Document(s) 7 of 20

The first international treaty bringing intellectual property (IP) into agriculture was drawn up in Europe to harmonize and support existing national systems that give commercial plant breeders' rights over the plant varieties they breed, and to promote the system in other countries. This system of plant breeders' rights was a newly created alternative to the US approach of allowing plant patents, itself a special regime designed to protect vegetatively reproduced ornamental and fruit varieties. Since the TRIPS Agreement, all WTO Members must provide some form of IP protection for plant varieties, but it is up to them how to do it. This chapter examines the development of plant breeders' rights and some issues that arise from them today.

Background and History

For almost all of human history, farming and crop improvement were carried out by the same people and in the same places, by farmers and indigenous peoples on their own land. The separation of the two activities is very recent, starting in the 19th century. In this chapter I explain what scientific breeders do (see also Chapter 6) and then briefly trace the history of this separation between farming and breeding, which began in Europe and North America, where the first professional breeders emerged, and farmers abandoned or were forced out of breeding as an activity. This separation is an ongoing process in many developing countries and in some areas has hardly even begun.

Since Neolithic times, farmers have set aside some of their harvested seeds for replanting. They selected such seeds, whether consciously or unintentionally, on the basis that the plants producing them possessed desirable traits such as high yields, disease resistance, or drought or frost tolerance. Over the generations, this practice resulted in ever increasing quantities of locally adapted varieties known as 'landraces', 'folk varieties' or 'farmers' varieties'.

Breeding as a science

This situation changed in North America and Europe from the late 19th century as the profession of farming became a separate one from seed production. The emerging seed producers started to select from the existing materials to increase their share in the market. This commercial crop improvement remained merely empirical and experimental but with a growing scientific basis in mathematics applied to selection methods. Very soon after the 1900 rediscovery of Mendel's insights into the laws of heredity, scientists sought to apply genetics to crop improvement. This led to the directed development of 'pure lines' of self-pollinating crops. Pure lines are uniform, breed true to type and contain consistent and identifiable traits that can be transferred to other plants. According to two experts on the politics of modern agriculture, Robin Pistorius and Jeroen van Wijk (1999), 'while Mendelian breeding allowed for a controlled mixing of genetic characteristics, pure line breeding offered a practical method to "fix" them in succeeding generations'.

Breeding new plant varieties is a laborious and time-consuming process. It takes about 7–10 years to get from the first cross to the marketable variety. The first task is to determine the objectives of the breeding programme. One obvious goal is to produce varieties with higher yields, but there are many other possible objectives such as the development of varieties with added or improved characteristics such as pest resistance, disease resistance or drought tolerance, compatibility with inputs such as fertilizers and pesticides, and improved consumption or food-processing characteristics. A major challenge for breeders is to respond both to the requirements of varying farming conditions and to the need to develop varieties that can be sold widely. Furthermore, they increasingly have to respond to the ever-changing demands of conglomerate seed and chemical companies, food processing companies and supermarket chains.

The basic conventional technique is known as 'crossing and selecting', which involves crossing two or more parent lines or varieties with desirable traits to produce multiple offspring. Of these, the best plants are selected and allowed to breed again. Again, the best ones are selected for breeding and the process is repeated a number of times. After 8 to 12 generations, an improved variety is produced that breeds true and is ready to be planted by farmers.

But breeding is rarely this simple. For one thing, a new variety may be derived from 50 or more parental lines. For another, a variety used in the breeding programme may be the source of only one desirable trait and many undesirable ones. So how does the breeder incorporate this single trait into his or her new variety while excluding the others? Very simply, let us call plants from the parent line or new variety into which the single trait is to be introduced 'Group A' and call members of the 'donor' plants (which could well be a wild or semi-domesticated relative) 'Group B'. These Group B plants are the source of just one desirable trait, among many unwanted ones, for which as little as one allele (a DNA sequence that codes for a gene) may be responsible. For the breeder to transfer this allele without the undesirable traits, he must first cross Group A and Group B plants and then 'back-cross' those offspring containing the trait with plants from Group A. This is repeated through the generations, selecting plants that retain the trait and back-crossing them with Group A plants. In time, the proportion of genes from Group B plants contained in the offspring goes down in conventional selection systems from 50:50 in the first generation to a negligible figure.

These approaches generally work well with crops like wheat, rice and sorghum that self-fertilize. These tend to be genetically stable and consequently breed true. But, as with humans and animals, inbreeding can be deleterious for cross-pollinators such as maize, pearl millet and cruciferous crops like cabbages and oilseed rape. This is not such a problem for plants that can reproduce asexually, such as vines, apple trees and potatoes, where the genetics are fixed through this reproduction system: once a new variety has been bred, it can be multiplied through vegetative forms of propagation, whether cuttings, grafts or tubers. But for cross-fertilizing seed crops, the breeder must find another approach.

Maize breeders in the early 20th century came up with a solution by applying the rediscovered principles of Mendelian genetics. George Shull, a breeder working at a US government research centre, managed to induce the characteristic of what he called 'heterosis' in the corn plants, resulting from his cross-breeding of inbred lines. This phenomenon, commonly referred to as 'hybrid vigour', is manifested in heightened yields. But because they are hybrids, the offspring cannot breed true and the maximum yield enhancements thus last only for a single generation. The additional advantage that hybrid varieties provide a uniform crop compared to the open-pollinated populations became apparent with large-scale agricultural mechanization. So while farmers stand to benefit from seeds providing this hybrid vigour, they need to buy seeds at the beginning of every planting season to enjoy equally productive future harvests. If farmers replant the seeds from hybrid crops, the resulting plants tend to be 'segregated', reflecting the characteristics of the grandparents. This necessity to buy seed was and continues to be a boon for the seed companies, which could correct a major risk factor in seed production, namely that seed markets are generally anti-cyclic, in other words after a good harvest – when the seed producer has good stocks – farmers save their seed, whereas the demand for seed is high when seed production conditions have been poor. Hybrids create a stable seed market.

The hybrid route to the breeding of better seeds is generally assumed to be a very good thing for farmers and for the development of the seed industry, but sceptics argue that the massive investments in the development of hybrid varieties that were made in the 1920s and 1930s could have been allocated to breeding based on more conventional techniques using open-pollination that would have achieved similar yield increases but without preventing farmers from being able to replant their harvested seeds (Lewontin, 1993).

Other breeding techniques, such as tissue and cell culture development, have also been used for several decades. These enable scientists to regenerate large numbers of plants that are genetically identical and free from disease. These techniques do not replace conventional breeding but can improve its efficiency. More recently, molecular biology introduced new opportunities in breeding, either to make conventional breeding more efficient and effective (marker-assisted selection) or by moving foreign genes into the breeding materials (genetic engineering), not just from other plant species, but sometimes from completely different forms of life. For example, scientists have succeeded in inducing insect resistance in crops like corn and cotton by inserting genes from a soil microbe called Bacillus thuringiensis (Bt) that is toxic for certain insects. These techniques include direct gene transfer into tissue cultures using bacteria or viruses as carriers of the foreign DNA, and such devices as high-velocity 'gene guns' which shoot DNA-containing 'bullets' into cell nuclei. The new science of genomics is being used to identify useful genes and the plants which contain them.

The emergence of the modern seed industry

During the 19th-century westward expansion of the US, the government sought to encourage settlement. One way to do this was to entrust the farmers themselves with the selection, breeding and multiplication of seed. To this effect, first the Patent Office, and then the US Department of Agriculture (USDA), provided farmers with free seed packets for them to experiment with. At the time the seed industry was small and insignificant. Farmers used these seeds and those introduced by the immigrants arriving in the US to breed varieties adapted to suit their own needs and the local ecological conditions. The number of such varieties increased enormously. Later these farmer-bred and selected crop varieties formed the basis of the public and private sector breeding programmes.

Cary Fowler (1994) argues that the separation of farming from breeding, the undermining of the customary practice of seed saving in the case of hybridized crops and the commodification of the seed cannot be explained by advances in plant breeding science and technology alone. When scientifically bred seeds came onto the market, subsistence agriculture had largely been replaced by commercial farming anyway. Mechanized harvesting and the consolidation of landholdings had made seed selection non-viable compared to the greater convenience of purchasing mechanically-cleaned seed from dealers. And, since most farmers were no longer improving seeds themselves, the attraction of selecting and replanting was declining even before scientifically bred varieties became widely available.

In 1890, 596 US firms were involved in commercial seed production. Having formed a business association called the American Seed Trade Association (ASTA) a few years earlier, they were becoming active in defending their interests. One of ASTA's early campaigns was to stop the government from providing farmers with seeds. This failed for lack of support from the public and Congress, many of whose members sent seed packets to constituents. However, during the first two decades of the 20th century, the government increasingly sent seeds only of the most common varieties to farmers, while passing on the more exotic germplasm to the government experimental stations and colleges. A later campaign by ASTA from the First World War onwards was to oppose the saving of seed by farmers.

Shortly after the First World War, the US Secretary of Agriculture decided that the USDA would henceforth support research aimed at the development of hybrids and ending farmer participation in breeding programmes. The Secretary's son, Henry A. Wallace, was sent by his father on a tour of experimental stations and recommended his father appoint a devotee of hybrids as head of research. Henry subsequently founded the Pioneer Hi-Bred Seed Company and was himself appointed Secretary of Agriculture in 1932 (Lewontin, 1993, pp55–56).

The implications of the emergence of corn hybrids for private-sector breeding cannot be underestimated. Several of the world's major 20th-century seed companies first came to prominence through their successful breeding of hybrid corn varieties. Many of these old seed companies are now owned by companies like Monsanto, Syngenta, Dupont and Delta & Pine Land, which was itself bought by Monsanto in 2006. According to Jack Kloppenburg (1988), 'hybridization is … a mechanism for circumventing the biological barrier that the seed had presented to the penetration of plant breeding and seed production by private enterprise'. This was well understood by some of the pioneering scientists involved in the development of hybrid corn, who realized that the absence of genetic stability in the harvested seed gave them a kind of virtual IP protection which they could back up by using trade secrecy law. Indeed, the determination of companies to prevent unauthorized access to their inbred parent lines could be very intense. Unfortunately for breeders (and presumably for farmers), though, hybridization does not work for some of the most economically important crops such as wheat. Clearly, this presents problems for breeders. Plants are self-reproducing. With no law to prevent it, there is nothing to stop farmers from replanting harvested seed, or even multiplying seed and selling it in competition with the breeder (assuming this would be more profitable for them than selling harvested produce). This is where IP rights come in.

As opposed to the US situation of expansion by settlers into new land for cultivation, albeit into land often taken from the indigenous peoples, virtually all the cultivable land in 19th century Europe had been farmed for a very long time. Most of the major European crops whose origins were exotic, like wheat, rye, maize, potatoes and tomatoes, had become well-established and integrated into local farming systems for centuries or even millennia. Although some crops were vulnerable to devastating diseases due to widespread genetic uniformity (most notoriously potatoes), European farmers developed a huge range of varieties over the centuries to suit local conditions. European governments generally did not find it necessary to encourage farmers to breed new varieties themselves as in the US case.

Agricultural intensification took place under different circumstances and with different aims. In Europe, land was in short supply but labour was plentiful, rather than the other way round as was the case in the US. Farms tended to be smaller and did not lend themselves so easily to mechanization. Attempts to increase productivity came through other techniques to grow more food on existing land.

Introducing new species and formal experimental breeding were carried out first by wealthy landowners, and from the second half of the 19th century by small family seed firms. These firms descended from farmers that made it their main business to provide seed for other farmers and who then started breeding programmes to better meet the requirements of their customers. As in the US, in the early 20th century public research institutions and universities were also carrying out breeding work, which benefited the emerging private plant breeding sector. By the time of the Second World War, Germany, the US, The Netherlands, Sweden and the Soviet Union were the leading plant breeding nations for field crops. In countries like Britain and France, government-supported research during the first half of the century was often directed to tropical agriculture rather than temperate-zone crops. This was to develop and improve the production of certain crops in the colonies. Both countries reoriented their breeding efforts as these colonies became independent, and France subsequently developed the world's second largest private seed sector. Britain, though, had few seed firms until the 1960s, and most breeding, especially of major crops like wheat, was left to the public sector. This situation has changed, but the seed sector is still much smaller than that of France – the birthplace of the UPOV Convention.

From the 1960s, the use of modern high yielding varieties of major food crops like rice and wheat became increasingly common in developing countries, particularly in Asia and Latin America. Nonetheless, the extent of private sector involvement in breeding for developing country farmers was quite modest during that era. Since then, though, US and European plant breeding companies have become much more active in the developing world. This is happening at a time when public sector agricultural research investments targeted at the needs of developing country farmers, especially those with few resources, are considered by many to fall well short of what is needed.

As already mentioned, several economically valuable crops do not lend themselves to hybridization. For these, breeders needed to find other means to control the use and production of their varieties. This is where lack of IP protection became an issue several decades ago, leading to the development of an international regime designed specifically to protect plant varieties whose seeds could otherwise be easily saved, replanted and sold, namely the UPOV Convention. Before looking into the particular solution devised, Table 2.1 places the discussion in context by presenting the legal and technological problems and solutions faced by companies seeking to capture the rewards from their investments in plant improvement according to how the plants normally reproduce (Dutfield, 2007).

Table 2.1 Appropriating plant breeding innovations: Legal and technological problems and solutions

The UPOV Convention

The form of plant variety protection (PVP) under UPOV, otherwise known as plant breeders' rights (PBRs), is commonly described as a 'patent-like' regime. In fact, this is not quite accurate, as an understanding of the background to the UPOV Convention should clarify. Admittedly, though, the increasing strength of the PVP right of recent years is beginning to approximate that of the patent.

The UPOV Convention was adopted in Paris in 1961 and entered into force in 1968 once it had been ratified by three countries – The Netherlands, the UK and West Germany. It was revised slightly in 1972 and more substantially in 1978 and 1991. The 1978 Act entered into force in 1981, the 1991 Act in 1998. All members, with the exception of Belgium, are parties either to the 1978 or the 1991 Acts. New members are required to accept UPOV 1991, although exceptions allowing membership on the basis of the 1978 Act have been made.

The Convention established an organization called the Union Internationale pour la Protection des Obtentions Végétales (UPOV). The official English translation is International Union for the Protection of New Varieties of Plants. UPOV has a close association with the World Intellectual Property Organization (WIPO – see Chapter 4) to the extent that the latter organization's director general is also secretary general of UPOV. As of April 2006, the Union had 60 member states plus the European Community. Unlike WIPO, UPOV is a lean organization with a small office in Geneva. The supreme decision-making body is the Council, which consists of representatives of each member and has one ordinary session a year.

The existence of UPOV can be attributed largely to two international organizations. One is the Association Internationale pour la Protection de la Propriété Industrielle (International Association for the Protection of Intellectual Property) (AIPPI), which was founded in 1897 and consists of activist legal practitioners, patent attorneys, trademark agents, scientists, engineers and corporations. The other is the Association Internationale des Selectionneurs pour la Protection des Obtentions Végétales (International Association of Plant Breeders) (ASSINSEL).

At the 1952 AIPPI Congress, the delegates, partly at the urging of ASSINSEL, discussed the issue of plant varieties. There was general agreement that plant varieties should be protected in some way. The most concrete ideas came from the German AIPPI group, which submitted a detailed technical report arguing that both patents and an alternative system should be available to breeders. As the authors, Franz and Freda Wuesthoff, explained, it is a normal requirement of patentability that other people skilled in the art should be able to reproduce the invention described in the specification (Wuesthoff and Wuesthoff, 1952). That is to say, following the instructions provided in the specification should result in the invention as claimed. But, as they explained, when it comes to plant breeding, being able reliably to reproduce the new variety from the beginning is difficult because it depends on natural processes over which breeders do not have total control and which are to some extent random. However, repeating the whole breeding process is not necessarily important or even necessary. What really matters is that the new plant that has been brought into existence can be directly propagated. For sexually reproducing plants, this means that they must breed true so that the offspring are identical to their parents.

As the two authors saw it, the solutions were to change the patent system by taking a permissive view of the reproducibility requirement and to extend the range of patentable subject matter in those countries where plants were not considered to constitute inventions, and to develop a new or modified IP system for the more incremental plant breeding-derived innovations. They considered that patents in their present form would accommodate a certain amount of innovation in plant breeding and should be made available to breeders, but that for many new varieties, workable IP protection would require a relaxation of the novelty and inventive step requirements so that varieties reflecting incremental improvements on existing ones and that were already known about could nonetheless be protected.

The AIPPI Congress could not reach a consensus on the means of protection, and a 1954 Congress also failed to do so. One of the main reasons was that some of the patent lawyer members of the AIPPI opposed the patenting of plant varieties on the grounds that doing so would stretch basic patent law concepts like inventiveness to the point of undermining the credibility of the patent system (see Chapter 3).

In the event, ASSINSEL's members decided at their own Congress in 1956 to abandon the patent route and to call for an international conference to consider the possibility of developing a new international instrument for protecting plant varieties. ASSINSEL requested the French government to organize what became the International Conference for the Protection of New Varieties of Plants. The Conference, which convened in May 1957 in Paris, established the basic principles of PBRs that were later incorporated into the UPOV Convention. Only European governments were invited to participate or attend as observers. A Committee of Experts was set up to:

• study the legal problems arising out of the protection of the breeder's right as defined by the Conference;
  
  
• give as precise formulations as might be appropriate of the basic technical and economic principles laid down by the Conference; and
  
  
• prepare the first draft of an international convention for submission to a later session of the Conference itself.

The Committee met twice before appointing a Drafting Group to develop a legal text. One of the important issues the Committee had to decide upon was whether the convention would be incorporated into the general framework of the Paris Convention, which dealt with indus­trial property, primarily patents, or whether a separate convention was necessary. It decided in favour of the latter, but recommended that the new office administering the convention should work closely with the Bureaux Internationaux Réunis de la Protection de la Propriété Intellectuelle (BIRPI – see Chapter 4), the forerunner to WIPO.

The second meeting of the International Conference for the Protection of New Varieties of Plants took place in November 1961, with 12 European countries invited along with BIRPI (now WIPO), the FAO, the European Economic Community, the Organisation for Economic Co-operation and Development (OECD), the AIPPI, ASSINSEL and two other business associations: the Communauté Internationale des Obtenteurs de Plantes Ornementales et Frutières de Reproduction Asexuée (CIOPORA) and the Fédération Internationale du Commerce des Semences (FIS). Since then, ASSINSEL, the FIS and the new International Seed Federation (ISF), along with CIOPORA and the International Chamber of Commerce, have played key roles in shaping the evolution of the UPOV Convention through its various revisions. UPOV was created and shaped by plant breeders for plant breeders and they have a strong sense of ownership of the Convention. Public interest organizations have had minimal involvement. The UPOV office is an active proselytizer of plant variety protection around the world and strongly defends the Convention from those who question its value to developing countries or its continued relevance in the age of biotechnology.

Just as the UPOV Convention was being adopted, the Council of Europe was actively working to promote the harmonization of patent rules, procedures and principles among the Western European countries. One key result of this was the signing of the 1963 Convention on the Unification of Certain Points of Substantive Law on Patents for Invention. The Convention had to accommodate wide differences in national patent rules relating to pharmaceuticals, food, agriculture and horticulture while encouraging states to harmonize their rules within a realistic time-frame at the level of the most expansive rights available at that time in any one country. Accordingly, parties were not required to grant patents in respect of '(b) plant or animal varieties or essentially biological processes for the production of plants or animals'.

The terms 'essentially biological' replaced 'purely biological' from an earlier version of the text. The Council's Committee of Experts on Patents, which was responsible for drafting the convention, changed the wording to broaden the exclusionary language to embrace such 'essentially biological' processes as varietal selection and hybridization methods even if 'technical' devices were utilized to carry out the breeding processes. The Convention's text reflects, as of course does UPOV, the decision in Europe made in the late 1950s to keep plant breeding out of the patent system. There is much similarity between the wording of the Convention and that of the European Patent Convention and the TRIPS Agreement (Chapter 3).

The Convention's Provisions and Issues Arising

Compared to some other important international agreements on intellectual property, such as TRIPS and the Paris Convention for the Protection of Industrial Property, the UPOV Convention's provisions are extremely detailed and specific. They deal with the plant varieties covered, the requirements for protection, the length of the protection term, the scope of protection, breeders' exemption, farmers' privilege, and whether or not both patents and PVP can be held on the same variety. They have been subject to several revisions since 1961. In order to join the Union, countries are supposed to have PVP regimes already in place, and these are normally scrutinized by UPOV to see that they are in harmony with the Convention's provisions.

The most substantial revisions took place in 1978 and 1991; these are discussed and compared below. Note, however, that the French word obtention in the name of the Union and the Convention is significant since it indicates that rights can be acquired not just by those who breed new varieties in the classic sense of creating new varieties by crossing and selecting sexually reproducing plants, but also by those who improve plants based on the discovery and selection of mutants or variants found in a population of cultivated plants. Thus UPOV 1991 clarifies that a breeder is a person 'who bred, or discovered and developed, a variety'. This is consistent with the original intent of the Convention to protect varieties that may not entirely be attributable to the application of scientific breeding. At the same time, however, it represents a divergence from patent law, which professes not to allow mere discoveries to be protected. Table 2.2 compares the key provisions of UPOV 1978 and 1991 and patent law.

What qualifies for protection?

To be eligible for protection under the UPOV system, plant varieties must be new, distinct, stable and uniform:

• To be new, the variety needs to be so not necessarily in the absolute sense, but not to have been offered for sale or marketed, with the agreement of the breeder or his successor in title, in the source country, or for longer than a limited number of years in any other country.
  
  
• To be distinct, the variety must be distinguishable by one or more characteristics from any other variety whose existence is a matter of common knowledge anywhere in the world, implicitly including among traditional farming communities. Compared to UPOV 1978, the requirement in the most recent version has been relaxed somewhat. It does this by dropping the phrase 'by one or more important characteristics' after the word 'distinguishable'.
  
  
• To be considered as stable, the variety must remain true to its description after repeated reproduction or propagation. In other words it must have a certain level of uniformity which avoids the change of the variety through genetic drift.

The uniformity requirement also shows the specific nature of the UPOV system, since this requirement cannot practically be the same for species with different ways of reproduction; self-fertilizing species can be much more uniform than cross-fertilizing crops. Therefore the uniformity requirement is made relative instead, in other words a new variety should be uniform when compared to the varieties of the same species. This means that when plant breeding techniques were refined, the uniformity requirement gradually increased, making it beyond the reach of farmer-breeders who may select in landraces, which are not genetically uniform, to develop new varieties. Unlike patents, there is no disclosure requirement. Instead, applicants are required to submit evidence that the variety meets the protection requirements (in the US, for example) or to submit the plant material for which protection is sought to the responsible governmental authority for testing to ensure that the above eligibility requirements have been met.

While it is logical to require that protected varieties are genetically uniform, there are dangers with widespread planting of crop varieties that are genetically similar to each other. Kloppenburg (1988, p93) provides a good illustration of what can go wrong when there is 'genetic vulnerability accompanying dependence on a narrow base of germplasm'. In 1970, '15 per cent of the corn crop in that year was lost to an epidemic of southern corn leaf blight. Corn prices rose 20 per cent, and losses to consumers and farmers totalled some US$2 billion.'

Table 2.2 Comparison of main provisions of UPOV 1978/1991 and patent law

What is a 'plant variety', and how may it be distinguished, for the purposes of IP protection, from a 'plant'? This is very important given the increased application of genetic engineering to crop research and the fact that in some jurisdictions, plants are patentable but plant varieties are not. The original 1961 version of the UPOV Convention defined 'plant variety' as including 'any cultivar, clone, line, stock or hybrid which is capable of cultivation'. The 1991 revision contains a more detailed definition, according to which a plant variety is:

a plant grouping within a single botanical taxon of the lowest known rank, which grouping, irrespective of whether the conditions for the grant of a breeder's right are fully met, can be:

• defined by the expression of the characteristics resulting from a given genotype or combination of genotypes;
  
  
• distinguished from any other plant grouping by the expression of at least one of the said characteristics; and
  
  
• considered as a unit with regard to its suitability for being propagated unchanged.

Scope of protection

UPOV 1978, which several countries are still contracting parties to, defines the scope of protection as the breeder's right to authorize the following acts: 'the production for purposes of commercial marketing; the offering for sale; and the marketing of the reproductive or vegetative propagating material, as such, of the variety'. The 1991 version extends the scope of the breeders' rights in two ways. First, it increases the number of acts for which prior authorization of the breeder is required. These include 'production or reproduction; conditioning for the purpose of propagation; offering for sale; selling or other marketing; exporting; importing; and stocking for the above purposes'. Second, such acts do not just concern the reproductive or vegetative propagating material, but also encompass harvested material obtained through the illegitimate use of propagating material and so-called essentially derived varieties.

Breeders' exemption

However, the right of breeders both to use protected varieties as an initial source of variation for the creation of new varieties and to market the resulting varieties without authorization from the original breeder (the 'breeders' exemption') is upheld in both versions. This represents a major difference with patent law, which normally has a very narrow research exemption. Many plant breeders are concerned about the effects of patents on free access to plant genetic resources, including varieties bred by others. One difference between UPOV 1978 and UPOV 1991 is that the latter extends rights to varieties which are essentially derived from the protected variety. So the breeder of PVP-protected variety A has the right to demand that the breeder of variety B secure his or her authorization to commercialize B if it was essentially derived from A. The main idea here is that breeders should not be able to acquire protection too easily for minor modifications of extant varieties produced perhaps through cosmetic breeding or genetic engineering, or free-ride without doing any breeding of their own, problems that the increased application of biotechnology in this field appeared likely to exacerbate.

PVP and patents

Beyond resolving these particular issues, but related to them, the provision on the scope of protection was also intended to ensure that patent rights and PVP rights operate in a harmonious fashion in jurisdictions where plants and their parts and genes are patentable and access to these could be blocked by patent holders. Such a practice would undermine one of the main justifications for PVP protection, which is that breeders should be able to secure returns on their investments but without preventing competitors from being able freely to access breeding material. An example here might be useful. Consider the case of a PVP-protected variety, variety A, and a patented genetic element owned by another company (Jördens, 2002, p6). The owner of a patent on this genetic element is free to use A to produce his or her variety B and, in the absence of the essential derivation provision, place B on the market with no obligations to the owner of A, despite the fact that B differs from A only in the addition of the patented genetic element. However, the owner of A would need a licence from the producer of B to use the patented genetic element in the breeding of further varieties. In such a situation, then, patents can have the effect of blocking the breeders' exemption that PVP rights normally provide. The PVP-issuing office, however, will not itself determine whether a variety is essentially derived from an earlier one. This will be left to the courts. So far, only one court, in The Netherlands, has been called upon to make such a determination and it found in favour of the defendant (Fikkert, 2005). According to the court, the general rule is that distinguishable varieties are normally independent, the essentially derived variety (EDV) provision being an exception to this rule that ought to be construed narrowly. Given that one of the two varieties at issue differed in several ways in shape and form from the variety from which they were allegedly essentially derived, the exception was not applicable. As for the other variety, no convincing case had in any case been made that it was an EDV, besides which the Community Plant Variety Office made no mention of its similarity to the already registered original variety or found any grounds to investigate such a possibility.

In the EU, the 1998 EC Directive on the Legal Protection of Biotechnological Inventions seeks to make PVP and patents operate more harmoniously by providing that where the acquisition or exploitation of a PVP right is impossible without infringing a patent, or vice versa, a compulsory licence may be applied for to allow for its use. If issued, the licensor party will be entitled to cross-license the licensee's patent or PVP right. Subsequent legislation in Germany and France restore the breeder's exemption in that it explicitly allows breeders to use genetic materials that include patented components for further breeding. When the new variety contains the patented component, however, consent has to be sought for the marketing of that new variety; when the patented component is 'bred out' of the material, the patent holder has no rights on the new variety.

Farmers' privilege

There is no reference in the 1978 version to the right of farmers to re-sow seed harvested from protected varieties for their own use (often referred to as 'farmers' privilege'). The Convention establishes minimum standards such that the breeder's prior authorization is required for at least the three acts mentioned above, namely the production for purposes of commercial marketing; the offering for sale; and the marketing of the reproductive or vegetative propagating material, as such, of the variety. Thus, countries that are members of the 1978 Convention are free to either uphold farmers' privilege or eliminate it. All UPOV member countries implemented the exemption for 'private and non-commercial use' under the UPOV Act of 1978 to include the re-sowing and in some cases the local exchange or sales of seed. However, this was not the case for ornamental crops in The Netherlands, where a stronger protection was deemed necessary. In the US this was interpreted very liberally, so that in practice sales of farm-saved seed were allowed provided that they contributed less than 50 per cent of total farm income. This resulted in large quantities of seed being 'brown bagged' to the detriment of the commercial interests of the breeder.

The 1991 version is more specific. Whereas the scope of the breeder's right includes production or reproduction and conditioning for the purpose of propagation (Article 14), governments can use their discretion in deciding whether to uphold the farmers' privilege which includes only the use of saved seed on the same farm (and thus excludes any type of exchange or sale of such seed). According to Article 15, the breeder's right in relation to a variety may be restricted 'in order to permit farmers to use for propagating purposes, on their own holdings, the product of the harvest which they have obtained by planting … the protected variety'. Even though the Act states that the legitimate interest of the breeder explicitly has to be taken into account, the seed industry generally dislikes farmers' privilege. The EC Regulation 2100/94 on Community Plant Variety Rights, which was adopted in 1994, restricts farmers' privilege to certain crops, and breeders must be remunerated through the payment of royalties unless the users of the farmers' privilege are small farmers, in which case they are exempted. Interestingly, the European Community's patent rules also require that farmers' privilege be provided and defined under the same terms as the above regulation. The US's PVP rules are less strict in this regard: seed saving must be restricted to the amount necessary for on-farm replanting, but it is not clear how the legitimate interests of the breeder are implemented since royalty payments on farm-saved seed are not required.

Length of protection and double protection

UPOV 1991 extends protection from at least 15 years to a minimum of 20 years. This later version is silent on the matter of double (that is, both patent and PVP) protection, whereas the 1978 version prohibited such double protection on the same variety. Allowing double protection without any restriction was to ensure the intellectual property practices of the US and Japan, which allowed such double protection, would be fully compliant with UPOV. Nonetheless, most countries, including all European countries, expressly forbid the patenting of plant varieties. In 1995, in Greenpeace v. Plant Genetic Systems NV, the European Patent Office's (EPO) Technical Board of Appeal ruled on an appeal against the upholding of a plant-related patent. The board determined that a patent claim for plant cells contained in a plant is unpatentable since it does not exclude plant varieties from its scope. This implied that transgenic plants per se were unpatentable because of the plant variety exclusion. Consequently, for the next four years, the EPO stopped accepting claims on plants per se. However, in December 1999, the EPO Enlarged Board of Appeal decided in Novartis that, while genetically modified plant varieties are unpatentable, 'a claim wherein specific plant varieties are not individually claimed is not excluded from patentability under Article 53(b), EPC, even though it may embrace plant varieties'. This reopened the door to the patenting of plants as long as the claims in the patent specification do not refer to individual varieties.

Changing membership

Until the late 1990s, the overwhelming majority of UPOV members were developed countries, reflecting the fact that in many developing countries, especially in Africa, private sector involvement in plant breeding and seed supply is quite limited. Moreover, in many of these countries small-scale farming communities are responsible for much of the plant breeding and seed distribution, as they have been for centuries. Consequently, until recently there would have been few domestic beneficiaries of a PVP system in these developing countries apart from the public institutes for agricultural research.

However, many developing countries are now joining UPOV. In many, if not most, cases, this is not because of any strong domestic demand for PVP, but because of their obligations under Article 27.3(b) of TRIPS (see Chapter 3) or trade agreements (see Chapter 7). The UPOV system is the only sui generis system for plant varieties that exists in international law and is currently being actively promoted worldwide by the organization itself, as well as by the US and the EU though bilateral free trade agreements that tend to require developing country parties to join UPOV. However, developing country WTO Members that prefer not to allow plant varieties to be patented do not need to join UPOV. In principle they can devise their own system without reference to UPOV's standards. Alternatively, they could simply use one of the UPOV Acts as a model but opt not to join the organization, an approach that many Asian countries currently follow, choosing to use the 1978 version of UPOV as a model, mainly because of the greater freedom to formulate farmers' privilege.

PVP versus patents

Despite the increasing membership of UPOV, the question arises of why breeders still tend to prefer PVP to patents, and also of whether this particular intellectual property right has a future. After all, patents provide much stronger legal protection (Table 2.2). Moreover, breeders nowadays tend to work not for small independent seed firms, but for large life science corporations that invest huge amounts of money in biotechnological research and hold massive patent portfolios. Probably the main reasons for this preference for PVP is the breeders' exemption, which allows them such broad access to breeding material, and their concerns that the patenting of biotechnological research tools may jeopardize this access. Breeders, especially those in smaller companies, also fear the complexity of rights in the patent system, compared to the simple 'one variety, one right' system of PVP, leading to complex legal battles for which they fear they do not have the financial or legal resources. This suggests that as long as there is profit-motivated plant breeding, PVP rights will continue to exist alongside patents, and sometimes in tension with them.

Critical Concerns

Concern has been raised that the UPOV system was drawn up mainly by European countries, and is designed to accommodate the specific characteristics of the capital-intensive large-scale commercial agricultural systems that generally prevail on that continent. As a result, it is often argued, the system is unsuitable for most developing countries. Among such critics, the current system of IPRs protection for plants has raised concerns over their impact on food security in three areas: (i) PVP and research priorities; (ii) the interests of poor farmers; and (iii) the availability of genetic resources for further breeding. Note, however, that, while a few studies have been carried out on the impacts of PVP in developing countries (see below), the overall effects of plant intellectual property in developing countries are difficult to discern conclusively and researchers find themselves having to rely to too great an extent on the experiences of developed countries.

PVP and research priorities

Does the UPOV Convention encourage breeders to investigate minor crops and to bring whole new species into cultivation? Empirical evidence casts doubt on whether PVP (as well as patents) does much to encourage investment in plant breeding except in just a few commercially important crop species such as wheat and soya bean and ornamentals, although UPOV's own studies are, perhaps unsurprisingly, much more positive about the overall impacts of PVP (UPOV, 2005a). Critics also argue that even if breeders did turn to neglected crops, many of the small farmers that grow them would not be better off if their freedom to use saved seed as they wished were diminished. In most developing countries a very large proportion of the farming population consists of smallholders, and for these people saving, selling and exchanging seed is common practice and essential for their survival.

Many smallholder farmers cultivate minor food crops that enable them to meet the nutritional needs of rural and urban communities much better than if major crops such as wheat, rice and maize alone were cultivated. In many parts of the world, farmers may grow more than 100 crop species and cultivated varieties. However, PVP does not encourage breeding related to minor crops with small markets. This is because the likelihood of good returns on breeders' research investment is small even with the legal protection provided by PVP. Rather, it encourages breeding targeted at major crops with significant commercial potential. Moreover, protected varieties of plants may not even be food crops. In Kenya, for example, from 1997 to 2003, out of a total of 611 PVP applications, 247 were for foreign-bred roses. This is not necessarily a bad thing, since such exports of cut flowers are a good source of foreign exchange, but some argue that the production methods used damage the environment and the health of the growers (War on Want, 2007).

It is conceivable, then, that PVP may contribute to a trend whereby traditional diverse agro-ecosystems, containing a wide range of traditional crop varieties, are replaced with monocultures of single agrochemical-dependent varieties, with the result that the range of nutritious foods available in local markets becomes narrower. Admittedly this trend is a global phenomenon whose beginning predates the introduction of PVP systems; nevertheless, it is one that the existence and increasingly widespread use of PVP may indirectly encourage.

PVP and smallholder farmers

In most developing countries a large proportion of the population depends on agriculture for employment and income. Many of these farmers are smallholders for whom the saving and across-the-fence and inter-community exchanging of seed are common practices. This is especially the case in countries and regions where neither the public nor private sectors play a significant role in breeding, producing or distributing seed. Although the UPOV system may allow on-farm replanting, its rules restrict farmers' freedom to buy seed from sources other than the original breeders or their licensees.

Seed companies argue in response that farmers do not have to purchase PVP-protected seed just because it is available. They point out that the farmers are free to continue cultivating seed that is not plant variety protected, including traditional local varieties, if they so wish, and that therefore their basic freedoms are unaffected by PVP. While this is likely to be true, traditional varieties are often disparaged and are likely to be excluded from government-approved seed lists that some countries maintain under their seed regulations. While finding non-PVP seed may not yet be a serious difficulty for developing country farmers, this situation may change. In some developed countries, it is becoming difficult for farmers to find non-PVP varieties of some crops.

Seed laws

PVP is not the only issue, as seed laws may sometimes unduly limit the choice of varieties that farmers are allowed to bring into commercial production. Seed regulations were introduced for very good reasons. From the late 19th century a number of European governments became alarmed about the unregulated nature of the seed trade and the extent to which poor quality seed got onto the market. This situation was problematic for farmers, legitimate breeders and governments, which had become concerned about the need to increase agricultural productivity. In the early decades of the 20th century, many governments responded first by establishing seed testing stations, and then by certifying seed. The latter also provided to a certain limited extent a kind of IP protection for breeders, and as such was a kind of barrier to market entry. Indeed, UPOV largely grew out of such seed certification regimes. In many developing countries, governments concerned about rural poverty and convinced, even if mistakenly, that traditional agriculture is unproductive support farmers in rural credit schemes by promoting particular crops and types of seed, such as hybrids, which tend to require expensive inputs and may not be suitable for the local agronomic conditions (see also Chapter 8, Box 8.9). Furthermore, seed aid is often used by providers as a way to promote the use of modern varieties that may not necessarily be the most appropriate ones to plant. As Sperling et al (2006) explain:

while formal sector varieties are referred to as 'improved' and the quality of the seed is certified, these varieties often yield poorly in many smallholder cropping systems. Such new varieties may not be adapted to the local agroecological conditions and farmers may not possess the management inputs (for example fertilizers and pesticides) crucial for their growth.

Seed saving, however, is not always a cost-effective option for farmers since saved seed deteriorates over generations. Moreover, seed is one among several agricultural inputs that farmers may have to pay for. Even poor farmers may decide to pay a higher price for better quality seed if they expect a bigger harvest to result.

IPRs and genetic resources for breeding

Plant breeders and other supporters of UPOV tend to stress the necessity of being able to freely access genetic material, including that which is IPR protected. This is why the UPOV Convention contains such a broad breeders' exemption. Patent law tends to have a much narrower research exemption, which is often limited to non-commercial scientific or experimental use. Moreover, while a protected plant variety is covered by a single title, plant-related biotechnological inventions are likely to be protected by a patent, in some cases by several patents. These patents may cover not just plants, but also seeds, genes and DNA sequences. The effect of patents is to restrict access to the patented 'products'. It has been argued that 'locking up' genetic resources with patents is a bad thing because innovation in plant breeding is cumulative and depends on being able to use as wide a stock of material as possible. This is a plausible concern and one that many plant breeders share with public interest NGOs such as GRAIN. It was to deal with this concern that the FAO International Treaty introduced a number of provisions; these are discussed in Chapter 6.

However, the restrictions on access to breeding material may have other causes than IPRs. For one thing, some countries have chosen to exclude certain categories of plant genetic resources they consider to be strategically important from the multilateral system set up under the International Treaty. Furthermore, some developing countries have been exercising their rights under the Convention on Biological Diversity (CBD – see Chapter 5) to regulate access to their genetic resources in ways that unduly restrict their movement. This may well be detrimental to long-term food security even in their own countries.

Research, development and ownership

Beyond these issues about how specific intellectual property rights privatize genetic material needed for breeding is the association of IPRs generally with the shrinkage of non-proprietarian public sector research, and the increased concentration of ownership of breeding material, research tools and technologies in the hands of a small number of giant corporations. While IPRs are not directly responsible for this shrinkage of public sector research, they do appear to contribute to this concentration effect. For one thing, the expense of acquiring large IPRs portfolios can operate as a barrier to market entry. For another, the advantages of owning lots of patents and PVP titles are such that large firms have an incentive to buy up or merge with rival companies that also own such rights (see also Chapter 8).

Are these trends a bad thing? One consequence is a reduction in the free circulation of breeding material. This may lead to a reduced level of welfare-enhancing breeding activity, especially if rights are asserted against the public sector. In addition, they may make public policymaking aimed at enhancing food security harder to put into practice. This is because it is much more difficult for governments to influence companies than the public institutions they partly or wholly fund. Cash-strapped governments, however, may have to reduce their research expenditures out of necessity, and the private sector can and sometimes does play a useful role in taking up the slack.

Empirical Evidence

This discussion on how PVP affects food security and nutrition in developing countries leads one to consider in more general terms the applicability of such an IPR to these countries. Unfortunately, few comparative empirical studies exist, but one such was conducted by UPOV (2005a). It covered five countries experiencing highly varied levels of development (Argentina, China, Kenya, Poland and South Korea) and argued that PVP brought economic benefits.

Two other studies were published in 1994 by the Inter-American Institute for Cooperation in Agriculture and the University of Amsterdam (Jaffé and van Wijk, 1995), and in 2005 for the World Bank by various researchers (Louwaars et al, 2005).

Taking the 1994 study first, the research aimed to examine the expected impact of PVP on developing countries in the areas of 'private investment in plant breeding, breeding policies of public institutes, transfer of foreign germplasm and diffusion of seed among farmers'.

Five countries were used as case studies, of which three (Argentina, Chile and Uruguay) had PVP systems already in place and two (Colombia and Mexico) were about to introduce them. These countries are similar in the sense that each has basically two seed markets. The hybrid seed market is controlled by transnational corporations, whereas the seed market for self-pollinating varieties is dominated by domestic firms.

However, Argentina differs from the others in that it is the only country in which owners of PVP rights have successfully enforced their rights to the extent that their control over seed supply for wheat and soya is comparable to that of their counterparts in the US. This leads the authors of the study report to conclude that, in all probability, PVP in that country has 'prevented the local wheat companies from reducing or even terminating their breeding activities and triggered the reactivation of some soya bean breeding programmes'.

For exotic genetic resources, there is little evidence to show that PVP has led to any significantly improved access for domestic seed companies to modern cultivars, special genetic stocks and genomic material from abroad. Moreover, companies with licences from overseas breeders to use proprietary varieties may sometimes have to contend with restrictions on where they can export to. For example, in 1994 Argentinean strawberry plant growers were prevented from exporting their plantlets to Europe because the US breeder and the European licensees did not want these plantlets to compete with those that were already produced in Europe.

In Argentina and Chile, public agricultural research centres are using PVP to secure income and collaborate with companies. According to the report, this is shifting the orientation of public research and reduces the public availability of their genetic resources.

How are farmers affected? First, Argentinean seed dealers must now pay royalties and taxes on the seed they trade. So far these costs have not been passed on to the farmers. Second, PVP legislation in the three countries where it is well established has not prevented the replanting of farm-saved seed. Third, as the report indicates, 'since many modern plant varieties are not appropriate for resource-poor farmers, PBRs predominantly favour plant breeding for those farmers who operate under relatively prosperous conditions'.

The study for the World Bank covered China, Colombia, India, Kenya and one country that still does not have a PVP regime, Uganda. Among the study's numerous findings is that the availability of PVP is not an absolute prerequisite for the existence of a thriving plant breeding sector. India had quite a large number of private plant breeding firms many years before PVP legislation was passed. Rather cautiously, among the report authors' findings are that 'the ease of implementing PVP seems to be overestimated. In all cases, the effectiveness of PVP is still being tested and refined, and the cases illustrate that establishing a PVP law and putting it into practice are two separate challenges.'

Moreover, as the authors conclude:

not only do IPRs in plant breeding have to be seen in the context of a wider range of agricultural policies, but IPR regimes themselves must be carefully tailored to specific situations. It is important that countries recognize that they have choices in designing legislation consistent with the TRIPS Agreement and that there are still opportunities for debating and interpreting the Agreement itself. The UPOV Conventions offer some important advantages for fulfilling the requirements for a sui generis system but they do not exhaust the possibilities.

Emerging Issues – Harmonization

The big emerging issue is harmonization. UPOV and the associations representing the plant breeders are very keen to see a situation in which PVP is not only available throughout the world but follows the same standards of protection. At present, many countries are still parties to the 1978 version of UPOV, and some countries (for example India) have laws that diverge from any version of UPOV. In the longer term, the associations would probably like to see more regional regimes, such as that in Europe, which has a Community Plant Variety Rights Office. The office, which was established under Council Regulation (EC) No 2100/94 of 27 July 1994, provides a single unitary right covering the whole EU. A certain amount of administrative harmonization may be a good thing for developing countries in terms of cutting the cost of managing the PVP system. But holding an incredibly diverse range of countries to the same substantive rules is inadvisable for similar reasons that harmonized patent rules tend to benefit countries that are 'leaders' and may well hold back the 'followers' (Dutfield and Suthersanen, 2005). Each country should be free to tailor IP systems to their economic conditions and in ways that promote their wider development objectives and strategies.

National Instruments

Realistic proposals for non-UPOV PVP systems have been few and far between. Most countries that do not want to join UPOV use legislation that is based on the 1978 version of the Convention. This is probably one reason why more developing countries are joining UPOV. Nonetheless, it is important to consider alternatives to UPOV so that informed decisions can be made.

To help countries devise an appropriate sui generis system, the International Plant Genetic Resources Institute (IPGRI – now called Bioversity International) came up with a list of key questions that decision makers should take into account:

• What kind of domestic seed industry exists?
  
• What kind of public breeding sector exists?
  
• What kind of seed supply system is in place?
  
• To what extent is farm-saved seed used in the country?
  
• What is the current capacity of breeders?
  
• What do local breeders want to do in the next 5–10 years?
  
• Are external inputs to agriculture low or high?
  
• What are the country's production needs and objectives?
  
• What is the country's biotechnology capacity?
  
• What are the goals and realistic expectations of the biotechnology sector?
  
• What kinds of strategic alliances will the country want to enter into in the next 5–10 years and how involved will other countries be?

The answers to these questions will vary widely from one country to another, which suggests that, as with patents, one size is unlikely to fit all. A detailed discussion on all the issues involved falls outside the scope of this chapter, but it is important at least to discuss the requirements for protection and the scope of the systems.

The sui generis clause in TRIPS (see Chapter 3) does give governments a certain amount of freedom to tailor their PVP systems to address such concerns. Thus, while an increasing number of developing countries are joining UPOV, some countries are devising alternative PVP systems that aim in part to strengthen food security. They do this, for example, by allowing farmers to acquire protected seed from any source and/or requiring protected varieties to display qualities that are genuinely superior to existing varieties.

The Indian parliament has passed legislation that would maintain farmers' freedom to save, sell and exchange all produce of a protected variety (Box 2.1), and the African Union (formerly the Organization of African Unity) has developed a model law, for the consideration of member governments, known as the African Model Legislation for the Protection of the Rights of Local Communities, Farmers and Breeders, and for the Regulation of Access to Biological Resources. In both cases, as much emphasis is placed on the interests of farmers as on those of breeders.

The Indian Act appears to reflect a genuine attempt to implement TRIPS in a way that supports the specific socioeconomic interests of all the various producer groups in India, from private sector seed companies to public corporations and research institutions and resource-poor farmers. However, India is in the process of joining UPOV under its 1978 Act, and the 2001 legislation may need to be modified. The Indian case also shows the interaction between PVP and seed legislation, since a new seed bill seems to restrict the same farmers' right to sell seed by requiring compulsory certification.

Conclusion

It is actually very difficult for developing countries to design and implement their own systems of PVP if, as is likely, these would diverge at all from the latest version of the UPOV Convention. Intellectual property reform has always been political as much as technical, never more so than in the present day. As this book and others amply demonstrate, both the EU and the US impose various forms of 'soft' and 'hard' pressure on developing countries to introduce IP rules that they approve of. And for PVP, UPOV provides the approved standards, with no alternatives acceptable to those standards. While India may be strong enough to resist pressure from the EU, US and UPOV, it seems most other countries are too small and weak to have much room for manoeuvre.

The importance of PVP globally, and pressures to introduce the UPOV model into developing countries, stems from the extension of IP requirements into agriculture through the TRIPS Agreement in the WTO, which is the subject of the next chapter.

Resources

Apart from UPOV, other key organizations of importance in this area are the International Community of Breeders of Asexually Reproduced Ornamental and Fruit Tree Varieties (CIOPORA), GRAIN, and the International Seed Federation (ISF) (see Appendix 1).

Document(s) 7 of 20