only reasonable interpretation of the maximum loss [to Commercial Factors] was $250,000." (R. at 2260.)