- This chapter is adapted from an article published in 2002 in Pacific Economic Review 7(2):319-34. It was prepared for the joint National Bureau of Economic Research-International Seminar on International Trade conference, held June 4-5, 1999. We wish to thank Jonathan Eaton and Damien Neven for insightful comments and Robin Koenigsberg for research assistance.
Parallel imports are products that, once placed into circulation in one country by the owner of a trademark, copyright, or patent, are sold in a second country without the authorization of the rights holder in the second market. For example, imagine that an authorized distributor of computer software in Thailand sells copies locally at a wholesale price below the retail price existing in Japan. If permitted to do so, a parallel trader could transport the copies to Japan and make a profit net of tariffs and shipping costs. Such goods are produced legitimately under trademark and are not unauthorized knockoffs or pirated products. Trade in such goods exists largely to profit from arbitraging against differential prices set by trademark owners in various markets, once control over their distribution escapes the original rights holder.
The legal treatment of parallel imports varies widely across countries and stems from each jurisdiction's choice of territorial exhaustion of intellectual property rights (IPRs). Under international exhaustion, rights to control distribution Page 190 expire upon first sale anywhere, and parallel imports are permitted. Under national exhaustion, first sale within a nation exhausts internal distribution rights, but IPR holders may legally exclude parallel imports or exports. Finally, a policy of regional exhaustion permits parallel trade within a group of countries but not from outside the region.
Because IPRs are traditionally the province of national or territorial policy, policies on parallel imports have been the purview of each country. U.S. efforts to incorporate a global standard of precluding parallel trade into the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) in the World Trade Organization (WTO) failed to reach consensus. As written, TRIPS (article 6) preserved the standard of national discretion, which was an outcome favored by numerous developing countries and several wealthy nations that tend to be net importers of intellectual property, such as Australia and New Zealand (Maskus 2000). More recently, U.S. negotiators have required countries to ban parallel imports when entering into a bilateral preferential trade agreement with the United States.
Most formal economic analysis of parallel imports treats them as a channel for overcoming third-degree price discrimination across countries (Malueg and Schwartz 1994). In Malueg and Schwartz's model, which focuses on price differences at the retail level and ignores distribution issues, countries differ in demand elasticities for homogeneous goods. Segmented markets permit discrimination, whereas parallel imports establish a uniform international price. Global welfare effects are ambiguous and depend on the balance of consumer surplus created and destroyed. Moreover, some high-elasticity (low-demand) nations might be eliminated as export markets under uniform pricing. Informal literature discusses the problems that exist when parallel importers free ride on the marketing and service investments of authorized wholesalers (Barfield and Groombridge 1998; Chard and Mellor 1989).
We argue that these two explanations for parallel imports ignore the main reason for their existence. Such trade arises endogenously in response to attempts by IPR holders to establish vertical price control by setting varying wholesale prices in unsegmented markets. The bulk of parallel trade exists by virtue of procuring goods at the wholesale distributor level. We study the incentives for parallel imports by comparing a regime of national exhaustion (segmented markets) with one of international exhaustion (integrated markets), considering price discrimination as a special case. The welfare effects of restricting parallel trade are ambiguous but informative for the debate about whether such a policy makes sense at the global, national, or regional level. Our results show that there is a significant link between declining trade costs and the gains from parallel imports, with such trade becoming more likely to improve welfare as trade barriers are reduced. Thus, Page 191 permitting parallel imports may be most advantageous among countries in a regional trade agreement with declining trade costs, such as the Association of Southeast Asian Nations or Asia-Pacific Economic Cooperation.
In the next section, we present arguments in the policy debate over parallel imports and review some empirical evidence on its existence. In the third section, we describe the main results of a simple model of price setting within a vertical international distribution framework. This model establishes key welfare tradeoffs among three features of parallel imports: pro-competitive trade between oligopolistic markets, resource cost of cross-hauling goods, and inefficient vertical pricing. The model establishes some distinguishing empirical predictions, which we examine econometrically in the fourth section. Those results confirm indirectly that our model is descriptive of actual behavior on international pricing. We conclude in the final section.
There is active debate over the question of whether to establish a global ban on parallel imports or to maintain national policy discretion. Three arguments are made in favor of permitting parallel trade. One claim is that restrictions on such trade essentially act as nontariff barriers (NTBs) to goods that have escaped the control of IPR owners. Because these barriers partition markets, they both violate WTO proscriptions against NTBs and forgo consumer gains from market integration. As trade economists might put it, if international price differences exist because of manufacturers' attempts to set market-specific prices, the situation would be no different from price differences coming from other demand or supply characteristics.
A second argument is that parallel imports help prevent abusive price discrimination and collusive behavior based on private territorial restraints. In this sense, a policy of international exhaustion complements competition policy and limits the scope of IPRs (Abbott 1998). The claim that buttressing territorial restraints with restrictions against parallel imports could generate collusion is consistent with past evidence in the United States (Hilke 1988; Tarr 1985). A final objection is that government enforcement of territorial rights invites rent-seeking behavior.
Several arguments can be made in favor of prohibiting parallel imports. First, price discrimination can raise welfare under certain circumstances (Varian 1985). Banning parallel trade partitions markets and supports perfect discrimination (Malueg and Schwartz 1994). In contrast, parallel imports push the global economy toward uniform international pricing, subject to transport and marketing costs. Thus, consumers in economies with inelastic demand should face higher Page 192 prices under price discrimination than under uniform pricing. If such countries are not significant developers of intellectual property, they are made worse off by price discrimination.
Countries with high demand elasticities should face lower prices under price discrimination. In the presence of parallel trade, such countries might not be supplied by foreign IPR owners because local demand might be insufficient under uniform pricing (Malueg and Schwartz 1994). In this view, international exhaustion could lower the well-being of developing economies through higher prices and lower product availability. Despite this possibility, most developing economies prefer not to restrict parallel trade (Abbott 1998). This position reflects concerns that banning parallel imports would invite abusive behavior in their markets on the part of foreign rights holders. Furthermore, many nations see opportunities for being parallel exporters. Indeed, foreign restrictions on parallel imports are sometimes seen as backdoor attempts by industrial countries to close markets through implicit NTBs.
A second complaint is that firms engaged in parallel imports free ride on the investment, marketing, and service costs of authorized distributors. These distributors incur the costs of building their territorial markets through advertising and postsale service activities. Thus, they require protection from parallel traders who procure the same goods without incurring similar costs. In this view, restrictions on parallel imports are a natural component of the right of IPR proprietors to control vertical markets. Such restrictions may be pro-competitive, both through increasing interbrand competition and through providing incentives to build markets and provide services.
Efficient international distribution could require a strong vertical control within an enterprise, and private contracts may be inadequate for this purpose. Exclusive distribution rights make it easier to monitor marketing efforts and enforce product quality. However, it may be difficult in foreign markets to enforce private contractual provisions that prohibit sales outside the authorized distribution chain. In this view, restrictions on parallel trade complement the existence of exclusive territories.1
From this discussion, it follows that whether regulating parallel imports is beneficial or harmful is an empirical issue that depends on circumstances regarding demand parameters, market structure, and innovation. Thus, it is not...