The economics of trips: International trade in information-intensive products
Publication: The George Washington International Law Review
Date: Monday, January 1 2001
The Agreement on the Trade Related Aspects of Intellectual Property Rights' (TRIPs) has become highly controversial. Although trade in the most directly affected products is still a small portion of overall international trade, it is becoming more and more important. U.S. pharmaceutical firms sell
Solid numbers in this area are extremely difficult to obtain, in part because data collection systems are more oriented toward trade in goods. It is also partly due to definitional problems. For example, the "real royalty" for a sale of a pharmaceutical in a foreign nation may appear as a royalty from the foreign firm that sold it, or it may appear as a higher price on the exported product itself. Economically, these two cases are similar but there is a significant difference in balance of payments accounting. Nonetheless, the flow of royalties and payments is significant-and strongly favors the United States, which received over $36 billion in royalties in 1998 and paid out less than a third of that. 6 Moreover, international trade in these areas affects fundamental values including cultural sensitivities and human health.
The World Trade Organization (WTO) has taken jurisdiction over these products, here called information-intensive products (liPs), through the Uruguay Round's General Agreement on Trade in Services,7 as well as through TRIPs. The debate over TRIPs has been unusually sharp, ranging from those who view the failure to observe U.S.-style intellectual property arrangements as piracy8 to those who see TRIPs as a form of neo-imperialism.9
As a result of TRIPs, the WTO panel process is now available for disputes in the area of IIPs. This process allows for resolution of international trade disputes by an expert panel that interprets the relevant treaty documents-its decisions are enforceable through trade sanctions.10 In 1999, there were sixteen WTO dispute settlement filings based on TRIPs, amounting to ten percent of all filings before the WTO panel.ll The United States brought eleven of these sixteen filings.12
Integrating trade law and intellectual property law creates problems because the economics of IIPs can be significantly different from that of the products traditionally the subject of international trade law. In general economic trade theory, we think of trade as equalizing prices near the world's lowest available marginal cost, which benefits the world's consumers in the form of lower prices.13 For IIPs, however, market prices are necessarily significantly different from marginal cost. Intellectual property protection exists precisely to maintain that difference, as an incentive for innovation and creativity.
At the same time, technology is a crucial accelerator to the global economy. It is responsible for at least one to two percent per year of the long-term growth in the U.S. economy,14 and is likely to have a comparable contribution to global growth. The advance of technology and authorship is supported by intellectual property law, which necessarily distorts market prices. However, a balance must be struck between encouraging future development and making the benefits of past development available at as low a price as possible.15
Consequently, as the WTO and the world move into this new area of regulation, it is important to examine the extent to which the non-zero-sum mutual benefit assumptions of traditional free trade theory are satisfied for IIPs. This paper attempts such an examination and identifies certain necessary and important modifications of the traditional free trade theory. The paper then explores the implications of this modified theory for several important WTO issue areas, including intellectual property protection itself, differential pricing, and market protection devices such as subsidies. Finally, this paper suggests ways in which the WTO might strengthen its supporting statistical and study programs to ensure that TRIPs actually leads to global benefit.
I. INFORMATION INTENSIVE PRODUCTS AND THEIR ECONOMICS
For the purposes of this paper, IIPs may be defined as products whose content or development requires a substantial intellectual expenditure in development or creation but can thereafter be reproduced at relatively low cost.16 The most obvious examples are movies, data, and software, whose creation is enormously expensive, but whose copying is essentially costless. In other examples, such as pharmaceuticals and seeds, the copying cost is not quite costless, but the research and initial creation costs are such an important part of the overall costs that the products are also considered information intensive.17 The ability to sell these products above marginal cost stimulates creativity and invention. Intellectual property rights in these products exist to protect the creator or inventor against third-party sales at marginal cost.18
To consider the economics of such products a new movie is a useful thinking example. The demand curve for a specific movie is a normal downward sloping curve, even though the curve is based on the assumption that there are already many other movies in the market. The producer has a certain amount of monopoly power because this movie is different from all other movies, and will attract some of its viewers on that basis.
When the movie goes onto the market, the production costs are sunk costs, and the cost of showing the movie to each additional viewer is small. If the producer must choose a single price, he or she might choose the price at which marginal cost equals marginal revenue. But this leaves profits on the table for the benefit of the viewers. The producer would rather maximize his or her profits through price discrimination. To accomplish this, the producer first shows the movie at a high-price first-run theater, followed by discount theaters, and on down to post-midnight TV showings and remaindered videocassettes. By doing so, the producer can achieve a higher overall return.
This higher overall return, however, does not simply take something away from the consumer for the benefit of the producers. By obtaining larger rents, more movie makers can afford to make better (or at least more expensive) movies and more movies. These two effects, better and more movies, deserve separate attention. First, the quality of a movie may correlate to some extent with the production values embodied in the movie, and thus the overall market for the movie. The production values include better scriptwriters, actors, actresses, sets, etc. According to economic theory, movie makers will seek to optimize their investment level, and arguably therefore the quality of the movie. In general, the investment level in a market will increase with the size of the market. Thus, higher returns for producers creates a larger movie market which will lead to the production of higher-investment (and higher quality) movies.
Likewise, the larger market will lead to more movies. But, as more and more movies come on the market, the demand curve for a movie moves down because each viewer can only watch so many each week. Ultimately, the rent available under the curve becomes too small for a producer to anticipate recovering costs, and the market will be in equilibrium with respect to the number of movies as well as with respect to the quality of each movie.19
This overall economics, and therefore the quality and number of movies available, is also likely to be shaped by marketing opportunities in third markets and by intellectual property rights. Of course, advertising that is used to encourage viewers to come to movies is an important cost factor and raises the actual cost to the viewer, probably without any increase in quality. But the movie may also be a form of advertising for third markets. Suppose the movie shows an airliner from an identified airline, in return for a payment from that airline? Suppose the movie can be coupled with advertising, in the theater, on TV, or on cassette? Suppose the movie shows an action figure that can be marketed separately? What is common here, and is seen in other forms of mass media including newspapers, television, and the internet, is that the cost of producing the central information-for which the viewer is willing to pay with at least his or her attention-can be shared with those who would like a share of that viewer's attention. These side payments or markets can help recover producer costs, and therefore contribute to a larger overall production of information goods.
A second significant issue is intellectual property protection. The analysis above assumes that the producer can recoup its costs in a predictable series of prestige theaters, neighborhood theaters, television runs and the like. If, however, the movie is easily reproduced and distributed by third-parties in taped or internet form, a portion of the movie maker's market will be lost from the lost rents. Realistically, the lost market share will only be a small portion, and only the portion that is available at lower prices. Nevertheless, the overall rent is decreased, and the existence and expansion of an uncompensated market will ultimately lead to a decrease in the quality and number of movies available.20 If intellectual property rights can increase the producer's return, the reverse should occur. It is, of course, not clear from the analysis just presented whether there is in fact too little or too much intellectual property protection. With less protection, there will be fewer new movies and more viewing of tapes and of older movies. With more protection there will be more new movies and probably higher salaries and dividends in the movie industry.
Much of the above analysis for movies is applicable to most IIPs, despite some differences from product to product. In the movie market, consumers are willing to watch many movies; one movie will not exclude others, although it will certainly compete with them. For many information products, this is not true-consumers generally want only one local daily newspaper or one computer operating system. In such markets, the declining cost characteristics may lead to the emergence of a dominant firm. When a firm has a great volume of business and the large investment cost is in producing the first copy, not the nth copy, that firm can price its product low enough that other firms are unlikely to be able to enter the market without facing a prohibitive entry cost. In the case of a computer operating system, there are network externalities as well. Consumers buy Windows because more application software runs on Windows, and more application software is written for Windows because more consumers have it installed on their computers. The fact that others run the system makes it more valuable for each consumer, so the demand curve may actually increase with the number of consumers rather than fall, as would be expected for a "normal" product.
The market dominance and network externality issues just described both create economic barriers to entry by new firms. In some cases, there may be additional entry barriers. There may be fixed channel capacity-shaped, for example, by the number of local television stations or by the type of internet interface demanded by an on-line cable internet supplier. In such cases, the firms controlling the access may be able to operate with a monopoly rent that is not driven down by the entry of alternate products. There can also be control over access to the related markets mentioned above-the movie maker with the strongest linkages to toy companies may be able to maintain an extra strong position in children's movies.21
II. IMPLICATIONS FOR TRADE
Based on this analysis of domestic information-intensive industries, it is possible to examine what happens when trade is opened. The movie industry continues to provide a basic analytic example, allowing one to consider the opening of trade between two nations, one with a substantial information industry sector and one with a much weaker information sector.22 It is, of course, unrealistic in today's world to think of a movie industry limited to one nation-firms are generally owned multinationally, use talent hired from around the world, film around the world, and market around the world. Yet, for the purposes of analysis, it is still possible to envision a hypothetical national industry marketing, at least initially, to the national market. To ensure that the model illuminates TRIPs, we will explore both the situation in which the nation with the weaker information sector has intellectual property rights and the situation in which it does not. The more symmetrical case in which both have effective intellectual property rights systems is the easier place to begin.
A. Trade with symmetrical intellectual p"erty rights
Suppose the United States and Canada have never traded movies with each other, but have evolved their own national movie industries, each serving only its own nation and not exporting to third nations. Then one day they open trade in movies. What will happen?
First of all, because the U.S. movie market is significantly larger than the Canadian, the demand curve is significantly higher in the United States than in Canada. It follows that the rent available to producers is greater in the United States, that the number of movies produced in the United States is also higher, and that the quality of the U.S. movies is higher (at least in terms of economically definable production values).
When trade is opened, both U.S. and Canadian film producers will see a larger market. And the consumers in each nation will have a greater variety of movies to choose from. Because of the greater number of movies, there is a chance that the prices of movies will fall somewhat, but this will not be at all similar to the strong trend toward price equalization that occurs with trade in "normal" products. It is important to point out that it is unnecessary for the theater prices to be the same in the different nations. Indeed, each firm will continue to price its movies against specific groups of consumers, possibly even attempting (at least on first-runs of movies) to use a Ramsey pricing of setting the prices different in the two national markets at levels proportional to the inverse of the demand elasticities in the two markets. Conceivably, however, the videocassette prices will equalize between the two nations. As with "normal" goods, they will almost certainly be at a price somewhere in between the initial Canadian and U.S. levels. Although there may be some trend toward price equalization toward the price previously in force in the lower-cost market, this trend is much weaker than for "normal" goods.
There are also important differences with respect to the net directions of trade and restructuring of the industry. The U.S. movies were made with higher production values-they will therefore, on net, be in a relatively stronger competitive position compared to Canadian movies in both Canada and the United States. Thus, exports of U.S. movies to Canada (as measured in numbers of viewings) will generally be greater than those of Canadian movies to the United States.23 With the U.S. exports to Canada being greater than those in the reverse direction, there will be a net flow of rents (in such form as royalties per-viewer fees, or per screening fees) from Canada to the United States.
Moreover, as the industries settle down, the production values of movies produced in either nation will increase, since the available markets, as viewed by each supplier, have increased, as has the level of competition. There will probably be fewer movies produced than were produced in the two nations combined before trade, because the suppliers are investing in higher production values for each movie.24 The consumer in each nation, however, will have the benefit of more and better movies, because the two markets combined will provide more movies than were available in either separate market. And, because of the net flow of royalties, and the U.S. industry's initial greater strength, it will probably end up relatively stronger during the consolidation to serve the new integrated market.
The features captured by this example are, with appropriate adjustments, paralleled in trade of other IIPs. First, domestic price discrimination is common for most IIPs, protected by some combination of the character of the product and of intellectual property rights. This price discrimination will likely continue, both within and between nations. For example, a touring orchestra will charge different prices for its performances in Europe and Latin America and for its performances in larger and smaller cities of each area, college textbooks will be offered in cheap editions in developing nations, and even drugs will be offered at different prices in different nations. And, since arbitrage is often difficult, international trade will have less tendency to equalize prices. Every producer will attempt to optimally price discriminate against the demand curves in a variety of different submarkets.
Second, trade will benefit consumers by offering a wider variety of products-more orchestras, more books, more pharmaceuticals. To the extent that this wider choice of products produces competition, there may be a decline in the overall prices. For example, before technology made trade possible, the Financial Times and the Wall Street Journal did not compete; today they do, and consumers have gained increased information and the benefit of competition in both product quality and price.
Third, the trade advantage generally goes to the firms from the nations with the initially stronger industry. Put most generally, in those industries marked by large research or information development costs, the effective cost per unit declines with increases in the level of production. The firm with the initially larger market is therefore able to invest more in product development or to market its product at a lower price per unit than its competitors. In the movie case, this is exemplified by the point that those firms are likely to have been making movies with greater per-movie investment. In the case of competing research-intensive products, such as cameras or computer applications software, the industry with the initially larger market is likely to have offered a larger range of products, appealing to more niches and submarkets. This industry is also likely to have been able to invest more substantially in the technology. In either case, the industry will do best as trade is opened.25 And in a product like computer operating systems, network externalities will strengthen this force. In the Windows example, unless barred by some form of protectionism or perhaps difficulties in modification to some linguistic systems or major technological change, it is likely to remain the world operating system. In many information contexts, vertical integration or the ability to enter into cost-recovery arrangements with third parties may give special market power to current leaders.
Thus, in contrast to normal products in which production costs are decisive, for IIPs, the industries with the comparative advantage are those that were initially stronger in terms of level of investment or variety of product. It must be recognized, of course, that these arguments are not deterministic: the unknown author may gain a world market, the theatrical worlds of different languages may remain isolated, or technological change may enable cable networks to displace traditional networks in news distribution. And, as between two pharmaceutical products that serve the same medical function, the less expensive or more effective will almost certainly take over the global market. But it is certain that there is a tendency that initial market strength-not marginal cost-is the basis of comparative advantage in trade in IIPs. National market size begets competitive prowess for the participants in that market.
B. Trade in which one nation lacks intellectual prop" -rights
The implications of intellectual property rights can now be explored, by assuming their absence in the previous scenario. Suppose we go back to movies and assume there is no intellectual property protection in Canada-at least none against unauthorized copying of completed movies.26 What will the industry be like in Canada, initially assuming autarchy? There will be some firms making movies whose income likely derives from subsidies and from a few first-line theaters, along with limited amounts from those who cannot afford VCRs. Canada's market, however, will be extremely shrunken because the producers will be unable to recoup income from many of the showings of their product. It follows that the Canadian industry will be significantly smaller than under autarchy with intellectual property rights.
Suppose trade to Canada is then opened. Now, through the unregulated Canadian market in videotapes of U.S. movies (a market called the gray market or parallel market in various contexts),27 Canadians will have effectively free access to U.S. movies. The U.S. firms will be uninterested in exporting officially to Canada, for they will see little or no economic return from such exports. Even the Canadian industry will see decreased incentive to produce for Canada, because the relatively weak market it already faced will be further shrunken by the U.S. videotape imports. On the other hand, the Canadian firms will see the U.S. market as quite appealing, and may seek to make movies for that market. Yet, the paying market is essentially the same as the old U.S. market, so there will presumably be consolidation between the U.S. and the Canadian firms as they seek to gain their shares of the U.S. market. In summary, the Canadian viewer benefits from the cheap videotapes; the U.S. firms are hurt by increased competition from Canadian firms; and the U.S. consumer will see no increase in the quality or quantity of the available movies, although some may begin to come from Canadian studios.
Thus, the effect of intellectual property rights is that Canadian consumers will see a significantly higher price, because they then have to pay much more for their videotapes. But their participation in the market encourages both Canadian and U.S. studios to invest in more and better movies. The consumers of both nations will benefit from this increased investment and competition. Finally, there is likely to be a net royalty flow from Canadian viewers to U.S. studios.
Again, the outlines of this analysis are applicable for any form of IIP. In the absence of intellectual property rights, there will be widespread local production and distribution of III's based on foreign information used without compensation. In fact, TRIPs was brought about largely due to the developing-world sales of entertainment material, computer software, and pharmaceuticals protected in the developed world. With intellectual property protection, these products will increase in price, raising costs for consumers and creating an international rent flow. The increased costs reflect a real decrease in consumer welfare and the rents are a harm to the nation paying for the product that it used to produce freely. These rents can be roughly estimated on the basis of the difference between marginal cost and sales price in the IIP. At the same time, this new source of payment may contribute to bringing a wider variety of products and innovation to the benefit of the customers. Whether it does depends on the magnitude of the market that they represent (with intellectual property), the magnitude of investment needed to create an IIP, and the extent to which rents are actually used in producing new IIPs.28
Developing world consumers are, for example, far more likely to benefit from focused private sector investment in books or in entertainment material than in pharmaceuticals.29 And, investment oriented toward this market may lead to benefits for consumers in other nations; producers will be able to invest more since they have a larger market against which to amortize their investment.30
III. IMPLICATIONS FOR THE PROTECTION OF INTELLECTUAL PROPERTY RIGHTS
The United States and other exporters of IIPs saw that they would benefit from higher global standards for intellectual property rights and sought such an agreement in the Uruguay Round. They were successful in achieving TRIPs, which sets significantly higher minimum standards for intellectual property protection than the various prior multilateral treaties. TRIPs arranges for the enforcement of these standards through the WTO panel process, which can impose trade sanctions against nations that do not comply with panel decisions.31 WTO panels have thus become the ultimate international authority on intellectual property standards.
There is a strong pragmatic argument for this evolution. The trade of IIPs is greatly growing in importance-for example, royalties received by U.S. firms have increased from roughly $8 billion in 1986 to roughly $37 billion in 1998.32 In addition, such trade brings many disputes of a type amenable to the WTO dispute settlement panel process. Further, the agreement clearly benefits several U.S. sectors, especially the audio-visual and pharmaceutical sectors, in a way parallel to that of other trade agreements. Finally, at least to these exporters, a failure to provide intellectual property protection appears very similar to a trade barrier.
As the above analysis shows, however, there is an important difference between a failure to provide intellectual property protection and a more traditional trade barrier. In the traditional trade barrier, such as a tariff erected by the European widget industry against cheap U.S. widget imports, the European consumer is clearly harmed. The United States, acting under the pressure of its widget exporters, will protest this barrier, and either negotiate its removal (presumably balanced politically by removing some U.S. trade barrier) or attack Europe before a WTO panel. Once the barrier is removed by negotiation or dispute settlement, the European consumer will get cheaper widgets, widget production will shift to the lower-cost U.S. producers, and the European widget industry will shrink. We can, under principles of free-trade theory, be quite confident that the world has moved to a Pareto-superior state in which, even for Europe viewed as an economy on its own, the benefits to consumers outweigh the costs to the European widget industry. The WTO dispute settlement process serves as an engine through which the achievement of free trade, as well as the concept of negotiating reciprocal reductions in trade barriers on a package basis, is more politically feasible.
The same analysis can next be considered in the case of an IIP, such as a pharmaceutical whose European patent was, according to the U.S. firm, not enforced. If the patent is enforced in response to trade negotiations or a WTO panel decision, the price of the product will go up in Europe and the European consumer will pay more. This result is the reverse of the widget tariff case. The U.S. industry gains additional rent, and therefore, increased incentive to innovate and invest in research and development.
Free trade theory tells us that, for "normal" products, the world, as well as each nation, will be better off with free trade. Under the theory, free trade is best because the price of goods will equalize closest to their lowest marginal cost, minimizing consumer cost. With IIPs, however, intellectual property rights increase sales prices away from the lowest marginal cost, creating an economic misallocation. But it is generally believed that intellectual property rights are necessary to stimulate innovation and creativity. Thus, in determining whether the world is better off with intellectual property rights, one must ask whether the intellectual property right produces a favorable trade-off between the short-term cost to consumers through higher prices and long-term benefit to consumers through increased innovation.33 In other words, the relevant question for determining global benefit when considering IIPs is whether the overall incentive benefits outweigh the price costs. This political question often depends on social evaluations of which types of IIPs should be encouraged. It can also depend on the extent to which the intellectual property rents are used for research or content creation rather than for marketing. Systematic differences in perspective between developed and developing nations may also exist, arising from differing discount rates in comparing present costs and future benefits.
There is a second question, however, as to what is best for each individual nation. With free trade, each nation is benefited by falling trade barriers because consumer prices are minimized. But this is not necessarily the case with intellectual property rights. In the pharmaceutical industry, the United States certainly benefits from free trade through increased exports and a larger pharmaceutical market that can lead to more pharmaceutical products. The imposition of intellectual property rights in Europe may also benefit the United States through increased consumer prices and royalty outflow. For the European, however, these are considered to be costs of imposing intellectual property rights because the importers must pay for the higher prices and royalties. The benefits, however, are increased incentives for the global industry to produce new pharmaceuticals that may benefit Europeans. The analysis here must consider two perspectives-the world and the European community. The world analysis compares the European price increases with the benefits of the increased research to the entire world. The European analysis compares the European price increases and the royalty outflow with the research benefits accruing to Europe alone.34 Through this analysis, it is quite possible for the European balance to be unfavorable, while the world balance is favorable.
It is clear then that the political dynamics of the international intellectual property system are different from those of the international trade system. The WTO panel process is institutionally adapted to resolving disputes in intellectual property, and the traditional WTO plaintiffs, generally the home nations of exporting industries, have a strong incentive to uphold international standards. But, one cannot be quite sure that those standards serve the global interest; and often they will not serve the local interest. However, even in the latter case, there may be a global benefit in avoiding free riders-the fewer the number of nations willing to pay for research or authorship, the fewer and lower-quality IIPs there will be.35
What does this mean for TRIPs and for dispute settlement panels enforcing it? First of all, it counsels care to ensure that the forms of intellectual property protection applied are actually beneficial. The issue is not whether the rents from an IIP will go to an entity that is politically powerful (i.e., software or movie industries) or to one that appears politically deserving (i.e., originators of traditional knowledge), but whether the long term benefits to innovation and authorship outweigh the costs to the world's consumers. This issue raises many questions which are the subject of substantial debate. What proportion of their rents is the movie or pharmaceutical industry using for innovative production or research? Should copyrights over software include protection against reverse engineering? Should database protection be enacted? The last two questions have been recent subjects of sharp domestic debate and of international trade discussion. In the software case, U.S. trade officials have sought to compel third nations to accept forms of intellectual property protection that have been rejected by U.S. courts.36 Such debates are important to consider in negotiating new international intellectual property agreements; and can be important to WTO panels to the extent that they are free to consider such arguments, as well as in interpreting TRIPs.37
At the same time, the free-rider point is solid. The world benefits from certain intellectual property protections that do not benefit specific nations. The integration of intellectual property issues into the WTO provides a mechanism for managing this free-rider problem and for compelling nations to do what is in the group interest if not in their individual interest were they acting alone. But if legitimacy and loyalty are to be maintained, the burdens must be reasonably balanced.
The real issue here is the global allocation of the authorship and innovation costs. For example, the costs of pharmaceutical research must be reasonably balanced across different economies, which means that the developed-world patient should pay a much larger per-capita share of those costs than does the developingworld patient. This must be done in the background of the definition and interpretation of provisions such as those governing compulsory license arrangements.311 It also affects the differential pricing analysis to be undertaken in the next section.39 If an importing nation fails to provide a form of intellectual property protection that is required by TRIPs, there is no reason not to adopt the standard measure. However, there is a real risk that the amount involved may be overestimated, as by assuming that all sales would be made at a world price or at a price comparable to that of the highest price sales in the particular nation. Clearly, reasonable principles must be applied in estimating the actual rent lost.
These are particularly important issues for developing nations. It seems unfair to ask the consumers of these nations to contribute to investments that are unlikely to benefit them. In addition, it is difficult to encourage investment for pharmaceuticals oriented specifically toward developing world needs because the great cost of developing a pharmaceutical is unlikely to be recoverable from developing world markets. Such investment may, however, be feasible with partial public-sector support. In some cases, product development may be undertaken with an expectation that costs will be recovered from the developed-world market. If there is developingcountry patent protection in such a case, the producer gets the additional benefit of the developing-world rents. Developing nations also often benefit from such pharmaceuticals, but there remains the question of whether they should contribute a share of the royalty and what that royalty should be.
Presumably, to the extent feasible, producers will also set Ramsey prices, in which the price is set inversely to the elasticity of the market. Thus, those who are willing to pay almost any price will do so, and those who are unlikely to buy without low prices will have the benefit of low prices. This seems fair for luxury products (such as some forms of entertainment), but may be unfair for pharmaceuticals.
IV. IMPLICATIONS FOR DIFFERENTIAL PRICING AND EXHAUSTION
The economic analysis presented above indicates that IIPs will be priced differently in different markets. This implicates two areas of WTO law: the law of exhaustion of intellectual property rights and the law of dumping. In addition, since trade in a product with little information content may be manipulated by use of intellectual property rights, it raises questions about achieving the gains of trade.
A. Exhaustion
As was implicit in some of the economic models examined above, intellectual property rights can be used to divide markets. The classical U.S. case, Boesch v. Graff,40 involved a lamp burner. The U.S. patent holder licensed the German counterpart patent to a German manufacturer, and then tried to bar imports from the German manufacturer.41 The Supreme Court upheld the bar. There is a principle (often called the "first sale doctrine" in the United States and called "exhaustion" internationally) under which the German manufacturer's sale of particular products "exhausted" their rights in these products under the German patent, which permitted the resale of the products within Germany without infringing the German patent.42 But it does not follow that the product can be resold in the United States without infringing the United States counterpart patent. Indeed, the Court said that the "[t]he sale of articles in the United States under a United States patent cannot be controlled by foreign laws."45 It thus permitted the U.S. patent holder to bar resale-the implication is that the imports can be prevented and a price differential can be maintained between the different markets.44
There is now debate as to whether such market differentiation should be permitted; this is the "exhaustion" issue.45 Assume that the intellectual property rights to a product are licensed to one firm for nation A, and to another for nation B. If the principle of "national exhaustion" is followed, a product sold in nation A may be freely resold there (because the intellectual property rights have been "exhausted"), but the product may not be sold in nation B, because those rights still remain. In contrast, under the principle of "global exhaustion," the product marketed in one nation can be freely sold anywhere, on the theory that all rights are exhausted. The future of this issue is still open under TRIPs, which includes an agreement to disagree on the issue.46
Clearly, national exhaustion attempts to give an IIP producer a larger rent, while global exhaustion attempts to facilitate trade and permit a flow of parallel imports. The use of a regional analogue of the global exhaustion principle has been especially strong within Europe, where there has been a strong desire to create a single European market as indivisible as possible, even by intellectual property barriers. This is reflected in a long line of cases holding, in effect, that it is a violation of European law to use intellectual property rights to divide the internal European market.47 In addition, in a very recent case, the European Commission was directed to consider applying its antitrust law to protect its consumers from a Microsoft effort to restrict arbitrage in French-language software, intended for the Canadian market and being resold into Europe.48
The result of the Uruguay Round agreement to disagree is that importing nations can decide whether they want the exporter's rights exhausted, and can thus decide whether to protect their market from a lower foreign market price. In other words, price discrimination is possible at the option of the higher-priced market. This was precisely one of the disputes between the United States and South Africa over the latter's effort to maintain a low price for certain HIV medications.49
There seems little question about the desirability of price discrimination for IIPs. It is essential to the successful development and marketing of these goods. With appropriate market segmentation, it can also benefit developing nations, where prices are generally likely to be lower. But the principles discussed here can be stretched too far, because intellectual property rights can potentially be used to prevent any trade in at least some non-IIP products. In one of the classic cases, U.S. firms sought to divide markets by placing copyrighted labels on products. Their efforts failed before the Supreme Court.50
Consider an example. Suppose the lower cost manufacturer of widgets is in the United States and the higher cost in Europe. When trade is opened, the U.S. firm will dominate the widget market and the price will fall in Europe, benefiting the European consumer. But, now, suppose that both the U.S. and European widgets are subject to intellectual property protection, perhaps a trademark or a copyrighted label, and license rights are manipulated to ensure that prices can be differentiated between the two markets.
The U.S. price will remain low, and the European price may be kept higher. There is a chance, of course, that competition between the widget manufacturers will lower the European price; but there is also a chance that parallel pricing by a small number of firms will leave the price above competitive levels. After all, the U.S. firm may be very comfortable with a higher markup for just a portion of the market; the European firm may still be able to make some profit at that price. In addition, third party arbitrage is prevented by use of the intellectual property rights. Clearly, these results are suboptimal.
So what is the right dividing line between the IIP and the nonIIP cases? How far should we allow price discrimination to go? Or put conversely, when do we want the gray market-the arbitrage market-that undercuts the seller's intended price discrimination? Generally, we want to allow price discrimination when we believe that the additional rents will serve a beneficial purpose, but in all other circumstances we want to permit arbitrage to provide the benefits of trade.51 For the IIPs analyzed in this paper, the key beneficial purpose is encouragement of the front-end research and authorship. Thus, the initial answer is that price discrimination should be allowed and the gray market discouraged in connection with the intellectual property rights that protect innovation and authorship-this generally includes the areas of patents and copyrights. From the perspective of this paper, the opposite conclusion follows for trademark protection because this right serves to demonstrate authenticity, which is not a large concern for gray market products. The economics are quite different for trademarks than for intellectual property rights that encourage innovation and authorship.52 The second part of the answer is that the market division should not be permitted where the information content is small compared to the other content of the product. The Supreme Court case noted above is correct-it should be possible to use copyright-based trade barriers to maintain price differences in computer programs, but not necessarily to maintain price differences in every product that is given a copyrighted label.
The analysis of this paper therefore supports price discrimination and opposes global exhaustion for IIPs. But, this principle should be applied only to goods that are genuinely information intensive-this can be defined in terms of the intellectual property right protected and the share of the product value that is constituted by information. For the remaining goods, unless other principles are relevant, global exhaustion is essential if the benefits of trade are to be realized.
Finally, we must consider a nation, especially a developing nation, that wants to permit parallel imports of IIPs to give its consumers a gray market providing the benefit of lower prices. Here, as in the previous section, the issue is the balance between the global interest in avoiding free riding (measured in terms of the difference in rents between the local market with intellectual property protection and no gray market imports and the local market with protection and gray market imports) and the fairness of alloeating this share of the cost of innovation or authorship to the particular nation. At least for pharmaceuticals, there is a strong argument for derogating from most-favored-nation principles to allow two or three global prices (as between developed and developing nations, with a possible middle income tier). In such an arrangement, gray markets would be wide open within each of the tiers, but the price differences between the tiers would be protected.
B. Dumping law
Anti-dumping law raises additional price discrimination issues. This body of law, although uniformly condemned by economists,53 is available in most nations to authorize the levying of an antidumping duty to protect local firms from "dumping."54 Dumping, in this context, is often envisioned as a form of price discrimination in which products are sold in a foreign market at a low price in order to prevent the emergence of competition, while fixed costs are recovered in the home market.55 In fact, the laws are frequently much broader, permitting levying of an anti-dumping duty against a sale made below "normal value," where "normal value" is defined in a way that ensures that a portion of fixed costs and a profit margin are included in the cost calculation; thus, making "normal value" some value above marginal cost.56 There are WTO provisions and an international WTO code governing these laws.
Although some IIPs could also be subject to anti-dumping efforts,57 the above analysis makes it clear that dumping principles should not be applied to information-intensive goods. As an example, consider the movie industry once again. Suppose the movie industry of nation A sets prices so low in nation B that nation B's movie makers are unable to compete. Since price discrimination is normal and reasonable with IIPs, arguments that a particular product is being marketed below "normal value" value are economically unsound.58 In the general economic context, economists see no harm in marketing products below "normal value," unless the price is actually below marginal cost. Pricing below marginal cost, where the pricing is so low that the only purpose can be to drive a competitor out of the market, is known as predatory pricing. But for IIPs, predatory pricing is extremely unlikely because the costs of innovation must be recovered in addition to the costs of production. Thus, it is only in the actual predatory situation that there should be a remedy for overly low pricing.59 But, note that, under the arguments developed above, a consumer should also sometimes have a remedy against artificially high prices maintained by intellectual property barriers to trade!
V. IMPLICATIONS FOR MARKET PROTECTION AND SUBSIDIES
The final set of trade issues to explore is that associated with protectionism and subsidies. One of the most important implications of the analysis above is that an industry from the larger market enters trade with a great advantage-the principle of comparative advantage becomes, in effect, one of comparative size.60 Nations without these advantages therefore seek to protect their IIP industries, especially where these industries have substantial cultural content. For example, Canada has attempted to protect its magazine industry against U.S. competitions and the European Union has attempted to restrict imported TV programs to a specific portion of air time.62 The latter protection was particularly controversial in the Uruguay Round and, depending on who is speaking, may or may not have led to an agreement to disagree.63 And, even where there is no direct protection, there is likely to be a subsidy. The classic example is the Boeing-Airbus battle, in which the United States and Europe, in different ways, subsidize the development of aircraft. Each fear that competition in this research-intensive product might otherwise lead to domination of the industry by a single firm.64 Subsidies to a nation's film industry are also common.65
Should protection in the IIP sector be treated any differently from protection in other sectors? Consider a hypothetical, and assume that Canada restricts the entry of U.S. movies (perhaps by a restriction on the number of screen hours of foreign films per year). This will raise the domestic demand curve for the Canadian industry from where it would have been with free trade in films, and that industry will make more films. Some of these films may market well outside Canada as well as in Canada. In this sense, protection works. But the costs are high. The Canadian consumer is deprived of choice, and industry outside Canada is deprived of a portion of market share that might encourage the creation of films of interest to all, including Canadians. We hesitate to authorize such protection because Canada may reasonably feel it more legitimate to protect against an industry that dominates as a result of its size than to protect against an industry that dominates because of its low costs.
On the other hand, we may feel more comfortable allowing the Canadian government to subsidize its movie industry. After all, the U.S. industry is already effectively subsidized by U.S. intellectual property rights, which give the industry a substantial market against which to recover the cost of making movies. In a real sense, intellectual property is nothing more than a subsidy dependent on the size of the appropriable market available to the product involved. Why should the nation with the smaller market not engage in a more explicit subsidy in response? The economic results are, in fact, the production of more IIPs (something generally good for the world) and greater variety for the consumer. Assuming that imports of U.S. films into Canada are permitted, the harms to the U.S. industry are likely to be relatively small-they will be the lost audience deriving from the increased quantity and production value of the subsidized Canadian movies. Given the character of the competition, this may encourage better U.S. movies, and, more generally, better lips. The judgement that subsidies should be permitted is strengthened by the fact that subsidies in this area may have positive spill-over effects that are not captured by the market. Although there are limits, it seems a reasonable use of public funds to subsidize research and the humanities. Further, there are important non-economic issues for certain IIPs, especially those related to culture. It is essential to human diversity that there be a variety of sources of entertainment, and it is essential to the advancement of technology that there be substantial subsidization of research.
There is, of course, a balance here, which need not necessarily be accomplished through permitting subsidies and prohibiting barriers. The economics imply that it is reasonable to provide some form of assistance to IIP industries at risk of being displaced by global behemoths, especially when the competitive basis of the strong industries is not that they are cheaper or better but that they are stronger. The choice of permitting subsidies and prohibiting protection appears reasonable. Both subsidies and protection somewhat weaken incentives for foreign industry, but subsidies give local and global consumers greater choice. However, they are also more difficult to impose politically because they require a budgetary expenditure, and are therefore less likely to be misused.
The balance suggested here implies that there should not be a cultural exemption permitting a nation to restrict IIP imports.66 On the other hand, at least some subsidies in the sector should be permitted; this means that the WTO agreement on subsidies67 should not prohibit them and that nations should not be allowed to attempt to nullify foreign subsidies through countervailing duties.68 Subsidies should be permitted without restriction by the international trade system especially when they support the production of goods that are culturally important and for which creation costs are a large component of the price (as is the case with software, movies, or books).
The use of research subsidies to produce goods in which the information content constitutes only a portion of the price raises a more difficult question, as does price discrimination for such goods. This has been an especially important issue in the competition between Boeing and Airbus, where the United States and Europe each argue that the other is giving its producer an unfair competitive advantage; the result has been specific agreement.69 There are other ways to face this issue, such as defining specific limits.70 A favorable aspect of subsidies is that they may allow the resulting scientific and technological developments to be immediately available for global application because they are not subject to restrictive intellectual property arrangements. In addition, subsidies may be more acceptable for less wealthy nations or nations with weak industries.
VI. OVERALL IMPLICATIONS FOR WORLD TRADE REGIME
Trade in intellectual-property based products is different from trade in "normal" products. It benefits consumers in both exporting and importing nations; but prices do not necessarily fall toward marginal cost. Moreover, the basis of comparative advantage is much more complex, and much more balanced in favor of those with existing market strength than for traditional products. The application of well-designed intellectual property protection systems in every nation is a clear benefit to the world, but the individual nation will not necessarily benefit. Therefore, the role of the global system is one of preventing free-riding, a role that requires thoughtful and fair allocation of the costs of creating the intellectual property.
The WTO is a reasonable home for this function. The WTO panel process can be applied reasonably and fairly. In addition, with appropriate adaptation, the various principles of setting standards for intellectual property, of managing the use of intellectual property rights to prevent trade, and of controlling market protection and subsidies can be integrated into the WTO.
In traditional trade, the WTO process is assisting nations in parallel moves towards a mutually beneficial goal of removing trade barriers. In intellectual property, in contrast, it is supervising a system which can provide global benefits, if the system is of the right strength. More intellectual property protection is not necessarily better. Moreover, the WTO is allocating the costs of creating intellectual property and avoiding free riders, which is very different from ensuring that the trade barriers agreed to be removed are actually removed.
Intellectual property is important to the future of the world and its encouragement necessarily involves trade; thus, the WTO involvement in TRIPs must be continued. This implicates many technical adaptations such as those discussed above. But, in addition, two new WTO activities-that might well be shared with the World Intellectual Property Organization (WIPO)-would be very useful in effectively integrating TRIPs:
(1) To collect statistics and conduct studies to help estimate whether specific international intellectual property regimes really are "working," in the sense that the net new research or authorship is really comparable to the net rents derived from the use of intellectual property law in different nations and its encouragement by TRIPs; and
(2) To collect statistics and conduct studies to estimate whether the cost of that protection is being reasonably allocated among different nations.
These are difficult tasks and may be possible only on a very approximate basis.71 Yet, the WTO is already an important compiler and distributor of trade statistics-and trade now includes intellectual property. The statistics suggested will help determine if the integration of TRIPs into the WTO was beneficial and will help us to make any necessary modifications. If they do show that the TRIPs system is globally justified, in the sense that reciprocal use of trade sanctions to avoid free riding really is in the global interest, they will contribute to the legitimacy of TRIPs in the global community.
FOOTNOTE1. Agreement on Trade-Related Aspects of Intellectual Property Rights, Apr. 15, 1994, Marrakesh Agreement Establishing the WTO, Annex IC, LEGAL INSTRUMENTS RESULTS OF THE URUGUAY ROUND VOl. 1 (1994) 33 LL.M. 81 [hereinafter TRIPS.
2. PhaRMA Annual Survey, Table 12, at http://www.phrma.org/publications/industry/profile00/phrma-Tables.pdf (last visited June 10, 2001).
FOOTNOTE3. Business Software Alliance, U.S.-Owned Software Industry Produces Trade Surpluses, at http://www.bsa.org/statistics/index99.html (last visited June 10, 2001) [hereinafter Trade Surpluses].
4. Compare U.S. Dep't of Commerce, Cross-Border Trade in 1998 and Sales Through Affilfates in 1997, 79 SuRv. CuRRENT Bus. 48, 48 & 64-65, tbl. I (Oct. 1999), at http:// www.ita.doc.gov/td/sif/USSTH-05.htm [hereinafter Gross-Border Trade], with Trade Surpluses, supra note 3, at Table F-I (overall U.S. exports are on the order of $1 trillion).
5. Trade Surpluses, supra note 3, at tbl. 1.
FOOTNOTE6. Id These royalties include everything from patent license fees to fees for copyrights and broadcasting rights to the right to distribute computer software.
7. TRIPs, supra note 1, art. 31, 33 LL.M. at 95.
8. See, eg., BENI:.DICTE CALLAN, PIRATES ON THE HIGH SEAS; THE UNITED STATES AND GLOBAL INTELLECTUAL PROPERTY RIGHTS 27 (1998) (discussing the failure to adopt more stringent intellectual property protections).
9. See, e.g., Keith Aoki, Neocolonialism, Anticommons Property, and Biopiracy in the (Not-SoBrave) New World Order of International Intellectual Property Protection, 6 IND. J. GLOBAL LEGAL STUD. 11 (1998) (noting the dangers of an overly aggressive intellectual property system).
FOOTNOTE10. Understanding on Rules and Procedures Governing the Settlement of Disputes, TRIPs, supra note 1, art. 31, 33 LL.M. at 95. See generally Matthijs Geuze & Hannu Wager, WTO Dispute Settlement Practice Relating to the TRIPs Agreement, 1 J. INT'L ECON. L. 347 (1998) (discussing WTO procedures for settling international disputes over intellectual property) [hereinafter WTO Dispute Settlement].
11. See WTO Dispute Settlement, supra note 10, at 348.
12. Id-; see also Canada, Term of Patent Protection, WT/DS170/R (May 5, 2000) (ruling on a complaint brought by the United States against Canada); Canada, Patent Protection of Pharmaceutical Products, WT/DSI 14/R (Mar. 17, 2000) (wherein the United States reserved third-party rights to intervene).
FOOTNOTE13. This is, of course, a great oversimplification. It ignores, for example, the roles of currency differences. Nevertheless, it is a reasonable comparison point for the analysis of this paper, which is, of necessity, more qualitative than quantitative.
14. See generally Robert Solow, Technological Change and the Aggregate Production Function, 39 REv. ECON. BC STAT. 312 (1957) (analyzing technological growth from 1909-1949); see also M. Boskin & L. Lau, Generalize Solow-neutral Technical Progress and Postwar Economic Growth (Dec. 2000), available at http://www.eee.nber.org/papers/w8023.
FOOTNOTE15. This paper explores the trade-intellectual property relationship under the assumption that intellectual property actually works as a monopoly creating an incentive toward innovation. That assumption is generally accurate for IIPS as defined in this paper. In many other areas, intellectual property rights may have a much more direct impact on industrial structure than on innovation. See also John Barton, Intellectual Property, Biotechnology, and International Trade: Two Examples, MICH. WoRLD TRADE F. (forthcoming 2000) (exploring the implications of the globalization of intellectual properties for industrial structure and suggesting antitrust responses to the more perverse of those implications); John Barton, New International Arrangements in Intellectual Property and Competition Law, Address Before the Swedish International Symposium on Economics, Law and Intellectual Property June 2000), (forthcoming 2001) [hereinafter New International Arrangements].
FOOTNOTE16. Although some of these goods are public goods, and the research or creation involved in any of them may be a public good, the liP definition is broader than the traditional public good definition. Public goods are typically defined in terms of "non-excludability," i.e., that no one can be excluded from use of the good, and "non-rivalry," meaning that one person's use of the good does not keep another from using it. These requirements are fulfilled with respect to the information contained in the IIP. Excludability may, however, be achieved through intellectual property protection, and non-rivalry may not be satisfied with respect to the practical distribution of the IIP (my purchase of a videotape keeps you from purchasing the same tape, albeit not from purchasing another tape).
FOOTNOTE17. There are also examples of products whose characteristics are somewhat less similar: computer chips containing software programs within automobiles or other products, and, at a further step, products like steel in which heavy fixed costs must be distributed over a product volume whose marginal production cost is relatively small. Yet, at least in the latter case, we do not provide an analogue of intellectual property protection; therefore, they will not be included as IIPs. As will be seen below, these products can present dilemmas for the international trade system, because intellectual property or intellectual property type arguments can be used to achieve market separation.
18. The intellectual property considered in this paper includes patent, copyright, trade secret, and a variety of more specific rights (e.g. semiconductor maskwork protection, database protection, and plant variety protection). They do not include trademark or geographic indications, which are covered in TRIPS, supra note 1, art. 31, 33 LL.M. at 95, but protect a completely different seller interest - an interest based on advertising and product reputation and service, rather than on initial product creation.
FOOTNOTE19. See, eg., Eli M. Noam, Media Americanization, National Culture, and Forces of Integra tion, in INTERNATIONAL MARKET IN FILM AND TELEVISION PROGRAMS 41, 41-52 (1988) (illustrating the number of films and the investment level per film are determined simultaneously and concluding that both number of films and investment per film vary roughly with the square root of the market size).
FOOTNOTE20. In the Napster litigation, arguments were made that unauthorized distribution provides a form of advertising for the authorized product, and that distribution from the artist to the public directly, even with weak payment mechanisms, brings the artists more than when there is a record studio intermediate. See A & M Records, Inc. v. Napster, Inc., No. C99-5183 MHP, 2000 WL 1009483, at *3 (N.D. Ca., July 26, 2000), aff'd in part, rev'd in part and remanded, 239 F.3d 1004 (9th Cir. 2001). Clearly, such arguments need to be taken into account should they prove economically significant.
FOOTNOTE21. Such linkages between particular content providers and distribution channels, however, can create very significant sources of market power, which could lead to antitrust concerns. See America Online & Time Warner Merger, FCC Dkt. # CS 00-30 (2000) (concerning the proposed merger between AOL and Time Warner); In re Eurovision [1999] 5 C.M.L.R. 1437 (1999) (regarding the allocation of sport television rights); In re Television par Satellite (TPS), [1999] 5 C.M.L.R. 168 (1999) (discussing the possibility of a joint venture for satellite pay TV).
FOOTNOTE22. Presumably, trade arising in the IIP sector will be balanced, at least in the long run, by reciprocal flows in other sectors. Trade in information products is important, but not to the extent that its analysis requires consideration of effects on the balance of payments or terms of trade.
FOOTNOTE23. See Steven Wildman & Stephen Siwek, The Economics of Trade in Recorded Media Products in a Multilingual World: Implications for National Media Policies, in INTERNATIONAL MARKET IN FILM AND TELEVISION PROGRAMS 13, 14 (Eli M. Noam & Joel C. Millonzi eds., 1993) (arguing that Anglo-American media products, and American products in particular, dominate trade flows in relationships between nations).
FOOTNOTE24. See id. (discussing the "square-root model," which states that the number and production values of the separate Canadian and U.S. industries were proportional, respectively, to the square roots of the sizes of the two markets. When the markets are integrated, the new levels will be proportional to the square roots of the sum - and mathematically, vC + vUS > v(C+US). This inequality can be proven by squaring each side.).
FOOTNOTE25. See, e.g., Paul Krugman, Scale Economies, Product Differentiation, and the Pattern of Trade, 70 Am. ECON. REv. 950 (1980) (discussing the fact that the economics of any form of inter-industry or declining-cost industry trade can be quite similar); see also David Greenaway & PA.M. Tharakan, Imperfect Competition, Adjustment Policy, and Commercial Policy, in IMPERFECT COMPETITION AND INTERNATIONAL TRADE 7 (David Greenaway & RICK Tharakan eds., 1986) (noting the role of specialization and of product variety in determining which of several competing nations will come out ahead).
FOOTNOTE26. With the increasing ease of Internet transmission in today's world, it may no longer be possible for copying to be prohibited in some nations and not others. Although this issue is important, it is separate from that considered in this paper, which attempts to explore the trade implications of intellectual property where that property system is reasonably effective.
FOOTNOTE27. See, eg., Jonathan Baker, The Gray Market in Video, Consumer Welfare, and Public Policy: An Economic Analysis, in INTERNATIONAL MARKET IN FILM AND TELEVISION PROGRAMS 83, 86-87 (Eli M. Noam &Joel C. Millonzi eds., 1993) (discussing the illicit importation and sale of trademarked or copyrighted materials, primarily with respect to video products) [hereinafter Gray Market].
FOOTNOTE28. See id. at 91-97 (developing a model to analyze this issue). Baker attempts to estimate the balance between the consumer benefits of lower prices and consumer harms deriving from the producers' decreased incentive to produce a quality product as the uncompensated market grows. See id. at 101-05. Although he argues that the consumer is usually, on net, harmed by the uncompensated market, the balance may not be quite so clear. It certainly, however, depends on the factors he identifies. See id. at 105. As the ratio between the official (intended) market price and the marginal cost of making a copy grows, the consumer is more likely to be favored by the gray market. See id. at 106. And, as the ratio between the producer's product development investment and the producer's sales increases, the consumer is more likely to be harmed by the gray market (because the loss of sales to that market is more clearly likely to reduce investment by the producer). See id. at 99.
FOOTNOTE29. Judith Chin and Gene Grossman analyze a closely related balance using a very different model, based on the incentive and pricing behavior of the members of a global duopoly, one firm in the developed world and one in the developing world. Judith C. Chin & Gene M. Grossman, Intellectual Property Rights and North-South Trade, in THE POLITICAL ECONOMY OF INTERNATIONAL TRADE: EssAYs IN HONOR OF ROBERT E. BALDWIN 90, 105-06 (Ronald W. Jones & Anne 0. Krueger eds., 1990). Under their model, the developed world firm considers whether to invest in research that will reduce production costs. They find that, under many realistic assumptions, the adoption of an intellectual property regime may hurt the developing world. The global benefits of an intellectual property system are likely to be greatest in highly innovative industries, i.e., those in which the cost savings associated with a specific amount of research are largest. Id. at 94-96.
FOOTNOTE30. Note that the impact on local research or authorship is unclear. Certainly, for the reasons discussed infra Part ILA, trade is likely to favor larger nations and thus, in the typical case of a developing nation applying intellectual property protection, hurt the local industry. But, in the context being analyzed in this paragraph, trade already exists, and the local industry is already suffering. If it attempts to operate on the basis of limited effective rights - a small movie industry that manages to survive on the basis of subsidies and first run showings, for example - it may benefit from the possible increased returns as intellectual property rights are applied. If it attempts to operate as a parallel market or generic industry on a basis that assumes that intellectual property rights are irrelevant - mass production of CD-ROMs of works copyrighted elsewhere or of pharmaceuticals patented else
FOOTNOTEwhere, for example - it must change its strategy through entering licensing arrangements based on its local production or distribution capabilities.
31. See CARLOS M. CORREA Bt ABDULQAWI A. YUSUF, INTELLECTUAL PROPERTY AND INTERNATIONAL TRADE: THE TRIPs AGREEMENT 88 (1998).
32. Cross-Border Trade, supra note 4, at 64, tbl. 1.
FOOTNOTE33. Note that different nations may have genuine reasons for different balances among these short and long-term factors. It was not unreasonable for India to decide, in 1970, that a low consumer cost for pharmaceuticals was more important than incentives to develop new pharmaceuticals. See The Patents Act, 1970, No. 39 (1970) (India).
FOOTNOTE34. If the European industry were included in the analysis, the benefits to it of the increased intellectual property protection would need to be taken into account as well. And in a full analysis of specific pharmaceutical cases, it would be necessary to take into account the fact that the European and U.S. industries are closely integrated in research and ownership - but even so, differential prices have significant welfare impact on the two regions.
FOOTNOTE35. But see Allan V. Deardorff, Should Patent Protection be Extended to All Developing Countries, in ROBERT M. STERN, THE MULTILATERAL TRADING SYSTEM; ANALYSIS AND OPTIONS FOR CHANGE 435 (1993). Deardorff suggests that, as one extends the intellectual property system to poorer and poorer nations, the consumer losses from the higher prices for products that would have been produced or developed anyway necessarily outweigh any incentive benefit. Id. at 440-43. This seems likely to be an empirical issue rather than a theoretical issue and, as suggested toward the end of this section, certainly seems likely to vary from sector to sector.
FOOTNOTE36. See Jonathan Band, Gunboat Diplomacy on the Pearl River. The Tortuous History of the Software Reverse Engineering Provisions of Hong Kong's New Copyright Bill, 15 COMPUTER LAw. 2, 8, 12 (1998).
37. Clearly, these panels must be bound by the language of the agreements before them. But there is certainly a difference in the context of interpretation. In traditional trade agreements (at least those governing the pure economics of trade), the language is generally chosen to support free trade insofar as politically feasible. The panel members know which of the alternative interpretations is more supportive of free trade, which they generally regard as a positive, but know also that the members had reasons for not going further. In the intellectual property area, they must recognize that the alternative readings represent alternative balances of the consumer and incentive aspects of intellectual property - and that we are usually much less confident of which balance is generally correct.
FOOTNOTE38. See, eg., TRIPs, supra note 1, art. 31, 33 LL.M. at 95.
39. There is also the issue of defining "damages" in the case of an intellectual-property associated trade dispute. The normal "damage" principles for WTO panels are based on the concept that a trade barrier restricts an export into the nation, and therefore deprives the exporting nation of the employment and profit benefits associated with producing those exports. A numerical estimate is made, where necessary, and the exporting nation is entitled to build trade barriers against a comparable amount of the exports of the nation not living up to WTO principles. Although economically sub-optimal - for the sanction pushes trade balances even further from those that would prevail in the absence of trade balances - this is an effective remedy for the purposes of encouraging compliance.
FOOTNOTE40. 133 U.S. 697 (1890). 41. 133 U.S. at 698-99. 42. 133 U.S. at 703.
FOOTNOTE43. 133 U.S. at 703. Note that the analogous bodies of law on trademarks and copyrights are not the same and can become quite complex. See, eg., Quality King Distrib., Inc. v. L'Anza Research Int'l, Inc., 523 U.S. 135 (1998) (holding that lawful owners of hair care products bearing copyrighted labels did not engage in copyright infringement by importing and reselling products without manufacturers' authority); KMart v. Carrier, Inc., 486 U.S. 281 (1988) (holding that foreign articles bearing a trademark identical with one owned by a United States citizen are subject to seizure as prohibited importations).
FOOTNOTE44. For a more complex form of market division, consider the case in which competitors license exclusive rights under all their patents to one firm in one market and to another firm in another market. See, e.g., United States v. Westinghouse Elec. Corp., 648 F.2d 642, 647 (9th Cir. 1981) (finding no violation where Westinghouse refused to grant Mitsubishi licenses in the United States and Canada because the Court found it was the patentee's "untrammeled right" to refuse to license a patent); United States v. Imperial Chem. Indus., 100 F. Supp. 504, 592 (S.D.N.Y. 1951) (finding a violation of the Sherman Antitrust Act where corporations engaged in the manufacture of explosives, exchanged patents and processes with the intent to divide the world market in explosives).
FOOTNOTE45. See, eg., Frederick M. Abbott, First Report (Final) to the Committee on International Trade Law of the International Law Association on the Subject of Parallel Importation, I J. INT'L ECON. L. 607 (1998) (discussing the costs and benefits of permitting intellectual property right holders to prevent parallel imports, and concluding that such barriers do not justify the potential impact on free trade).
46. Article 6 states, with certain exceptions not relevant to the examples in text, that "[f] or the purposes of dispute settlement under this Agreement .... nothing in this Agreement shall be used to address the issue of the exhaustion of intellectual property rights." TRIPs, supra note 1, art 6, 33 LL.M. at 86.
FOOTNOTE47. The Court's case law shows that Article 30 of the Treaty of Rome allows derogations from the fundamental principle of the free movement of goods within the common market only in so far as such derogations are justified in order to safeguard the rights which constitute the specified subject-matter of the industrial and commercial property in question. See, eg.,Joined Cases 427, 429/93 & 436/93, Bristol-Myers Squibb v. Paranova A/ S, 1996 E.C.R. 1-3457, [1997] 1 C.M.L.R. 1151, 1 42 (1996); Glaxo Group Ltd. v.
FOOTNOTEDowelhurst Ltd., [2000] 2 C.M.L.RP 571 (2000), 2000 WL 191144 (referring questions regarding the importation of trade-marked pharmaceuticals subsequently relabeled to comply with national legal regulatory requirements to the European Court of Justice), Case 15/74 Centrafarm BV v. Sterling Drug Inc., 1976 E.C.R. 1-1147, [1974] 2 C.M.L.R. 1, 6-8 (1974) (discussing key patent principles). See generally Jurg M. Ammann, Intellectual Property Rights and Parallel Imports, 26 LEGAL ISSUES EUR. INTEGRATION 91 (1999); Irini A. Stamatoudi & Paul L. Torremans, International Exhaustion in the European Union in the Light of `Zino Davidoff Contract Versus Trade Mark Law?, 31 INT'L REv. INDUS. PROP. COPYRIGHT L. 123 (2000) (arguing that international exhaustion rules need to be changed to stop improper infringement actions, price discrimination and unfair trade).
FOOTNOTE48. Case 198/98, Micro Leader Bus. v. Commission, 2000 All E.R. (EC) 361 (2000), [2000] 4 C.M.L.R. 886 (1999).
49. Emily Miao, TRIPS Agreement Impacts Pharmaceutical Sector, NAT'L LJ., at Cl 1, July 24, 2000.
50. See Quality King Distr., 523 U.S. at 152.
FOOTNOTE51. See generally Gray Market, supra note 27, at 91-97 (highlighting a model comparing the rents lost through the gray market and the consumer benefits of having those rents available for product or service improvement).
52. There are several relevant arguments that fall outside the scope of this paper. Market separation, for example, may be intended to protect a markup used to provide better customer service and repair in one nation. Alternatively, the products for the different markets may be different, albeit genuine. See, e.g., Original Appalachian Artworks v. Granada Electr., 816 F.2d 68 (2d Cir. 1987) (discussing Cabbage Patch dolls with "birth certificates" oriented toward a Spanish-language market). These interests are clearly relevant in some contexts, but are of a type different from those considered in this paper.
FOOTNOTE53. See, e.g., Alan V. Deardorff, Economic Perspectives on Antidumping Law, in ANTIDUMPINC L. PRAc. 23 (John H. Jackson & Edwin A. Vermulst eds., 1991) (asserting that economists believe dumping is harmless in the absence of "predation") [hereinafter Economic Perspectives].
54. See, e.g., Tariff Act of 1930 (Smoot-Hawley Act), 731ff (coded as amended at 19 U.S.C. 1673ff (2001)).
FOOTNOTE55. Presumably, some form of protectionist barrier is used to prevent the arbitrage that would even out the prices.
56. 19 U.S.C. 1677b. Note, in particular, 1677b(b), which ensures that the "normal value" will not be below a constructed amount, even if all actual sales in the home nation are below that amount.
57. See, e.g., Mary E. Footer & Christoph B. Graber, Trade Liberalization and Cultural Policy, 3J. INT'L ECON. LAw 115, 140 (2000) (maintaining that dumping occurs in the field of cultural goods, which includes films and television programs, because of the cost of supplying export markets at prices below those being charged in domestic markets instead of merely selling below cost) [hereinafter Trade Liberalization].
FOOTNOTE58. Note that industry A can use intellectual property rights to protect against the return flow of its own products, and thus to maintain the price differentiation. It cannot however protect itself against the flow of B's products, which can therefore have some competitive impact in nation A. Obviously, the analysis might be different if, as is emerging in some industries, there are patents covering the entire industry. See New International Arrangements, supra note 15 (forthcoming Summer 2001).
59. See Economic Perspectives, supra note 53, at 23-24, 35-36.
FOOTNOTE60. Note that antitrust principles may sometimes assist the weaker industries in protecting themselves against domination by the stronger industries. This is an extremely important issue, especially in the information industry (as distinguished from the pharmaceutical industry). It is subject, even without further international agreement, to certain WTO disciplines, in particular those of Article VIII of the General Agreement on Trade in Services, which requires nations to ensure that monopoly service providers do not abuse their position. See General Agreement on Trade In Services, April 15, 1994, art. VIII, reprinted in 33 LL.M. 1174, 75. The issues involved in this analysis diverge so fundamentally from those discussed in this paper that they are not included.
FOOTNOTE61. See, eg., Trade Liberalization, supra note 57, at 125 (discussing attempts by Canada in 1995 to introduce an excise tax on advertising in split-run magazines); see also Term of Patent Protection WTO Dispute Settlement Panel, Canada - Certain Measures Concerning Periodicals, AT/DS31/AB/R (June 20, 1997) (finding Canada's actions inconsistent with the provisions of GATT).
62. See, e.g., Trade Liberalization, supra note 57, at 122-26 (describing examples of a variety of national cultural policy measures, including subsidies, domestic content regulation, market access restriction, tax measures, intellectual property protections, restrictions on foreign investments and ownership, and film co-production agreements); see also JeanLuc Renaud, International Trade in Television Programs: Quota Polices and Consumer Choice Revisited, in INTERNATIONAL MARKET IN FILM AND TELEVISION PROGRAMS 151, 153-55 (Eli M. Noam &Joel C. Millonzi eds., 1993).
FOOTNOTE63. Compare Ivan Bernier, Cultural Goods and Services in International Trade Law, in The Culture/Trade Quandary: Canada's Policy Options (Dennis Browne ed. 1998); Sandrine Cahn & Daniel Schimmel, The Cultural Exception: Does it Exist in GATT and GATS Frameworks? How Does it Affect or is it Affected by the Agreement on TRIPS? 15 CARDozo ARTs & ENT. LJ. 281, 297 (1997).
FOOTNOTE64. Thomas L. Boeder & Gary J. Dorman, The Boeing/McDonnell Douglas Merger; The Economics, Antitrust Law and Politics of the Aerospace Industry, 45 AN=Rus-r BULL. 119, 131-33 (2000).
65. See generally Cahn & Schimmel, supra note 63 (discussing the effect of various international trade agreements on the film industries in Europe and the United States).
FOOTNOTE66. See Trade Liberalization, supra note 57, at 139-40. The economic analysis presented here is not relevant to the question of cultural imports viewed as politically threatening, offensive to a community, blasphemous, or obscene. These issues raise very different questions.
FOOTNOTE67. See Agreement on Subsidies and Countervailing Measures, available at http://www. wto.org/english/docs-e/legal_e/final-e.htm (last visited June 10, 2001) [hereinafter Agreement on Subsidies].
68. See Tariff Act of 1930 (Smoot-Hawley Act), 731ff (codified as amended at 19 U.S.C. 1673ff (1994)) (imposing a countervailing duty if an American industry would be "materially injured" or "materially retarded" by importation or sales of imported merchandise).
FOOTNOTE69. See Agreement Concerning the Application of the GATT Agreement on Trade in Civil Aircraft on Trade in Large Civil Aircraft, July 17, 1992, E.C.-U.S., State Dept. No. 92199, reprinted in 9 Int'l Trade Rep. (BNA) No. 30, at 1273 July 24, 1992).
70. See generally Agreement on Subsidies, supra note 67 (setting specific prohibitions on government subsidies by members and providing remedies for members injured by subsidies).
FOOTNOTE71. See genera/y WARWICK A. RoTHNIE, PARALLEL IMPORTS (1993) (discussing the difficulties in collecting data and producing studies that measure the effectiveness of international property regimes).
AUTHOR_AFFILIATIONJOHN H. BARTON*
AUTHOR_AFFILIATION* George E. Osborne Professor of Law, Stanford University, Stanford, CA 94305, jbarton@leland.stanford.edu. I want to extend a special thank you to Charles Buck, whose 1999 seminar paper, The International Motion Picture Marketplace: An Explanation of U.S. Dominance and Recommendations for Future Trade Agreements, led me to a number of sources and ideas on the economics of the international movie industry.
In addition, make sure to read these articles:
- Government Contracts for Small Businesses
- Registering as a Contractor with the Government
- What Government Forms Do I Need New Employees to Fill Out?
- Statement by IPI President Regarding Next Round of Trade-Related Aspects of Intellectual...
- House reprimands USTR for inaction on intellectual property protection.
- IACC Cites Countries for Failing to Protect Intellectual Property Rights.
- Intellectual property rights: who owns the knowledge?
- Trade-Related Intellectual Property Rights and Product versus Process Innovations.
- The Economics of Intellectual Property in a World without Frontiers: A Study of Computer...
- American Biophysics Corp. Awarded Intellectual Property Rights for Lentek `Mosquito Traps'.
- Callaway Golf Continues to Enforce Valuable Intellectual Property Rights.
- Shall the Northern optimal R&D subsidy rate inversely respond to Southern Intellectual...
- Intellectual Property, International Market Growth and Business Management Themes for Day...
Email
Digg It
del.icio.us
