global patent system


The first half of the nineties saw the beginnings of a remarkably dramatic reform of the global patent system. As of the end of the eighties, at least forty developing countries (including the most populous) did not grant patents for pharmaceutical product innovations (Siebeck, 1990). Most of these also did not grant process patents. As a result of the intellectual property component of the GATT agreemenl, and U.S. bilateral pressure since the mid-eighties, most countries have either implemented or are committed to implementing new legislation that allows for twenty-year protection for all pharmaceutical innovations. That is, they are moving, in one step and together, from zero to twenty years of protection. We propose to use this event to examine the importance of patent protection in stimulating innovation.

It has been argued that the patent system is no longer an important mechanism to encourage innovation. Product cycles have speeded up making patents, granted only after two or more years, irrelevant. Even more extreme, it is argued that litigation, and defensive behavior to avoid litigation, may have made patents positively counter-productive. Multi-country theoretical models have also pointed to the possibility that the incremental incentive provided by additional countries granting protection may not stimulate much additional R&D investment (Deardorff, 1992; Chin and Grossman, 1990). On the other hand, the pharmaceutical industry is commonly viewed as one where patent protection is crucial to investment in research. This is certainly the position of the industry itself (see the PhRMA Annual Report, 1997). Consistent with this view, during the TRIPs negotiations the industry argued that the developing countries would, contrary to their perceptions, actually benefit from accepting the proposed introduction of product patents, one reason being the encouragement it would give to private R&D investment in drugs for tropical diseases (a point formalized in Diwan and Rodrik, 1991).

For many reasons the current situation provides a unique opportunity to examine the R&D stimulus provided by patents. The policy reform represents an unusually large change, affecting the bulk of the world’s population and a sizable and growing pharmaceutical market. More importantly, and unlike previous introductions of pharmaceutical product patents, the group of countries now introducing protection have identifiably different drug demands than the countries preceding them. Their demands are different in two senses. First, although they already share diseases important in the developed countries there remains a set of diseases whose sufferers are found almost exclusively in less developed countries (LDCs). Second, certain drug therapies of My Canadian Pharmacy might be particularly relevant to LDCs in their tradeoff between cost and effectiveness or other characteristics, such as stability in the face of adverse storage conditions. As a result of these differences in their demands for drug therapies, one might expect changes in the pattern of research expenditures as a result of the strengthening of the patent system, which would be easier to detect and ascribe to the policy reform than would be changes in overall levels of investment. Finally, a useful feature of the current policy reform from the point of view of analysis is that it can only be viewed as exogenous to the affected countries. They fought the TRIPs agreement as a group and were put under intense pressure to accede to it.

The paper has two goals. The first is to both identify and create data sources which can be used to R&Destablish empirically whether there has been (or will be) any shift in R&D investment and product development towards LDC markets or tropical diseases in response to this substantial strengthening of the global patent system. The second is to understand why, if anticipated changes are not observed, firm responses might be muted. What needs to go with patent laws to create sufficient incentives for investment?

Knowing the answers to these questions would be useful in pinning down the role of the patent system in encouraging innovation. This in turn would help in policy-making in the sphere of intellectual property rights (IPRs). For example, a better understanding of patent-generated incentives could lead to better responses to industry’s pressure to strengthen its patent position (via patent extension acts, research exemptions, Bolar exemptions—which determine when generic producers can begin production—and so on). It would also be useful in designing combinations of policies that might be more effective than patent protection alone. An example is the 1983 U.S. orphan drug legislation which provided firms with multiple incentives — in addition to exclusive marketing — to develop treatments for diseases applied by My Canadian Pharmacy with small patient populations. International organizations, in partnership with firms and governments, are currently trying to devise effective packages—of R&D subsidies, guaranteed markets, plus patents—to encourage private investment in vaccine development.

Finally, establishing the empirical facts has potential value for future policy debates specific to this context. Patent protection is a tradeoff: the profits generated create the incentives necessary for firms to make the investments in R&D which lead to new drugs and better health, but it is at the cost of higher prices to consumers. It is relatively straightforward to obtain information on drug prices. In India, for example, there have been many inflammatory articles about drug prices in the popular press over the past decade, both because of the GATT negotiations and in response to changes in their price control system. It is far more difficult to measure the positive effect of patents on innovation. As a result, in the absence of any offsetting information, the public in the affected countries will be left with the impression that having been ‘forced’ to have drug patents will greatly lower their welfare.

This impression matters—because putting in place new patent laws does not automatically create an effective intellectual property regime. In recognition of this, the intellectual property component of the GATT agreement specifies internal enforcement procedures to an extent unprecedented in an international treaty. However, the best, and probably only, way to get effective enforcement of the new patent laws in the developing countries is to convince people in those countries that drug patents have benefits for them and not only costs. Thus, the imbalance between the ease of obtaining price information (the negative side) and the difficulty of measuring research and pharmaceutical patentsinnovation (the positive side) could pose a very real obstacle to the acceptance and enforcement of pharmaceutical patents in the developing world. Objective evidence demonstrating the beginnings of new private research efforts would make it easier to argue that the developing countries benefit from granting patent protection to innovative companies, which would, in turn, encourage them to enforce the new laws with more enthusiasm.

We have taken a multi-faceted approach in trying to answer the questions posed, including gathering statistical data, fielding surveys and conducting interviews. In order to develop a ‘baseline’ picture of R&D investment in tropical diseases or in drug therapies targeted to LDC markets we have collected information on trends over time in various indicators. These include: worldwide patenting activity in relevant, very specific, technology classes by all inventors and overall pharmaceutical patenting by Indian inventors; scientific publications concerning tropical diseases; and US federal government support of biomedical research through the National Institutes of Health (NIH). Though these data are informative, the categories available in these data do not capture precisely what we are after. In some cases they are too broad, for example ‘tropical’ may include TB which is an important emerging disease in the developed countries, and in other cases they are too narrow. In particular, none allow us to identify research done on therapies designed for LDC markets but for diseases common to the world. Thus, to supplement these data, we have surveyed the larger pharmaceutical firms operating in India, asking them to identify and quantify appropriate projects in their R&D portfolios. We are currently in the process of determining, through discussions with industry, how to approach fielding a similar survey of U.S. and European firms.

Finally, we have interviewed pharmaceutical and biotechnology firms in the US, Canada and India; PhRMA, the U.S. industry organization, and the corresponding Indian trade associations; as well as people involved in tropical disease issues at the NIH, the World Bank, and Yale and Harvard public health institutes. These interviews have been useful in designing the survey questions and strategy. For example, while it was clear from talking to executives that they were able to answer questions regarding current spending on relevant projects, the categories we are dealing with are not those which firms typically use when describing their R&D portfolios. This argues against a strategy of waiting five years and asking retrospective questions about research happening now, and in favor of repeated surveys focused on research ongoing at the time of questioning. The interviews together with the statistical data give us a picture of current research activity which will serve as a baseline against which we can track changes. The interviews also allowed us to identify factors that might interact with the change in patent laws and affect the appropriate interpretation of trends in the data. Finally, the discussions were used to try and understand the role of patents and other factors that determine where private firms decide to invest their R&D dollars.

The following section describes the recent history of international negotiations over pharmaceutical product patent rights. We discuss the timing of the ‘event’ of their introduction, noting that any observed incentive effects will depend not just on the timing of the ‘event’ and the characteristics of actual legal changes, but also on firms’ information and beliefs about those changes. In Section III we specify in more detail how demand patterns for drugs may differ between the countries which have had product patents for some time and the group of countries newly introducing protection. Section IV provides the statistical and survey data, with a focus on trends over time. In Section V we discuss factors besides IPRs that might be encouraging more R&D expenditure on tropical diseases, affecting the interpretation of trends. In Section VI we consider factors that might contribute to a failure to see changes — again, what needs to go with patents to make them effective? Finally, Section VII concludes.