PATENT RIGHTS AND INTERNATIONAL TECHNOLOGY TRANSFER THROUGH DIRECT INVESTMENT AND LICENSING
Keith E. Maskus, Department of Economics, University of Colorado at Boulder
Kamal Saggi, Department of Economics, Southern MethodistUniversity
Thitima Puttitanun, Department of Econonomics, San DiegoStateUniversity
Revised draft: 30 June, 2003
Paper prepared for the conference, "International Public Goods and the Transfer of Technology after TRIPS," DukeUniversityLawSchool, April 4-6, 2003.
Contact information: Maskus, Department of Economics, UCB 256, University of Colorado at Boulder, Boulder CO 80309-0256; Telephone 303-492-7588; Fax 303-492-8960;
- Introduction
A central concern among policymakers is to understand how the shift toward stronger protection of intellectual property may affect access of developing countries to advanced proprietary technologies from firms in developed countries. Developing countries, including least developed countries, place considerable hope in the power of foreign technology to improve the productivity and growth performances of their economies. Indeed, a key plank of the Doha Declaration calls on developed countries to find means of encouraging technology transfer to poor nations, as specified in TRIPS Article 66.2.
The extent to which international technology flows would increase depends importantly on the state of access to such information. Such access is determined by a variety of factors. Impediments may come from many sources in the recipient country, including weak domestic absorption capacities, poor infrastructures, restrictions on inward technology, trade, and investment flows, and inadequate regulatory systems. In this context, strengthening intellectual property rights (IPRs) could play a positive and important role in mitigating the costs such factors raise for investors and thereby expanding technology flows. It should be evident from this brief description, however, that simply strengthening IPRs alone cannot suffice to improve access significantly. Rather, the intellectual property regime needs to be buttressed by appropriate infrastructure, governance, and competition systems in order to be effective.
At the same time, however, patents can block technology transfers under certain circumstances. Firms may choose to withhold technological information from particular countries for competitive reasons, a strategy that is facilitated by globalized IPRs. The specter of anticompetitive deployment of patents and patent pools in order to discourage local firms from learning technologies through imitation and reverse engineering surely looms large in the context of weak competition enforcement in most developing economies. Thus, as ever in the area of IPRs, there is a balancing act to pursue in linking policy on technology protection to the needs of economic development.
Our aim in this paper is to make two contributions to this question. First, we provide an extensive discussion of the economic literature on this subject in order to bring out important themes for policymakers. Second, we add to the literature by positing a simple model in which a multinational firm can choose between transferring a technology abroad through either foreign direct investment (FDI) or licensing. The decision is affected in two ways by a tightening of patent protection. On the one hand, by raising the costs of imitation on the part of local firms, this policy change makes the firm more likely to engage in both forms of transfer. On the other hand, by reducing the relative fixed costs of reaching and enforcing licensing contracts, stronger protection should shift incentives toward that form of transfer and away from FDI. However, we identify a subtle interaction in these two impacts that implies a different response, depending on whether the firm resides in a low-innovation or a high-innovation sector. We also test this model with U.S. contract data and find good support for it.
- Economics of Access to Foreign Technology
We provide an overview of the central literature from economics on the relationship between intellectual property protection and access of enterprises in technological follower nations to foreign proprietary technology. This is a complex question that can only be highlighted here.[1]
A central point is that successful technology transfer is generally costly and complex. A complete transfer that takes place through formal channels involves the shift of codified knowledge (blueprints, formulas, management techniques, customer lists), tacit knowledge (know-how, information gained from experience), and contractual obligations (payments, territorial restrictions, conditions on use, profit-sharing, tax liabilities). Such channels include FDI, licensing, joint ventures, and various mixed forms. The costs of making such transfers vary among channels, as do the subsequent costs of production and performance monitoring. Indeed, these costs are often at the heart of decisions among these modes.
Technologies are also transferred through informal channels. Most evidently, much technological information flows across borders through international trade in advanced inputs, reverse engineering of technology embodied in goods and services, and replication of production processes in published patents. It is clear that much information spills across borders through such uncompensated channels, with a positive impact on local productivity.[2] Indeed, this productivity enhancement may be due at least as much to efforts of local firms to make improvements upon technologies from abroad.
Although copying of books, videos, and CDs receives most of the attention regarding conflicts over IPR protection, imitating most products is not straightforward.[3] Empirical evidence indicates that imitation is a costly activity for a wide range of high-technology goods, such as chemicals, drugs, electronics, and machinery. One study found that the costs of imitation average 65% of the costs of innovation (and very few products are below 20%).[4] However, these estimates are outdated and in such areas as pharmaceuticals and biotechnology imitation costs have declined relative to original innovation.[5]
The likelihood that firms in a particular country will have access to, and successfully absorb, technological information through either informal or formal channels, depends on characteristics of the country in which they are located, capacities of the firms themselves, and the nature of technologies in question.
Recipient Country Characteristics
In this context, the recipient nation’s ability to attract and absorb technology depends on a wide variety of factors. Most obviously, the size and expected growth of an economy’s market acts as a major incentive for inward licensing and FDI.[6] Besides the demand-side factor, large economies provide scope for plant-level scale economies, distribution economies, and agglomeration advantages.
The impacts of openness to trade and investment are more subtle. Liberal economies have greater access to imported goods and capital, with embodied technologies that may be learned and diffused through the economy. However, FDI, at least of a horizontal nature, is attracted by restraints on trade in order to benefit from protected markets. Vertical FDI seems more attracted to open economies because such investment is generally designed to produce exports. There is not much evidence available on the relationship between openness and licensing.
Labor skills are a central element, in terms of both the nature of goods traded and of capital and technology imported. Evidence suggests that direct investment generally seeks locations with an abundance of skilled and semi-skilled workers rather than economies in which low wages reflect lagging productivity.[7] A significant skill basis is important for absorbing know-how successfully and for deploying a work force that effectively complements modern technologies. While labor-intensive sectors are more likely to seek low-wage locations, even these cost advantages are attenuated by low productivity, high absenteeism, and the like.
Numerous studies demonstrate that adequate physical and telecommunications infrastructure and a significant supply of producer and financial services are positive determinants of inward FDI and technology contracts.[8] Political and economic stability are important also, as is a transparent governance structure. In this context, an important task for encouraging technology transfer is the development of local capacity for providing public and private services and for efficient public administration.
The role of geographic distance in limiting trade flows and FDI is well-established. Because of this factor, the ability of countries that are remote from major markets to achieve technological spillovers through trade and investment is severely limited.[9] Perhaps equally important is the technological distance at which a country is located from the information frontier. Countries are more likely to absorb technology and learn from it when they are engaging in innovation themselves. In this regard, policies regarding education, capital accumulation, R&D incentives, integration of research done in universities and public laboratories with private-sector commercialization, and other aspects of the innovation system can be significant.
An important reason that some degree of domestic innovation capacity is critical is that imitation, learning, and innovation are themselves dynamic and cumulative processes in which information is incrementally improved and diffused through competition.[10] In this context, the design of appropriate incentives for competition and small-scale innovation, in addition to successful absorption of foreign technology, is particularly important for developing countries.
Firm Characteristics
Evidence suggests that for technology transfer to succeed the recipient firm may need to undertake some R&D on its own.[11] The essential reason is that without an understanding of the research and experimentation process the transactions costs in acquiring and deploying the technology may be high and uncertain, making effective absorption infeasible. More directly, firms undertaking R&D presumably are better positioned to engage in effective reverse engineering and imitative production. Technology management is also a complex activity that improves with experience.[12]
Intellectual Property Rights and Technology Transfer
To provide focus for what would otherwise be an overly broad discussion, we focus on the theory and empirical evidence in the economics literature about the relationship between intellectual property protection in developing countries and flows of technology through FDI and licensing.
The theoretical literature has often investigated the effect of IPR enforcement on technology transfer and FDI in endogenous growth models.[13] Several of the papers are linked through their use of the quality-ladders model and the love-of-variety model developed extensively by Grossman and Helpman (1991). In a product-cycle model with exogenous rates of Northern innovation, Helpman (1993) showed that a decline in the amount of effort devoted to imitation by firms in developing countries would promote FDI to the South. By associating a reduction in imitation with stronger IPRs, Helpman provided the first detailed welfare analysis of patent enforcement in the South. He showed that a strengthening of IPR protection was not in the interest of developing countries and could in the long run reduce global innovation, thereby reducing global welfare. Lai (1998) extended Helpman’s model to allow for FDI and showed that both FDI and innovation would be encouraged if the South were to strengthen its IPR protection. The common weakness of both models is that stronger patent enforcement is modeled as an exogenous decline in the rate of imitation.
Yang and Maskus (2001a) studied the effects of Southern IPR enforcement on the rate of innovation in the North as well as on the extent of technology licensing undertaken by Northern firms. A key assumption in their model is that strengthened IPR enforcement would increase the licensor's share of profits and reduce the costs of enforcing licensing contracts, thereby making licensing more attractive. Thus, in their model, both innovation and licensing would rise as patent protection in the South becomes stronger.
Glass and Saggi (2002a) analyzed the implications of Southern IPR protection in a comprehensive product-cycle model of trade and FDI. In their model, Southern imitation targets both multinational firms producing in the South and purely Northern firms producing in the North. They treated stronger patents as an increase in imitation cost stemming, say, from stricter uniqueness requirements in the South. They found that FDI actually would fall with a strengthening of Southern IPR protection. This result arises because an increase in the cost of imitation would crowd out FDI through tighter Southern resource scarcity.
It is clear from this discussion that the theoretical literature does not give an unambiguous prediction regarding the effects of stronger Southern IPR protection on the extent of technology transfer through FDI and licensing. Empirical evidence provides some insights, though because much of it exists at the aggregate level and relies on survey evidence it should be treated cautiously. Surveys of U.S. multinational firms found that such firms are more willing to invest in countries with stronger IPR protection.[14] These studies also suggested that IPRs are important for location decisions in FDI, though the importance varies by type of investment. A further result was that perceived weakness of IPRs was significantly and negatively related to investment decisions across countries, suggesting that countries that strengthen their patent regimes could attract additional FDI inflows.
The preceding theoretical models suffer from a fundamental problem: either FDI or licensing is the only channel through which Northern firms are allowed to produce in the South. A more complete treatment of FDI requires that Northern firms be given options over how to transact technology through the market. For example, FDI could increase with IPR enforcement but this policy could more readily encourage licensing by lowering the risk of opportunism in market transactions. Studies that ignore the possibility of licensing (or joint ventures) are likely to overstate the effect of IPR enforcement on inward FDI.
In fact, a more subtle analysis may be needed. Increased IPR enforcement by a developing nation may indeed make it a more attractive location for production. However, the technologies transferred for that purpose might flow through licensing rather than FDI, so that the net effect on FDI is ambiguous. Later in this paper, ignoring the effect of increased IPR enforcement on the magnitude of technology transfer, we develop a simple model that focuses on the effect of IPR enforcement on the choice between licensing and FDI.
Using data for 1982 on U.S. exports and sales of overseas affiliates of U.S. firms, one study attempted to identify the cross-country determinants of both exports and sales of multinational affiliates of U.S. firms.[15] The most interesting finding was that U.S. firms export higher-than-expected volumes to their affiliates in countries that have weak IPR regimes. The author suggested that this result may reflect attempts by U.S. firms to limit technology leakage to their rivals abroad by confining production within the United States. This interpretation fits well with a central theme of this paper: multinational firms adjust their strategies to optimize against policies and market conditions they face in various host countries, casting doubt on the conclusions of empirical (or theoretical) work that treats FDI as given.
Empirical evidence indicates that the level of IPR protection in a country may affect the composition of FDI in two different ways.[16] First, in industries for which IPRs are crucial (pharmaceuticals, for example), firms may refrain from investing in countries with weak IPR protection. Second, regardless of the industry in question, multinationals are less likely to set up manufacturing and R&D facilities in countries with weak IPR regimes, and more likely to set up sales and marketing ventures because the latter run no risk of technology leakage. This finding is consistent with results in a careful econometric study that distinguishes between incentives for a multinational enterprise to transfer technology abroad within a firm and those for independent licensing outside the firm.[17] In countries where local imitation capacity is significant, multinational firms are more likely to shift from FDI to licensing as intellectual property rights are made more secure.
These studies present useful findings, but are unable to address perhaps the most central question of all. Does a country's IPR regime affect its economic growth through enhancing inward technology flows? Although there are several theoretical analyses of this question, empirical studies are scarce. One such study used cross-country data on patent protection, trade regime, and economic fundamentals.[18] The authors found that IPR protection, as measured by the degree of patent protection, is an important determinant of economic growth. Somewhat more interestingly, they found that a strengthening of IPR protection is more conducive for growth when it is accompanied by a liberal trade policy.
A possible interpretation of this finding is that, by increasing foreign competition, trade liberalization not only curtails monopoly power granted by patents but also ensures that such monopoly power is obtained only if the innovation is truly global. If firms in other countries can export freely to the domestic market and have better products or technologies, a domestic patent cannot provide strong monopoly power. Furthermore, trade liberalization itself can improve productivity.[19] Thus, the results suggest that IPR enforcement matters over and above trade orientation, and that they may have mutually reinforcing effects.