Valuing Biodiversity: An Application of ''Green Accounting''
R. David Simpson
Roger A. Sedjo
Resources for the Future
As concern for environmental degradation grows there has been a corresponding growth in interest in "green accounting": constructing measures of economic activity that accurately reflect the costs of environmental degradation. The effects of such degradation are not immediately apparent. Rapid extraction of petroleum reserves, overfishing, or excessive logging, to give three examples, may result in high short-term values of earnings. Excessive demands on other types of resources may also generate short-term surges in earnings. Production of goods and services may be more plentiful in the short term if the processes of production are allowed to foul air and water, or diminish biological diversity. When viewed over the long run, however, each of these phenomena may result in a decline in future well-being. Accounting measures that do not incorporate such considerations can, then, yield very misleading results.
The realization that short-term flows do not necessarily measure long-term well-being is reflected in the principles of personal, business, and—perhaps to a lesser extent—national accounting. Accounting measures of "income" include corrections for changes in capital stocks. Rather than purporting to measure year-to-year flows of gross production or consumption, they reflect attempts to capture changes in long-term profitability, production potential, or well-being. These long-term changes are induced by changes in the ability to produce goods and services. The stock of capital (in combination with other factors of production) determines this ability to produce.
The measurement of capital can be a difficult problem. Capital is something that is used to transform one (or more) good into another (or others) without itself being transformed. This definition is, of course, too simple. If it were literally true that capital is not transformed with use (or if it did not deteriorate with age), one of the more difficult accounting questions would be side-stepped: we would not need to be concerned with the measurement of depreciation. The essence of the measurement problem, however, is that capital is not itself consumed or transformed. Since it is the services of capital, as opposed to capital itself, that is being consumed in production, we must measure the value of capital in terms of its ability to provide a stream
of services, and measure the depreciation of capital as a deterioration in that ability to provide services.1
Measurement of the value of capital and the losses arising from its depreciation could, in general, be rather difficult exercises. The value of a capital asset is the expected net present discounted value of the stream of services that it will provide over its useful life, plus whatever salvage value it might have at the end of that period. It is important to realize that this value is computed "at the margin." That is, when we are talking about valuing capital, we consider the net present discounted value of the stream of services provided by one incremental unit of capital. A common error is to confuse the average value—total value divided by total capital stock—with the value of an incremental unit. Doing so can give rise to what are sometimes referred to as diamonds-and-water paradoxes: how can it be that something that is so useful and essential (water for example) is so much cheaper than something that is of such limited practical importance (diamonds)? The answer is that the former is very common and the latter very scarce. We will suggest that many natural assets are like water in their abundance, their total value, and their low value at the margin.
Performing the calculations necessary to value an incremental unit of capital could be a daunting exercise. The value of the stream of services is a derived quantity. In order to compute it, one would have to know both the values to be assigned to the outputs that are produced with the use of capital and other factors and what the contribution of an additional unit of capital is to the production of the outputs. What might one expect the value of the stream of future services to be? What assumptions on future earnings are built into such an estimate? How do we factor in uncertainties about future market conditions and the life of the asset? What is the appropriate interest rate by which to discount future returns?
There is, however, a powerful tool to assist us in valuing many capital assets. If one is willing to assume the existence of more-or-less competitive markets for capital assets, he can use the information provided by these markets in assigning the assets value and in inferring the rates at which value is lost through depreciation. An accountant might avoid painstaking reviews of the many factors that determine the expected net present discounted value of capital assets if she is willing to assume that someone (or many someones) has performed such calculations, and they are reflected in the market prices of capital assets. The accountant can then assume that the value of a new capital asset is what it costs to purchase it, while the rate of depreciation is revealed by the difference between the prices of new and used assets.
This discussion of the measurement of capital and the use of market information to aid in doing so has been something of a digression. The issue at hand concerns the measurement of natural capital. Great difficulties arise in assigning values to natural capital precisely because the markets in which it is traded are limited. The point we want to make is not so much that it would be a good thing if there were more and better markets for trading in natural capital
(although it most certainly would be a good thing if there were), but rather, that unless or until there are such markets, "green accounting" exercises must remain extremely speculative.
This conclusion begs the question of the purposes for which environmentally accurate national accounting is to be used. It is true that properly constructed national accounts give a snapshot view of national income, although there might still be substantial disagreement as to whether or not what is being measured is in fact the—or even an—appropriate measure of national well-being. It is not clear that changes in national accounting practices will necessarily lead to improvements in economic performance, however. One might think of some cases in which more accurate accounting would feed back to better performance; overseeing the management of public lands, or monitoring the performance of international lending programs are examples that come to mind. In general, however, economic performance will be enhanced by getting incentives right. The national income accounts are a report card from which we can infer performance after the fact, but to modify behavior we must intervene at a much earlier stage. Nor does it seem likely that we can mollify proponents of growth at the expense of environmental amenities by cooking the books. It is unlikely that an increase in measured net national product would spur elation if the trend it plotted were contradicted by other aggregate (e.g., unemployment, wages) and individual (my job, my salary) indicia.
Having said this, there is, of course, no reason not to get the national accounts right, at least if doing so can be accomplished at a socially acceptable cost. We are only advising caution in the expectation of benefits to be achieved by doing so. An environmentally correct system of national accounts would, at best, indicate only the success or failure of efforts to allocated natural resources appropriately; other than serving as a warning that this is not being done, more accurate accounting cannot be expected to induce change in and of itself.
In the following section we very quickly review some principles of national income accounting and .discuss some of its controversial aspects. After that we discuss some of the special issues likely to arise in the depletion of natural assets, and, particularly, of what are called non-rival goods. We discuss a particular example in which national income accounting might be extended to biodiversity in the third section following. A final section concludes.
NATIONAL ACCOUNTING: A VERY BRIEF OVERVIEW
This paper will be largely devoted to the discussion of an example of the possibilities and potential difficulties of including a particular type of natural capital—genetic diversity—in national income accounts. For a discussion of the fine points of the conceptual rationale and practical details of constructing environmentally accurate national income accounts, the reader is referred to the excellent summaries prepared by other authors (see, e. g., Solow, 1992; Dasgupta and Maler, 1991; Ahmad, E1 Serafy, and Lutz, 1989; the papers collected by Costanza, 1991; and Krautkraemer, Pezzey, and Toman, 1993). We will, however, attempt self-containment by alluding to some basic principles of national product accounting.
A good starting point is with what it is that national accounting purports to measure. Regrettably, a narrow definition of even such a basic point as this is unlikely to be agreed upon by everyone. We will beg the reader to bear with us as we call this goal "long-run consumption
possibilities."2 Now "long-run consumption possibilities" may differ from "current consumption" as the latter may not recognize the depletion of resources necessary to maintain future consumption.
We will gloss over a great many important considerations by merely citing in a schematic form a result due to Weitzman (1976): national product, if properly measured, indicates the "stationary equivalent of future consumption" (Weitzman, 1976, p. 160). In other words, if Y* is this year's national product, the total net present discounted value of all future consumption would be the same as that arising from consuming Y* every year. Excessive consumption in the present—using up resources in consumption, or consuming more rather than replacing depreciated assets—results in a decline in the level of consumption we could enjoy in all future periods.
Another important concept is due to Hartwick (1977, 1978; see also Solow, 1986): constant consumption levels may be maintained even if some resources are exhausted if the value of the exhaustible resources consumed is offset by new investments in capital. Thus we have "Hartwick's Rule": intergenerational equity3 can be achieved by investing the scarcity rents arising from the depletion of natural resources in the acquisition of man-made capital assets.4 Hartwick's rule is only meaningful in circumstances in which man-made capital can be substituted for natural assets, however. This may well be the case for fossil fuels and minerals; Daly (1991; see also Daly and Cobb, 1989) and others have argued eloquently that it is not the case for biological assets and ecosystems.
We should expect, however, that if man-made capital is not substitutable for natural capital, and the latter is of vital importance in the production of the goods and services on which we rely for our continued well-being, then the value of an incremental unit of natural capital would increase as its scarcity increases. The problem is again that we do not have much information on which to infer such values. To make these inferences, we would have to have either markets in which natural capital is traded, or be able to infer the value of natural capital inputs in the production of other goods.
The question, then, is how one might go about measuring the value of, or the decline in, natural capital, and incorporate such results in the types of analyses suggested by Weitzman, Hartwick, and others. It is this issue that we will address in the remainder of the paper. In doing so, we put aside more complex—and possibly metaphysical—matters, such as whether or not the objectives implicit in these authors' analyses are in fact appropriate.5 While one might address the matters in either order, suggestions for the measurement and interpretation of national accounts must be both conceptually valid and practically feasible if they are to be of value.
GREEN ACCOUNTING AND ECOLOGICAL ECONOMICS
Most accounting exercises may be fairly mechanical. One observes the values of things that are bought or sold, adds up these values, and records the result. Some careful distinctions may need to be made in determining which items belong on which side of ledger lines, but the basic notion is one of recording rather than imputing. As we have said above, even more difficult calculations involving capital depreciation may be facilitated by reference to market transactions.
It is precisely in calculating the figures of greatest interest that we lose the guide of market experience, however. While there might be some argument that the way in which existing markets allocate most goods is fair, equitable, or even efficient, the motivation for "green accounting" largely concerns the treatment of resources that are not traded in existing markets. Even the exceptions prove the rule: measured harvest rates are deemed excessive either because those performing the harvest do not consider the damage they cause to other resources, or because property rights to the resources harvested are not well defined and a "tragedy of the commons" ensues.6 The former phenomenon describes the collateral damage
to watersheds and fisheries when logging occurs. The latter might describe the state of both fisheries and forests in many nations.7
Resources not traded in markets give rise to externalities: benefits or costs not reflected in the prices at which they are available to or from those receiving the benefits or incurring the costs. Present effects of environmental externalities will be reflected in current consumption and production. If, for example, pollution has degraded a fishery or acid rain is killing a forest, harvests from both resources will be diminished. In order to infer the loss of value of the capital asset, that is the fishery or the forest, however, one must know what are the effects of the externalities on the ongoing ability of the resource to produce valuable commodities.
If one can observe consequent declines in, for example, timber leases or other assets used in fishing, she might be able to infer the induced loss of value in the underlying asset. Again, however, it is often impossible to impute the value of environmental assets from observable indices. This is particularly true when we regard ecosystems as assets.
The value of a capital asset is determined by its contribution to the production of value in the form of other goods. Gross payments for the use of a capital asset will be determined by the incremental contribution of that asset in the production of other goods or services; the rental charged for the use of a shoe machine should be equal to the value of the additional shoes produced as a result of renting the machine. Net payments consist of gross payments less depreciation (or plus appreciation) in value. The purchase price of such a machine should be equal to the discounted present value of the incremental stream of payments that would be forthcoming from its use.
Shoe machines and natural assets are different types of "machines," however. Consider attempting to quantify the value of a rainforest and the loss in value as a result of deforestation. Recall that we need to compute the losses in future ability to produce other items as a result of deforestation. 'There are, of course, a myriad of "products" of a rainforest. One might be timber. To the extent that the values created by the forest in this context are reflected in marketed products and, moreover, that these values can be captured by the owners of the forest, we might hope that computation of the contribution of these elements to the national accounts would be little more than a report of efficient resource use.8 This would be unlikely to be the case with other rainforest "products." Carbon sequestration, watershed protection, and preservation of biodiversity are all aspects of rainforest performance that are not only not directly compensated, but involved in so complex an array of other production processes as to make the separation of its contribution virtually impossible.
Absent the development of markets in these areas—a deus ex machina for which we might earnestly hope, but for which realistic prospects are virtually nil—the best we might
expect are meaningful contributions in what is becoming known as the discipline of ecological economics. The study of economics has always lay at the intersection of the psychology of wants and the mechanics of production possibilities. From an economist's perspective, the most pressing need in order to do meaningful green accounting is for more concise statements from the natural sciences—and especially from ecologists—as to how ecological degradation diminishes production possibilities.
This is admittedly a very tall order. Inasmuch as the essence of the problem is its pervasiveness—the impacts of global warming and diminished biodiversity might be felt in all sectors of the economy—attempts to incorporate these factors in meaningful national accounts might prove quixotic. We cannot even say if it would be wise to make such attempts. If it is one's goal to complete national accounting in this way, however, it is important to recognize the enormity of the task.
VALUING RESERVES OF BIODIVERSITY
We are going to step back from the broad view to consider a much narrower issue. We have argued that accurate accounting is made much more difficult in the absence of functioning markets, and, by implication, market prices, for natural assets. This is clearly a long way off for many assets. It may be instructive to take a look at one area in which markets are coming into existence, however. This is, on one hand, an encouraging phenomenon: the establishment of property rights and the realization of economic value leads to incentives to protect natural assets rather than allowing them to be squandered. To the extent that protection is in fact taking place, we should suppose that little deterioration of natural capital would be reflected in national accounts. If, on the other hand, substantial deterioration were recorded, it would point to a need to husband the assets more carefully.
We will see in this example, however, that the fact that transactions are beginning to be observed is no guarantee that values will be identified and reported immediately and unambiguously. The appearance of property rights and sales, at least in their incipiency, does not assure that meaningful statistics will be forthcoming. The example we present is also interesting in that it demonstrates some of the areas in which economists—and national income accountants—might gain from greater instruction from ecologists.
The example we will use is that of indigenous genetic resources.9 Plants and animals produce a myriad of chemicals to enhance reproductive success, resist infection, overcome prey, and thwart predators. These chemicals can be of potentially enormous value in agricultural, industrial, and, particularly, pharmaceutical applications.
We refer to these naturally produced chemical compounds as genetic resources, as the recipe for their creation is encoded in an organism's genetic instructions. Useful compounds can either be extracted and employed in their naturally occurring forms, or can serve as "blueprints" for synthetic molecules. In either case, the natural molecules are important inputs in the
research and development process. Millions of generations of evolution by trial and error may result in combinations more ingenious than the designs of synthetic chemists. Random screening of natural compounds occasionally yields extremely valuable leads.10
In order to earn a return for the preservation and provision of naturally occurring molecules, countries in which they are found must enforce property rights in them. Just as intellectual property (patents, copyrights, and trademarks, for example) exclude non-payers from the use of inventions or innovations, a number or recent national statutes and international agreements are coming to establish the rights of states. to their genetic resources and the requirement of their prior informed consent before genetic resources are appropriated.11
Commercially useful compounds are most likely to be found in the areas in which biological diversity is greatest. It has been estimated that half of the world's extent species are to be found on only six percent of its land area: that covered by the tropical rainforests (Wilson, 1988). To the extent that destruction of rainforest habitat results in the extinction of unique genetic resources, then, the felling of these forests represents an irreversible loss of natural assets.12
As we have noted, however, these resources are coming to be assigned values, and it is to be hoped that this realization of values will motivate greater conservation. To say that the value of genetic resources is coming to be appreciated is not to say that there is an emerging consensus on the amount, or even the order of magnitude of these values, however. While several agreements for the sale of rights to access of genetic resources have been finalized in recent years, it is difficult to infer from these arrangements the monetary value of the resource being exchanged in the particular transaction, let alone the value of other undiscovered resources in situ.
A number of factors prevent the observation of the value of resources being exchanged in particular transactions. First, particular details are not divulged in public documents. Merck and Company, for example, has made public the fact that it has made an up-front payment of $1 million to Costa Rica's Insituto Nacional de Biodiversidad (INBio) for access to samples provided by INBio. Neither Merck nor INBio has revealed the number of samples or the rate at which royalties are to be paid in the event that a commercially valuable product is developed, however. Published sources have varied by an order to magnitude in their estimates of both
quantities. Similar uncertainties surround the values being exchanged in other, less celebrated, arrangements. Even if royalty rates were known, of course, one would have to know the probability distribution of expected values in order to infer the value of the resource in situ.
The second source of uncertainty in ascertaining the value of genetic resources in situ arises from the combination of inputs represented in a typical transaction for sample materials. Rarely does a pharmaceutical, agricultural, or industrial laboratory purchase the rights to access to a certain collection of samples only. More commonly, the good purchased is a group of samples that has been collected, classified, dried, ground, and/or extracted by the seller. The compensation offered for the material reflects not only the value of the resource in situ, but also payments for the labor, materials, and machinery involved in the other operations.
For these reasons, it is extremely difficult to infer the value of genetic resources from what is observed concerning existing transactions. In fact, existing efforts to impute such values have not concentrated on transactions, but rather on attempts to infer values from the profitability of successful products and the probability of finding a product that proves to be successful (Pearce and Puroshothaman, 1992; Principe, 1989).
We might attempt such a valuation exercise as follows. We will think of prospecting for genetic resources as a process of trying to find a "cure" for a certain condition. The "cure" might be analogous to the search for a particular industrial product or an agricultural strain. Inasmuch as we should be concerned with the preservation of unknown genetic combinations—uncatalogued species—it may be reasonable to assign to each the same independent unconditional probability of containing the cure. Then if we believe there to be n unknown species that might be tested to find the cure, the probability that at least one of them does contain the secret is equal to one minus the probability that none of them do:
1 -(1 -p)n.
What is the value of preserving one additional species as a possible source of genetic resources? Differentiating the probability of finding at least one cure with respect to n yields -1n(1 -p)(1 -p)n. Since p is likely to be a very small number, we can use the approximation to say that the contribution to success of the "marginal" species is approximately p(1 -p)n.
We do not want to take this example too far, but, for the sake of illustration, phytochemicals—chemicals produced by plants—are thought by many to be unique in chemical structure and for this reason particularly worthy of preservation. There are about 270,000 species of higher plants (Wilson, 1992). Taking the often-cited figure of one in 10,000 (see, e. g., Roberts, 1992) as the probability of success, the incremental contribution of the "marginal" species of flowering plant would be on the order of 1016. This figure must be multiplied by the value of such a discovery (net of research and development costs), and we should also allow for the possibility that the same organism will be tested at regular intervals for further applications. Nonetheless, even if we assign a fairly large number to the value in perpetuity of having extent genetic resources to be explored, the resulting figure for the value of a marginal species is unlikely to be large.
We have been discussing the value of a "marginal species," but conservation of unexplored species must generally take the form of habitat preservation. We might then ask how much of this form of natural capital we would lose by clearing a square kilometer of natural habitat. For help here we might appeal to the theory of island biogeography (MacArthur and
Wilson, 1967; Wilson, 1992). It predicts that the species diversity of a region varies as approximately the fourth root of its area. Let us suppose that this relationship is true and that the proportion of endemic species in an area is n. Then the endemic-species-to-area relationship might be expressed as
where Si is the number of endemic species, Ai the area, ßi is a constant giving the region's capacity to support species, and α a constant on the order of one-quarter. Differentiating with respect to area, we find that
(i. e., endemic species loss is proportional to initial density).
To get some idea of the relative magnitudes involved, we might consider the species densities in areas defined by Myers (1988, 1990) as biodiversity "hot spots." In only two of these (Western Ecuador and Southwestern Sri Lanka) does the density of endemic higher plants exceed one per square kilometer of primary forest preserved. Continuing in the same vein of heroic assumptions, then, even if we are talking about the preservation of a species of higher plant for which tens of billions of dollars in earnings might be anticipated, the expected value lost by losing a square kilometer of habitat for the purpose of preserving higher plants for pharmaceutical research might be negligible.
The point of this exercise has not been to argue that the value of natural capital involved in genetic resources is high or low. While our own guess is that it is relatively low at the margin, the real issue is the compounding of imprecision inherent in any attempt to measure these values. We could very well generate estimates of the value of the "marginal" square kilometer of plant habitat for use in pharmaceutical research that would differ by several orders of magnitude, by changing the underlying assumptions in not implausible ways.
These estimates are at best a small part of the picture, however. In our example we have considered only higher plants. Inasmuch as there are on the order of one hundred species of other organism per higher plant species, estimates of value would be rendered still less precise by considering their contribution. We also have not considered the relationship between numbers of species and genetic diversity. While economists have done some preliminary work in an attempt to elucidate these relationships (see, e. g., Polasky and Solow 1993), this is another area in which we need guidance from natural scientists.
Moreover, we have confined out attention very narrowly to values in pharmaceutical and other research applications only. It is not unreasonable to suppose that this is just the tip of the iceberg. Perhaps the greatest contribution of biological diversity lies in its ability to maintain itself. Some evidence suggests that diversity breeds diversity: complex webs of obligate relationships between species evolve, and the demise of one species may doom many. This
raises the spectre of ''ecosystem collapse.'' The curious layperson may be justified in wondering whether such a "collapse" is a profound tragedy or merely an occurrence that, while regrettable, is of no great lasting consequence. More to the point, the layperson may wonder if she should cast her own lot with those who take the former view or those who take the latter. In short, economists attempting a valuation of biodiversity have been given very little guidance from ecologists as to the magnitude of the losses, or even the risks, involved.
To recap, national product accounts should reflect both current production (or consumption) and changes in capital stock that reflect changes in our ability to produce and consume in the future. The valuation of capital stocks is often a difficult exercise. It may be made considerably easier if the accountant can impute values from observed market transactions. In the absence of direct market prices, the accountant may be forced to attempt an indirect imputation of values from estimates of the product of an incremental unit of capital. This exercise may become practically impossible if we are dealing with assets such as climate, ozone protection, and biological diversity, whose contributions are important in virtually all production processes and are, at present, unpriced in most. The accountant's task is rendered still more difficult by the unpredictability of the effects of global externalities. When even the probability distribution of outcomes is unknown, and most likely unknowable, we have to wonder if any prescriptions for valuation will be of any practical use.
We have presented as examples some of our own findings concerning the commercialization of genetic resources and the valuation of those assets. It might be suggested that developments in this field represent an encouraging trend which, furthermore, might demonstrate the practical irrelevance of green accounting. An argument to this effect might go as follows. Genetic resources are so plentiful as to have been, for most of history, of negligible value at the margin. As they are becoming more scarce and our technology for exploiting them is becoming more sophisticated, they are being protected by both national laws and international agreements, and their value is coming to be incorporated in market prices. As this process continues, transactions in these resources will come increasingly to be drawn into the national accounts.
As the values of other, still less conventional, natural assets come to be realized, markets for their allocation and preservation will also come into existence. This will happen at the points at which the establishment of property rights is efficient. Thus, again, we may expect efficient outcomes, and correct national product accounting would record economic activity accurately, but would be of little help in suggesting reallocations of resources.
This view is certainly a Panglossian caricature. Given the imprecision with which markets are now valuing such relatively concrete goods as genetic resources, it seems very unlikely that markets will soon be effective in valuing global environmental externalities. While institutions and instruments for allocating and conserving environmental and ecological assets should certainly be encouraged, it is vanishingly unlikely that we will see an allocation of such assets that can be defended as "optimal" in any meaningfully concise way. We are a long way, then, from being able confidently to construct indices of economic well-being by simply recording transactions.
The best policy might combine a faith in the ability of markets to do many things right, a great deal of caution in dealing with resources that are not, and perhaps cannot soon be, traded in markets, and healthy skepticism concerning the reliability of the entire national accounting apparatus. Accounts might reasonably be expanded to include losses in natural capital in those instances in which values are recorded or might reasonably be backed out from observed statistics for closely related assets. In other, more speculative areas, we might maintain the hypothesis that these assets are not recorded in the accounts because they have yet to realize sufficiently high values on the margin to warrant the establishment of markets. At the same time, however, it would be prudent to recognize that national product estimates should be taken with several grains of salt. Further deterioration of natural assets should be allowed only to the extent that relatively large tangible gains are very likely.13
Abroad, Y. J., S. E1 Serafy, and E. Lutz (eds.), 1989, Environmental Accounting for Sustainable Development (Washington, D. C.: World Bank).
Costanza, R. (ed.), 1991, Ecological Economics: The Science and Management of Sustainability (New York: Columbia University Press).
Daly, H. E., 1991, "Elements of Environmental Macroeconomics," Chapter 3, 32-46 in Robert Costanza (ed.) Ecological Economics: The Science and Management of Sustainability (New York: Columbia).
Daly, H., and J. Cobb, 1989, For the Common Good (Boston: Beacon Press).
Dasgupta, P., and G. Heal, 1979, Economic Theory and Exhaustible Resources (Cambridge: Cambridge University Press).
Dasgupta, P., and K. G. Mäler, "The Environment and Emerging Development Issues," Paper prepared for the World Bank's Annual Conference on Development Economics, 1990.
Hardin, G., 1966, "The Tragedy of the Commons," Science 162, 1243-48.
Hartwick, J. M., 1977, "Intergenerational Equity and the Investing of Rents from Exhaustible Resources," American Economic Review 67, 972-74.
Hartwick, J. M., 1978, "Substitution Among Exhaustible Resources and Intergenerational Equity," Review of Economic Studies 45, 347-54.
Krautkraemer, J., J. Pezzey, and M. A. Toman, 1993, "Economic Theory and 'Sustainability,'" in Daniel Bromley (ed.), forthcoming, Handbook of Environmental Economics (London: Basil Blackwell).
MacArthur, R. H., and E. O. Wilson, 1967, "The Theory of Island Biogeography" (Princeton: Princeton University Press).
Myers, N., 1988, "Threatened Biotas: 'Hot-spots' in Tropical Forests," The Environmentalist 8, 187-208.
Myers, N., 1990, "The Biodiversity Challenge: Expanded Hot-Spots Analysis," The Environmentalist 10, 243-256.
Pearce, D., and S. Puroshothaman, 1992, "Protecting Biological Diversity: The Economic Value of Pharmaceutical Plants," CSERGE Discussion Paper GEC 92-27.
Pindyck, R., 1991, "Irreversibility, Uncertainty, and Investment," Journal of Economic Literature 29, 1110-1148.
Polasky, S., and A. Solow, 1993, "Measuring Biological Diversity," working paper, Woods Hole Oceanographic Institute.
Principe, P., 1989, "Valuing the Biodiversity of Medicinal Plants," 79-124 in O. Akerele, V. Heywood, and H. Synge (eds.), The Conservation of Medicinal Plants (Cambridge: Cambridge University Press).
Reid, W., (ed.) Biodiversity Prospecting: Guidelines for Using Genetic and Biochemical Resources Sustainably and Equitably (Washington: World Resources Institute).
Sedjo, R. A., 1991, "Tropical Forests, Property Rights, and Environmental Values: Economic Concepts and Real World Complexities," working paper, Resources for the Future.
Sedjo, R. A., 1992, "Property Rights, Genetic Resources, and Biotechnological Change," Journal of Law and Economics 35, 199-213.
Simpson, R. D., and R. A. Sedjo, 1993, "The Commercialization of Indigenous Genetic Resources: Values, Institutions, and Instruments," working paper, Resources for the Future.
Solow, R. M., 1974, "Intergenerational Equity and Exhaustible Resources," Review of Economic Studies Symposium, 29-45.
Solow, R. M., 1986 "On the Intergenerational Allocation of Natural Resources," Scandinavian Journal of Economics 88, 141-49.
Solow, R. M., 1993, "An Almost Practical Step Toward Sustainability," Invited Lecture, Resources for the Future, 8 October 1993.
Weitzman, M. L., 1976, "Welfare Significance of National Product in a Dynamic Economy," Quarterly Journal of Economics 9, 156-162.
Wilson, E. O., 1988, "The Current State of Biological Diversity," Chapter 1 in E. O. Wilson (ed.) Biodiversity (Washington: National Academy Press), 3-18.
Wilson, E. O., 1992, The Diversity of Life.