National Income accounts, which provide the framework for analyzing the performance of an economy, are one of the most significant social inventons of the twentieth century. Their political and economic impact can scarely be overestimated. Consider the most famous component of these statistics, gross national product (GNP). When quarterly GNP figures emerge, policymakers stir. Should the numbers be lower, even marginally, than those of the preceding three months, a recession is declared, the strategies and competence of the administration are impugned, and public debate ensues.

The current system, first published in 1942, reflects the Keynesian model that dominated macroeconomic thought at the time. It carefully defines and measures the great aggregate concepts of Keynesian analysis; consumption, savings, investment, and government expenditures. But Keynes and his contemporaries were preoccupied with the Great Depression and the business cycle&emdash;they wanted to explain why an economy could remain at less than full employment for such a long time. The least of their worries was a scarcity of natural resources. Just as Keynesian analysis largely ignores the productive role of these resources, so does the United Nations system of national accounts, which most countries follow closely. As a result, a nation could exhaust its mineral reserves, cut down its forests, erode its soils, pollute its aquifers, and hunt its wildlife to extinction; all without affecting measured income.

The difference in the treatment of natural resources and other tangible assets provides false signals to both economists and politicians. It reinforces a false dichotomy between the economy and the environment that leads them to ignore or destroy the latter in the name of development. It confuses the depletion of valuable assets witli the generation of income. And it promotes and validates the idea that rapid economic growth can be achieved and sustained by exploiting the resource base.

The consequences can be illusory gains in income and permanent losses in wealth. The situation is especially serious for low-income countries, which are typically heavily dependent on natural resources for employment, revenues, and foreign exchange. As long as these governments use a system for national accounting and macroeconomic analysis that almost completely ignores their principal assets, their future is in jeopardy.

No Free Gifts

National income accounts have become so much a part of our life that it is hard to remember they are scarcely 50 years old. It is no coincidence that this half-century has been one in which the governments of most countries have taken responsibility for the growth and stability of their economies.

Before the mid-1800s, classical economists regarded income as the return on three kinds of assets: natural resources, human resources, and invested capital, land, labor, and capital, in their vocabulary. However, natural-resource scarcity played little part in neoclassical economics, from which most contemporary economic theories derive. In nineteenth-century Europe, steamships and railroads were markedly lowering transportation costs, while grains and raw materials were flooding in from North and South America, Australia, Russia, and the imperial colonies. What mattered to England and other industrializing nations was the pace of investment and technological change. Thus, neoclassical models concentrated almost exclusively on labor and invested capital.

The result is a dangerous asymmetry in the way we think about and measure the value of natural resources versus other assets. Buildings and equipment, for example, count as productive capital, and their depreciation is subtracted from the value of production. Consumption that diminishes the stock of capital is seen to reduce the sustainable level of income. But naturalresource assets are not so valued, so their loss entails no debit that would suggest a decrease in future production.

Should a farmer cut and sell timber to pay for a new barn, the farm's private accounts would reflect the acquisition of one asset, the barn and the loss of another, the timber. The farmer is better off only if the barn is worth more to him or her than the timber. In national accounts, on the other hand, income and investment would rise as the barn was built and as the wood was cut. Nothing would reflect the loss of a valuable asset.

Such anomalies come from the implicit, and inappropriate, assumption that natural resources are so abundant that a small loss has no economic cost. But natural resources make important contributions to economic productivity and face increasing pressure from human activities. Strictly speaking, they are economic assets.

Natural resources are also assumed to be "free gifts of nature,' entailing no investment costs to be written off. However, the real value of an asset is its income potential, not its investment cost. For example, brilliant ideas are the principal assets of companies worth billions of dollars. The Polaroid camera, the Apple computer, and the Lotus spreadsheet are worth vastly more than what their inventors spent to develop them.

It is important to remember that the common formulas for depreciating investment costs are just convenient rules of thumb and, in many cases, artifacts of tax legislation. The true measure of depreciation is the amount that future income will decline as an asset decays or becomes obsolete. Just as machines depreciate as they age, soils depreciate when their fertility diminishes.

The fundamental definition of income does, in fact, encompass the notion of sustainability. Accounting and economics textbooks alike say that income is the amount someone can consume now without reducing future consumption. Business income is the dividend a firm can pay without reducing net worth. Depreciation recognizes that unless physical assets are maintained, future consumption will decline. The failure to extend this concept to naturalresource assets is a major inconsistency.

Keeping Score Correctly

For resource-based economies, this failure seriously distorts evaluations of economic performance. Indonesian data illustrate the shortcomings of traditional accounts.

Over the past 20 years, Indonesia has drawn heavily on its considerable natural-resource endowment to finance development. Revenues from oil, gas, hard minerals, and timber and other forest products have offset a large share of government expenditures. The country's recent economic performance is generally judged to have been successful. Only a handful of countries have exceeded its growth in per capita gross domestic product (GDP), which averaged 4.6 percent per year from 1965 to 1986.

But a closer look reveals how much this evaluation is affected if the score is kept more correctly. Compare the growth of Indonesia's GDP with that of its "net" domestic product, which we have derived by subtracting estimated depreciation for soil erosion on Java and for petroleum and timber (see table). While GDP increased 7.1 percent per year from 1971 to 1984, the net...

Erosion's economic consequences include lost nutrients and soil fertility and increased downstream sedimentation in reservoirs, harbors, and irrigation systems. By ignoring these costs, income accounts significantly overstate the growth of agricultural income in Indonesia's highlands. Although upland crop yields have improved as farmers have used better seed and more fertilizers, the farm income that will be lost owing to the annual depreciation of soil fertility is about 4 percent of the value of crop production&emdash;equal to the annual production increase. In other words, in Indonesia's uplands, farm output is increasing at the expense of future output.

Such an accounting of natural-resource depletion should flash an unmistakable warning that a country is on an unsustainable course. A system that does not highlight dangers like these is a deficient tool for analyzing resource-based economies.

Of course, nations legitimately draw on natural resources to fund economic growth. This is especially true in resource-dependent countries, where the revenues pay for education, increasing industrial capacity, and developing infrastructure. To help planners invest resources wisely, a reasonable accounting system would identify when one kind of asset is exchanged for another.

Revising the Accounts

Fortunately, a large and growing body of experts has recognized the need to reform the national accounting systems. For example, in 1985, the Organization for Economic Cooperation and Development (OECD) issued its report, Declaration on Environment: Resources for the Future, which endorses steps to ensure "longterm environmental and economic sustainability" and commits OECD nations to developing "more accurate resource accounts." Similarly, Our Common Future, the 1987 study by the U.N.-sponsored World Commission on Environment and Development, observes that "in all countries, rich or poor, economic development must take full account in its measurement of growth of the improvement or deterioration in the stock of natural resources."

Some developed countries have already established environmental accounting systems. In both Norway and France, extensive resource accounts supplement the national economic accounts. The West German government recently announced that it will include resource and environmental degradation in its national income accounts. These systems reflect two types of approach to natural-resource accounting.

On the one hand, planners can register the stocks of natural resources&emdash;and changes in those stocks&emdash; in physical units. Opening stocks plus all additions less all reductions equals closing stocks. Considers for example, timber resources. Additions to the timber stock can originate from growth and regeneration of the initial supply, from reforestation, and from planting new forest. Production (harvesting), natural degradation (for instance, fire), and deforestation by humans would reduce the stock. Separate accounts might be established for categories such as virgin production forests, logged (secondary) forests, protected forests, and plantations. In temperate forests, which have relatively few tree types, stocks could be further classified by species. In each case, cubic meters of available wood is proba bly the most important measure, since a substantial part of a nation's standing timber cannot be profitably harvested with current technology or sold under today's market conditions.

However, physical accounting by itself has considerable shortcomings. First of all, it does not lend itself to amassing many different accounts into a single useful number. Combining the cubic meters of various tree types obscures wide variations in the economic value of different species. Similarly, aggregating total mineral reserves in tons obscures the vast differences that grade and recovery costs make in the value of deposits. But maintaining physical accounts in fine detail yields a mountain of hard-to-manage statistics.

Further, physical accounts must be expressed in monetary terms before planners can directly integrate them into economic decisions, presumably the point of the exercise. (The necessary calculations rely on the concept of economic rent, which is broadly equivalent to the net price. For example, if a barrel of crude oil can be sold for $10 and costs a total of $6 to discover, extract, and market, each barrel has a rent of $4. Natural-resource rents arise from factors such as the scarcity and location of particular stocks.)

Like physical accounts, monetary valuation has its limits, set mainly by how tightly a resource is tied to the market economy. Some resources, such as many minerals, are relatively easy to value in monetary terms, but others, such as noncommercial wild speciesj can be valued only through quite roundabout methods involving numerous, somewhat questionable assumptions.

Thus, Norway's resource accounts are tabulated in physical units, such as tons or cubic feet, and are not directly integrated with the national income accounts. Norway compiles accounts for "material" resources such as fossil fuels and other minerals, "biotic" resources such as forests and fisheries, and "environmental" resources such as land, water, and air. However, the Central Bureau of Statistics does express some important resource accounts, especially those for petroleum and gas, in monetary terms for macroeconomic planning and projection models.

France's "natural patrimony accounts" have been designed to provide a comprehensive framework for monitoring changes in all resources that can be affected by human activity. Since 1971, French statisticians have been developing the methodology for these accounts and compiling empirical estimates of the stocks of specific resources. The patrimony accounts now cover the same range of resources as Norway's: nonrenewables, the physical environment, and living organisms. As in Norway, the basic accounting units are physical, with monetary values for resources that are sold or contribute directly to producing marketable goods.

Some developing countries, recognizing their dependence on natural resources, have also become interested in a better accounting framework. The World Resources Institute is working on pilot studies with government researchers and statistical agencies in Indonesia, Costa Rica, and the People's Republic of China. The Philippines' government has recently begun compiling resource accounts, and the World Bank and the U.N. Environment Programme are planning to carry out limited experimental studies in Thailand, the Ivory Coast, Argentina, and possibly some other countries.

The Role of the United Nations

Since most developing countries must think of natural resources as productive assets, the first priority is to document these reserves in a way that gives due emphasis to the costs as they disappear. Here, the U.N. Statistical Office has an important role to play. Its system of national accounts (SNA), which includes privately owned assets used in the commercial production of goods and services, supplies a standard that most countries follow closely, at least in its main accounts. The Statistical Office is also a worldwide source of economic expertise and guidance in designing and using national accounts.

Already, the SNA is more complete with respect to natural resource accounting than the systems most nations derive from it. The framework provides for "reconciliation accounts" corrections to the main balance sheets that cover changes in the value of natural-resource stocks due to both price shifts and material alterations, such as growth, discoveries, depletion, and extraction. Reconciliation accounts cover not only reproducible assets, such as tree plantations, but also non-reproducible assets, such as farmland and subsoil minerals.

But dissatisfaction with the SNA stems from its many inconsistencies and omissions. For example, it largely fails to look at goods and services outside the market sector, notably those that households produce. Thus, when a person buys an apple pie at the supermarket instead of baking it, national income rises. Also, the SNA ignores important capital assets, such as education and workforce training. Furthermore, it only imperfectly measures goods and services that governments create. It treats as income-generating many activities undertaken only to avert or remedy the disadvantages of modern industrial society, such as pollution.

These and many other deficiencies have led to a long agenda of suggested improvements, and the U.N. Statistical Commission, advised by a number of expert working groups, is preparing to modify the SNA, as it does once every 20 years or so. However, the commission has evidently already decided to make no fundamental changes, even though deliberations will continue until 1991. Rather, the expert committees have proposed encouraging countries to link natural resource accounts to conventional national-income measures through "satellite accounts." In other words, natural-resource depletion would simply be an addendum to the main tables.

In a sense, the U.N.'s existing methodology is protected by its very inadequacy: wholesale reform is a massive task, and improvement limited to some aspects is hard to justify when so many other problems would remain. At both the national and international level, statisticians, who are typically short on staff, money, and raw data, resist changes, especially since so much is yet to be done to imp]ement the existing SNA.

Nonetheless, events of the past decade; such as coastal pollution, tropical deforestation, and the accumulation of greenhouse gases&emdash;demonstrate the importance of bringing natural-resource considerations into the main national income accounts as early as possible. Certainly another 20 years is too long to wait for reforms that are already overdue. Only when the basic measures of economic performance, codified in an official framework, conform to a valid definition of income will economic policies be influenced toward sustainability. While virtually all countries calculate na-tional income accounts, few have implemented past U.N. recommendations with respect to satellite tables because, with limited resources, they have "stuck to the basics." Moreover, politicians, journalists, and sophisticated economists continue to treat the GNP as the prime measure of economic performance. Even the "basic indicators" table that leads off the World Bank's annualWorld Development Report cites GDP, GDP growth per capita, and rate of inflation, but no net figures.

Since the burdens on statisticians are relevant concerns, we instituted the Indonesia study in part to obtain firsthand experience about the effort needed. We found that we could make reasonable estimates from existing information, so compiling and reorganizing it were the main tasks. In this pilot study, without the access to data that a government statistical office would have, a modest effort&emdash;some 12 person-months, mostly in the United States, shed substantial light on Indonesia's growth performance over more than a decade.

There is ample time before the U.N. Statistical Office announces a revised SNA to fully explore the implications of extending the concept of depreciation to natural-resource assets. That office should use this time to prepare for a change in the main accounts. Certainly the reforms could be put in place within three to five years.

At the same time, key international economic institutions; including the World Bank, the International Monetary Fund, and the OECD&emdash;need to begin to compile, use, and publish figures for net national product and income. And these institutions should ready themselves to assist the growing number of national statistical offices that are deciding to make and use such estimates themselves.

ROBERT REPETTO is director of the Economic Research program of The World Resources Institute in Washington, D.C. He is the author of several books, including The Global Possible (Yale University Press, 1985) and Public Policies and the Misuse of Forest Resources (Cambridge University Press, 1988).

Back...