Sustainability, Income Measurement, and Growth
by Salah El Serafy
Sustainability
Sustainability is a concept that has figured prominently in the Brundt-land Report, though it has proved difficult to define without ambiguity. Within the Brundtland Report itself we find more than one definition, but the one that has since been most quoted is the following: "Sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs." The Report goes on to clarify sustainable development:
It contains within it two key concepts:
• the concept of "needs," in particular the essential needs of the world's poor, to which overriding priority should be given; and
• the idea of limitations imposed by the state of technology and social organization on the environment's ability to meet present and future needs.1
The reference to limitations of technology and social organization and to meeting "essential needs of the world's poor" in the above quotation, and a later statement that "concern for social equity between generations . . . must logically be extended to equity within each generation," while appealing to many readers, emphasizes the complexity of Brundtland's sustainability, both as a concept and as a pragmatic guide to policy action. As discussed later in this chapter, the vagueness of definition of Brundtland's sustainability should not detract from its valid concern for addressing distributional issues, which are viewed rightly as an integral part of the environmental problem. This ambiguity is by no means confined to Brundtland. A more recent attempt to clarify what sustainability meant to different authors yielded a bewildering array of definitions.2
The search for a precise meaning of sustainability has remained elusive, with a growing awareness now that for practical purposes sustainability should be perceived in approximate terms only.3 It is certainly evident that the use of the expression "sustainable growth" has become more frequent in recent development literature, replacing the older unqualified "growth," in an apparent attempt to impart the notion that growth should be kept within environmental limits. The Brundtland Report represents one of the early attempts at this usage. It is true, however, that such environmental limits remain undefined in a manner conducive to practicable policy guidelines, but I return to this point later.
Brundtland's Impact
In retrospect it seems that while the Brundtland Report made a great impact on world leaders and environmentalists alike, its impact on economists has been rather modest. This is not to deny, however, some influence it has had on economic policy, indirectly through the political forces it has motivated." The attention that has been given to global environmental issues since the publication of "Our Common Future" may be a product of its political impact.5 There is also the growing coverage of environmental issues in economic work practically everywhere, which may be traced back, at least in part, to Brundtland's publication.
Environmental Accounting for Sustainable Development
While Brundtland was in gestation, an initiative was developing, spurred by the United Nations Environment Program (UNEP) and the World Bank, to revise national income calculations in order to reflect in them environmental concerns. The coincidence in timing is remarkable between the World Commission on Environment and Development, which began its work in December 1983 and reached its conclusions in mid-198 7, and the UNEP—World Bank workshops, which sought improved national income measurements. This parallel effort also began in 1983, reached a crucial stage in 1988, and is still progressing in a number of directions.6
During the past two decades most countries have been calculating their national income according to guidelines, issued in 1968 by the United Nations Statistical Office, generally known as the System of National Accounts (SNA). These guidelines paid practically no attention to the fact that, in order to reckon income properly, the SNA must account for natural resource erosion and environmental degradation. The old system treated much of the antipollution expenditures as final expenditures that would raise income, instead of regarding them as necessary intermediate costs that should be charged against the final products. It also failed to take account of environmental disasters when they occurred. It treated natural resources, particularly those emanating from the public sector, as a free gift from nature, reflecting in the accounts mainly their direct extraction costs and any valuation, over and above extraction cost, that the uneven and heterogeneous free market deigned to attach to them.
Worst of all, the SNA failed to distinguish between value added by factors of production and sale of natural assets such as forestry products and petroleum. Through income measurements patterned on the SNA, many natural-resource-based developing countries were made out to have higher income than they actually had and to be growing at rates that obscured their true economic performance. Besides, the accounts failed to reflect the fact that the current levels of prosperity they were enjoying would not last, since the basis for such prosperity was progressively being eroded. False accounting resulted from mixing in the flow accounts elements of natural capital that should have been kept separate from current income. Such income measurements, where they occurred, covered up economic weaknesses that needed urgent attention, thus misdirecting economic policy. Countries where natural resources contributed significantly to fiscal and external balance failed to make essential adjustments and ended up allocating to consumption too much of the receipts they obtained from selling their natural assets. Many of them assumed too much external debt for their own good. Domestically, relative prices moved against tradeable goods, resulting in a lamentable shrinkage of non-natural, resource-based activities. Little wonder that so many resource-rich developing countries that should have benefited from the exceptional improvement of their terms of trade in the 1970s found themselves in the 1980s hardly better off than they had been before.7
At the UNEP-World Bank workshop held in Paris in November 1988, experts from various national statistical offices met with economists and others who had been investigating the topic of environmental national accounting, and for the first time a consensus was reached that natural resources and the environment were indeed important and likely to become more so in the future; that natural accounts should reflect the stress on the environment that had become increasingly evident; and that a set of environmental satellite accounts needed to be elaborated and attached to the new SNA core accounts, with the view of reflecting environmental considerations. That 1988 meeting was a watershed from which significant developments were to flow. Further work since then, conducted in cooperation with the United Nations Statistical Office, has led to the acceptance of the notion that when the revised SNA (expected in 1993) came out, it would recommend compiling a set of satellite environmental accounts showing to the extent possible the changes that occur from year to year in the state of the environment and attempting a recalculation of national income to reflect such changes. This national accounting adjustment initiative, which still continues, has provided a bridge between some of the objectives of the environmentalists and the work of the economists.
Sustainahility and Income
If properly measured, income is sustainable by definition. From an environmental angle, errors in measuring income can be viewed as coming largely from wrongly mixing in income certain elements of natural capital, and from confusion of inventory liquidation with depreciation of fixed assets.8 A person or a nation cannot continue to live at the same material level if present enjoyment is obtained at the cost of liquidating capital. As capital is eroded, the ability to maintain the same level of consumption into the future is undermined. That is why, from its inception, the accounting profession has insisted that for profit and loss calculations, whether for individuals or corporations, capital must be "kept intact." To the accountant, keeping capital intact never meant that capital should be preserved in its original state (the preservationist argument), but only that allowance be made out of current income in order to restore capital to the extent that it has eroded. Unless capital is "maintained," future income would inevitably decline. By extension of the same argument to the area of national accounting, keeping capital, including environmental capital, intact for accounting purposes requires adjusting income to reflect capital deterioration. Again, this does not mean that the accountant is advocating that capital should be kept undisturbed, or in the language of some environmentalists, that it should be "preserved" in its existing state, since the very essence of sustaining economic activity relies on utilizing capital to generate future profits or income. There is little disagreement now on extending the same principles that apply to manmade capital to environmental capital, save on the application of those principles to the special case of de-pletable resources, which cannot be renewed or recycled, but whose stock steadily dwindles as it is used up in the productive process.
That the environment can be viewed as natural capital is easy to perceive, both as a sink for wastes and a source of materials and energy.9 Wastes have been dumped in rivers and seas, buried on land, and dispersed in the atmosphere in the belief that such natural receptors had an unlimited capacity to receive them. As production has grown, this capacity has clearly been seen to be limited and has also become limiting. There is thus growing acceptance of the notion that the polluting activities should bear the full costs to society of their pollution. If standards are set for acceptable levels of pollution, the cost of achieving such standards, even if not actually incurred, can be used as a measure of environmental deterioration on account of pollution and be charged against income as depreciation.
As a source of materials, the environment should also be brought into income calculation. A distinction is clearly needed between resources that can be regenerated and others that cannot. Nature, and society in cooperation with nature, can amend, restore, or regenerate fish stocks, forests, soils, and the like. Where such regeneration falls short of theoretical or practical rates that would maintain such capital intact (that is, at its original level at the beginning of each accounting period), shortfalls should be deducted, as depreciation, from gross income calculations. Some problems of valuation would present themselves, but the guiding principle throughout should be pragmatism and approximation, since precise measurement is still, and likely to remain, an unattainable goal. Ecologists, likewise, should attempt measurements of sustainable yield in the same spirit of providing pragmatic and prudential estimates, instead of letting their quest for precision become an obstacle that would render their measurements irrelevant for policy.
With respect to depletable minerals such as fossil fuels, which cannot be meaningfully restored once they are used, applying the same approach of depreciation as in the case of renewable resources would be inappropriate. Such resources represent known wealth that can be liquidated over a variable time span depending on their owners' needs, their expectations of future prices, and the state of the market. While productive capacity is depreciated, existing inventories are used up or liquidated, and it would be wrong conceptually to include the proceeds from selling inventories in gross income. And it is equally wrong to believe that, in order to correct for their inclusion in gross income, all that is needed is to deduct the decline of the stock from the wrongly calculated gross income to arrive at a correctly measured net income. If such an approach is adopted, neither the gross nor the net income will be correctly measured. The gross will be inflated by asset sales that do not represent value added, and the net will be underestimated, since the whole contribution of the exploitation activity to income is removed as capital consumption or depreciation. If, on top of such erroneous accounting we add windfalls from upward reestimation of reserves, and deduct from income downward adjustments of these reserves, we arrive at very dubious and gyrating estimates of income that are as meaningless as they are useless, either for gauging economic performance or for guiding economic policy. A depletable resource's contribution to income requires special handling.
Accounting for Depletable Resources
In as much as the reserves of depletable resources are ascertained, they should be treated as inventories, not as fixed capital. Inventories can be drawn down to exhaustion if that is perceived by their owners as economically desirable. The proceeds from their exploitation in any one accounting period should, as a first step, be viewed as proceeds from asset sales, not as value added. If the owners draw down all their known reserves in one year because they believe this to be best in light of their assessment of future prices, it would obviously be wrong to include all such proceeds in their gross income for that year and to deduct the diminution of the asset, equivalent to the same amount that had been included in gross income, so that net income from this activity is shown as zero. Now that the owners have substituted for the subsoil asset, say, a bank account, true income is the interest that can be earned on the new account. Alternatively, the owners may sink the proceeds from selling the mineral assets in new material investments whose returns would represent true income. In this way capital liquidation would be kept, as it should, outside the flow accounts.
Following a proposition by the late Professor Sir John Hicks, which he put forward half a century ago,10 it was possible for me to calculate that part of the proceeds from a wasting asset that must be reinvested in alternative assets so that the yields obtained from such reinvestments would compensate for the decline in receipts from the wasting asset. Using a discount rate and the amount extracted from the reserves in any one year relative to total reserves, I was able to indicate the proportion of the proceeds that can be reckoned as true income, the remainder—a kind of a Keynesian user cost—having to be set aside and reinvested to produce an aggregate stream of constant future income. The user cost part is a capital element that should be expunged from the gross domestic product (GDP) or gross income and therefore would not appear in the net domestic product or net income either. If fresh deposits are located, these would affect the flow accounts only indirectly through the change of the reserves-to-extraction ratio—that is, providing a longer lifetime of the asset so that the income part rises and the user cost part falls."
This proposal, which is slowly gaining ground among economists, is still by no means generally accepted, either by them or by the national income statisticians.12 Many of the latter, even if convinced, would still prefer to preserve old-time series of erroneously calculated GDP along conventional lines on the argument that all that is required is to deduct natural resource "depreciation," equivalent to the entire diminution of stock, from the gross product to show a more sustainable net product that would amount to nil. The conceptual confusion implied by such procedures has already been mentioned. If one must persist with this confusion for the sake of preserving old-time series, the user cost, as explained above, would be the appropriate estimate of "depreciation."
The Limited Function of Accounting
Accounting, by its nature, has a limited function. It is essentially a backward-looking activity attempting to sort out from the behavior of economic units during a past period elements from which an arithmetical history is compiled. This usually takes the form of a snapshot at a point in time (a balance sheet of assets and liabilities) and a flow, during a certain period (most commonly a year), of net results of the economic activity concerned: profits and loss for an individual or a corporation and value added for a nation. Economists have often misunderstood the functions of the accountant, and his concern—perhaps obsession— with keeping capital intact, often challenging the accountant's meaning of keeping capital intact and the accuracy of his measurements, since such a concept of capital maintenance inevitably refers to the future. The Hicksian definition of income itself, whose author insisted that it was merely a rough guide for prudent behavior, has wrongly been criticized on the economist's usual ground of concern with precision and his (the economist's) forward—rather than the accountant's backward—orientation. Hicks's income has been said to be incapable of being "directly measured" and even that it is "not suited to an accounting of what happened in the past" either.13 Whereas Hicks stressed the accountant's quest for at.proximately defining a level of income that can be devoted to consumption with concern for a sustainability built around the reuse of capital in the future, other economists have tended to hanker after a precise level of sustainability that the Hicksian approach, with its emphasis on future income sustainability, obviously cannot meet, partly because the future will always remain unknown.
Economists and accountants have different, but perfectly reconcilable, objectives. In their measurements the accountants seek approximations, assume constant technology, and posit that the future will be a continuation of the past. In practice, technology does change, and the future is a little different from the past. But this does not matter much, however, since the accountants' accounting period is seldom more than one year, and every new year brings with-it new facts and some fresh technology that the accountants have to, and certainly do, take in their stride.
Businesses and Governments
The approach I have proposed for estimating income from depletable natural resources, which relies on setting aside part of the proceeds from the sale of natural capital to be sunk in alternative investments so that they may yield a constant stream of future income, begs the question as to what kind of alternative investments are available, and whether for the sake of sustainability such investments will always be available. Here we leave the ex post world of the accountant and enter the realm of ex ante analysis.
Individual owners of depletable resources usually see to it that part of their receipts, whether in the form of depletable allowances or set-asides, are reinvested so that the owners can continue in business. Whether or not their new investments should be in the same line of business they are already in, or diverted toward other lines, depends on many factors. If the price of the natural resource they own rises in reflection of its growing scarcity, thus indicating the opportunity for investment to produce substitutes based on renewable resources, and if such a course is economically feasible, the owners may well continue in the same line of business. But frequently the market would fail to reflect the resource's growing scarcity, and its price would fail to rise. Besides, technologies for producing substitutes may not be available, and if available may not be economic at the prevailing set of prices. Thus we often observe a tendency for diversification away from one-product business on the part of large corporations that exploit natural resources.
Some environmentalists would prefer that the user cost entailed in the exploitation of a depletable natural asset be invested in a "twin" project that would supply a renewable substitute for the same depletable source.14 But in light of the considerations just mentioned, such "twinning" or "pairing" may not be attractive to private owners. On the other hand, there is nothing against society as a whole indicating its desire to raise the overall level of savings and investment so that these become consistent with the objective of future income sustainability and also subsidize pioneering and experimental ventures in pursuit of finding renewable sources to replace the diminishing ones. This can be done by insisting, through appropriate monetary and fiscal policies, that the user cost of depletable resource exploitation should be added to current investments. The extra investments would be guided to socially desirable ventures, such as natural resource restoration and maintenance, through a carefully designed system of taxation and subsidies.
User Cost and Income Identities
Consider what happens to the usual identity that income, Y, is the sum total of consumption, C, and investment, /. Denoting user cost by the letter U, we can write: Sustainability, Income Measurement, and Growth 7 3 Y=C + I |1| Adjusting for user cost, equation (1) becomes: y- u = {c-u\ + i (2) If the user cost is devoted to fresh investments, income rises and we get: Y = (C- U) + [I+U) (3) Equation (3) is thus seen to be identical to equation (1) except that consumption is lower and investment is higher. Equation (2), however, depicts the correct level of income if the user cost is not reinvested. But if C remains unchanged, then the true level of investment that has been attained is only 1-U since U represents a disinvestment. In this latter case we have: Y - U = C + (/- U) (4)
Policy and the Problem of Scale
While the approach of sinking part of the proceeds into new investments seems perfectly valid for individuals, businesses, and even small countries, which also have the option of acquiring foreign investments if profitable domestic opportunities are not available, is it workable if it is done on a large scale so that significant portions of global natural capital might be liquidated to be substituted for by manmade capital formation?
Once the problem is posed in this way, the realization of the objective of creating a permanent income stream from wasting assets becomes questionable. Individuals, corporations, and even nations can run out of a natural resource—even if their livelihoods depend materially on it—in the knowledge that future income may be generated through carefully selected new investments. When considering better accounting for depletable resources, my focus was on the income of their owners. It did not matter what form the new investments would take, provided they guaranteed for the owners a constant stream of future income. The form of the new investments would be guided by the market, and if the market indicated that the new investments should be in the same line of business, so be it. However, if the problem is considered not just as one of better accounting for the resource owners, but in a forward context as a guide to economic policy on a global scale, we have to face the issues raised by Brundtland and the various constraints and propositions we find there for future environmental directions. We also encounter the problems of scale and of ultimate substi-tutability between natural resources and manmade capital to which Herman Daly has been drawing our attention.
If we perceive the problem globally, then it is clearly necessary to replace, for example, dwindling natural energy sources, not just with other sources of income, but with other sources of energy that are renewable, and the issue of "twinning" becomes relevant. If the market fails to signal rising energy prices to justify investing in renewable energy sources, then society may wish to give the market a helping hand through appropriate policy. Viewed globally, society should have a broad interest in the creation and application of new technologies that would substitute renewable sources for diminishing, nonrenewable ones.
But what should be done about the search for an equilibrium between the state of the environment and global economic activity? The world economic organization has been functioning cm the basis of economic agents seeking perpetual economic growth, a pursuit that has traditionally been seen not only as desirable for raising material welfare all around but also as essential for energizing the development of the less developed countries and thus assisting in the alleviation of poverty. If technology could be organized so that it gave us substitutes for natural resources through the instrument of manmade capital formation, we would be able to continue "business as usual," hoping that the market would reflect scarcities into higher prices and thus guide this process of substitution. This certainly appears to be one of Brundtland's fundamental assumptions. However, we have reached a stage where the state of the environment has become so stressed, and technology and social organization have clearly lagged, at least so far, that some drastic alternative solution deserves to be explored.
Brundtland offered one solution, which leans toward maintaining the current emphasis on growth while using the fruits of growth to lessen the material throughput in economic activity, to repair the environment, and also to redistribute income, both intranationally and from the richer to the poorer nations, with the objective of alleviating poverty. I join with the other contributors to this volume in contending that this strategy is questionable—partly because much of the damage to the environment caused by indiscriminate growth is irreversible; partly because the process of substitution of manmade capital for natural resources is slow and erratic; and also in view of the enormous increase projected for global economic activity as compared with the advanced state of environmental stress already reached. If we are serious about saving our planet, we must seek a steady state for the economies of the rich, while the poor grow and develop so that poverty is eradicated and income disparity, which is the source of so much environmental damage, is reduced. Meanwhile technology development and dissemination should be accelerated and population growth urgently halted.
If the Brundtland path is rejected as impractical, can the proposal to arrest growth in much of the world economy be viewed as anything short of Utopian? It is difficult specifically to perceive the sociology and political economy of maintaining a steady level of income in the richer countries. Such countries rely primarily on free market forces to guide the allocation of economic resources. In these countries, the essential profit motive is geared unavoidably to business expansion in search of opportunity. The impact of the richer countries' economic expansion on developing countries has also often been seen as benign in an "empty world" context of nonbinding environmental constraints. In fact, every time growth slows down in the richer countries, the poorer ones appear to suffer from depressed incomes and adverse terms of trade. And yet the richer countries use the bulk of the world resources to support a minority of the world population. If the rich are to grow richer merely to provide markets for the poor, not only are there more economical ways to achieve the same objective, but such a course would accelerate international income inequality.
Clearly something drastic has to take place in social and industrial organization and in the modalities of international relations if a steady state of economic activity, involving a constant level of throughput, is to prevail in the developed countries. Drafting a blueprint for this vision of the future is essential. Its economic content will have to address the problem of obtaining growth and/or development in the poorer countries simultaneously as the economies of the richer countries are kept on an even keel. In addition, the richer countries would be asked to transfer to the less developed countries the resources necessary to redress the negative effect of the richer countries' arrested growth and to alleviate poverty. Furthermore, it is necessary to plan for the kind of economic policy that would have to apply in the richer countries to produce the target of a steady state: as some activities will have to expand, others must contract. What criteria would be used to modulate aggregate activity in a free market economy that also has to be managed in pursuit of many other policy objectives? The issues this scenario raises will have to be faced by the advocates of such a strategy.15 The Brundtland Report avoided all these complex issues and opted instead for a nonrevolutionary, rather optimistic, but seemingly untenable course.
Conclusion
Finally, a word about the importance of proper income accounting, since it is income measurements that will indicate what kind of growth or expansion of economic activity is being experienced and projected. Today's income changes, which probably lie behind Brundt-land's projections of growth, relate to the gross domestic product (GDP) as conventionally measured and as valued at factor cost.16 But if we shift the focus from the gross product to an environmentally more sustainable net product (from which the user cost of depletable resources has been eliminated), put a value on natural disasters and deduct this from income, and develop the habit of valuing activities at their full environmental cost when prices reflect true scarcities, we are bound to get a very different reading of income and its growth. In which case it might well turn out that the five-to-ten-times expansion in economic activity, as envisaged by Brundtland and stressed by McNeill, will be less.17 A hint of this is to be found in the contribution by Tinbergen and Hueting in this volume (see Chapter 4), but clearly much work is needed to clarify this issue.
Notes
1. World Commission on Environment and Development, "Our Common Future" (The Brundtland Report) (Oxford: Oxford University Press, 1987), 43.
2. Definitions of sustainability are surveyed by J. Pezzey in Appendix 1, "Definitions of Sustainability in the Literature," of Economic Analysis of Sustainable Growth and Sustainable Development (Environment Department Working Paper 15) (Washington, D.C.: World Bank, 1989).
3. As the reader will note, presently the search for a precise meaning of "sustainability" is akin to defining "income" in exact terms. No unanimity is possible, since both concepts depend upon one's vision of the future. For practical purposes, however, and as a guide for prudent behavior, we must be content with some useful degree of approximation.
4. Whether it was the Brundtland Report itself, or the political forces that have been gathering momentum independently in various parts of the richer nations, it is remarkable how the impact of the Green Movement has been reflected in the declarations of recent economic summits of the Group of Seven leading industrial nations and through the latter's influence has given vent to a number of environmental initiatives. The July 1989 Economic Declaration of the G-7 Economic Summit (section 37) contained the statements: "In order to achieve Sustainable development, we shall ensure the compatibility of economic growth and development with the protection of the environment" and "We encourage the World Bank and regional development banks to integrate environmental considerations into their activities."
5. The July 1990 Economic Declaration of the G-7 Economic Summit referred to global environmental stress (ozone depletion, climate change, marine pollution, and loss of biological diversity) and stated that "one of our most important responsibilities is to pass on to future generations an environment whose health, beauty and economic potential are not threatened."
6. It is interesting thai in their initial stages the UNEP-World Bank work shops, after establishing national physical indicators of environmental stress, were seeking to combine these eventually into one national index that would reflect the state of the environment, but participants very quickly realized that a system of "weighting" (or valuation) was necessary to produce such a single index. This moved the concern of the workshops quite early in the direction of reforming national income measurement. Cf. S. El Serafy's "Rapporteur's Report of the October 1985 Paris Meeting" (Washington, D.C.: World Bank, 1986), mimeographed.
7. Cf. Alan Gelb and Associates, Oil Windfalls: Blessing or Curse! (World Bank Research Publication) (Oxford: Oxford University Press, 1988).
8. I am abstracting here from a number of activities that have traditionally been excluded from national income reckoning, such as household services by family members. That the environment can be viewed as capital, contributing to the productive process, is a notion that is entirely in harmony with neoclassical economic thinking. See Salah El Serafy, "The Environment as Capital," Ecological Economics: The Science and Management of Sustainability, ed. R. Costanza (New York: Columbia University Press, 1991).
9. See El Serafy, "The Environment as Capital," in Ecological Economics, op. cit.
10. J. R. Hicks, Value and Capital, 2d ed. (Oxford: Clarendon Press, 1946), 187.
11. Hicks's all too brief coverage of this topic in Value and Capital shows that he regarded such a user cost as an allowance for depreciation. In a personal communication in 1987, however, he indicated approval of my line of thinking and that I had "made good use of the income chapter in Value and Capital."
12. A qualified acceptance of this approach is to be found in M. A. Adelman, Harindar De Silva, and Michael F. Koehn, User Cost in Oil Production (Cambridge, Mass.: MIT Center for Energy Policy Research, October 1990). This work uses the calculations of El Serafy to adjust national income for a number of countries in support of arguments made in the text, but states that "El Serafy ... err(s| in supposing that production can proceed at a constant rate, then abruptly cease. The decline rate stands at the center of every reservoir engineering calculation. Moreover the rate of extraction is limited by sharply rising marginal costs .. . However, this correction would not basically change the problem." It should be mentioned, however, that Adelman belongs to the camp that sees no scarcity developing in the supply of minerals, which he views correctly as inventories, but believes that "only a fraction of the minerals in the earth's crust, or in any given field, will ever be used (op. cit., p. 11. The approach I have been advocating is one that relies on a standard accountant's rule of thumb that estimates inventory use out of a given stock in an attempt to approximate reality. I stated in my 1981 Journal of Energy and Development article that factors such as the ones mentioned by Adelman et al. could be accommodated under the approach I proposed. The so-called reservoir engineering rule of always keeping a constant ratio between reserves and extraction is of dubious reliability and not essential for the calculations in any case. See Salah El Serafy, "Absorptive Capacity, the Demand for Revenue and the Supply of Petroleum," Journal of Energy and Development 7 (Autumn 1981): 73-88.
13. See David F. Bradford, "Comment on Scott and Eisner," Journal of Economic Literature 28 (September 1990): 1184. This was a comment on Maurice Scott's "Extended Accounts for National Income and Product: A Comment" and Robert Eisner's "Reply" to Scott in the same issue.
14. Ecologists tend to define substitutes more narrowly than economists, who appear to favor a broad definition that allows the market freedom to define what a substitute is. David Pearce, Anil Markandya, and Edward Barbier, in their Blueprint for a Green Economy (London: Earthscan Publications, 1989), advocated "pairing" or "twinning," but within a program of many projects rather than for each project at a time.
15. A vision of a possible course is offered in Herman E. Daly and John B. Cobb, For the Common Good (Boston: Beacon Press, 1989). Many aspects of such a course, however, need to be much more carefully examined, as the authors urge.
16. The convention of valuing GDP at factor costs and not at market prices derives from the presumption that taxes and subsidies represent deviations from genuine values produced by the market and that they should provide weights for the various activities that make up the domestic product. But if a new set of environmentally inspired taxes and subsidies is viewed as necessary to correct the market's failure to put proper values on the services of natural resources, then we should regard the new, environmentally adjusted "market prices" as better weights than factor costs for the purpose of estimating income in the present context.
17. James McNeill, "Sustainable Development, Economics and the Growth Imperative" (paper presented to the Workshop on the Economics of Sustainable Development) (Washington, D.C.: U.S. Environmental Protection Agency, January 1990).