6
Natural Resource and Environmental Accounting in U.S. Agriculture1

Pierre Crosson

Resources for the Future

Washington, D.C.

ISSUES IN ASSESSING AGRICULTURAL PERFORMANCE

Introduction

Between 1950 and the early 1990s U.S. agriculture had impressive success in reducing the economic costs of production—measured by declining real prices of farm commodities— despite an almost two-fold increase in crop and animal production. The combination of lower prices and expanded output conveyed substantial economic benefits to consumers of American farm output, both at home and abroad. The high productivity of the nation's agricultural land, greatly increased use of water for irrigation, and major advances in agricultural science and technology and in the managerial skills of farm people were the key elements in U.S. agriculture's strong economic performance.

But over the last decade or so, questions increasingly have been asked about the adequacy of this assessment. The questioning has run along two lines. One asks whether farm commodity prices fully reflect the on-farm costs of long-term depletion of the land and water resources used in agricultural production. Erosion-induced losses of soil productivity are of special concern; but the ''mining" of groundwater for irrigation—withdrawing more water than is replenished by infiltration from the land surface—also excites unease.

The other line of questioning reflects recognition that U.S. agriculture generates a variety of costs and benefits which are not reflected in the prices of farm output because they occur off the farm. The scenic amenity values of a well-tended agricultural landscape provide an example of an off-farm agricultural benefit. The values are enjoyed by farmers and nonfarmers alike, but they are not reflected in farm commodity prices. On the other side of this coin, losses of scenic amenity values when farmers sell their land for conversion to urban sprawl are an example of an off-farm cost not reflected in commodity prices.

1  

The research underlying this paper was funded by a grant from the U.S. Department of Agriculture.



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 111
Assigning Economic Value to Natural Resources 6 Natural Resource and Environmental Accounting in U.S. Agriculture1 Pierre Crosson Resources for the Future Washington, D.C. ISSUES IN ASSESSING AGRICULTURAL PERFORMANCE Introduction Between 1950 and the early 1990s U.S. agriculture had impressive success in reducing the economic costs of production—measured by declining real prices of farm commodities— despite an almost two-fold increase in crop and animal production. The combination of lower prices and expanded output conveyed substantial economic benefits to consumers of American farm output, both at home and abroad. The high productivity of the nation's agricultural land, greatly increased use of water for irrigation, and major advances in agricultural science and technology and in the managerial skills of farm people were the key elements in U.S. agriculture's strong economic performance. But over the last decade or so, questions increasingly have been asked about the adequacy of this assessment. The questioning has run along two lines. One asks whether farm commodity prices fully reflect the on-farm costs of long-term depletion of the land and water resources used in agricultural production. Erosion-induced losses of soil productivity are of special concern; but the ''mining" of groundwater for irrigation—withdrawing more water than is replenished by infiltration from the land surface—also excites unease. The other line of questioning reflects recognition that U.S. agriculture generates a variety of costs and benefits which are not reflected in the prices of farm output because they occur off the farm. The scenic amenity values of a well-tended agricultural landscape provide an example of an off-farm agricultural benefit. The values are enjoyed by farmers and nonfarmers alike, but they are not reflected in farm commodity prices. On the other side of this coin, losses of scenic amenity values when farmers sell their land for conversion to urban sprawl are an example of an off-farm cost not reflected in commodity prices. 1   The research underlying this paper was funded by a grant from the U.S. Department of Agriculture.

OCR for page 111
Assigning Economic Value to Natural Resources The wildlife habitat values of farm woodlands and wetlands are another example of typically off-farm benefits. That is, habitat values are enjoyed by hunters, birdwatchers, and even by people who get pleasure from the mere knowledge that wildlife are provided habitat for their continued existence. But the values are not reflected in farm commodity prices, and only partially captured in agricultural land prices.2 Symmetrically, the costs of losing habitat values when the land is cleared and drained for crop production are not reflected in commodity prices. The off-farm benefits and costs of agriculture are not always symmetrically balanced, as in these examples. The benefits of resources which increase the productivity of farm operations, such as pesticides and fertilizers, are reflected in lower prices of food and fiber. But use of these resources may also impose a variety of off-farm costs, for example, damages to human health from ingestion of contaminated groundwater, to wildlife (from pesticides) and to aquatic ecosystems (from pesticides and fertilizers). Similarly, sediment eroded from farmland and deposited in floodplains and river deltas can increase soil productivity, and hence may generate benefits reflected in lower farm commodity prices. But sediment may also impose off-farm costs in the form of losses of recreational values because of murky water in streams, lakes and reservoirs, damages to fish spawning areas covered by silt, and higher costs of cleaning the water for residential and industrial uses. Clearly, a full assessment of the performance of U.S. agriculture must take account not only of on-farm economic costs and benefits but the off-farm costs and benefits as well. In such an assessment, Crosson (1992) concluded that the off-farm costs of U.S. agriculture rose relative to the off-farm benefits from the end of World War II to the early 1990s. The implication is that off-farm costs rose also relative to on-farm economic costs (since the latter, reflected in farm commodity prices, declined). But because of the lack of reliable data for off-farm costs and benefits, Crosson (1992) made no judgment of whether these costs rose enough to offset the decline in on-farm economic costs. Sustainability Issues Consideration of the on-farm and off-farm costs of U.S. agricultural production raises the issue of the sustainability of the production system. Sustainable agriculture is the subject of intense discussion and increasing research in the agricultural and environmental communities, not only in the U.S. but also in other countries and in international agencies such as the World Bank. Definitions of sustainable agriculture vary widely, but the key element in all of them, implicit if not explicit, is concern about the long-term social costs of agricultural production. The most hard-bitten environmentalist would join with the most hard-nosed agriculturalist in agreement that the U.S. agricultural system is sustainable if each were satisfied that the system could indefinitely accommodate demands for food and fiber at satisfactory social costs. 2   Many farmers charge fees to hunters and bird watchers for access to their land, and the resulting net income must be reflected in land values, at least to some extent. But not all farmers do this, and they have no way of capturing the existence value of habitat on their land.

OCR for page 111
Assigning Economic Value to Natural Resources Where the participants in the sustainability discussion part company, therefore, is not about the key importance of costs but about criteria for judging whether costs are, or are likely to be, satisfactory into the indefinite future. This is not the place to discuss differences with respect to these criteria. It seems clear, however, that many of the differences reflect differences in deeply held values, e.g., about more equal income distribution or the intrinsic worth of wildlife relative to low-priced food. But many of the differences must also reflect lack of information about the relative importance of the various costs, now and in the future. That is, it likely would be easier to narrow present differences about the sustainability of the U.S. agricultural system if we had more quantitative information about the magnitude of the various on-farm and off-farm costs, now and in the future. For example, it likely would be easier to get agreement between environmentalists and agriculturalists about the importance of protecting on-farm wildlife habitat relative to increasing crop production if we had reliable estimates that on a national scale the annual social costs of habitat loss because of clearing and drainage are × percent of crop production costs, plus or minus 20 percent. The objective of this paper is to indicate how a system of accounts for agriculture could be developed to display quantitative estimates of the effects of agricultural production on natural resource capital used in agriculture and on environmental costs and benefits. Over time, such a system of accounts should help to reduce differences between agriculturalists and environmentalists concerning the sustainability of U.S. agriculture. AN ACCOUNTING FRAMEWORK FOR AGRICULTURE Introduction It was asserted above that environmentalists and agriculturalists could agree that the U.S. agricultural system is sustainable if each, in their respective judgments, believed that the system is capable of satisfying long-term demands for food and fiber (the benefits of the system) at acceptable on-farm and off-farm costs. The advantage of an accounting framework for agriculture is that it provides a mechanism for systematic tracking of the production system's benefits and costs, thus promoting better-informed judgments about the sustainability of the system. This section discusses issues in constructing such an accounting framework. Net Income as an Indicator of Sustainability The net social income provided by the agricultural production system is the difference between the value of gross production and the on-farm and off-farm production costs. Net social income, therefore, is a key indicator of the sustainability of the production system. If over time production costs rise relative to the value of total production, net social income will fall, arousing doubts about sustainability. If the proportion of costs to production remains the same or declines, net income will be maintained or increased, indicating that the system is on a sustainable path.

OCR for page 111
Assigning Economic Value to Natural Resources Maintaining a constant or rising stream of net social income—achieving sustainability— depends on the capacity of the production system to transform inputs into outputs desired by people at social costs which do not increase in proportion to production. The capacity of the system, in turn, depends on the quantity and productivity of the resources employed in production—the land, the water, the people, the fertilizers, farm machinery and so on.3 Should the quantity of these resources decline and their productivity (output per unit of resources) remain the same, then the capacity of the system to maintain the stream of net social income would be reduced. Similarly, a decline in resource productivity would reduce capacity unless there were an offsetting increase in the quantity of resources. The distinction between the quantity and productivity of resources is recognized, albeit implicitly, in the literature on accounting, whether for a firm (e.g., a farm), a sector (e.g., agriculture) or for an entire regional or national economy. All modern accounting systems are based on the principle of double-entry bookkeeping. The principle is that for any production unit the quantity of resources employed (the inputs), measured in dollars, must be precisely equal to the quantity of outputs produced, also measured in dollars. This is to say that the bookkeeping system "accounts for" all the inputs by insisting that the sum of them must equal the sum of the outputs. Changes over time in the productivity of resources typically are measured by changes in the ratio of total output to total resource inputs.4 This measure of productivity change cannot be derived from double-entry accounting systems, however, because, as just noted, in such systems the quantity of resources employed, measured in dollars, is always equal to outputs produced, i.e., the ratio of outputs to inputs is always and necessarily equal to one. Changes over time in the resource productivity of a production unit nonetheless can be inferred from examination of the unit's accounts with respect to net income. In the accounts net income is the residual difference between the value of resources employed and the value of outputs produced. Thus resources and outputs will always be equal, but net income may rise or fall, depending on an increase or decline in resource productivity. 3    We deal here with resources directly employed on the farm and with off-farm resources directly affected by farm operations, e.g., streams used as a dump for sediment or agricultural chemicals from farmers' fields. However, agricultural production draws also on a much larger and more heterogenous set of resources such as the transport and communications infrastructure; the social, legal, and political institutions which shape attitudes toward farming and define and enforce agricultural property rights; and stocks of accumulated social knowledge about how to make the economic system, of which agriculture is an integral part, work effectively. We make no effort to capture the role of this very broad but less agriculture-specific set of resources in shaping agricultural performance. For a discussion of these resources and the problems of reflecting them properly in a set of national income accounts, see Juster et al., 1979. 4   The U.S. Department of Agriculture, for example, measures changes in the total productivity of American agriculture as changes in the ratio of an index of total farm output to an index of total farm input (USDA, 1991).

OCR for page 111
Assigning Economic Value to Natural Resources To understand this, consider an example. Suppose that over time soil erosion reduces the productivity of the land by removing nutrients, but that production does not fall because farmers exactly compensate by increasing applications of fertilizer. Since the value of total output by assumption remains the same, the value of total input must also be unchanged. However, the composition of inputs has changed. Inputs of fertilizers have increased, reducing, by exactly the same amount, the proceeds available as net income. The decline in net farm income—equal to the increased outlays for fertilizer—measures the cost of the erosion-induced decline in soil productivity. In this example farm productivity, measured by net income, declines because erosion increases on-farm production costs in the form of higher outlays for fertilizer. However, increased off-farm costs also reduce the productivity of the system, measured by social (farm plus off-farm) net income. Suppose that the erosion posited above does not affect soil productivity, but that the soil moved is carried to streams and lakes where it reduces water quality, imposing losses in recreational values of the water. Total marketed farm output is unchanged, so inputs also necessarily are unchanged. Net farm income, therefore, also is unchanged; but net social income is reduced by the cost of sediment damages to recreational uses of surface water. With total on-farm costs unchanged and off-farm costs higher, the net social income from farm operations necessarily is reduced, measuring a decline in the productivity of the system. Conceptually, the effects of off-farm costs and on-farm costs on the productivity of the agricultural system are the same; and a properly constructed accounting framework will incorporate both kinds of costs. From the standpoint of farmers, however, the two kinds of costs are quite different. Farmers pay on-farm costs. They do not pay off-farm costs, absent policies or institutions which require or induce them to do so. This difference in the two kinds of costs is critically important in understanding and assessing the performance of the agricultural system. It also is critically important for policies to ensure the sustainability of the system. In this paper, however, we deal only with an accounting framework to incorporate the two kinds of costs. Limitations of Existing Accounting Systems Table 6-1 shows data for gross and net farm income in the U.S. for selected 3-year periods since the late 1940s-early 1950s. Gross income is the value farmers received from selling the crops, animals, and other services they produced, the value of their consumption of farm produce, the change in value of farm-held inventories of crops and animals, and payments from the federal government under various price and income support programs. Production expenses are the amounts farmers paid to purchase fertilizers, pesticides, seed, and a whole range of marketed inputs used in production; to hire labor; as interest on debt and on farmer-owned equipment and buildings; plus an amount representing depreciation of the farmer's human-made capital: principally equipment such as tractors, cultivators, combines and so on, and barns, fences, silos, and other built structures.

OCR for page 111
Assigning Economic Value to Natural Resources TABLE 6-1 Average Annual Farm Income, Various Periods (Billions nominal and 1987 dollars) Time Period Grossa Net   Total Gov't payments Gov't percent of total Production expensesb Totalc Ex-gov't payments Gov't Percent of Total Net in Total 1987 dollars Ex-Gov't. 1949/1951 34.1 .252 0.7 19.9 14.2 13.9 1.8 64.9 63.7 1959/1961 39.0 .968 2.5 27.7 11.3 10.3 8.6 43.5 39.8 1969/1971 59.1 3.549 6.0 44.6 14.5 11.0 24.4 41.5 31.4 1979/1981 155.4 1.531 1.0 131.9 23.5 22.0 7.0 32.8 30.5 1988/1991 186.6 11.600 6.2 139.5 47.1 35.5 24.6 43.4 32.7 a Sales of crops and livestock, value of produce consumed on the farm, government payments, and changes in the value of farm-held inventories. b All purchases of goods and services used in production, wages of hired farm labor, interest on real estate and non-real estate debt, non-income taxes, and depreciation and other consumption of farm capital. Depreciation is the money needed to replace the capital used up in production annually. Depreciation of natural resource capital is not included. c The difference between gross income and production expenses. SOURCES: Gross and net income, government payments, and production expenses: 1949-57 -USDA (1960); 1959-61, 1969-71 -USDA (1974); 1979-81, 1988-90 -USDA (1991b). Net income in 1987 dollars: Office of the President (1992). Nominal net income was deflated by the price deflator for gross domestic product, 1987 = 100.

OCR for page 111
Assigning Economic Value to Natural Resources The depreciation charge against gross farm income does not include consumption of natural resource capital incurred by the production process, e.g., losses of soil productivity because of soil erosion. Nor does gross farm income include the off-farm benefits of farming or production expenses of the off-farm costs of environmental damage. Net income is the difference between gross income and production expenses. We argued above that movements of net farm income over time could be taken as an indicator of the sustainability of the agricultural system. However, this interpretation is ambiguous in Table 6-1 because of the role of government payments in net farm income. These payments are made under various federal government programs designed to protect farm income against the negative effects of market forces. As a percent of net farm income government payments rose from 1.8 percent in 1949/51 to 8.6 percent in 1959/61, to 24.4 percent in 1969/71, fell to 7.0 percent in 1979/81, reflecting the relative farm prosperity of the 1970s, then rose sharply to 24.6 percent in 1988/90 as U.S. export markets and crop prices fell in the 1980s. One interpretation of government payments is that they reflect efforts to compensate for weaknesses in the agricultural system, the higher the payments, the greater the weaknesses. Because of the weaknesses, farmers are insufficiently competitive in agricultural markets to earn satisfactory incomes. Under this interpretation, the rise in government payments as a percent of net farm income over the last 40 years would raise questions about the sustainability of the system. Another interpretation of government payments, however, is that they reflect a politically determined decision that the benefits bestowed by agriculture on American society are not fully captured by market processes. In this view the payments reward farmers for this otherwise uncompensated contribution to the general welfare. We do not discuss the merits of these two interpretations, but we believe that net income exclusive of government payments is the better indicator of performance. This is based on the assumption that a sustainable system for agriculture (or for any economic sector) is one which over the long-term can "earn its own way," that is that it can earn a socially satisfactory income for the people engaged in the system without the help of government payments. As noted, other interpretations of government payments are possible; but we believe the one adopted here is closest to what most people have in mind when they think about sustainable agriculture. By this criterion, net income exclusive of government payments is the best indicator of agricultural performance. Specifically, this indicator adjusted for inflation (the last column in Table 6-1) is most relevant. In the 20 years from 1949/51 to 1969/71 this measure of agricultural performance fell from $63.7 billion (in 1987 dollars) to $31.4 billion, then remained about the same over the next 19 years. Over three-fourths of the $31 billion decline from 1949/51 to 1988/90 had occurred by 1959/61. The decline in real net farm income after 1949/51 was accompanied by—indeed reflected—a massive shift of people out of farming. Between 1950 and 1989 the total farm population decreased about 75 percent. Although the amount of land in farms declined some 15 percent, the total number of farms fell 60 percent.5 Consequently, real net farm income per 5   The 60 percent decline in farm numbers is overstated because in 1975 the definition of a farm changed to eliminate approximately 10 percent of the rural operations previously classified as farms.

OCR for page 111
Assigning Economic Value to Natural Resources farm, exclusive of government payments, was roughly 25 percent higher in 1988/90 than in 1949/51, and about 50 percent higher than in 1959/61.6 There is a widely held view, which most Americans probably share without bothering themselves about its philosophical underpinnings, that if farmers are better off—and not at the expense of nonfarmers—then society generally is better off. The view is based on the philosophy of utilitarianism, first developed by British philosophers in the 18th century and still alive and well at the end of the 20th century. Crudely, but not inaccurately put, utilitarians argue that the welfare of society is the sum of the welfare of the individuals in the society. It follows that if the welfare of some members of the society—e.g., farmers—increases without detriment to the welfare of other members, then the welfare of society as a whole is increased. If one holds this view, and is willing to accept net income as measured in Table 6-1 as a satisfactory, if very rough, measure of welfare, then one could conclude that over the last 40 years fanning in America has enhanced the welfare not only of farmers but of society generally.7 But whether or not one is a utilitarian one may, and many do, object to the estimates of net income in Table 6-1 as measures of ''true" net farm income. There are two reasons: (1) the measures of production costs may underestimate on-farm costs because as noted, they do not include the draw-down—the consumption—of natural resource capital, principally land and water, used in production; (2) the estimates of gross farm income and production costs do not include the off-farm benefits and costs of farming. These are serious issues. If consumption of natural resource capital is significant, then agriculture has not done as well in generating net income for farmers as Table 6-1 indicates. Indeed, if in fact there was a significant, but unmeasured, loss of natural resource capital, then American farming may be on an unsustainable course. If off-farm costs are significant, then net farmer income, as measured in Table 6-1, has been sustained in part by the imposition of costs on nonfarmers. In this case, net income in Table 6-1 overstates the contribution of farming to the general social welfare. If off-farm benefits are significant, then to that extent the Table 6-1 estimates of net income understate farming's welfare contribution. The rest of this part is devoted to discussion of how to deal in an income accounting sense with these issues of natural resource depletion and off-farm costs and benefits. The Consumption of Natural Resource Capital The value of any capital asset is the sum of the discounted stream of net income yielded by the asset over time. The income stream may be measured in any increments of time, but for convenience here we assume annual increments. Given the costs of producing the output of the 6   The numbers in this paragraph on farm population and number of farms are from various issues of USDA's Agricultural Statistics. 7   The increase in farm output and decline in farm commodity prices, noted in the Introduction, provide a more direct indicator of the contribution of farming to the general welfare over the last 40 years.

OCR for page 111
Assigning Economic Value to Natural Resources asset, the size of the net income stream depends on the annual quantities of goods or services produced by the asset, times the prices of these outputs over however many years the asset remains in service. Consequently, given the life of an asset, its value can change because the quantity of its output changes, or the prices of the output changes, or the discount rate changes, or some combination of these. Capital is consumed, or "used up" whenever the discounted stream of its net income declines. Apart from increases in the discount rate, this can happen because the quantity of output of the capital declines, reflecting, perhaps wear and tear, as on a machine which experiences more downtime for repairs. The decline in capital value may also occur because the prices of the capital's output decline, or because newer, more productive machines are developed which economically displace the older machine. If in the process of production capital is consumed—the net income stream of the capital declines—then sustainability requires that investments be made in new capital that generates enough value of output to replace the loss of net income. This required amount of investment is called a capital consumption allowance because it is the amount needed to replace the capital consumed. Perhaps paradoxically, the value of an asset can increase even if the quantitative output of the asset declines. This will happen if the price of the output increases proportionally more than the decline in the quantity of output, (holding the discount rate constant). For example, if the demand for food were rising faster than the supply, forcing food prices up relative to the general price level, then the real net income stream to the land, hence real land values, would rise even if the quantity of food production were declining because of severe soil erosion. The physical productivity of the land would fall but its economic productivity would rise. In such cases the capital value of assets is not "consumed" or "used up." On the contrary it is increasing and the increase is counted as income to asset owners. This consequence is entirely consistent with the concept of capital and with understanding of how real capital values change over time for any given sector of the economy. But the situation depicted cannot apply to capital values in the economy as a whole. In the example cited, in which we are dealing with inflation adjusted prices, if the real price of food goes up, then the real prices of some other goods and services must come down. The real value of the capital employed in producing those goods and services would also decline. This is another way of expressing the common sense notion that the real income of the economy as a whole cannot rise, or even stay even, if the physical productivity of all the capital assets in the economy are declining. The discussion indicates that in principle there would not necessarily be anything paradoxical in rising real prices of agricultural land in the U.S. combined with declining physical productivity of the land. In such a case, the notion of a capital consumption charge against the land would seem to make little sense since the economic capacity of the land is not consumed but in fact is rising. Yet if the question concerns the long-term sustainability of the U.S. agricultural system, this conclusion excites unease. Should, in fact, U.S. agriculture combine rising economic value of the land with diminishing physical productivity it almost surely would be because of rising foreign demand for U.S. agricultural output. Foreign demand currently takes 25-30 percent of the value of U.S. output, and most of the changes in demand over the last couple of decades

OCR for page 111
Assigning Economic Value to Natural Resources have been in foreign demand. This is expected to continue in the future (Crosson, 1992). In the situation depicted, therefore, U.S. consumers would be confronted by an increase in real prices of food, not of their making, simultaneously with a decline in the physical productivity of the nation's agricultural land. Would such a situation be perceived as consistent with U.S. concerns about the sustainability of agriculture? I believe that it would not. Consequently, I regard the situation as in principle a possibility (although no such situation has occurred in U.S. agriculture for at least the last 60 years), but one that, should it occur, would be considered inconsistent with agricultural sustainability and would prompt public action to control erosion-induced losses of soil productivity. I return, then, to the assertion that sustainability requires that losses of the physical productivity of natural resources should be reflected in a capital consumption charge against gross output; and that the charge should equal the amount of investment needed to restore the net income stream of the capital to its previous level. Notice, however, that the money spent to replace capital consumed does not have to be spent on the same assets that are consumed. It is the total capital of the enterprise (or of the sector, or of the nation) which must be maintained, not the capital represented by any particular asset. Thus a farmer may decide to use the capital consumption allowances for a moldboard plow to purchase the equipment needed to shift to some form of conservation tillage system. In a more extreme case, the farmer may cut back sharply over time On his investment in crop production, using the capital consumption allowances for crop production capital to expand his investment in animal production. The important point is that the capital consumption allowances charged against gross farm income must be sufficient to maintain the total farm capital, and hence the stream of net farm income over time. Given this condition, the precise form the capital takes is not important. Clearly, the consumption of capital is a cost of production. Since net income is the difference between gross income and production costs, consumption of capital must be taken into account if net income is to be properly measured. All modern accounting systems recognize this. However, the systems of national income accounts used by the United States and other nations do not include an allowance for consumption of natural resource capital. There are a variety of arguments for this exclusion. One of the most important is that natural resources are free gifts of nature so consumption of them is costless. This argument misses the point that although natural resource capital may be "free," it generates social value which may be diminished through use of the resource. Another argument is that estimates of natural resource capital are subject to large short-term fluctuations, reflecting sudden shifts in market conditions. For example, the capital value of global oil reserves rose steeply in 1973 and 1974 when the Organization of Petroleum Exporting Countries raised its prices, then declined for several years, then rose steeply again in 1979, only to decline again in the 1980s. Changes in capital value are part of national (and global) income, so inclusion of the value of oil reserves in national income accounts would result in large, and so far at least, temporary swings in national income estimates. This would "distort" long-run trends in national income, runs the argument for excluding natural resources from national income accounting. There is a substantial literature on the conceptual and measurement problems of including natural resource capital in national income accounts. (A sampling of this literature is cited below in footnote 8). Suffice it to say here that among the experts in national income account

OCR for page 111
Assigning Economic Value to Natural Resources ing, there now is agreement in principle that consumption of natural resource capital should be counted as a cost of production, just as is consumption of human-made capital. By this principle, the estimates of farm production costs in Table 6-1, which are part of the U.S. national income accounts, are underestimated—and net farm income is overestimated by the same amount—to the extent that, over the period shown, the capital value of land and water resources was in fact diminished by use of them. Off-Farm Costs and Benefits The systems of national accounts are designed to measure the production, cost, income and expenditure consequences of transactions in markets. This quantifies the consequences in dollars and permits size comparisons among sectors of the economy and measurement of trends over time, both by sector and for the economy as a whole. But many transactions involving the use of resources are not registered in markets. Hence they are unpriced and are not reflected in the national income accounts. Unpriced uses of ''environmental" resources are a major but not the only category of these excluded transactions. Unpriced use of the land for wildlife habitat, of surface water as a dump for runoff from farmers' fields, and of the atmosphere as a sink for methane emitted by farm animals are examples from agriculture of unpriced transactions in environmental resources. These unpriced transactions make up a major pan of what, in this paper, I call off-farm benefits and costs of agricultural production. As noted above, there is an extensive and growing literature on the failure of systems of national income accounts to incorporate natural resource depletion and unpriced transactions in environmental resources; on the consequences of this failure for assessment of performance over time by individual sectors and for the economy as a whole; and on measures to remedy the failure. It is not necessary for our purposes here to review this literature.8 Instead we draw on a portion of the literature to illustrate how natural resource depletion and off-farm costs and benefits could be included in an agricultural sector system of accounts which would be appropriate for the farm level as well as for agriculture as a whole. The system also would be consistent with national accounts as they might be revised to include natural resource depletion and unpriced transactions in environmental resources. A Schematic Accounting Framework Table 6-2 displays an accounting framework consistent with currently used national income accounting practices, but which incorporates both consumption of natural resource capital 8   For a representative sampling of this literature see the following: Ward (1982); Levine (1991); Peskin with E. Lutz (1990); Norgaard (1989); Repetto et al. (1989); Repetto (1992); Faeth et al. (1991). Also the following chapters in Ahmad, E1 Serafy and Lutz (eds., 1989): Peskin (1989a and 1989b); El Serafy and Lutz (1989); Daly (1989); El Serafy (1989); Harrison (1989); Hueting (1989); Blades (1989); Bartelmus (1989); Lutz and El Serafy (1989).

OCR for page 111
Assigning Economic Value to Natural Resources and environmental costs and benefits. The framework—taken with a slight modification from Peskin (1989b in Ahmad et al., 1989)—is designed to represent the inputs and outputs of a private sector firm as they would be displayed for use in the national accounts. In this discussion we assume that the firm is a farm. In this, as in any accounting system based on double-entry bookkeeping, the inputs must equal the outputs. Specifically, everything the farm sector produces (the output side) must be accounted for somewhere on the input side of the ledger. Before adjustment for environmental damages, the output side of the farm sector is readily understood. Everything farmers produce is sold to industry (e.g., wheat for milling to flour), to households (e.g., fresh vegetables purchased from a farmers' market), for export (e.g., grain to Japan), to government (e.g., under farm price support programs) and for gross investment which, in the case of agriculture, would be mostly changes in inventories of crops and animals. The corresponding output entries in Table 6-2 are 12a, b, and c; 13 and 14. The entries on the input side require more explanation. Note that profits, item 3, are calculated after subtracting capital consumption, item 9. Subsidies, item 8, are entered negatively because they are a free good. They can be thought of as a balancing item. If the subsidies are for purchases of inputs, say fertilizer, then the price of the inputs will not be fully reflected in the prices of commodities farmers sell. Consequently, the value of inputs purchased will exceed the value of outputs sold unless the subsidy is subtracted from the input side. In American agriculture, however, subsidies show up as government payments added to net farm income (Table 6-1). In Table 6-2 these subsidies are included in item 3. However, the subsidies are not reflected on the output side of the farm ledger, that is they are not attributable to the goods and services farmers produce and sell. Consequently, the farm income subsidies must be subtracted from the input side to assure balance between the inputs and outputs. The inclusion of a charge for consumption of natural resource capital (item 9b) assures that this cost of farm output is reflected on the input side just as is the consumption of human-made capital. Recall that profit (item 3) is calculated after accounting for capital consumption. To the extent that farming "consumes" natural resource capital, farm profit is reduced, as it should be.9 Items 10 and 15 respectively are the environmental inputs and outputs of the farm enterprise. The environmental inputs (item 10) are entered negatively because they are free to the farmer, that is, they are a subsidy to farm production. For example, the unimpeded use of 9   In fact, consumption of natural resource capital may also show up as reduced rental income to the farm (item 2). For present purposes it does not matter whether profits or rental income is charged with consumption of natural resource capital so long as the charge is fully reflected on the input side of the ledger.

OCR for page 111
Assigning Economic Value to Natural Resources TABLE 6-2 An Accounting Framework for the Farm Sector Inputs Outputs Gross farm input before adjustment for environmental inputs Gross farm output before adjustment for environmental outputs 1. Purchases from other sectors 2. Compensation of employees and proprietors, including rental income 3. Profits after inventory evaluation and capital consumption adjustment 4. Net interest paid 5. Imports 6. Transfer payments 7. Indirect taxes 8. Subsidies received (-) 9. Capital consumption adjustment a. Human-made capital b. Natural resource capital 12. Current account sales to the private sector a. to industry b. to households c. for export abroad 13. Sales to government 14. Sales for gross investment Gross farm input adjusted for the environment Gross farm output adjusted for the environment 10. Environmental inputs (-) a. Land b. Water c. Air 11. Net environmental benefit (15-10) 15. Environmental outputs a. Land b. Water c. Air   SOURCE: Adapted from Peskin (1989b in Ahmad et al., 1989).

OCR for page 111
Assigning Economic Value to Natural Resources streams by farmers as dumps for sediment and chemicals in runoff is a free service. The value of the service is the loss of income to farmers if they had to eliminate these discharges. Environmental outputs, item 15, may be either negative or positive. For example, the value of reduction in recreational uses of streams because of sediment and agricultural chemicals is a negative environmental output (a "bad" of farming). The aesthetic value of a well-tended farm landscape is a positive environmental output (a "good"). Item 11, a balancing entry, is the absolute difference between items 15 and 10. For convenience it is called a measure of net environmental benefit, meaning of course that it may be positive, negative, or zero. An example will help to explain the environmental entries in the farm product account. Suppose that the sums of gross farm inputs (items 1-9) and outputs (items 12-14) before adjustment for environmental inputs and outputs are each equal to $1 million. Now suppose that the value of environmental inputs to farms (e.g., the savings from being able to dump sediment in streams free-of-charge) is $10,000, and that this is the only environmental service received by farmers. Item 10b in Table 6-2 would therefore have a value of-$10,000, as would item 10 as a whole. Gross farm input adjusted for the environment then would be $990,000. Suppose also that the sediment reduces recreational benefits of the receiving streams by $15,000 and that the scenic amenities of the farm landscape generates benefits of $10,000. Item 15a would then show an entry of $10,000 and item 15b an entry of-$15,000. Total environmental outputs then would be-$5000, and total gross farm output adjusted for the environment would be $995,000. Net environmental benefits, the absolute difference between items 10 and 15, would be $5000, raising total gross farm input also to $995,000. (Table 6-3 presents these hypothetical numbers.) The interpretation of this way of handling the environmental account runs along the following lines. The $10,000 of free inputs provided by streams receiving sediment is a benefit either to farmers (profits are higher) or to society (commodity prices are lower), or to both farmers and consumers. In either case, the $10,000 in environmental benefits is shown in the input account, and subtracted to avoid double counting. On the output side the positive $10,000 in scenic amenities are added to the negative $15,000 sediment-induced losses of recreational values to give a $5000 loss of environmental outputs. Net environmental benefits, therefore, the absolute difference between environmental inputs and outputs, are $5000, and total gross farm input and output, adjusted for environmental transactions, is $995,000. Notice that the adjustment for environmental inputs does not change total gross input. It merely reallocates the inputs to make explicit the contribution of the environment. To see this, suppose that before adjustment the $10,000 in free environmental inputs is reflected in some combination of higher profits (item 3 in Table 6-2) and in lower prices to purchasers of farm output (items 12-14 in Table 6-2). The $10,000 in environmental benefits thus is buried in both the unadjusted input and output accounts, each of which total $1 million. Suppose also, to keep things simple, that there are no environmental outputs (item 15 in Table 6-2 is zero). This implies that the quality of the water in the streams receiving the sediment from farms is not seriously impaired for users of the water. So item 10 in Table 6-1 is-$10,000, item 15 is zero, net environmental benefits (item 11), the absolute difference between items 10 and 15, are $10,000 and gross farm input and output is $1 million, as it was before the environmental

OCR for page 111
Assigning Economic Value to Natural Resources TABLE 6-3 A Hypothetical Account of Gross Farm Product Gross farm input before adjustment for environmental inputs: Gross farm output before adjustment for environmental outputs:   $1,000,000   $1,000,000 Environmental input (-)   Environmental output   Land - 0 - Landb +$10,000 Watera -$10,000 Waterc -$15,000 Air - 0 - Air - 0 -   Total-$10,000   Total-$5,000 Gross input minus environmental inputs $990,000     Net environmental benefit $5,000     Adjusted gross farm input $995,000 Adjusted gross farm output $995,000 NOTES: a To dispose of sediment; b Scenic amenities; c Loss of recreational values because of sediment damage to water quality. SOURCE: See text. adjustment. The adjustment nevertheless is important because it shows the $10,000 where it properly belongs, in the environmental input account, not hidden in the profit account or in lower prices of farm output. The implication of this argument is that if the adjustment for environmental transactions changes the estimate of gross farm input and output, the change will show up in the environmental output account (item 15 in Table 2). This is evident in Table 6-3. Scenic amenities increase output by $10,000, but damages to water quality reduce it by $15,000. Gross adjusted output thus falls by $5000 to $995,000, as does gross adjusted input. That adjustments for environmental transactions which affect gross input and output should show up in the environmental output account makes sense. We are accustomed to thinking that agricultural production generates a set of both good and bad environmental outputs. If the bads—damages or costs—exceed the goods—benefits—then the adjustment for environmental transactions should reduce the value of gross farm output. And vice versa if the goods exceed the bads. This interpretation of the role of environmental outputs in adjusting the gross farm product account probably will raise few eyebrows. The argument that inputs of environmental

OCR for page 111
Assigning Economic Value to Natural Resources services—item 10 in Table 6-2—generate benefits may not be so readily accepted. The idea that when farmers use the environment (e.g., neighboring streams) to freely dispose of farm effluents (e.g., sediment in runoff) they generate private and social benefits goes against the grain. Yet unbiased reflection indicates that these benefits are real, showing up, as indicated above, either in higher net farm income or in lower farm commodity prices, or in some combination of both. There are two points to keep in mind when considering the treatment of inputs of environmental services as benefits. One is a point made above: this treatment does not increase gross farm input; it merely reallocates it to make explicit the contribution of environmental services. The second point is that the benefits of environmental inputs are not necessarily net. The free use of the inputs may well generate environmental costs which exceed the benefits. For example, the recreational values lost because of sediment damages to water quality may be greater than the benefits of the inputs to farmers. This is the situation represented in Table 6-3. Indeed, the marginal social costs of this use of streams almost surely will exceed the marginal benefits except in the unlikely case that the social value of the water is insensitive to sediment damage. The reason is that farmers have incentive to use the water as a sediment dump to the point where the marginal benefit received by them is zero. This is because the marginal cost of the water to the farmer is zero. If the marginal value of the water in other uses, for example recreation, is greater than zero, then the marginal net social benefits of the farmers' use are negative and the pattern of use is socially inefficient. That is, the marginal social cost of the sediment damage is greater than the marginal benefits to farmers of using the stream to dispose of sediment. The pattern of resource use may also be inequitable if the water available to downstream users formerly had been unpolluted by sediment. Thus the treatment of the free services of the environment to farmers as a benefit on the input side is designed to give an accurate account of one aspect of the role of the environment in farming. The treatment does not imply that the benefits are greater than the costs (measured on the output side), and it is quite consistent with the argument that the free use of environmental services by farmers (or anyone else) is socially inefficient and possibly inequitable as well. In summary, the accounting framework presented in Table 6-2 offers a way of dealing with the two principal limitations of current representations of the farm sector in the national income accounts: (1) it specifically includes an item for consumption of on-the-farm natural resource capital; (2) it specifically accounts for the inputs of environmental services to farm production and for the outputs of environmental benefits and costs. The inclusion of consumption of natural resource capital assures, in principle, a more complete accounting of on-farm costs; and the inclusion of the environmental inputs and outputs does the same for off-farm costs and benefits. Adoption and implementation of the accounting framework, therefore, would provide more complete information than the existing system for assessing the sustainability of current agricultural practices.

OCR for page 111
Assigning Economic Value to Natural Resources CONSUMPTION OF NATURAL RESOURCE CAPITAL: HOW TO REPRESENT IT10 Recall the definition of capital consumption: it is the amount of money which must be withdrawn from annual11 gross income and reinvested to replace the capital used up in production. The reinvestment, however, need not be in the "used up" assets. If the capital consumption allowance is not reinvested but is spent for consumption, then the capital of the enterprise (or the sector, or the nation) will not be maintained and productivity, measured by net income of the activity, will decline. In agriculture the natural capital of principal interest is the nation's stocks of land and water resources used in production.12 Soil degradation, mainly by wind and water erosion, is the principal threat to the productivity of the nation's agricultural land resource. The "mining" of aquifers and impaired surface water quality, primarily from rising salt loadings in irrigation return flows, are the main threats to the water resource. The data available for analyzing erosion-induced declines in the productivity of the soil resource are much more ample than those for analyzing the comparable threat to the water resource. And the principles for incorporating these threats in a farm income accounting system are the same for water as for land. For these reasons the discussion here deals with the' land resource. It is clear that if soil erosion reduces the productivity of the soil resource, proper measurement of farm net income requires that the productivity decline be reflected on the input side of the farm, and farm sector, income and product accounts. In terms of Table 6-2, the money required to compensate for the loss of soil productivity must be entered in one of the input-side accounts and subtracted from item 3, the profit account. 10   In a longer version of this paper I deal also with measurement of environmental costs and benefits of agricultural production. Space limitations preclude dealing with those issues here. 11   As noted earlier, I assume here that the accounting period is annual. This is a convention. The period could be more or less than a year. 12   The plant and animal gene pool and the climate also are important parts of the natural capital employed in agriculture. And their productivity could be degraded in the course of agricultural production. For example, the development of a few high yielding crops may result in neglect, and eventual loss, of some of the genetic variety represented by wild, lower yielding representatives of the favored crops. The evidence does not suggest that this has, in fact, happened in the U.S. But the example illustrates a form of possible consumption of agricultural natural capital. Methane emissions from farm animals contribute to global warming, hence to possible depreciation of the climate resource available to agriculture. However, U.S. agriculture is a minor contributor to global warming (Council for Agricultural Science and Technology, 1992).

OCR for page 111
Assigning Economic Value to Natural Resources But ambiguities arise with respect to where on the input side the soil productivity decline should be represented because farmers have several alternatives for dealing with the effects of erosion on soil productivity. They can adopt tillage practices or crop rotations which reduce erosion enough to eliminate the productivity effects of erosion and over time restore soil productivity by building up soil organic matter. They can invest in the building of terraces, grass waterways, strip cropping or other ''structural" approaches to controlling erosion and restoring soil productivity. If the productivity effect of erosion is to reduce soil nutrients farmers can respond by putting on enough fertilizer to compensate. Farmers may decide not to control erosion but to use the capital consumption allowance to invest in new technology, e.g., higher yielding crop varieties, to compensate for the loss of soil productivity. This response typically would also require increased outlays for fertilizer. Farmers may decide to gradually withdraw some or all of their capital from agriculture, investing the allowance for consumption of soil capital in the stock market or some other nonfarm enterprise. In this case the capital, and stream of net income, of the farmer is maintained, but the capital invested in the soil resources of the farm declines, as does net income from farming. All of these alternatives are open to farmers. All of them reflect the negative effects of erosion-induced losses of soil productivity on net farm income. Presumably farmers choose among the alternatives, or among some mix of them, to minimize the loss of net farmer income. But how would the alternatives best be represented in a farm accounting framework such as that displayed in Table 6-2? Recall that capital inputs are represented in the table solely in the capital consumption account, items 9a and 9b. Outlays for new machinery and equipment are not included, only the necessary annual consumption charges against these assets. But all of the alternatives likely would involve some measure of current account expenses, all of which must be included on the input side of Table 6-2. The first alternative, for example, likely would require more labor than the erosion-inducing practices it would replace. These labor expenditures would be included in item 2 of Table 6-2. Alternatives 3 (more fertilizer) and 4 (higher yielding seed varieties) would require increased expenditures in item 1. Only alternative 2 would appear to involve predominantly, although not exclusively, capital outlays, to be accounted for in item 9a, consumption of human-made capital. The question is this: if the capital consumption allowance for the erosion-induced decline in soil productivity is spent on current account inputs (e.g., item 1 in Table 6-2), should the same amount also be charged against the capital consumption account for natural capital, item 9b? The answer would seem to be "no" because to do so would involve double-counting of the productivity effect of soil erosion. I am assuming that the outlays, under alternatives 1-4, exactly compensate the net income effect of erosion.13 The output side of Table 6-2, therefore, 13   I do not further consider alternative 5 because I am interested only in situations in which farm capital is maintained.

OCR for page 111
Assigning Economic Value to Natural Resources is unaffected by erosion, which means, of course, that total input also is unaffected. The increase in current account outlays in, for example, item 1 must, therefore, be matched against an equal decline in some other input account, probably item 3, profits (although it could be in item 5, imports). If the capital consumption account, 9b, also is charged, this too would be reflected in item 3, profits. In this case profits would be charged twice for the erosion-induced decline in soil capital, which seems incorrect. The way out of this conundrum is to follow this principle: if the alternative, or combination of alternatives, chosen to deal with the effect of erosion on soil productivity would indefinitely maintain the same level of output, then no capital consumption charge should be levied against soil capital. However, if in time additional expenditures to deal with erosion-induced losses of productivity would be needed beyond those required by the chosen alternative, then a capital consumption charge sufficient to accumulate the additional funds by the time they are needed must be entered in the natural capital consumption account (item 9b in Table 6-2). For example, suppose that the economically optimum choice for the farmer "today" is alternative 3—additional fertilizer to offset erosion-induced losses of soil nutrients. Suppose also, however, that after 10 more years of erosion, topsoil depth will have been reduced enough to significantly diminish soil water-holding capacity. Much research indicates that when this happens, additional fertilizer will no longer suffice to offset erosion-induced productivity losses. To avoid these losses the farmer will have to shift to erosion-control practices (perhaps alternatives 1 or 2) before soil water-holding capacity has been diminished to the critical point. To accumulate the funds needed to finance the shift to erosion control practices when that becomes necessary (10 years in this example), an annual charge should be entered against the capital consumption account for natural resources, beginning "now," and subtracted from farm profits (item 3 in Table 6-2). This way of dealing with consumption of natural capital will assure that the costs of consumption are always reflected in the farm, and farm sector, income and product accounts, either in the capital consumption account for the natural capital itself or in one of the operating cost accounts. Accordingly, net farm income (or profit) will always be less by the amount of the costs of capital consumption, as it should be. The current failure of the national agricultural accounts to reflect consumption of natural resource capital thus would be corrected. CONCLUSION The sustainability of U.S. agriculture can be judged by whether over time the net social income per unit of resources employed by the system is maintained. In making the judgment it is essential that all of the outputs generated by the system, both on and off the farm, and all the inputs employed by the system, also both on and off the farm, be counted. The various outputs and inputs can be represented in a system of accounts, as in Table 6-2 for example. Such an accounting system has the virtue of representing both the marketed outputs and inputs of agriculture, as in the present system of accounts, but also the unmarketed outputs and inputs now left out of the present system. In this paper the focus is on how to represent on-the-farm inputs of natural capital in a revised system of farm sector accounts. The discussion dealt with this issue with respect to

OCR for page 111
Assigning Economic Value to Natural Resources agricultural land. There can be no doubt that if the productivity of the land is reduced by soil erosion or any other form of land degradation, the effect of this must be reflected in the farm sector and national accounts to assure that the contribution of agriculture to social income is accurately measured. How best to represent the productivity decline in the accounts, however, is not so clear. Much depends on how farmers respond to the erosion-induced loss of productivity. If they do nothing, the output side of the accounts will decline and a corresponding capital consumption charge must be made on the input side and subtracted from farm income. The negative income effect of the loss of soil productivity thus will be properly reflected in the accounts. But farmers may take a variety of measures to counter the erosion effect, for example, by adding more fertilizer to maintain production. The output side of the accounts, therefore, remains the same but net farm income nonetheless falls, as it should, by the amount of the increased outlays for fertilizer. In this circumstance to also enter a capital consumption charge against the natural resources account and subtract it from net income would seem to double count the income effect of the erosion-induced loss of productivity of the land. The principle is clear: judgments about the sustainability of the agricultural production system require that any negative (or positive) effects of the system on the productivity of the natural resource base must be represented in the farm sector accounts. How best to represent those effects, however, depends on how farmers respond to them—or so it seems to me. REFERENCES Abroad, Y., S. El Serafy, and E. Lutz. 1989. Environmental Accounting for Sustainable Development, a UNEP-World Bank Symposium, The World Bank, Washington, D.C. includes the following: Bartelmus, P. 1989. Environmental Accounting and the System of National Accounts Blades, D. 1989. Measuring Pollution within the Framework of the National Accounts . Daly, H. 1989. Toward a Measure of Sustainable Social Net National Product. El Serafy, S. 1989. The Proper Calculation of Income from Depletable Natural Resources. El Serafy, S. and E. Lutz. 1989. Environmental and Resource Accounting: an Overview. Harrison, A. 1989. Introducing Natural Capital into the SNA. Hueting, R. 1989. Correcting National Income for Environmental Losses: Toward a Practical Solution. Lutz, E. and S. El Serafy. 1989. Recent Developments and Future Work. Peskin, H. 1989a. Environmental and Nonmarket Accounting in Developing Countries. ———. 1989b. A Proposed Environmental Accounts Framework. Council for Agricultural Science and Technology. 1992. Preparing U.S. Agriculture for Global Climate Change, Task Force Report No. 119, Ames, Iowa.

OCR for page 111
Assigning Economic Value to Natural Resources Crosson, P. 1992. United States Agriculture and the Environment: Perspectives on the Next Twenty Years. Unpublished manuscript, Resources for the Future, Washington, D.C. Faeth, P., R. Repetto, K. Kroll, Qi Dai and G. Helmers. 1991. Paying the Farm Bill: U.S. Agricultural Policy and the Transition to Sustainable Agriculture, World Resources Institute, Washington, D.C. Juster, F., P. Courant, and G. Dow. 1979. Social Accounting and Social Indicators: a Framework for the Analysis of Well-Being, Survey Research Center, Institute for Social Research, University of Michigan, ISR Code #8007, East Lansing. Levin, J. 1991. Valuation and Treatment of Depletable Resources in the National Accounts, Fiscal Affairs Department, International Monetary Fund, Washington, D.C. Norgaard, R. 1989. Three Dilemmas of Environmental Accounting, Ecological Economics, vol. 1, pp. 315-334. Office of the President. 1992. Economic Report of the President 1992. Government Printing Office, Washington, D.C. Peskin, H. with E. Lutz. 1990. A Survey of Resource and Environmental Accounting in Industrialized Countries, Environmental Working Paper no. 37, Sector Policy and Research Staff, World Bank, Washington, D.C. Repetto, R., W. Magrath, M. Wells, C. Beer, and F. Rossini. 1989. Wasting Assets: Natural Resources in the National Income Accounts, World Resources Institute, Washington, D.C. Repetto, R. 1992. Accounting for Environmental Assets, Scientific American, June, pp. 94-100. U.S. Department of Agriculture. 1991. Economic Indicators of the Farm Sector: Production and Efficiency Statistics, Economic Research Service. ———.1960, 1974, 1991b. Agricultural Statistics. Government Printing Office, Washington, D.C. Ward, M. 1982. Accounting for the Depletion of Natural Resources in the National Accounts of Developing Countries, Organization for Economic Cooperation and Development Centre CD/R (82), 3010, Pads.