A bitter decade-long dispute over pharmaceutical patent protection in developing countries has become exceptionally costly. The clash between the pharmaceutical industry and advocates for the poor may not only hinder research on diseases endemic to the developing world but also call into question the entire system of supporting research through patent rights.
Setting patent standards was one of the most contentious issues in negotiations leading to establishment of the World Trade Organization in 1995. After heated debate, WTO founders agreed to require all member countries, including those in the developing world, to issue patent rights on pharmaceutical products. The decision sparked angry criticism, which has only intensified with growing awareness of the AIDS crisis and the development of effective, but expensive, drug therapies to combat the disease. Although the pharmaceutical industry insists on the benefits of strong patents on drug innovations, others are adamant that inventors should have only limited rights to control the sale of pharmaceuticals in the developing world, or no rights at all. Pharmaceutical firms' current strategyto negotiate price and licensing terms product-by-product and country-by-country for the developing worldhas left them vulnerable to criticism by advocates for the poor. The negative attention focused on the industry has implications that go far beyond development, and compromise on this highly charged issue is urgently needed.
The Roadblock of Uncertainty
Patents exist to stimulate research. But the current uncertainty about the form of future patent rights in the developing world prevents patents from being of any use as a research incentive there. Why would a pharmaceutical firm invest now in a product for a poor country when it can have no way of knowing what patent protection will be available there in 10 years once product development is finally complete and the product is ready to market? This fallout of the long-running dispute is particularly unfortunate because it hinders research on drugs for diseases, such as malaria, that kill millions of people in the developing world each year. These diseases receive few research dollars, and private incentives to invest have been largely absent. Until a more predictable and reliable global system is set up, patent rights can do nothing to help bridge this gap.
What would a predictable and reliable system look like? For patent rights to be predictable, the rules must be clear. This means avoiding the ambiguities that give rise to disputes over interpretation. For patent rights to be reliable, the global system must be viewed as "fair." Otherwise, inventors will continue to risk attack if they try to exercise their patent rights.
For the global system to be seen as fair, it will surely need to recognize differences in the development level of countries. Too many people remain unconvinced that the poorest countries should be required to set up a U.S.-style patent system. In fact, many of the developed countries themselves did not protect pharmaceutical innovations until they reached a far higher level of income than the poor countries have today. Greece, Norway, Portugal and Spain, for example, introduced pharmaceutical patents only in 1992, when GDP per capita exceeded $12,000 (in 2001 dollars).
Disagreement over the appropriate form of global patent rights has gone on for a long time and has become very public. Positions on both sides are entrenched. But finding a resolution remains imperative. How can a patent system recognize differences in countries' well-being and, at the same time, encourage the private sector to become involved in creating new products for the developing world?
One solution would be to establish a system where patent protection in poor countries differs across diseases depending on the importance of those countries' markets as a potential source of research incentives. No one, for example, would argue that African nations are an important source of incentives to do cancer research. The search for a cure for cancer is driven by demand from the big Western markets. Demand in Africa, however, might be sufficient to spur more malaria research. Under the system I describe below, patent protection would be minimal in the poorest countries and would broaden gradually to cover more diseases, starting with those, like malaria, that are particularly prevalent there. Protection would grow as a country became richer and its market significant, but not before. Such an approach automatically tailors the patent system both to development levels and to the characteristics of different drug markets.
Pricing and Research Incentives
Supporting pharmaceutical research and development through a patent system entails an unavoidable trade-off between lower prices and new products. A patent right confers a monopoly that raises the price that consumers (and governments) face in purchasing a pharmaceutical drugand thus may put the drug out of reach of some disease sufferers. Yet this monopoly right also provides companies with incentives to invest in research for new drugs that may one day save millions of lives. Weaker patent rights allow more competition and lower prices, benefiting patients today. More extensive patent rights raise prices, giving firms greater reason to invest in creating frontier products that may cure future disease sufferers.
Extending patent rights to new countries, however, does not inevitably lead to significant new research investment. When profitable markets exist elsewhere and the countries introducing patent protection are small, new rights will have little effect, for they will add only a small increment to an already large market. But, as suggested, the global drug market consists of two different marketsone for diseases specific to developing countries and another for global diseases. Taking into account these different markets offers a way to use patents to support research on diseases specific to developing countries while allowing competition to lower prices on products developed for rich-country markets.
Why Have Patents in Poor Countries At All?
No one disputes the need for more investment to fight diseases that cripple developing countries while sparing rich countries. Diseases like malaria, Chagas' disease, and leishmaniasis afflict millions of people, yet have received almost no research investment from the private sector and little from the public sector. Only 8 of 1,233 drugs licensed from 1975 to 1997 (less than 1 percent) were developed specifically for tropical diseases in humans. For every thousand references to diseases in scientific papers in 1998, just 15 were to developing country-specific diseases. Only one-half of 1 percent of pharmaceutical patents issued in 1998 were related to those diseases. Clearly, past policies that have not included patents in the developing world have largely failed to create the incentives for products for the particular health needs of the poor.
Can new patent rights change this situation? No one views the weak incentives for investment as stemming solely from the lack of patent protection in the developing world. Investment is weak because people in the developing world are poor. The World Health Organization estimates that in 17 of the poorest countries, no more than $10 per person was spent in 1998 on all health care, not just pharmaceuticals. In such small markets, merely introducing patent protection will not suffice to reorient the research priorities of private firms. But global private-sector R&D in pharmaceuticals topped $30 billion in 2001and is still growing. If patents can shift even a tiny fraction of that investment in the direction of products to treat diseases endemic to poor countries, those funds would offer a significant complement to ongoing public and philanthropic investment. New patent laws in the developing world on drugs for diseases specific to those countries, therefore, may encourage greater research.
But what about patents for drugs aimed at diseases common worldwide? These diseases also claim millions of victims in poor countries. In fact, people in developing countries lose four times more disease-adjusted life years from just two global diseases, cancer and heart disease, than from malaria. The high incidence of these diseases in developing countries is reflected in the large share of their drug spending targeted on global diseases. In 2000 spending on cardiovascular drugs alone was 8 percent of total drug spending in Mexico, 13 percent in Mexico, Argentina, and Brazil together. Because poor countries have limited resources, however, their spending on drugs is a tiny part of pharmaceutical spending worldwide. For example, six poor countries with 46 percent of the world's populationChina, India, Indonesia, Pakistan, the Philippines, and Egyptare estimated to account for less than 2 percent of global spending on drugs for cardiovascular disease.
The disparity between rich and poor countries in spending on drugs for global diseases is important. It means that the profits pharmaceutical companies derive from having a patent-based monopoly over sales in poor countries contribute relatively little toward those companies' worldwide profitsand therefore toward their incentive to invest in research. High demand in rich countries drives R&D investment for pharmaceuticals treating global disease. At the same time, patents that increase drug prices even a little can cut millions of people in poor countries off from these drugs.
In short, extending patent protection in poor countries to increase research on diseases such as malaria makes sense; extending it to encourage research on global diseases is less obvious.
Proposal for a Multi-Market System
I have proposed a system of pharmaceutical patent protection that would take into account this reality of the global pharmaceutical market. Drugs for diseases specific to developing countries would enjoy patent protection worldwide, thus giving the greatest support possible to R&D where incentives are weak. For drugs for global diseases, patent owners would be required to choose their market. Would they rather have patent protection in the rich countries or in the poor countries? Given this choice, they would typically choose the rich countriesthe source of their profitsand thereby allow competition to dampen prices in the poor countries.
The system would not be difficult to implement. It would be consistent with WTO membership and would require no change in international treaties. It would require only minor changes to patent law in the United States and other developed countries with research-based pharmaceutical firms. It would impose no regulatory burden on developing countries.
It would work as follows (from the perspective of implementation by the United States). By law, someone inventing in the United States must request a "foreign filing license" from the U.S. patent office before filing patent applications abroad. The mechanism simply requires that a patentee petitioning for this license sign a declaration along these lines: I, the undersigned, request a license to make foreign patent filings covering the invention described in U.S. patent application no. X, with the understanding that this permission will not be used to restrict the sale or manufacture of drugs for [a list of global diseases] in [a list of poor countries] by suing for patent infringement in [a list of poor countries].
How would the mechanism work? Suppose a firm has begun selling its patented drug in a poor country. It has three options if a competitor begins to sell the same drug. The firm can compete, exit the market, or sue for infringement. If it competes or exits, it does not make use of its patent rights in the poor country. If it chooses to sue, it protects its poor-country market on the basis of its patent there. If the firm's product is for a listed global disease, the firm will, by filing a suit, have falsified its declaration to the U.S. patent officecausing its U.S. patent to be unenforceable and allowing generics firms to enter the U.S. market. If the product is not for listed global diseases, the suit will proceed and the firm will have patent protection in all countries.
Because filing a suit would put a firm's U.S. market in jeopardy, firms would rarely file suits. Monitoring would be done by generics firms eager to enter the U.S. market. Almost no government spending would be needed for administration or enforcement. Should a suit arise, information that was already available could be used to determine whether a patentee had falsified its declaration. For example, standard U.S. Food and Drug Administration procedures could be used to determine whether a generic product sold in the poor-country market was for a listed global disease. The generics firm seeking to enter the U.S. market would apply to the FDA for an abbreviated new drug approval of its product, claiming its equivalence to one already marketed in the United States for a listed global disease. The procedure would be precisely the same as that already followed for any generic on the expiration of a patented product. The use of the generic product would be established by the FDA with the issuance of a report confirming bioequivalence.
Identifying Developing Countries and Global Diseases
A transparent and objective procedure would be established to allow the U.S. Patent and Trademark Office to determine the lists of countries and diseases to include in the declaration to obtain a foreign filing license. The country and disease lists would be updated periodically and would change as markets evolved and countries developed.
One possible approach would be a two-step procedure for listing poor countries and corresponding global diseases. It might in fact be practical to specify several sets of countries, beginning with a small set of extremely poor countries and moving toward larger and more inclusive sets. Just as an example, the first set of countries (set A) would include the roughly 40 countries, primarily in Africa but also India, with GDP per capita less than $500 (constant U.S. dollars). The next larger set (B) would include countries (including all in set A) with GDP per capita less than $2,000. That set would number roughly 75 countries and would include such countries as Indonesia and China. The third set (C) would include countries (including all in B) with per capita less than $4,000. That set would number roughly 95 countries and would include such countries as Brazil and South Africa.
Once the country lists had been established, the patent office would draw up a list of diseases for each. To do that, it would identify diseases for which the share of total potential sales coming from each group's markets represents less than some threshold share. Pharmaceutical sales data are already collected regularly and serve as a reasonable indicator of the share of potential profits. So, for example, if sales of drugs to treat a disease in a country group are less than 2 percent of total global sales, the disease would fall under the policy. Disease list A, then, would include all diseases for which drug sales for country set A are less than 2 percent of world sales, and similarly for disease lists B and C.
For countries in set A, spending on drugs would be so low that all diseases would probably qualify; effectively, no protection would be afforded pharmaceuticals. The number of countries in set B gets larger and the countries are also somewhat richer. Their spending on drugs would be greater, so sales on drugs for some diseases would exceed the threshold of 2 percent of global sales; those diseases would no longer qualify and patent protection would become available on them. For countries in set C, just a few diseases would qualify and patent protection would widen further. The above procedure is illustrative onlyother GDP cutoffs and more country groups could be chosen, and a market share other than 2 percent might be appropriate.
With this system, the effective patent rights available to a firm with respect to a particular innovation are determined by the content of the license when it is signed. These rights remain the same throughout the life of the patent and the firm can make its marketing decisions accordingly. At the same time, the content of the declaration evolves. A country starting out in group B, for instance, would move to group C as it grew richer and eventually would not be included in the license declaration. The regime would automatically evolve in line with the development level of countries and the importance of different product markets.
Moving Beyond the Roadblocks
Today positions on pharmaceutical patents are hardened and stakes are high. Firms are reluctant to depart from a strategy of crisis management to entertain systemwide reforms. Advocates for the poor see people dying and have little appetite for reforms that take effect over time and without fanfare. Everyone is hesitant when confronting new ideas.
The proposed new system provides a middle way. It would extend patents to pharmaceutical firms for products targeted at diseases that afflict poor developing countries. It provides a way to introduce patents gradually in developing countries for drugs to treat global diseases. With a global patent system designed to recognize the dual nature of the world's pharmaceutical markets, we can give a big welfare boost to poor countries while supporting the full implementation of patent systems in the developing world.