The Merger Guidelines of the
Oliver E. Williamson*
These remarks have their origin in a conversation I had with Edwin Zimmerman about two years ago. Ed observed, and I agreed, that Donald Turner's role in bringing economic analysis to bear on antitrust enforcement was under-valued, even ignored. It was Ed's view and it is mine that an accurate record of the progressive introduction of economics into antitrust should make prominent provision for the "early years" from 1965-1968 when Turner was the Assistant Attorney General for Antitrust.
The object of my remarks is not to diminish the importance of his successors, most importantly, William Baxter. Rather it is to make the case that initiatives taken during the Turner administration not only had immediate consequences for changing antitrust enforcement as we had come to know it, but these also provided the foundation for successor administrations to build on.
I begin with some background on prevailing attitudes within the antitrust enforcement agencies and the courts toward the uses of economic reasoning in the early 1960s. I then describe three important changes that I associate with Turner: a change in the tenor of the antitrust enforcement dialogue that was brought about by Turner; new hires and staffing decisions made by Turner and Zimmerman that had the purpose and effect of bringing economics more prominently to bear on strategic decision making within the division; and the issuance of the 1968 Merger Guidelines. Two of the cases in which I was involved when I served as Special Economic Assistant are briefly discussed. Concluding remarks follow.
1. Pre-Turner antitrust enforcement
(a) industrial organization
Legislation from the New Deal Era(1) and hearings held under the auspices of the Temporary National Economic Committee set a populist tone for antitrust enforcement in the immediate post-World War II era. Arguably the most important development in the field of industrial organization in the 1950s was Joe Bain's book Barriers to New Competition (1956), which was then followed by his textbook on Industrial Organization (1959) and the book by Carl Kaysen and Donald Turner on Antitrust Policy (1959). This body of work ushered in the structure, conduct, performance (SCP) approach to industrial organization, with special emphasis on barriers to entry. Franco Modigliani's survey of "New Developments on the Oligopoly Front" (1958) gave added weight to entry barrier arguments. Entry barriers took many forms, some real and some imaginary. Albeit unintentionally, entry barrier arguments were quickly adopted by, and gave impetus, even legitimacy, to, populist antitrust.
To be sure, there were other views. Aaron Director and his students and colleagues at Chicago were grave skeptics of "the new SCP learning." Be that as it may, both Harvard and Chicago were price theoretic constructions (Coase, 1972, p.62). Both, for example, described the firm as a production function (a technological construction) for anti-trust enforcement purposes. Also, except as Chicago gave prominence to the allocative efficiency benefits that purportedly resulted from vertical market restrictions, examining contract and organization through the lens of price theory often ended with a monopoly explanation -- as Ronald Coase pithily observed (1972, p .67):
One important result of this preoccupation with the monopoly problem is that if an economist finds something--a business practice of one sort or other--that he does not understand, he looks for a monopoly explanation. And as in this field we are very ignorant, the number of ununderstandable practices tends to be very large, and the reliance on a monopoly explanation, frequent.
The possibility that nonstandard practices sometimes had economizing purpose and effect was ignored or, worse, efficiency was regarded as the source of an unfair competitive advantage. Thus, the Federal Trade Commission held in Foremost Dairies that the necessary proof of violation of Section 7 "consists of types of evidence showing that the acquiring firm possesses significant power in some markets or that its over-all organization gives it a decisive advantage in efficiency over its smaller rivals."(3) Although Turner, among others, was quick to label that as bad law and bad economics (1965, p. 1324), in that it protects competitors rather than promoting the welfare benefits of competition, the Commission carried its reasoning forward in Procter & Gamble and linked it with barriers to entry in the following way:(4)
In stressing as we have the importance of advantages of scale as a factor heightening the barriers to new entry into the liquid bleach industry, we reject as specious in law and unfounded in fact, the argument that the Commission ought not, for the sake of protecting the "inefficient" small firms in the industry, proscribe a merger so productive of "efficiencies." The short answer to this argument is that, in a proceeding under Section 7, economic efficiency, or any other social benefit resulting from a merger is pertinent only insofar as it may tend to promote or retard the vigor of competition.
The emphasis on entry barriers and the low regard accorded to economies also appear in the Supreme Court's opinion. Thus the Court observed that Procter's acquisition of Clorox may(5)
... have the tendency of raising the barriers to new entry. The major competitive weapon in the successful marketing of bleach is advertising. Clorox was limited in this area by its relatively small budget and its inability to obtain substantial discounts. By contrast, Procter's budget was much larger; and although it would not devote its entire budget to advertising Clorox, it could divert a large portion to meet the short-term threat of a new entrant. Proctor would be able to use its volume discounts to advantage in advertising Clorox. Thus, a new entrant would be much more reluctant to face the giant Proctor than it would have to face the smaller Clorox.
The low opinion and perverse regard for economies went so far that beleaguered respondents disclaimed efficiency gains.(6)
This upside-down assessment of economies was bound to change, and it did--but not before Justice Stewart, in a dissenting opinion in 1966, remarked that the "sole consistency that I can find is that in [merger] litigation under section 7, the Government always wins."(7) Such was the state of antitrust enforcement when Turner became the Assistant Attorney General for Antitrust.
2. The Turner regime
Donald Turner was the first Ph.D. economist to head the Antitrust Division. After receiving his Ph.D. from Harvard in 1947, he went directly to Yale where he received his law degree in 1950. Turner then clerked at the Supreme Court and was an associate at Cox, Langford, Stoddard & Cutler before joining the Harvard Law School Faculty in 1954, where he specialized in antitrust.
As described in the Forward (and elaborated in the Preface by Edward S. Mason) of the famous Kaysen and Turner book on Antitrust Policy (1959), Harvard was in the vanguard for reshaping the field of industrial organization in the 1950s. Moreover, the structure-conduct-performance approach to industrial organization took shape with applications to antitrust enforcement foremost in mind. Kaysen and Turner, for example, took the "limitation of market power" (1959, p. 82), rather than consumer benefits, to be the chief public policy purpose of antitrust.
Such an orientation had the unintended effect of stoking the fires of activist antitrust enforcement. With the help of concepts such as "potential competition," it was easy to find problematic structures and practices lurking under nearly every rock. One of Turner's concerns was that good but delimited arguments were being taken to extremes. Part of his mission when he was appointed Assistant Attorney General for Antitrust in 1965 was to restore perspectives.
To my knowledge, Turner never announced the objective of putting antitrust enforcement on sounder economic foundations. This objective can nevertheless be inferred from a series of steps taken during the Turner administration. Modest though each of these steps may appear, they set in motion changes that had lasting effects. The Turner-Zimmerman era was the beginning of what would turn out to be a vast reorientation of antitrust in which more and better economic reasoning made progressive headway--a process that continues even to this day.
That the initial changes were subtle and modest was partly because this was the only practical way to proceed. Forcing a change was not only unrealistic, but could be counterproductive. Not only would the career staff be there long after Turner had left, but every administration depends on the career staff to initiate new cases and manage ongoing cases. Lest this be done in a perfunctory manner, change had to be implemented in a way that the career legal staff could buy into and ascribe merit. Also, activist antitrust enforcement had its supporters in both Congress, where the Celler-Kefauver Amendment to the Clayton Act had been passed in 1950, and the Court, which celebrated the populist rhetoric from the legislative history of that Amendment. Additionally, the Economics Section in the Antitrust Division had come to understand its function to be that of litigation support. Changing that would likewise take time.
(a) management and proselytizing
So Turner changed direction of the ship of state gradually. Problematic cases that were already in progress were rarely dismissed but were reshaped, thereby to replace incorrect or vague arguments with more careful and nuanced arguments. New cases that could be brought and won under current law, as inferred by recent decisions, were not approved if they relied on mistaken or contrived economic thinking.
Turner's reliance on economic reasoning was evident, moreover, in his weekly conference with section heads, where the discussion of cases-in-progress was extended to include the economic merits (or lack thereof). The economic merits were also featured in meetings with outside counsel. Turner also made a point of updating the antitrust bar on his antitrust thinking in his frequent speeches. And he was actively involved in the preparation of briefs for the government that were headed for the Supreme Court.
(b) supporting staff
Turner could not do it all. Key figures in his administration who played important roles in transforming antitrust enforcement included Edwin Zimmerman (whose title was that of First Assistant and who became Turner's successor), Lionel Kestenbaum (who headed up the Evaluation Section), Robert Hammond (as head of Policy and Planning), and Stephen Breyer (who was Turner's Special Assistant). Also, Richard Posner was in the Solicitor General's office, where he specialized in briefing and arguing antitrust cases before the Supreme Court. And Turner created a new position: Special Economic Assistant, which reported directly to him. He also staffed the Evaluation Section with an exceptional group of recent law school graduates, many being his students and Zimmerman's students from Harvard and Stanford.
As previously remarked, the Economics Section in the Antitrust Division was being used mainly for litigation support. The creation of a Special Economic Assistant who reported directly to Turner signaled that economic analysis would now be featured more prominently in the decision to bring cases and in the manner in which the cases were argued. William Comanor was the first Special Economic Assistant. I was his successor.
Lou Marcus, who headed the Economics Section, retired in 1973, at which time Thomas Kauper was the Assistant Attorney General for Antitrust and George Hay was his Special Economic Assistant. Kauper and Hay took the process to its "natural" completion. The position of Special Economics Assistant was abolished, the Economics Section was transformed into the Economic Policy Office (which put it on a parity with the Office of Operations, at least on the organizational chart), George Hay was appointed Director of the Economic Policy Office, and the active recruiting of Ph.D. economists ensued (four were recruited that first year.) These changes served to ratify the upgrading of economic analysis in the enforcement of antitrust that had been begun by Turner.
Taken together, the creation of the position of Special Economic Assistant and the decision to staff the evaluation section with young lawyers who bought into the idea that economic reasoning should be featured more prominently in antitrust enforcement were important "organizational innovations." For those who were a part of this transition, these were exciting times.
(c) Merger Guidelines
We are assembled here today to celebrate the 20th anniversary of the 1982 Merger Guidelines and the leadership of William Baxter, who inspired and oversaw the revision. The 1982 Merger Guidelines (and the 1984 Guidelines that followed) signaled a new and important era in which economic analysis and economic criteria have played more important roles in antitrust enforcement. Public policy is much the better for it.
But note that I refer to a revision in the Merger Guidelines. The idea for merger guidelines go back at least to Kaysen and Turner's early sketch of such guidelines in their 1959 book (pp.132-136). The first Department of Justice Merger Guidelines were issued on Turner's last day (May 30, 1968) as Assistant Attorney General.
The 1968 Guidelines are more stringent than both the earlier Kaysen and Turner "bench marks" and the later 1982 Guidelines. In large measure, the 1968 Guidelines adopted market share limits that could be inferred from recent merger decisions by the courts.
The first draft of the 1968 Guidelines was written by Turner and was circulated and discussed in the Spring of 1967. The general response among the career staff was one of caution and even disfavor.
One of the consequences of issuing Guidelines is that firms considering merger would face less antitrust enforcement uncertainty. Inasmuch, however, as such uncertainty served mainly to deter mergers, many of the career staff viewed uncertainty as an "advantage." Of greater concern is that the issuance of Guidelines would serve to introduce a floor, below which the burden of bringing a case would have to be borne by the Division. The previous practice of progressively ratcheting down admissible market shares would be brought to a halt. That the 1968 Guidelines were stringent can thus be thought of as a compromise. To propose more permissive market shares would, in effect, concede error--by the antitrust enforcement agencies and the courts--in earlier cases. Not only did Turner and Zimmerman need the continuing cooperation of the staff, but they did not need a squabble on Capitol Hill.
But the 1968 Guidelines also reflect the intellectual heritage that came out of the structure-conduct-performance school of which Turner was a part. Thus the 1968 Guidelines are preoccupied with real or imaginary entry barriers, to which detrimental effects purportedly accrue by reason of (1) product differentiation and promotional expenses, (2) price or supply "squeezes," and (3) reciprocal buying from which competitors are excluded. More generally, the theory of the firm on which the 1968 Guidelines are based is that of the firm as production function--the requisite background for which was relative prices and technology, or "economics and industrial engineering" (Kaysen and Turner, 1959, p. 119), but not organization theory. Albeit broadly in accord with the prevailing price theory, viewing the firm as a production function is also self-limiting.
For some purposes, of which antitrust is one, firms are more than a black box for transforming inputs into outputs according to the laws of technology. Firms have internal structure, which structure is not adventitious (Arrow, 1999, p.vii). Viewing the firm as a governance structure (which is an organizational rather than a technological construction), to be compared with simple and complex modes of contracting, has thus turned out to be instructive. For one thing, the opening of a new window on firm and market organization the lens of comparative contractual analysis invites a more detailed and nuanced examination of the activities to be organized and of the properties of alternative modes of governance. Second, and related, the presumption that nonstandard and unfamiliar contracting practices and organizational structures have monopoly purpose and effect is revealed to be simplistic. If organization (to include the design of complex contracts) has efficiency consequences, then antitrust thinking needs to be expanded to take that into account. The narrowness with which economies are viewed in the 1968 Guidelines is partly a byproduct of the prevailing technological (production function) conception of the firm.
3. Two Cases
I learned a lot about antitrust enforcement during the year that I spent as Special Economic Assistant to Turner. One of the things that I learned was to have deep respect for the leadership, the career staff, and those who were passing through the Antitrust Division on the way to careers in law firms and academia (and, sometimes, economic consulting). This was an extraordinary group of highly motivated, hard working, and talented people.
I also learned a lot about how the Division functioned and witnessed both good economic arguments and some economic misconceptions in action. Of the approximately 40 different "projects" that came across my desk, about 10 involved interagency matters (many with Defense and NASA) and 30 were cases in various stages of development.
I like to believe, moreover, that I occasionally brought some value added to the enterprise even if, sometimes, it took a while for the value added to register. The two cases that I have selected for discussion here are chosen not because they are representative, but because they illustrate the need for new thinking about the uses and misuses of economics in antitrust. The first of these is the Schwinn case,(8) which illustrates the benefits of going "outside the [neoclassical black] box." The second involved the proposed merger of two daily newspapers in a small community, where the benefits of economic reasoning accrue from working within an orthodox applied welfare economics setup.
Turner asked me to take a look at Schwinn soon after I arrived in September 1966, and I received a draft of the brief from Richard Posner shortly thereafter. Both Turner and Posner regarded Schwinn as an important opportunity to bring the "then prevailing thinking of the economics profession [to bear] on restricted distribution" (Posner, 1977, p.3).
Alas, what Turner and Posner took to be the then prevailing thinking of the economics profession was deeply confused. As I have discussed elsewhere (Williamson 1979; 1985, pp. 183-189), the prevailing thinking was self-limiting in three respects: (1) there was little appreciation for the possibility that product differentiation (as opposed to homogeneous product market exchange) might be the source of economic benefits, (2) there was even less appreciation for the possibility that that the integrity of a distribution system could be compromised by subgoal pursuit among the parts (in this case, the individual franchisees), and (3) there was a preference for internal organization (hierarchy) over market organization (interfirm contract) if vertical restrictions, for whatever reason, were to be applied.
Such misconceptions were the product of combining firm as production function thinking with the structure-conduct-performance school of which Turner was a part and to which he had contributed. Thus Turner viewed "customer and territorial relations not hospitably, in the common law tradition, but inhospitably in the tradition of antitrust."(9) Unsurprisingly, the brief argued that restrictions in Schwinn were per se unlawful, and that is what the Court decided.
To be sure, and to their credit, both Turner and Posner revised their views ten years later, when GTE-Sylvania came before the court.(10) As of 1966, however, my efforts to place a more favorable construction on Schwinn's restrictions and to reshape Schwinn got precisely nowhere. The difference is that whereas I viewed contract and organization from a nonneoclassical perspective in which organization (as well as technology) mattered, which was a byproduct of my (unorthodox) training at Carnegie, Turner and Posner were confident that SCP thinking on which they relied for authority was entirely sufficient.
That has changed and many of us like to believe that those years are behind us forever. Even if they are, however, new issues of complex contracting and economic organization will arise for which new apparatus will sometimes be needed. Antitrust economics cannot settle for the quiet life but needs to remain in the vanguard.
The second case was the proposed merger of two daily newspapers in a small community. Given the clear prospect of increased market power, this was an easy case: disallow the merger.
But there was a further complication: the merger might also yield economies. The question was how should a merger that simultaneously yielded economies and market power be evaluated? I discovered to my surprise that the allocative efficiency consequences of such a merger had never been worked out. The prevailing intuition, however, was that any increase in market power, however small, would trump the benefits of any cost saving. Was that intuition correct?
What I have referred to as the "naïve model" (Williamson, 1968) revealed that, over a wide range of demand elasticities, the allocative efficiency benefits of a small reduction in average costs could be offset only by a large increase in prices--easily on the order of 10:1. That result surprised me and it also surprised Turner.
To be sure, such a simple model should not, and did not, carry the day--in that an economies defense did not immediately become available to respondents in merger cases. The demonstration, however, that the earlier intuition--that any shred of monopoly power, however small, was decisive on reaching informed decisions on the merits--had it exactly wrong, had an immediate and salutary effect, in that it served to delegitimize the animus with which economies had previously been held. Thus, although the 1968 Merger Guidelines scarcely invite an economics defense, the Guidelines nevertheless viewed economies favorably (which is also the way that economies are treated throughout by Kaysen and Turner (1959)).
The 1982 Merger Guidelines can be viewed as being more deferential to economies, in that market share criteria for challenging a merger are increased. The 1982 guidelines are nevertheless cautious about admitting an efficiency defense except in close cases (Antitrust Bulletin, 1982, p.163, n.53), mainly because the magnitudes would be "extremely difficult to determine" (p.163). The 1984 Guidelines, by contrast, introduce the discussion of efficiencies with the following statement: "The primary benefit of mergers to the economy is their efficiency-enhancing potential, which can increase the competitiveness of firms and result in lower prices to consumers" (Antitrust Bulletin, 1984, p.761). Thus, although the 1984 Guidelines "are designed to proscribe mergers that present a significant danger to competition," the Department will nevertheless consider "clear and convincing evidence [of] efficiencies in deciding whether to challenge [a] merger" (p.163). As discussed by William Kolasky and Andrew Dick (2002), the merits of economies, both in general and in antitrust enforcement, eventually prevailed.
It is appropriate that we are assembled today to celebrate the 20th Anniversary of the 1982 Merger Guidelines that were issued by the Department of Justice when William Baxter was the Assistant Attorney General for Antitrust. Not only the 1982 Guidelines but the "Baxter years" more generally are regarded as a proud chapter in the history of antitrust enforcement. I join in this celebration with alacrity.
But I also want to remind you that there were some important antecedent events. Not only did the 1968 Merger Guidelines break new ground upon which the 1982 Guidelines could build, but the Turner administration deserves credit for bringing economic reasoning to bear on antitrust in a much more forceful and systematic way than had been done previously. If my interpretation is correct, the vast upgrading of the role of economics within the Division--from litigation support in the pre-Turner years to analysis of the merits of proposed antitrust actions and active participation in shaping antitrust arguments today--is partly attributable to organizational changes that were made by Turner. Specifically, the staff that now reports to the Deputy Assistant Attorney General--Economics is very much a consequence of two organizational initiatives taken by the Turner administration: creation of the position of Economic Special Assistant and staffing of the Evaluation Section with young lawyers that shared the Turner-Zimmerman vision of what antitrust enforcement was all about. "Tall oaks from little acorns grow." We are here to admire the oak, but I suggest that the seeds planted during the Turner administration warrant more than a passing nod.
Arrow, Kenneth.1999. "Forward" in Glenn Carroll and David Teece, eds. Firms, Markets, and Hierarchies. New York: Oxford University Press, pp.vii-viii.
Bain, Joe. 1956. Barriers to New Competition. Cambridge, MA: Harvard University Press.
Bain, Joe. 1958. Industrial Organization. NewYork: John Wiley & Sons.
Bork, Robert J. 1978. The Antitrust Paradox: A Policy at War With Itself. New York: Basic Books.
Coase, Ronald H. 1972. "Industrial Organization: A Proposal for Research," in Victor Fuchs, ed. Policy issues and Research Issues in Industrial Organization. New York: National Bureau of Economic Research. pp. 59-73.
Fisher, Alan, and Robert Lande. 1983. "Efficiency Considerations in Merger Enforcement," California Law Review, 71 (December): 1580-1696.
Kaysen, Carl, and D. F. Turner. 1959. Antitrust Policy. Cambridge: Harvard University Press.
Kolasky, William J., and Dick, Andrew R. 2002. "The Merger Guidelines and the Integration of Efficiencies into Antitrust Review of Horizontal Mergers." Available from the U.S. Department of Justice, Antitrust Division.
Modigliani, Franco. 1958. "New Developments on the Oligopoly Front," Journal of Political Economy, 66 (June): 215-32.
Posner, Richard A. 1977. "The Rule of Reason and the Economic Approach : Reflections on the Sylvania Decision," University of Chicago Law Review, 45 (1): 1-25.
Turner, Donald F. 1965. "Conglomerate Mergers and Section7 of the Clayton Act," Harvard Law Review, 78 (May): 1313-95.
White, Lawerence J. 2002. "Are Giant Companies Taking over the US Economy?" Milken Institute Review, 4 (2): 60-62.
Williamson, Oliver E. 1968. "Economies as an Antitrust Defense: The Welfare Tradeoffs," American Economic Review, 58 (March): 18-35.
Williamson, Oliver E. 1979. "Assessing Vertical Market Restrictions," University of Pennsylvania Law Review, 127 (April): 953-93.
Williamson, Oliver E. 1985. The Economic Institutions of Capitalism. New York: The Free Press.
* The Author was Special Economic Assistant to the Head of the Antitrust Division of the U.S. Department of Justice in 1966-67 and is currently Edgar F. Kaiser Professor of Business, of Economics, and of Law at the University of California, Berkeley. The helpful advice of Edwin Zimmerman is gratefully acknowledged.
1. See Lawrence J. White, "Are Giant Companies Taking over the US Economy?" Milken Institute Review, 2002, 4(2): 60-62.
2. This subsection is based on Williamson (1985, pp. 366-367).
3. In re Foremost Dairies, Inc., 60 F.T.C. 944 1084 (1962), emphasis added.
4. Quoted from Robert Bork (1978, p. 254).
5. Federal Trade Commission v. Procter and Gamble Co., 386 U.S. 568, 574 (1967).
6. Thus Procter and Gamble insisted that its acquisition of Clorox was unobjectionable because the government was unable to definitively establish that any efficiencies would result:
[The government is unable to prove] any advantages in the procurement or price of raw materials or in the acquisition or use of needed manufacturing facilities or in the purchase of bottles or in freight costs. [T]here is no proof of any savings in any aspect of manufacturing. There is no proof that any additional manufacturing facilities would be usable for the production of Clorox. There is no proof that any combination of manufacturing facilities would effect any savings, even if such combination were feasible.
This disclaimer of efficiencies appeared in Procter and Gamble's brief as respondent in the Clorox litigation. See Alan Fisher and Robert Lande (1983, p. 1582, n.5).
7. United_States_v. Von's Grocery_Co., 384 U.S. 270, 301 (1966) (Stewart, J., dissenting).
8. United States v. Arnold, Schwinn & Co., 388 U.S. 365 (1967).
9. The quotation is attributed to Turner by Stanley Robinson, 1968, New York State Bar Association Antitrust Symposium, p. 29.
10. Continental T.V. Inc. et al. v. GTE Sylvania Inc., 433 U.S. 36 (1977). Bother Turner and Posner participated in amicus briefs that helped defeat the per se rule in Schwinn.