Biopolicy Journal Pontificia Universidad Católica de Valparaíso ISSN: 1363-2450 Vol. 4, Num. 1, 2001

Douglas Pachico

Centro Internacional de Agricultural Tropical (CIAT), AA 67-13, Cali, Colombia
d.pachico@cgiar.org

Received: December 22nd 2000
Accepted: May 15th 2001
Published: May 22nd 2001

With the advent of biotechnology and the emergence of property rights in genetic materials, the distribution of benefits from genetic resources is an issue of growing importance. Farmer's Rights is a concept that has emerged to seek to insure appropriate compensation to farmers for their contributions to crop improvement and genetic conservation. This paper evaluates crop genetic resources from two points of view: first in terms of royalty incomes that could be earned from seed sales of improved germplasm; and second, in terms of the potential increase in agricultural productivity that could result from improved germplasm. A model to calculate the upper limit of potential royalty incomes is developed and applied to the case of common beans (Phaseolus vulgaris L.). The paper finds that high-income gene-poor countries in the North would indeed make payments under a royalty model system to low-income gene-rich countries in the South, but the magnitude of these payments is quite modest. Much greater payments would flow between gene-poor regions of the South to a few gene-rich countries. The main gainers from the system for beans would be Mexico, Peru and Ecuador, while Brazil and sub-saharan Africa would bear the brunt of the payments. Most countries in the south would have far more to gain from increases in productivity due to utilizing germplasm, than they would from receiving royalty payments for the ownership of germplasm. This model could serve as a basis for calculating benefits sharing formula for the International Fund for Plant Genetic Resources as envisioned by FAO as the mechanism for compensating for Farmers' Rights.

Key words: Farmers' Rights, value of genetic resources, distribution of benefits from genetic resources, beans.

Crop genetic resources are the foundation of modern agriculture. When the original farmers in Asia, Africa, Europe and the Americas first learned to cultivate wild food plant species, they began a long process of plant improvement. Farmer selection increased the yield of crop species, changed plant architecture and adapted crops to new growing environments (Gepts, 1988). Millennia of farmer selection have endowed today's world with a wealth of crop genetic resources commonly called farmer land races.

In the last century, by building upon the inheritance of farmer land races, the science of genetics and plant breeding has vastly accelerated the process of plant improvement, leading to huge increases in crop productivity through breakthroughs like hybrid maize and semi-dwarf rice and wheat. This progress has been heavily dependent on the availability of crop genetic resources. It has been estimated that up to one half of the gains in U.S.A. agricultural yields from 1930-1980 can be attributed to genetic resources (Office of Technology Assessment 1987).

Despite some efforts to control ownership, for example by preventing the export or rubber tree seeds from Brazil, to a very considerable degree plant genetic resources were long treated as a common global heritage, freely accessible by anyone. More recently, however, there has been a trend towards establishing ownership rights in genetic resources. In 1930 the U.S.A. passed the Plant Patent Act establishing intellectual property rights (IPR) in some improved plants that are asexually produced. Subsequently in Europe in the 1960s and then in 1970 in the U.S.A., legislation established breeders' ownership rights in improved sexually reproduced plants. This was followed by U.S.A. court decisions in the 1980s that expanded IPR coverage to include patents over plants and animals (National Research Council 1993).

This process of privatization of what previously had been common resources led to a questioning of the propriety of treating farmer improved genetic resources, mostly from the South, as a freely available public resource while scientist improved genetic resources, mostly from the North, were increasingly protected as private property (Mooney 1979).

As a result, in 1989 the concept of Farmers' Rights was endorsed by the Food and Agriculture Organization International Undertaking on Plant Genetic Resources. This concept recognizes the enormous contribution of farmers to the conservation and development of plant genetic resources, and posits that farming communities should be compensated for the use of their genetic resources (Pistorius, 1997; Rural Advancement Foundation International 1997). Farmers' Rights are considered to be vested in the international community as trustees for present and future farmers (Food and Agriculture Organization 1994). An International Fund on Plant Genetic Resources was envisioned as the mechanism through which Farmers' Rights would be implemented.

It is generally accepted that low-income countries would be net beneficiaries from a system of compensation for Farmers' Rights in germplasm. Analysis at the level of regional groupings has confirmed this tendency (Kloppenburg and Kleinman, 1987). Yet it is clear that most countries are dependent on germplasm that originated outside their borders creating a situation of mutual interdependence among countries for their crop germplasm (Flores-Palacio, 1999). These previous studies have examined this issue in terms of mega-centers of diversity each of which include a number of countries. However, use of the mega-centers as a unit of analysis conceals what may be considerable differences among countries associated with a mega-center in terms of its contribution to genetic diversity. Moreover, a fuller understanding of the distribution of benefits among countries is needed to assist in the implementation of Farmers' Rights (Food and Agriculture Organization 1994).

 A variety of indicators are currently being assessed for their suitability as guides to mechanisms to share the benefits of germplasm (Food and Agriculture Organization 1999). Among the considerations proposed are the amounts of benefits that have been or might be obtained from plant genetic resources as well as the amount and kind of plant genetic resources that have been provided.

This paper seeks to assess in more detail the international distribution of benefits that might guide the implementation of Farmers' Rights in land race germplasm. It develops a conceptual framework for estimating such benefits, and the framework is applied to common beans (Phaseolus vulgaris L.) as a model crop for this analysis. The paper first attempts to quantitatively estimate the potential magnitude of benefit flows that would accrue to countries of origin of germplasm based on a market derived approach to calculating benefits. Second, it breaks down benefit flows by individual countries rather than regions, thereby revealing significant intra-regional variation. Next, estimates are made of the potential productivity gains that could come to different countries from improved bean germplasm. These potential benefits are contrasted with evidence about gains that have already been made from germplasm improvement. Finally, some implications are drawn and directions for future research are pointed out.

 A Royalty Model to Estimate Payments for Farmers' Rights

 Typically, incomes from intellectual property rights are accrued through a market based system of royalty payments. This reflects the willingness of users to pay for the use of the intellectual property while the holder of the IPR obtains a return based as a per cent of the value purchased in the market. To estimate the benefit flows associated with Farmers' Rights in land race germplasm this paper develops a simple model of the incomes that could accrue if payments were based on what would in principle accrue from a royalty system.

 (1) Y_i = A_g * S * P * R * D_i- E_i + O_i

where

 Y_i is the royalty receipts income from land race genetic resources in country i.

 A_g is the global area planted with seed using the genetic resources of country i.

 S is the amount of seed planted per unit area

 P is the price of seed

 R is the royalty rate earned for rights in land race germplasm

 D_i is the share of land race genetic diversity of country i

 E_i is the transaction cost of enforcing ownership rights in land race germplasm

 O_i is the spillover benefits from the use of land race germplasm to improve other crops

This model uses a market derived royalty approach to estimating an appropriate level of benefit sharing for farmers' rights in germplasm. In this model payments by users of germplasm are assumed to be made to countries of origin of germplasm as a per cent of the value of the seed price. Thus, income generated by the use of germplasm is the product of the quantity of seed sold that is based on the germplasm, the price of the seed and the proportion of the seed price that is paid as royalties.

Countries will not only "export" their land race germplasm for use elsewhere, but also countries will be "importing" land race germplasm for their own use. Even for countries with a wealth of genetic resources, self-sufficiency would be unattractive. It has been found, for example, that the anthracnose pathogen (one of the most important bean pathogens in Mexico) has co-evolved separately with its host in the two major centers of primary diversity. Mexican strains of anthracnose can attack Mexican bean germplasm, but Andean germplasm is highly resistant to Mexican anthracnose. Hence, Mexico's best strategy for dealing with several of its most important pathogens is to import novel resistance alleles from the Andean center of diversity (Pastor-Corrales et al., 1994). Thus as a simplifying premise, it is here assumed that most countries will be "exporting" germplasm in proportion to the amount of genetic diversity they have while they will be "importing" germplasm proportional to the area they cultivate of a particular crop.

 Consequently, the net income flows to a country of a royalty system in land race germplasm would be

 (2) B_i = Y_i - C_i

 where

(3) C_i = A_i * S * P * R*(1-D_i)

 B_i is the net benefits to country i from royalties in land race germplasm

 C_i is the payment made by country i for use of land race germplasm of other countries

 A_i is the area in country i planted with land race germplasm of other countries

This model will be first used to simulate a situation which provides a reasonable upper limit of maximizes the benefits that could accrue to countries for compensation for their role in developing and conserving land race germplasm, consistent with the concept of Farmers' Rights. The purpose of this initial analysis is tries to appraise in the best of circumstances, what the upper limit maximum of benefits that could be generated by such a system using a simulation of what a market based flow of royalty payments that would be roughly consistent with similar market phenomena

Thus, it is assumed that all the global area of beans, A_g, is planted with purchased seed that has been improved through the use of land race germplasm and consequently pays royalties for the use of this germplasm. This is necessarily an overestimate of what is likely to in fact occur because many farmers save their bean seed and do not purchase new seed each year, and thus would not pay royalties each year. Calculations are made on the basis of the average area planted to beans for each country in the period 1997-99. The seed rate, S, is taken conventionally at 50 kg./ha. and a standard seed price, P, of US $7.00/kg is assumed.

In actual practice, royalties, R, would accrue both to both owners of the germplasm that is used to generate the genotypes sold as seed, and also to those who performed the research that produced the new genotypes. It is likely that the relative shares of these two classes of royalties would be the subject of negotiation among the involved parties which would be the country owning the land race germplasm and the entity commercializing the final product of improved germplasm that has utilized the land race germplasm.

Typically, in such circumstances the relative contributions and the relative risks of the two parties are taken into account. These are reported to vary between 1-5% for genetic resources that are essentially wild and have not been previously characterized. For genetic resources that have had some ethnobotanical characterization, royalties may reach 5%. When there is information on the specific commercial character characteristics, for example, specific chemical compounds contained in the resource and their level of efficacy, royalty rates can reach as high as 15% of final product value. Royalties will range between 5% and 15% depending on the degree of specific information about the material (Laird 1993). The case of farmer land race crop genetic resources would appear to exceed that of ethnobotanically characterized wild plants because farmers have actively selected materials for demonstrated performance. Between 7.5% and 10% may be a reasonable range for farmer land race genetic resources, and here the figure of 10% will be used to represent the best of circumstances for payments for farmers' rights.

If there were asymmetry in the bargaining strength of the negotiating parties, for example, due to cost of enforcement or the superior market access of the seed companies, royalty shares could be biased in favor of the stronger party which might well be the seed company. This paper abstracts from any such bargaining process, and assumes that the countries of origin of germplasm receive royalties equal to ten per cent of the value of seed sold. This may be an overestimate of what might actually be earned by the countries in a market-based system.

Transactions costs, E_I, for the enforcement of intellectual property rights can be substantial and in fact have been cited as a barrier that disfavors small scale enterprises, communities or individuals from being able to establish protect their intellectual property rights. In order to portray what the maximum potential flow of royalties from these rights in land race germplasm might be, it is initially assumed that these costs are zero. This again clearly leads to an overestimate of the real net returns that these rights might yield.

For simplicity it is assumed here that there are no transgenic applications made with bean germplasm, that it, that bean genetic resources are used only to improve bean production. Treating, O_I, spillovers to other crops as zero is consistent with the call for a moratorium on the deployment of transgenic crops due to lack of certainty about their environmental or health risks (Oxfam Policy Department 1999). It is also an essential simplifying assumption because it is impossible to foresee the extent to which bean genetic resources might be used to improve other crops.

Here it assumed that royalty receipts are garnered by countries for their ownership rights in germplasm, proportional to the degree that their germplasm would be used in genotypes sown by farmers, D. Again in actual practice, negotiations on this point could be complicated because modern varieties have increasingly complex pedigrees, with ancestors typically coming from several countries. Although payments could be calculated relatively straightforwardly based on the per cent of ancestry coming from each country, this might not be fully acceptable. For example, germplasm from a given country might constitute only one of 16 ancestors, but contribute genes of particularly high value. In contrast, genes from another country might overcome only an occasional minor constraint, making it illogical that both countries receive the same royalty.

The procedure used here is to abstract from any negotiating problem and conduct a simulation as if countries received royalties proportional to the amount of genetic diversity they have. This in turn is based on the assumption that useful alleles are randomly distributed. In this conceptual model for common beans, then, a country's royalty receipts are equal to its share of useful diversity which is equal to its share in total diversity of land races.

Historically, most gains in crop improvement have come from land races. However, there are recent examples of introgression of useful traits from wild ancestors into cultivated materials. It is also known that genetic variability in wild materials is often greater than that present in domesticated germplasm (Gepts 1988). Consequently over time, particularly with improved techniques for gene transfer, the contribution of currently undomesticated germplasm to agricultural productivity, could be as greater or greater in importance as further use of land races. However, this paper deals only with land race genetic resources that have been selected, cultivated and conserved by farmers.

It must also be noted, that while wild materials are only found in the centres of origin of crop species, land races have a much wider dispersion. In the case of common beans, it is known that it has they have two centres of origin, Mexico and the Andes (Singh 1989). However, since the encounter of 1492, beans have spread throughout Africa, Asia and Europe, giving rise to the emergence there of unique diversity in the land races in the secondary centres of origin. While previous papers have considered only land race genetic diversity in the center of origin (Kloppenburg and Kleinman 1987), this paper takes a more complete approach by including diversity from the land races of both the primary and secondary centres of diversity.

The share of diversity in different regions and countries is measured by relying on prior work that selected a core collection for common beans (Tohme et al 1994). The core collection was devised to represent the distribution of diversity in common bean germplasm based on knowledge of variability in morphological, agronomic, and molecular characters as well as geographical data on the origin of accession. For this paper, then, the amount of variability in a country or region, is taken to be equal to its per cent participation in the common bean core collections for cultivated land races.

Based on the proceeding royalty based model for the value of Farmers' Rights, the estimated flow of annual benefits for common beans by major continental groupings was calculated. As noted above, the model is used in this paper to approximate the highest level of royalty flows as a basis for computing possible payments for Farmers Rights. It is assumed that all bean farmers annually purchase seed on which royalties are paid. It is assumed that there are no transaction costs and complete compliance with the system. A royalty rate is assumed that includes some valuation for the information contained in land race germplasm that is recognized as having been historically improved by farmers. Any actual market based system of royalty payments would certainly lead to lower benefit flows than calculated here. The model simulates a pattern of payments for these new genes that is proportional to the share of diversity in land races and wild ancestors by country or region of origin.

On this basis, Tropical America would owe the largest payments for Farmers Rights in bean germplasm since it has the largest area sown in beans, while sub-saharan Africa would owe the second largest amount (Table 1). More importantly, though, as the major source of variability in land races, Tropical America would earn the vast bulk of royalties, $48.2 million, so on balance would receive a positive net income of $11.7 million from common bean genetic resources.

Generally it is anticipated that the high income North would end up paying substantial royalties for agrobiodiversity to the gene rich South. This is indeed the case for the high- income countries of North America, which would have to make net payments under a royalty system for bean seed. Even taking into account receipts accruing from useful genes found in the diversity of their secondary center land race germplasm, the USA/Canada would pay $2.9 million annually. Somewhat surprisingly Europe would emerge as a net gainer from a royalty system in beans. Because beans are not an important crop in Europe, royalty payments would be quite low, only $1.5 million while due to its place as a secondary center of germplasm diversity, Europe would earn $3.0 million for a net income of $1.5 million. These results do not fully confirm the expected view that the wealthy industrialized countries would end up paying in a system of ownership rights for land race germplasm.

The biggest net payments under this system, through, would come from sub-saharan Africa. As the most important of the secondary centers of land race diversity, it would receive 21.9% of total royalties from land race genes, that being the region's estimated share of land race diversity. These royalty receipts of $5.5 million are, however, considerably overshadowed by projected payments $12.9 million, leaving Africa with a net bill of $7.4 million. The other two groupings of low income countries - East/South Asia and West Asia/North Africa, would also make net payments under a system of royalties for ownership rights in germplasm for common beans.

Thus, this analysis is consistent with the conventional view that in the aggregate, the high-income countries would pay out, while low income countries would earn net positive receipts. However, not all developing regions would gain. Only in the center of highest biodiversity will there be net gains. Further disaggregation by country illustrates that this pattern also repeats itself at the regional level. Table 2 shows estimated annual royalty flows for selected countries and sub-regions of Latin America and the Caribbean. These data portray remarkable disparities in net royalty income even within the region.

At the positive extreme, is Mexico which is the world's greatest source of bean genetic diversity. Although Mexico would earn $16.5 million annually, it would net only $7.4 million because as the world's second greatest bean producer, it would be paying royalties of $9.1 million for germplasm that is brought into Mexico. Peru, as the second most important home of common bean genetic diversity, reaps the second highest level of royalty earnings, $14.1 million. Since it would make very minor royalty payments due to the relatively small area sown to beans in Peru today, it emerges as gaining overall from royalties even more than Mexico. Ecuador, Colombia, Central America, and the Caribbean would also be net winners from royalty payments.

Brazil, and to a lesser extent Temperate South America, represent the opposite extreme. Brazil is the world's largest producer of beans, but it has quite a narrow genetic base in its land races. As a major grower of beans, and thereby a major user of bean seed and bean genetic resources, projected royalty payments from Brazil reach $20.6 million, with a net outflow of $19.2 million.

Clearly there are very substantial potential gains to access to genetic resources. Historically, there has been essentially open access to these resources and their potential benefits. In the context of increasing assertion of property rights over germplasm, the following analysis shows that, for the case of beans, payment of an access free would be worthwhile to producing countries if this was needed to secure continued access

At first glance, then, developing countries of Asia and Africa as well as Brazil and the Caribbean islands, would emerge as losers from a system of payments to compensate farmers for conserving bean genetic resources. Such a view overlooks the gains from increased productivity that would accrue to these countries due to improved germplasm. Several studies have shown that improved bean varieties can increase yields 30-40% (Janssen et al. Pachico and Borbon). Here, the conservative assumption is made that improved germplasm results in a 10% increase in production.

The benefits of increased productivity due to new germplasm are far greater than royalty payments. For example, although Brazil would pay $19.2 million in royalties, improved germplasm would lead to production gains conservatively worth $134.7 million, 7 times the amount of the royalty payments (Table 3). Total net benefits to Brazil would be $115.5 million annually. Similarly, Africa would gain over ten times as much from improved productivity as it would have to pay in royalties.

For countries that are net recipients of royalties as well as bean growers, the two classes of benefits are additive. Even for most of these countries, the benefits from improved productivity in beans considerably outweigh those from royalties. Only in Peru and Ecuador would royalty receipts exceed projected productivity gains. These countries have the unusual features of a wealth of bean genetic diversity combined with fairly small bean production. Since bean production is modest in these countries, productivity gains are slight. At the same time potential returns from germplasm are high due to substantial genetic diversity.

Firm policy guidance can not be judiciously drawn solely from the case of beans which is a crop of secondary importance. Consequently, this conclusion will first summarize the implications of this study of beans as a model crop, then it will touch upon the wider context.

A system that recognized farmers' contributions in developing and conserving common bean land race germplasm would generate income flows for those countries that are major sources of bean agro-biodiversity. Among the high income, gene-poor regions of the north, North America would indeed have to make payments to low income, gene-rich countries in the south, but Europe would actually be a net recipient of royalty payments. This occurs largely because many countries in the south are poorly endowed in bean genetic diversity. In particular, Africa and Asia would be net payers for germplasm, as also would be Brazil and the Caribbean.

Notwithstanding payments for the use of germplasm, the north would very much emerge as gaining from access to germplasm. Even on conservative assumptions for productivity gains, and liberal assumptions for payments as estimated from a simulation of a royalty system, the projected gains in productivity in the north would far exceed payments. This implies that economically it would be worthwhile for the north to pay for germplasm to the degree that breeding gains are dependent on access to wide genetic variability. If agreements about enforcement of farmers rights' in germplasm could be made workable, perhaps through multi-national agreement for an International Fund for Plant Genetic Resources, the system would be viable economically. It must be noted that it appears that many private sector breeders today rely almost exclusively on privately held germplasm and rarely access new germplasm. Since these firms would not face clear market incentives to participate in a germplasm royalty system, they would need to be forced to do so. Lack of political will to run counter to these interests has not facilitated the implementation of an International Fund for Plant Genetic Resources.

While such a system would clearly address fairness concerns of compensation for the domestication, improvement and conservation of agro-biodiversity, it would not automatically provide improved incentives for continued conservation. Moreover, just as patents provide ownership rights for a specified time period, it would have to be clarified whether farmers' rights would similarly be time bound or would exist in perpetuity.

A system based on what the market flow of royalties would be in bean land race germplasm would impose costs on low income countries that are poor in bean diversity and are major bean growers. This might seem disadvantageous to Africa, Asia, Brazil and the Caribbean. Nonetheless these countries would gain substantially from improved productivity, and these gains would be 7 to 25 times greater than projected royalty payments. It would, in principle, be worthwhile for gene poor low income countries to pay for access to genes.

Alternatively, rather than a uniform system of ownership rights, the system could be discriminatory and offer preferential access to germplasm among low income countries while charging full prices for germplasm to those who can afford it. In practice, germplasm exchange may be negotiated bilaterally and such a system could evolve. It might, though, raise complicated issues of enforcement of rights with respect to subsequent transfer of germplasm to third countries. Due to these considerations, probably the lowest cost system would be to assess payments to the International Fund for Plant Genetic Resources based on crop area while receipts from the Fund would be based on the share of genetic diversity.

Moreover, preferential germplasm access for low-income countries would do away with the bulk of the potential benefit flows in the case of beans. If bean germplasm were exchanged freely in the south without any charges, total payments from Europe and North America would amount to less than $6 million. This is such a small amount as to make the system hardly worthwhile.

In any case, most countries in the south have far more to gain from increasing the productivity of their bean crop than from a royalty system in bean germplasm. This suggests that they have most to gain from investing in agricultural research to develop improved germplasm. Not only is this generally true for almost all countries, but it is likely to be a good strategy even for the most gene rich countries. As noted above, royalties on property rights in improved germplasm include both compensation for genetic resources as raw materials, and compensation for the knowledge needed to identify and use these genetic resources.

Typically raw material prices are a fraction of final product prices, with most income going to value added in processing. Therefore, it would be highly advantageous for countries, which are origins of genes, to also develop them. This would not only contribute to their earning a larger share of royalties, but also could enable them to insure the development of their genes. Some valuable alleles may not be unique to a single country, and sometimes different alleles may produce a similar expression of a desired trait. Thus, it is in the interests of gene possessing countries to be able to expedite development of their genes. In addition, this would also contribute to their production agriculture, and in many if not most cases, this may yield benefits much larger than those from ownership rights in germplasm.

Since this paper deals specifically with the case of beans, it is useful to assess the degree to which this case is likely to be representative of other major crops. Royalty incomes from beans are modest in comparison to potential increases in productivity in low-income countries due to improved bean germplasm. To a significant degree, this is due to the fact that 79.6% of beans are produced in low-income countries. Obviously, the higher the share of production of a crop that occurs in low-income countries the relatively less important royalty revenues become compared to productivity increases. Many other major crops, like beans, are grown principally in low-income countries. Such crops include rice, sugar cane, cassava, banana, sweet potatoes, yam, millet, coconut and sesame. Significant revenues from ownership rights to this germplasm could only come about through horizontal transfers among low income countries, but not to any important degree as revenues from the north to the south.

In contrast, some major crops are grown mainly in developed countries: wheat (64%), maize (64%), potato (78%), barley (86%) and soybean (66%). Revenues from germplasm rights to these commodities would be relatively more important, compared to productivity gains than in the case of beans. It is likely, though, that as in the case of beans, royalties would accrue principally to a few countries in the south that are particularly rich in genetic variability. Thus, ownership rights in germplasm may be a significant benefit to some selected countries, for example, Mexico and Peru in the Americas, and Ethiopia in Africa. The rest of Africa and Latin America could expect little income from ownership rights royalties in agricultural crops.

Nonetheless, a system of ownership rights in germplasm could be in the common interest if it serves to provide improved incentives for the conservation of genetic diversity. Moreover, ownership rights do reflect a concern for fairness in the distribution of benefits as property rights are increasingly asserted in germplasm. Such a system could produce important income flows for some poor countries. Finally, though, most low-income countries, even the few with a wealth of genes, have far more to gain from a research capacity that enables them to more effectively utilize genetic diversity.

The author owes a debt to Geoff Hawton for the initial stimulus to pursue this topic. Joe Tohme provided the core collection data which were ably managed by James Garcia. Useful comments were received from Cary Fowler, Ian Sheldon and two anonymous reviewers. This research was conducted with the support of the CGIAR (Consultative Group on International Agricultural Research). Neither CIAT nor the CGIAR are responsible for any representation of fact, interpretations or views expressed in this paper which are the sole responsibility of the author.

Brush, S.R. (1992). "Farmers' Rights and Genetic Conservation in Traditional Farming Systems." World Development. 20(11):1617-1630.

Flores-Palacio, X. (1999). Contribution to the Estimation of Countries' Interdependence in the Area of Plant Genetic Resources. FAO, Rome.

Food and Agriculture Organization Commission on Plant Genetic Resources (1994). Issues for Consideration in Stage II: Access to Plant Genetic Resources and Farmers' Rights. FAO, Rome.

Food and Agriculture Organization Commission on Plant Genetic Resources (1999). Possible Formulas for the Sharing of Benefits Based on Different Benefit-Indicators. FAO, CGRFA-8/99/8, Rome.

Gepts, P., ed. (1988). Genetic Resources of Phaseoulus Beans: Their Maintenance, Domestication, Evolution and Utilization. Kluwer Academic Publishers, Dordrecht, Holland.

Janssen, W., S.M. Teixeira and M. Thung (1992). "Adocao de Cultivares Melhoradas de Feijao em Estados Selecionados no Brasil." R. Econ. Sociol. Rural, Brasilia 30(4):321-338.

Kloppenburg, J. Jr. and D.L. Kleinman (1987). "The Plant Germplasm Controversy." Bio Science. 37(3):190-198.

Laird, S. (1993) "Contracts for Biodiversity Prospecting" in W. V. Reid Biodiversity Prospecting. World Resources Institute, Washington, D.C.

Lesser, W.H. (1994). Attributes of an Intellectual Property Rights System for Landraces. International Academy of the Environment, Geneva. Working Paper No. RIOW

Mooney, P. R. (1979) Seeds of the Earth: A Private or a Public Resource. International Coalition for Development Action, London.

Mooney, P.R.. (1993). "Exploiting Local Knowledge: International Policy Implications." In W. de Boef, K. Amanor, K. Wellard and A. Bebbington, Cultivating Knowledge: Genetic Diversity, Farmer Experimentation and Crop Research. Intermediate Technology Publications, London.

National Research Council (1993) Managing Global Genetic Resources. National Academy Press, Washington, D.C.

Office of Technology Assessment (1987) Technologies to Maintain Biological Diversity. U.S.A. Government Printing Office, Washington, D.C.

Oxfam Policy Department (1999). Genetically Modified Crops, World Trade and Food Security http://www.oxfam.org.uk/policy/papers/gmcrop.htm.

Pachico, D. and E. Borbon (1987), "Technical Change in Traditional Small Farm Agriculture: The Case of Beans in Costa Rica." Agricultural Administration and Extension 26:65-74.

Pastor-Corrales, M.A., M.M. Otoya, A. Molina, and S.P. Singh (1994). "Patterns of variation in common bean for reaction to Collectotrichum lindemuthianum from Middle America and Andean South America and Sources of Resistance." Plant Disease 79(1):63-67. forthcoming.

Pistorius, R. (1997) Scientists, Plants and Politics International Plant Genetic Resources Institute, Rome.

Rural Advancement Foundation International (1997) Enclosures of the Mind: Intellectual Monopolies RAFI, Winnepeg, Canada.

Singh, S.P. (1989). "Patterns of Variation in Cultivated Common Bean (Phaseolus vulgaris)." Economic Botany 43(1):39-57.

Tohme, J., P. Jones, S. Beebe and M. Iwanaga (1994). "The Combined Use of Agroecological and Characterization Data to Establish the CIAT Phaseolus vulgaris core collection." In, A.H.D. Brown, T.J.L. van Hintum, J. Hodgkin and E.V. Morales. Core Collections of Plant Genetic Resources. J. Wiley and Sons, London.

Copyright remains with the author.

Published by Biostrategy Associates
Editorial Office: biopol@biostrat.demon.co.uk