A proposal for antitrust merger enforcement reform: repudiating Judge Bork in favor of current economic learning.
The ideas of former Judge Robert Bork, as set forth in The Antitrust Paradox,(1) continue to shape federal merger enforcement policy for horizontal mergers. However, the time is ripe to reexamine the current reliance on Judge Bork's views. The reasons for this include (1) the increasing pace of merger activity, as indicated by Walt Disney Co.'s proposed acquisition of Capital Cities/ABC, Inc., and TimeWarner Inc.'s proposed acquisition of Turner Broadcasting System, Inc.; and (2) new developments in the study of the relationship between seller concentration and market price, principally reflected in Leonard Weiss', Concentration and Price.(2)
As explained by Audretsch and Siegfried, Concentration and Price provides the following "almost unequivocal answer" to the question: Does seller concentration in a market raise price?
Not only was concentration found generally to raise price, but the relationship holds across a broad spectrum of settings involving different product and geographic markets, centuries and time periods.(3)
These findings have broad implications for merger enforcement policy and are the basis for the proposed enforcement policy set forth here.
The Department of Justice (DOJ) first set forth its merger enforcement guidelines under section 7 of the Clayton Act(4) in 1968 during the Johnson administration.(5) For horizontal mergers, these 1968 Guidelines were essentially consistent with the presumptive approach to section 7 enforcement taken by the Supreme Court in United States v. Philadelphia National Bank,(6) where the Court held that a merger that "produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration in that market . . ." is presumptively illegal.(7)
The Reagan administration replaced the 1968 Guidelines with new DOJ merger guidelines issued in 1982,(8) and again in 1984.(9) Also, on April 2, 1992, during the Bush administration, the DOJ and the Federal Trade Commission (FTC) jointly issued Horizontal Merger Guidelines,(10) which supersede the horizontal provisions of the DOJ's 1984 Merger Guidelines.
As explained in the 1992 DOJ and FTC Statement,(11) the 1982 Guidelines "departed dramatically" from the 1968 Guidelines and purportedly reflected current "developments in legal and economic thinking about mergers."(12)
The Reagan administration's 1982 Guidelines provided a more liberal approach to merger enforcement than the 1968 Guidelines, and the 1984, and 1992 amendments continue this liberalizing trend. More specifically, when compared to the 1968 Guidelines, the 1982, 1984 and 1992 Guidelines set forth a much higher threshold for identifying concentrated markets, give less significance to concentration analysis, and give more weight to an analysis of other factors, such as entry and efficiencies.
The more liberal concentration thresholds adopted in the 1982, 1984, and 1992 Guidelines are premised on suggestions for merger enforcement set forth by Bork in The Antitrust Paradox. As explained by Kovacic:
The 1982 . . . Merger Guidelines raised the threshold of illegality for horizontal mergers significantly above levels contained in the . . . 1968 Guidelines, but stopped short of Bork's more permissive Standards. . . [However,] in practice . . . the Reagan Administration tended to allow transactions yielding combined market shares that substantially exceed the guidelines stated limits.(13)
This more liberal approach has resulted in a substantial decline in enforcement actions.(14) The Clinton administration apparently has no plans to amend the 1992 Horizontal Guidelines,(15) and the enforcement of these Guidelines by the Clinton administration seems to be no more aggressive than enforcement by the Bush administration.(16)
It is the purpose of this article to examine whether the policy prescriptions for horizontal mergers proposed by Bork and essentially embodied in the 1992 Horizontal Guidelines are consistent with current economic learning.
In the words of section 7, this article addresses the issue of whether in an appropriately defined line of commerce (i.e., product market) and section of the country (i.e., geographic market) the "effect [determined by the applications of microeconomic reasoning] of [the] acquisition may be substantially to lessen competition or to tend to create a monopoly?" Thus, this article seeks to determine the appropriate application of the "effects" clause of section 7. In answering this question the focus is purely on an economic approach, as distinguished from a social or political approach.(17)
Section II sets out Judge Bork's prescription for the effects test, and section III surveys the theoretical and empirical research, including Concentration and Price, dealing with the correlation between concentration and noncompetitive pricing. Although it is not possible to examine all of the literature on this subject, a review of the principal authorities is sufficient for developing the suggested enforcement rules developed in sections IV and V.
Section IV addresses the proper scope of the effects test under section 7, and section V uses the principles discussed in sections III and IV in developing a suggested enforcement policy for horizontal mergers, that is, a proposed section 7 enforcement standard.(18) Finally, section VI provides concluding remarks.
II. Judge Bork's merger prescriptions
In The Antitrust Paradox, Bork sets out prescriptions for horizontal merger enforcement that were largely accepted by the Reagan and Bush administrations, and seem to be accepted by the Clinton administration.(19) Although the 1982, 1984, and 1992 Guidelines do not go the full way in adopting Bork's prescriptions, his positions have essentially been followed by both the DOJ and FTC in practice.(20)
Judge Bork frames the issue under section 7 for horizontal mergers as follows:
In applying Clayton 7's injunction to judge mergers, what rules should the court evolve so as to avoid monopoly and preserve competition?(21)
This starting point understates the scope of section 7, the purpose of which is to prevent any merger where, in the relevant market, "the effect of such acquisition may be substantially to lessen competition or tend to create a monopoly." Thus, the manner in which Bork frames the issue ignores the incipiency concept embedded in the words "may be substantially to lessen competition or tend to create a monopoly." He also frames the issue solely in economic terms, which also is the approach taken in this article.
He then makes it clear that in propounding an enforcement rule the focus should be on the market shares of the merging firms, which is the approach followed by the Supreme Court:
In choosing the market shares that may be created by horizontal merger, we are dealing with a spectrum, and it must be confessed that the proper place to cut it is not entirely clear.(22)
This statement acknowledges the uncertainty associated with the assessment of the concentration level resulting from a merger.
A. Polar positions
In attempting to get at the problem he starts with polar positions. He first posits that "[w]hen companies each having 1% of a fragmented market merge, they cannot be supposed to have monopoly profit in mind. Their motivation and hence the presumed effect of their action, must be either increased efficiency or some effect irrelevant to antitrust."(23) He thus correctly concludes that such a merger should be lawful. This view is embodied in the 1968, 1982, 1984, 1993, and NAAG Guidelines and also in both the Supreme Court cases and the legislative history of section 7.
The other polar case he posits involves a merger to monopoly by two companies each having 50% of a market. He concludes that the "motivation and effect of their act are not free from doubt," and "[t]he law is justified in striking down such a merger on the theory that restriction of output is certain and an offsetting increase in efficiency, so far as the law enforcement officers can judge, is not."(24) He goes on to point out that the policy of prohibiting this type of merger to monopoly presents a "problem of a welfare trade-off [and] [t]he policy may incur some efficiency Costs."(25)
The welfare trade-off he speaks of here is between the dead-weight loss resulting from the output restriction associated with monopoly and the potential productive efficiencies that may result from the monopoly.
B. Focus exclusively on output restriction
In condemning mergers to monopoly, Bork focuses only on the output restriction resulting from monopoly, and ignores the transfer of wealth from consumers to producers that results from the increase in price.(26) This approach of ignoring the distribution of income between producers and consumers is what Judge Bork refers to as one of the two "primary characteristics" of the Chicago school of antitrust: "the insistence that the exclusive goal of antitrust adjudication, the sole consideration a judge may bear in mind, is the maximization of consumer welfare."(27) The second Chicago principle is the rigorous application of economic analysis to "test the propositions of law and to understand the impact of business behavior on consumer welfare."(28)
Bork classifies profit-maximizing behavior into three different categories:
A business firm may seek to increase its profits  by achieving new [productive] efficiencies (beneficial),  by gaining monopoly power and restricting output (detrimental), or [3) by some device not related to either productive or allocative efficiency, such as taking a bookkeeping advantage of some wrinkle in the tax law (neutral).(29)
On the basis of this classification system, he argues that the "task of antitrust is to identify and prohibit those forms of behavior whose net effect is output restricting and hence detrimental . . . [and to] leave untouched behavior that is beneficial or neutral."(30)
Bork's view of consumer welfare is different from what one might expect. He views both producers and consumers as consumers, and he is not at all troubled by the wealth transfer from consumers to producers that results from monopoly. The only infirmity with monopoly in his view is the reduction in production that results in the welfare triangle or the deadweight loss associated with monopoly. Bork explains that the formation of a monopoly shifts income from consumers to the monopoly and its owners who are also consumers. He further argues:
This is not deadweight loss due to restriction of output but merely a shift in income between two classes of consumers. . . . [A] decision about . . . [this shift in income] requires a choice between two groups of consumers that should be made by the legislature rather than by the judiciary.(31)
This limited view of consumer welfare is certainly not what Congress had in mind when enacting section 7, and sections 1 and 2 of the Sherman Act. As early as 1910 in the Standard Oil case the Supreme Court set out three reasons under the common law for opposing monopolies and indicated that these reasons were also a basis for adoption of sections 1 and 2 of the Sherman Act. The first reason given by the Supreme Court was the "power which the monopoly gave to the one who enjoyed it to fix the price and thereby injure the public."(32) Thus, the focus is on the adverse effect of higher prices on the public, that is, the general body of consumers. Apparently, nowhere in the history of the antitrust laws does Congress purport to treat the owners of firms as the equivalent of consumers. Thus, its seems clear that in enacting sections 1 and 2, Congress was concerned with protecting "public" consumers from higher prices that would result from a monopoly. Since that was one of the purposes of sections 1 and 2, it is also one of the purposes of section 7, which is designed to arrest monopolies in their incipiency.(33)
Bork may be correct when he says that a shift in "income distribution does not lessen total wealth, and a decision about it requires a choice between two groups of consumers [i.e., the public and the owners of the monopoly firm] that should be made by the legislature rather than by the judiciary."(34) What he does not acknowledge, however, is that in enacting the Sherman Act in 1890, the legislature has made the income distribution choice in favor of the public consumers.
Indeed, it is intuitively obvious that the legislature would make just such a judgment because as Arnold Harberger explains "monopoly profits can be treated as a privately imposed and privately collected tax."(35) Would any legislator take a neutral position on privately imposed taxes?
Bork's argument that distributional effects of monopoly should be ignored in the antitrust law is an attempt to ascribe to the legislatures that enacted these laws one of the basic tenets of applied welfare economics, which Harberger specifies as follows:
[W]hen evaluating, the net benefits or cost of a given action (project, program, or policy), the costs and benefits accruing to each member . . . [or] group . . . should normally be added without regard to the individual(s) to whom they accrue.(36)
Harberger goes on to explain that distributional effects "may be exceedingly important perhaps even the dominant factors governing any policy decision . . ."(37) He argues, however, that distributional effects "are not part of the package of expertise that distinguishes the professional economist from the rest of humanity."(38)
This tenet of welfare economics gives conservative economists an avenue of escape from one of the most important issues facing any society: the distribution of income. This approach to welfare economics could result in the adoption of policies that lead to significant distortions in the distribution of income, which could severely restrict economic growth. It seems obvious that countries in which income distribution tends to be more equal, such as in northern Europe, Canada, Japan and the U.S., enjoy more economic progress than countries with significant distortions in income distribution such as in many countries in Latin America.
Bork's argument for a distributional neutral antitrust policy not only pretends that Congress has not addressed the issue, but also is an abdication of a major responsibility of anyone who attempts to think seriously about antitrust policy. The following words from Richard and Peggy Musgrave in their public finance text seem to be spoken directly to Bork:
Policy measures . . . have distributional repercussions . . . . [G]iven the close bearing of distributional issues on questions of economic policy, economists who are concerned with public policy can hardly detach their thinking from equity issues.(39)
C. The intermediate case
Bork posits that "[a]s the law moves down the spectrum from the 100% merger, the problem of the trade-off between restriction of output and efficiency rapidly becomes severe."(40) By this he means that the less the concentration associated with the merger, the less the deadweight loss or loss in allocative efficiency and the greater the potential productive efficiencies. He refers to this as a "mixed case" that produces both output restrictions and productive efficiencies. With regard to this mixed case, which he says arises primarily in horizontal mergers, he correctly argues that a prohibition against all horizontal mergers, no matter how small, will deter productive efficiencies, and a rule allowing all horizontal mergers, no matter how large, would deter allocative efficiencies. He adds:
Somewhere on the spectrum between large and small mergers lies a range of mixed cases, and in these we do not know with certainty whether the efficiency or the trade restraint element predominates.(41)
Further, on this same point Bork correctly observes that "[i]n some mergers there will be a deadweight loss and no efficiency gain; in others there will be an efficiency gain and no deadweight loss; and in some both these effects will be present."(42)
1. Rejection of Economies Defense Bork rejects Williamson's efficiency argument to the effect that the trade-off between loss in allocative efficiency and gains in productive efficiencies can be properly determined in an antitrust case.(43) Williamson argues that even a small decrease in cost, that is, small productive efficiencies, can outweigh the deadweight loss or the decrease in allocative efficiency. This trade off is illustrated by the figure. Williamson correctly points out that the savings from the productive efficiencies indicated by the darkened rectangular area can outweigh the deadweight loss represented by the welfare triangle.
On the other hand, Judge Bork persuasively argues that a "[p]assably accurate measurement of the actual situation is not even a theoretical possibility,"(44) and that the "quantification of the productive efficiency factor . . . [is] utterly insoluble."(45) He thus rejects an efficiencies defense in the context of horizontal mergers:
The trade-off problem arises primarily in the context of horizontal mergers, and there we can take it into account by framing rules about allowable percentages that reflect the probable balance of efficiency and restriction of output.(46)
On this point, he further says "[i]f we are in fact dealing with a spectrum - if, that is, some non-collusive restriction of output actually occurs in markets with `few' firms - then general rules can at best be expected to do more good than harm on balance."(47)
Bork notes that De Prano and Nugent have argued that very large productive efficiencies would be required to offset the associated price increase resulting from a merger that created both market power and productive efficiencies.(48) Bork quotes the following conclusions reached by De Prano and Nugent: "the required cost reductions would have to be so large that they are very unlikely to be forthcoming" and for this reason, "more rigorous activity on the part of antitrust authorities would be likely to serve both the goal of increasing net economic welfare, and that of equity in income distribution."(49) He summarily rejects their concerns with the adverse price effects on the grounds that he has already shown that "antitrust should never concern itself with equity in income distribution."(50)
On this important issue of the significance of efficiencies, Bork seems to be closely aligned with the general approach of the Warren Court in focusing on market shares in merger cases and in generally rejecting an economies defense. As indicated above, in United States v. Philadelphia National Bank,(51) the Warren Court adopted a presumptive test of illegality under section 7 if the merger "produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market. . . ."(52) And, in F.T.C. v. Procter & Gamble, Co.,(53) with Justice Douglas writing, the Court said:
Possible economies cannot be used as a defense to illegality. Congress was aware that some mergers which lessen competition [i.e., in Judge Bork's terms, reduce output, thereby resulting in a decrease in allocative efficiency] may also result in [productive] economies but it struck the balance in favor of protecting competition [i.e., allocative efficiency].(54)
As will be seen below, Bork and the Warren Court part company, however, on the level of concentration that should give rise to a presumption of illegality under section 7.
2. Rationale for Enforcement policy in mixed cases Judge Bork frames the issue for those mixed mergers that may produce both losses in allocative efficiency and increases in productive efficiencies as follows: "What size horizontal mergers should be banned?"55 He rejects the stringent merger policy argued for by De Prano and Nugent on the grounds that their analysis applies only to mergers to monopoly and "tell[s] us nothing of the wisdom of permitting or opposing mergers that create market shares of 10, 30, or even 75%."(56) He argues that the trade-off analysis does not show whether restrictions in output will occur at such levels of production and does not predict the amount of any output restriction.(57) Therefore, it is impossible to predict the amount of the deadweight loss, if any.(58) He argues correctly that the needed data "must be estimated according to our theory of the effects on output of particular degrees of concentration."(59) On this point, Bork seems to be on all fours with the Supreme Court's presumptive test in Philadelphia National Bank.
Bork then sets out his theory on the effects concentration has on output. He first points out that he is "not persuaded that [noncollusive oligopolistic] behavior occurs outside the economics textbook . . .," and he believes that in any event such behavior "rarely results in any significant ability to restrict output."(60) Along this same line he says that he "doubts that there is any significant output restriction problem arising from the concentration of any industry."(61) He argues that the "available evidence strongly suggests that oligopolies do not generally result in substantial or significant restrictions of output."(62) He further says that it is possible that oligopolies result in no output restriction "but the evidence does not permit one to make that statement with complete confidence."(63)
He refers to various concentration studies that found a correlation between high concentration and high profits and observes that these studies have been discredited.(64) He points out that high rates of return are consistent with superior efficiency as well as with concentration, a point first made by Harold Demsetz.(65) Judge Bork thus concludes that these concentration-profit studies "prove nothing of interest to antitrust policy."(66) The assessment of the many concentration-profit studies discussed by Richard Schmalensee in the Handbook of Industrial Organization(67) paints a much more balanced picture of these studies than the short shrift Bork, who is not a professional economist, gives to them.
The observation by Demsetz of a dual possible explanation for high profitability in concentrated industries is certainly true. It would appear, however, that any benefits of the efficiencies realized by the high profit firms are not being passed on to consumers in the form of greater production and lower prices, which would be present if the industry were in long-run competitive equilibrium. Consequently, any industry in which substantial high profits are present would probably depart substantially from a competitively organized industry.
Professor Weiss, one of the principal proponents of the concentration-profit thesis argues that "Demsetz' empirical work on the subject was not convincing."(68) Notwithstanding his skeptical view of the Demsetz theory, in an attempt to construct a "test of oligopoly theory [that] can convince those who find Demsetz' argument persuasive or at least disturbing,"(69) Weiss has studied the relationship between concentration and price. This study, which found a high correlation between concentration and price, is examined, along with other such evidence, in section III, infra. Bork does not address this evidence in either the Introduction or the Epilog to the 1993 edition of The Antitrust Paradox. Thus, as will be demonstrated in section III, Bork's first justification for his view that oligopoly rarely results in output restrictions is not persuasive.
The two other bases for his conclusion are that observations from antitrust cases and from studies of regulated industries indicate that oligopoly does not produce results approximating monopoly.(70) He summarized the evidence from antitrust cases as follows: "(1) a large price drop occurs when one firm appears to challenge an established monopolist; (2) oligopolists are frequently discovered in overt collusion; and (3) even overt collusion among oligopolists frequently breaks down, so that prices drop to competitive levels."(71) First, as demonstrated in section III, infra, recent studies of certain high concentration horizontal merger cases that were not challenged have found that many of these mergers have resulted in higher prices. Second, the fact that oligopolists are frequently discovered in overt collusion is a reason for opposing oligopoly even though overt collusion may break down.
With respect to regulated industries he says that economic studies indicate that regulation of monopolies produces little change in price levels, and regulation of oligopolies produces substantial increases in price levels.(72) These observations merely show the failure of regulation and emphasize the benefits to be derived from competitively organized industries; they do not demonstrate that output restrictions do not occur in concentrated markets.
Finally, Bork argues that "conventional oligopoly theory predicts results that are not observable in markets to which the theory should apply if it applies anywhere,"(73) and he goes on to conclude that "we are entitled to be highly skeptical of the entire theory of oligopolistic interdependence. . . ."(74)
As will be discussed in greater detail in section III, infra, Weiss, who has found "overwhelming" evidence of the relationship between concentration and price, points out that "almost all oligopoly theory makes predictions about price, and with few exceptions, those predictions are that price will rise with concentration."(75)
3. Specific Prescription for enforcement policy On the basis of the rationale discussed in the preceding section, Bork concludes that "[s]ince the amount of restriction of output seems to decrease greatly from 1-firm markets to 2-firm markets, . . . and since mergers may very well create substantial new efficiencies, we are in an area of uncertainty when we ask whether mergers that would concentrate a market to only two firms of roughly equal size should be prohibited."(76) In making this assessment he points out, incorrectly that "we do not know that there is any significant non-collusive restriction of output in 2-firm markets . . . ."(77) He goes on to conclude that it is his "guess" that output restrictions would not result in two-firm markets and that, therefore, "mergers of up to 60 or 70 percent of the market should be permitted . . . ."(78) He notes that this threshold "resembles the old Sherman Act merger rule and Judge Hand's dictum in Alcoa."(79)
He backs off of this 60% or 70% threshold "[p]artly as a tactical concession to current oligopoly phobia and partly in recognition of Section 7's intended function of tightening the Sherman Act rule. . . ."(80) He concludes:
Competition in the sense of consumer welfare would be adequately protected and the mandate of Section 7 satisfactorily served if the statute were interpreted as making presumptively lawful all horizontal mergers up to market shares that would allow for other mergers of similar size in the industry and still leave three significant companies.(81)
He points out that this merger rule would have a maximum attainable market share by merger of 40% in a fragmented industry.(82) If a company already had more than 40%, the maximum would be scaled down. As an example of this scaling he says that if a firm already has 50% of the market it could not engage in horizontal mergers and no other merger could produce a firm with more than a 30% market share.(83)
As pointed out above, although the 1982, 1984, and 1992 Guidelines do not explicitly adopt Bork's prescription, these Guidelines are significantly more liberal than the 1968 Guidelines, and "in practice both the Antitrust Division and the FTC during the Reagan Administration tended to allow transactions yielding combined market shares that substantially exceed the guidelines stated limit."(84) Thus, the enforcement policies adopted by both the Reagan administration and the Bush administration were consistent with Bork's prescriptions.
III. Survey of findings of concentration studies
A. Significance of concentration studies
The significance of the study of the relationship between concentration and anticompetitive market performance for purposes of merger analysis is illustrated in Judge Posner's opinion in United States v. Rockford Memorial Corp.(85) In this case, the merger between two nonprofit hospitals was challenged by the government on the ground that the transaction violated section 1 of the Sherman Act. Posner indicated that the government could also have challenged the transaction under section 7 but was procedurally barred from doing so. He said, however, that the judicial interpretations under section 1 and section 7 have "converged" and that the court "doubts whether there is a substantive difference today between the standard for judging the lawfulness of a merger challenged under Section 1 of the Sherman Act and the standard for judging the same merger challenged under Section 7 of the Clayton Act."(86)
Posner then noted the presumption that present law applies upon a finding of a high level of concentration resulting from a merger:
[O]nce the government showed that the merger would create a firm having market share approaching, perhaps exceeding, a common threshold of monopoly power - two-thirds (United States v. Aluminum Co. of America, 148 F.2d 416, 424 (2d Cir. 1945) (L. Hand, J) - it behooved the defendants to present evidence that the normal inference to be drawn from such a market share would mislead.(81)
A merger resulting in a firm controlling two-thirds of the market would be illegal even under the prescriptions of Bork.
Posner then goes on to discuss the significance of empirical studies of the relationship between concentration levels and anticompetitive market behavior: It is regrettable that antitrust cases are decided on the basis of theoretical guesses as to what particular market-structure characteristics portend for competition, but to place on the government the insuperable burden of proof is not the answer. We would like to see more effort put into studying the actual effects of concentration on price in the hospital industry as in other industries. If the government is right in these cases, then, other things being equal, hospital prices should be higher in markets with fewer hospitals. This is a studiable hypothesis, by modem methods of multivariate statistical analysis, and some studies have been conducted correlating prices and concentration with hospital industry. Unfortunately, this literature is at an early and inconclusive stage, and the government is not required to wait the maturation of the relevant scholarship in order to establish a prima facie case.(88)
This statement by Posner, that there is not enough "empirical guidance from industrial organization economists" has been described as the Posner lament. This lament falls short of Bork's statement that there is no convincing evidence of a relationship between concentration and anticompetitive pricing.
Posner's lament was addressed in a 1991 article entitled Policy Adequacy of the Empirical Concentration Price Studies: The Posner Lament.(89) In the article, Charles E. Mueller, the editor of the Antitrust Law & Economics Review, published the views of forty-seven industrial organization economists on the following questions:
Is Judge Posner simply inadequately informed as to the true inventory of studies on the concentration-price relationship or is there in fact an empirical gap such as he describes?(90)
In explaining the importance of the empirical questions, Mueller poses the following general policy questions:
Does high concentration lead to higher-than-competitive prices . . . ? If so, is there some "critical" level of concentration at which
those higher-than-competitive prices regularly appear? . . .
If supercompetitive prices commonly appear in concentrated industries, is the underlying price-raising mechanism oligopolistic collusion (express or tacit) or . . . unilateral market power . . . ?(91)
He goes on to ask whether the four-firm concentration ratio should be replaced with a one-firm figure. The issue concerning he possibility of a one-firm critical concentration ratio was addressed directly by Mueller in another survey entitled Market Power and the Critical Concentration Level: Need for a 1 -Firm Threshold Number?(92) This latter article publishes the views of thirty-eight industrial organization economists on this question.
In the introduction to the Posner lament article, Mueller criticizes Judge Posner's adoption of the two-thirds rule from the Alcoa decision as the "common threshold of monopoly power":
[This threshold] is rather dramatically inconsistent, however, with the empirical findings of industrial-organization economists as reported in standard text, . . . that higher prices generally begin when the 4-finn concentration ratio reaches about 50% (CR-50).(93)
In support of this proposition, Mueller cites Weiss' book, and a book by Scherer and Ross.(94) Mueller also points out that the "rather extensive nature of . . . [the) empirical literature on the concentration-price/profits relationship" is addressed by Richard Schmalensee in a chapter in the Handbook of Industrial Organizaion.(95)
These three books were also mentioned by several of the surveyed economists in their responses to the questions posed. Also cited was a paper by Gregory Werden, an economist at the Justice Department's Antitrust Division, entitled A Review of the Empirical and Experimental Evidence on the Relationship Between Market Structure and Performance.(96)
In the introduction to the Posner lament article, Mueller also says:
[I]f Posner . . . [is] using a CRI-64% threshold of market power without realizing that empirical economic science evidently puts it at e.g. CR4-50% (implying a 1-firm share or perhaps well under 30%) - bringing policy and science into alignment should be an easily remediable problem, namely calling the matter to everyone's attention.(97)
In the introduction to the one-firm concentration level article, Mueller says:
Scherer and Ross ... note, for example, that, as a very . . . general rule, if evenly matched firms supply homogeneous products in a well-defined market, they are likely to begin ignoring their influence on price when their numbers exceeds 10 or 12, which of course implies individual market shares of . . . 8% or 10% and a 4-firm share of about 40% at the recognition boundary.(98)
Mueller also points out that Scherer and Ross, citing a 1982 study of manufacturing firms by Kwoka, say that the study finds that the average market share of the leading firm was 17.5%, with a range up-to 68.7% and that the average four-firm concentration ratio was 39.8%.99 Thus, the average leading firm held 44% of the average four-firm concentration level. Mueller then asks:
If the average No. 1 firm holds a share that is 44% of the average 4-firm total . . . - and if CR4-50% is . . . empirically-valid . . . - does that mean that, on average, such price-raising market power is likely to be present when . . . [a] single firm has a share of 22% . . . ?(100)
In his letter to the surveyed economists, Mueller specifically asks whether the critical concentration ratio needs to be expressed as a one-firm rather than as a four-firm number.(101)
This section analyzes the views expressed by the economists in both the Posner lament article and in the White et al. article with the purpose of determining whether there is common ground on these issues. In particular, this section seeks to determine whether the surveyed economists generally are in agreement with Mueller's conclusion that the critical four-firm concentration ratio is 50%, implying that the critical one-firm ratio is 30%. Significant agreement on Mueller's conclusion would be a basis for incorporating these findings in the horizontal merger guidelines.
B. Difference between studies of concentration-profits and
In attempting to determine if there is a critical concentration ratio, economists first studied the relationship between concentration and profits. One of the best survey articles of the findings of these studies was one by Weiss entitled The Concentration-Profits Relationship and Antitrust.(102)
Weiss first shows that dominant firms have higher than average profits,(103) and points out that most oligopoly theories generally "predict a positive effect of concentration on profits."(104) He then surveys the concentration-profits studies and concludes that the "bulk of the studies show a significant positive effect of concentration on profits or margins,"(105) and he generally criticizes the studies that do not find such a relationship.
The findings of Weiss were criticized by Demsetz in his paper Two Systems of Belief About Monopoly.(106) He asserts that "not one of the oligopoly theories [Weiss] described deals with the problem posed for colluding oligopolists by potential and actual entry."(107) Demsetz summarizes his view of the theoretical and empirical evidence concerning the effects of concentration as follows:
[T]he theoretical support of the market concentration doctrine . . . is weak or nonexistent. [M]ore [empirical] studies reveal a positive correlation between profits and concentration than do not. [E]nough of those . . . fail to show such a correlation, however, that the policymaker ought not suppose that conclusive evidence of the statistical relationship exists.(108)
Demsetz then goes on to show that the studies that demonstrate a relationship between concentration and profits can be explained by the fact that "some products are more efficiently produced by firms possessing a large share of the market," and that "[t]hose firms that first act on the belief that large scale is an advantage . . . will possess a competitively secured advantage in timing and in obtaining early customer acceptance that will be difficult to overcome in a short period of time."(109) He further argues that "[i]f industries are persistently concentrated, the fundamental reason will be the presence of large scale advantages."(110) He explains:
Such firms will record higher profits because the market has not yet grown sufficiently; . . . to bring smaller less efficient firms into production, prices will need to be high enough to cover . . . unit cost, and this means that price will . . . be high enough for [large firms to realize] accounting profits . . .(111)
Demsetz further argues that "after the size of the firm is taken into account, no significant correlation between profit rate and market concentration is apparent. . . "(112)
Weiss characterizes Demsetz' theory as "indicating that the high profits of large firms in concentrated industries arise from economies of scale, superior products, or good management which the small firms cannot duplicate."(113) Weiss then argues that if Demsetz is correct there should be no reason why this "same effect should not be true in unconcentrated as well as concentrated industries."(114) Weiss goes on:
Moreover, the weak or generally insignificant negative effect of concentration on small firms' profits does not undermine the notion that high concentration affects profits by making tacit or explicit collusion more effective. The expectation that high profits will attract inefficient firms that could not survive in a highly competitive industry is an old one.(115)
Further, Weiss goes on to say that even if Demsetz is correct in concluding that "high concentration is derived from low costs, superior products, or superior management that does not preclude the leading firms from benefitting from an oligopolistic meeting of minds that is made possible by that high concentration."(116)
In response to Demsetz' critique of the concentration-profits studies, Weiss and others switched to the study of the relationship between concentration and prices. Weiss explains:
What is clearly needed is a hypothesis that follows directly from oligolopoly theory and will not succumb to [Demsetz'] alternative interpretation.(117)
Weiss goes on to explain the virtues of studying the relationship between concentration and price:
The obvious hypothesis is that price will rise with concentration. Predictions about price are less equivocal results of oligopoly theory than those about profit.(118)
The following sections analyze the views of economists on the issue of whether the current state of empirical research, including Weiss' book, establish a concentration-price relationship. It is the state of this research that is at the core of Posner's lament, and the best way to assess this issue is to examine the views of the experts.
C. Concentration proposition
Sections 1, 2, 3, and 4 analyze the views expressed by the economists surveyed in the Posner lament article.
1. SUPPORTING VIEWS AND OTHER EVIDENCE In his response to Mueller's question Stephen Rhoades of the Federal Reserve Board first points out that "[a]s a result of increasing questions . . . in connection with concentration-profit studies, a rather large number of concentration-price studies have been conducted in a fairly short period of time."(119) And, he goes on to cite Weiss, Concentration and Price.(120) He says that Weiss' "summary judgment of the literature is what there is overwhelming' evidence of a concentration-price relationship."(121)
Mr. Rhoades goes on to indicate that he is in agreement with this conclusion:
[I]f you give me a monopoly position, I am not going to charge a price that just covers cost. If I have only a small number of competitors, [it is] in my best interest to get along with my "competitors" well enough to maximize joint profits. This suggests pricing above Cost.(122)
He goes on to say that he thinks there have been enough concentration-price studies, "especially when combined with standard economic theory, to establish the concentration-price relation as a basic economic relationship."(123)
Basically agreeing with the views of Rhoades, Takeo Nakao of Doshisha University, Kyoto, Japan, says that "the proposition that an increase in market share raises the price-cost margins seems to be empirically verified.(124) Ellen Magenheim of Swarthmore College says that there is no "empirical gap" regarding concentration-price relationships and in support of this assertion quotes Richard Schmalensee's chapter in the Handbook o Industrial Organization(125) in which he writes that in "cross-section comparisons including markets in the same industry, seller concentration is positively related to the level of price and that the relation between concentration and price seems much more robust statistically than that between concentration and profitability (pp. 987-988)."(126)
Citing Weiss(132) William Baldwin of Dartmouth College says that the "preponderance of evidence . . . is that there is a positive relationship between either concentration or individual firm market share, on the one hand, and price on the other."(128)
Lucile Keyes of Washington, D.C. says that Werden(129) convinces her that "while the evidence is certainly inconclusive, there is probably enough to establish a prima facie case for a positive relationship."(130)
Kenneth George of the University College of Swansea, United Kingdom, adds that for the U.S. "there is a lot of evidence to support a positive link between structure (height of the concentration ratio) and price-cost margins and stronger evidence on a positive link between price-cost margins and the market share of the leading firm."(131)
Citing among other things Weiss(132) and Scherer and Ross(133) as the support for the general proposition, James Brock of Miami University also refers to his forthcoming antitrust play, Antitrust Economics on Trail: A Dialogue on the New Laissez-Faire, which he says "dramatically highlights the absurdities of the Posner/Bork/Chicago school."(134)
Alfred Kahn of Cornell University says: "Every study I have seen of pricing in the airline industry over the last decade or so . . . has established a significant positive relationship between concentration and pricing."(135) And, John Kwoka of George Washington University adds that his "overall assessment is that the concentration-price studies clearly establish the association between industry structure and performance in a manner not subject to the Chicago school or any other criticism of profits studies." He further says: "Perhaps Posner should be sent a copy [of Weiss' book Concentration and Price]."(136)
Also, citing Weiss' book, William Shepherd of the University of Massachusetts says that the book "reaffirms that there is a consistent pattern which reflects market power."(137) Richard Caves of Harvard University says he sees "no substantial problems" with the qualitative adequacy of the studies in Weiss' book Concentration and Price.(138) and, William Comanor of the University of California at Santa Barbara writes: "I do not understand Posner's comment, particularly in light of Len Weiss' new book on just this relationship . . . ."(139) After citing the Weiss book, Werner Sichel of Western Michigan University says: "I don't think one can say that there is an empirical gap but of course one can always want more and also be dissatisfied with the quality of the work."(140)
In addition to the above thirteen economists, eight of the other economists surveyed also agreed with the proposition, with many citing Weiss' Concentration and Price, which was the principal source of support relied on by these supporting economists.
2. CONTRADICTORY VIEWS AND OTHER EVIDENCE Harold Demsetz of UCLA, who, as a Chicago school economist, was one of the principal critics of the concentration-profits studies, writes: "Everything is relative and, relative to the concentration-profit studies, there is relatively little systematic work of quality on the concentration price relationship."(141) He refers to Weiss' Concentration and Price and says:
Beyond this, there are some industry specific studies. Taking, these studies into account, the quality adjusted sum total of this work - given the data quality problems and the assumptions needed to pretend these problems do not exist - is meager.(142)
Edward Ray of Ohio State University writes: "[M]y very imperfectly informed opinion is that Judge Posner is correct, that the empirical connection between market structure and competitive behavior has not been established well enough to provide much predictive power about the potential restraint on competitive forces associated with particular market structures."
Although he cites Weiss' Concentration and Price, Donald Alexander adds: "[T]he empirical generalizations linking concentration and price-which would provide judges with a reliable guide to use in deciding cases-is as yet uncovered."
Paul Auerbach of Kingston Polytechnic, United Kingdom, says that he is "extremely dubious about the possibility of encapsulating the process of competition in a simple concentration parameter." Samuel Baker adds: "Posner is correct, that the relationship is not well established."
Citing Concentration and Price, which he refers to as state-of-the-art, William Hogan of Southwestern Massachusetts University says: "Though I do not consider myself an expert in this area, I believe that the connection between concentration and price has not been adequately explored in general."
Thomas Chirikas of the University of Southern Florida says that "Judge Posner is right about the lack of current empirical estimates of the concentration-price relationship in local hospital markets." Professor Chirikas did not comment on the general state of concentration price studies.
Mark Schupack if Brown University says: "I must reluctantly agree with Judge Posner that this `literature is at an early and inconclusive stage.'" He goes on to add: "I should emphasize again that I have not been able to study in detail the few references I found where the concentration price relationship may have been covered." He apparently was not aware of Weiss' Concentration and Price.
Thomas Wilson of University of Toronto says "one could not rely on the general body of evidence relating to price-cost margins to concentration to infer the impact of concentration on prices charged by the hospitals."
Finally, Colin Aislabie of the University of Newcastle, Australia, says that in a case like Rockford Memorial, "we would be inclined to urge in the strongest terms that what might happen to price should be studied on a case-by-case basis, as any broad generalization is unlikely to have policy significance."
3. EQUIVOCAL VIEWS AND OTHER EVIDENCES Joseph Jadlow of Oklahoma State University says that while "it is true" that many of the concentration price studies "find a positive relationship, [the] magnitude of the effect is small in some of these studies and some other studies find either no significant relationship or a negative effect of concentration on price."
Craig Newmark of North Carolina State University adds: "There are many studies and almost all find a significant positive relationship but even staunch proponents of those studies-professor Leonard Weiss, for example, - now concede that the proper interpretation of them is unclear."
Ronald Cummings of the University of New Mexico says: "While I am not prepared to support any position, my feeling is that Judge Posner's statement is in one sense, much too general but in another probably correct."
4. SUMMARY OF RESULTS A summary analysis of the views of the above experts can be informative in assessing the state of the current economic learning.
Of the forty-seven economists whose positions on the concentration-price relationship are published in the Antitrust Law & Economics Review, the breakdown of general views is as follows:
(1) Support the proposition 21
(2) Disagree with the proposition 11
(3) Equivocal 5
(4) No opinion 10
Thus, the number of surveyed economists that agree with the proposition that there is a relationship between concentration and price outnumber those who believe that there is no such relationship by a margin of almost two to one. Further, many of the supporting economists cite to authorities supporting the proposition. On the other hand, none of the economists who disagree with the proposition cited any evidence in support of that position, and none of them specifically criticized any of the authorities cited by those supporting economists.
Section III.E. infra, reviews the authorities cited in support of the proposition.
D. One-firm concentration proposition
None of the thirty-eight economists surveyed in White et al. are of the opinion that such a critical concentration ratio exists.
Christina Kelton of the University of Minnesota, one of the contributors to Weiss, Concentration and Price, says: "Critical concentration ratios differ among industries, so it is undoubtedly hard to `pick up' a single critical ratio with our model. She adds, however, that she does not think this is a `dead' research area." Frederic M. Scherer of Harvard University adds: "[W]e know very little about critical 1-firm concentration ratio thresholds. I suspect in any event that the relationship if we could measure it statistically, would prove to be very complex, depending among other things on several additional variables."
Thus, although there is substantial support for the four-firm concentration ratio, there seems to be no support for the one-firm ratio. We do know, however, that in a study of 314 manufacturing industries, John Kwoka found that the market share of the leading firm was 17.5% on average. Also, George Stigler, of the University of Chicago and a Nobel laureate in economics, has proposed that horizontal mergers creating firms with 20% or more market shares be presumed to be in violation of section 7. Further, the 35% market share rules in the unilateral effects section of the 1992 Horizontal Merger Guidelines recognize the potential for anticompetitive behavior by dominant firms. The unilateral effects section sets out three different circumstances in which a merger of firms with combined market shares of 35% may result in a challenge. The Guidelines explain that these mergers may lead to the "unilateral" "elevat[ion] of price and the suppress[ion] of output."
E. Review of the principally cited empirical evidence and of
This section briefly reviews the authorities that were principally cited by the economists in the Posner Lament article for the proposition that there is a strong positive relationship between concentration and price. The authority cited more than any other was Weiss, Concentration and Price. Also cited were Schmalensee, Inter-Industry Studies, Scherer and Ross, Industrial Market Structure, and Werden, Review of Empirical Evidence.
It is significant that not one of the surveyed economists offered any specific criticisms of these authorities or cited to any studies critical of the findings expressed by these four authorities.
1. WEISS, CONCENTRATION AND PRICE In a summary of his findings, Leonard Weiss reports "overwhelming support to the concentration-price hypothesis" and that he has found "very strong evidence that concentration raises prices." He reports that the "effects of concentration seem to be minor at levels below CR4 of 50." He further says that although the relationship between an increase in concentration and an increase in price may be small, a small price increase could result in large profits. He says that "if profits were 10 percent of price in a competitive market, [a] 12-point rise in CR4 might raise price by only I percent, say, but this could double total profits before taxes." Along this same line he says:
Even a series of changes that raise concentration from 20 to 70 would raise price by only 5 percent. The direct welfare loss due to a 5 percent rise in price is pretty small, though the rise in profits could be huge.
He also says that he doubts that much weight should be given to any of his measures of a critical concentration ratio (CCR) because he suspects that "some industries have CCRs and others do not, and that in those that have them, they often change over time."
Although he finds no substantial evidence of a strong relationship between concentration and price below a four-firm concentration ratio of 50, he points out that certain "cases might conceivably imply that concentration has a significant effect on price well below CR4=50."
2. SCHERER AND ROSS, INDUSTRIAL MARKET STRUCTURE In a discussion of the conditions limiting oligopolistic coordination, Scherer and Ross say that the number and size distribution of sellers has an "obvious influence on coordination." They add: "Generally, the more sellers a market includes, the more difficult it is to maintain prices above cost, other things being equal."
They further point out that "as the number of sellers increases and the share of industry output supplied by a representative firm decreases, individual producers are increasingly apt to ignore the effect of their price and output decisions on rival reaction and the overall level of prices."
They go on to set forth the following proposition:
As a very crude general rule, if evenly matched firms supply homogenous products in a well-defined market, they are likely to begin ignoring their influence on price when the number exceeds ten or twelve. 
This proposition implies a four-firm concentration ratio of 40%, which is close to the 50% four-firm concentration ratio that Weiss finds as generally applicable.
3. SCHMALENSEE, INTERINDUSTRY STUDIES Richard Schmalensee reports that "[i]n samples of U.S. firms or business units that include many industries, market share is strongly correlated with profitability. . ." Although he concludes that the relationship between "seller concentration and profitability is weak . . .," he finds "strong support" for the proposition that "[i]n cross-section comparisons involving markets in the same industry, seller concentration is positively related to the level of price." Thus, this review, which was published before the publication of Professor Weiss' book, provides substantial support for the concentration-price proposition.
4. WERDEN, REVIEW OF THE EMPIRICAL EVIDENCE In his Review of the Empirical Evidence, Werden reports that most of the studies of the relationship between concentration and price "found a statistically significant, positive relationship between concentration and price, but the effect was often small."
On the other hand, he reports that there have been "substantial criticisms of the data used in the concentration-profit studies and of at the major variables." He further says that "by 1987 Weiss came to believe that the literature proved nothing and stated that `the concentration-profits relationship now seems a weak reed on which to hang a major branch of economics.'"
F. Policy implications
A review of the empirical evidence leads one to conclude that a strong case has been presented for a positive relationship between concentration and price. There appears to be substantial agreement within the economic profession on this point. Thus, while more evidence would always be better, there does not appear to be a sound basis for Judge Posner's lament.
Also, there appears to be no empirical economic evidence supporting Judge Bork's view that "noncollusive oligopolistic behavior . . . [does not] exist[s] outside of economics textbooks. . ." Bork's merger prescriptions are based on his erroneous view of the economic evidence, and it is, therefore, appropriate to reject his prescriptions.
The findings here are used in developing the structural effects test proposed in Section V, infra.
IV. Development of an appropriate effects test
under section 7
A. Scope of the effects test of section 7
Under the effects test of section 7, an acquisition is unlawful if in an appropriately defined product and geographic market the "effect of such acquisition may be substantially to lessen competition or to tend to create a monopoly."
The statute does not prohibit acquisitions where the effect will be substantially to lessen competition or tend to create a monopoly; the statute prohibits acquisitions where the effect may have such results. Thus, in applying microeconomic principles to section 7, one is not searching for certainty. As the Supreme Court said in Brown Shoe Co. v. United States:
Congress used the words "may be" substantially to lessen competition (emphasis added), to indicate that its concern was with probabilities, not certainties.
In a footnote to the above quote, the Court set forth the following discussion from the legislative history of the Clayton Act: "The use of... words ["may be"] means that the... [statute] would not apply to the mere possibility but only to the reasonable probability of the pr[o]scribed effect." The Court also said that the purpose of section 7 was to "arrest mergers at a time when the trend to lessening of competition . . . was still in its incipiency." Thus, as even Bork has recognized, section 7 is an incipiency statute; not a deconcentration statute.
This incipiency concept is consistent with microeconomic theory that teaches that a competitive market is far superior to both a monopoly market and an oligopoly market. Since this is so, it is eminently sensible to have an antitrust policy designed to prevent movements of competitive markets toward oligopolistic markets and movements of oligopolistic markets toward monopoly.
In 1986 the Reagan administration proposed legislation that would have amended section 7 to, inter alia, replace the words "the effect of" with "there is a significant probability that" and replace "may" with "will." Also the fundamental competitiveness bill of 1992 would replace the effects clause in section 7 with the following standard: "there is a significant probability that such acquisition will substantially increase the ability to exercise market power." Also this bill defines the term "ability to exercise market power" to mean the ability of one or more firms profitability to maintain prices above competitive levels for a significant period of time. The bill provides that "all relevant economic factors" are to be considered in making this determination. In focusing exclusively on the ability to exercise market power, the bill would prohibit the consideration of noneconomic factors in merger enforcement.
These proposals, which have not been enacted, would severely erode the incipiency concept currently embedded in section 7.
The effects test in section 7 reaches acquisitions that may substantially lessen competition or tend to create a monopoly. From a microeconomic standpoint, the "substantially lessen competition" standard would appear to encompass the "tend to create a monopoly" standard. Any acquisition the "effect of which may be substantially . . . to tend to create a monopoly" would seem also to be encompassed by the substantially to lessen competition standard.
The application of microeconomic theory to determine whether the effect of an acquisition may be substantially to lessen competition leads to the following reformulation of the section 7 standard: An acquisition is unlawful if in an appropriately defined product and geographic market the effect of the acquisition may be substantially to increase price above or to reduce output below the levels that exist in the premerger market. Thus, in a microeconomic sense the inquiry is whether the acquisition may substantially erode the price and output performance of the premerger market. This reformulation is consistent with the exclusive focus in the fundamental competitiveness bill of 1992 on the ability to exercise market power.
This leads to the question of how much of a price increase or output restriction would be substantial. There is no guidance on this point in the legislative history of section 7. It would appear that this substantiality issue would be dependent upon the particular product and market. A relatively small increase in price on a high price product may not be substantial, whereas such an increase on a low price item might indeed be substantial. One way of getting at this issue is to focus on the question of a price increase that would not be substantial, that is one that would be de minimis. Thus, the question could be whether the effect of the acquisition may be to increase price or decrease output by more than a de minimis amount above the levels in the premerger market.
This is a difficult question. If the premerger market is perfectly competitive and the merger does not produce a firm that could dominate the market, the answer seems clear that the acquisition would not have the proscribed effect. On the other hand, if the market is less than perfectly competitive and the merger would make the market even less so, then the question cannot be answered with precision because the pricing and output levels in such situations are inherently indeterminate. However, as demonstrated from the survey of the theoretical and empirical evidence on concentration in section IV infra, it is clear that the greater the concentration, the more likely it is that the merger may have the effect of substantially increasing price or decreasing output.
In developing the effects test, it is appropriate to distinguish between markets that are concentrated and ones that are not, and this is the approach taken in section V, infra.
B. Factors to be considered in structuring an effects test
In constructing an effects test the first question to be addressed is the factors to be considered in applying the test. Both the 1968 and 1992 Guidelines use a structural test based on the degree of concentration in the market and the percentage of the market controlled by the merging firms. This approach is consistent with microeconomic theory.(195) The 1968 Guidelines use nonmarket share standards only for the purpose of deciding whether to challenge an otherwise lawful merger,(196) and these Guidelines did not consider possible efficiency gains, except in extraordinary circumstances.(197)
On the other hand, the 1992 Horizontal Guidelines use the market shares and concentration standard only as the starting point for analysis. These Guidelines consider many other nonmarket share factors in determining whether to challenge a merger. These other relevant factors can be divided into three broad categories: (1) factors affecting the significance of market shares and concentration, (2) ease of entry, and (3) efficiencies. Consideration of these other relevant factors has a decided bias in favor of a finding that a merger does not violate section 7.
1. GENERAL DYNAMICS DEFENSE Although, as demonstrated below, the case for considering factors other than concentration and market share is weak, it is appropriate to take account of factors that may overstate or understate the nominal market shares of the firms as was the situation in General Dynamics (U.S. v. General Dynamics Corp., 415 U.S. 486 (1974)). The 1992 Horizontal Guidelines essentially codify the General Dynamics defense in providing for consideration of factors affecting the significance of market shares. These factors focus on changing market conditions, such as new technological developments.(198) These elements can be used to establish that a merger that nominally exceeds the market share and concentration standard does not in reality exceed the standard.
It seems reasonable to consider these factors because they relate specifically to the structure of the market and, therefore, help determine the actual market shares and concentration. These factors should, however, avert a challenge only in rare circumstances when a market is moving in the direction of the competitive model.
2. THE EFFICIENCIES DEFENSE Efficiencies should rarely be a defense to an otherwise illegal merger. First, Scherer and Ross find that there is very little evidence that mergers lead to efficiencies.(199) In fact, some evidence indicates that mergers lead to a decline in efficiency.(200) Thus, efficiency claims are likely to be overstated and unlikely to be fully realized. Second, as pointed out by Scherer and Ross, because of the size of U.S. markets, most efficiencies can be realized at market share levels that would not give rise to concerns under section 7.(201)
Further, even if efficiencies are attained, the price charged to consumers may remain high or may even increase because of the increase in concentration. In such cases, the "efficiency gains" merely mean more profits for the firm. Thus, the parties should have to establish that the benefit of the efficiencies will likely be passed on to consumers and will not result in higher prices. This type of passing on requirement is contained in the NAAG Guidelines, which require that the parties "demonstrate that the efficiencies will ensure that consumer prices will not increase despite any increase in market power due to the merger."(202) This standard does not require that prices be expected to decrease as a result of the efficiencies, which would be the effect of a pure passing on requirement.
Professor Pitofsky, who is now Chairman of the FTC, has criticized this passing on requirement on the ground that with such a requirement "efficiency defenses would only be permitted in highly competitive markets, and in such markets mergers would be unlikely to be challenged anyway."(203) Although a passing-on requirement is not explicitly contained in the 1992 Horizontal Guidelines, apparently both the DOJ and FTC may in certain cases insist that the benefit of efficiencies be passed on.(204)
The approach taken by the 1968 Guidelines in considering efficiencies only in "exceptional circumstances"(205) seems appropriate. Further, the parties should be required to meet the standard in the NAAG Guidelines of demonstrating "by clear and convincing evidence that the merger will lead to significant efficiencies."(206) This clear and convincing standard was contained in the 1984 Guidelines,(207) but does not appear in the 1992 Horizontal Guidelines.(208) In addition, the parties should have to demonstrate that comparable efficiencies could not be achieved through other means. This requirement is in both the 1992 Horizontal Guidelines(209) and in the NAAG Guidelines.(210)
The proposal here for a limited role for an efficiency defense is consistent with Judge Bork's suggestion that the efficiency defense not be permitted because the "elements of the trade-off between output restriction and efficiency gain cannot be studied directly."(211) As seen in section II, supra, however, Bork's proposal would allow mergers that create firms with much larger market shares than proposed here in section IV.
The approach to efficiencies proposed here is also consistent with the proposals of Porter who has found in his study if international competitiveness that domestic rivalry is important in stimulating international commercial success.(212) In a review of Porter's book, Brodley describes Porter's position on the efficiencies defense as follows:
The claim that a competition-reducing merger . . . is justified because
it reduces [costs] leads to the most common and most profound error
in government policy. . . . Even a rarely attainable 10 percent cost
saving, is swamped by rapid innovation and "[w]asteful" and "excessive"
competition is . . . the essence of national advantage.(213)
3. THE EASE OF ENTRY DEFENSE Under the 1992 Horizontal Merger Guidelines, an otherwise illegal merger may not be challenged if entry into the market is timely, likely and sufficient, that is, easy.(214)
Entry into a market may appear to be easy; however, if entry is in fact easy, the market should reflect the long-run stability associated with a purely competitive market. In the short run, firms can earn economic rents even in a competitive market. A competitive price will result only if sufficient firms enter in the long run. If a purely competitive market can, in the short run, have a shortage of firms, then it would appear that it is highly unlikely that entry will ever be easy in a noncompetitive market. Indeed, even if entry would otherwise be easy, the firms in the market (particularly a dominant firm) may engage in limit pricing in order to discourage entry.(215)
Reasoning from the microeconomic models it would seem that entry will rarely be easy in a market that substantially departs from the competitive model. Thus, the case for considering ease of entry is not supported by the economic models. Therefore, ease of entry should not be a factor in merger analysis.
This approach to entry seems to be consistent with that of Porter who has found in his study of competitiveness that local rivalry is important in stimulating international success.(216) In a 1991 interview Porter elaborated on this point:
[A] potential rival is not the same as an actual rival. . . . I have trouble
with . . . the notion that if there is a potential competitor . . .
then there is no reason to be worried about the merger. I do not think
this reflects the actual process by which dynamic improvements take
4. TEACHINGS OF THE FINANCIAL THEORY OF THE ACQUISITIONS PROCESS Modern finance theory of the acquisition process has ramifications in the development of an appropriate effects test, and, therefore, this theory is briefly outlined here.
Microeconomic theory teaches that departures from the competitive model can produce economic rents (i.e., profits above the cost of capital) for firms operating in the market. Thus, the only way for a firm to realize returns above its cost of capital is to put itself in the position to earn economic rents.
Modern finance theory teaches that in making an acquisition the acquiring firm should seek economic rents. This rent seeking is inherent in the net present value (NPV) model that is widely used in corporate finance for making acquisition decisions.(218) NPV is a way of determining if the cost of an investment is worth more or less than the value of the investment.(219) NPV is found by subtracting the initial cost (IC) of an investment (i.e., an acquisition) from the present value (PV) of the investment. Thus, NPV can be expressed as follows:
NPV = PV - IC
The present value is determined by discounting the expected future cash flows from the acquisition back to present value.(220) The discount rate generally should be determined through the use of the capital asset pricing model.(221) If the NPV of a proposed acquisition is negative, the acquisition should not be made because a negative NPV indicates that the firm cannot be expected to earn its cost of capital. The more positive the NPV the more desirable the acquisition.
Brealey and Myers explain that if a project has a "positive NPV then in a classical economic sense the project produces economic rents," that is, profits in excess of the firm's opportunity cost of capital.(222) These rents may be either persistent, indicating that the firm has monopoly or market power, or temporary (i.e., quasi rents), indicating that the market is not in long-run equilibrium.(223) A positive NPV is nothing more than the present value of the economic rents, and, therefore, as a practical matter whenever there is a positive NPV the firm should have some identifiable advantage over other firms in the market.(224)
Thus, a properly informed acquiring firm is on a mission to make a positive NPV investment and thereby realize economic rents. The financial goals of the acquiring firm are designed to seek precisely what section 7 is designed to prevent. This fundamental conflict between the financial goals of the acquiring firm and the public policy reflected in section 7 should be taken into account in structuring merger standards under section 7. The rules should be structured with the realization that every properly informed acquiring firm is on a rent seeking mission.
Finally, in addition to the rent seeking that is inherent in the NPV model, business strategists have suggested that firms target industries with noncompetitive characteristics. For example, Porter has suggested that an attractive industry is one in which "entry barriers are high, suppliers and buyers have only modest bargaining power, substitute products or services are few, and rivalry among competitors is stable."(225)
5. SUMMARY OF FACTORS TO BE CONSIDERED In summary, the standard to be applied in determining the effect of an acquisition under section 7 should be based on the level of the postmerger concentration in the market and market shares of the merging firms. The only other factor to be considered should relate solely to the question of whether the nominal market shares and concentration ratio significantly overstate or understate the likely future competitive significance of a firm or firms in the market. In the case of an overstatement there should be a showing that the market is moving decidedly in the direction of the competitive model. Finally, since a well-advised acquiring firm is on a rent seeking mission, the structural effects test should be conservatively structured.
This type of structural test is developed in section V.
V. Proposed structural effects test
In developing a structural test, the approach taken here is first to determine, on the basis of microeconomic reasoning, the minimum characteristics of a market that is effectively competitive for purposes of section 7, that is, to identify a market that has a degree of concentration but that nevertheless is essentially competitive. This is referred to here as a base case competitive market.(226) Once the minimum standards for a base case competitive market are established, this section then identifies a market that is substantially more concentrated than a base case competitive market. This is referred to here as a section 7 concentrated market.(227) This section 7 concentrated market becomes the trigger point for the imposition of an enforcement action under section 7. Thus, the only role of the base case competitive market is to assist in the development of the section 7 concentrated market. These proposed enforcement standards are referred to here as the proposed section 7 enforcement standards.
In developing the enforcement standards, the four-firm concentration ratio and the Herfindahl-Hirschman index (HHI) are used. As explained by Schmalensee, "received theory does not dictate the choice of concentration measure."(228) Thus, there is no good reason to think that the HHI measure used in the 1992 Horizontal Guidelines is superior to the four-firm measure. In any event, both measures are used here.
A. Base case competitive market
The problem here is to draw the line between markets that are clearly competitive and ones that are not. The best view on this point seems to be the observation of Scherer and Ross that "if evenly matched firms supply homogeneous products in a well defined market, they are likely to begin ignoring their influence on price when their number exceeds 10 to 12."(229) This conclusion is essentially confirmed by the finding of Weiss that there is "overwhelming support [for] the concentration-price hypothesis,"(230) and that the "effects of concentration seem to be minor at levels below CR4 of 50."(231)
Further, as discussed in section III, supra, there is broad support among economists for these propositions, and there do not appear to be any significant challenges to these propositions. As explained by Schmalensee in his review of interindustry studies of concentration: "In cross-section comparisons involving markets in the same industry, seller concentration is positively related to the level of price."(232)
Thus, it would appear that a rule that defined for section 7 purposes a competitive market as one in which there were no fewer than twelve evenly matched firms would reach the correct conclusion in most cases. Some departure from the evenly matched concept should be allowed, however, subject to the caveat that the departure not permit a possibility of dominant-firm price leadership. It would appear that a market in which the leading firm has no more than 20% of the market is one in which dominant-firm price leadership is unlikely to be a concern. Although there does not appear to be empirical support for a one-firm concentration ratio,(233) the average size of the leading firm in most markets appears to be approximately 17.5%,(234) and as indicated, Stigler has proposed that mergers of firms with market shares of 20% or more be presumed to be in violation of section 7.(235)
Thus, the rule proposed here for distinguishing between those markets that are clearly competitive in a section 7 sense and those that may not be is as follows:
Base Case Competitive Market
A base case competitive market has (1) at least twelve firms, (2) a four-firm concentration ratio of no more than 32% or an HHI of no more than approximately 750,(236) and (3) no firm with more than 20% of the market.
It is unlikely that either collusion or dominant-firm price leadership would exist in a base case competitive market. Consequently, section 7 should clearly not prohibit mergers in such markets as long as the postmerger market does not substantially depart from a base case competitive market. Thus, in this type of market two 10% firms could merge to form one 20% firm as long as the market did not become a section 7 concentrated market as defined in section V.B. below.
B. Section 7 concentrated market
If a postmerger market is substantially more concentrated than the maximum concentration allowed in a base case competitive market, then a merger resulting in a significant increase in concentration should be challenged. The following type of market, which is referred to here as a section 7 concentrated market, is substantially more concentrated than a base case competitive market:
Section 7 Concentrated Market
Any postmerger market with a four-firm concentration
ratio of at least 50% (or an HHI of at least 1250)(237) or
with a postmerger dominant firm with a market share of
at least 25% is a section 7 concentrated market.
Any postmerger market with these characteristics is subject to the section 7 enforcement standards set forth in the next section.
The minimum concentration level in this type of market is generally at the point at which the firms in the market are likely to begin to take account of rival firms in making price and output decisions(238) and is consistent with the findings of Weiss regarding the relationship between concentration levels and price.(239) Also, the 25% dominant-firm provision is consistent with concerns expressed by Stigler.(240) Thus, the price and output decisions in this type of market may substantially depart from such decisions in a base case competitive market. It is appropriate, therefore, to target these markets for the enforcement standards set forth below.
Concerns about anticompetitive pricing in some industries may, however, arise at a four-firm concentration ratio that is either higher or lower than the 50% level or at a dominant-firm market share that is either higher or lower than 25%. For example, Weiss finds that the critical concentration ratio for the cement industry is 60% to 80%.(241) The enforcement authorities should be able to take such findings into account in specifying the section 7 concentrated market for a particular industry. Thus, for a particular industry, the 50% and 25% market share percentages in the definition of a section 7 concentrated market could be either increased or decreased upon a showing by clear and convincing evidence that different market share rules should apply.
In the following discussion of the proposed section 7 enforcement standards, it is assumed that the regular 50% and 25% market share rules apply.
C. Proposed section 7 enforcement standards
It is proposed here to treat a postmerger section 7 concentrated market as a trigger point for enforcement of section 7. Thus, the enforcement policy proposed here would apply whenever the concentration in a postmerger market is at least equal to the concentration in a section 7 concentrated market. The proposed policy is as follows:
Proposed Section 7 Enforcement Standards
Any acquisition, other than an acquisition of a de minimis
firm in the market, producing a firm with more
than 10% of a postmerger section 7 concentrated market
(or resulting in an increase in the HHI of at least 50
points) will be challenged. In determining the concentration
level in the market and the market shares of the
merging firms, consideration will be given to whether
the data significantly understates or overstates
the likely competitive effect of a firm or firms in the
These enforcement standards rely solely on an analysis of market structure. There is no role for an analysis of other factors such as entry and efficiencies. The reference to an overstatement or understatement of a firm's market shares is designed to incorporate into market definition the General Dynamics principle.(242)
Although, a section 7 concentrated market is defined as one with a four-firm concentration ratio of 50%, mergers in such a market are permitted as long as the resulting firm does not have a market share of more than 10% (or the HHI does not increase by more than fifty points).
This rule would prevent a premerger market with a four-firm concentration ratio of 40% from becoming a postmerger market with a four-firm concentration ratio of 50%, because no resulting firm in the market could have more than a 10% share. For example, assume that the top five firms in a market had the following market shares: 15%, 10%, 10%, 5%, and 5%. The four-firm concentration ratio is 40%, and the market is not a section 7 concentrated market. However, under the proposed section 7 enforcement standards, none of these firms could merge, because any such merger would cause the market to become a section 7 concentrated market.
The purpose of the de minimis exception is to permit the acquisition of small firms that are not a procompetitive force in the market. This exception is to be applied on a facts and circumstances basis, and the exception should not be interpreted to permit a dominant firm to eliminate a small firm that is a particularly procompetitive force as in the Alcoa case.(243)
Even if the four-firm concentration ratio is less than 50% and the HHI is less than 1250, a section 7 concentrated market includes a market with a dominant firm that has at least a 25% market share. Under the proposed section 7 enforcement standards, the dominant firm in such a market could not grow by a merger unless the de minimis exception applied, and no other mergers could result in a firm with 10% or more of the market.
The economic justification for the proposed section 7 enforcement standards is as follows: A section 7 concentrated market is significantly concentrated and departures from competitive performance can be expected. In this type of market any merger that produces a firm with at least a 10% market share (or resulting in at least a fifty point increase in the HHI) can be expected to add substantially to the noncompetitive performance of the market because such mergers could significantly reduce the number of firms in the market, thereby substantially increasing the concentration in the market.
Empirical evidence concerning concentration ratios is generally supportive of this approach. For example, there is empirical evidence supporting the proposition that the critical concentration ratio (CCR) for manufacturing firms is a four-firm CCR of about 50%.(244) The studies show that once this CCR is reached, price begins to exceed cost and firms begin to show superior profits.(245)
The case for excluding consideration of other factors, such as entry and efficiencies is quite strong. In the words of Porter: "[A] potential rival [i.e., a potential entrant] is not the same as an actual rival."(246) Further, there is little evidence that mergers result in substantial efficiencies,(247) and it would appear that most scale efficiencies can be attained with market shares of around 5%.(248)
The rule regarding markets with a 25% dominant firm is consistent with the proposal by Stigler that horizontal mergers creating firms with 20% or more market shares be presumed to be in violation of section 7.(249)
Also, finance theory teaches that acquisitions should not be made unless they produce positive net present values, that is, economic rents. Thus, it can be expected that acquiring firms are on rent seeking missions. Any such rent seeking involving substantial firms in concentrated markets should be prohibited.
D. Comparison of the proposed section 7 enforcement standards
with Judge Bork's prescriptions and with the 1968 and 1992
The proposed section 7 enforcement standards are much more restrictive than those proposed by Judge Bork.(250) He says that "mergers up to 60 or 70 percent of the market should be permitted."(251) As a concession to the purpose of section 7, he would, however, allow merged firms to obtain no more than 40% of the market in fragmented industries.(252) He bases his recommendation on the assumption that "the amount of restriction on output seems to decrease greatly from 1-firm to 2-firm markets," and "mergers may very well create substantial new efficiencies."(253) Both of these assumptions are at odds with the evidence. As indicated above, the empirical evidence shows that noncompetitive pricing becomes evident when the four-firm concentration ratio reaches 50%,(254) and Scherer and Ross find that generally a market needs at least ten to twelve evenly matched firms in order for the firms in the market to ignore the actions of rivals.(255)
The 1968 Guidelines have one standard for highly concentrated markets (i.e, markets with a four-firm concentration level of 75%) and a less rigorous standard for less highly concentrated markets (i.e., markets with a four-firm concentration level of less than 75%).
The 1992 Horizontal Guidelines have separate standards for three different levels of market concentration (i.e., HHI below 1000, HHI between 1000 and 1800, and HHI exceeding 1800). The proposed section 7 enforcement standards have just one threshold: The section 7 concentrated market, that in the absence of a clear and convincing showing otherwise, is a postmerger market with a four-firm concentration ratio of at least 50% (or an HHI of at least 1250) or with a dominant firm with a market share of at least 25%. Once these thresholds are crossed, all significant acquisitions (defined as acquisitions resulting in a firm with at least 10% of the market or an increase of at least fifty points in the HHI) are prohibited, subject to the de minimis exception.
The proposed standards depart from the 1968 Guidelines in disallowing many acquisitions by larger firms of smaller firms, unless the de minimis exception applies. Thus, for example, under the 1968 Guidelines, if the concentration ratio is 50%, a 10% firm may acquire a 3% firm. Under the proposed standards, however, such an acquisition would be prohibited, unless the 3% firm satisfied the de minimis exception, which would not generally be the case.
On the other hand, adoption of the proposed section 7 enforcement standards would not lead to a challenge to a merger in cases like Von's(256) or Pabst,(257) because the markets in those cases were not section 7 concentrated markets.
Whereas the 1992 Horizontal Merger Guidelines provide for analysis of mergers for potential anticompetitive effects when the postmerger concentration level exceeds an HHI of 1000 (implying a four-firm concentration ratio of approximately 40%),(258) the proposed section 7 enforcement standards do not provide for the examination of mergers until the postmerger HHI reaches 1250, implying a four-firm concentration ratio of approximately 50%.(259) Thus, the starting point for analysis under the proposed section 7 enforcement standards is more liberal than the 1000 HHI starting point provided in the 1992 Horizontal Merger Guidelines.
On the other hand, once the threshold under the proposed section 7 enforcement standards is exceeded, it is much more likely that a merger will be challenged than is the case under the 1992 Horizontal Guidelines. Also, the decision on whether to challenge will be based solely on a proper analysis of the postmerger concentration level and the increase in concentration. Factors such as entry and efficiencies, which play a large role under the 1992 Horizontal Merger Guidelines, have no role under the proposed section 7 enforcement standards.
The 25% dominant firm provision of the proposed section 7 enforcement standards is much more straightforward than the complicated three-tier 35% dominant-firm rules in the unilateral effects section of the 1992 Horizontal Merger Guidelines.
Judge Bork's views on merger enforcement are not supported by the current state of economic learning. This learning is supportive of a much more aggressive antitrust policy than is reflected in the 1982, 1984, and 1992 Guidelines.
The proposed section 7 enforcement standards set out here is consistent with this learning. Under this standard, if a postmerger market has a four-firm concentration ratio of 50% or more (or an HHI of at least 1250), a merger of firms resulting in a 10% or more market share (or an increase in the HHI of at least fifty points) would be challenged, unless the de minimis exception applied.
The proposed standards would lead to more investigations under the Hart-Scott-Rodino Premerger Notification Act and to more challenges to acquisitions. The question is whether the increase in enforcement actions and the discouragement of other acquisitions would be worth it.
The answer is a clear yes for the following principal reasons. First, both microeconomic theory and empirical evidence clearly show that departures from a competitive market lead to a misallocation of resources. The proposed standards, which are based on the current state of the economic learning, should have the effect of preventing the use of mergers to accomplish such departures.
Second, it is unlikely that the deterrent effect of the proposed section 7 enforcement standards would result in any significant loss in efficiency, because it does not appear that mergers are particularly successful in attaining efficiencies. In the words of Scherer and Ross, the "statistical evidence supporting the hypothesis that profitability and efficiency increase following mergers is at best weak."(260)
Finally, the adoption of the proposed section 7 enforcement standards should have the effect of tilting firms in the direction of internal expansion. On a microeconomic level, as Porter points out, this is good for competition. On a macroeconomic level, building new plant and equipment adds more directly to the gross national product than does the acquisition of old plant and equipment. AUTHOR'S NOTE: I would like to thank Mary Coombs, the Associate Dean for Research at the University of Miami School of Law, for her helpful comments on this article. Also, I would like to thank the following for their helpful comments on earlier drafts: Professor Mark Grady of the UCLA School of Law; Kevin O'Connor, Assistant Attorney General of the State of Wisconsin; and Tyler Baker, former Special Assistant to the Assistant Attorney General of the Antitrust Division of the U.S. Department of Justice. Finally, I would like to thank my research assistant, Michelle Austin, a third-year student at the University of Miami School of Law, for her very diligent and helpful work on this article.
(1) Robert Bork, The Antitrust Paradox: A Policy at War With Itself (1978, reissued 1993).
(2) Leonard Weiss, Concentration and Price (1989). See also, Richard A. Miller, Book Review: Concentration and Price, 36 Antitrust Bull. 709 (1991); David A. Audretsch & John J. Siegfried, Leonard W. Weiss and Industrial Organization, 10 Rev. Ind. Org. 121 (1995); F.M. Scherer, Leonard Weiss' Contribution to Research in Industrial Organization, 10 Rev. Ind. Org. 127 (1995).
(3) Audretsch & Siegfried, supra note 2, at 124.
(4) Clayton Act [sections] 7, 15 U.S.C. [sections] 18 (1988).
(5) Department of Justice Merger Guidelines (May 30, 1968) [hereinafter 1968 Guidelines].
(6) 374 U.S. 341 (1963).
(7) Id. at 363.
(8) Department of Justice 1982 Merger Guidelines (June 14, 1982) [hereinafter 1982 Guidelines]. For a discussion of various aspects of these 1982 Guidelines, see, e.g., Eleanor M. Fox, The New Merger Guidelines - A Blueprint for Microeconomic Analysis, 27 Antitrust Bull. 519 (1982); and Neil B. Cohen & Charles Sullivan, The Herfindahl-Hirschman Index and the New Antitrust Guidelines: Concentrating on Concentration, 62 Tex. L. Rev. 453 (1983).
(9) Department of Justice 1984 Merger Guidelines (June 14, 1984) [hereinafter 1984 Guidelines]. For a comprehensive discussion of the impact of the 1984 Guidelines on horizontal mergers, see ABA Antitrust Section, the 1986 Horizontal Merger Guidelines: Law and Policy (1986).
(10) Department of Justice and Federal Trade Commission Horizontal Merger Guidelines (April 2, 1992) [hereinafter 1992 Horizontal Guidelines]. THE 1984 Guidelines, supra note 9, continue to govern nonhorizontal mergers, that is, conglomerate and vertical. See, Statement Accompanying, Release of Revised Merger Guidelines (April 2, 1992) [hereinafter 1992 DOJ and FTC Statement]. See also, Dissenting Statement of FTC Commissioner Mary L. Azcuenaga on the Issuance of Horizontal Merger Guidelines (April 2, 1992); Charles A. James, Overview of the 1992 Horizontal Merger Guidelines, 61 Antitrust L.J. 447 (1993); and Kevin J. Arquit, Perspectives on the U.S. Government 1992 Horizontal Merger Guidelines, 61 Antitrust L.J. 121 (1992).
In 1987, the National Association of Attorneys General (NAAG) promulgated horizontal merger guidelines (National Association of Attorneys General, Horizontal Merger Guidelines (March 20, 1987)), and on March 30, 1993, partially in response to the issuance of 1992 Horizontal Guidelines, NAAG issued amendments to its guidelines (Horizontal Merger Guidelines of the National Association of Attorneys General (March 30, 1993) [hereinafter NAAG Guidelines]). The enforcement standards in the NAAG Guidelines are similar to, but not as liberal as, the enforcement standards in the 1992 Horizontal Guidelines.
(11) 1992 DOJ and FTC Statement, supra note 10.
(13) William E. Kovacic, The Antitrust Paradox Revisited: Robert Bork and the Transformation of Modern Antitrust Policy, 36 Wayne L. Rev. 1413, 1453 (1990). See also, Thomas G. Krattermaker & Robert Pitofsky, Antitrust Merger Policy and the Reagan Administration, 33 Antitrust Bull. 211, 226-28 (1988); Walter Adams & James W. Brock, Reaganomics and the Transmogrification of Merger Policy, 33 Antitrust Bull. 309 (1988); and Walter Adams & James W. Brock, Revitalizing Structural Antitrust Policy, 39 Antitrust Bull. 235 (1994).
(14) F.M. Scherer & David Ross, Industrial Market Structure and Economic Performance 191 (3d ed. 1990). See also, Robert Pitofsky, Proposals for Revised United States Merger Enforcement in a Global Economy, 81 Georgetown L. Rev. 195, 196 (1992).
(15) Report from Officialdom, 63 Antitrust L.J. 951, 955 (1995).
(16) Id. at 951-72.
(17) See, e.g., Robert Pitofsky, The Political Content of Antitrust, 127 U. Pa. L. Rev. 1051 (1979); and David W. Barnes, Nonefficiency Goals in the Antitrust Law of Mergers, 30 Wm. & Mary L. Rev. 787 (1989). Also, this article does not address the question of market definition.
(18) For another suggested approach for revising the 1992 Horizontal Merger Guidelines, see Lucile S. Keyes, The Horizontal Merger Guidelines of 1992, 10 Rev. Ind. 143 (1995).
(19) See, e.g., supra note 13, at 1453.
(21) Bork, supra note 1, at 218.
(22) Id. at 219.
(27) Id. at New Introduction to reissued volume, p. xi.
(29) Id. at 122.
(31) Id. at 110-11.
(32) Standard Oil Company of New Jersey v. United States, 31 S. Ct. 502, 512 (1910). The other two reasons for opposing monopoly were: "(2) the power which it engendered of enabling a limitation on production; and (3) the danger of deterioration in quality of the monopolized article which it was deemed was the inevitable resultant of the monopolistic control over its production and sale." Id.
(33) Brown Shoe Co. v. United States, 370 U.S. 294, 316 (1962).
(34) Bork, supra note 1, at 111.
(35) Arnold C. Harberger, Taxation and Welfare 17 (1974).
(36) Arnold C. Harberger, Three Basic Postulates for Applied Welfare Economics, 9 J. Econ. Lit. 785 (1971) reprinted in Harberger, Supra note 35, at 5.
(37) Harberger, supra note 35, at 5-6.
(38) Id. at 6. (39) Richard A. Musgrave & Peggy B. Musgrave, Public Finance in Theory and Practice 84-85 (4th ed. 1984). (40) Bork, supra note 1, at 219. (41) Id. at 124. (42) Id. at 220. (43) Oliver Williamson, Economies as an Antitrust Defense: The Welfare Tradeoffs, 58 Am. Econ. Rev. 18 (1968). (44) Bork, supra note 1, at 125. (45) Id. at 126. (46) Id. at 128. (47) Id. at 219. (48) Michael E. De Prano & Jeffrey B. Nugent, Economies as an Antitrust Defense: Comment, 59 Am. Econ. Rev. 947 (1969).. (49) Bork, Supra note 1, at 220 (quoting De Prano and Nugent, supra note 48). (50) Id. at 220. (51) 374 U.S. 321 (1963). (52) Id. at 363. (53) 386 U.S. 568 (1967). (54) Id. at 580. (55) Bork, supra note 1, at 220. (56) Id. at 220-21. (57) Id. at 221. (58) Id. (59) Id. (60) Id. (61) Id. at 178. (62) Id. at 180. (63) Id. (64) Id. at 180-81. (65) Harold Demsetz, Two Systems of Belief About Monopoly, in Industrial Concentration: The New Learning 164 (Harvey Goldschmid, H. Michael Mann & J. Fred Weston eds., 1974). (66) Bork, supra note 1, at 181. (67) Richard Schmalensee, Inter-Industry Studies of Structure and Performance, in 2 Handbook of Industrial Organization 966-67 (Richard Schmalensee & Robert D. Willig eds., 1989). (68) Weiss, supra note 2, at 7. (69) Id. at 8. (70) Bork, supra note 1, at 181. (71) Id. (72) Id. (73) Id. at 181. (74) Id. at 191. (75) Weiss, supra note 2, at 4-5 and 268. (76) Bork, supra note 1, at 221. (77) Id. (78) Id. (79) Id. (80) Id. (81) Id. at 221-22. (82) Id. at 222. (83) Id. (84) Kovacic, supra note 13, at 1453. (85) 898 F.2d 1278 (7th Cir. 1990). (86) Id. at 1281. (87) Id. at 1285-86. (88) Id. at 1286. (89) Harold Demsetz et al., Policy Adequacy of the Empirical Concentration Price Studies: The Posner Lament, 1991 Antitrust Law & Econ. Rev. 19 (1990) [hereinafter Posner Lament]. (90) Id. at 26. (91) Id. (92) Lawrence J. White et al., Market Power and the Critical Concentration Level: Need for a 1-firm Threshold Number?, 1991 Antitrust L. & Econ. Rev. 15 (1991). (93) Posner Lament, supra note 89, at 23. (94) Weiss, Supra note 2, at 27; and Scherer & Ross, supra note 14, at 422-23. (95) Posner Lament, supra note 89, at 30. See, supra note 67. (96) Gregory J. Werden, A Review of the Empirical and Experimental Evidence on the Relationship Between Market Structure and Performance (Economic Analysis Group Discussion Paper 91-3, May 8, 1991. (97) Posner Lament, Supra note 89, at 24-25. (98) White et al., supra note 92, at 17. (99) Id. at 21. (100) Id. (101) Id. at 22. (102) Leonard W. Weiss, The Concentration-Profits Relationship and Antitrust in Industrial Concentration: The New Learning (Harvey Goldschmid, H. Michael Mann & J. Fred Weston eds., 1974). (103) Id., at 186-87. (104) Id. at 192. (105) Id. at 202. (106) Demsetz, supra note 65. (107) Id. at 166. (108) Id. at 174. (109) Id. at 176-77. (110) Id. at 177. (111) Id. (112) Id. at 177-78. (113) Weiss, supra note 102, at 225. (114) Id. (115) Id. (116) Id. (117) Weiss, supra note 2, at 8. (118) Id. at 8 & 9. (119) Posner Lament, supra note 89, at 31. (120) Weiss, supra note 2. (121) Posner Lament, supra note 89, at 31. (122) Id. at 31-32. (123) Id. at 32. (124) Id. at 28. (125) Schmalensee, supra note 67. (126) Posner Lament, supra note 89, at 35. (127) WEISS, supra note 2. (128) Posner Lament, supra note 89, at 37. (129) Werden, supra note 96. (130) Posner Lament, supra note 89, at 38. (131) Id. at 40. (132) Weiss, supra note 2. (133) Scherer & Ross, supra note 14. (134) Posner Lament, supra note 89, at 41. (135) Id. at 47. (136) Id. at 49. (137) Id. at 50. (138) Id. at 52. (139) Id. at 43. (140) Id. (141) Id. at 27. (142) Id. 143 Id. at 32. 144 Id. at 40. 145 Id. at 42. 146 Id. at 45. 147 Id. at 58. 148 Id. at 54. 149 Id. at 80. 150 Id. 151 Id. at 61. 152 Id. at 36. 153 Id. at 33. 154 Id. at 42. 155 Id. at 51. 156 White et al., supra note 92. 157 Id. at 30. 158 Id. 159 Id. at 25. 160 John E. Kwoka, Jr., Regulating and Diversity in Firm Size Distributions in U.S. Industries, 34 J. Econ. & Bus. 391 (1982). 161 George J. Stigler, Mergers and Preventive Antitrust Policy, 104 U. Pa. L. Rev. 176, 181-82 (1955). 162 1992 Horizontal Merger Guidelines, supra note 10, at [section] 2.2. 163 Id. 164 Weiss, supra note 2. 165 Schmalensee, supra note 67. 166 Scherer & Ross, supra note 14. 161 Werden, supra note 96. 168 Weiss, supra note 2, at 268. 169 Id. at 270. 170 Id. at 276. 171 Id. at 281. 172 Id. 173 Id. 174 Id. at 32. 175 Id. at 277. 176 Scherer & Ross, supra note 14, at 277. 177 Id. 178 Id. 179 Id. 180 See, White et al., supra note 92, at 17. 181 Schmalensee, supra note 67, at 984. 182 Id. at 976. 183 Id. at 988. 184 Werden, Supra note 96, at 6. 185 Id. at 16. 186 Id. at 17 (citing Leonard W. Weiss, Concentration and Price-A Progress Report, in Issues After a Century of Competition Policy, 318 (Robert L. Wills, Julie A. Caswell & John D. Albertson, eds., 1987). 187 Bork, supra note 1, at 221. 188 For reasons stated above, this article does not deal with the market definition issue. 189 Brown Shoe Co. v. United States, 370 U.S. 294, 323 (1962). 190 Id. at 323 n.39. 191 Id. at 317. 192 Bork, supra note 1, at 206. 193 The Merger Modification Act of 1986. 194 Fundamental competitiveness bill of 1992 (H.R. 5229, May 21, 1992). (195) See, section IV.A., supra. (196) 1968 Guidelines, supra note 5. (197) Id. [sections] I.10. (198) 1992 Horizontal Guidelines, supra note 10 [sections] 1.52 1. (199) Scherer & Ross, supra note 14, at 174. (200) Dennis C. Mueller, U.S. Merger Policy and the 1982 Merger Guidelines, 8 Rev. Ind. Org. 151, 153 (1993). (201) Scherer & Ross, supra note 14, at 165. (202) NAAG Guidelines, supra note 10, at [sections] 5.3. (203) Pitofsky, supra note 14, at 221. (204) Id. at 207 (DOJ policy); and Steve Newborn, Interview, 6 Antitrust 17, 22 (1992) (FTC policy). (205) 1968 Guidelines, supra note 5, [sections]] I. 10. (206) NAAG Guidelines, supra note 10, [sections] 5.3. (207) 1984 Guidelines, supra note 9, [sections] 3.5. (208) 1992 Horizontal Guidelines, supra note 10, [sections] 4. (209) Id. (210) NAAG Guidelines, supra note 10, [sections] 5.3. (211) Bork, supra note 1, at 127-29, 219. See also, Pitofsky, supra note 14, at 210-11. (212) Michael E. Porter, The Competitive Advantage of Nations 662 (1990). (213) Joseph F. Brodley, Antitrust and Competitive Advantage in World Markets, Antitrust, Fall/Winter 1990, at 38, 39. (214) 1992 Horizontal Merger Guidelines, supra note 10, at [sections] 3. (215) See Dennis W. Carlton & Jeffrey M. Perloff, Modern Industrial Organization 195-96 (1990). (216) Porter, supra note 212. (217) Interview with Michael Porter, Antitrust, Spring 1991, at 5, 8. (218) R. Brealey & S. Myers, Principles of Corporate Finance 13 (4th ed. 1991). (219) Id. (220) Id. (221) Id. at 162. (222) Id. at 248. (223) Id. (224) Id. (225) Michael E. Porter, From Competitive Advantage to Corporate Strategy, Har. Bus. Rev., May-June 1987, at 43, 46. (226) See infra, section V.A. (227) See infra, section V.B. (228) Schmalensee, supra note 67, at 966. (229) Scherer & Ross, supra note 14, at 8. (230) Weiss, supra note 2, at 268. (231) Id. at 276. (232) Schmalensee, supra note 67, at 988. (233) See section III.D., supra. (234) Kwoka, supra note 160. (235) Stigler, supra note 161 at 181-82. (236) A four-firm concentration ratio of 32% would indicate a market with an HHI of no more than 768 (i.e., twelve equal 8% firms, 8 x 8 x 12 = 768). (237) A market with eight equal size firms would have an HHI of 1250 (i.e., 12.5% x 12.5% x 8 = 1250). (238) See, Scherer & Ross, supra note 14, at 8. (239) Weiss, supra note 2, at 283. (240) Stigler, supra note 161, at 181-82. (241) Weiss, supra note 2, at 36. (242) United States v. General Dynamics Corp., 415 U.S. 486 (1974). (243) United States v. ALCOA, Inc., 84 S.Ct. 1283 (1964). (244) Weiss, supra note 2, at 27. (245) Id. (246) Porter, supra note 212. (247) Scherer & Ross, supra note 14, at 162-67. See also, Dennis C. Mueller, supra note 200, at 158-60. (248) George Stigler, The Economies of Scale, 1 J.L. & Econ. 1 (1958). (249) Stigler, supra note 161, at 181-82. (250) Bork, supra note 1, at 221-22. (251) Id. at 221. (252) Id. at 222. (253) Id. (254) Weiss, supra note 2, at 27. (255) Scherer & Ross, supra note 14, at 277. (256) United States v. Von's Grocery Co., 384 U.S. 270 (1966). (257) United States v. Pabst Brewing Co., 384 U.S. 546 (1966). (258) Assuming a market with ten equal firms, the four-firm concentration ratio is 40% and the HHI is 1000 (i.e., 10 x 10 x 10). (259) Assuming a market with eight equal size firms (i.e., 12.5% each), the four-firm concentration ratio is 50% (i.e., 12.5% x 4) and the HHI is 1250 (i.e., 12.5% x 12.5% x 8). (260) Scherer & Ross, supra note 14, at 174.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Robert H. Bork|
|Author:||Thompson, Samuel C., Jr.|
|Date:||Mar 22, 1996|
|Previous Article:||Competitive effects of cable copyright law.|
|Next Article:||Hospital mergers and antitrust policy: arguments against a modification of current antitrust law.|