An economic analysis of Matsushita revisited.
The U.S. Supreme Court's decision in the case of Zenith v. Matsushita(1) continues to incite debate among economists and lawyers. The case involved an allegation by the U.S. Color Television (CTV) manufacturers Zenith and NLTE that a group of Japanese firms, among which was Matsushita, had engaged in predatory below cost pricing as part of a collusive plan to drive U.S. firms out of the market. The case goes back at least to 1968 when Zenith filed an antidumping complaint against the Japanese firms. Despite a positive finding of dumping by the Treasury, the U.S. administration was slow to impose antidumping duties. Accordingly, Zenith decided in 1974 to bring a private suit under U.S. antitrust law (invoking also the little-used 1916 antidumping act, based on antitrust principles). The case against Matsushita et al. went before the district court in 1974, and in 1986, after almost 20 years of litigation and procedural wrangling, the Supreme Court finally dismissed the case on summary judgment without full examination of the evidence. Meanwhile, U.S. trade laws had been tightened to allow redress in cases such as this.(2)
The Supreme Court's decision was heavily based on theoretical economic reasoning: the Justices said, after a review of the evidence, that the allegations implied such economic irrationality on the part of the Japanese firms that predation was unlikely. The evidence brought forward, however, did demonstrate concerted action by Japanese firms on both the U.S. and the Japanese markets. Several observers have voiced the opinion that these irregularities should have warranted a full examination of the evidence, while others agree with the Supreme Court that there was no antitrust case to answer.(3) We agree with the former. After discussing at some length the explanation put forward by Blair et al.(4) we offer an alternative explanation for the behavior of Japanese firms in the industry that is based on modem economic theories of oligopolistic competition. In our explanation, the behavior of Japanese firms had anticompetitive effects and harmed U.S. producers, while providing no corresponding gains to U.S. consumers. This is not due to a conspiracy to monopolize the market but rather a spillover of collusive conduct and entry barriers on the Japanese market.
The analysis in this article highlights the difficulties in applying antitrust laws in the international marketplace. Although the announcement of the Department of Justice in April 1992 that anticompetitive conduct abroad harming U.S. producers will also be subject of litigation in U.S. courts may be seen as an attempt to counter these difficulties, the inevitable overlap with jurisdictions of other countries is likely to make enforcement unattainable. We will argue that the establishment of an international antitrust authority or at least harmonization of antitrust laws and enforcement policies would be welfare-improving and should be pursued.
In the following, we will review the key facts pertaining to the Japanese and U.S. CTV industries in the 1960s and 1970s and provide a brief overview of the litigation in the case. After discussing alternative explanations including those provided by Blair et al. we provide our own explanation of Matsushita. We conclude with a discussion of the implications of our findings.
Il. Stylized facts
From 1969 to 1977, the U.S. market share of the seven Japanese CTV producers rose dramatically from 12% to 44%.(5) Several sources indicate that during this period CTV prices in the U.S. were significantly lower than those in Japan.(6) There was no direct evidence, however, that Japanese export prices were below marginal costs. One of Zenith's expert witnesses, Horace DePodwin, produced calculations that supported this view but had to rely on aggregated constructed data that were rejected by the Supreme Court as of no evidential value.(7)
The seven Japanese CTV producers were engaged in two schemes that apparently were intended to limit competition between them on the U.S. market. Firstly, under the auspices of the Japanese Machinery Export Association, they agreed to each limit the number of U.S. distributors to a maximum of five (the so-called five-company rule). Secondly, they agreed on minimum export prices to the U.S. market; these so-called check prices were administered through the Japanese Ministry of International Trade and Industry (MITI). While the minimum prices were adhered to at U.S. customs, the Japanese firms secretly sought to give substantial rebates to their U.S. distributors by transfers to Swiss bank accounts.
In addition, and crucial to understanding Matsushita as we shall see, there was powerful evidence of the existence of a cartel agreement by the seven CTV producers in the Japanese market. Anticompetitive behavior in the Japanese market was facilitated by weak enforcement of the Japanese Antimonopoly Law and lack of credible entry threats related to the firms' control over retailing. In order to understand Japanese firms' behavior, it is useful to describe the structure of the Japanese CTV industry in some detail.
In the 1950s, Japanese consumer electronics firms established vast networks of wholesalers and retailers in Japan (the so-called distribution Keiretsu) which were to deal exclusively in the firms' own brands. The largest Japanese manufacturer, Matsushita, pioneered this strategy and came to control some 25,000 retail outlets all over Japan; Toshiba built up a network of 14,000 stores, Hitachi 10,000, Sanyo 6000, Mitsubishi 5000, Sharp 3500, and Sony 3000.(8) The cost of setting up the huge distribution network could be paid out of the surplus profits earned on the Japanese market which was protected until the end of the 1960s by quotas and high tariffs. During the 1960s and 1970s, almost all electric appliances in Japan where sold through manufacturer-controlled wholesalers and retailers.
Forward integration in distribution in the Japanese electronics industry had two major consequences for competition in the industry. Firstly, it constituted a major barrier to entry for foreign firms and potential Japanese entrants. It is argued that the retail networks were set up exactly for this purpose(9) although retailers' lack of financial means in the 1950s and the manufacturers' strategy to control the quality of promotion and after-sales service were also important. In any case, since the 1960s the seven major manufacturers mentioned above have dominated the Japanese CTV market, and no entry or exit has occurred in three decades.(10)
Secondly, control over a vast network of retail outlets led to tight vertical restrictions that directly facilitated horizontal anticompetitive behavior. The CTV producers used their control over retailers to fix retail prices, retail market shares and margins and to prevent retailers from selling competitors' products. These practices almost certainly inhibited forms of downstream procompetitive behavior that could have destabilized the manufacturers' horizontal cartel arrangements. The Japanese Fair Trade Commission (JFTC), the body that is given the task of enforcing Japan's Antimonopoly Law, several times found evidence of vertical price fixing and other restrictions in breach of this law. For instance, in 1971, it found that Matsushita assigned geographic territories to wholesalers and fixed minimum prices for retailers.(11) However, lack of independence of the JFTC and political realities apparently made the JFTC refrain from using formal sanctions. The JFTC instead relied on administrative guidance in the 1971 case against Matsushita and in other cases in which firms were asked to promise to change their behavior. Since it was highly unlikely that any sanctions would follow in case of noncompliance, firms were very often found to repeatedly breach the law.(12) Moreover, Japanese courts have been very reluctant to accept JFTC findings as definitive proof of wrongdoing.(13)
The possibility of engaging in vertical restrictions and vertical price fixing, the absence of credible threats of entry, and weak enforcement of the Antimonopoly Law all contributed to horizontal anticompetitive behavior by the CTV manufacturers. There is powerful evidence that the CTV producers were effectively cartelizing the market, collusively setting prices and producer market shares until well into the 1970s.(14) The JFTC investigated the industry in 1956 and 1966 and these investigations produced unambiguous evidence that a cartel existed. In 1956, Japanese producers were found to have set up a "market stabilization council" which in effect established fixed prices for CTVs; disobedient retailers were boycotted. In 1966, a draft decision of the JFTC stated that six producers held regular meetings under the auspices of the Electronic Industries Association of Japan (EIAJ), exchanged detailed information on prices, margins, production, and inventory, and coordinated price setting and sales volumes.(15) Although the Japanese firms pleaded nolo contendere--they did not deny the charges--the case never came to a final decision. In 1977, 11 years later, it was dropped.(16)
III. The Matsushita litigation
A. Litigation history
In 1968, the U.S. firms began their action by seeking complaints under the antidumping laws.(17) Faced with what appeared to be a reluctance by the administration to act decisively, Zenith and NUE in 1974 filed a case against Matsushita and the six other Japanese CTV producers under the Sherman and Robinson-Patman antitrust laws and the 1916 (Wilson) antidumping act.(18) They claimed that these firms had been engaged in a conspiracy since 1953 designed to keep prices high in Japan and low in the U.S., in order to drive U.S. firms out of the market. They alleged that the check price system and the five-company rule provided circumstantial evidence of a wider cooperation between Japanese firms to monopolize the U.S. market through predatory pricing.
This line of argument had several weaknesses. Firstly, the agreement on price setting and distribution in the U.S. aiming at reducing the intensity of competition was incompatible with a conspiratory scheme of predatory pricing. Secondly, whereas the existence of some form of concerted action was indisputable, it was extremely hard to show that the aim was to secure a joint monopoly to raise prices since the Japanese firms (1) never gained more than 45% market share, and (2) were not found to have raised prices during 20 years.
In the first round, the district court accepted a request by Matsushita and the other defendants to have the charges thrown out on summary judgment on the grounds that there was no case to answer. The court judged that most of Zenith's evidence was inadmissible, and what was admissible could not reasonably sustain a conspiracy charge even if accepted. This judgment was reversed by the appellate court who proposed sending the case to trial and accepting most of the excluded evidence. Finally the Supreme Court, by a 5-to-4 majority overturned the appeals court judgment, and reinstated the dismissal on summary judgment. The Justices cited the work of McGee, Easterbrook, and Bork to argue that predation was a very rare phenomenon and went on to suggest that it was particularly unlikely to have been possible in this case. The majority said:
If the factual context renders respondents' [i.e. Zenith's] claim
implausible--if the claim is one that makes no economic sense--respondents
must come forward with more persuasive evidence to support
their claim than would otherwise be necessary.(19) . . . Finally, if
predatory pricing conspiracies are generally unlikely to occur, they are
especially so where, as here, the prospect of attaining monopoly power
seem slight.(20). . . Two decades after their conspiracy is alleged to
have commenced, petitioners appear to be far from achieving this
goal: the two largest shares of the retail market are held by RCA and
the respondent Zenith.(21). . . The alleged conspiracy's failure to
achieve its ends in the two decades of its asserted operation is strong
evidence that the conspiracy does not in fact exist. . . . As presumably
rational businesses, petitioners had every incentive not to engage
in the conduct with which they are charged, for its likely effect would
be to generate losses for the petitioners with no corresponding gains.(22)
...In sum, in the light of the absence of any rational motive to conspire,
neither petitioners' pricing practices, nor their conduct in the
Japanese market, nor their agreements respecting prices and distribution
in the American market, suffice to create a "genuine issue for
The majority concluded that Zenith's accusations that Japanese CTV industry was attempting to drive the U.S. industry out of business in order to monopolize the U.S. CTV market, not only were untrue but could not possibly have been true. We argue that this conclusion is based on too many a priori assumptions. Not being in a position to have reviewed every single aspect of the arguments and evidence, we cannot argue that a reasonable fact finder would necessarily have found predation to have occurred. However, we would suggest that a reasonable fact finder might well have concluded that it did occur. There is room for dispute about what the appropriate definition of predation is and there is room for dispute about whether or not the Japanese firms' conduct could fall within such a meaning.
The U.S. definition of predation has evolved with circumstances, as the debate over the adoption of the Areeda-Turner criterion shows. The Supreme Court majority chose one specific definition of predation, which involves actually losing money in order to drive rivals out of business. But the appellate court and the Supreme Court minority would probably have chosen a broader definition within which anticompetitive behavior could have been identified. On the Supreme Court's definition of predation, it is highly unlikely that predation was occurring. We would argue, however, that if a broader definition of predation had been admitted, then the prima facie evidence, in the words of the appellate court, "creates a genuine issue of fact as to whether defendants conspired to dump color televisions in the United States with the specific intent to injure or destroy the industry in the United States."
Although the Supreme Court's definition of monopolization is not unreasonable, Scherer and Ross highlight the multiple meanings of the term "monopolize" that can be invoked under the Sherman Act.(24) The Japanese firms between them could be said to have obtained a dominant position in the world market. It did indeed eventually happen that all but one U.S.-owned producer of CTVS went out of business, though CTV production continues by foreign investors.
We argue that the Supreme Court prejudged the outcome of a full investigation by looking only at a certain class of economic models. It is indeed likely that if a like-minded court had subsequently examined the case in full, they would also have found no predation, so the majority decision was internally consistent. But the recorded opinions of the appellate court and the Supreme Court minority indicate that other interpretations were admissible on the basis of an initial reading of the facts. Because the minority opinion argued that the facts did lend themselves to an interpretation of predation, it seems that there were reasonable people who might have found for Zenith on a more detailed inspection of the evidence.(25)
B. Alternative explanations
The Court's decision provoked considerable dispute as to its wisdom and many lawyers and economists have given differing opinions on the case. We will review in turn the leading economic hypotheses that have been offered to explain what undoubtedly was unconventional conduct. We argue that there are a number of explanations of the facts that are consistent with modem industrial economics construing the Japanese action as rational conduct likely to drive U.S. firms out of business in a way that was facilitated by the collusive exercise of market power in Japan.
1. SALES V. PROFIT MAXIMIZATION First of all, the Supreme Court minority, in the dissenting opinion, accused the majority of simply imposing their own view of rationality and ignoring alternatives, thus trespassing on the territory of the "fact finder." Justice White, for the minority, attacked the assumption of profit maximization inherent to the traditional analysis of predatory pricing:
Discussing the improbability of a predatory conspiracy assumes that petitioners valued profit maximization over growth. In light of the evidence that petitioners sold their goods in this country at substantial losses over a long period of time, I believe that this is an assumption that should be argued to the fact finder, not decided by the Court.(26)
In fact, recent work by economists on the governance of Japanese firms suggests that the relative power of "stakeholders" (managers, employees, firms and banks within the same business grouping) versus stockholders leads to a stronger pursuit of market share at the cost of profits.(27) This suggests that a restriction of the economic analysis to models of profit maximization may not explain the behavior of Japanese firms in full, although we do not suggest dismissing this concept lightly.
2. STRATEGIC INVESTMENT BEHAVIOR The minority also argued that without the concerted action on the Japanese and U.S. markets, CTV production by U.S. firms would have been higher. They cite with approval expert witness Horace DePodwin's opinion:
As it was, however, the influx of sets at depressed prices cut the rates of return on television receiver production facilities in the U.S. to so low a level as to make such investment uneconomic.(28)
In other words, Japanese firms might have been engaged in strategic behavior to deter investments by U.S. firms. Such behavior can be profitable even if no full monopolization of the market occurs. Dixit has shown that by being the first to build capacity ahead of demand, firms may be able to deter entry by their rivals.(29) Once the capacity is in place, the firm is precommitted to incurring only marginal costs from extra production. A similar argument applies if incumbent firms have to decide about investment in new product lines or cost-reducing manufacturing processes and investments in additional capacity in growing markets. In general, structures that transform variable costs into fixed costs can provide a strategic advantage.(30) Such structures may not even be a conscious strategic decision by firms, but the result of institutions, such as long-term employment systems that make labor into a fixed cost for the firm. Modem game theoretic analysis of strategic behavior by firms in imperfectly competitive markets has also shown that there are more kinds of predation than those leading to full monopolization.(31) Firms may invest in excess capacity or set low prices in order to deter entry or to persuade rivals in accepting lower market shares. This may succeed without prices being lower than average variable costs such that predation will not be identified by the Areeda-Turner rule. In sum, there exists a rich menu of strategic behavior and the possibility of behavior to deter investments by U.S. firms cannot be dismissed ex ante.
3. LEARNING AND SCALE ECONOMIES Scherer and Ross argue that the definition of predation was set too narrowly by the majority. They agree with the minority that there was a case to be examined, but they base their argument on dynamic scale economies, arguing:
. . .if there were enduring economies associated with learning by doing (which seems likely) a high-domestic-price, low-export-price strategy would have been profitable even when continued for more than a decade as capacity was expanded. It would not have been necessary for the Japanese producers eventually to raise U.S. prices to realize the pot of gold at the end of the predatory-pricing rainbow.(32)
Thus, the main argument here is that in the presence of learning effects, predation losses could have been recouped without raising prices.(33) This was not raised by the claimants Zenith and NUE and hence was not taken into consideration by the Supreme Court.
4. REPUTATION EFFECTS Blair et al. offer one of the most recent and comprehensive explanations for the behavior of Japanese firms. Like the Supreme Court majority and also Elzinga, Blair et al. dismiss a predatory pricing conspiracy on the grounds that predation losses could never have been recouped.(34) They also dismiss the possibility that the cartel in Japan was implicitly allowed by MITI as an instrument of strategic trade policy through which profits generated in Japan could subsidize sales in the U.S. They argue that, given the alleged loss-making pricing below marginal costs, such a policy would allow exploitation of Japanese consumers without providing corresponding benefits (profits) to producers.
Instead they offer an alternative explanation, suggesting that the primary motive for the lower prices in the U.S. was to take advantage of a "reputational externality." With Japan-made products suffering from a low-quality image in the U.S., every high-quality Japanese CTV sold would make demand conditions for Japanese CTVs, but also for other Japan-made products, more favorable. MITI, representing the interest of Japanese industry in general, accordingly bad a (legitimate) interest in subsidizing sales in the U.S. with profits generated in Japan, to benefit from a positive externality for other industries. Blair et al. suggest that a simple scheme set up by MITI to encourage sales in the U.S. could have generated the observed behavior. MITI could have allowed the individual Japanese firms a share of the profits in Japan proportional to their market share in the U.S. Firms accordingly had an incentive to lower prices in the U.S. even below marginal costs in order to increase sales, since the marginal losses on the U.S. market would be automatically compensated by greater profits in Japan. Since under this scheme price competition in the U.S. by Japanese firms would ultimately lead to destabilization of the cartel, MITI needed to establish the minimum price system. This in turn gave individual firms an incentive to increase U.S. market shares by secretly giving rebates to U.S. buyers. Blair et al. conclude that, since MITI had a legitimate interest in pursuing positive externalities and there was no predatory intent, there was no breaching of antitrust laws, although U.S. producers were clearly and maybe "unfairly" harmed.
While at first glance the above explanation seems to incorporate all the stylized facts of Matsushita, we find it unconvincing. Firstly, it is implausible that there were many U.S. consumers in the 1970s who still had a negative image of Japanese products, since U.S. consumers already had experience with the reliability of Japanese calculators and black and white televisions.(35) It is unlikely that the extra reputation effect through subsidized sales of CTVs was so large as to warrant a costly encompassing interventionist scheme set up by MITI as implied by the analysis of Blair et al.
Secondly, Blair et al. also stretch their explanation too far by arguing that the check price system was part of a grander scheme by MITI to balance concerted action, implying that firms giving secret rebates were trying to cheat MITI. This implies that the cheating firms expected that they would not be detected: if they could increase market share by giving secret rebates, MITI would reward them with greater profits in Japan. However, given the degree of concerted action and the intensity of information exchange, it is highly unlikely that a firm could post sudden sales increases this way and be rewarded for it without other firms or MITI noticing it.36 The existence of the check price system can be explained easily without resorting to conspiracy schemes. MITI has always seen its role as a protector of Japanese industry by aiming to minimize conflicts with major trading partners and so preserving a free flow of Japanese exports. In case of Japanese CTV exports, it feared that unbridled exports of cheaply priced CTVs to the U.S. would eventually lead to actions under antidumping or trade acts. By establishing minimum prices, the existence of which was in fact public knowledge,(37) it reduced the chance of dumping findings and showed its good will to the U.S. administration.(38)
Thirdly, the rejection by Blair et al. of the cartel as an instrument for strategic profit-shifting hinges on the assumption that Japanese firms were making losses on sales below marginal costs. However, precisely in the case of important reputation effects as analyzed by Blair et al., pricing below marginal costs, or "forward pricing," is rational for profit-maximizing firms. If reputation effects are important, setting low prices in early periods helps to secure brand loyalty of more consumers and will lead to higher market shares and profits in later periods, more than offsetting the initial losses. Such reputation effects naturally will be much greater at the level of the individual brand than at the country level. Furthermore, as discussed above, pricing below marginal costs is compatible with rational profit-maximizing behavior if there are cumulative learning effects on the production side. Thus, even if we assume that the available evidence substantiates below marginal cost pricing (which is disputable), this does not necessarily imply that profit-maximizing Japanese firms could only be found behaving like this under the Miti-administered scheme suggested by Blair et al.
5. OUR EXPLANATION The above considerations suggest that the Supreme Court based its decision to dismiss the case on summary judgment on a narrow class of economic models. We argue that the most important omission was neglecting the possibility that anticompetitive behavior in the Japanese market might affect Japanese firms' behavior in the U.S. market. In the next section, we offer our explanation, based on such spillovers of anticompetitive conduct from Japan to the U.S., which is compatible with the stylized facts of Matsushita. We obtain our results while maintaining the assumption of profit-maximizing firms and without including learning and reputational effects. We show that what may have mattered most for Japanese firms' behavior is the asymmetry in market conditions: intense competition in the U.S. market compared with a cartel and insurmountable entry barriers on the Japanese market. The fact that the Japanese firms could cartelize their home market and reap surplus profits there, could eventually have driven out U.S. firms and may moreover have eventually harmed U.S. consumers. We call this "market share predation without monopolization." Our analysis has much wider implications for antitrust than those concerning the Supreme Court's handling of the case only. We will argue that the most important lesson to be learned from Matsushita is that in international markets, competitive conditions in a country, which are ultimately the result of antitrust rules and enforcement of these, can have a major impact on competition abroad. Antitrust policy itself can be used to give home firms a strategic advantage over foreign firms in the international marketplace.
IV. A model of international rivalry in the CTV industry
The evidence from the Matsushita-Zenith case does not prove that Japanese CTV producers were engaged in predatory pricing in order to monopolize the U.S. market. However, Japanese firms were undoubtedly price discriminating, selling CTVs in the U.S. at prices substantially below prices charged in Japan, and thus reducing U.S. firms' market shares. This entails dumping by modem antidumping law and raises issues of "fair competition," but the absence of a proven monopolistic intent makes an antitrust case hard to mount unless international price discrimination per se is banned.
The idea which we try to incorporate in our analysis of the CTV industry is that the behavior of Japanese firms could have been the result of differences in the structure of the Japanese and U.S. CTV industries, rather than the result of a monopolization strategy with eventual increases of U.S. prices to Japanese levels in mind. We show that the assumptions of cartelization of the Japanese market and effective control over the distribution system blocking access to U.S. firms and potential Japanese entrants alike, are sufficient to explain dumping and the exit of U.S. firms.(39) under these conditions, normal profit-maximizing Japanese firms will be found exporting at prices below Japanese prices and below full production costs in the long term, but export prices will still be above marginal Costs.(40) Below we will summarize the model and its analytical results.(41)
We consider two markets for CTVs, one in the U.S. and one in Japan, characterized by linear demand functions P = a-bQ and P* = a*-b*Q*, respectively. Asterisks refer throughout to variables related to Japan. There are N identical firms producing in the U.S. and N* identical firms producing in Japan. Production costs involve a fixed cost F of acquiring the technology to produce and setting up production facilities, which is assumed to be the same for U.S. and Japanese firms. Unit variable costs (c and c*) are invariant with respect to output. Variable costs may differ between U.S. and Japanese firms, which allows for a possible efficiency advantage of Japanese firms. In addition, there are fixed per unit transport costs t associated with exports.
The U.S. CTV industry is oligopolistic such that prices exceed variable costs of production in the short term and firms may earn excess profits.(42) However, in the long term, excess profits eventually attract new entrants which will bid profits to zero. In longterm equilibrium in the U.S. industry, firms earn just enough of a surplus over variable costs to cover the fixed cost. If prices rise above this level there will be entry, and exit occurs if prices fall below it. Price is above variable costs but in equilibrium no one can afford the fixed cost of entry to take advantage of this. This long-term equilibrium situation on the U.S. market before the entry of Japanese firms is represented by figure 1. The U.S. firms between them sell Q CTVs at a price P which is strictly higher than variable costs c. However, the variable profits earned by the industry, indicated by the area A, are equal to the fixed cost incurred by the firms: A = NF, such that net profits in the industry are zero.
In contrast to the U.S. industry characterized by free entry and exit, a fixed number of established firms, N*, is active in Japan, while entry is prohibited. To highlight the effects of differential market structure in our analysis, we assume that cooperative behavior by Japanese firms takes the extreme form of a production cartel designed to reap monopoly profits, which are shared equally by the N* firms. This situation is represented by figure 2. Since the U.S. market is roughly twice as large as the Japanese market, we assume for convenience that b* = 2b and a* = a: for any given price, demand in the U.S. is twice demand in Japan. The Japanese cartel behaves as a monopolist and sells Q* CTVs at the monopoly price P*. Variable profits, indicated by the area B, are greater than the fixed cost: B>N*F. This situation remains unchanged in the long term since no entry occurs.
The Japanese firms can enter the U.S. market, but U.S. firms cannot enter the Japanese market. Profit-maximizing Japanese firms seek to cover variable costs in the U.S. plus transport costs, but they do not have to incur the fixed cost a second time. Selling in the U.S. does not affect variable costs (we abstract from learning effects) but spreading the fixed cost over more units lowers average costs. Entry by the Japanese firms in the U.S. market can be shown to have the short-term effect of depressing prices in the U.S. and increasing market volume, thus benefiting U.S. consumers.
Figure 1 The US Market Before Japanese Entry
Figure 2 The Japanese Market Under a Cartel
However, at depressed prices, U.S. firms are not covering the fixed cost, some must exit, and joint output of U.S. firms falls. With the number of U.S. firms adapting to the entry of Japanese firms, output from each of the surviving U.S. firms in the new equilibrium is at the preimport level, which guarantees that they can stay in business. With U.S. firms' output now reduced, the price comes back to its preimport level, so import competition does not actually benefit consumers in the long term. The new equilibrium is illustrated by figure 3, in which the new equilibrium values are indicated by a prime. The new price, P', is equal to the previous price P. Japanese firms reap variable profits on the U.S. market amounting to Q*(P'-c*-t), indicated by the area C. We assumed in figure 3 that Japanese firms have an efficiency advantage that is greater than transport costs, c*+t < c, though this is not crucial to the analysis. U.S. firms are not able to capture the same output and variable profits as before with Q' < Q and A' < A. This means that the fixed cost of fewer U.S. firms can be supported and N' < N.
It can be shown that the ability of the Japanese to cartelize their home market and to charge monopoly prices means that it is likely that dumping will be observed in the sense of export prices being lower than home prices. The Japanese can afford to sell below their average costs since they can recoup the fixed cost through their profits on sales on the Japanese market only: B > N*F. There is no traditional price predation here, in the sense of aiming to monopolize the market and raising prices to consumers. Japanese firms will not sell at U.S. prices below variable costs. They aim at maximization of profits and entry in the U.S. market strictly increases profits by C. In any case, there is no prospect of monopolization of the U.S. market because there is assumed to be free entry in the U.S. However, Japanese firms do sell at a price below average total costs in the U.S. even in the long term and their behavior harms U.S. producers. The existence of anticompetitive behavior in the Japanese market causes behavior in the U.S. market that increases Japanese total sales above what they would have been with fully competitive markets.
The above analysis thus shows that the assumptions of free entry on the U.S. market and a cartel barring any entry in Japan are sufficient to explain dumping by Japanese firms associated with the exit of U.S. firms. Of course, we acknowledge that these are somewhat restrictive assumptions. Free entry on the U.S. market implies a more competitive market than is the case in reality. In the model, this assumption assures that profits and prices are kept low and are unaffected by Japanese imports. Likewise, the assumption of a cartelized Japanese market will exaggerate the extent of collusive practices on the Japanese market. Nevertheless, as long as competitive conditions in the U.S. and Japan are anywhere near these stylized concepts, the results will hold and only the magnitude of the effects will differ.
It is worthwhile also to consider the assumptions concerning fixed and variable costs, which are important to the kind of analysis presented here. We assumed that the fixed cost only has to be incurred once. This enables firms to exploit scale or scope economies abroad. To the extent that CTV producers are involved in continuous rounds of R&D spending to introduce new and better models this assumption is justified: the fruits of R&D can be transferred abroad to export markets at near zero marginal Costs.(43) Following this logic, the analysis is also applicable if we take into account that since early 1973 several Japanese CTV producers have been investing in manufacturing facilities in the U.S. This is because Japanese firms retain the advantage of covering the fixed R&D cost in the Japanese market. Finally, we assumed variable costs to be constant, while variable costs may be a function of cumulative production if firms "learn by doing." In this case, the effects of asymmetric industry structure will only be strengthened: exclusive access to the Japanese market will give Japanese firms greater leverage over U.S. firms and will enable them to further increase their U.S. market share.(44)
Policy options for the U.S.
The implications of the differences in industry structure and market access between the U.S. and Japan can be seen more clearly by considering what would happen under liberalization of the Japanese market, leading to some form of reciprocal access. There are three possible cases:
1. Under U.S. pressure entry of U.S. firms into the Japanese market as members of the cartel is allowed, while the cartel remains in existence;
2. U.S. firms are allowed to enter the Japanese market, though not potential Japanese entrants--U.S. entry causes the cartel to collapse;
3. Any Japanese and U.S. firm is allowed to freely enter the market.
We will discuss the results of the second case in which U.S. firms are granted privileged access to the Japanese market, presumably by the promotion of U.S. CTVs through the Japanese distribution network.(45) Japanese firms are not able to operate the cartel under these conditions and competition on the Japanese market becomes characterized by oligopolistic rivalry just as in the U.S. market.
The new equilibrium on the Japanese market is illustrated by figure 4 (equilibrium variables are now indicated by a double prime). The entry of U.S. firms brings prices in Japan down to P*". Variable profits of Japanese firms decrease to C", but for convenience we assume that these are still sufficient not to force existing Japanese firms out of business. U.S. firms are now able to earn variable profits on the Japanese market indicated by the area D.
The entry of U.S. firms on the Japanese market has positive consequences for competition in the U.S. market, as is illustrated by figure 5. Since U.S. firms now earn variable profits on the Japanese market as well, more U.S. firms are able to earn sufficient profits to cover the fixed cost of entry. Profits in Japan and the U.S. are bid down by new U.S. entrants until N" F = A"+D, with N" > N'. The intensified competition by a greater number of U.S. firms will cause the price of CTVs in the U.S. to fall: P" < P'. Because U.S. firms are now able to sell in Japan, they have higher total production than before, spread the fixed cost over more sales, and are able to tolerate the lower home price.
Reciprocity would thus bring gains to consumers compared to the status quo. In this sense therefore the sum total of Japanese behavior in both markets is harmful to U.S. consumers as well as producers. Given that this harm arises from anticompetitive behavior by Japanese firms albeit in the Japanese market, Japanese behavior could be said to fall within a broad notion of "predation," which we have termed "market share predation without monopolization."
The results are quite robust to the choice of reciprocity. If we assume reciprocity takes the form of U.S. producers joining the Japanese cartel (case 1), results would be even more favorable for U.S. firms and U.S. consumers, as long as we impose that U.S. entrants (attracted by high profits) are also allowed to join. If reciprocity takes the form of entirely free entry for Japanese and U.S. firms alike (case 3), U.S. firms will still benefit, but less than in the other two cases, since Japanese entrants would take a share of the world market. The price in the U.S. market would still come down, while the price in Japan evidently would decline even more.
V. Conclusions and implications
We will first consider conclusions that follow directly from our analysis of the international CTV industry and then discuss the broader implications of Matsushita for antitrust policy.
A. The implications of the model
We have presented a highly simplified model that suggests that asymmetric market access caused by asymmetric competition policies can induce profit-maximizing firms to export at prices below average costs and below domestic prices. In our simple model, exports by Japanese firms to the U.S. market force some U.S. firms to exit, driving down prices in the short term, while the equilibrium price in the long term is unchanged. Reciprocal access for U.S. firms on the Japanese market would give U.S. firms the opportunity to reap export revenues. This would not only enable more U.S. firms to survive, but would also lead to increased competition on the U.S. market and lower U.S. prices. U.S. consumers would thus benefit from the opening of the Japanese market.
The latter result contrasts with the conventional view that dumping, as long as there is no eventual monopolization, while harming U.S. producers, at least benefits consumers. Our analysis shows that depressed prices may just be a short-term phenomenon. This is a direct consequence of the exit of U.S. firms in the long term, and not the result of a deliberate monopoly pricing plan. The analysis suggests that an assessment of dumping and price discrimination should take long-term effects on entry and exit into account. Japanese firms' behavior, while harming U.S. producers, may ultimately harm U.S. consumers as well, as compared to a situation of full reciprocity. Thus, anticompetitive behavior abroad can create negative "spillover" to other countries' industries in global markets, and U.S. antitrust authorities may have a legitimate interest in prohibiting such anticompetitive behavior.
If our model does indeed approximate reality, a major message evolving from the analysis here is that differences in competition policy can have effects similar to strategic trade policies, giving asymmetric advantages to firms competing on international markets. MITI may have had the motive to allow Japanese firms to cartelize their home market to reap economies of scale and scope while prohibiting entry by foreign firms. In fact these effects are quite similar to modern analysis of strategic trade policy under increasing returns to scale.(46) Nonenforcement of antitrust laws, by allowing firms to collude and restrict entry, may function as a tool of strategic policy. While this increased profits of Japanese producers, Japanese consumers obviously lost. The Japanese government had a motive to relax antitrust rules if it valued the positive effects for Japanese industry greater than the losses to consumers. For instance, it may have taken into account that growth of the Japanese CTV industry can have positive spillover effects through increased R&D activities and the acquisition of manufacturing skills through learning by doing.
In our model, by taking a few salient characteristics (cartelization and prohibited entry in Japan, free entry and exit in the U.S., and some kind of economies of scale and scope), we can generate results that are consistent with the stylized facts of Matsushita. In this view, anticompetitive behavior in the U.S. takes the form of "market share predation without monopolization." This means that Japanese market shares are being increased in the U.S. and injury is being inflicted on U.S. firms (with no gain to consumers) as a result of the incentive structure created by anticompetitive behavior in the Japanese market. The behavior follows from cartelization of the Japanese market, even if independent profit-maximizing behavior applies in the U.S. market. Although no predatory intent or collusion needs to be assumed in the U.S. market, the consequences of Japanese firms' pricing below average total cost can be anticompetitive and harmful to U.S. interests. This departs from the orthodox view of predatory pricing that predation should only be seen as harmful if there is an intent to monopolize.
Although we do not claim that our analysis is the only explanation for the observed behavior and stylized facts, we do argue that it is a distinct possibility. If our explanation is correct, it is hard to predict whether a full examination of the evidence would have led to a finding of predation. This is because it seems unlikely that the behavior of Japanese firms would lead to a finding of price predation under the Areeda-Turner criterion. On the other hand, one can argue that the cartel in Japan was set up willingly to provide Japanese firms with surplus profits that could finance long-term below-average-cost pricing to increase market share at the cost of U.S. producers. The resulting price discrimination could have been found violating the Robinson-Patman Act if the courts had been willing to give that Act an international dimension. While we would suggest that on the basis of our analysis a full investigation of the evidence would have been warranted, we argue that the broader implications of our explanation of Matsushita are of much greater importance.
B. Broader implications of the Matsushita case
Our review of alternative explanations for Matsushita including our own explanation would lead one to doubt the economic reasoning behind the Supreme Court's decision to dismiss the case on summary judgment. But at the same time one could still argue that litigation of this kind is bound to fall at some point in the judicial process, just because there is an inherent ambiguity in the evidence. A predation case is almost always speculative and ultimately unprovable, so why open the way for potentially competition-deterring litigation even if it means allowing a few scoundrels to flourish? Where competing economic models each offer a different alternative counterfactual scenario, any antitrust court is going to have to take a position on the basis of inherently unobservable data on what would have happened if the alleged offender had acted otherwise. The difficulty of doing this has led courts to be very doubtful about making predation decisions. This skepticism is even partly shared by economists who do believe in the possibility of predation. Milgrom and Roberts have argued that, while predation is more likely than McGee supposed, it is nevertheless almost always unprovable.(47)
The Matsushita case may seem all the more difficult to solve since we argued the case was basically one of market share predation without monopolization, which is likely to pass traditional predation tests such as the Areeda-turner criterion.(48) However, if we are correct in our analysis of Matsushita, anticompetitive effects on U.S. industry can stem from weak antitrust laws and enforcement in Japan. This implies that this class of anticompetitive behavior needs to be combated by applying more uniform antitrust standards to globalized industries.(49)
The announcement by the U.S. Department of Justice in April 1992 that anticompetitive behavior abroad affecting U.S. producers (exporters) would be the subject of U.S. antitrust litigation indeed shows a growing awareness of the possible spillover effects of anticompetitive behavior abroad. However, this type of extraterritorial application of U.S. antitrust law will generate conflicts with antitrust policies of foreign countries.(50) It is clear that international codes governing alleged transnational anticompetitive or predatory behavior are called for. Such rules would presumably in the first instance affect states, in that they would require certain degrees of harmonization of antitrust law, or at least removal of incompatibilities. They would forbid countries to exempt export cartels from antitrust laws, as has in the past been the case in Japan, the U.K. and Germany. They would specify rules under which countries, users, or alleged victims of predation could act against firms whose behavior in one market was affecting outcomes in another.
The need for international antitrust enforcement is all the more apparent if one considers another reality of global competition: the fact that firms all too often take recourse to trade protection laws such as antidumping and countervailing duties. The same evidence pertaining to Matsushita led to a finding of dumping under the 1921 Antidumping Act in 1971, but the U.S. Treasury Department delayed levying duties. In 1974, domestic producers also applied for import relief under the 1974 Trade Act and in 1976 they petitioned for higher import tariffs under the 1930 Tariff Act. In 1977 all these pending cases finally were closed after the signing of an "orderly market arrangement" in which Japanese firms through MITI promised to limit CTV exports and maintain minimum prices.
One might argue that by refusing to apply antitrust law to the Zenith case, the Justices left no recourse other than trade law or outright protectionism to aggrieved domestic competitors. This is important because it has been shown that the application of trade laws often harms consumers and may even have serious anticompetitive effects. Antidumping law, for instance, is ostensibly based on the same economic notions of predation for monopolization and one might suppose that the same criteria would apply. But in practice the system is very different.(51) Antidumping law does not accord symmetrical rights to all producers and neglects the interests of consumers and user industries.(52) The nature of the procedures and burden of proof are quite different. Many observers of antidumping practices in the U.S. and the E.C. argue that procedures are biased toward a finding of dumping. Although dumping duties may only be levied if they are causing "material injury" (such as declining profits and market shares) to the petitioning industry, a genuine proof of causality between "injury" and the alleged dumping is usually not required. Antidumping practices are thus found to discriminate against foreign firms and to be open for protectionist abuse.(53)
Furthermore, Prusa has shown that U.S. producers may make strategic use of antidumping actions to induce foreign importers to agree on undertakings restricting price setting and sales.(54) Such private undertakings are allowed as settlements of antidumping cases and the firms involved are exempted from antitrust persecution. Stegemann argues that undertakings frequently settling E.C. antidumping cases provide opportunities for illegal price fixing beyond the purposes of antidumping law.(55) Messerlin and Stegemann show that E.C. industries with a history of collusive conduct tend to petition more for antidumping protection while antidumping cases involving these industries are relatively often settled through undertakings.56 Thus, trade laws, rather than acting against predatory or anticompetitive behavior, may facilitate anticompetitive behavior.
It is rather ironical that while economists and lawyers are going through sophisticated debates on the economic logic of antitrust policy, the reality has rapidly surpassed them and has led to recourse to trade instruments with much more harmful effects to consumers, whose interests competition policy is supposed to protect. We suggest that at least one class of problems, those stemming from differences in antitrust enforcement, can be dealt with through the harmonization of antitrust laws and the enforcement of these. This is likely to provide welfare and efficiency gains worldwide while at the same time reducing the possibility to excuse disguised protectionism and anticompetitive conduct by alleging "unfair" behavior by foreign firms.
[Figures 1 to 5 ILLUSTRATION OMITTED] (1) Matsushita Electric Industrial Co., Ltd. v. Zenith Radio Corporation, 106 S. Ct. 1348 (1986). (2) Parallel trade policy complaints continued in the meantime. The 1974 Trade Act was passed with this industry largely in mind, and the alleged loopholes in this act prompted further moves in Congress. (3) Examples of the former are David Schartzman, The Japanese Television Cartel (1993); F. M. Scherer & D. Ross, Industrial Market Structure and Economic Performance (1990); J. A. Ordover, A. O. Sykes & R. D. Willig, Unfair International Trade Practices, J. Int'l L. Pol. 323 (1983), P. Milgrom & D. Roberts, New Theories of Predatory Pricing, in Industrial Structure and the New Industrial Economics 112-37 (G. Bonano & D. Brandolini, eds., 1990); examples of the latter are: D. Blair et al., An Economic Analysis of Matsushita, 36 Antitrust Bull. 355 (1991); Kenneth G. Elzinga, Collusive Predation: Matsushita versus Zenith, in The Antitrust Revolution (Kwoka & White eds., 1989); Kenneth G. Elzinga & David E. Mills, Testing for Predation: Is Recoupment Feasible?, 34 Antitrust Bull. 869 (1989). (4) Blair et al., supra note 3. (5) A similar increase in market share was recorded for black and white televisions. Throughout this article, however, we will limit the analysis to CTVs. (6) See Blair et al., supra note 3, and Kozo Yamamura, Caveat Emptor: The Industrial Policy of Japan, in Strategic Trade Policy and the New International Economics 169-210 (Paul Krugman ed., 1986). (7) Blair et al., supra note 3, at 362 n.27 regard the DePodwin estimates as a "reason for suspicion" of below marginal cost pricing. Schartzman, supra note 3, at ch. 6, regards the DePodwin data as evidence that U.S. prices were below average total costs. (8) See D. Flath, Vertical Restraints, in JAPAN AND THE WORLD ECONomy 187-203 (1989) and D. Flath & T. Nairu, Returns Policy in the Japanese Marketing System, J. Japan & Int'l Economies 49 (1989). (9) See K. Van Wolferen, The Enigma of Japanese Power (1989). (10) Yamamura, supra note 6. (11) See JFTC, 17 JFTC Decision Reporter (Kousel Torihiki Iinkai Shinketsu Shuu) 187 (1971). (12) See Schwartzman, supra note 3, at ch. 5; Flath, supra note 8; and Kenji Sanekata, The Enforcement of Competition Policy in Japan: Its Sicio-Politico Background and Enforcement System (University of Hokkaido 1989). (13) Sanekata, supra note 12. (14) See Yamamura, supra note 6 and Schwartzman supra note 3, at ch. 5, for detailed descriptions of the evidence. (15) The six firms were Sanyo, Toshiba, Sharp, Hitachi, Matsushita, Mitsubishi; Sony was not implicated. See JFTC, JFTC Decision Reporter (Kousei Torihiki Iinkai Shinketsu Shuu) 37 (1978). (16) What may have played a role in this decision is that in the same year, the last of the regular meetings by the Japanese manufacturers was reportedly abolished (see Schwartzman, supra note 3, at 87). Also in 1977, greater awareness of the harmful effects of cartels led to a surcharge system that enabled the JFTC to fine firms engaged in cartelization. The Japanese firms had never faced a formal financial sanction until then. (17) For fuller details of the litigation including the antidumping aspects see Schwartzman, supra note 3. (18) See Blair et al., supra note 3, at 356. (19) 106 S. Ct. at 1356. (20) Id. at 1358. (21) Id. (22) Id. at 1359. (23) Id. at 1361. (24) Scherer & Ross, supra note 3. (25) Jorde and Lemley argue that the Court chose to apply a particularly strict criterion of proof at summary judgment because the allegations were of a predatory conspiracy which is per se illegal whereas if only one firm had been involved a rule of reason would have left further discretion in a full trial. This issue does not concern the economic analysis but highlights the potential economic significance of different criteria of proof applied by courts. See T. M. Jorde & M. A. Lemley, Summary Judgment in Antitrust Cases: Understanding Monsanto and Matsushita, (36) Antitrust Bull. 271 (1991). (26) U.S. Supreme Court Minority Opinion in Matsushita. 106 S. Ct. at 1365. (27) See Hiroyuki Odagiri, Growth Through Competition, Competition Through Growth (1992); and Masahiko Aoki, Information, Incentives, and Bargaining in the Japanese Economy (1988). (28) Horace DePodwin cited by U.S. court of appeals in Matsushita. 106 S. Ct. at 1364. (29) Avinash Dixit, The Role of Investment in Entry Deterrence, 90 Econ. J. 95 (1980). (30) See also Alasdair Smith, Strategic Investment, Multinational Corporations, and Trade Policy, 31 European Econ. Rev. 89 (1987). (31) See for an overview of the literature: Milgrom & Roberts, supra note 3. (32) Scherer & Ross, supra note 3, at 470. (33) This is true even without any predation strategy: in the presence of learning effects, firms have an incentive to sell below marginal costs to increase sales early on, learn by doing, and profit later from lower marginal costs. See, e.g., Howard K. Gruenspecht, Dumping and Dynamic Competition, 25. J. Int'l Econ. 225 (1988). (34) See Blair et al., supra note 3 and Elzinga, supra note 3. (35) See also Ordover, Sykes & Willig, supra note 3 on this argument. (36) Another point of critique is that the scheme suggested by Blair et al. implies that firms with the largest sales in the U.S. would get most profits in Japan, while in reality this was rather the reverse: large sellers in the U.S. such as Sony and Sharp reaped relatively few profits in Japan. (37) See also Gene Gregory, Japanese Electronics Technology: Enterprise and Innovation (1986). (38) Although the avoidance of trade friction was in the collective interest of Japanese firms, individual firms had an incentive to cheat, at least as long as the check prices were above marginal costs. Since the arrangement was informal, MITI did not have a direct way to punish firms if it found evidence of cheating. Moreover, it is conceivable that MITI was not too preoccupied with punishing cheating firms as long as U.S. Customs did not find out about the discounts and so was not able to substantiate dumping. (39) The model presented here takes as the starting point the analyses in James Brander & Paul Krugman, A Reciprocal Dumping Model of International Trade, 15 J. Int'l Econ. 313 (1983) and Anthony Venables, Trade and Trade Policy with Imperfect Competition: The Case of Identical Products and Free Entry, 19 J. Int'l Econ. 1 (1985). Brander Krugman show that in oligopolistic industries dumping can be in accordance with normal profit-maximizing behavior if exporting firms have a smaller market share abroad than at home and so perceive a greater demand elasticity on export markets. In a symmetric two-country model a situation of "reciprocal dumping" can arise, with firms in both countries dumping their products in exports markets. Such dumping may be welfare enhancing by increasing competition if transport costs are not too high. Venables extends the analysis to long-term equilibrium by allowing profits to attract entry. We alter these models by introducing anticompetitive behavior and entry barriers in the Japanese market. In this case dumping becomes unilateral, while its procompetitive effects are limited. (40) This implies Japanese firms were not making strategic losses on exports. We note that, while price discrimination was indisputable, there has been no direct evidence of pricing below marginal costs. The fact that the institutional traits of Japanese firms such as lifetime employment systems tend to transfer variable costs into fixed costs, may have caused estimates of Japanese marginal costs to be systematically too high. Furthermore, any pricing below marginal costs that did occur, may have been "forward pricing" behavior that is consistent with profit maximization. (41) The full mathematical description of the model can be obtained from the authors upon request. (42) We assume that equilibrium is of the Nash-Cournot form: firms simultaneously set production quantities and prices adjust to clear the market. (43) The fixed cost can also be taken as constant at the manufacturing level as long as there is excess capacity on the home market, such that exports can be accommodated by an existing manufacturing plant. In the long term, the notion of a fixed outlay necessary for production implies that there have to be economies of scale in production with average costs declining as output increases. Since CTV manufacturing is characterized by economies of scale with a minimum efficient plant size of 500,000 units, Japanese firms can only profit from economies of scale in manufacturing by exporting to the U.S. as long as their sales in Japan are lower than this minimum plant size. This, however, probably only applied to the early exporting experience of smaller firms such as Mitsubishi and Sharp. We see the assumption of a fixed cost more as representing economies of scope in R&D. (44) See Paul Krugman, Import Protection as Export Promotion: International Competition in the Presence of Oligopoly and Economies of Scale, in Monopolistic Competition and International Trade 180-93 (Henryk Kierzkowski, ed., 1984). (45) There are at least two recorded efforts by U.S. CTV manufacturers to sell CTVs in Japan through Japanese firms. In Detrouzos, Kester Solow, Made in America: Regaining the Productive Edge (1989), it is reported that Zenith tried to sell CTVs in Japan through C. Itoh, one of the general trading companies, in 1962. MITI, however, refused to grant C. Itoh the right to use dollars to import the sets. In 1973, Motorola tried to sell its Quasar CTVS through Aiwa, a subsidiary of Sony at prices a third lower than Japanese sets. The following year Matsushita bought Motorola's CTV operations and imports apparently stopped. See also Schwartzman, supra note 3, at ch. 2 for a discussion of barriers to entry on the Japanese market U.S. firms have been facing. (46) Krugman, supra note 44. (47) See Milgrom & Roberts, supra note 3, and J. McGee, Predatory Pricing: The Standard Oil (N. J.) Case, 1 J. L. & Econ. 137 (1958). (48) We did not suppose Japanese firms were making losses by pricing below marginal costs. (49) Schwartzman, supra note 3, at ch. 4, argues that the Areeda-Turner principle and most analyses of predation pay inadequate attention to the international dimension. Nevertheless his conclusion that all export prices below home prices should be considered predatory goes much further than our argument. (50) In fact, in the case of Japan, which is still judged to be a country with very weak antitrust enforcement, the U.S. government has followed another route by urging it to toughen antitrust laws and enforcement. Sanekata, supra note 12, speaks of an antitrust "renaissance" in Japan startin, in the early 1990s. In 1991, the FTC increased penalties on cartels from 1.5% of sales made during the period of cartelization, to 6%. In the same year, for the first time in 17 years, the JFTC brought charges against the conspirers of a cartel under criminal law. (51) See Phedon Nicolaides, Predatory Behavior (Maastricht: European Institute of Public Administration 1992) and J. Michael Finger, Antidumpring How It Works and Who Gets Hurt (1992) at ch. 2. (52) Consumers and user industries have no role in antidumping procedures in the United States (Finger, supra note 51). The EC Commission is obliged to consider the "Community Interest" in deciding to levy duties but has in practice more or less equated the interests of the Community with the interests of the petitioning industries. See Ivo van Bael & Jean-Francois Bellis, Anti-Dumping and Other Trade Protection Laws of the EEC (Bicester: CCH Editions 1990) and Alasdair Bell, Anti-Dumping Practice of the EEC: The Japanese Dimension, 2 Legal Issues of European Integration 1 (1987). (53) See, among others, Bell, supra note 52, van Bael & Bellis, Supra note 52; Finger, Supra note 51; and Richard Boltuck, Joseph F. Francois Seth Kaplan, The Economic Implications of the Administration of the U.S. Unfair Trade Laws, in Down in the Dumps: Administration of the Unfair Trade Laws 152-99 (R. Boltuck, Richard & R. Litan, eds., 1992). (54) See Thomas J. Prusa, Why Are So Many Antidumping Petitions Withdrawn?, 33 J. Int'l Econ. (1992). (55) See Klaus Stegemann, EC Anti-Dumping Policy: Are Price Undertakings a Legal Substitute for Illegal Price Fixing?, Weltwirtschafliches Archiv 268 (1992). (56) See Klaus Stegemann, Settlement of Anti-Dumping Cases by Price Undertakings: Is the EC More Liberal Than Canada?, in Policy Implications of Anti-Dumping Measures (Paul Tharakan, ed., 1991) and Patrick A. Messerlin, Anti-Dumping Regulations or Pro-Cartel Law? The EC Chemical Cases, 13 World Economy 465 (1990).
Rene Belderbos Senior Lecturer in Economics, School of European Studies and Sussex European Institute, University of Sussex.
Peter Holmes Associate Professor of Economics, Institute for Economic Research, Hitotsubashi University, Tokyo and Honorary Fellow, Science Policy Research Unit, University of Sussex.
AUTHORS' NOTE: The authors are grateful for research assistance by Su Geng and comments from Alan Cawson, Mike Hobday, Motoshige Itoh, Dan Kovenock, Mitsuo Matsushita, Leo Sleuwaegen, Alasdair Smith, Kotaro Suzumura, Friedl Weiss, and the anonymous referees. Reni Belderbos received support from the Japan-German Berlin Centre, and the Tinbergen Institute at Erasmus University Rotterdam; Peter Holmes' research has been funded by the UK Economic and Social Research Council European Initiative grant L113251007.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||international antitrust laws|
|Author:||Belderbos, Rene; Holmes, Peter|
|Date:||Dec 22, 1995|
|Previous Article:||Commercial bribery and the antitrust laws.|
|Next Article:||Overlap and antitrust: fixing prices in a smoke-filled classroom.|