accounts thereby create the illusion of rising national income when in fact national wealth is being depleted.
Indonesia provides an illustration of the potential for NRA, following the depletion method developed in a World Resources Institute (WRI) report (Repetto et al., 1989). Over the past 20 years, Indonesia has drawn heavily on its considerable natural resource endowment to finance development expenditures. Revenues from production of oil, gas, hard minerals, and timber and other forest products have offset a large share of government development and routine expenditures. Resource extraction contributes more than 43 percent of gross domestic product (GDP), 83 percent of exports, and 55 percent of total employment. Indonesia's economic performance from 1965 to 1986 is generally judged to have been successful: its per capita GDP growth averaged 4.6 percent per year, a rate exceeded by only a handful of low- and middle-income countries and far above the average for those groups. Gross domestic investment (GDI) rose from 8 percent of GDP in 1965, at the end of the Sukarno era, to 26 percent of GDP (also well above average) in 1986, despite low oil prices and a difficult debt situation (World Bank, 1988).
Estimates derived from the Indonesian case study illustrate how much this evaluation is affected by keeping score more correctly. Table 1 compares the GDP at constant prices with the net domestic product (NDP), derived by subtracting estimates of net natural resource depreciation for only three sectors: petroleum, timber, and soils. It is clear that conventionally measured GDP substantially overstates net income and its growth rate, because it does not account for consumption of natural resource capital. In fact, although the GDP increased at an average annual rate of 7.1 percent from 1971 to 1984, the period covered by this case study, the adjusted estimate of NDP rose by only 4 percent per year. If 1971, a year of significant additions to petroleum reserves, is excluded, the respective growth rates from 1972 to 1984 are 6.9 percent and 5.4 percent per year for gross and net domestic product, respectively.
The overstatement of income and its growth rate may actually exceed these estimates considerably because the estimates cover only three natural resources—petroleum, timber, and soils—on only one island, Java. Other important exhaustible resources that have been exploited over the period, such as natural gas, coal, copper, tin, and nickel, have not yet been included in the accounts, and neither has the depreciation of such renewable resources as nontimber forest products and fisheries. When complete depreciation accounts are available, they will probably, on balance, show a greater divergence between gross output and net income.
Other important macroeconomic estimates are even more distorted. Table 2 compares estimates of gross and net domestic investment (NDI), the latter reflecting depreciation of natural resource capital. NDI is central to economic plan-