The Environmental Data Dilemma

While environmentalists and their critics are girding for a new round of confrontations, they might give some thought to a problem of mutual concern: the glaring inadequacy of data about environmental quality. Nationwide monitoring of environmental quality has proved to be an extraordinarily difficult task, and so has the development of indexes that measure the impact of environmental changes on human health and on the entire ecosystem.

As is emphasized elsewhere in this issue in the article on Global 2000, anecdotal evidence and ad hoc studies abound, but comprehensive and dependable statistical data are scarce. Much of the first-hand data about pollution or environmental degradation tend to be fragmentary—limited to brief periods of time, to small geographic areas, or to selected pollutants. Reasonably reliable and comprehensive data about population growth and industrial output are available and are sometimes used as indicators of the impact of human activity on the ecosystem, but these data are likely to be more useful for framing questions than answering them. On many matters there are really no useful data available. Moreover, there is no conceptual framework to guide data collection and analysis and little coordination among agencies engaged in environmental monitoring and statistical activities.

Under these circumstances, environmentalists and critics alike—as well as the public at large—can hardly feel confident about reports on the seriousness of specified hazards or the progress of efforts to protect the ecosystem and human health. Even more uncertain are estimates of costs and especially of benefits of environmental protection programs, which are notable more for their ingenuity than for their conclusiveness.

The importance of good data for the effectiveness of an environmental protection program has been reiterated periodically for more than a decade, for example:

• As early as 1970 the Environmental Studies Board of the National Academy of Sciences urged the creation of a comprehensive federal program to monitor and evaluate environmental quality.
• In 1975—noting that little progress had been made on this recommendation —a committee of the same board proposed a number of recommendations "in the interest of accelerating progress toward the development of environmental indices."
• In 1977 a report on environmental monitoring, prepared by a committee of the National Academy of Sciences, called for a greater commitment to scientific principles and procedures in designing and operating programs to monitor sources of pollution, concentrations of pollutants in air, water, and soil, and the effects of pollution on human health and the ecosystem.
• In March 1980 an Interagency Task Force on Environmental Data and Monitoring that had been convened under the auspices of the Council on Environmental Quality (CEQ) called for the establishment of a National Environmental Statistics Program. Also during 1980 the National Center for Health Statistics submitted to Congress a report, Environmental Health, that recommended ways to improve data collection, with emphasis on the need for achieving better coordination among the various participants in the statistical system.

The recurrent complaints of data gaps, poor-quality information, and the lack of coordination do not suggest that what is available is useless or that no progress has been made. Rather, they underscore the view that existing information is inadequate for our needs given the high priority the public attaches to environmental quality, the extent of environmental regulation, and the high costs of both environmental damage and compliance with regulations.

The limitations of environmental data are illustrated by two publications of the CEQ: the Council's annual report, Environmental Quality-1980 and Environmental Statistics, 1978 (an updated version is scheduled to be published early in 1981, along with a new publication entitled "Environmental Conditions and Trends"). The annual report has scores of tables on specific environmental measurements, but carries very few data series that enable one to trace change in environmental conditions and their effects over time.

Environmental Statistics, on the other hand, offers many well-developed series, but a substantial proportion of them—perhaps half or more, depending on one's point of view—are demographic tables and indirect measures of stress on the environment, such as housing, construction, and industrial outputs. The coverage of the environmental series is spotty, partly because of the need to be selective, partly because the quality of some available data was too poor to merit inclusion, and partly because no data are available. Thus, the volume is instructive both because of what it omits and what it includes.

These limitations are largely the consequences of the CEQ's dependence on other sources for the data it assembles. These data generally have been gathered for other purposes, such as enforcement programs that are directly related to the mission of the gathering agencies. They do not represent the results of a statistical program created to meet broad policy needs or the requirements of public information. Although the CEQ has a mandate to gather and analyze environmental information, it has had neither the staff nor the resources to mount a statistical program comparable to what is conducted by federal statistical agencies such as the Census Bureau or the Bureau of Labor Statistics. Nor does it have the leverage to require other agencies to gather the information it deems essential. Indeed, no federal agency has such authority for all environmental data.

The government's slowness in developing comprehensive and dependable statistics on environmental quality is partly attributable to delays in giving the effort high priority, but in large part is is a reflection of the enormity of the problem. Although the national income accounts, employment statistics, price indexes, and various scientific measures have been cited as prototypes for an environmental statistical base, the analogy is misleading. The economic statistics are based on records of transactions and stocks that all participants have incentives to keep (most importantly, for tax purposes). Moreover, the dimensions of economic data are not comparable to the scope and complexity of environmental programs. The numbers of potentially harmful substances in the air, water, and soil, the problems of monitoring discharges and ambient conditions, and the difficulties in linking exposures to subsequent symptoms of illness or to ecological changes all point to a problem of far greater complexity than any other data operation now being conducted by the federal government.

Much of the discussion about improving environmental information has been focused on gaps in data, the specifics of monitoring, questions of data quality, and coordination of statistical efforts. But the underlying question—though not always considered explicitly—is, data for what purpose? What does the public want to know, especially the people whose comfort or health is most likely to be affected by pollutants? What do policymakers and government administrators need to know? Answers to such questions are clearly relevant to decisions about what information to seek.

The identification of purpose is only a first step, however, as is suggested in the companion piece on the Pollutant Standard Index.

The shortcomings of data about environmental conditions are not unique to the United States. The 1979 Organisation for Economic Cooperation and Development report, The State of the Environment in OECD Member Countries, makes it apparent that other—in this case, mostly European—nations are struggling with the same problems, both in scientific measurement and the construction of indicators.

Nor is the uncertain quality of data peculiar to the environmental field. It has been noted by others that government regulations require measurements of almost every physical, chemical, or biological phenomenon, but that many of these measures lack dependability.

The fact that similar difficulties confront other countries and are not limited to the environmental field reinforces the view that the job of establishing a comprehensive, coordinated, and reliable U.S. environmental information system will be especially difficult and will entail costs that will require close scrutiny before new programs are started. But it does not weaken the conviction that improving environmental information is an essential task. After more than a decade of discussion and piecemeal advances, a program to improve and coordinate environmental data ought to unite all who are interested in environmental policy.