associated nonmarket service flows. Greater emphasis should be placed on measuring effects as directly as possible, particularly on measuring actual human exposures to air and water pollutants. (Recommendations 4.3 and 5.9)
True public goods, such as biodiversity, species preservation, and national parks, present major conceptual difficulties for incorporation into a national accounting system. More work will be needed on techniques for measuring production flows and values for the assets and services of true public goods in order to make them compatible with the prices and quantities used in the core accounts.
Notwithstanding the awesome difficulties that arise in accounting for air quality, this is likely to be the single most significant sector in environmental accounts. Creating accounts for sectors such as clean air is an essential component of efforts to develop a comprehensive set of nonmarket accounts. However, the construction of air-quality accounts transcends the present scope and budget of BEA and will require further research on the underlying physical phenomena, measurement methods, and economics.
The cost to BEA and other agencies of developing and maintaining a set of augmented accounts will depend on the intensity and extent of the effort. The costs would be small for a minimal program of incremental improvements limited to a few natural-resource sectors. Estimates provided by BEA indicate that the cost of a small activity, including reinstatement of the pollution abatement survey, would be approximately $1.5 million annually. Developing a comprehensive set of environmental and augmented accounts would require more funds over a longer period. Although the cost of a comprehensive accounting system will depend on the extent to which BEA is able to draw on data and expertise from other agencies, a preliminary estimate is that a full set of accounts would require incremental outlays for BEA and other agencies of about $10 million per year for a decade or more.
In weighing future directions for environmental and augmented accounting in the United States, the panel concludes that developing a comprehensive set of nonmarket accounts is the most promising approach. Because of the high cost and low return involved, reliable nonmarket accounts will not be supplied by the private sector. In a country as large,