Antitrust sanctions: deterrence and (possibly) overdeterrence.
Antitrust enforcement in the United States is an odd mixture of public and private efforts. The Antitrust Division of the Department of Justice (DOJ) has a limited enforcement budget and pursues relatively few cases each year. In its enforcement arsenal are monetary sanctions (i.e., fines), injunctions, and imprisonment for those found guilty of the most egregious antitrust violations. (1) Successful private plaintiffs recover treble damages plus the cost of the suit including a reasonable attorney's fee. (2) The twin goals of this system are to punish antitrust violators in an effort to deter such conduct by others and to compensate antitrust victims.
In its April 2, 2007, report, the Antitrust Modernization Commission (AMC) made only one major recommendation for change regarding public and private enforcement. (3) Specifically, the AMC recommended that Illinois Brick (4) and Hanover Shoe (5) be overturned to the extent necessary to allow indirect purchasers to be able to have standing in federal court to pursue overcharges passed on to them by direct purchasers. (6) All other recommendations regarding antitrust sanctions were simply endorsements of the status quo. (7) These recommendations, however, deserve some examination.
With respect to civil and criminal remedies, the AMC considered four main areas, including private remedies, civil remedies, indirect purchaser litigation, and criminal prosecution. In this article, we confine our discussion mainly to the recommendations involving private enforcement and criminal prosecution. (8) With respect to private enforcement, the AMC highlighted five main purposes of treble damages: (1) deterring anticompetitive behavior, (2) punishing violators, (3) depriving the violators of their ill-gotten gains, (4) providing full compensation to the antitrust victims, and (5) providing a powerful incentive for private enforcement of the antitrust laws. (9) While alternatives to the treble damages rule had been suggested, no clear evidence was presented justifying a change, and the AMC therefore recommended no modification to the existing private damages provisions. (10) There were no changes recommended to the award of attorney's fees or prejudgment interest. The AMC did recommend a new provision under joint and several liability that would improve fairness. (11) In terms of criminal prosecution, no change was recommended to the criminal penalties for violations of the antitrust laws. The AMC did, however, support the continued harsh prosecution of anticompetitive business behavior, such as price fixing, which has no procompetitive justification and is per se illegal. (12) Overall, the majority of the AMC's recommendations for private enforcement and criminal prosecution were not recommendations for change but rather recommendations for continuing the status quo.
In what follows, we address the AMC's continued support of private enforcement using treble damages and the recommendations to continue criminal prosecution of "naked" agreements among competitors. In part II, we review the economic role of sanctions in deterrence by analyzing choices under uncertainty using the expected utility model. Part III considers the specific case of price fixing as an application of the model of deterrence. Part IV examines the possibility of overdeterrence under the current system and the relevance of overdeterrence to the case of price fixing and considers some exceptions to the per se rule for horizontal price fixing as well as some issues in vertical restraints. In part V, we offer some concluding remarks. Overall, the AMC's recommendations on sanctions and continued prosecution of per se violations appear sound, with a few exceptions in the case law.
II. THE ECONOMIC ROLE OF SANCTIONS (13)
The economic purpose of sanctions is not to punish undesirable behavior as a means of revenge or retribution. Rather, the purpose is to deter such behavior so punishment will not actually be necessary. The economic role of sanctions, then, is to make objectionable behavior unprofitable or otherwise unattractive to the potential wrongdoer. The idea is to induce individuals to choose to behave as society wants them to behave. The way that this works can be seen in a simple economic model of decision making under uncertainty. Although the model that we employ has general applicability, we will develop it in the context of a potential antitrust violation.
A. Expected utility analysis
Whenever a firm violates the antitrust laws, it engages in risky behavior. The outcome is uncertain because there is a gain if the crime goes unpunished, but there is a penalty if the crime is detected and successfully prosecuted. This risky situation can be avoided by not committing the crime in the first place. To analyze such choices in the face of uncertainty, we employ the expected utility model of yon Neumann and Morgenstern. (14)
An antitrust violation is a risky prospect that yields a profit of [[PI].sub.1] with probability p or [[PI].sub.1]- F with probability (1 - p), where F is the punishment. According to yon Neumann and Morgenstern, the decision maker values this risky prospect as a probability weighted average:
U([[PI].sub.1] [[PI].sub.1] - F; p) = pU([PI].sub.1]) + (1 - p)U ([[PI].sub.1 - F)
where U(.) is the decision maker's utility function. Thus, the value of the risky prospect is the expected utility of the possible outcomes. Expected utility analysis is based on the assumption that decision makers will make choices in an effort to maximize expected utility.
Suppose that the firm can earn a profit of [[PI.sub.0] without committing an antitrust violation. This will generate a utility value of U([[PI].sub.0]). Violating the antitrust laws will result in [[PI].sub.1], which is larger than [[PI].sub.0], if the firm gets away with it, but [[PI].sub.1] - F if it does not. (15) The firm will find it to be economically rational to violate the antitrust laws if doing so makes it better off. Put differently, if the E[U([PI])] exceeds the U([[PI].sub.0]), then we should expect the firm to violate the antitrust laws. Conversely, if the E[U([PI])] is less than U[[PI].sub.0]), there will be no violation. Thus, we construct a deterrent function (D):
D = U([[PI].sub.0]) - E[u([PI].)].
If D > O, violations will be deterred; if D < O, violations will not be deterred.
C. Attitudes toward risk
Generally, decisionmakers can be classified according to their attitudes toward risk. Those who are risk neutral focus only on the expected profit. Their utility functions are linear and, as a result, if the expected profit, E[PI], is less than [[PI].sub.0], they will be deterred, but if the E[[PI]] exceeds [[PI].sub.0], they will not be deterred. If they are deterred, it is because on average antitrust violations are unprofitable in both monetary and utility terms. Thus, it is economically rational for them to obey the law. The converse is also true: when the E[[PI]] exceeds [[PI].sub.0], antitrust violations are profitable on average and, therefore, they will be committed.
Risk averters do not like risk and will try to avoid it. They are willing to pay something or sacrifice something to avoid risk. This does not mean that a risk averter will never bear risk. If the E[U([PI])] exceeds U([[PI].sub.0]), then the risk averter will bear the risk and commit the crime. In essence, if the reward is large enough and the probability of success is high enough, then the risk averter will take the chance. In this case, the firm is better off in utility terms on average. The risk is worth bearing. This result can be illustrated graphically.
[FIGURE 1 OMITTED]
In figure 1, U([PI]) is the decisionmaker's utility function. It is positively sloped because more profit is always better than less profit, but it increases at a decreasing rate (that is, it is concave). This individual is risk averse because he will reject all actuarially fair gambles. (16) Risk can be avoided by not committing an antitrust violation, staying at ([[PI].sub.0], and enjoying a utility of U([[PI].sub.0]). If an antitrust violation is committed, the outcome will be either [[PI].sub.1] or [[PI].sub.1] - F. The way to measure the expected utility in this graphical model is to connect the points on the utility function corresponding to [[PI].sub.1] and [[PI].sub.1] - F with a chord. Now, find the expected profit associated with a violation,
E[[PI]] = p[[PI].sub.1] + (1 - p)([[PI].sub.1] - F),
and find the height of the chord at E[[PI]]. That height corresponds to an expected utility level of E[U([PI])]. (17) If the U([[PI].sub.0]) is equal to the E[U([PI])], then the individual will be indifferent between committing the violation and not doing so, i.e., D = 0 in that case. For the risk averter, we can see that the E[[PI]] must be well above [[PI].sub.0] for the individual to be indifferent. If the probability of escaping detection and conviction is high enough to make the expected profit higher than E[[PI].sub.1] in figure 1, then the E[U([PI])] will exceed U([[PI].sub.0]). In that event, the crime is "profitable" in monetary terms because the E[[PI]] exceeds [[PI].sub.0], but it is also "profitable" in utility terms since the E[U([PI])] exceeds the U([[PI].sub.0]). In other words, crime "pays" and even the risk averter will commit the crime.
Alternatively, in figure 2 the utility function is convex--it is positively sloped and it increases at an increasing rate. This is the utility function of a risk seeker, i.e., a person who will accept all actuarially fair gambles. The values for [[PI].sub.0], [[PI].sub.1], and [[PI].sub.1] - F are the same as in figure 1. In this case, however, the individual is indifferent when the expected profit is as low as E[[[PI]].sub.2], which is well below [[PI].sub.0]. Again, if p is high enough to push expected profit past E[[[PI]].sub.2], the crime will be committed. For a risk seeker, crime need not be profitable in monetary terms. As long as the E[U([PI])] exceeds the U([[PI].sub.0]), crime will be profitable in utility terms.
[FIGURE 2 OMITTED]
These results are interesting because we can see that deterrence is influenced by attitudes toward risk. Suppose, for example, that the values [[PI].sub.1], p, and F were such that the expected profit associated with an antitrust violation were equal to [[PI].sub.0]. The risk averter would be deterred because the U([[PI].sub.0]), which is the height of U([PI]) at [[PI].sub.0], exceeds the E[U([PI])], which is the height of the chord at [[PI].sub.0] in figure 1. The risk seeker, however, will embrace that gamble because the E[U([PI])], which is the height of the chord at [[PI].sub.0] in figure 2, exceeds the U([[PI].sub.0]) under the assumed conditions. Consequently, the enforcement efforts and sanctions that will deter a risk averter may not be sufficient to deter a risk seeker, This fact may pose some interesting policy problems.
III. SANCTIONS IN PRICE FIXING CASES
Deterring undesirable business behavior depends on both the expected fine or penalty for engaging in an antitrust violation as well as the probability of detection and successful prosecution. Section I of the Sherman Act prohibits "every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade." (18) This includes price fixing and other naked restraints of trade that serve no procompetitive purpose such as output restriction, bid rigging, market division and the like. A section 1 violation can be penalized with monetary sanctions, imprisonment, or both. Corporations face fines up to $100 million; individuals face up to $1 million in fines and up to 10 years in jail. In some cases, these provisions will not be sufficient to deter anticompetitive behavior. Large firms that engage in price fixing could reap benefits far greater than the maximum $100 million penalty. In an effort to address this problem, the Criminal Fines Improvement Act of 1987 amended these provisions allowing the court discretion to impose a fine equal to twice the pecuniary gain resulting from the violation or twice the loss suffered by victims. (19) Ultimately, the amount of the total penalty for violating section 1 will influence the extent of price fixing activities.
A. Treble damages actions
Price fixing and other anticompetitive conduct can also be disciplined through private enforcement. Direct purchasers from price fixing firms have standing to sue for treble damages under section 4 of the Clayton Act, which holds in relevant part that "[a]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor ... and shall recover threefold the damages by him sustained and the cost of the suit, including a reasonable attorney's fee." (20) Currently, under the rules of Hanover Shoe (21) and Illinois Brick, (22) only direct purchasers have antitrust standing. Indirect purchasers do not have standing to bring private damage actions under section 4. (23)
In many states, private damage actions are limited to direct purchasers only. Some states, however, have passed Illinois Brick repealer laws, allowing indirect purchasers an opportunity to sue for damages under those state statutes. In states with repealer laws, conspiring firms face expanded liability in private damage actions. In other words, both direct and indirect purchasers have standing to bring private claims to recover damages from a price fixing scheme. This can increase the exposure of the price fixing firm to duplicative damage claims. The damages will be trebled if the litigation is successful in federal court. The defendant then faces an indirect purchaser suit for the same antitrust violation. We should note that the AMC recommended the repeal of the Illinois Brick and Hanover Shoe rules, which would allow both direct and indirect purchasers to bring private damage actions in federal court. This would eliminate duplicative recovery.
With respect to price fixing, direct purchasers can bring a private suit for the full amount of the overcharge sustained as a result of the collusive activity. For an individual consumer, the overcharge equals the amount the consumer actually paid (the anticompetitive or collusive price) minus the price that would have been paid in the absence of the conspiracy (the "but for" price) times the number of units purchased. The total overcharge as a result of a price fixing conspiracy would aggregate this calculation over all consumers. In this way, the potential liability that a price fixing firm faces is not only the potential penalties if detected and prosecuted by the DOJ, but also private damages that, if successful, would be trebled by the court. We must consider both of these components when modeling the decision to engage in anticompetitive behavior.
B. Model of deterrence extended
We explored the basic model of expected utility theory in the previous section. Now, consider the expected utility model in the specific case of price fixing focusing on a risk averse firm. Assume as before that [[PI].sub.0] represents the profits associated with no antitrust violation, p is the probability of the violation going undetected, and (1 - p) is the probability of the violation being detected and successfully prosecuted. Assume further that the profits associated with a price fixing scheme are given by [[PI].sub.1] if the violation is undetected, but are [[PI].sub.1] minus the total sanction if the violation is detected and successfully prosecuted.
[FIGURE 3 OMITTED]
In the specific case of price fixing, the total sanction can be made up of several components. Figure 3 reproduces much of the same information as figure 1, but in this special application of price fixing sanctions. In order to deter potential anticompetitive conduct, it is necessary to make the behavior unprofitable. First, consider the possibility that the firm may face only monetary fines from public enforcement authorities. If a price fixing violation is detected and successfully prosecuted by the DOJ, the antitrust violator could be responsible for up to $100 million in monetary fines (or up to $1 million for an individual). To the extent that judges are reluctant to levy the maximum fine, potential violators will take into account the expected fine rather than the maximum fine. Denote this expected punishment as F. The height of the chord connecting the corresponding points on the utility function represents the expected utility associated with the price fixing violation. Let E[[PI]] represent the expected profit given this particular risky decision:
E[[PI]] = p[[PI].sub.1] + (1 - p)([[PI].sub.1] - F).
Now, introduce the additional possibility that the price fixing violation would lead to a private damage action where the overcharges sustained by consumers would be trebled. Under the Clayton Act, the antitrust violator would also be required to pay a reasonable attorney's fee. The amount of attorneys' fees can be substantial depending on the duration and complexity of the case. The potential exposure here is D, which incorporates treble damages plus attorneys' fees. For the price fixer, this leaves the upside ([[PI].sub.1]) unchanged, but it reduces the downside from [[PI].sub.1] - F to [[PI].sub.1] - F - D. This causes the chord to rotate downward, which means that the E[U([PI])] will be lower for all values of the expected profit. Figure 3 illustrates the rotation of the chord when the profit associated with the risky prospect is reduced by both F and D. The expected profit is given by
E[[PI]] = p[[PI].sub.1] + (1 - P)([[PI].sub.1] - F - D).
Because the profit associated with price fixing ([[PI].sub.1]) does not change, the chord rotates as we increase the penalty from F to F + D. This improves deterrence.
An antitrust violation will be deterred if the utility from the certain profit, or U([[PI].sub.0]), exceeds the expected utility of the risky prospect, E[U([PI])]. To see the effect of adding private enforcement, we start from a position where the potential violator is indifferent between committing the antitrust violation and not committing it. Put differently, we start where U([[PI].sub.0]) = E[U([[PI])].sub.F], where E[U([PI])]F represents expected utility with only public enforcement. This is shown in figure 3. Note that the height of the utility function at [[PI].sub.0] is precisely equal to the height of the chord at E[[PI]]. Given the values of [[PI].sub.0], [[PI].sub.1], F, and p that are assumed in figure 3, the potential price fixer is indifferent. The deterrent function is equal to zero: D = U([[PI].sub.0]) - E[U([[PI])].sub.F] = 0. NOW, introduce the additional penalty of private damages (D). The new end point is reduced to [[PI].sub.1] - F - D. Holding all other factors constant, this additional penalty will reduce E[U([PI])] for two reasons. First, the variance in the outcome is increased which makes the violator worse off at any value of the expected profit. Even if the probability of detection and successful prosecution were changed so that the expected profit remained the same (E[[[PI]].sub.F] = E[[[PI].sub.]D]), the expected utility would fall because the new chord lies below the original chord. Second, if the probability of detection and successful prosecution remained the same, the expected profit would fall. A lower expected profit would have resulted in positive deterrence without an increase in the penalty. The combination of these two effects is to reduce the expected utility of wealth to, say, E[U([[PI])].sub.D]. The violation will now be deterred because the utility from not breaking the antitrust laws exceeds the expected utility of doing so.
To extend this analysis further, consider the possibility of jail time if convicted. The AMC has recommended continued harsh punishment, including prison sentences, for violations like price fixing. By monetizing the possibility of spending time in prison as a result of the violation, the chord would rotate further and reduce the expected profit associated with the risky prospect as well as the expected utility associated with the risky prospect. This would make it even more likely that the violation will be deterred.
IV. THE PROBLEM OF OVERDETERRENCE
As we have seen above, the economic role of antitrust sanctions is to deter antitrust violations. But these sanctions can be fairly severe, which raises the possibility of overdeterrence. For naked restraints of trade, i.e., those that have only anticompetitive consequences, "overdeterrence" is essentially meaningless and, therefore, of no concern. (24) But there are many business practices that are not clearly anticompetitive and may on balance be procompetitive. These practices might be deterred by the possibly heavy antitrust sanctions, especially for risk averters. Overdeterrence occurs when procompetitive or competitively neutral conduct is avoided for fear of antitrust prosecution. This leads firms to choose second-best practices that are less effective or more expensive, but are safe from antitrust suits. This outcome is inconsistent with the promotion of social welfare. A few examples will illustrate our point.
A. Horizontal restraints
For many horizontal restraints, the competitive consequences are fairly obvious. In most cases of price fixing, bid rigging, market division, and customer allocation, the results will be increased prices and reduced output. Unless there are some significant cost savings that cannot be obtained without the horizontal agreement, these practices will reduce both consumer welfare and social welfare. As a general proposition, we want the punishment to fit the crime, but we still want the punishment to be severe enough to deter those practices that are clearly anticompetitive.
Suppose, however, that there are significant cost savings that outweigh the social welfare losses attributable to the increased price. (25) This is illustrated in figure 4, in which D is demand and [AC.sub.1] is industry average cost with competition among the firms. The competitive price and output are [P.sub.1] and [Q.sub.1], respectively. Collusion leads to a higher price equal to [P.sub.2], a reduced output equal to [Q.sub.2], and a cost saving of [AC.sub.1] - [AC.sub.2] per unit. Now, the effect of collusion is to reduce consumer surplus by area [P.sub.2]ab[P.sub.1]. There will be no reduction in social welfare, however, if the cost saving exceeds the usual welfare triangle. That is, if area [P.sub.1]cde is larger than area abc, social welfare is actually improved. Consequently, we do not want to deter this agreement, but the antitrust sanctions may do just that. Putting aside any public sanctions, the consumers of this product may file a private suit for damages. In this event, the consumers of this product will point to an overcharge of ([P.sub.2] - [P.sub.1])[Q.sub.2] as their damages. If the treble damages of 3([P.sub.2] - [P.sub.1])[Q.sub.2] exceed the enhanced profit of ([P.sub.2] - [AC.sub.2])[Q.sub.2], a socially desirable agreement may be deterred. (26)
[FIGURE 4 OMITTED]
There is some precedent for taking efficiencies into account. In both BMI (27) and NCAA, (28) the Supreme Court recognized that some horizontal agreements may be procompetitive on balance, and therefore should not be condemned on a per se basis. In BMI, the Court found that blanket licenses issued by rights organizations, such as BMI and ASCAP, provided the only efficient way for copyright holders in the music industry to be compensated. This compensation was deemed necessary to encourage the production of creative works. On balance, the Court reasoned that the blanket licenses caused output to expand rather than to contract. Accordingly, the Court rejected the plaintiff's claim.
In its NCAA decision, the Court found that the product at issue was athletic competition on the field. Some cooperation off the field among the members of the NCAA was necessary to produce that competition on the field. As a result, it ruled that per se treatment was not appropriate. On balance, however, it found the restraints at issue in that case to be unreasonable and, therefore, violative of the antitrust laws.
B. Vertical restraints
The nature of vertical restraints has been misunderstood and resulted in largely unwarranted judicial hostility. For many years, vertical price restraints were treated as harshly as horizontal price fixing agreements. Maximum resale price fixing was condemned in the Supreme Court's Albrecht (29) decision. This decision was decidedly unfortunate since maximum resale price fixing is invariably used to cope with successive monopoly. The downstream monopolist is prevented from exploiting its monopoly position by the maximum resale price restraint. The upstream firm engages in this practice to enhance its own profits, but this effort also benefits consumers as well since the resale price is lower and output is higher. (30)
This ill-advised decision was overruled nearly 20 years later in the Court's Khan (31) decision. So one might say "all's well that ends well." During that 20 years, however, firms that might have engaged in maximum resale pricing to the benefit of consumers had two choices. First, they could just live with the reduced profits that would inexorably follow from monopoly pricing by the downstream firm. Second, and more probable, the upstream firm could turn to economically equivalent alternatives. (32) Since these alternatives were not the first choice, they were presumably not completely equivalent in the financial sense. If they involved higher transaction costs or were less effective, the Albrecht rule imposed economic losses on everyone except perhaps the downstream monopolist.
In the case of resale price maintenance (RPM), i.e., minimum resale price restraints, the Supreme Court's Dr. Miles (33) decision made RPM illegal per se for over 90 years. This decision was based on no economic analysis at all. Despite blistering criticism, this unfortunate per se rule remained in force for some 97 years. RPM is now subject to a rule of reason analysis when it is challenged as an antitrust violation. (34) Even so, however, this can lead to overdeterrence.
There are promotional uses of RPM that may have ambiguous welfare effects. (35) Promotional uses of RPM lead to an increase in both price and output. Under a rule of reason analysis, however, the increased price suggests that consumers are worse off. The increased output suggests that consumers sufficiently value the product plus the promotion to pay the higher price and, therefore, they are not worse off. Proof of the actual effect is problematic. (36) The risk that the supplier faces is that RPM will be deemed unreasonable and the damages will be equal to the "overcharge"--that is, the difference between the pre-RPM price and the RPM price. As an economic matter, this would be incorrect because the customer gets only the physical product at the pre-RPM price but gets the product plus value-enhancing promotions at the RPM price. (37) Facing some chance of treble damage actions, a supplier might decide that RPM is too risky. But forgoing RPM may make consumers worse off and will surely make the supplier worse off. In such instances, a legitimate, procompetitive business practice would be deterred.
C. Tying arrangements
Tying arrangements provide another example of a vertical restraint that may well be procompetitive, but be deterred by the antitrust sanction. If a seller conditions the purchase of one good on the buyer's agreement to buy a second good, the transaction involves a tying arrangement. Under some circumstances, tying arrangements are illegal per se. If the seller has monopoly power in the tying good market and the volume of commerce is not de minimis, the tying arrangement will be per se illegal. (38) The competitive concern is market foreclosure: rivals in the tied good market are foreclosed from selling to those who purchase the tying good. Since the buyer must be coerced into buying the tied good, the buyer is presumably injured in some way. Perhaps the quality of the tied good is otherwise unacceptable or the price of the tied good may exceed the competitive level. In either event, the buyer is in some sense overcharged.
As with other vertical restraints, tying has many uses and many of them are either procompetitive or completely neutral. Franchisors, for example, may use tying arrangements to prevent quality degradation. (39) Uniform quality across all outlets in a chain is crucial to the chain's success. Each franchisee, however, has an incentive to use cheaper ingredients and thereby improve its profits. But there is an externality here that the franchisee ignores. Consumers expect uniform quality from a chain. When the consumer experiences inconsistent quality, demand for the brand falls at all outlets. The location that reduces quality gets all of the benefit of the cost saving while the adverse effect on demand is spread over the entire chain. Thus, one franchisee increases its profit at the expense of the chain. If that practice becomes widespread, the entire chain will experience a possibly large decrease in demand. The franchisor can reduce the opportunity for such behavior by requiring the franchisees to buy everything from the franchisor itself or from designated sources of supply. Used as a means of ensuring quality, tying is procompetitive. On balance, everyone is better off than without the quality control.
It is certainly true that quality can be controlled in other ways. The franchisor could stop franchising and vertically integrate. Alternatively, it could closely monitor the day-to-day operations of each franchisee to make sure that each unit adheres to the franchisor's quality standards. Since the franchisor selected neither of these options and chose tying instead, tying is presumably a more efficient way of achieving the desired quality control. Although plaintiffs have not fared well in recent franchise tying cases, (40) some franchisors could be deterred from tying inputs to the franchise license. If so, we are worse off for it.
Tying can also be used by an upstream monopolist when its output is employed in variable proportions by a competitive downstream industry. (41) Absent tying, the monopoly price on one of the inputs causes the downstream industry to substitute other inputs. This is a sensible course of action, but it results in a socially inefficient input mix. The per se illegality of tying may dissuade the input monopolist from tying.
D. Other Vertical Restraints
There are a variety of other vertical restraints that are subject to rule of reason treatment under the antitrust laws. Exclusive dealing, exclusive territorial assignments, and customer restrictions, among others, may all be procompetitive or at least competitively neutral. Those firms that are nervous about antitrust suits and the heavy sanctions that may result may be deterred from employing the restraint. Once again, there are alternatives, but these are presumably less efficient (i.e., more costly or less effective) alternatives. In such cases, the antitrust sanctions overdeter business conduct.
V. CONCLUDING REMARKS
For the most part, the AMC endorsed the status quo with respect to antitrust sanctions. As we have discussed, these sanctions may be quite severe and may serve to deter undesirable naked restraints of trade. If the threat of heavy sanctions deters such conduct, this is socially desirable: there is no violation, there is no welfare loss, and no need to actually impose the sanction. But what will be condemned as an antitrust violation is unclear in many cases. There are many business practices with anticompetitive and procompetitive potential. The threat of heavy sanctions and the uncertainty of antitrust treatment may deter some socially desirable practices. In that event, the sanctions overdeter legitimate conduct. The AMC did not address this problem, but should have done so.
(1) See Sherman Act, 15 U.S.C. [section][section] 1 & 2 (2007) for fines and imprisonment. Provisions for injunctive relief are found at 15 U.S.C. [section] 26 (2007).
(2) The provision for private remedies is contained in [section] 4 of the Clayton Act, 15 U.S.C. [section] 15 (2007).
(3) This recommendation is not the focus of our article; however, it is the subject of another article in this symposium: William H. Page, Class Interpleader: The Antitrust Modernization Commission's Recommendation to Overrule Illinois Brick, in this issue of THE ANTITRUST BULLETIN.
(4) Ill. Brick Co. v. Illinois, 431 U.S. 720 (1977).
(5) Hanover Shoe, Inc. v. United Shoe Mach. Corp., 392 U.S. 481 (1968).
(6) ANTITRUST MODERNIZATION COMMISSION, REPORT AND RECOMMENDATIONS, III.B. at 267 (2007) [hereinafter AMC REPORT].
(7) For another analysis of the AMC's recommendations in this area, see John E. Lopatka, Missed Opportunity: The Enforcement Recommendations of the Antitrust Modernization Commission, in this issue of THE ANTITRUST BULLETIN.
(8) The AMC suggested no changes to existing antitrust authority for civil fines or monetary relief. AMC REPORT, III.C. at 285-88.
(9) Id., III.A. at 246.
(10) Some of the suggested changes included providing treble damages only in cases where conduct is clearly unlawful and has no competitive benefit, allowing for a different rule for per se versus rule of reason cases, allowing for a different rule in covert versus overt cases, and giving judicial discretion on the amount of the multiplier of damages. Id. at 243-48.
(11) The proposed change would grant nonsettling defendants the right to claim reduction by the amount of the settlements among settling defendants as well as allow claims for contribution among nonsettling defendants. Id. at 252.
(12) All section 1 and 2 violations could be subject to criminal prosecution, but the focus of criminal prosecution should continue to be limited to "naked" restraints, such as price fixing, bid rigging, and market division.
(13) The modern economic analysis of crime can be traced to Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. POL. ECON. 169 (1968).
(14) The modern economic analysis of decisions under uncertainty began with the pathbreaking insights of JOHN VON NEUMANN & OSKAR MORGENSTERN, THE THEORY OF GAMES AND ECONOMIC BEHAVIOR (1945).
(15) It is possible for [[PI].sub.1] - F to be larger than [[PI].sub.0], but we will assume in what follows that [[PI].sub.1] - F is less than [[PI].sub.0].
(16) The economics of uncertainty is examined more fully in ROGER D. BLAIR & LAWRENCE W. KENNY, MICROECONOMICS WITH BUSINESS APPLICATIONS 385-427 (1987).
(17) Id. at 406-07 provides a graphical proof of this proposition.
(18) 15 U.S.C. [section] 1 (2007).
(19) 18 U.S.C. [section] 3571(d). Similarly, for individuals the fine can be twice the gain to the individual or twice the injury experienced by the victims.
(20) 15 U.S.C. [section] 15 (2007).
(21) Hanover Shoe, Inc. v. United Shoe Mach. Corp., 392 U.S. 481 (1968).
(22) Ill. Brick Co. v. Illinois, 431 U.S. 720 (1977).
(23) This limitation applies only to standing in federal court.
(24) Overdeterrence refers to the possibility of deterring socially desirable conduct. There is not much that is socially desirable in restraints of trade that have only anticompetitive consequences. This, of course, is circular as is usually the case with definitional matters.
(25) This analysis was put forward by Oliver E. Williamson, Economies as an Antitrust Defense: The Welfare Tradeoffs, 58 AM. ECON. REV. 18 (1968).
(26) Deterrence will depend on attitudes toward risk and the probability of detection and conviction.
(27) Broad. Music, Inc. v. Columbia Broad. Sys., 441 U.S. 1 (1979).
(28) NCAA v. Bd. of Regents Univ. Okla., 468 U.S. 85 (1984).
(29) Albrecht v. Herald Co., 390 U.S. 145 (1968).
(30) This analysis was spelled out in some detail in Roger D. Blair & David L. Kaserman, The Albrecht Rule and Consumer Welfare: An Economic Analysis, 33 U. FLA. L. REV. 461 (1981).
(31) State Oil v. Khan, 522 U.S. 3 (1997). For an analysis, see Roger D. Blair & John E. Lopatka, Albrecht Overruled--At Last, 66 ANTITRUST L.J. 537 (1998).
(32) For an analysis of alternatives, see Roger D. Blair & Amanda K. Esquibel, Maximum Resale Price Restraints in Franchising, 65 ANTITRUST L.J. 157 (1996).
(33) Dr. Miles Med. Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911).
(34) Dr. Miles was overruled by Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 127 S. Ct. 2705 (2007).
(35) When RPM is used to support a horizontal cartel agreement among suppliers or among dealers, the welfare effects are clearly negative. In promotional cases, however, F.M. Scherer, The Economics of Vertical Restraints, 52 ANTITRUST L.J. 687 (1983) and William S. Comanor, Vertical Price Fixing, Market Restrictions and the New Antitrust Policy, 98 HARV. L. REV. 990 (1985) have shown that consumer welfare may decline even though output expands.
(36) See, e.g., Roger D. Blair, The Demise of Dr. Miles: Some Troubling Consequences, 53 ANTITRUST BULL. 133 (2008).
(37) See, e.g., Roger D. Blair, Jill Herndon, & John Lopatka, Resale Price Maintenance and the Private Antitrust Plaintiff, 83 WASH. U. L.Q. 657 (2005), for a more detailed explanation.
(38) For a brief overview of the law and economics of tying, see Roger D. Blair & David L. Kaserman, ANTITRUST ECONOMICS 391-421 (2nd ed. 2008). For an extensive development, see Phillip Areeda & Herbert Hovenkamp, ANTITRUST LAW [paragraph][paragraph] 1700-79 (2nd ed. 2004).
(39) For the technical details, see Roger D. Blair & David L. Kaserman, A Note on Incentive Incompatibility Under Franchising, 9 REV. INDUS. ORG. 323 (1994).
(40) Principe v. McDonald's Corp., 631 F.2d 303 (Cir. 1980); Krehl v. Baskin-Robbins Ice Cream Co., 664 F.2d 1348 (9th Cir. 1982); Queen City Pizza, Inc. v. Domino's Pizza, Inc., 124 F.3d 430 (3d Cir. 1997).
(41) For technical details, see Roger D. Blair & David L. Kaserman, Vertical Integration, Tying, and Antitrust Policy, 68 AM. ECON. REV. 397 (1978).
AUTHORS' NOTE: We thank our respective institutions for financial support.
ROBER D. BLAIR, Department of Economics, University of Florida.
CHRISTINE PIETTE DURRANCE, Department of Public Policy, University of North Carolina--Chapel Hill.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Symposium: The Antitrust Modernization Commission|
|Author:||Blair, Roger D.; Durrance, Christine Piette|
|Date:||Sep 22, 2008|
|Previous Article:||Unilateral refusals to deal and the Antitrust Modernization Commission Report.|
|Next Article:||Missed opportunity: the enforcement recommendations of the Antitrust Modernization Commission.|