Introduction*
Multinational corporations (“MNCs”) operate today in an increasingly open global trade environment. While tariff barriers have collapsed dramatically, several states and numerous scholars have raised concerns that the benefits of trade liberalization are undermined by various non-tariff barriers (“NTBs”) to trade, including the anticompetitive business practices of private enterprise. As a result, demands to link trade and antitrust policies more closely by extending the coverage of the World Trade Organization (“WTO”) to incorporate antitrust law have gathered momentum over the last decade.
Most advocates of a WTO antitrust agreement base their normative claims on largely intuitive assumptions about the necessity or desirability of international rules. The existing literature contains few examinations of the strategic situation that characterizes international antitrust cooperation and, as a result, has either completely ignored or largely mischaracterized the collective action problem that has impeded the efforts to negotiate any multinational antitrust rules. With the help of insights developed in game theory, this Article seeks to fill the gap in the current debate by analyzing the strategic interactions underlying states’ attempts to seek convergence of their antitrust laws. Understanding why attempts to generate formal international antitrust cooperation have thus far been unsuccessful is a critical prerequisite for designing a normatively desirable international antitrust regime. By offering a more accurate descriptive account for the failure to reach a binding international antitrust agreement, the Article can also be expected to inform the normative debate on optimal international antitrust governance.
The prevailing perception is that recent trade liberalization and the reduction of tariff barriers, which governments have historically employed to protect or promote their national interest, will induce states to make use of antitrust laws to pursue the same objectives. Thus, states are expected to apply domestic antitrust laws strategically to advance their national interest at the expense of global welfare. Andrew Guzman, for instance, has claimed that states attempt to “externalize the costs and internalize the benefits of the exercise of market power across borders” in an effort to maximize their national interest. According to this theory, states have the incentive to either under-enforce or over-enforce their antitrust laws depending on trade flows. Specifically, if a state is a net importer, it has an incentive to employ stricter antitrust standards than what would be globally optimal as it fails to internalize costs born by foreign producers. If a state is a net exporter, it has an incentive to enact laxer antitrust laws than it would in a closed economy, externalizing costs to foreign consumers. Guzman concludes that an international antitrust agreement is needed to overcome these sub-optimal domestic antitrust laws. Recognizing that such an agreement would be difficult to reach as net exporters and net importers disagree on the optimal content of an international antitrust regime, he argues that the negotiations ought to take place under the auspices of the WTO, which allows for the formation of strategic linkages across issue areas and hence the compensation of losing states by prospective winners. While there have been no systematic efforts to formalize the strategic situation characterizing international antitrust regulation in the existing literature, Guzman, among other commentators, seems to imply that the collective action problem underlying international antitrust resembles a Prisoner’s Dilemma (“PD”), where the dominant strategy for each player is to behave opportunistically and deviate from the mutually optimal policy. While sharing Guzman’s assumptions about states as rational actors who seek to maximize their national interest, this Article contests the extent to which domestic antitrust laws and enforcement are characterized by national bias. Instead of deliberately deviating from the mutually optimal antitrust policy, states appear to engage in under- and over-enforcement of domestic antitrust laws largely for less opportunistic reasons and would be willing to align their antitrust policies but either have found it too costly or have not yet managed to agree on the precise content of cooperation.
This Article argues that a binding international agreement on antitrust has been difficult to reach as states find themselves facing one of two distinct game theoretic situations that inhibit effective cooperation. First, cooperation has at times failed because states have perceived the political and economic costs to exceed the expected benefits of cooperation. In these “Deadlock” situations, pursuing the convergence of domestic antitrust regimes has not represented a Pareto-superior alternative and has thus failed on the merits. Retaining the status quo has therefore been the single rational and optimal equilibrium of the game.
Second, in situations where states have perceived cooperation to Paretodominate non-cooperation, the pursuit of binding cooperation has been obstructed by the simultaneous existence of a distributional problem and an informational problem. The distributional problem arises as states assume that the costs and the benefits of an international antitrust agreement would be unevenly distributed among them. The informational problem arises as states are unable to predict the value of various solutions and have difficulties ex ante identifying which country would fare better under an international agreement and, therefore, who should compensate whom and by how much. This type of uncertainty regarding the magnitude and the division of payoffs among states is further complicated by uncertainty at the domestic level. Costs and benefits arising for domestic actors from an international antitrust agreement are likely to be diffuse, case-specific, and exceedingly difficult toforecast. When the distributional consequences of the agreement are uncertain, the status quo is likely to persist. These latter interactions are modeled as a Coordination Game with Distributional Consequences (“CGDC”), where the game is modified to include the informational problem.
While the WTO could conceivably solve either the distributional or the informational problem in isolation, the coexistence of the two problems exacerbates the cooperation dilemma and increases the likelihood that cooperation will fail. The distributional problem creates an incentive for states to dissemble or misrepresent information in the hope of obtaining a more favorable solution. The distributional conflict thus thwarts the honest sharing of information, which again would be necessary to mitigate the informational problem. With the persisting uncertainty regarding the distribution of gains and losses under the agreement, it has been difficult to devise reciprocal commitments and rely on cross-issue bargaining in the WTO, removing the essential foundation on which the argument for linking trade and antitrust rests.
Conceptualizing the strategic structure of state interaction predominantly as a Deadlock or a CGDC has important policy implications. First, the Deadlock situations call into question the rationale for any binding international antitrust agreement, challenging the prevailing presumption that the pursuit of international convergence of domestic antitrust laws would at all times be desirable. Second, the CGDC situations suggest that international coordination of antitrust policies, when desirable, is unlikely to necessitate extensive enforcement provisions in the majority of antitrust issues. Once the agreement has been reached in a CGDC setting, it is largely self-enforcing as neither party has an incentive to deviate from it. Accordingly, the case for incorporating antitrust into the WTO seems less compelling as the organization’s ability to facilitate linkages is contested and its capacity to enforce compliance is not called for by the underlying strategic structure of the game. With uncertain benefits to any binding multilateral antitrust regime and little advantage in using the WTO to reach or enforce an agreement, this framework explains why such negotiations have thus far failed to show any meaningful progress and why states have resorted, and will likely continue to resort, to informal cooperation instead.
This Article proceeds as follows. Part I explains why the majority of the cooperation problems in the antitrust domain may more fruitfully be characterized as a Deadlock or a CGDC than as a PD. Part II discusses the difficulties in reaching an international agreement in a Deadlock situation. It disputes the assumption that cooperation always Pareto-dominates non-cooperation and describes how cooperation has at times failed as states have perceived the political and economic costs to exceed the expected benefits of cooperation. Part III examines the CGDC situation, showing that even when cooperation is perceived Pareto-optimal, it has at times been unsuccessful due to the simultaneous existence of the distributional problem and the informational problem relating to the consequences of the prospective agreement. Part IV discusses why the WTO has been unable to mitigate the cooperation problem even in situations where the benefits have been perceived to exceed the costs of cooperation. The conclusion summarizes the argument and outlines some possible normative implications that follow from the discussion. . . .
* This excerpt does not include citations. To read the entire article, including supporting notes, please download the PDF.