Global Food Fight
An intense drive to patent agricultural biotechnologies may hurt those who would benefit most: people in developing countries.
More than one million children die each year because of a chronic lack of vitamin A. Millions more suffer disease. Many of these children live in developing nations where rice is the main staple. To help solve this problem, scientists have genetically engineered a variety of rice that is rich in beta carotene, an important source of vitamin A. Dubbed golden rice because of its yellow color, it could help improve millions of lives in developing countries, as well as improve the nutrition of legions of people in developed countries. But a careful study shows that anyone wanting to produce golden rice might have to secure licenses for more than 30 groups of patents issued to separate entities.
The long-term challenge for agriculture is daunting. Earth's population is expected to rise by 50 percent over the next half century. The current agriculture system simply will not be able to feed this world. We will need another Green Revolution to provide adequate food without seriously damaging the environment. Despite recent consumer skepticism, genetically modified crops such as golden rice are one of the only ways to drive such a revolution. Many scientists say that these seeds offer a safe route to crops that are more productive, that better resist plant disease and stress, and that provide improved nutrition. Research projects offer not just golden rice but crops that are resistant to viruses and insect pests. Drought- and salt-resistant crops are possible as well. But beyond the problem of public acceptance, there is the barrier of patents on genetically modified seeds, the biotechnology techniques for creating them, and the gene sequences of plants themselves. The patent system, designed to foster innovation, may be slowing it for some of these applications.
The first Green Revolution grew from an international public research system that began in the 1940s with support from the Rockefeller Foundation and expanded to include 16 research centers, including the International Rice Research Institute (IRRI) in the Philippines and the Centro Internacional de Mejoramiento de Maiz y Trigo (CIMMYT), the corn and wheat research center in Mexico. These centers collaborate through the Consultative Group on International Agricultural Research (CGIAR), a consortium of donors including foundations, national governments, United Nations institutions, and the World Bank. These centers have long conducted research and breeding to develop new crop varieties, sometimes on their own and sometimes in cooperation with national agricultural research systems in developing nations. The centers evolved in a world without intellectual property rights, in which seeds and breeding procedures were free for all to use and were distributed without charge to seed and farming groups throughout the developing world. This system increased rice yields in South and Southeast Asia by more than 80 percent and led to plant varieties that have served as parents for one-fifth of the U.S. wheat crop and more than two-thirds of the U.S. rice crop.
These research institutions are now facing increasingly pervasive ownership of intellectual property rights. Simply to conduct research, the centers must consider the risk of infringing patents. This is a situation in which the patent system has worked to encourage private research but has at the same time greatly complicated crucial applications of the new technology.
The problem goes much further than the legal scope of patents. Universities in developed nations, such as U.S. land grant universities, which are so critical to healthy U.S. agriculture and which for decades have collaborated closely with CGIAR and developing world research institutions, are themselves pursuing intellectual property rights. As a result, they may refocus their research away from developing-world needs. Furthermore, out of fear of offending their developed-world donors, the international research institutions may be hesitant to use technologies patented by private firms in those nations, even though the technologies are unpatented in the developing nations.
How the patents evolved
Until about 1980, the only intellectual property protection available for crop plants was Plant Breeders' Rights (PBRs), a relatively weak form of legal protection that prevailed in most developed nations. A country's department or ministry of agriculture issued a PBR certificate to a seed owner that prevented competitors from selling seeds or breeding material from the owner's specific seed variety. However, the PBR allowed competitors to use the protected varieties as sources of subsequent seed variation in their own breeding programs. But then the United States began to permit regular patents on living organisms such as plants and seeds, as well as on genes and a variety of other biological plant components and diagnostic materials, all of which are much more restrictive than PBRs. These broadened intellectual property rights have helped create the new biotechnology industry.
Other nations are adopting similar rules, partly to encourage their own industries and partly because of the Trade Related Intellectual Property (TRIPS) agreement. Under this agreement, signed in 1994 as part of the Uruguay Round of trade negotiations, all nations, including developing nations, committed themselves to an intellectual property regime that would protect plant varieties. As this agreement is implemented, a company in almost any nation will be able to obtain exclusive rights to a particular seed or variety and keep others from selling it. Other developed and developing nations are also following the U.S. lead and are beginning to provide regular patent protection on various genes, plants, seeds, and biological procedures.
The other factor that spawned the move toward more restrictive intellectual property rights was the 1980 Bayh-Dole Act, which gave universities the right to obtain patents on and to commercialize inventions created under government grants. This legislation was supported by the argument that important inventions would languish in the absence of such intellectual property rights. Although many university patents and a number of successful products have resulted, the law has led to legal wrangling as universities argue over rights to use one another's very basic patented inventions in research. A recent National Institutes of Health study indicated that proprietary rights to basic research procedures and reagents may seriously slow the flow of scientific information and therefore potentially hinder the progress of science. Nevertheless, the pursuit of royalties is spreading to universities and government research institutions in Europe, Japan, and the developing world, as many nations consider and adopt similar laws. Their research institutions hope that license revenue can be an income source in periods of shrinking government support. However, the resulting patents may also slow and complicate the application of biotechnology to meet the developing world's food needs.
These trends are creating a patent problem. The first indicator is that the number of patents in many areas of basic agricultural research is growing exponentially. For example, U.S. patents related to rice remained well below 100 per year through 1995. But in 1999 and 2000, more than 600 patents were issued annually. There will be many more for crops such as corn, which have greater commercial interest in the West. Further evidence of the rapid patenting of basic agriculture comes from a recent survey published in Nature, which found that about three-quarters of plant DNA patents are in the hands of private firms, with nearly half held by 14 multinational companies; virtually no such patents existed before 1985.
The United States permits the broadest variety of agricultural patents. It has issued regular patents for entire plant lines, such as specific lines of herbicide-resistant rice. Such varieties are probably unpatentable in most nations, where only PBR protection is available for plant lines. The U.S. patentability of plant varieties was upheld early in 2000 in an appellate case, Pioneer v. J.E.M. Agric. Supply, which the Supreme Court is now reviewing. The claims of these patents typically extend to the progeny of the plant and its seeds. The claims clearly are designed to keep other breeders from using the protected seed for breeding material, which will restrict its use in U.S. research for developing-world applications.
There are also very broad U.S. and European patents on groups of plant varieties, such as the U.S. Agracetus patents that seek to cover all transgenic cotton and soybeans. These patents, if valid, could give Monsanto, which has acquired Agracetus, control of all transgenic varieties of these crops.
Of most importance to plant breeders, however, are patents covering specific technical procedures used in agricultural genetic engineering. Technology to create hybrid rice, for example, was developed largely in China, where hybrid seed provides a substantial portion of the country's rice. Although the China National Seed Corporation's early patents are no longer in force, the company patented certain aspects of the technology in the United States. These patents deny breeders access to research tools that could be useful in developing new varieties of many crops. Patents have also been granted on other ways to produce hybrid seed.
Further limitation on research could come from a U.S. patent for the gene gun, one of the most common means for inserting genes into plants. It was issued to Cornell University, which licensed it to DuPont. Similarly, Monsanto holds a patent on the 35S promoter, a portion of DNA that is often inserted with a plant gene to encourage its expression. If breeders cannot use such tools or need licenses to use them, it will be substantially more difficult and expensive for them to produce superior seeds.
Patenting genes and DNA
Genes themselves are now routinely patented, typically with claims that cover the isolated gene, various constructs that include the gene, plants transformed with those constructs, and the seed and progeny of those plants. Plants that naturally contain a given gene are not novel and therefore the patent does not apply to them or to breeding with them. But any other use of the gene, its constructs, seeds, or progeny may be prohibited. One example is the University of California patent on the Xa21 Kinase gene, which makes grains resistant to disease. Work done at IRRI was important to identifying the gene, and the university arranged to protect IRRI's right to use the gene. However, the rights to some other genes are securely in private hands, with no commitment to make them available. This is the case for some of the patents for inserting into plants the genes that code for viral coat proteins, which confer resistance to plant viruses.
It is also the case for many of the patents for Bacillus thuringiensis (Bt) technology, in which bacterial genes inserted into plants code for toxic proteins that kill insects. Loose granting of Bt claims has led to hundreds of often overlapping patent rights that have been the subject of substantial litigation. At least four different companies, for example, have laid claims to Bt-transformed maize. It is almost impossible for a researcher to find ways through this patent thicket.
Genomic information is typically protected through trade secrecy practices. In this system, a company that creates a substantial database or map of a genome provides access only under agreed terms, which might include a mechanism for compensation. This model is also the basis for important international nonprofit cooperation. For example, because rice is so important to the world's poor and its genome is smaller than that of some other cereals, a global genome sequencing effort is being carried out by Japan, Korea, China, the United States, the European Union, and the Rockefeller Foundation through the International Rice Genome Sequence Working Group. Information will be placed into public databases, and the participants have agreed not to file patent applications on the sequences. Monsanto has developed a sequence of its own and has agreed to make its genomic rice information available for public breeding in developing nations. Syngenta and Myriad Genetics completed a rice sequence in January and have promised to provide information and technology for developing world subsistence farming, but they are not putting their sequences in the public domain. Moreover, many of the important rice genes may be patented, and it is not clear that other genomes or the genomes of major pathogens will be as readily available.
This patenting trend is paralleled by an enormous concentration of agricultural biotechnology. Five large companies--Aventis, Dow Chemical, DuPont, Monsanto, and Syngenta--now control a substantial piece of the agricultural patent portfolio. These firms have been purchasing smaller biotechnology companies in order to obtain the technologies those companies have developed, have merged with chemical and pharmaceutical companies for access to production capacity and chemical markets, and have bought seed firms throughout the world to improve their ability to market new products. In the process, they have assembled broad intellectual property portfolios. As the concentration of the industry is growing, the amount of agricultural research is shrinking. The reduction may in part be a response to recent environmental and consumer criticism of bioengineered foods, but it may also stem from decreased incentive because of industry consolidation.
In the past few years, several of these large firms have actually begun to take an interest in developing world markets. The interest is strongest in soybeans and the major grains (maize, wheat, and rice), where developing-world markets are large and where there may also be major export potential, but it also extends to rice, the seed of which was viewed until recently as a fundamentally noncommercial product, supplied by public institutions on a free or low-cost basis. During the Green Revolution, better varieties such as IR-16 and IR-64 were developed under donor funding at IRRI. The institute freely transferred new varieties and innovative breeding materials to national research centers in the major East Asian nations. They, in turn, further bred varieties that were optimized to local growing conditions and released them to national systems for production and distribution to farmers.
These public varieties dominate in Asia, but companies are moving in. Pioneer, now owned by DuPont, has established research programs in India. Private hybrid rice breeders such as Mahyco also have emerged there. Monsanto has undertaken collaborative research with the Indian Institute of Science. Japan Tobacco became interested in rice seed. And the developing-world components of Cargill had already begun a hybrid rice-breeding program before being acquired by Monsanto. Global patent searches show that these and other agricultural majors are seeking to protect their intellectual property positions in large developing nations, including China and Brazil. Even though these nations may not issue the full panoply of legal protections available in the United States, important research procedures, tools, and gene constructs are likely to be patented in at least some of them.
The private sector's interest in providing rice seed to developing nations reflects the growth of substantial commercial markets there. The total value of the rice produced in the two leading Asian markets is easily more than that of the U.S. maize crop that has induced so much private research. This does not immediately convert into a seed market, because harvested rice can generally be used as seeds. Private-sector investment will depend on some form of proprietary position: successful hybrids or plants protected by either intellectual property rights or by a "terminator" technology that makes the rice plants infertile. There may be difficulties in achieving this position, but the Asian rice potential is big enough for companies to want to try.
The firms also have a commercial interest in marketing chemicals. By transferring into national crop lines the genes necessary for herbicide resistance, a firm can create a larger market for the herbicide. China has already made intellectual property rights available on herbicides. India has granted exclusive marketing rights and its laws require granting full patents by 2005.
When the multinational firms enter markets such as the Asian rice seed market, they will probably come with seeds that are better than those now available. This is good. And many scientists argue that the use of herbicide-resistant plants is environmentally better than the alternative ways of fighting weeds. But the private-sector seeds will probably be developed only for the larger commercial markets; it will be a long time before the private sector improves small crops or serves subsistence farmers. More important, there is a very serious possibility that, because of patent rights and the small number of large companies, the multinational industry will hold a monopoly or oligopoly on transgenic seeds, keeping out competitors and even the public sector. Prices will therefore also be higher than current prices. Finally, it may be impossible or at least very expensive or difficult for the public sector to gain access to patented technologies or to use protected varieties for research in developing new applications for the smaller crop or subsistence farmer.
How we could respond
Three kinds of responses to the dangers of overly restrictive intellectual property rights deserve consideration: for national governments to change their patent laws, for the public and private sectors to negotiate a global licensing system that makes new biotechnologies available, and for public research institutions to obtain rights to technologies on a case-by-case basis.
Redesigned patent laws. Developing nations are generally responding to the 1994 TRIPS agreement, in which all countries committed to protecting work on crops by adopting as low a standard of protection as possible, typically PBRs only. This protects specific varieties but does not provide very significant incentives for biotechnology advances such as new genes or new transformation methods. Hence, multinational and even national firms are likely to press national governments to adopt stronger intellectual property protection. Developing nations will be held back, however, by the fear that such legal changes will increase royalty costs to their farmers, breeders, seed companies, and research groups, and give even greater advantage to the multinationals.
Nations might be able to help resolve this dilemma by fine-tuning their patent systems. For example, a stronger standard for rejecting patent applications for inventions that are "obvious" would slow the rise in the number of patents. Many patents currently issued in the United States may satisfy the patent law's "nonobviousness" requirements as judged by lawyers, but they appear obvious to most scientists or engineers. A stronger standard would not affect important inventions that are really nonobvious, but it could decrease the risk that large firms might freeze others out by patenting numerous minor inventions.
Furthermore, to decrease the risk that a company can block others from large areas of science, the scope of patents could be narrowed. Use of a strong requirement that the invention be genuinely useful, rather than just an abstract concept, could help prevent patents from preempting broad areas of research. So could provisions permitting the experimental use of patented inventions, notably the use of patented materials in breeding processes. Or there might be dependency license systems that permit subsequent inventors to use prior inventions on a reasonable royalty basis. These issues apply to many countries beyond the United States, including those in the developing world. An institution such as the World Intellectual Property Organization or the World Bank should sponsor serious study and dialogue on whether such changes in patent laws might wisely balance the need for research incentives with the fact that researchers--especially those working for the needs of the poorest--must build on the work of previous researchers.
Another tack is for nations to develop and use their own fair-competition laws to maintain a strong defense against monopoly in the seed supply sector. Even though the industry oligopoly is evolving at the global level beyond the control of developing nations, these countries might still be able to discourage the takeover of a local firm or use compulsory licensing in response to monopolistic practices.
The chief barrier to these approaches is that the policy issues involved are technically difficult, and few nations have a staff or resources that allow them to define and implement the necessary policies. Educational and expertise-sharing programs among patent offices or other national bureaucracies would help. And breeders themselves should be heard from on the design of patent systems that affect plants.
Global licenses. A second plausible approach is to grant developing-nation institutions a license to all or many technologies from the private sector. A new institution or clearinghouse could be created that would acquire the necessary legal rights by license and then license them forward for developing-world needs. A consortium of electronics companies that have patents related to digital video disks have already put such arrangements in place within the developed world, as has the American Society of Composers, Authors, and Publishers, which issues licenses that provide an economical mechanism for collecting royalties for certain musical and recording performances. Presumably, a developing-nation license would apply only to the poorest nations and to subsistence farmers in the middle-income nations. It is important to note that unless markets can be divided in this way, it is unlikely that the multinational firms would be amenable, because the license would otherwise threaten their most lucrative markets in the developed world. This market division is not as easy as it would have been a few years ago, because nations such as Brazil and Mexico increasingly have both commercially important markets and many subsistence farmers. But the approach would certainly be possible for crops such as cassava, in which there is unlikely to be any commercial interest, and in situations where markets can be divided by climatic or soil conditions.
The real question is whether the private sector will be motivated to provide such a license other than in contexts such as cassava. After all, many of these firms are hurting financially and are worried about recouping the agricultural research investments they have already made. The motivations that have underlain other broad license systems are absent here. For example, the pharmaceutical industry recently formed a SNP (single-nucleotide polymorphism) consortium to ensure that a large number of these gene sequences would stay in the public domain for research use by all. The industry therefore created a cooperative research procedure to identify the SNPs and implement legal arrangements to ensure their free use. Neither this cross-license motivation nor the desire to facilitate the collection of royalties are yet present in agricultural biotechnology. Collection of royalties is currently more easily done through vertical integration or arrangements with seed distributors.
The most likely motivation for global licensing today is that the large seed firms may decide that they themselves need a cross license to gain freedom to operate. Some semiconductor companies have agreed to these kinds of cross licenses, in which each of the firms was infringing on many patents held by the others. If the agricultural biotechnology industry, which may be facing a similar pattern of cross infringement, does create such a cross license among the large firms, antitrust considerations may compel openness to other firms and possibly to the international public sector.
Public funding of licensing is also possible. Many donors and research funding institutions might be able to condition their grants on a commitment by the recipient to license the technology for developing-world applications. Moreover, in the face of the current environmental and consumer concern about agricultural biotechnology, leading companies are becoming concerned about their image, and they may be willing to facilitate licenses to developing nations in order to garner positive public relations.
Public-sector research rights. As noted, the public agricultural research sector has provided developing countries with enormous benefits for many years and until recently was able to conduct biotechnology-based research without constraints imposed by the intellectual property system. Because life is no longer that simple, the public sector has to find a way to coexist with the private sector within the developing world itself.
In this context, the public sector must rethink its focus. One option is to move upstream from crop seeds and concentrate on the development of more environmentally sustainable agricultural technologies, which would then be applied in cooperation with the private sector. Another approach is to concentrate on subsistence crops, such as cassava, and on varieties of commercial crops, such as upland rice, that appeal primarily to subsistence farmers. For basic crops such as rice and corn, however, it is important to keep good-quality public-sector seeds available, even if they do not have the advantages of the newest seeds from the private firms. These seeds serve as competition to keep down the price of private-sector seed and thus make it more likely that poorer farmers can have sophisticated multinational technology at a reasonable price.
Fortunately, there will not always be conflict. Many of the most important patents have been issued only in developed nations and thus far do not directly affect research or domestic agriculture in developing nations. Many developing nations already have exemptions in their patent laws that permit patented inventions to be freely used in certain forms of research, so that some of the public-sector research may not be an infringement. And many of the patents that are most important to the private sector, such as the terminator patent or patents on particular inbred lines, are essentially irrelevant to the public sector. Moreover, the multinationals will be concerned about the public relations costs of restricting work in poor nations.
When the public sector does need the private sector's patented technology, its best current approach is through collaboration--for example, a cooperative program with a private firm to place the firm's proprietary Bt gene for disease resistance into a public-sector variety that is bred at IRRI or CIMMYT. The private sector brings the new gene and associated technology. The public sector brings important varieties and an understanding of local growing conditions, pathogens, and agronomic conditions that are important to the success of the variety.
To date, these collaborations have taken two forms. In the first, the public institution, typically IRRI or CIMMYT, acquires a specific technology from a particular firm. The firm may be motivated by public relations, and the real costs may be small. But the firm may also be paid, usually through funds raised for this purpose from global donors. Developed nations may be especially willing to provide this form of indirect subsidy to their own national firms. In these agreements, the products of the collaboration are typically made available to the developing nations on royalty-free or reasonable royalty terms but are kept off the developed-world market or made available to that market only on terms that protect the commercial interests of the private company.
In the second form of agreement, the international institution has started a line of research that is of interest to the private firm. In this case, the company may be willing to subsidize the institution's research or assist in developing a project if it is given some commercial exclusivity in the resulting technology. Clearly, the public-sector institution cannot, consistent with its charter, permit such exclusivity to apply to the developing-world poor. But it can permit it to apply in the developed world. Here again, there must be discrimination among markets. A good example is the arrangement organized through the German company Greenovation, under which golden rice, developed at a Swiss public research institution, was licensed to Zeneca (now part of Syngenta) for assembly of the necessary patent rights and development for both developed and developing markets, with the latter receiving preferential treatment.
These collaborations enable the public sector to benefit from the patent position of the private partners. To carry out the arrangements and to have a strong bargaining position to enter the arrangements, public institutions may need to obtain patents themselves, as was done with golden rice. This is clearly appropriate. It is also appropriate for public institutions to build a portfolio of patents to use as bargaining chips that give them freedom to operate with their own technologies, which may, even unintentionally, infringe on particular patents. For bargaining chip purposes, the most useful patents are those that the multinationals will want, and the most useful place to obtain patent coverage is in the developed world. It will be difficult for the public sector to obtain many such patents, but even a few important ones could strengthen their position. In the short term, this step-by-step, institution-by-institution, agreement-by-agreement strategy is essential.
The private sector may or may not be able to reach the broad or individual agreements that will make advanced agricultural technologies available to developing nations, and the private sector itself will provide some of the technologies to some of the developing nations. But for the public sector's research, so critical to the developing world and the future of the human food supply, the patent system is causing enormous complexity and may be slowing the development of needed technology. The United States and other national governments, together with institutions such as the World Intellectual Property Organization and the World Bank, must together figure out how to adjust national and international patent systems and research and competition policies so that they actually encourage the global application of essential agricultural technology.
K. Fischer et al., "Collaborations in Rice," Science 290 (October 13, 2000): 279280.
N. Harl, "Possible Consequences of Concentration in Input Supply in Agriculture," testimony presented to the Committee on Agriculture, Nutrition and Forestry, United States Senate, Washington, D.C., October 6, 1999.
R. David Kryder et al., The Intellectual and Technical Property Components of pro-Vitamin A Rice (GoldenRice™): A: Preliminary Freedom-To-Operate Review (ISAAA Brief No. 20, 2000).
S. Thomas et al., "Plant DNA Patents in the Hands of a Few," Nature 399 (June 3, 1999): 405406.
B. Wright, IPR and International Research Collaborations in Agricultural Biotechnology, presented November 16, 1999, at the meeting on Agricultural Biotechnology in Developing Countries: Toward Optimizing the Benefits for the Poor, November 1516, 1999, Bonn, Germany.
John H. Barton (firstname.lastname@example.org) is professor of high technology and international law at Stanford Law School, Stanford, California. Peter Berger (email@example.com) is an associate in Preston, Gates & Ellis's technology and intellectual property department in Seattle, Washington.