One of the arguments often made for the maintenance of genetic diversity among agricultural crops and their wild relatives is that such diversity acts as a sort of "insurance policy" against the effects of unanticipated risks. New pest infestations or changes in climate, for example, can result in large crop losses or other reductions in yield unless resistance or adaptability can be bred in. In addition to this insurance function, there is also the more mundane matter of agricultural improvement. The broader is the set of materials from which breeders can draw to improve cultivated varieties, the more valuable those varieties will be.

The basic economic principle involved in the valuation of genetic material is straightforward, but reducing this principle to actual implementation is far from simple. Rather than launching immediately into a discussion as to why this is "far from simple," I will defer these matters to the final section of this chapter. I will begin instead by discussing some highly simplified, but illustrative models. I will present two basic models because, as I noted above, two considerations motivate concern for maintaining genetic diversity in agriculture. First, dramatic events could largely or perhaps totally wipe out a crop. Pest infestations or diseases are examples. Resistance to such threats is a qualitative characteristic. Qualitative characteristics often are related to the presence or absence of a single gene. Second, crops can be improved or adapted to small changes in circumstances by "optimizing" the combination of genes they contain. Characteristics involved in such incremental improvement or adaptation are generally linked to large numbers of genes. Such characteristics are categorized as quantitative characteristics, and involve attributes such as height and weight.'

In what follows I will develop two very simple models of the valuation of genetic resources. The first will treat considerations with respect to the potential to provide qualitative attributes, and the second, quantitative attributes. Some basic principles are the same between each. First, the social value of genetic improvement, either qualitative or quantitative, is related to the change in benefits arising from the particular improvement. Second, the contribution of a larger pool of genetic resources is, in expectation at least, to enhance the characteristics of those individual organisms chosen for cultivation. The principle of marginal analysis that underlies economic valuation, then, implies that the social value of genetic resources is related to the expected difference in attributes between the best individuals drawn from larger as opposed to smaller sets.

I will illustrate these points in the sections that follow. My intention here is more illustrative than descriptive. Thus, while I will provide an example drawn from some empirical work with which I have been involved, I will not attempt to provide an actual estimate of value. Again, I will defer a discussion of "how things really work" to a later section, and confine myself to a stylized example. That example is as follows. Consider a situation in which a pool of individual organisms of size N is being evaluated for their potential to develop a superior variety for commercial cultivation. I will suppose that:

1 The actual observed values of quantitative attributes also depend on environmental circumstances. I will abstract from these for present purposes.

1. A single parent is identified, and any number of offsprings can be developed from this single parent organism. Moreover, the attributes of the parent are replicated exactly in the commercial offspring. Note that I am abstracting from both sexual reproduction and environmental variation in making these assumptions.

2. The parent is selected for a single attribute (although this might be a complex index incorporating a number of dimensions). In other words, I will not be considering a situation in which different organisms are selected for different purposes or growers care about diversifying their risks by planting different varieties.

3. I will suppose that the selection is made for purposes of planting a single generation of offspring. As I will argue later, one of the most complicated aspects of valuing genetic resources concerns the weight to be assigned to the contribution of one generation to the propagation of others. For the purposes of expositional clarity, I am going to abstract from this consideration for now.

Much of what I write here may be interpreted as applying to situations in which crops are propagated by conventional methods (although assumption 1 above might only literally be accomplished via large-scale cloning). With the advent of the gene-splicing methods of biotechnology, new methods of inserting valuable attributes into commercial varieties might be adopted. Rather than simply propagating the "best" of a set of parent plants, one might pick and choose among the genes of many members of the set, inserting only the "best of the best" in the commercial cultivars. This may presume, however, greater knowledge of genetic function than is now, or will likely soon be, available.

More generally, the effects of developments in biotechnology on the economic value of genetic diversity may be difficult to predict. On one hand, the potential scope over which the useful genes of one organism might be applied is greatly increased when genes can be inserted, rather than having to be bred into only those organisms that are sexually compatible. On the other hand, however, the same principle works in reverse: The number of potential sources of genetic material for use in the improvement in any one particular crop is greatly expanded.

In the next section I discuss the basic economic value of genetic resources in crop improvement. The second section that follows illustrates the application of this valuation framework to qualitative improvements. The third section illustrates its application to quantitative improvements. Following that, I present an application using data from "provenance trials" of teak trees in Thailand. The fifth section discusses the impediments to applying such approaches to "real world" valuation problems, and the sixth briefly concludes.

0 0

Post a comment