57252

The largest auto company in the world and the third largest auto company in the world are proposing to join together to limit competition in a critical segment of the U.S. market. Competition in the U.S. has already been choked off by the Japanese Voluntary Restraint

Agreement. Since 1981, when the VRA took hold, the average retail selling price

for all new cars sold in the U.S. has risen by 18.8 percent or $1741 more than the consumer price index.

Consummation of this joint venture will reinforce this upward pressure on new car prices.

Battalions of neo-classical economists dancing on the head of a pin cannot obscure the threat that this marriage of competitors poses to the American consumer, nor the fact that this joint venture is a plain and unambiguous violations of the antitrust laws. The Commission's settlement, requiring Toyota and GM to abide by the precise terms of their illegal agreement, hardly qualifies as antitrust enforcement.

The suggestion that GM needs this joint venture to learn the secrets of Japanese auto production is not credible. Ford, Chrysler, GM, and the UAW are already revolutionizing domestic auto production with learned Japanese knowhow and that process will continue unabated. Even if GM's plea for Japanese help is genuine, linking up with the dominant Japanese importer is one of the most anticompetitive ways to get it.

The Commission's settlement provisions leave the essential structure of the venture intact and allow the parties to carry it out in the way they had contemplated from the beginning The settlement does not prevent coordination of output and pricing decisions nor does it (even assuming strict compliance) forbid anticompetitive exchanges of information. Accepting it as a resolution of these fundamental antitrust flaws demonstrates once again

I have been advised by the General Counsel's Office that my prepared statement of December 22 1983 contained references to nonpublic material While I disagree with that conclusion. I have decided to delete those portions of the statement which have been identified as containing nonpublic material in order not to delay publication and dissemination of the current agreement, pending further resolution of the issue.

the Reagan administration's antipathy to antitrust law enforcement

In addition to the serious antitrust violation this joint venture appears to represent, it signals an ominous trend of American manufacturers conceding that they cannot compete on their own in small cars. Our major government policies toward the automobile industry-the Voluntary Restraint Agreement, the "CAFE" requirements for an overall mpg average for domestic manufacturers, and even this

administration's misguided "regulatory relief' for the auto industry in lowering safety and environmental standardshave all been intended at least in part to

help American manufacturers sell cars which can compete on an even footing with foreign manufacturers, particularly the Japanese. GM has now found a way to sell small cars, not by manufacturing them itself, but by assembling cars which are made primarily in Japan. The staff estimates that sixty percent or more of the value of the TVX (the joint venture vehicle), including the engines and transmissions, will be built by Toyota. Not only does this represent a white flag by GM, but there will be powerful incentives for Ford and Chrysler to follow. The result of this trend is likely to be the precise opposite of what is desirable for the long run health of the industry and for vigorous competition-production of high quality small cars in the US.

The Risks to Competition

Even if this joint venture were viewed in a world market it would present competitive problems. However, the staff concludes, I believe correctly, that the appropriate geographic market is the United States and Canada. Widespread trade barriers, including our own. require that we evaluate competition based on U.S. (or at most U.S. plus Canadian) sales.

General Motors is by far the largest manufacturer of cars in the United States, with 44 percent of US and Canadian sales, as well as the largest in the world. It is the largest seller of small cars in the US. with a 27 percent share and the third largest seller of subcompacts in the United States and Canada. Toyota is the second largest seller of subcompacts in the U.S. and Canada, the fourth largest seller of small cars in the U.S., and the fourth largest seller of all cars in the U.S. and Canada. If we view the United States and Canada as a geographic market, and all

My conclusions are necessarily tentative, based on the evidence before us. My final determination would depend upon review of a full adjudicatory record.

new cars as a product market, the market is "highly concentrated" under the Justice Merger Guidelines (with a Herfindahl exceeding 2400). The automobile market has high barriers to entry. Consequently, we are faced with the type of market most susceptible to harm from collaboration among competitors.

A joint venture between two major competitors in the same market as the parents, the type of joint venture presented here, threatens to reduce competition by changing the incentives for the two parents to compete between themselves and with their joint venture and by facilitating information exchange. The ultimate result is the promotion of cooperative, rather than competitive, behavior.

These effects are inherent in this joint venture, regardless of the mostly cosmetic and unenforceable limits on information exchange provided in the settlement. For example, GM and Toyota will establish jointly the price GM will pay the joint venture for the car, in practice the principal determinant of what price GM will charge its dealers. The parties have agreed to base the price on a "market basket" formula, consisting of the weighted increases over the past model year (or fraction of it) for the top selling subcompacts. Although GM and Toyota would like us to view this provision as taking all discretion out of their hands. (akin to the annual Price, Waterhouse calculations of Academy Award winners), the effects of the formula are not so simple. They will constitute a percent of the market basket. They will be part of it, too. Thus, any increase in price of these two cars will weigh

heavily in determining the price of the TVX. Moreover, GM and Toyota have traditionally been the price leaders, GM setting the domestic industry benchmark with price announcements in the fall. and Toyota serving as the primary price leader for Japanese imports. Consequently, the market basket price changes are, to a large extent, Toyota and GM's price changes from the

If the parent companies are in competition, or might compete absent the joint venture, it may be assumed that neither will compete with their progeny in its line of commerce US. v. Penn-Oil Chemical Co., 378 U.S. 158, 169 (1964). "Of all joint ventures, the horizontal is inherently the most anticompetitive, because it involves the formation of a joint venture in the markets in which the parents operate. Under such circumstances. antitrust compliance and enforcement problems are scute. if the arrangement is allowed to operate at all the parents, through their representatives in the joint venture, will necessarily agree on prices and output in the very market in which they themselves operate." Brodiey "Joint Ventures and Antitrust Policy." 95 Harvard Law Review 1523, 1552 (1982).

36-253 0-84-30

previous year. A price increase by either will be assured of playing a large role in determining the TVX price.

·

Moreover, there is looseness in the way these components are calculated. Our staff concluded that the determinations will inevitably depend to some extent on "potentially imprecise judgments and predictions of market value" instead of factors outside the control of either company. Any discretion the companies have in calculating the market basket will, of course, provide the opportunity for strategic use of the provision. It is impossible to believe that this cooperative process in setting prices will not modify to some extent the incentives for the companies to compete with each other on the basis of price. Cooperation between the two price leaders, with a combined share of 50 percent of the new car market, represents a serious antitrust risk by any standards but this administration's. Moreover, the VRA reinforces the oligopolistic structure of the domestic industry by restoring much of the immunity from foreign price competition historically enjoyed by the domestic manufacturers."

In addition to the effects of the market basket formula on GM and Toyota's incentives, the agreement provides that the TVX price to GM will not be determined by the formula, and must be renegotiated between GM and the joint venture, if in GM or the joint venture's view, the price is at "significant variance with then current market conditions." This open-ended concept means that the price of the TVX (which in turn is likely to be a price leader for other small cars) is essentially a negotiated price between GM and Toyota. While the companies argue that the joint venture somehow has independent incentives to be profitable (again, the Price, Waterhouse image comes to mind). there is no escaping the fact that the joint venture is most fundamentally a GM-Toyota partnership, with each receiving half the profits. We can try to view the joint venture as somehow a separate company with separate incentives, but the image lasts only until we consider why and how decisions will be made.

Toyota asserts all price increases included in the market buske: are histoncal, rather than predicted but because of periodic changes during the model year. It is difficult to understand how the formula would work without some discretionary judgment

Professor Kwoka points out that the market shares in small car sales. efter excluding the Japanese, is strikingis similar to the traditional market positions of the "Big Three." with GM having 44.6%. Ford 28.35 and Chrysler 22.75market structure that was not known for vigorous price competition.

In addition to opportunities to cooperate on price and to influence each other's output, the parties are going to learn a lot about each other in ways that can reduce competition. For example, Toyota will know the transfer price to GM for the TVX well before it is announced (and before Toyota announces Corolla's price). If GM renegotiates the price of the TVX with the joint venture, there will be opportunities (and, of course, strong incentives) for GM to learn about Toyota's price plans before GM announces its own. The agreement also contemplates that GM will receive advance notice of design changes in the Corolla and the similar Sprinter (a version of the Corolla presently sold in Japan and likely to be the TVX). Uncertainty on the part of the dominant American firm of just what technical advances the leading Japanese importer has up its sleeve-up until now-did wonders for burning midnight oil in the R&D and Marketing Departments. Conversely, increased certainty about what the competitor will be offering helps make complacency the preferred strategy.

It is important to recognize that exchanges of highly sensitive competitive information are not prohibited under the consent agreement because they are inherent in the structure of the venture, that is, the parties consider some information exchange essential for the joint venture to work. I concede that these exchanges may be essential to carry out the joint venture in the way the parties have structured it, but this necessity only reinforces the fact that the risks to competition are inherent in the structure of the arrangement and cannot be cured by the Commission's settlement.7

Further, the joint venture results in increased opportunities and incentives for information exchange, even though they may violate the consent agreement. There have already been questionable exchanges between the two companies

which may well have blunted competition.

57253

Procompetitive Effects and Less
Anticompetitive Alternatives

I think it is beyond dispute that there are potential anticompetitive effects from this joint venture. The limitations insisted upon by the Commission (even though they have little substance) suggest general agreement that there is some prima facie showing of harm. The key question then becomes the magnitude of the potential competitive benefits. GM says the principal one is first-hand observation of how Toyota makes small cars. On close examination this "learning" efficiency turns out to be modest and, in any event, all or most of it can be achieved in a less anticompetitive way.

The TVX, certainly in the early years of its life and perhaps always, will be made primarily in Japan, not in Fremont. Toyota estimates the value added in the U.S. to be -percent. but the BC staff estimate the Japanese component to be over 60 percent. Consequently, the "efficiencies" that GM will be observing first-hand are limited mainly to assembly and stamping processes. GM points to the great importance of learning about the "kanban" process"just in time" delivery by suppliers. (It is a sign of growth in our national life that this is one of the first of a stream of Japanese words we are certain to adopt.) However, a June 1983 article in Ward's Auto World quotes a GM official as saying the implementation of this type of component delivery system is well underway at a Buick City plant. indicating GM has become familiar with the process on its own. The reality is probably that "kanban" is hard to implement by GM, given the historical relationship of American suppliers and the automakers, but it's much more likely to be a matter of convincing suppliers to change their ways, rather than learning special technology from the Japanese. Similarly. GM claims that learning about plant layout is a major

57254

benefit. But it seems safe to say that it would not take the extreme step of a joint venture with the major Japanese importer to adopt new and better plant design.

GM also argues that it needs a joint venture to serve as a catalyst in reorganizing its work force, e.g., by reducing the number of worker classifications. I concede that it may be difficult or impossible to change (at least over the short run) fundamental labor arrangements that have developed over many years without a new venture in a new (or newly reopened) plant. But this argues only for a new venture, not for one with the major Japanese importer.

I have no doubt that GM can learn from the Japanese, and that it would prefer learning from Toyota rather than other Japanese manufacturers or by recruiting management consultants. But in any proposed merger or joint venture, one firm will be a better manager than the other. We must be careful about justifying combinations among competitors on the grounds that the better manager will lift the other firm to its standard. Traditionally, we have relied on the weaker firm's incentives to improve its own operations, rather than to join its stronger rival in order to adopt its management techniques.

Production joint ventures have traditionally been thought to be justified, not because they allow a transfer of management techniques, but because they produce a new product, something "extra" not now in the market or production efficiencies which cannot be achieved without the joint venture. It is conceded by GM and Toyota that the TVX, though it may well be an excellent vehicle, is essentially the Toyota Corolla with minor, mostly stylistic, changes. Consequently, this is not a joint venture to produce an electric car, or one that is crash-proof, or even one that is significantly different in design from one now on the market. The joint venture enables GM to sell a Japanese-quality car by selling a Japanese car already being sold. Further, the evidence clearly suggests that the TVX is at least a partial substitute for the Chevette in the short run, and that the joint venture will

"The joint venture is in some respects a 'quasimerger where cooperation between formerly independent companies cfter acts to benefit and spur competition. The combined capital assets, or knowhow of two companies may facilitate entry into ne markets and thereby enhance competition. or may create efficiencies or new productive capacity urechievable by either alone." Brunswick Com FTC 1174. 1265 (1979), aff'd end modified ub rom Yamaha Mr Co. v. FTC. 637 F23 871 sh Car. ?), cert denied, 432 U.S. 915 (2982).

hasten the termination of the Chevette line. Far from a net addition to small cars in the U.S., the joint venture is more likely to reduce small car sales in the long run by driving up prices.

Consequently, even if we concede all of GM's arguments concerning the need to learn Japanese technology, the justification for the serious anticompetitive problems in the joint venture turn on the greater benefits of a venture with Toyota rather than a smaller Japanese firm. I find it exceedingly unlikely that the marginal gain of collaborating with Toyota rather than anyone else could provide this justification.

Our economists argue that one major "procompetitive" benefit of the joint venture is that it allows Toyota to import cars and avoid the Voluntary Restraint Agreement import quota. One can object to the VRA, one can argue it is harmful to competition, even harmful to the auto industry in the long run. But to argue that evading it can be counted on the "procompetitive" side of the ledger is an unacceptable exercise in second-guessing other national policies. The fact is that we are likely to experience some form of import restraints for some time because of the substantial Japanese cost advantage. The idea behind the restraints is to allow our domestic industry time to develop competitive cars, the very development the joint venture allows GM to avoid by selling a car made primarily in Japan. We would be turning the policy behind the VRA totally on its head by justifying the joint venture as a way to get around it.

Conclusion

The American automobile industry. until the 1970's, was a complacent oligopoly. Product innovation was too sluggish. prices were too high, promotional expenses became bloated, and management and labor inefficiencies became entrenched. The development that undercut this oligopoly and started us on the painful, but ultimately beneficial, road toward competitiveness has been the imports of foreign, particularly Japanese. cars. The Voluntary Restraint Agreement began a reversal of this trend by creating a wall against further Japanese imports. This trade barrier has inevitably been a step back toward the historic American oligopoly and one that increases the power of GM to set prices for the industry. Under these circumstances, it is a fundamentally misguided decision to sanction this type of cooperative

arrangement between the dominant domestic manufacturer and the leading importer. To justify such risks to competition we should require a substantial showing of true procompetitive benefits. These are not present. The cosmetic limitations placed on this joint venture by the consent agreement represent an antitrust policy based on crossing fingers and looking the other way.

In October 1981, GM embarked on a spirited "Beat Toyota Project." In 1982 the project was placed on hold. As of today, GM has a green light to embark on this "Join Toyota" project. That may be just fine for GM and Toyota, but it's bad news for the consumer. Addendum

Dissenting Statement of Commissioner
Patricia P. Bailey GM/Toyota Joint
Venture, File No. 821-0159
December 22, 1983.

It is a matter of serious concern to me that I am advised by the FTC General Counsel that section 6(f) of the FTC Act prevents me from providing in this Statement specific mention of certain information which, ultimately, was important to my decision in this matter. It is not entirely clear to me that certain of the information to which I would make reference should, in fact, not be disclosed pursuant to section 6(f). Nonetheless, I have deleted it and, thus, if references from time to time throughout this Statement appear vague, it is for that reason.

The Commission majority has today voted to accept a consent agreement with the General Motors and Toyota Motor Corporation which does not cure the antitrust infirmities of their proposed joint venture. I have, therefore, dissented from that decision.

I am acutely aware of the arguments favoring this joint venture. Certainly any knowledgeable observer would agree that American car companies, facing stiff foreign competition in the United States market, need to improve production techniques in order to strengthen their competitive positions into the future. The decision for this Commission, however, is whether a joint venture such as that proposed by these companies is sanctioned by the nation's antitrust laws. I do not believe by any stretch of the imagination that it is. Whether it should be is not for me to say. That argument should be posed in another forum.

In any event, to claim that the consent agreement accepted today, which allows a partial combination of the first and

third largest car companies in the world. solves any perceived antitrust problems with the venture, is simply, in my view, not the case. Indeed, both companies have acknowledged publicly that the consent merely restates the essential conditions of their original agreement.

The reasons for my decision in this matter are summarized below

Effect of precedent

There should be no mistake about the effect of the Commission's decision today. The principles of legality for this joint venture cannot be limited to one hermetically sealed experiment in Freemont, California. This joint venture is between the largest U.S. car producer and the largest Japanese car producerboth price-leaders for their makes of cars; thus, any similarly-structure joint venture between any other members of the industry must be sanctioned. How could we deny to other companies what we have authorized for the industry giants? In effect, this is rule-making for the industry.

It is predictable that several features of this joint venture will result in a reduction of competitive vigor between GM and Toyota. Concern about that should deepen when the strong likelihood that these features will be copied in "me-too" joint ventures between the remaining domestic car companies and foreign partners is considered. This joint venture, then. must be seen as a prototype for the industry that may well produce changes which are quantitatively more significant than those caused by it alone. The auto industry is clearly undergoing a concentration trend: the question is whether the Federal Trade Commission should accelerate that process by an action which will almost inevitably touch off a reactive pattern of strategic pairing between car

manufacturers. That is especially a troubling concern since the purpose behind these cooperative ventures would not be the creation of a new competitor, but rather a decrease in the overall number of market participants. leading to increased likelihood of tacit. if not actual. collusion.10

•See, eg. New York Times. December 21, 1983, p. Di If it gives them (the FTC] some comfort and vesis the deal. then it's OK." (quoung General Motors Chairman Ruger Smith Washington Post. December 21, 1983 p. D1 The precise terms of the coder.. are likely to include no more than a when agreement to abide by three elements of the venture that have already been publicly announced." (Acording to Toyola's US Counsel)

Professor Profsky has observed that a market Neng with numerous joint ventures rains pominular antitrust concerns. Profsky Ju at Verules Under The Antitrust Laws. Some

Nature of the transaction

Some joint ventures can be highly procompetitive, although this is not likely to be one of them. Particularly prized are ventures where the combination of the parent firms' resources achieves what neither can manage alone: an increase in pure research, a technological breakthrough, product innovation, or entry into a new market. This joint venture has none of those outputenhancing features. Manifestly, neither GM or Toyota is a new entrant into the automobile market. The car to be produced by this joint venture likewise is nothing new: it is a derivative of Toyota's Corolla. The design differences between the two models are "modest" and beneath the sheet metal the cars will be "essentially identical" (BC staff memo. I, 10)

On its face the GM/Toyota arrangement falls into the most suspect category of joint ventures:

Of all joint ventures, the borizontal is inherently the most anticompetitive because it involves the formation of a joint venture in the markets in which the parents operate. Under such circumstances, antitrust compliance and enforcement problems are acute: if the arrangement is allowed to operate at all, the parents, through their representatives in the joint venture, will necessarily agree on prices and output in the very market in which they themselves operate Brodley, supra. 95 Harv. L. Rev. at

1522

Or, as another commentator puts it: when one or both parent firms actively compete in the same product and geographic market as the joint venture, the inevitable coordination of competitive activities between parent and partly-owned subsidiary and the resultant stifling of aggressive behavior of the joint venture should be treated under typical cartel rules. Pitofsky, supra, 82 Harv. L. Rev. at 1035-1036.

Initial concerns about the joint venture's anticompetitive potential are only intensified when it is analyzed in its market context. Our economic and legal staffs have calculated the Herfindahl indices for various probable markets. They range from a low of 1262 (dollar sales, subcompact cars) to a high of 2413 (unit sales, all cars). (BC staff memo VI, 9; BE memo. Appendix H). This means that a plausible market is at best moderately concentrated, and at worst highly concentrated-but in any event structured in a way which mandates a very hard look at any

Reflections on the Significance of Penn-Olin, 82
Harv. L Rev 1007, 1033 (1969).

"US Department of Justice Antirust Guide Concerning Research Joint Ventures. 46 CCH Trade Reg Reports 35 (Decembe: 1. 1980). Brodley Joint Ventures and Antitrust Policy 95 Harv. L Rev., 1523 (1682x Pitolsky, op. CIL

combination of competitors. Entry barriers to this market are obviously quite high, consisting of economies of scale in production and distribution and, for foreign car manufacturers, import limitations. (BC staff memo VI, 22, 26). Within this oligopolistic market GM holds the longstanding leading market share (44% as compared with the 16.7% of its closest rival, Ford) and is the price leader among domestic auto producers. (BC staff memo, VI, 10, 12) 2 Toyota holds the same price leader position among Japanese importers. (BE staff memo. VI. 15). Toyota is the fourth largest car manufacturer in the U.S. and the third largest in the world. (BC staff memo. III. 1), 13

In short, this is a market which is prone to effective collusion, and a collaboration between two major competitors resembles a partial merger .more than a true joint venture. In these circumstances the degree of anticompetive risk and the genuine need for the venture must be stringently examined. See, eg., U.S. v. Penn-Olin, 378 U.S. 158, 170-72 (1964); Brunswick Corp., 94 F.T.C. 1174, 1265-66 (1979). off'd and modified on other grounds sub. nom., Yamaha Motor Co. v. FTC, 657 F.2d 971 (8th Cir. 1981), cert. denied, 102 S. Ct. 1768 (1982).

Anticompetitive Risks

The two principal aspects of the joint venture which I fear will lead to blunted competition between the two companies are the transfer price formula and the ongoing exchange of a broad range of product planning, engineering design, and marketing information.

The price which the joint venture will charge GM for the car is calculated by a formula which consists of a weighted average of wholesale prices of competitive small cars. Toyota's Corolla is given special weight in the formula. Simply between GM and Toyota this formula reduces price competition. because any price cuts Toyota gives its dealers must be passed on to GM, with a corresponding reduction in Toyota's joint venture profits. Consequently.. Toyota's incentives are to raise the Corolla price, knowing that such a price rise is incorporated into the cost of the joint venture car to GM; and knowing. moreover, that both it and GM are the

1 Genera! Motors is clearly the price leader among domestic auto producers, both because it announces prices first and because its prices virtually dictate Ford and Chrysler decisions (BC sia!! memo. V. 12)

13 In the aut compact portion of the U.S. marker which is most directly affected by this joint store,

Ford. Toyota and GM are ranked respectively Best, second and third, with the following market shores: 19:10, 16.065, 14.41% (PC staff memo. VI. 95!