مقررات سهم بازار؟

In the 1950s and 60s, Japanese and US antitrust authorities occassionally used the degree of concentration to regulate industries. Does regulating firms based on their market shares make theoretical sense? We set up a simple duopoly model with stochastic R&D activities to evaluate market share regulation policy. On the one hand, market share regulation discourages the larger company's R&D investment and causes economic inefficiency. On the other hand, it facilitates the smaller company's survival, and prevents the larger company from monopolizing the market. We show that consumers tend to benefit from market share regulation. However, the social welfare including firms’ profits would be hurt if both firms are equally good at R&D innovation. Nonetheless, if the smaller firm can make innovations more efficiently, then protecting smaller firms through market share regulation can improve the social welfare. We relate our analysis to a case study of Asahi Brewery's introducing Asahi Super Dry to become the top market share company in the industry.

Traditionally, antitrust authorities have monitored the degree of concentration (or the market shares of firms) of an industry as an important measure of the market's competitiveness. In the US, the Department of Justice and the Federal Trade Commission monitored the degree of concentration of each industry to protect consumers against firms’ colluding in more concentrated industries. This view, that a high degree of concentration leads to collusion, is based on the old industrial organization theory (the “Harvard School” or “structuralists”) that proposes a framework to analyze industries—the Structure-Conduct-Performance Paradigm (the SCPP: see Bain, 1959 for details; Viscusi et al., 2005, pages 62–69, for a summary). The SCPP emphasizes the role of market structure (the degree of concentration, the condition for entry to the market, etc.), market conduct (pricing strategies, investment decision, etc.), which in turn determines market performance (efficiency, fairness, etc.). Thus, the theory presumes that the market structure of an industry determines the performance of a market. Bain (1959) provided a series of empirical evidence showing that in highly concentrated markets the profit ratios are high, and concluded that if the market structure is highly concentrated, collusions tend to occur, hurting the market's performance. The structuralists suggested that the antitrust authority should use structural regulation, that is, divide dominant firms in an industry if the market structure of the industry satisfies a certain set of conditions.1 In contrast, the Chicago School, especially Demsetz (1973), argued that the high profit ratios in highly concentrated markets may be caused by the cost efficiency of firms in the industry, and that having many smaller firms in the industry may result in inefficiency.2 Demsetz also cautioned that regulations based on the degree of concentration may cause efficiency damages in the market by discouraging firms’ R&D investments.3 Following the Neal report, in late 1960s and early 1970s, influenced by the structuralists’ view, the US Department of Justice and the Federal Trade Commission sued corporate giants such as IBM, ATT, Xerox, and Kelloggs based on the fact that these firms had exceedingly high market shares in their industries. However, these court battles revealed that using market shares as a measure of competitiveness of industries is not accurate and not very useful. After these court battles, the Chicago School's view defeated the Harvard structuralists’ view in practice, and the market shares of firms in industries per se are no longer regarded as an important measure in US antitrust policies. However, in some countries the antitrust authorities still seem to be concerned with the market shares of firms. Pressure may be put on to the dominant firm by the antitrust authority's explicitly referring to the country's antitrust law, or by implicit threats from the general public to firms that wield monopoly power in the markets. For example in Turkey, the antitrust division still gives guidance to dominant firms, although they no longer announce direct market share limits since the EU accession process has started. It appears that if the antitrust authority observes a firm with over 50% of the market share, they regulate the dominant firm. If the dominant firm's market share is 55%, then it is not allowed to have a contract with a retailer such that it sells only the dominant firm's products. However, if its market share is 35%, it is allowed to make such an exclusive contract.4 The antitrust division of Turkey has given high priority to complaints against firms with more than 50%, and it is known that the fines are harsher for these firms. In addition, if a firm has a market share more than 50%, it must ask the antitrust division for permission when purchasing another firm. In the Turkish mobile phone industry, the dominant firm, Turkcell, was under heavy pressure from the antitrust division until recently when its share reduces to 50%.5 However, are these structural regulations by the antitrust authorities ever useful? If firms know that they can be punished if their market shares exceed some threshold level, then they would try to keep their market shares below that level. Such response by a firm would cause inefficiency in resource allocation as long as the dominant firms are more cost efficient than other firms, as Demsetz (1973) pointed out.6 On the one hand, the dominant firms should be more cost-efficient than other firms, and forcing them to reduce their market shares by a market share regulation may cause resource misallocations. Moreover, even if a small firm exits the market, the resource that is used to pay the fixed operation cost could be saved.7 A market share regulation could do even more harm if we consider firms’ R&D activities. A successful R&D investment may improve the quality of product, or may reduce production cost through improvements in production process. In either case, it will increase the market share of the firm. It can be particularly harmful for society if the most cost efficient firm is discouraged from engaging R&D investment for the society.8 On the other hand, however, there is also a good reason to fear that the dominant firm will exercise monopoly power after the smaller firms exit the market. These acts must harm consumers quite a bit, and the dominant firm indeed may attempt a “predation” exercise by improving its technology through R&D investment in order to totally monopolize the market, raising market price. This predation practice through R&D investment can be considered a ”non-price monopolization practice” (Motta, 2004, page 454), but it is hard for the anti-trust authority to prove the motivation of investment. If structural regulation provides a temporary relief to a small firm that can grow into a competitive rival of the dominant firm through their R&D activities, preventing this non-price monopolization practice, then the increased competition may improve resource allocation. In this paper, we will investigate this possibility. We employ a model of an oligopolistic market with stochastic R&D and fixed costs of operations to evaluate the welfare effects of regulation on R&D decisions. Although the concentration measure of an industry is no longer considered significant in Japan's antitrust policies, at one time the structuralists’ principle was more thoroughly applied there than in the US. In the post-war Japan, right after the World War II, under the rule of the US military, many zaibatsu conglomerates were divided into many pieces, and some dominant firms were also divided into smaller companies after being cited as the monopolists with very high market shares. In the next subsection, we will look in detail into episodes that the Japanese beer industry went through under a market share regulation policy conducted by the Japanese government. This case study appears to provide a best supporting case for a market share regulation policy and will highlight the possible pros and cons of a structural regulation policy. 1.1. A case study: history of the Japanese beer industry After World War II, the Allied Powers General Headquarters (GHQ) demilitarized Japanese society, democratized the political process, and decentralized the wealth and power in the first phase of the military occupation of Japan between 1945 and 1947. In decentralizing wealth and power, the GHQ engaged in breaking up Japan's zaibatsu conglomerates, fostering the growth of labor unions and carrying out a rural land reform program. Subsequently, General MacArthur pressured the Japanese Congress to pass the Law for the Elimination of Excessive Concentrations of Economic Power, which authorized dismantling any company that so dominated a particular market that potential newcomers were unlikely to survive ( McClain, 2002, Chapter 15). Enjoying 75% share of the market, the Dai-Nippon Brewery was divided into Asahi and Nippon (later Sapporo) Breweries in 1949. Due to dysfunctional organization and unnecessary rivalries between the two newly created companies, Asahi and Sapporo lost their market shares over the years to a third smaller company Kirin Brewery, and Kirin rapidly became the leading company. In 1973, when Congress proposed an amendment of the Antitrust law to give Kousei Torihiki Iinkai (the Japan Fair Trade Commission) the power to divide monopolistic companies, Kirin Brewery had a very good reason to be afraid of being divided into smaller companies by observing Dai Nippon Brewery's fate and Kirin's own success. 9 From then on, Kirin Brewery stopped advertising their products completely for a few years, and tried not to expand their market share. 10 With this effort, Kirin's market share stayed around the low 60s for 15 years. In this period, beer companies started to provide a new variety of beers (Japanese beer was rather homogeneous for many years until then). In 1987, Asahi Brewery introduced Asahi Super Dry to the market, which was an instant huge success: the market share of Asahi Brewery steadily increased from 10% to 45% over the next decade, surpassing Kirin's market share in the beer industry. 11 On the one hand, for the Asahi Brewery, the implicit market share regulation gave them a chance to conduct R&D to develop its new products, and they were lucky enough to win the gamble. So, it may appear that the market share regulation helped to improve efficiency of the market. On the other hand, however, Kirin Brewery's investment was perhaps discouraged by this policy. If there were no market share regulation, then Kirin might have succeeded in innovation, and this success might have greatly improved efficiency. Of course, it also could have led to Kirin's monopolization of the Japanese beer industry, but the efficiency improvement from Kirin's low-cost production could surpass the loss from the monopoly pricing (Table 1).1.2. The rest of the paper In the following, we present our simple duopoly model with heterogeneous marginal costs of production (a dominant low-cost firm and a small high-cost firm) and stochastic cost-reducing R&D.12 In the presence of fixed cost in operation, we assume that the smaller firm might exit dependent on parameter values. With this model, we will try to explain the experience of the Japanese beer industry. When the small firm does not exit the market, we first show that the dominant firm has more incentive to invest in R&D than the small firm,13 and identify the set of parameter values in which only the dominant firm invest in R&D (Proposition 1). With the possibility of the small firm's exit, the incentive of R&D is further encouraged for the dominant firm, and is further discouraged for the small firm (Proposition 2). Then, we identify the set of parameter values in which (i) the dominant firm only invests without market share regulation, while (ii) the smaller firm only invests with market share regulation (Proposition 3). We provide a numerical example that satisfies the parameter restrictions in Propositions 2 and 3 (Proposition 4 and Example 1). Then, we conduct welfare analysis: we evaluate market share regulation by consumer price and social welfare (total surplus). Although the expected consumer price is lower under market share regulation, (expected) social welfare, taking firms’ fixed and production costs into considerations, is always harmed by adopting market share regulation, assuming that firms are symmetric in the technology of R&D innovation (Propositions 5 and 6). This is because it is more beneficial for society for the dominant firm to succeed in R&D than for the small firm to succeed in R&D, given symmetric R&D technology. That is, without market share regulation, Kirin Brewery may succeed in R&D resulting in Kirin's monopoly, but social welfare is improved more than it is with Asahi's success in R&D. This is partly because Asahi's fixed cost of operation is saved under Kirin's monopoly.