13
Genetic Resources: Assessing Economic Value

The improvement of plant and animal genetic resource is increasingly an international activity. Improvements in the techniques for breeding and applied genetics (modern biotechnology) enhance the value of genetic resources. The difficulty is in measuring the value. This chapter reviews evidence that provides general support for the proposition that investment in the collection, preservation, and management of genetic resource to support crop improvement is economically sound.

Genetic resources are valued as public consumption goods in the same way that scenic lakes and mountains are. Therefore, their conservation and preservation, usually in in situ states, are also valued. They are also of value in plant and animal breeding programs. This is an indirect or derived value, because it is derived from their contribution to economically improved plants and animals. This value is termed their producer good value. (The distinction between the consumption good and producer good value is effectively the same as the nonuse value and use value distinction in the valuation literature.)

VALUING GENETIC RESOURCES

The fundamental difficulty in measuring the value of genetic resources is that genetic resources are seldom traded in markets. Hence, prices cannot be observed except under exceptional circumstances (usually when proprietary rights have been established). Analysts



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies 13 Genetic Resources: Assessing Economic Value The improvement of plant and animal genetic resource is increasingly an international activity. Improvements in the techniques for breeding and applied genetics (modern biotechnology) enhance the value of genetic resources. The difficulty is in measuring the value. This chapter reviews evidence that provides general support for the proposition that investment in the collection, preservation, and management of genetic resource to support crop improvement is economically sound. Genetic resources are valued as public consumption goods in the same way that scenic lakes and mountains are. Therefore, their conservation and preservation, usually in in situ states, are also valued. They are also of value in plant and animal breeding programs. This is an indirect or derived value, because it is derived from their contribution to economically improved plants and animals. This value is termed their producer good value. (The distinction between the consumption good and producer good value is effectively the same as the nonuse value and use value distinction in the valuation literature.) VALUING GENETIC RESOURCES The fundamental difficulty in measuring the value of genetic resources is that genetic resources are seldom traded in markets. Hence, prices cannot be observed except under exceptional circumstances (usually when proprietary rights have been established). Analysts

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies use two basic methods to place value on such nonmarketed goods. The first is the contingent valuation method, in which the analyst seeks to elicit indicators of what people would be willing to pay for a public consumption good such as germplasm. This method is suited to measuring germplasm's value as perceived by the general public, such as in nature reserves. The second is the hedonic pricing method (the productivity method), in which the economic value of the nonmarketed good is estimated or inferred from the value of the marketed good in which it is contained or to which it contributed. This method is suited to germplasm uses in breeding. Few evaluation have been made for the value of the public consumption good of genetic resources. It appears likely, however, that the demand for conserving genetic resources may be strongly affected by income level. Thus, in general, poor people (as expressed by their organizational leaders) might be less willing to pay for the pure conservation of genetic resources. This would be particularly true in the poorest countries where other concerns rank higher. By contrast, if the source of value for genetic resources is based on them as producer goods, the picture is much different. Because the proportion of the income of poor people that is devoted to food is high, poorer people generally place a high value on plant and animal improvements that could increase production and hold down prices. Hence, the genetic resources that help to produce such improvements should be highly valued by the poor. Because plants and animals are generally traded in markets, an economic value can be placed on them. Furthermore, the value of improvements in these plants and animals can be determined directly. Hedonic pricing methods can be used to relate the value of such improvements to the genetic resources and other activities that were used to produce them. In this chapter, hedonic pricing methods are used to estimate the value of rice germplasm resources in a specific geographic setting as producer goods; their value as consumer goods is not estimated. These estimates are then used to consider the likely similar values of genetic resources in other crops and animals. The extension of the rice analysis to other commodities or other settings is subject to considerable uncertainty. THE PROBLEM OF LACK OF MARKETS The fundamental characteristic of germplasm resources that differentiates them from other natural resources or goods is that they are usually replicable at a very low cost. This means that the scarcity

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies of a particular genetic resource may be transient. A genetic source of resistance to a plant disease (for example, grassy stunt virus in rice) may be very valuable, because more rice can be produced when the resistance gene is incorporated into rice varieties. Once the source is discovered (in this example, in Oryza nivara, a wild species in the International Rice Research Institute [IRRI] collection), it can be replicated easily and cheaply in rice plants. Thus although initially scarce, it may quickly become abundant. A private enterprise might not be able to earn a substantial income from selling germplasm resources under these conditions of unrestricted replicability. A firm might, for example, collect landraces and wild species of rice, identify particular traits and characteristics, then seek to sell particular germplasm resources to plant breeders. However, once the genetic resources (via seeds or other reproducible materials) are available to a single breeder, they will be reproduced and distributed. Thus their price will fall. This is the justification for supporting collection and management of genetic resources as well as plant and animal breeding and research in the public sector. The simple economics of demand for improved seed indicates that as the price declines, more seed is demanded. Two types of costs are associated with supplying improved seed. The first are those of developing the improved genetic combination (the variety). These are the initial costs of locating, developing, testing, storing, and classifying the genetic resource and are fixed, in that they do not depend on the number of units of seed sold. Average fixed costs then decline with growth in sales. Costs of the second type are those of replicating, treating, and packaging the improved genetic combination in seed units. If this replication is easy (for example, if farmers can save their own seed), the marginal variable cost, that is, the cost of replicating the genetic combination in an additional seed unit, is very low. If many suppliers of seed are competitively selling the seed, they will not be able to charge a price that is above this marginal cost. If this price is below average fixed costs, no private firm could capture a return to the plant breeding program through seed sales. Remedies for Market Failure There are two remedies for this market failure. The first is to limit sales of the improved genetic material and charge a higher price to buyers. If replicability can be controlled through natural means (for example, seeds produced by hybrid corn plants themselves produce

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies plants of much less value than the original hybrid), the supplier of genetic characteristics, including landraces and advanced lines, could charge a price sufficiently high to cover costs (see Chapter 12). In the absence, or even in the presence, of strong intellectual property rights, the second, and historically more common, remedy to this problem is to establish public sector institutions to engage in plant and animal improvement and in the collection, classification, preservation, and use of germplasm resources associated with this activity. Naturally occurring germplasm resources are generally not subject to intellectual property rights protection. Even when markets for crop varieties exist, markets for germplasm resources do not exist. Thus, most germplasm resources are collected, catalogued, and maintained in public sector collections. Many private sector collections exist, but these are usually composed of parts of public sector collections and specific advanced breeding materials developed by a private firm. PRICE EVALUATION METHODS Improved plants and animals are produced by systematic breeding (genetic recombination) and selection activities. Germplasm resources constitute a type of capital or source material which, when combined with other forms of capital (fields, buildings, and equipment), research labor (crossing and selection), and breeding technology, produces superior plants and animals. The process is subject to uncertain outcomes (stochastic) and has a trial-and-error aspect. (Evenson and Kislev [1975] have presented a stochastic model for agricultural research.) Nonetheless, these processes are systematic, and the role of germplasm resources is relatively well understood. The relevant germplasm resources can be classified into two groups: (1) the stock of naturally occurring genetic resources (these include landraces, wild species, and related materials) and (2) recombinations or advanced stocks of materials that have been made in past systematic breeding programs. These recombinations may be commercial varieties of different vintages or advanced lines that may or may not be commercially developed. Both types of resources are important. The recombinations may represent years of systematic breeding work and development. Such recombinations are usually based on a small subset of the naturally occurring germplasm resources and usually form the parent materials in breeding programs.

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies Farmers of the plains of northern Thailand transplant upland rice seedlings. Credit: Food and Agriculture Organization of the United Nations. Modes for Introducing Germplasm Resources into Advanced Cultivars There are effectively two modes by which genetic resources are selected for incorporation into advanced breeding stocks. The first mode is through the building of breeder collections, that is, of working collections used for strategic crossing purposes. Most breeders maintain a relatively small set of cultivars for crossing purposes. The local breeding program typically keeps some landrace material and a set of advanced materials. Local breeders rely on information from national and international sources and from the managers of germplasm collections to identify promising new materials. Chang (1976) and Hargrove (1978) have described this process for rice. National breeding programs may maintain larger numbers of germplasm resources (landraces and wild species). They seek to produce new varieties

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies suited to fairly large regions, but they also produce advanced germplasm materials for more local breeding programs. International breeding programs concentrate even more on germplasm development and seed to provide national systems with advanced materials (Chang, 1976). The second mode is the incorporation of distinctive genetics traits into breeder cores. For example, several serious pest outbreaks of rice such as tungro virus and green leafhoppers (Nephotettix spp.), affecting rice variety IR-8 in 1969 and a grassy stunt virus outbreak that occurred in 1977, precipitated collectionwide searches of germplasm resources by IRRI (Chang et al., 1975) and others for genetic traits that conveyed resistance to these diseases. The Hedonic Pricing Method The hedonic pricing method used is a statistical procedure in which the value of plant or animal improvements is statistically associated with the germplasm and breeding stocks and other inputs that contributed to the improvement. The value of the improvement is measured in terms of relative amount of product per unit of input (for example, the yield per hectare of a new variety relative to those of older varieties is an index of the value of the new variety) or in terms of the change in the costs of production of a unit of the commodity. The genetic resources can be quantified in terms of the complexity of advanced breeding lines and the number of landraces in the pedigree of the variety. Other inputs into the improvement process include the labor associated with the process and capital equipment. The usual procedure entails a statistical multiple regression analysis to ascertain the value of the improved plants or animals as a result of the contributing inputs, that is, germplasm stocks, breeding and selection, labor, and other research program resources. From this regression, the marginal contribution of inputs can be estimated in value terms. From these estimates, the value of collecting and maintaining genetic resources stocks can be inferred. For example, if it was found that improved crop varieties were essentially based on recombinations of existing advanced materials and that no new germplasm resources were included in successful varieties, this would indicate that there would be little value to expanding existing collections of landrace and wild species. On the other hand, if new germplasm material is being incorporated into improved varieties, this would indicate that there is value to maintaining germplasm resource collections. If the genetic resources in question were found in the unusual or fringe materials in the collection,

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies this would indicate that there is value to an expanding collection including such materials. RICE: AN EXAMPLE OF PRICING METHODS Pricing methods have been used to establish a value for rice germplasm resources in India (Gollin and Evenson, 1990). This section summarizes that application. Institutional Background Cultivated rice falls into two species, Oryza sativa and O. glaberrima. The former is the common Asian cultigen, whereas the latter accounts for a small fraction of African rice production. In addition to these two cultigens, the genus Oryza includes about 20 wild species, although some scholarly disagreement remains on the exact number (Chang, 1985d). It is estimated that about 140,000 cultivars or types of rice exist in the world today (Chang, 1985d). These include cultivated varieties of O. sativa and O. glaberrima as well as landraces and wild species. About 85,000 of these are in a long-term storage facility at IRRI. Seeds of these germplasm resources are catalogued according to their agronomic and genetic characteristics, and they are kept under conditions of low temperature and humidity (Chang, 1976, 1987). Unimproved materials from the collection are evaluated by different disciplines at IRRI and are also sent out freely to scientists around the world. In addition, new germplasm sources may be incorporated into the pedigrees of improved lines that are sent out from IRRI through a variety of testing programs (Hargrove, 1978). Other rice germplasm collections are maintained by various national programs and by some regional centers (as in India). India has three national-level breeding programs that work on rice improvement: the Directorate of Rice Research in Rajendranagar, the Central Rice Research Institute (CRRI) in Cuttack, and the Indian Agricultural Research Institute (IARI) in New Delhi. In addition, there are 21 state programs, some of which have branch stations. The national-level programs generally perform the more technically difficult work of making crosses involving landraces and wild species. The national programs also tend to perform more experimental scientific work and to screen larger quantities of material for certain useful characteristics, such as disease resistance and pest tolerance. The national stations have the largest collections of germplasm in India. CRRI maintains a collection of about 15,000 accessions of rice. Of an estimated 60,000 to 65,000 accessions held in different collections

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies in India, perhaps 18,000 accessions are held at Rainpur, another main germplasm storage location. The National Bureau of Plant Genetic Resources, which has recently been given a mandate to collect and classify the germplasm resources being held in different locations around India, has perhaps 12,000 accessions at present (Gollin and Evenson, 1990). Analysis of Released Rice Varieties in India From 1965 to 1986, Indian rice breeders released a total of 306 rice varieties for planting in India. (These included varieties developed in earlier years.) This number also includes 27 high-yielding varieties requiring more intensive chemical inputs (so called green revolution varieties) that were actually developed at IRRI but that were released in India. These 306 varieties were the result of approximately 20,000 crosses made by Indian (or other) breeders since independence from the United Kingdom in 1947. Pedigree Analysis A pedigree analysis of each released rice variety was undertaken. It traced lineage back to the original genetic resources in the variety (in most cases landraces, but in some cases wild species). The characteristics emphasized in the development of each variety were recorded. These included disease and insect resistance, stress tolerance, and agronomic and grain quality. Many varieties emphasized more than one characteristic. Disease resistance tended to be the most sought after characteristic; this was followed by insect resistance, but no single characteristic dominated the breeding strategies. The analysis led to a quantitative description of varieties in terms of year of release, releasing institution, characteristics emphasized, parent and grandparent combinations, number of landraces in the pedigree, and the number of generations from crosses of landrace materials. Varietal Releases A steadily increasing trend in varietal releases was seen from 1965 to 1975, with approximately constant releases since then (Table 13-1). There were 27 rice varieties developed originally at IRRI and 6 varieties developed in other foreign breeding programs that have been distributed evenly over time.

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies TABLE 13-1 Varieties of Rice Released, by Year   Number Released   Average Number of Year of Release India IRRI Other Foreign Landraces Generationsa 1965 2 0 1 1 2 1966 2 1 0 2 1 1967 2 0 0 2 1 1968 6 0 0 2.5 1.5 1969 6 1 0 2 0.9 1970 12 1 1 3.8 2.4 1971 8 1 0 3.6 2.2 1972 17 5 0 4.2 2.6 1973 14 0 0 4.1 2.9 1974 7 0 0 4.0 2.7 1975 14 3 1 5.4 3.4 1976 17 2 1 4.1 2.7 1977 12 0 0 4.2 2.8 1978 11 4 0 4.3 2.9 1979 15 1 0 5.8 3.3 1980 20 0 0 4.3 2.7 1981 11 3 0 6.7 3.9 1982 23 2 1 6.1 4.0 1983 14 0 0 4.8 2.9 1984 7 0 1 5.5 3.6 1985 31 1 0 6.1 3.9 1986 22 2 0 8.7 4.6 NOTE: IRRI, International Rice Research Institute. a Generations refers to the average number of generations since introduction of the oldest landrace in each pedigree. The average number of landraces in each pedigree and the average number of generations since the oldest landraces in each pedigree by year of varietal release are shown in Table 13-1 (the number of generations is a useful index of genetic complexity). These data show steady growth over time in pedigree complexity. The early, high-yielding rice varieties released before 1970 had relatively simple pedigrees. Recent varietal releases are much more complex, with as many as 27 landraces and as many as 12 generations of crosses going back to original landrace material. Foreign germplasm is found in almost all varieties released in India, even though the country is the source of a large share of the world's rice germplasm resources. (IRRI is the major source of introduced germplasm.)

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies Institutional Releases The Central Variety Release Committee releases rice varieties deemed to have broad regional potential. Releases by the states tend to be more location specific. Most major state programs released 10 or more varieties from 1965 to 1986. Approximately 100 of the 306 varieties released since 1966 were planted in 1984. Of these 306 varieties, 118 were released after 1980. Landrace Appearance Table 13-2 tabulates landraces by the year of their first appearance in a released variety. These data show that the landrace content for Indian varieties has been expanding (and, by inference, that if TABLE 13-2 First Appearance of Various Landraces in Indian Rice Pedigrees Between 1965 and 1986   Landraces Appearing in Pedigrees of Released Varieties for the First Time Year Number Percent of Total Released Total Number, 1965 to 1986 1965 1 (0.6) 1 1966 5 (3.0) 6 1967 3 (1.8) 9 1968 6 (3.6) 15 1969 7 (4.2) 22 1970 8 (4.8) 30 1971 7 (4.2) 37 1972 17 (10.1) 54 1973 7 (4.2) 61 1974 2 (1.2) 63 1975 11 (6.5) 74 1976 11 (6.5) 85 1977 6 (3.6) 91 1978 8 (4.8) 99 1979 5 (3.0) 104 1980 13 (7.7) 117 1981 5 (3.0) 122 1982 11 (6.5) 133 1983 5 (3.0) 138 1984 4 (2.4) 142 1985 12 (7.1) 154 1986 14 (8.3) 168

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies breeders were constrained to work only with the landrace materials being used in breeder collections as of about 1970, they would have been far less productive). Origins and Inclusion in Varieties For purposes of further analysis, these landrace materials are classified according to origin (Indian and foreign) and further categorized into pre-1975, post-1975, and specialized groups, with the percentage inclusion in varieties noted (Table 13-3). Pre-1975 and post-1975 landrace categories entered varieties through the building of breeder collections, that is, working collections developed for strategic crossing purposes. Regression Analysis of Data The price evaluation analysis used entails a statistical regression relating a measure of varietal improvement in farmers' fields to factors expected to be associated with varietal improvement as described above. District-level measurements of rice yields are available for India. Yields are general productivity indexes and may be influenced by both varietal and nonvarietal factors. Accordingly, nonvarietal research activities and other increasing investments in rural infrastructure must be considered. For India a two-stage regression analysis was pursued by using data for 240 districts. The first stage was designed to estimate the relative contribution that overall varietal improvement made to productivity TABLE 13-3 Proportion of Pre- and Post 1975 Materials of Indian or Foreign Origin Included in Released Varieties Landcare Proportion of Planted Area Indian origin   Pre-1975 0.62 Post-1975 0.11 Specialized 0.12 Foreign origin   Pre-1975 0.84 Post-1975 0.07 Specialized 0.03

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies growth in rice. If it cannot be established that modern high-yielding varieties actually contribute to productivity growth, there is little point in attempting to identify germplasm resource effects, although use of such resources may be essential just to maintain, rather than increase, productivity in the face of new insect or disease pressure. (This maintenance effect is included in the productivity estimate discussed below.) However, if it is shown that varietal improvement does affect productivity, one can proceed to the second stage, where germplasm content variables can be incorporated into the analysis. The first stage estimates showed that varietal improvement was a significant determinant of rice yields for Indian districts from 1959 to 1984. The dependent variable was the rice yield for the district and year relative to the average rice yield for the district from 1957 to 1960 (Gollin and Evenson, 1990). The following independent explanatory variables were included: The proportion of land planted to modern varieties released since 1966. This variable was included to test whether varietal improvement actually affected productivity. By 1986, 60 percent of the rice area in India was planted to modern varieties. Indian agricultural research, a stock variable reflecting the contributions of nonvarietal public agricultural research. Indian private sector research and development relevant to agriculture. Agricultural extension services. Literacy of farmers. Roads, a road density variable. Markets, a measure of regulated market infrastructure. Irrigation investments. A specialized program providing additional extension and infrastructure to farmers. The estimates showed that varietal change contributed more than one-third of the rice productivity gains realized over the period after the green revolution (1972 to 1984). Having shown that rice varietal improvement did contribute to rice productivity, the second-stage analysis can be justified. In this stage, variables measuring the genetic content of varieties actually planted by farmers were substituted in the analysis for the variable representing genetic content of modern varieties. The analysis was undertaken only for the most recent 5-year period (1979 to 1984), because the relevant data for earlier periods were not available. The dependent variable, rice yield, was indexed relative to the average

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies yield from 1972 to 1974. Thus, the analysis focused on yield changes in the post-green-revolution period. It sought to determine whether rice yield gains after the 1972 to 1974 period were systematically related to the germplasm contents of the varieties planted by farmers. For each district, germplasm content variables were defined for the land actually planted. These germplasm content variables were defined for five clusters of variables: source of breeding materials, varietal characteristics, parental origin, pedigree complexity, and landrace content. Separate regression analyses were then undertaken for each cluster to estimate the impact of each cluster of genetic variables on yield. From these estimates were derived the percent change in yield as a result of a 1 percent increase in the variable at the expense of its reference group. Details of the analyses are given in Gollin and Evenson (1990). The analyses of source variables indicated that the varieties released by the Central Variety Release Committee had a higher impact than those released by the states. Varieties of foreign origin that were not released directly by IRRI were also associated with higher productivity. Varieties released directly by IRRI had lower impacts on productivity. Many of these were the early green revolution varieties, or more recent varieties released in response to specific problems thus designed more to maintain than to increase productivity. The varietal characteristic variables indicate that varieties stressing grain quality, specific agronomic characteristics, or stress tolerance had higher yields than those with no selection strategy for these characteristics. By contrast, varieties with resistance to disease and insects appeared to have lower yields. However, this may reflect pest or disease incidence as these were not monitored. The impacts of parental origin were particularly complex. In general varieties with mixed foreign and Indian parentage appeared to have better yields than varieties in which both parents were advanced and of foreign origin. Analysis of pedigree complexity indicated that varieties with higher landrace contents and more generations of development had higher yields. This supports the contention that landrace genetic resources are valuable breeding materials. Estimates for the impact of landrace content in varieties more directly address this implicit value of genetic resources. The data were compared to material based exclusively on old (pre-1975) national landraces and showed that new landrace resources generally had significant positive impacts on productivity of released varieties.

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies National landrace materials (which entered pedigrees after 1975) also contributed to increased yields. This suggests that systematic and strategic incorporation of more landrace materials into breeders' active lines has a payoff. The estimated impacts of landrace materials obtained from specialized national or international searches were quite large. This has implications for genetic resources management. The genes identified in such searches are typically uncommon and found in materials with few or no other valuable traits. The probability of discovering accessions with such uncommon traits is increased if collections are large. The evidence of their value to increasing yield indicates that the general activities of collection, preservation, and maintenance of landraces and related wild species are of value. Economic Analysis Forty-one percent of the Indian rice acreage in 1984 was planted to varieties containing pre-1975 international breeding materials. While this acreage includes both dryland and paddy rice, in practice only irrigated and lowland rainfed paddy rice varieties have been genetically improved. These were, effectively, the original green revolution genetic resources. The analysis showed that a 1 percent expansion in the use of these old international materials would still increase average rice yields in India by 2 percent. The more pertinent question, however, is what the post-1975 materials have contributed to yields. This can be calculated from the 1984 levels for the variables examined. The calculation indicates that yields were higher for all of India by 5.6 percent than they would have been had only the germplasm resources present at the time of the original green revolution been available to breeders. The total increase from modern varieties in the period was 13.4 percent. It can be inferred then that reworking of the original (pre-1975) genetic materials alone would have contributed 7.8 percent to yields (Gollin and Evenson, 1990). The 5.6 percent added yield increase realized from the newest added germplasm resources up to 1984 is likely to continue to increase as the area planted to the newer varieties expands. If the 5.6 percent is conservatively treated as having been realized at a 0.5 percent rate over the 11 years leading up to 1984, a cost-benefit or rate of return analysis can be undertaken, under the conservative assumption that no further gains after 1989 will be realized. To do this, an estimate of the time lag between incurring costs and realizing the yield gain benefits is required. With that time lag, the present value

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies of the stock of genetic resources can be computed, given a discount rate. Present Value of Rice Germplasm Stocks The Indian study indicated that the time period from initial breeding strategy until resultant varieties were fully adapted by farmers was about 9 years. Thus in 1975, the availability of the genetic resources enabled this contribution of 0.5 percent of Indian rice production (worth US$50 million in 1990 dollars) to be realized 20 years later (9 years from initial breeding to varietal release plus 11 years to full adaption). Since this contribution is maintained from 1995 and thereafter, it can be treated as a benefit "stream" and its present discounted value (as of 1975) computed. Using a 10 percent discount rate, this value is US$75 million. (At 5 percent this value is US$377 million.) These values can then be compared with the costs of maintaining and operating the genetic resources collections. They can also be compared with the costs associated with developing a larger collection. The costs of maintaining the larger world collection of rice material at IRRI are roughly US$700,000 annually (Chang, 1989). The costs of maintaining the Indian collections are roughly US$300,000. Thus, the economic value of genetic resources in India vastly exceeds the costs of maintaining them. If India were to invest, for example, US$20 million over a 10-year period to expand its collections further, this would add to the annual costs of maintaining the collection by the amortization of the US$20 million plus added costs. Even if this raised annual costs by US$3 million per year, the value produced by such additional resources would more than justify the expenditures. Indeed, since much of the estimated value of new materials emanates from the fringe materials, the value of a nearly complete collection relative to its cost is probably higher than the value of the present collection relative to its costs. This presumes, however, that the difficulties of managing very large collections are addressed (see Chapter 5). Contributions of International Collections Germplasm collections of rice are international and are exchanged internationally. Collections at IRRI and elsewhere contributed to productivity gains in India. Conversely, Indian germplasm has contributed to productivity growth in other countries. A global calculation for irrigated rice using the Indian estimates shows that a 0.5 percent increase in output per unit of input would be value at about US$400

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies million. Using the same 20-year time lag applied to the Indian data, the present discounted value of this benefit's stream is US$594 million (discounted at 5 percent, it is US$3.015 billion). The values can be compared with the current annual costs of rice germplasm maintenance of perhaps US$10 million. If the current rice collections were brought to near-complete status over the next 10 years at a cost of US$10 million per year, this could, when amortized and adjusted for the larger collection sizes, raise the annual costs of these collections to US$30 million per year. This is well below the estimated present annual value of the benefits. Even if the Indian estimates overstate the impact of new genetic resources by a factor of 10, the economics of moving to a near-complete collection justify doing so. However, as the collection approaches completion, the increased effort and cost needed to get the last few samples would probably be unjustified. IMPLICATIONS FOR OTHER CROP SPECIES The study of rice in India given above indicates the fundamentals of placing economic value on genetic resources as producer goods. For rice, a distinction was made between a general strategic search by breeders for new genetic materials and a specialized search for genetic materials to address specific problems associated with vulnerability or lack of diversity. This distinction is important, because the general strategic search for new resources provides the case for the economic value of genetic resources already in collections. This value can justify the maintenance and preservation of most materials. The economic case for the collection and preservation of fringe genetic resources, that is, wild species and wild relatives, is based on the second search motive. For rice, the evidence shows that both general search and specialized vulnerability-based strategies have paid off. It is useful to maintain this distinction in considering other commodities. In fact, the calculations for rice suggested that if there is an economic case for maintaining an ex situ collection, the case for maintaining a near-complete collection is actually stronger than the case for maintaining a partial collection, unless the costs of maintaining collections of fringe materials are extraordinarily high. The economic calculation for rice showed that the present value of genetic resource impacts 20 years hence was V × I × D, where V is the value of the commodity, I is the impact estimate (.005 for rice), and D is the discount factor when returns are discounted at 10 percent (1.486 = 1/6.73 × 10). The estimates for rice may be unrepresentative for other commodities.

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies It appears to be reasonable, if not unduly conservative, however, to consider the value .005 to be an upper bound and the value .002 to be a lower bound (this is approximately the lower bound of statistical significance) on a more general estimate of I. Given this, the estimated genetic research collection and maintenance costs can be compared with the commodity value. The discounted range of values for the ratio of annual costs of maintaining and operating the genetic research system to the value of the commodity for which the system is economically justified is thus .0029 to .0074 (.002 × 1.486 = .0029 and .005 × 1.486 = .0074). How do actual costs to value ratios compare with this range? For rice, the ratio of current costs to commodity value is roughly (US$15 million/US$100 billion) = .00015, which is well below the is roughly lower bound, .0029. Doubling of annual costs to US$30 million to achieve a near-complete collection raises the ratio to .0003, but this, too, is below the lower bound. Calculations for wheat and maize appear to be very similar to those for rice. The economic case for moving to a near-complete collection may be strong for these crops as well. For the less important cereal grains, the ratio in question depends on the size of the potential collection and its replication in different locations. Barley appears to have a large number of potential accessions relative to the crop value. On the other hand, most accessions have already been acquired for barley. For other crops for which the commodity value is some fraction of that for the big three crops (rice, wheat, and maize), the economic case for near completion of collections is just as strong, as long as the collection's costs are the same fraction of the costs for collection of the more important crops. The case for moving to near-complete collections is strong for most grains of significant economic importance. The case for roots and tubers is somewhat less certain. Essentially the same arguments apply. Collection costs are higher because the low-cost seed storage options that allow for low-cost grain systems are not available. It is likely, however, that the major root crops have economically viable ratios. RECOMMENDATIONS The calculations made here provide general support for the proposition that investment in the collection, preservation, and management of genetic resources as producer goods provides a good return. Rates of returns are likely to be high, probably on the order of those

OCR for page 303
Managing Global Genetic Resources: Agricultural Crop Issues and Policies for research in general. This is likely to be the case for all major agricultural commodities. For minor commodities, the case may not hold if costs are high. Yet, for many commodities of low economic importance, conservation costs may also be low. For very minor commodities (annual production costs of US$50 million), the case for conservation cannot be made unless the maintenance costs are very low. There well may be other compelling arguments for conserving genetic resources of minor crops, however. Biotechnology has opened more options for using germplasm and minor crops may well contain some important genes. National and international investments in research related to collecting, managing, and using genetic resources should be increased. The analysis suggests that with reasonable collection costs, if it makes sense to maintain a collection it makes even more sense to achieve a near-complete collection in a cost-efficient manner. For most commodities, uncollected materials probably have more value than most materials currently in collections. However, it does not make economic sense to incur extremely high costs in the search of the last few resources. For a well-designed program of collections, however, costs are likely to be reasonable. Improvements in the techniques for breeding and applied genetics enhance the value of genetic resources. Furthermore, these developments likely enhance the value of fringe materials most. They may also expand the species that can be considered as genetic resources considerably beyond the present state. The improvement of crops and livestock genetic resources is increasingly an international matter, because most valuable genetic resources cross international borders. International agricultural research centers, such as those of the Consultative Group on International Agricultural Research, and other agencies are critical institutions because of their role in facilitating this transfer. With the growth of private sector collections and intellectual property right protection, free exchange of genetic resources may be impeded. These developments could increase the importance of such public institutions for promoting international exchange of genetic resources. The study on which this analysis is based is the first of its type. Given the economic impacts of genetic resources, it is important that further economic studies be made to develop a more complete body of evidence on which to base policy.