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 163
Beyond the Market: Designing Nonmarket Accounts for the United States 8 The Environment Among the earliest, and now most thoroughly investigated, applications of nonmarket accounting were efforts to monitor change in the natural environment. While neglect of the natural environment in economic measurement concerned economists even in the nineteenth century, serious work to modify the conventional accounts did not begin until the late 1960s. Since then, there have been many official (governmental) and unofficial attempts at environmental accounting—those by the United Nations and the Bureau of Economic Analysis (BEA) have been among the most visible. The United Nations (1993) produced a Handbook of National Accounting: Integrated Environmental and Economic Accounting (referred to as SEEA).1 In the United States, BEA began intensive work on integrated environmental and economic satellite accounts (IEESA) in 1992, but in 1994 Congress directed BEA to suspend that work (see Bureau of Economic Analysis, 1994a). That such efforts have been undertaken reflects the view that there are benefits to be gained from developing natural resource and environmental accounts. A previous report, Nature’s Numbers, which describes and critiques the major environmental accounting efforts to date, identifies these benefits (National Research Council, 1999, p. 2): [Environmental and natural resource accounts] provide valuable information on the interaction between the environment and the economy; help in determining 1 An updated version (2003b), which has not undergone final editing and reproduction, is available at: http://unstats.un.org/unsd/environment/seea2003.pdf.
OCR for page 164
Beyond the Market: Designing Nonmarket Accounts for the United States whether the nation is using its stocks of natural resources and environmental assets in a sustainable manner; and provide information on the implications of different regulations, taxes, and consumption patterns…. More generally, augmented NIPA that encompass market and non-market environmental assets and production activities would be an important component of the U.S. statistical system, providing useful data on resource trends. This panel agrees with the basic messages of Nature’s Numbers, particularly the more overarching recommendations, which we strongly endorse. Recommendation 8.1: A concerted federal effort should be made to identify and collect the data needed to measure changes in the quantity and quality of natural-resource and environmental assets and 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 (National Research Council, 1999, pp. 7-8). The obvious place for new environmental accounting work to recommence is within the BEA, though the effort might equally well be housed at the Environmental Protection Agency, conducted with expert guidance from the BEA. Given this panel’s general agreement with recommendations published in Nature’s Numbers, the discussion of environmental accounting in this chapter is brief, focusing on the analysis, conclusions, and recommendations from that study. DEFINITION AND SCOPE OF COVERAGE Environmental accounting generally focuses on the measurement of natural or environmental wealth and the goods and services generated by this wealth. Accounting efforts have encompassed both current accounting and capital accounting—current accounting of the generation of pollutants or of the consumption of natural resources and capital accounting of changes in the stock of natural resources or the condition of the natural environment. For the most part, natural and environmental wealth excludes assets that are man made. Some environmental accounting systems—in particular, the French patrimony accounts—also include certain structures of special historical or cultural importance, such as ancient monuments. Examples of environmental goods generated by natural wealth are timber products and minerals. Examples of environmental services include recreational services, life support of animal and plant species, and esthetic services. One of the most important services is waste disposal, a positive benefit that is often accompanied by a “bad” or “disservice,” conventionally referred to as pollution. Valuing nonmarket natural assets is not an easy matter. Some environment-related goods, such as forestry products or mined minerals, are sold in markets and therefore pose no major problems; but others, such as clean air and water or
OCR for page 165
Beyond the Market: Designing Nonmarket Accounts for the United States unproven reserves of natural resources, are not priced in markets. A full accounting of natural resources and the environment must recognize that these assets, even those with market production costs of zero, have value, and that their value can be affected by human activities (Nordhaus, 2004, p. 12). Market and Nonmarket Factors At the outset, it is important to note that environmental and natural resource accounting is not necessarily the same as nonmarket environmental and natural resource accounting. Much environmental wealth, and the goods and services this wealth generates, is already measured in conventional economic accounts. Measurement is quite likely when legal, institutional, or technical arrangements permit the marketing of assets and any generated goods and services. Such marketing can be expected when the assets or the goods and services are privately owned. But marketing can take place even when the asset is in the public sector. A forest may be a public asset in one location and a private asset in another, but the harvested timber is usually marketed regardless of ownership. The timber, through leasing arrangements, may belong to the private sector even if the forest does not. While not as obvious as the sale of timber on public lands, the government can also market disposal services—many of which generate significant externalities (such as polluting air or water resources)—by establishing permits and fees. Without fees, the value of waste disposal services provided by a water body or an air shed typically will not be fully reflected in the conventional economic accounts. The extent to which it is reflected depends on who bears the costs associated with the disposal—that is, on whether the created externalities affect only other market producers (in which case it will be reflected in their input costs or output values) or if it impacts valued nonmarket activities or assets (in which case it will not be reflected). We return to this issue below. Clear rights of ownership, whether by private or public entities, do not guarantee coverage in conventional economic accounts. Regardless of ownership, a particular natural asset may generate a number of different goods and services, some of which are marketed and some of which are not. In our timber example, one could imagine that, if the nonmarket (non-timber) services—such as its contribution to watershed storage, its support of flora and fauna, or its support of recreation—were measured in monetary terms, the asset and output value of a forest could be many times the value determined by timber sales alone. One of the objectives of nonmarket accounting is to capture these kinds of output values. Given the often complex ownership arrangements for natural and environmental wealth, and that some of the generated goods and services are already captured in conventional accounts, it is not surprising that various accounting schemes differ in their coverage. Some cover only nonmarket activity that is excluded from the conventional accounts: natural resource depreciation, non-
OCR for page 166
Beyond the Market: Designing Nonmarket Accounts for the United States marketed goods and services, and pollution. Often the intent of these schemes is to augment the conventional accounts for the purpose of generating a superior (“greener”) measure of gross domestic product (GDP). Others have chosen a more comprehensive approach, covering not only nonmarket activity, but also market items (such as timber and mineral sales and the costs of pollution reduction) that are already covered in the conventional accounts. As set out elsewhere in this report, the panel favors the development of satellite accounts that include both the market and the nonmarket components of the activity in question. For the environment, an inclusive approach would more fully highlight the economic importance of the environment and natural resources and serve to disaggregate market activity so as to identify costs and expenditures that are related to the environment and to its protection.2 To reiterate, however, comprehensive sectoral accounts cannot simply be added up, either with each other or with the conventional income and product accounts—perhaps with the intent of generating a green GDP—since there would be double-counting. Subsoil Resources Currently, the NIPAs capture the value of minerals and other subsoil resources that are extracted and sold in the market economy. The NIPAs do not reflect changes in the stock of nonreproducible (or extremely slowly reproducible) assets such as oil or mineral reserves (Nordhaus, 2004, p. 13). Changes in these assets are excluded, not because they have no value, but because they have not yet flowed through markets. This is not unlike the current treatment of changes in the stocks of human or health capital explored in Chapters 4 and 5. The BEA’s expanded IEESA effort produced a “full and well-documented” set of subsoil mineral accounts that did value reserves (National Research Council, 1999, p. 4). The IEESA accounting of net investment in subsoil assets includes a negative entry for depletions or extractions of subsoil assets and a positive entry for additions to reserves. The IEESA methodology is similar to the NIPA treatment of conventional capital formation. We note that this approach is not universally accepted: the SEEA, for example, does not account for new discovery of minerals; only negative changes in the value of subsoil assets are possible in that account. We support the IEESA approach to accounting for nonreproducible asset 2 This task can be very difficult since environmental expenditures are often hard to separate from nonenvironmental expenditures. A firm may buy new equipment for reasons of profitability. Coincidentally, this equipment may be “cleaner.” In addition, some environmental control costs are not reflected in any increased expenditures. For example, a paper company may choose to reduce pollution simply by eliminating the production of bright paper. This choice has real economic costs for the firm, but it is not reflected in any increases in the costs of pollution-control equipment.
OCR for page 167
Beyond the Market: Designing Nonmarket Accounts for the United States reserves, as it is more analogous to the sensible way in which assets are treated elsewhere in the accounts. Cast in an input-output framework, the inputs reflect the net investment in subsoil resources, primarily spending on resource discovery and extraction. The output side reflects changes in the stock of reserves, valued at current prices, that have not yet been marketed. This calculation is in addition to the value of extractions that are sold in markets already captured as output in the NIPAs. Although natural resource discovery and depreciation are not reflected in conventional economic accounts and while attempts to do so involve an interesting controversy, accounting for these activities would not represent nearly so quantitatively significant a change as most of the others discussed in this report. This is because, empirically, resource discoveries and resource depletions are largely offsetting. Furthermore, because of their near-market character, including discovery and depreciation in national economic accounts (conventional or satellite) has been considered one of the least difficult nonmarket accounting recommendations to implement. Indeed, methods for estimating depreciation for both mines and forests exist (and are discussed in Nature’s Numbers) and have been used. This is not to say that asset depreciation is completely straightforward in practice. Even if there is only one product generated by an asset, such as a particular mineral, determining its natural lifetime is tricky. The gross stock of a mine is often uncertain and can even be hard to define. Similarly, in valuing discoveries of additional reserves, minerals that cannot be extracted economically do not have an equivalent value to those that can. That mines have been shut down due to lack of economic viability and then reopened after price increases, is indicative of this phenomenon. Ideally, proved and unproved (and easy and difficult to exploit) reserves should be priced differently (see Nordhaus, 2004, p. 15). Renewable Environmental Resources In terms of the magnitude of unmeasured nonmarket activity, renewable environmental resources and pollution are much more significant than subsoil assets. Their quantification and valuation involve more difficult methodology and data issues as well. In part because of this difficulty, BEA’s IEESA program had not yet begun to develop this part of the account at the time of its termination. The most interesting accounting issues relate to air and water pollution, for which externalities carry potentially very high values. The value of goods and services that can be produced from environmental resources are clearly linked to changes in the level of pollution. Some of the effects of pollution are captured in the market accounts, but some are not. For example, a reduction in the amount of sulfur dioxide (SO2) emissions or ground-level ozone may result in reduced worker absence due to illness (a market effect) but also in nonpriced health gains. Likewise, factors contributing to the quality of the environment may or may not be manifest in market expenditures. The cost of catalytic converters is directly
OCR for page 168
Beyond the Market: Designing Nonmarket Accounts for the United States reflected in automobile prices, but a firm’s choice of production technology may not fully embody the costs of pollution associated with different options. The extent to which the effects of pollution are captured in the NIPAs depends on who bears the costs. Nordhaus (2004) points out that there are two relevant cases involving externalities. In the first case, the entire impact of an externality flow is reflected in the market accounts, even though no market transaction occurs. If a chemical firm pollutes a nearby water source, and the sole harm that arises from that action is that a farmer’s crop yield (sold at market) is reduced, the flow takes place within the market. For accounting purposes, this case is a concern only if we want to disaggregate production accurately by sector—here, chemicals and agriculture. The second case, in which externalities flow across the market boundary, is important for aggregate output measurement. If pollution from the chemical plant impinges on nonmarket recreational opportunities or the population’s health status, then failure to account for these effects will distort total output and welfare measures (Nordhaus, 2004, p. 8). Standard accounting methods in the NIPAs would need revision to account properly for the second case since externality disaggregation changes the value added in both the nonmarket and market sectors. The relevance of this kind of information to policy is fairly obvious. Accounting data on externalities would assist policy makers charged with setting taxes or permit fees for emissions of pollutants or disposal of industrial waste. While market incentives are not yet widely used in environmental regulation in the United States, there is precedent; one important example is the marketable permits used in the acid rain trading program. If properly set, taxes and fees closely approximate the costs of damage associated with a harmful activity, thereby encouraging socially optimal decisions about production processes. But, whether or not fees and charges reflect the true value3 of the air and water services provided (such as the positive value of waste disposal services or the negative value of the pollution associated with waste disposal), a society that charges firms for the right to pollute will, by conventional market measures, look different from an otherwise similar society that is laissez faire regarding externalities. 3 The term “value” in this chapter refers to the amount individuals in society would be willing to pay for the services of the environment and to avoid any associated detrimental effects such as pollution. This valuation concept takes as given that individuals are the proper arbiters of value and that the environmental services (or damages) can, in principle, be substituted for equivalent amounts of other goods and services (including money) such that the level of value prior to the substitution can remain unchanged. These two assumptions are at the core of the concept of economic value as developed in the modern theory of neoclassical welfare economics; however, “economic value” may not be equivalent to other philosophical concepts of value such as intrinsic value or religious value. For an excellent discussion of economic value and how the concept relates to the techniques of benefit and welfare measurement, see Freeman (1993).
OCR for page 169
Beyond the Market: Designing Nonmarket Accounts for the United States In an accounting framework, there are two ways to handle environmental improvement or degradation that is tied to market production. One can think of pollution created by a firm in the course of its production of goods as a negatively valued output—the firm is producing goods, but it is also producing harmful emissions. Also, one can think of pollution-related environmental damage as a cost of production—to produce, the firm needs workers, equipment, and the environment for waste disposal. In either case, pollution can in many cases be quantified in terms of particulate levels or other physical units. It should be noted that pollution damage and the input of waste disposal services are not alternative measures of exactly the same thing: pollution associated with production. In fact, they are usually unequal in dollar terms and, indeed, waste disposal values can be quite high even when pollution damage is near zero (the reverse is also possible). For this reason, environmental accounting systems should keep these concepts distinct. Valuing degradation as it affects nonmarket outputs (e.g., health and recreation) is difficult because the link between pollution and health is not well understood and because valuing health increments is controversial, though the development of such valuations is a clear goal. Ideally, methods for valuing changes in output associated with changes in the quality of the environment would parallel those developed in the other nonmarket accounts. CURRENT ACCOUNTING APPROACHES Formal accounting systems, such as the System of National Accounts (SNA) or the NIPAs, provide a useful way of viewing the interactions between the inputs and outputs that characterize the modern economy. Indeed, one definition of a formal accounting system is that it is a summary of a process—a relationship that transforms inputs into outputs. Ideally, the accounting structures proposed and described in this report would contain information that would help researchers better understand underlying production functions for areas of nonmarket activity. A principal weakness of conventional accounting systems is that they ignore those inputs and outputs that do not trade in ordinary markets. In other words, they misspecify the economy’s “true” production function. Nonmarket accounting can be viewed as an effort to correct this misspecification. Concerns with the particular failure to measure nonmarketed environmental inputs and outputs has a history that parallels the concern with the failure of conventional economic output measures to reflect deteriorating environmental quality.4 There is widespread agreement that the conventional accounting systems, in spite of their neglect of environmental factors, have served their users well. Thus, 4 Leontief (1970) is an important early article on supplemental input-output accounting for the environment.
OCR for page 170
Beyond the Market: Designing Nonmarket Accounts for the United States most environmental accounting systems are designed as satellite accounts, as recommended throughout this report for other nonmarket areas. Still, environmental accounting efforts can be developed in close proximity with the conventional economic accounts. For example, the ENRAP system developed for the Philippines augments the conventional production inputs by including environmental waste-disposal services and the conventional production outputs by including recreation, esthetic, and biological support services, as well as the negative outputs associated with pollution (Peskin and delos Angeles, 2001). Thus, ENRAP has the appearance of an SNA-type input-output accounting structure with more rows and columns. While certain systems attempt to maintain close consistency with the SNA (with respect, for example, to sector definitions or valuation measures), the reliance on satellite accounts lessens the need for such consistency. Thus, for example, the Dutch NAMEA system (as well as the early Leontief system) relies on physical measures rather than the monetary valuations of the conventional accounts (Keuning, 1995). The Philippine ENRAP system relies on monetary measures, but its valuations often include consumer and producer surplus and are thus inconsistent with the market-type valuations of the SNA. The proponents of ENRAP do not consider this lack of consistency with the conventional accounts a major defect since the deficiency is driven by the type of additional information needed by users of the system. The proponents of other systems, most notably the U.N.’s SEEA, appear more concerned with consistency with the conventional SNA, but the accounting consistency that has been attained with SEEA is not costless. The SEEA 2003 comprises four major components: flow accounts for pollution, energy, and materials; environmental protection and resource management expenditure accounts; natural resource asset accounts; and valuation of nonmarket flow and environmentally adjusted aggregates.5 In an effort to maintain exact SNA definitions of productive sectors, SEEA cannot cover nonmarketed productive services of the natural environment, such as those associated with recreation, flora and fauna support, or waste disposal. In addition, current implementations of SEEA measure environmental damage and resource depletion by the cost of restoration. While the authors of SEEA recognize that such valuations are not justified in theory, the use of cost valuations is consistent with most SNA practice; for example, the SNA measures the value of governmental services at their cost. As a result, however, SEEA data cannot be used to assess the efficiency of prospective policies to protect the environment. The benefits of such policies are, by definition, equal to their cost. In spite of its weaknesses, SEEA has provided the framework for many 5 Full details of these accounts can be found at: http://unstats.un.org/unsd/environment/seea2003.pdf [accessed October 14, 2004].
OCR for page 171
Beyond the Market: Designing Nonmarket Accounts for the United States environmental accounting efforts worldwide. The suspended U.S. effort within the BEA, the IEESA system, was also an extension of the SNA (Bureau of Economic Analysis, 1994a). The overall structures and scope of SEEA and IEESA are similar. Both systems neglect nonmarket services of the natural environment. Perhaps the most important difference between them is that SEEA does not consider mineral discoveries as additions to capital stock (based on the argument that the minerals were already there in nature—they were simply unknown to man), while IEESA does count them. FUTURE DIRECTIONS For environmental accounting, broad agreement on approach or on implementation techniques has not yet emerged—again, in part because different users require different kinds of information. Experience has shown that it is technically easier to develop accounts for specific natural-resource sectors, such as mining and forestry, than for more general areas of environmental concern, such as air and water quality. As a result, many environmental accounting efforts have taken a staged approach, concentrating first on sector-by-sector natural-resource accounting and then, if financial resources permit, shifting focus to more general areas of environmental quality. While Nature’s Numbers (National Research Council, 1999) was sympathetic to this staged approach, it nevertheless recommended a more comprehensive approach: directing the effort to all areas that may be of environmental significance. The problem with the staged approach is that efforts tend to be directed towards problems that are easier and not necessarily those that are of more policy significance. The comprehensive approach is, of course, not without problems as well. First, it requires coordination. Ideally, there should be a commitment to a common framework but, as noted, there is no general agreement as to which of the several alternative environmental accounting approaches used around the world is the best. While Nature’s Numbers was critical of some features of certain accounting approaches, it made no recommendations regarding the best choice of accounting system. Nature’s Numbers did strongly back elements of the IEESA—for example, supporting the use of monetary (or valuation) systems, as opposed to the physical accounting approaches. Second, a comprehensive approach can make great demands on available expertise and accounting resources. Choice of approach may also involve tradeoffs among desirable attributes, such as accounting accuracy, policy relevance, and consistency with market accounts. Estimation Approaches and Practice While Nature’s Numbers was not intended to be a handbook on environmental accounting techniques, it did provide a general discussion of estimation methods,
OCR for page 172
Beyond the Market: Designing Nonmarket Accounts for the United States with more detail regarding estimates in the minerals and forestry sectors. This panel very much agrees with that book’s conclusion that, to the full extent possible, monetary estimates should rely on observed market prices and actual market behavior, rather than surveys projecting human behavior (e.g., contingent valuation studies). In addition, Nature’s Numbers expressed a preference for marginal valuations as opposed to total valuations that include consumers’ surplus. A major reason for this preference is to ensure that natural and nonmarket capital is treated in a manner consistent with the market-oriented NIPAs, in part so that valuations obtained for the former do not dwarf the value of capital for market sectors (Nordhaus, 2004, p. 13). The Nature’s Numbers panel recognized, however, that many existing valuation techniques, such as the travel-cost method to value nonmarket recreation services, provide estimates that contain consumer surplus. In actual practice, budget, staffing, and the lack of available data limit the ability to undertake de novo data development. Experience has demonstrated that making just one nonmarket estimate—such as the monetary value of a particular recreation area or the impact in monetary terms of water pollution in the Chesapeake Bay—can cost many thousands of dollars (Grambsch et al., 1993). Imputing the many similar estimates that would be required for a full set of environmental accounts would quickly exhaust many existing research budgets. Thus, the usual practice has been to “borrow” estimates made elsewhere or for other purposes. For example, an important source has been analyses undertaken by the U.S. Environmental Protection Agency (EPA) in connection with the development of its air and water regulations. The procedure for using these estimates requires an appeal to assumptions and a reliance on economic theory. The process can be illustrated with the ENRAP environmental accounts. An important entry on the “input” side of the ENRAP accounts is an estimate of the value of waste-disposal services provided by the environment to various industrial sectors (as well as agriculture, households, and governments). In the ENRAP system, this value is intended to equal what these sectors would be willing to pay for these services. In the absence of direct observations of willingness to pay, inputs were measured by estimating what the cost to a particular sector would be were it denied access to the waste-disposal service. These costs, in turn, were approximated by EPA data on the costs of meeting pollution abatement regulations. A number of assumptions are involved when accepting that the EPA estimates approximate the intended willingness-to-pay measure. For example, one must assume that the EPA cost estimates are efficient—that they represent the least-cost way to reduce pollution. In addition, one must also assume that the regulations (and the accompanying cost estimates) actually would eliminate all uses of the environment for waste-disposal purposes. In fact, many of the regulations are more practical; often the target is about 90 percent reduction from current levels—not the 100 percent that would be consistent with the intended
OCR for page 173
Beyond the Market: Designing Nonmarket Accounts for the United States willingness-to-pay measure. As discussed in Breyer (1993), the marginal cost of cleaning up “the last 10 percent” (e.g., from hazardous waste sites) can be very large. With this reliance on EPA data, ENRAP developers were able to confine new estimation to a minority of sectors for which it was believed that the EPA estimates were either totally inappropriate or simply not available. This practice greatly reduced the overall developmental costs. Nevertheless, the cost of the entire accounting process was not trivial. The Philippines ENRAP accounts, for example, cost about $4 million. This figure did cover several illustrative policy analyses considered important to the account development effort. When practical concerns dictate some reliance on borrowed data, it becomes unlikely that a final set of environmental accounts will meet ideal objectives. Data gaps and inconsistencies (especially with the conventional income and product accounts) should be expected. An important question is whether, in the presence of these deficiencies, the nonmarket accounting effort is still worthwhile. Past nonmarket environmental accounting efforts have addressed this question by considering whether the additional information—albeit imperfect—make the benefits of nonmarket accounting worth its costs. The answer depends largely on the purpose motivating the effort. At least with respect to environmental accounting, if the purpose is merely to obtain a measure of an environmentally adjusted GDP, there is, indeed, some question of whether the effort is worthwhile. Not only is it very difficult to obtain ideal measures, the less-than-perfect adjustments to GDP seem relatively small (though this may not be the case with other nonmarket activities covered in this report). Nevertheless, the information content of the satellite environmental accounts would be valuable in managing the nation’s assets and for improving regulatory decisions. As concluded in Nature’s Numbers: “improved natural-resource and environmental accounts can provide useful information on natural assets under federal management…. In the case of environmental resources such as air and water quality, a comprehensive set of environmental accounts would provide useful information on the economic returns the nation is reaping from its environmental investments” (National Research Council, pp. 31-32). One major finding in the environmental area concerns the relative importance of industrial versus household and agriculture pollution. Although policy has focused on the former, the environmental accounts have suggested that, for many pollutants, the major problem lies with the latter two sectors. Linkage with Other Nonmarket Accounting Efforts Efforts to build satellite environmental and natural-resource accounts have not typically been carried out alongside similar efforts in other areas of nonmarket accounting. The ENRAP effort was an exception since it originally was part of the Measurement of Economic and Social Performance project (MESP) funded by the National Science Foundation (National Bureau of Economic Research,
OCR for page 174
Beyond the Market: Designing Nonmarket Accounts for the United States 1978). Since the vast majority of environmental accounting efforts have been independent projects, there is little experience regarding how these efforts would have been affected were they part of a larger nonmarket accounting program. One can speculate, however, that, if work were progressing in other areas of nonmarket accounting, useful exchanges of information and methodology might occur. For example, estimates of the value of changes in morbidity and mortality developed in connection with health accounts could provide useful input into estimates of the value of those changes that are due specifically to changes in pollution levels.6 Such estimates have been used to measure the willingness to pay to reduce pollution—an important entry in a number of environmental accounting systems. Of course, information can flow in both directions. Those developing health accounts would probably find it useful to inspect the many estimates of mortality and morbidity that can be found in the environmental quality literature. Besides exchanges of data and information, one could promote coordinated research on valuation methods. The various techniques for valuing nonmarket services of the environment and associated damages that result from use of these services—hedonic estimation, travel cost methods, and the direct willingness-to-pay estimates typically used to value mortality and morbidity—could play a role in nonmarket accounting generally. In fact, one could imagine joint estimation efforts. For example, a well-designed property value study could support the development of those partial price elasticities needed to value the benefits of improved schools, better policing, improved infrastructure, and cleaner air and water—factors that not only help to determine the value of property, but also are nonmarket outputs of the government and environmental sectors. Of course, whether such joint estimation is efficient largely will depend on the available data. Estimating the various parameters associated with the above outputs assumes a rich data set free of strong collinearities. Lacking the variation in the data required to identify the parameters of interest, it might be more efficient to develop data specific to the parameter of interest. Thus, the effect of air pollution might be ascertained by looking at property values in regions that demonstrate sharp differences in air quality but not necessarily in school quality, while the effect of schools might be studied with data from areas that had sharp differences in school quality but not necessarily in air quality. Data Needs While the accounting objective requires monetary valuations, Nature’s Numbers did discuss the need for more complete physical information, especially data 6 See National Research Council (2002b), which summarizes research that estimates the public health benefits of proposed air pollution regulations.
OCR for page 175
Beyond the Market: Designing Nonmarket Accounts for the United States relating to environmental quality and its possible effects on human health and other activity. The emphasis on the need for better physical information reflects a reality familiar to those undertaking environmental benefit-cost analysis: knowledge about the methods of monetary valuation of environmental changes often exceeds knowledge of the physical changes themselves. For example, there are techniques that allow valuation of adverse health effects due to air and water pollution. Far less is known about the connection between any particular measured ambient air and water quality parameter and how this level of quality may affect human populations. Not only is there uncertainty as to effects on particular individuals, it is not clear which individuals are actually exposed. Indeed, data on actual ambient conditions can be very spotty. Water quality data, for example, are collected in relatively few locations in a timely and consistent manner, even though conditions can vary tremendously over time due to changes in water flows, temperature, and pollution discharges. Air pollution measurement also can be very spotty, especially in non-urban areas. Information on the stock, growth, and depletion of natural resources may be relatively more complete, but Nature’s Numbers cited needs for improvement even in this area, especially for resources under federal domain and for fisheries. Since much of the physical information needed to develop environmental and natural resource accounts is collected by various federal agencies as part of their administrative functions, cooperation with these agencies is essential. Nature’s Numbers (National Research Council, 1999, pp. 196-201) contains a useful appendix describing sources of physical and valuation data on natural resources and the environment. It includes details on the scope of activities and resources valued, and on valuation methods used; it also provides a good starting place for establishing where new data are needed. A NOTE ON THE SOCIAL ENVIRONMENT In this chapter we have briefly described how environmental resources factor into productive activities, whether market oriented or not. The environment, however, consists of not just physical factors, but social ones as well, and the social environment can affect the population’s productivity and well-being in ways that are at times analogous to influences of the physical environment. What is the social environment? People’s social networks consist of family, friends, and community. As discussed in Chapter 4, social factors influence the effectiveness with which families can develop human capital in children. This is very obvious, for example, in populations where household violence is prevalent. We also observe the importance of safety and level of support in school environments in the production of educated individuals, the subject of Chapter 5. While significant progress has been made in the area of physical environmental accounting, comparatively little has been done on the social side. One possible reason that environmental accounting research has generally neglected
OCR for page 176
Beyond the Market: Designing Nonmarket Accounts for the United States the social environment is that some aspects of it, while not lacking value per se, lack economic value in the sense that they are scarce. Justification for omitting social factors from environmental accounting might follow the reasoning for ignoring, say, energy production from the sun, which is valuable but not economically scarce. In this report, we do not address the social environment in detail, but we acknowledge that a comprehensive set of satellite nonmarket accounts would include information on the social environment, covering such factors as the degree of social cohesion, the stability and effectiveness of political systems, and levels of safety and security. An account ideally would link these largely nonmarket inputs to various market and nonmarket output aggregates. The social environment produces an output commonly referred to as social capital (the term is defined and briefly discussed in Chapter 1). Often, social capital is evaluated at specific geographic levels of detail, such as the community. One can observe many community-level variables that might relate to measures of social capital—such as membership in local organizations, volunteering, crime, and such public goods as education and local environmental amenities. But the quantity of social capital, as embedded in these indicators, is hard to quantify, let alone to price. One dimension of the social environment for which data and concepts are relatively developed is in the area of crime. For crime, one can easily observe quantities, though not prices (a feature common to many local public goods). There are four basic approaches that could be used to attempt to account for crime. One would be to take all of the income and product in the conventional NIPAs that relate to security systems, alarms, locks, etc., and subtract that out from standard market accounting data. Unfortunately, this tells us little if anything about how societal well-being changes as crime rises or falls. Another option would be to look at changes in murder rates and use value-of-life estimates as the price in a way that parallels valuation in health. The problem with this approach is that it helps us with only one, albeit an important, class of crime. An encouraging fact about local public goods is that they should be reflected in housing prices. We have discussed how housing prices might be used to help value the output of schools and even aspects of the local physical environment. This observation suggests two further approaches for accounting for crime: use of hedonic methods to value willingness to pay for reductions in crime; or of use discrete choice econometric models—in which people are presumed to choose from a discrete and finite set of options—to estimate that willingness to pay. Hedonic techniques have been applied by Smith and Huang (1995), Chay and Greenstone (2004), and Banzhaf (2002) to examine the value of air quality. Papers that have used housing hedonics to estimate quality-of-life indexes include Blomquist et al. (1988) and Kahn (1995). Banzhaf (2002) incorporated air quality, crime, and education in his index for Los Angeles for the period 1989-1994. Chattopadhyay (1999) and Palmquist and Israngkura (1999) estimated the value of air quality using discrete choice models. Banzhaf (2002) used discrete
OCR for page 177
Beyond the Market: Designing Nonmarket Accounts for the United States choice models to estimate the value of air quality, crime, and education in Los Angeles. There are still many problems with these methods, such as how to aggregate across households (see Pollak, 1989, on social Laspeyres indexes). Relative to hedonics, the discrete choice estimation approach has the advantage of allowing for heterogeneity in tastes and for allowing the “filling in” of the product space with new alternatives. But hedonic methods are computationally more tractable, and statistical agencies have had experience applying them. While we do not yet know whether in practice the two approaches would consistently yield similar results, Banzhaf (2002) found that the percent change in the cost of living implied by the results from his hedonic regressions were similar to those from his discrete choice models. The hedonic and discrete choice models raise some of the same issues. Both require that all public goods be included. Otherwise, an estimate of the cost of living is still biased. Both models also would require frequent estimation. Not only are quantities changing over time, so too are relative prices. For example, Cragg and Kahn (1997) find that people’s willingness to pay for a temperate climate has increased over time. Costa and Kahn (2003) find that value of life has increased over time. Most valuation work does not take price changes into account (e.g., Chay and Greenstone, 2004), but these can be very important and at least in the methodologies outlined above there is nothing that prevents doing so. Both hedonic and discrete choice models also raise the issue of what is a neighborhood. What level of detail in terms of geography and household characteristics do we need to work with to obtain good results? These are all areas for further research.
OCR for page 178
Beyond the Market: Designing Nonmarket Accounts for the United States This page intentionally left blank.
Representative terms from entire chapter: