The opinion
CLARK, Circuit Judge,
dissenting:
The majority concludes that the district court’s analysis of the procedural and substantive antitrust issues in this case is “thorough, complete, and correct.” I dissent. 1
1. Facts
The facts in this appeal are relatively simple. From 1976 to the present, defendants BRG of Georgia (BRG) and its owner Ronald Pelletier have offered a comprehensive bar review course in Georgia. 2 The course reviews both the multistate and Georgia portions of the Georgia bar exam through written materials and live and videotaped lectures by attorneys and law professors. During 1979, BRG was utilizing free multistate materials that West Publishing Company was test marketing.
*1429 Harcourt Brace Jovanovich Legal and Professional Publications (HBJ) sells comprehensive bar review courses in forty states and is the largest provider of bar review materials and lecture services in the country. 3 HBJ produces and uses its own multistate materials. HBJ began offering a Georgia review on a limited basis in 1976. During the period 1977-79, BRG competed directly with HBJ in the provision of comprehensive bar review courses in Georgia. 4 BRG and HBJ were the two dominant providers of bar review courses in Georgia at this time.
Beginning in 1979, BRG conducted an intensive advertising campaign against HBJ which HBJ alleges included false representations about the HBJ course. Vigorous price competition between BRG and HBJ resulted in driving the price of their bar review courses into the range of about $150 per student. The rivalry between BRG and HBJ was rancorous and vitriolic. Each accused the other of engaging in illegal tactics. According to Richard Conviser, head of HBJ’s Legal and Professional Publications, HBJ incurred a loss of $45,-000 on revenues of $120,000 in the 1979 price war. At some point during 1979, West Publishing decided to begin selling its multistate materials in bookstores; West therefore decided to charge BRG for the formerly free multistate materials.
Conviser asserts that HBJ unilaterally decided to withdraw from the Georgia market at some future unspecified date after sustaining the 1979 losses and following the death of the attorney who had operated HBJ bar review courses in Georgia. No documentation of this decision, however, exists. Conviser also asserts he was unaware of West’s withdrawal of its free multistate materials from BRG at the time HBJ purportedly decided to withdraw from Georgia.
In early 1980, Conviser and Pelletier got together during a time when BRG and HBJ were still selling bar review courses in Georgia. 5 Within a few months, they entered a written agreement on April 22, 1980 (“1980 agreement”) which gave BRG the exclusive license to market HBJ’s multistate materials in Georgia and the exclusive right to use the Bar/Bri tradename in Georgia. The agreement also contained a provision which required that HBJ not compete with BRG in Georgia and that BRG not compete with HBJ outside of Georgia. 6 Under the agreement, HBJ receives $100 per student enrolled and 40% of all revenues over $350. Soon after their agreement, BRG and HBJ announced in their advertising literature that because of their “combination, Bar/Bri Georgia students will, in essence, have the ‘best of both worlds.’ ”
Immediately following the 1980 agreement, the price of BRG’s course increased from about $150 to over $400. Law students who enrolled in the BRG course and paid the increased bar review course prices *1430 during the period April 22, 1980 to June 15, 1984, filed an antitrust suit against BRG, HBJ, Conviser, Pelletier, and others. This class action resulted in a settlement which provided class members with partial refunds. While the lawsuit was pending in 1982, BRG and HBJ modified their 1980 agreement by removing the express covenants not to compete and granting BRG the nonexclusive right to use Bar/Bri materials in Georgia (“1982 agreement”). BRG, however, retained the exclusive right to use the Bar/Bri tradename in conjunction with its bar review course in Georgia.
Despite these modifications, HBJ has not competed with BRG in Georgia and has not licensed its multistate materials for use by any other bar review course in Georgia. In addition, BRG has not competed against HBJ outside the Georgia market. The price of the BRG review course has continued to increase also. In 1985 the list price escalated to approximately $825. A group of law students who enrolled in the BRG course in Athens, Georgia from June 15, 1984 to the present initiated this class action alleging that BRG, HBJ and Pelletier engaged in various antitrust violations. 7
The law-student plaintiffs in this antitrust action could prove that BRG and HBJ violated the antitrust laws in a number of ways. First, they could demonstrate that the 1980 agreement between BRG and HBJ was a per se violation of the antitrust laws and that the conspiratorial agreement continued despite the defendants’ 1982 contractual modifications. Next, if the per se claims fail, they could demonstrate that the 1980 and modified 1982 agreements were unreasonable under the rule of reason. They could also establish that BRG and HBJ conspired to or attempted to monopolize, or succeeded in monopolizing, the Georgia market for Georgia bar review courses.
The district court held that the 1980 agreement was not illegal under the per se rule; it therefore did not decide whether the BRG-HBJ conspiracy continued. The district court also found that neither the 1980 agreement nor the modified 1982 agreement violated the rule of reason. 8
II. Standard of Review
My disagreement with the majority begins with its adoption of an improper standard of review in this summary judgment action. In its recitation of the law on antitrust summary judgment, the majority relies on Matsushita Elec. Indust. Co. v. Zenith Radio Corp., 475 U.S. 574 , 106 S.Ct. 1348 , 89 L.Ed.2d 538 (1986), a predatory pricing action. In Matsushita the Court held that in an action based on a predatory pricing theory “conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy.” Id. at 588 , 106 S.Ct. at 1357 . The plaintiff must therefore provide additional evidence demonstrating that the defendant’s conduct was inconsistent with competition. Matsushita’s holding is founded on the practical difficulties of differentiating between legitimate and illegitimate business practices (i.e. competitive pricing versus predatory pricing) when the plaintiff’s asserted antitrust theory is speculative.
Matsushita is one of two recent Supreme Court decisions which have clarified the applicable standard of review in antitrust summary judgment actions which are based on speculative economic theories. See also Monsanto v. Spray-Rite Service Corp., 465 U.S. 752 , 104 S.Ct. 1464 , 79 *1431 L.Ed.2d 775 (1984) (vertical price fixing action brought by a terminated distributor). Both Matsushita and Monsanto presented the Court with the difficult task of defining the proper evidentiary standard for determining under what factual circumstances an illegal antitrust conspiracy may be inferred from parallel conduct and other circumstantial evidence where direct evidence of conspiracy is lacking. 9
Both Matsushita and Monsanto, however, are distinguishable from the instant action. Neither case involved a consumer action based upon well-established horizontal per se and rule of reason theories where direct evidence establishes an explicit written agreement between two competitors allocating markets and interfering with independent price setting. Instead, the antitrust claims in Matsushita and Monsanto relied on competitors’ predatory pricing and vertical price maintenance theories, respectively, which antitrust scholars and judges have criticized for ensnaring legitimate business practices. 10 It is therefore doubtful whether the standards announced in Matsushita and Monsanto apply in situations, such as the instant action, where the direct evidence of concerted action is manifest in explicit written agreements between dominant firms allocating and monopolizing the market and interfering with independent price setting. The 1980 BRG-HBJ agreement is a classic example of the most blatantly anticompetitive practices in which horizontal competitors can engage. 11 There is no credible scholarly debate that geographic market allocation and horizontal price-fixing between two dominant firms in an industry are legitimate procom-petitive business practices. Thus, the substantive antitrust presumptions that controlled the Court’s holding in Matsushita and Monsanto are inapplicable to these types of horizontal agreements whose pro-competitive virtues are minimal or non-existent.
The standard of review in this summary judgment action is that announced in Celotex v. Catrett, 477 U.S. 317 , 106 S.Ct. 2548 , 91 L.Ed.2d 265 (1986) and Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986). Under Fed.R.Civ.Proc. 56(c), this court must grant summary judgment for defendants if the plaintiffs “fail to make a showing sufficient to establish the existence of an element essential to [their] case, and on which [they] will bear the burden of proof at trial.” Celo-tex, 477 U.S. at 322, 106 S.Ct. at 2552-53. In determining whether the plaintiffs have met their burden, we must view all the evidence, and inferences drawn from such evidence, in the light most favorable to the plaintiffs and resolve all justifiable inferences in their favor. Anderson, 477 U.S. *1432 242, 253 , 106 S.Ct. 2505, 2513 , 91 L.Ed.2d 202 (1986); Helicopter Support Systems v. Hughes Helicopter, Inc., 818 F.2d 1530, 1532 (11th Cir.1987). One particular piece of evidence, viewed in isolation, may not be sufficiently probative to justify sending a case to the jury; a review of all the evidence collectively, however, may justify such a result. In addition, we may not weigh conflicting evidence to resolve disputed factual issues. Instead, if a genuine issue of material fact exists, we must deny summary judgment. Helicopter Support Systems, 818 F.2d at 1532 (moving party has burden of demonstrating the absence of a genuine issue of material fact viewing the evidence in the light most favorable to the non-moving party; “A court must deny summary judgment if reasonable minds could differ as to the factual inferences to be drawn from undisputed facts”).
A thorough review of the entire record indicates that the plaintiffs have produced sufficient evidence to survive summary judgment on their price-fixing, market allocation, rule of reason, and monopolization claims under applicable Supreme Court precedents. Each of these claims relies upon well-established economic theories. 12 Further, they support their claims with direct evidence of concerted action (including an explicit written agreement) that excludes the possibility of legitimate independent action. In addition, they have provided substantial expert economic analysis which demonstrates the adverse economic consequences of the defendants’ conspiratorial agreement.
III. Per Se Claims
The district court’s analysis of the plaintiffs’ per se claims 13 is contrary to well-established antitrust case law. Under binding Supreme Court precedent, the 1980 agreement between HBJ and BRG falls within one of two recognized per se categories: market allocation and price-fixing. 14 The district court, however, misapplied the proper test on both claims. In addition, there is sufficient evidence of an unlawful agreement and continuing conspiracy between the defendants to survive a summary judgment motion.
A. HBJ and BRG are Horizontal Competitors
As an initial matter, HBJ and BRG argue that the plaintiffs’ horizontal per se theories do not apply to them because they are in a vertical supplier/retailer relationship. They further urge that HBJ’s asserted unilateral decision to withdraw from the Georgia market in early 1980 instantaneously transformed HBJ’s relationship from a horizontal to vertical one thus precluding horizontal per se theories. These arguments, however, are simply disingenuous and mer-itless.
It is firmly established that entities in a seemingly vertical relationship may be capable of horizontal restraints if they are actual or potential competitors. 15 During *1433 the period 1976-1980, both HBJ and BRG were direct, fierce and antagonistic competitors in the Georgia bar review course market. Because the HBJ and BRG courses were separate and distinguishable brands of comprehensive bar review courses, BRG and HBJ were horizontal competitors engaging in inter-brand competition. 16 In addition, HBJ and BRG have always been and continue to be at least potential competitors. 17 Simply stated, BRG and HBJ cannot have it both ways — either they are horizontal competitors because HBJ can reenter the Georgia marketplace and compete with BRG or they are co-conspirators acting in concert to prevent such competition between themselves.
Further, HBJ and BRG’s assertion that HBJ’s “withdrawal” from the Georgia market somehow altered BRG’s and HBJ’s status as actual or potential competitors is disingenuous. Their argument would essentially nullify the per se rule because horizontal competitors could avoid antitrust liability by simply entering into anticompet-itive agreements that have vertical aspects. Instead, a court must view an arrangement’s economic substance rather than its form.
The rationale for each per se rule is an economic analysis of the agreement, an analysis of the potential economic advantages which might motivate the parties to a particular type of agreement. A per se rule is applicable to a particular case if and only if the economic analysis which justifies the rule applies to the particular case.
Abadir & Co. v. First Mississippi Corp., 651 F.2d at 426. Given that HBJ vigorously competed and then joined forces with BRG following meetings between Pelletier and Conviser, the indisputable conclusion is that BRG and HBJ were, and continue to be, capable of engaging in per se horizontal restraints.
B. Market Allocation
The district court erroneously concluded that to prove a per se violation under a geographic market allocation theory the plaintiffs had to show that BRG and HBJ have sub-divided some relevant market in which they previously competed. 18 Because BRG and HBJ did not divide the Georgia marketplace between themselves, the trial court reasoned that no Sherman section 1 violation occurred. The court, however, made an erroneous legal distinction between the terms “divide” and “allocate.”
Well-established case law demonstrates that the allocation of markets or sub-markets by competitors is a per se antitrust violation. In United States v. Topco Associates, Inc., 405 U.S. 596 , 92 S.Ct. 1126 , 31 L.Ed.2d 515 (1972), the Supreme Court held that agreements between competitors to allocate territories to minimize competition are per se unlawful:
One of the classic examples of a per se violation of § 1 is an agreement between competitors at the same level of the market structure to allocate territories in order to minimize competition.... This Court has reiterated time and time again that “ [horizontal territorial limitations ... are naked restraints of trade with no *1434 purpose except stifling of competition.” Such limitations are per se violations of the Sherman Act.
Id. at 608, 92 S.Ct. at 1133-34 (emphasis added) (citations omitted). See also E. Sullivan & J. Harrison, Understanding Antitrust and Its Economic Implications, § 4.14 at 111 (1988) (“A horizontal market division is created when competitors agree not to compete in a designated market.”). The defendants in Topco did not divide a market in which they had previously competed; they simply agreed to allocate markets. See also Gainesville Utilities v. Florida Power & Light Co., 573 F.2d 292, 299-300 (5th Cir.1978) (horizontal market division per se violation), cert. denied, 439 U.S. 966 , 99 S.Ct. 454 , 58 L.Ed.2d 424 (1978). Thus, market division is simply a subset of market allocation both of which are per se antitrust violations. 19
In this action, HBJ and BRG allocated designated markets: BRG received the Georgia market while HBJ received the balance of the United States. Each agreed not to compete in the other’s respective market. The district court therefore erred in failing to classify this market allocation agreement as a per se violation.
C. Price-Fixing Claim
The plaintiffs also claim that BRG and HBJ entered into their 1980 agreement with the primary purpose of eliminating price competition between themselves in the Georgia marketplace. They assert that the license fee and revenue-sharing agreement between BRG and HBJ is the mechanism that facilitates the defendants’ purpose in raising the price of bar review courses and distributing the resulting profits. BRG and HBJ disavow any intent to restrain trade and claim that their agreement is “nothing more than an ordinary copyright royalty arrangement” which courts have “routinely sustained.” The district court held there was no per se violation because the agreement neither explicitly set prices nor was it inherently anti-competitive. 20
The district court’s analysis of the plaintiffs’ price-fixing claim is incomplete in at least two important respects. First, the fact that the agreement between the defendants did not explicitly address pricing and HBJ did not have the express right under the agreement to be consulted about the prices BRG charged for bar review courses, does not exclude the possibility of a price-fixing violation. Second, the district court erred in finding that the revenue-sharing agreement was not inherently anticompetitive because the record establishes that the purpose and effect of the agreement was to increase the price of bar review courses.
It is clear that a horizontal price-fixing violation does not require that the defendants literally fix prices. Instead, “[u]nder the Sherman Act a combination formed for the purpose and with the effect of raising, depressing, fixing, pegging, or stabilizing the price of a commodity in interstate or foreign commerce is illegal per se.” United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 223 , 60 S.Ct. 811, 844, 84 , L.Ed. 1129 (1940). For instance, price-fixing results when competitors agree to some non-price restrictions that directly and adversely affect the market prices for their goods or services. 21 Such indirect methods of *1435 raising prices injure consumers as much as direct, overt price setting.
In contrast to such indirect methods, the combination of the defendants’ licensing and revenue-sharing agreements has a direct upward effect on price because it requires BRG to set the price of its bar review course above $350 before revenue-sharing begins. It also totally eliminates HBJ’s incentive to compete with BRG in Georgia. First, HBJ has little incentive to compete with BRG because any resulting price competition would reduce HBJ’s profits under the agreement’s revenue-sharing provision. Second, BRG has the exclusive right to use the Bar/Bri tradename in Georgia thus precluding HBJ from using its own multistate materials and its own tradename. The combination of the licensing and revenue-sharing provisions have the effect of stifling competition and maintaining BRG’s (and indirectly HBJ’s) dominance in Georgia.
The second error in the district court’s analysis relates to the possible justifications for the defendants’ conduct. The district court correctly noted that because some horizontal agreements have procom-petitive potential, not all horizontal agreements among actual or potential competitors that have an impact on price are per se violations. For example, some joint ventures and other cooperative arrangements are not per se violations because an agreement on price is necessary to market the product. 22
In determining whether a challenged practice is a per se violation or subject to a rule of reason, a court must inquire into whether the practice always or almost always tends to raise price and restrict output or instead is likely to assist the creation of economic efficiency. Rigid line drawing must be avoided and close attention given to procompetitive, efficiency-creating integration that is accomplished as the result of an outwardly anticompetitive, yet ancillary, restraint. National Bancard Corp. v. VISA, U.S.A., Inc., 779 F.2d 592 (11th Cir.1986), cert. denied, 479 U.S. 923 , 107 S.Ct. 329 , 93 L.Ed.2d 301 (1986).
Here, in contrast, the record is devoid of any evidence that the agreement between HBJ and BRG was ancillary or that it had some procompetitive, efficiency-creating potential. Furthermore, there is no evidence that it was necessary for HBJ and BRG to combine their resources to provide a bar review course in Georgia particularly when they had competed for several years prior to their price war. The use of restrictive covenants in the 1980 agreement further buttresses the inference that the defendants’ purpose in allocating the Georgia market was to reduce competition between themselves and ultimately to raise prices and reduce output. 23
It is worth reiterating that the 1980 agreement totally eliminated price and inter-brand competition between HBJ and BRG in the Georgia market and there are no apparent economies of scale or other cost savings to counterbalance this anti-competitive result. HBJ could have sold its materials to BRG without entering such a blatantly anticompetitive agreement. 24 Instead, the two principal competitors in the Georgia market have accomplished through their written agreement and subsequent actions what the antitrust laws were designed to prevent: injury to the consumer.
The evidence establishes that the 1980 agreement between BRG and HBJ is per se illegal under binding Supreme Court prece *1436 dent. The written agreement has the effect of reducing price competition in Georgia and markets into which BRG might have otherwise entered absent the agreement. The agreement also has no redeeming procompetitive virtues. Thus, the trial court erred by too narrowly construing the rule against horizontal price fixing and should have condemned the defendants’ agreement as being per se illegal. 25
IV. Rule of Reason Claim
Although the 1980 agreement is a per se violation, the plaintiffs may also show that the 1980 agreement and its subsequent 1982 modifications violate the rule of reason. The district court found that neither the 1980 nor the reformulated 1982 agreements violated the rule of reason because the plaintiffs failed to demonstrate any actual anticompetitive effects. The district court erred in its findings and also failed to apply the proper analysis under the rule of reason.
Under the rule of reason, a “factfinder weighs all of the circumstances of a case in deciding whether a restrictive practice should be prohibited as imposing an unreasonable restraint on competition.” Continental T.V. Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 49 , 97 S.Ct. 2549, 2557 , 53 L.Ed.2d 568 (1978). Although the rule of reason will often require a careful balancing of a challenged restraint’s harms, benefits, and alternatives, a court can readily condemn a challenged restraint on summary judgment when the restraint directly limits competition on price or output and has no procompetitive justification. P. Areeda, Antitrust Law, ÍÍ1508, at 403 (1986); see also NCAA, 468 U.S. at 109, 104 S.Ct. at 2964 . Under the rule of reason, a sliding scale has developed which permits antitrust plaintiffs to demonstrate a lower threshold level of anticompetitive effect the more egregious the defendants’ challenged practice or restraint. A demonstration of anticompetitive effect or market power, however, may not be necessary in cases where the challenged restraints are sufficiently “naked” in the sense that they directly restrain price or output.
A. The Rule of Reason Under Indiana Federation of Dentists
Under the rule of reason as enunciated in FTC v. Indiana Federation of Dentists, 476 U.S. 447 , 106 S.Ct. 2009 , 90 L.Ed.2d 445 (1986), 26 the plaintiffs in this action can *1437 demonstrate an antitrust violation in one of two ways. The first requires that the plaintiffs demonstrate that the 1980 agreement between BRG and HBJ imposed a “naked" restriction on output or price; in this instance, the plaintiffs would not have to define a relevant market or show market power. BRG and HBJ, however, would have to provide a countervailing proeom-petitive virtue to avoid liability. Under the second method, if the agreement is not sufficiently “naked,” the plaintiffs can show actual detrimental effects, such as a reduction of output or increased price, instead of an inquiry into market power.
B. BRG-HBJ Agreement is a “Naked” Restraint
Under the first method, the 1980 BRG-HBJ agreement is a blatant “naked” restriction on output because it allocated markets and thus prevented HBJ from selling its bar review materials to any course provider other than BRG in the Georgia market. This along with the exclusive tradename provision perpetuated BRG’s, and indirectly HBJ’s, economic dominance in the Georgia bar review course market. The plaintiffs therefore did not have to provide evidence of a relevant market or market power to prevail on their claim. Instead, the burden shifted to defendants to provide evidence of some procompetitive efficiency-enhancing virtue to their agreement, which they failed to do. The district court erred in holding that the BRG-HBJ agreements were competitive under the rule of reason.
C. Anticompetitive Effects
Under the second method, the plaintiffs must prevail because they have demonstrated actual detrimental effects: 27 the price of bar review courses has increased dramatically with no sufficient cost-based explanation and BRG and HBJ’s non-competition agreement has reduced consumer choice and reduced output.
The plaintiffs introduced evidence of a dramatic price increase which immediately followed the defendants’ 1980 agreement. The district court rejected the conclusion that this price increase was an anticompeti-tive result of the defendants’ agreement. 28 A review of the record, however, indicates that a genuine issue of material fact regarding actual and continuing anticom-petitive effects exists thus precluding summary judgment for the defendants on the plaintiffs’ rule of reason claims. 29
1. Price Increases
The record indicates that the price of BRG’s bar review course in Georgia has increased from $150 in 1979 to a list price of $825 in 1985. This increase in price is obviously detrimental to consumers of Georgia bar review courses yet the district court held that the increase does not necessarily establish anticompetitive effect because BRG’s costs may have increased because West Publishing no longer made its *1438 free multistate materials available. 30 The court’s analysis, however, neglects the anti-competitive effects inherent in the defendants’ licensing fee and revenue-sharing provisions contained in both the 1980 and 1982 agreements.
The district court is correct in concluding that the possibility of anticompetitive effect exists and that the price increase of BRG’s bar review course may be attributable, in part, to the cost of multistate materials that BRG must now bear due to West Publishing’s decision to charge for its previously free materials. 31 These increased costs, of course, are increased revenues for HBJ. Under their agreements, BRG pays HBJ $100 per student enrolled and BRG and HBJ jointly share any price increases above $850. Thus, any price increase imposed on BRG bar review consumers above $350 results in increased revenues to both BRG and HBJ. For instance, for each bar review course BRG sells at the 1985 list price of $825, HBJ receives $190 (40% of $475) plus the $100 per student licensing fee. At the $825 list price, HBJ receives a total of $290 per course sold while BRG receives the balance of $535 per course. The revenue-sharing provision, therefore, minimizes HBJ’s incentive to compete with BRG because any resulting price competition would reduce HBJ’s profits under the agreement. Similarly, the revenue-sharing provision also minimizes HBJ’s incentive to license its multistate materials to other bar review courses in Georgia because such courses would compete with BRG and further reduce HBJ’s profits.
Thus, it is a disputed factual issue whether the price increases resulted from reduced competition following HBJ’s “withdrawal” or from BRG having to pay additional production costs and salaries. It may be that a large portion of BRG’s increased “costs” may simply be HBJ’s share of the increased revenues resulting from less price competition. The defendants failed to offer any evidence to rebut the strong inference that the price increases resulted in large part from the defendants’ agreements and their adverse effects on competition in the Georgia bar review markets.
2. Bar/Bri Tradename
In addition to price increases, BRG’s retention of the exclusive right to use the Bar/Bri tradename on Georgia materials under the 1982 agreement effectively prevents HBJ from competing with BRG or licensing its multistate materials to any other potential providers of bar review courses in Georgia. In fact, the Bar/Bri tradename has displaced the BRG trade-name making BRG and Bar/Bri synonymous as a common business entity in the Georgia marketplace. The defendants’ argument that another competing bar review in Georgia could use the Bar/Bri materials under another tradename such as “HBJ Georgia” and compete effectively is highly speculative.
D. Summary
In sum, the plaintiffs have demonstrated that the 1980 agreement and its modifications impose sufficiently “naked” restraints on price and output making an elaborate inquiry into market analysis unnecessary. Even if the 1980 agreement and its 1982 modifications were legally not *1439 sufficiently “naked,” the plaintiffs have demonstrated sufficient actual, sustained detrimental effects on competition to survive summary judgment.
V. Sherman Section Two Claims
The plaintiffs allege that the defendants conspired to monopolize, 32 attempted to monopolize, 33 and monopolized 34 the Georgia market (and relevant submarkets) for bar review courses that prepare law students for the Georgia bar. The district court granted summary judgment in defendants’ favor on these counts because it held that the plaintiffs failed to properly identify the relevant product and geographic markets. The court refused to consider additional economic evidence on these claims despite considering such evidence under the plaintiffs’ rule of reason claim.
A. Monopoly Power
Under Sherman section 2, a plaintiff must demonstrate that the defendants) have either monopoly power or substantial market power in a relevant product and geographic market. 35 In its attempt to evaluate what degree of economic power the defendants exercised in the relevant market, the district court discounted most of the plaintiffs’ totally uncontested evidence. 36 The district court held that “the entirety of Plaintiffs’ effort to identify the proper product and geographic markets is contained in the affidavit of Dr. William Henry_ Doctor Henry’s affidavit is insufficient to identify or justify a proper product or geographic market.” Order, January 9, 1987 at 12. The court held that Dr. Henry’s conclusions were “valueless” because he defined the product market in a “narrow and seemingly artificial manner without economic justifications.” Id. The court also rejected Dr. Henry’s conclusion that Athens, Georgia was a relevant sub-market. The court stated that Dr. Henry's analysis did not address “that area in which the product is offered, but rather limits its consideration to the place where the course is administered during the school year.” Id.
The district court did not include in its market analysis the deposition of expert witness Prof. David Kamerschen (including exhibits), 37 the depositions and affidavits of members of the law student class, and the survey data and affidavit provided by Prof. Ponsoldt. The court also neglected Dr. *1440 Henry’s second affidavit and the defendants’ own marketing literature and deposition testimony. These additional materials provide a vivid description of the product and geographic market structure issues relevant to the resolution of the claims in this case.
1. Geographic Submarkets
The court erred in failing to recognize Athens, Georgia as a relevant submarket for Sherman section 2 analysis. The district court correctly notes that BRG offers its review course on a statewide basis. 38 The district court, however, rejected the plaintiffs’ contention that Athens, Georgia was a relevant submarket 39 despite uncontested expert testimony and survey data which indicates that law student preferences and high transportation costs separate Athens from the Atlanta and Macon markets. In addition, 1986-87 survey data show that approximately 90% of University of Georgia third year law students take the BRG course during the winter term when they are still enrolled in school. 40 Yet, the district court discounted all this evidence.
Further, the record indicates that BRG and HBJ combined for at least 80% of the Athens market prior to their agreement. Pelletier Deposition at 130. The district court confirms this fact stating that the “vast majority” of University of Georgia law students who took a bar review course conducted in Athens, took either the “Harc-ourt or the BRG course” in 1979. Subsequent survey data of University of Georgia law students, who basically constitute the entire Athens market, indicates that BRG has over 95% of the Athens market. Pon-soldt Affidavit, Record Vol. 3, Tab 31, at 3.
2. Relevant Product Market
The plaintiffs presented evidence that the relevant product market is comprehensive bar review courses including lectures and written materials that prepare students to take the Georgia bar exam. 41 The district court, however, held that the plaintiffs failed to establish a relevant product market.
In Brown Shoe Co. v. United States, 370 U.S. 294 , 82 S.Ct. 1502 , 8 L.Ed.2d 510 (1962), the Court stated:
The outer boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it. However, within this broad market, well-defined submarkets may exist which, in themselves, constitute product market for antitrust purposes. The boundaries of such a submarket may be determined by examining such practical indicia as industry or public recognition of the submark-et as a separate economic entity, the products’s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.
Id. at 325 , 82 S.Ct. at 1523-24 . The district court’s analysis is contrary to Brown because the plaintiff presented evidence indicating that (1) University of Georgia law students perceived no reasonable alternatives to the comprehensive BRG review course; (2) third year law students (in particular winter semester third year students) are distinct consumers, and (3) law student consumers were insensitive to price in *1441 creases because they to the BRG course even after large price increases. The defendants’ own advertising literature and deposition statements also reflect their perception that comprehensive bar review courses that prepare students for the Georgia bar constitute a relevant product market.
This uncontroverted evidence provides substantial support for the plaintiffs’ Sherman section 2 claims. Thus, the district court erred in granting summary judgment for the defendants because the plaintiffs’ evidence raises a genuine issue of fact regarding whether the defendants violated Sherman section 2.
VI. Conclusion
In conclusion, the district court failed to adequately analyze the plaintiffs’ per se price-fixing and market allocation claims. The plaintiffs have demonstrated that BRG and HBJ conspired to restrain trade and entered an explicit written agreement in 1980 which was per se illegal. It remains a factual issue whether the defendants have withdrawn from this conspiratorial agreement. The trial court also erred in failing to properly analyze the plaintiffs’ rule of reason claim. The plaintiffs have presented sufficient evidence that the defendants entered an agreement that directly restrained competition in the Georgia bar review market. The plaintiffs have also demonstrated actual anticompetitive effects arising from the agreement. This same economic evidence supports the plaintiffs’ Sherman section 2 claims. I would reverse the district court’s summary judgment and remand to the district court for a trial on all of the foregoing issues.
. I concur in Part II of the majority opinion (upholding district court’s denial of Rule 11 sanctions). I disagree, however, with the majority’s conclusion in Part I that the district court’s denial of class certification was not an abuse of discretion. The premise of the district court's order was that the plaintiffs "may lack financial commitment sufficient for adequate investigation and trial preparation in light of the court's ruling [granting summary judgment on all counts except the rule of reason claim].” The record reflects that the plaintiffs’ counsel was qualified and experienced, and prepared this case from its inception in a competent and professional manner. The plaintiffs’ alleged lack of financial commitment was simply not a factor in their trial counsels' presentation of their case. I believe that the district court’s misperceptions regarding the evidence necessary to establish the asserted antitrust violations, discussed infra, resulted in the court erroneously denying class certification based on the plaintiffs’ purported "lack of financial commitment.” I would therefore reverse the district court on this issue.
. Prior to 1979, BRG’s predecessor, defendant Bar Review Group, Inc., offered the bar review course in Georgia. During 1979, Pelletier formed two corporations: BRG of Georgia (BRG) and BRG Publications. For purposes of this opinion, references to the defendant BRG includes Pelletier as well.
. HBJ originally retailed its bar review courses directly to consumers but has since modified its retailing strategy to include about 19 licensees who have the exclusive or non-exclusive right to provide HBJ bar review courses. HBJ does not concurrently offer its course at retail in competition with any of these licensees.
. The review courses were administered in three cities, Atlanta, Athens, and Macon, where law schools were located.
. Their first meeting was arranged by one of Convi ser’s attorneys in Chicago who had contacted Pelletier by phone. Conviser flew to Atlanta, Georgia and met Pelletier at the Coach & Six restaurant where they discussed "in passing” the possibility of Bar/Bri purchasing BRG or BRG buying HBJ's Georgia operations. Record, Vol. 1, Tab 2, at 10-11. Pelletier later flew to Chicago where he and Conviser discussed a licensing arrangement or buy-out. Id. Pelletier returned to Atlanta where a third meeting was held to discuss a possible licensing arrangement. Id.
.The 1980 agreement contained two provisions one termed a “Covenant Not to Compete” and the other entitled “Other Ventures.” Record, Vol. 2, Tab 13 at E10, E15. The former required HBJ not to directly or indirectly "own, manage, operate, join, invest, control, or participate in or be connected as an officer, employee, partner, director, independent contractor or otherwise with any business which is operating or participating in the preparation of candidates for the Georgia State Bar Examination." Id. at E10. The latter required BRG not to compete against HBJ in states in which HBJ currently operated outside the state of Georgia. Id. at E15.
. The nine count complaint alleges the defendants engaged in: (1) horizontal price-fixing; (2) geographic market allocations; (3) concerted refusals to deal; (4) an unreasonable joint venture; (5) a conspiracy to monopolize bar review written materials and lecture services in Georgia; (6) an attempt to monopolize bar review written materials and lecture services in Georgia; (7) monopolization of bar review written materials and lecture services in Georgia; (8) a conspiracy to monopolize bar review written materials and lecture services nationwide; and (9) an attempt to monopolize bar review written materials and lecture services nationwide.
. In its January 9, 1987 order the court granted the defendants’ motion for summary judgment on counts 1-9. The court, however, permitted the plaintiffs to recast counts 1-4 under the rule of reason.
. The common concern in Matsushita and Monsanto is that a court’s denial of summary judgment based merely on speculative inferences from ambiguous evidence could deter or penalize perfectly legitimate and procompetitive conduct. Because legitimate price-cutting is the "very essence of competition," allowing "mistaken inferences” to result in antitrust liability chills "the very conduct the antitrust laws are designed to protect.” Matsushita, 475 U.S. at 594 , 106 S.Ct. at 1360 (citing Monsanto). The Supreme Court therefore adopted more exacting evidentiary standards to separate permissible procompetitive business practices from proscribed anticompetitive practices in antitrust actions based on business conduct that facially supports either practice. Consequently, both cases reinvigorated the use of summary judgment to summarily dismiss certain antitrust actions based on circumstantial evidence and speculative economic theories.
. Criticism of the broad application of predatory pricing theories abounds. See, e.g., R. Blair & D. Kaserman, Antitrust Economics, 121-128 (1985) (predatory pricing may preserve or enhance, but cannot create, market power); R. Bork, The Antitrust Paradox: A Policy at War with Itself, 144-59 (1978) (debunking most predatory pricing theories); R. Posner, Antitrust Law: An Economic Perspective, 184-96 (1976) (narrowly limiting predation to practices designed to exclude an equally or more efficient competitor). Vertical price-fixing has also been severely critiqued. See, e.g., R. Blair & D. Ka-serman, Law and Economics of Vertical Integration and Control, 157-61 (1983) (law condemns vertical price maintenance despite potentially beneficial economic effects); R. Bork, supra, at 280-98 (virtually all vertical restraints benefit consumers and should be lawful).
.Notably, the plaintiffs in this action allege a conspiracy between two corporations and their respective officers. This situation is unlike the predatory pricing scheme involving twenty-one defendants that the Supreme Court found implausible in Matsushita.
. The plaintiffs’ price-fixing and market allocation claims are the types that courts and economists have long condemned as per se violations of the Sherman Act. Their rule of reason and monopolization claims are similarly founded upon widely accepted economic principles.
. Per Se rules are appropriate only for conduct which is manifestly anticompetitive and should be "invoked when surrounding circumstances make the likelihood of anticompetitive conduct so great as to render unjustified further examination of the challenged conduct." National Collegiate Athletic Ass'n v. Board of Regents of University of Oklahoma, 468 U.S. 85, 103-04 , 104 S.Ct. 2948, 2961 , 82 L.Ed.2d 70 (1984).
. The Supreme Court has applied the per se rule to geographic market allocation and horizontal price-fixing because such practices are so plainly anticompetitive that no elaborate study of an industry is necessary to establish their illegality. National Soc. of Professional Engineers v. United States, 435 U.S. 679, 692 , 98 S.Ct. 1355, 1365 , 55 L.Ed.2d 637 (1978).
. Abadir & Co. v. First Mississippi Corp., 651 F.2d 422, 426-27 (5th Cir.Unit A July 1981) (competitors not allowed to turn an otherwise horizontal agreement vertical by setting up a licensing corporation to impose market allocation agreements); see, e.g., U.S. v. Topco Associates, Inc., 405 U.S. 596 , 92 S.Ct. 1126, 1134 , 31 L.Ed.2d 515 (1972) (cooperative buying association which allocates territories violates horizontal per se rule); U.S. v. Sealy, Inc., 388 U.S. 350 , 87 S.Ct. 1847, 1850 , 18 L.Ed.2d 1238 (1967) (characterizing licensor-licensee relationship as horizontal where licensees own substantially all stock of licensor).
. Agreements between horizontal competitors that affect inter-brand competition Eire the "primary concern of antitrust law.” Continental T.V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36 , 52 n. 19, 97 S.Ct. 2549 , 2558 n. 19, 53 L.Ed.2d 568 (1977).
. Both provide the same type of service, bar review courses, and both are capable of providing their services in the same geographic locations (absent their agreements).
. The district court stated that:
Neither agreement between BRG and [HBJ] constitutes the sort of market or customer allocation agreement which has been recognized as a basis of per se liability. Cf. Topco, 405 U.S. 596 [ 92 S.Ct. 1126 ]. This is not a situation where competitors divided up a market in which both have previously done business, each taking a portion of a market. BRG has never done business outside the state of Georgia, and nothing in the record suggests that it ever intended to do so. [HBJ] as noted has done business virtually nationwide, but withdrew from the Georgia market following the 1980 agreement between BRG and [HBJ]. Thus, the only market ever claimed by both Defendants is the state of Georgia, and clearly the state of Georgia was not divided up under either the 1980 or 1982 agreements.
Order, January 9, 1987 at 8 (emphasis added).
. To use the district court’s parlance, BRG and HBJ divided the Georgia market by allocating 0% for HBJ and 100% for BRG and divided the national market by allocating 100% for HBJ and 0% for BRG.
. The district court held that:
the agreement is not a classic form of price fixing where two competitors agree what price they will charge for their products. On the contrary, neither agreement explicitly addresses the factor of price, and [HBJ] has never had any right under either the 1980 or 1982 agreement to be consulted about the price of the BRG course. While it is true that the agreement of two competitors to "pool" their products may exert an upward influence on price, such an agreement is not inherently anticompetitive, and in a given setting could actually be procompetitive.
Order, January 9, 1987, at'7-8.
. See, e.g., Catalano Inc. v. Target Sales, Inc., 446 U.S. 643 , 100 S.Ct. 1925 , 64 L.Ed.2d 580 (1980) (horizontal agreement to terminate the practice of giving credit falls squarely within the traditional per se rule against price-fixing); National Society of Prof. Eng. v. U.S., 435 U.S. 679 , *1435 98 S.Ct. 1355 , 55 L.Ed.2d 637 (1978) (agreement that interferes with setting of price by free market forces is illegal on its face).
. See, e.g., National Collegiate Athletic Ass’n v. Board of Regents of University of Oklahoma, 468 U.S. 85, 101 , 104 S.Ct. 2948, 2960 , 82 L.Ed.2d 70 (1984) (horizontal restraints on competition necessary if product is to be available); Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 21-22 , 99 S.Ct. 1551, 1563 , 60 L.Ed.2d 1 (1979) (horizontal licensing arrangement created product which was "greater than the sum of its parts”).
. See Citizens Publishing Co. v. United States, 394 U.S. 131, 134-36 , 89 S.Ct. 927, 928-29 , 22 L.Ed.2d 148 (1969) (market allocation agreement designed to prevent commercial rivalry and support price fixing conspiracy).
. See infra note 31.
. Because the district court erred in finding that the 1980 agreement did not constitute a per se violation, it did not address whether the 1982 modified agreement constituted a withdrawal from or abandonment of the conspiracy. In United States v. Kissel, 218 U.S. 601 , 31 S.Ct. 124 , 54 L.Ed. 1168 (1910), the Supreme Court held that antitrust conspiracies may continue in time beyond the original conspiratorial agreement until either the conspiracy’s objectives are abandoned or succeed. Id. at 608-09 , 31 S.Ct. at 126 . For example, a conspiracy continues so long as one or more conspirators continue to receive payments under a collusive contract because the conspirators continue to realize a central and obvious objective of the conspiracy (i.e. illicit profits). United States v. Dynalectric Co., 859 F.2d 1559, 1563-69 (11th Cir.1988); United States v. Aquafredda, 834 F.2d 915, 919 (11th Cir.1987), cert. denied sub notn., Agostino v. United States, — U.S. -, 108 S.Ct. 1278 , 99 L.Ed.2d 489 (1988). In the absence of proof of withdrawal, the conspirators are presumed to continue in their conspiracy particularly where they continue to share the fruits of their conspiracy, in this case through the revenue-sharing provision in the 1980 and 1982 agreements. Dy-nalectric, 859 F.2d at 1563-69 ; Aquafredda, 834 F.2d at 919 (antitrust defendants’ continued sharing of revenues and professional and social affiliations permits inference of continued conspiracy). Thus, it is an unsettled factual issue whether the conspiratorial objectives manifest in the 1980 agreement between HBJ and BRG have continued despite the supposedly ameliorative modifications made in 1982.
. In Indiana Federation of Dentists, 476 U.S. 447 , 106 S.Ct. 2009 , 90 L.Ed.2d 445 (1986) a group of dentists decided to stop supplying x-rays of their patients’ mouths to the insurance company paying the bill. The Supreme Court held that this practice is an illegal horizontal agreement that withheld from dental customers a particular service the customers desired. "Absent some countervailing procompetitive virtue — such as, for example, the creation of efficiencies in the operation of a market or the provision of goods and services, such an agreement limiting consumer choice by impeding the ‘ordinary give and take of the market place’ cannot be sustained under the Rule of Reason." 476 U.S. at 459 , 106 S.Ct. at 2018 (citations omitted). Additionally, the Court reiterated that the FTC's absence of specific findings regarding market definition and market power does not *1437 preclude, as a matter of law, a finding that the dentists unlawfully restrained trade. Id. at 460, 106 S.Ct. at 2018-19. Instead, actual proof of detrimental effects, such as a reduction of output, can obviate the need for an inquiry into market power. Id. at 460-61, 106 S.Ct. at 2019.
. Because they have shown anticompetitive effects, the plaintiffs would not have to establish a relevant market or that the defendants had market power.
. Instead, the district court held "it is as reasonable to assume that the price increase reflected [additional multistate] costs as it is to assume that the higher prices were merely to gouge the consumers.” Order, July 8, 1987 at 5. The trial court also held that the plaintiffs failed to demonstrate that the defendants’ modified contractual arrangement had a "present anti-competitive effect in a relevant geographic and product market." Order, July 8, 1987, at 7 (emphasis added).
.Plaintiffs in a rule of reason case do not have to demonstrate a present anticompetitive effect in a specifically-defined product and geographic market if they can establish the existence of anticompetitive effects or purpose during the time period that the anticompetitive practices were allegedly in effect. These anticompetitive effects can be inferred from the price increases following the HBJ-BRG agreement (and the resulting reduction in the number of bar review courses sold), the licensing and revenue-sharing provisions, and the elimination of consumer choice through HBJ’s withdrawal from the Georgia market.
. The district court concluded that ”[u]nder the circumstances, it is fair to conclude that the possibility of anticompetitive effect exists, but that such effect cannot be presumed from the contract itself.” Order, July 8, 1987, at 4.
. An important factual issue is whether BRG’s and HBJ's agreements are necessary for either’s competitive survival. First, no evidence exists to explain why BRG had to use Bar/Bri’s multi-state materials much less have the exclusive right to the Bar/Bri tradename. Other sources of multistate materials existed. For instance, West Publishing’s materials were still available albeit not free. Pelletier Deposition, at 73. Second, although HBJ would forego income by not selling its multistate materials to some other bar review course in Georgia, no legitimate reasons exist to explain why it was necessary for HBJ to form an agreement with its only rival in Georgia that contains blatantly anticompetitive licensing, revenue-sharing, and market allocation provisions. A plausible explanation is BRG's and HBJ's collective desire to profit by eliminating their rivalry and sharing the increased revenues that would result from BRG's new-found dominance in the Georgia market.
. A conspiracy to monopolize requires (1) proof of concerted action deliberately entered into with the specific intent to accomplish the unlawful result of achieving a monopoly; and (2) the commission of at least one overt act in furtherance of the conspiracy. Under a conspiracy theory, the defendants do not have to attain monopoly power nor would they have to obtain monopoly power if their conspiracy had been successful. L. Sullivan, Handbook on the Law of Antitrust, 132-34 (1977).
. Attempted monopolization is an offense distinct from monopolization which requires (1) a specific intent to monopolize and (2) a dangerous probability of success. Lorain Journal Co. v. United States, 342 U.S. 143, 153 , 72 S.Ct. 181, 186 , 96 L.Ed. 162 (1951). The most important evidence of defendants’ specific intent to monopolize is anticompetitive conduct. A dangerous probability of success is measured by a defendant’s possession of substantial market power.
. The offense of monopoly under section 2 of the Sherman Act has two elements: (1) the possession of monopoly power in a relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of superior product, business acumen, or historic accident. United States v. Grinnell Corp., 384 U.S. 563, 570-71 , 86 S.Ct. 1698, 1703-04 , 16 L.Ed.2d 778 (1966).
. Monopoly power is defined as the power to control price or exclude competition and is measured by reference to the relevant market. United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 391 , 76 S.Ct. 994, 1005 , 100 L.Ed. 1264 . Market share is the traditional measure of market and monopoly power. United States v. Grinnell Corp., 384 U.S. 563, 571 , 86 S.Ct. 1698, 1704 , 16 L.Ed.2d 778 (1966) (90% share of relevant market constitutes monopoly power, over 80% share a “substantial monopoly”). The determination of the relevant product and geographic markets (and a defendant’s market share), however, is not an exact science and mathematical precision may, at times, yield to administrative convenience.
. The defendants submitted no evidence to rebut the plaintiffs’ market structure analysis.
. Kamerschen’s deposition was the principal economic evidence upon which the first class action was based.
. "[T]he record discloses that BRG’s course is marketed statewide, and is offered at various statewide locations both during the school year and following the close of the school year.” Order, January 1, 1987 at 12. The court however failed to consider whether Georgia is a relevant geographic market despite evidence suggesting such a conclusion.
. The Supreme Court in United States v. Grinnell Corp., 384 U.S. 563 , 86 S.Ct. 1698 , 16 L.Ed.2d 778 (1966), stated that under Sherman section 2 "there may be submarkets that are separate economic entities.”
. Prof. Ponsoldt conducted the survey of all third year students at the University of Georgia. The survey indicated it was being conducted in "context of a pending federal antitrust case against certain bar review course companies." Record, Vol. 2, Tab 31 at 8. 63 of the 170 third year students responded. Id. at 3.
. See, e.g., Henry Affidavit, Record, Vol. 2, Tab 17.