Measuring Environmental Values and Environmental Impacts: Going from the Local to the Global

Measuring the impact of global change depends critically upon our ability to measure the impacts of local changes and our ability to aggregate these across individuals, sectors, regions, and time. Difficulties are presented at each of these stages. There are strong philosophical and practical concerns in measuring environmental, health, social, and economic impacts even at the very local and individual level. Aggregating these across commodities, individuals, sectors, regions, and time runs afoul of our inability to make interpersonal and intergenerational comparisons, requiring significant and weighty value judgments to be made.

Rothman began by describing two studies he was involved with in recent years. The first, with the World Resources Institute, looked at the productivity of electric utilities and pulp and paper mills in the U. S. The researchers sought to point out that arguments implying that the rate of growth of productivity in these sectors had been declining due to environmental regulations were misleading because they considered only the "good" output of these sectors (electricity, paper, etc.) and ignored the "bad" output (air and water pollutants). To make this argument in the context of standard measures of productivity, they faced a particular quandary: what is the monetary benefit of reducing a unit of air or water pollution? In the second case, Rothman and colleagues were asked to prepare a report for Environment Canada answering the question: what would be the costs to Canada from climate change if no international agreement were reached and emissions of greenhouse gases continued unabated? Not surprisingly, and probably much to the chagrin of those who asked the question, their answer had a large element of "it depends."

These examples illustrate a common problem faced in the study of environmental and global change. Aside from the fact that we often do not know precisely what the impacts of our actions are on natural and human systems, even if we did, how do we translate these into measures upon which social and political decisions are commonly made? These measures tend to be, although they are not exclusively, monetary. Compounding this problem is that different people, as individuals, residents of a particular region, members of a particular economic sector, or members of different cultural groups, may very well value the same impact quite differently. This is further exacerbated when the impacts spread across generations. How then does society add all of these values up? And is the total value merely the sum of the individual values?

Different people may very well value the same impact quite differently. How then does society add all of these values up?

Rothman briefly described some of the key issues in the measurement of values, particularly for those things to which a price, as normally conceived, cannot be attached. He pointed out, in particular, the ways in which the scale at which goods and services are considered can affect the value attached to them. In this context, he addressed issues of interpersonal comparisons of utility and measures of social welfare. Utility here refers to some overall measure of an individual's well-being, which is determined, in large part, by the values attached to particular goods and services. Social welfare represents some aggregation of these individual utilities across all members of a society.

He then addressed an issue of particular concern to global change research - the aggregation of values across time. The key issues here are those of changing values, and the concept of discount rates and their relationship to intergenerational equity. He also commented on the problems posed by scaling for assessments of the impacts of environmental and global change. He stressed the importance of being open and honest about these difficulties and the inherently normative and subjective nature of addressing them.

What is to be Valued and How is it to be Measured?

Categories of Values

Over the years, economists have identified five different kinds of value: direct use values, indirect use values, option values, existence values, and other non-use values . Ideally, all of these values would be represented in the price of a good or service sold in an open, competitive market, in which the external impacts of the production and use of the item have been taken into consideration. Under these conditions, the price of a good would provide an accurate estimate of its economic value. In reality, these conditions are not met more often than they are - many items are not bought and sold, many markets are neither open nor competitive, and many externalities are not considered in the pricing of goods or services.

Economists have developed a variety of techniques for assigning economic values to goods and services for which there are no price-fixing markets. Economic value can be estimated through the price of proxy goods or services, often the cost of replacing the specific good or service of interest. Techniques to assess value via indirect markets exist, such as the travel cost method, which derives monetary estimates from data on costs of traveling to visit a particular recreational site or engage in a particular activity, and hedonic pricing methods, which attempt to decompose property and wage differentials into components that capture the use value of non-market goods such as air quality or mortality risk. Direct methods, such as contingent valuation, have also been developed in which artificial markets are created by directly asking people what they would pay for normally unpriced goods or be willing to accept as compensation for their removal.

The scale at which goods and services are considered can affect the value attached to them.

Though accepted by many within economics and other disciplines, these valuation techniques remain problematic in many situations. Many, if not most, existing markets are not free and do not consider externalities. For price to accurately represent economic value also requires that other, fairly strict requirements be met. These include the existence of a world of rational agents, who are perfectly knowledgeable about the consequences of their actions and able, therefore, to act consistently in a way that maximizes their individual utility. Critics complain that human behavior hardly ever lives up to this assumption. People do not understand the systematic effects of their behavior or the costs associated with various trade-offs. Other, non-economic values, such as family, community, and religion, are important as well, but economic valuation, as opposed to other, more qualitative forms of value elicitation, is not well suited to representing them. These other considerations make economic valuation very difficult, if not offensive to many.

This difficulty is reflected in the work that has been done to date on estimating the effects of global and environmental change. Of the five types of economic value mentioned above, the farther one moves away from direct use values, the less effort has been put into including these values. Some are even categorically excluded. For example, the forthcoming study by Mendelsohn and Neumann, which in many ways represents the state of the art, omits "non-market impacts, such as health effects, aesthetics, and some ecosystem impacts." These categories include some of the most severe effects, and an economic valuation that omits them will systematically underestimate the costs of climate change, providing an incomplete and distorted basis for policy making.

Incommensurability and Problems with the Notion of a Single Yardstick

Valuation generally implies the reduction of values to a common metric. In many cases, this metric is dollars or some other form of currency. The use of a common metric contains within it the very significant assumption about the substitutability of very different items. Differences of opinion about substitutability underlie many of the debates about long-term development and environmental degradation. This issue has been extensively discussed in the sustainable development literature and was raised in the IPCC Second Assessment Report.

Many, if not most, existing markets are not free and do not consider externalities.

The notion that everything can be made commensurable and reduced to a single currency is quite an anathema to many. If a dollar value can be placed on both a barrel of oil and a human life, does this mean that a sufficiently large number of barrels of oil can substitute for a human life? Can the extinction of a species be compensated for by placing the assets earned from its extinction in a bank and accruing interest on these assets? Klassen and Opschoor (1991) argue that, "values go beyond wants: values form a hierarchical system of separate and partly connected values at different levels. This hierarchy implies that values cannot be reduced to a single yardstick."

Other Aspects of Scale in Measuring Values

Another important scale-related concern, also associated with substitutability, relates to the physical scale at which the analysis is undertaken. This commonly arises in the case of trying to estimate the value of a particular site for recreation, such as a lake. Depending upon the presence or absence of viable substitutes, i. e. other lakes, which depends in large part on their physical proximity and whether they are included in the scale of the analysis, the value of the lake can differ markedly. In general, the more substitutes, the lower the value. Similarly, in analyses of global change, the impacts of forest loss or declines in agricultural productivity will differ greatly depending upon whether the analysis takes a local, national or global perspective.

Because economic valuation of global and environmental change is so difficult, it has only been attempted for a limited set of goods and services and a limited set of values in a limited number of areas. In particular for climate change, most of the work has focused on the United States. Many of the estimates for other regions are based upon transferring U. S.-based estimates to these very different locales. The process of extrapolating values to places other than where the original analysis was performed is called "benefits transfer" in the economic literature. A limited amount of work has been done to explore techniques to take into consideration differing biological, societal, and economic structures, as well as cultures and values. However, especially in the case of global environmental change, these structures, as well as the extent of the actual physical changes, fall well outside the range over which the original estimates were made.

Going from Individual to Societal Measures of Value

There has been a long-running debate in economics over the ability to develop aggregate societal measures of welfare that are consistent with notions of individual utility. Three key elements that present a difficulty in doing this are interpersonal comparisons of utility and the existence of societal values that lie outside of measures of individual utility. Less from an individual, and more from a sectoral perspective there are also interactions that may be missed in simple aggregations.

Interpersonal Comparisons of Utility

For a wide range of reasons, different individuals may place quite different values on the same impact. Similarly, they may rank order the same set of choices, e. g., bundles of goods or policy actions, very differently. More importantly, there is no clear means by which to make quantitative comparisons between individuals' utilities. These present significant problems for any attempt to aggregate from individuals to a larger group to determine either a measure of welfare or a rank ordering of choices for the group as a whole.

Can the extinction of a species be compensated for by placing the assets earned from its extinction in a bank and accruing interest on these assets?

Kaldor (1939) and Hicks (1939), in two famous articles, presented the notions of Pareto improvements and Potential Pareto improvements, which seemed to provide a way around these dilemmas. A Pareto improvement represents a change in which no individual is made worse off and at least one is made better off. In a Potential Pareto improvement, certain persons can be made worse off, as long as the potential exists for a redistribution to occur after the change such that no individual is made worse off and at least one is made better off. This holds regardless of whether the redistribution is actually made or not. Hicks is careful to note that the costs of redistribution must be considered in the overall analysis of the efficacy of any change.

This very utilitarian approach holds sway even to this day and is at the heart of traditional cost-benefit analysis. By explicitly separating the positive, i. e., objective, aspects from the normative, i. e., subjective aspects, economists can avoid the need to make interpersonal comparisons of utility and/or judgments on distributional issues - "it is quite impossible to decide on economic grounds what particular pattern of income-distribution maximizes social welfare" - and focus on issues of efficiency.

Arrow (1950), among others, has raised doubts about the value of the Kaldor-Hicks criteria for the purposes of making policy choices and measuring social welfare. Pareto optimality is almost useless for policy as rarely will there be a case where at least one person is not made worse off. Potential Pareto optimality is problematic, as was noted by Hicks, because of the potential for an infinite number of redistributions and therefore an infinite number of outcomes. Also, since other Pareto optimal outputs could be reached if there were an initial redistribution of assets, and these are not considered, the criteria sanctifies the status quo, which in itself implies an ethical choice.

Societal Values Outside of Individual Values Missing Interactions

There are important values that society may hold dear that transcend the individual. These values may be related to issues of equity or they may reflect goods and services that have a more common character, such as culture and particular aspects of nature and nature's services. Klaasen and Opschoor (1991) state that "Society may have values that deviate from (aggregated) individual values, e. g., on the basis of society's much longer life expectancy, society as a whole may thus value environmental quality more highly than individuals do." Arrow (1950) addresses these when he states that "In general, then, there will be a different ordering of social states according to the direct consumption of the individual and the ordering when the individual adds his general standards of equity (or perhaps his standards of pecuniary emulation). We may refer to the former ordering as reflecting tastes of the individual and the latter as reflecting his values."

On the basis of society's much longer life expectancy, society as a whole may value environmental quality more highly than individuals do.

Missing Interactions

By estimating economic values at a small scale and then aggregating these, important interactions and interdependencies are lost. In examining the higher-order effects of climate change, Tol (1996) notes that the "cost of impacts together is larger than the sum of the individual impacts." A prime example of this is seen in how agriculture has been dealt with in studies evaluating the effect of climate change. Since agriculture comprises only 1 to 2 percent of GDP in developed countries such as Canada, a complete destruction of Canadian agriculture would only result in a 1 to 2 percent GDP loss in standard measures of the effects of climate change. This ignores, however, the extreme dependence of other sectors of Canadian society on agriculture.

Aggregating Values across Time

Global and environmental changes tend not only to be large in spatial scale, but also long in temporal scale, spanning across more than one, if not many, generations. This brings with it the problems of how to estimate future values and how then to aggregate these across generations. Two specific aspects of these problems are changing values and discounting and intergenerational equity.

Changing Values

Estimates of value for future generations are sensitive to assumptions about future biological, societal, economic, and cultural systems. How vulnerable will these societies be? How much and how successfully will they be able to adapt? Perhaps more importantly, how will their values and preferences differ from those of members of the present society.

By estimating economic values at a small scale and then aggregating these, important interactions and inter-dependencies are lost.

Tol (1994) shows that, by considering the increased willingness to pay for environmental goods that is assumed to come with projected future increases in income, the value of preventing climate change may be significantly higher than common estimates. Furthermore, while such adjustments account for the rising relative preference of future generations for greater quantities of presently valued environmental goods and services, they do not take into account the possibility that future generations may value the environment differently than we do. Take the example of coastal salt marshes, which are now threatened by rising sea-level. Largely ignored today, except for their amenity value and the provision of wetland habitat, a century ago, diked salt marshes were among some of the most valuable agricultural lands in Canada, providing the hay that fueled horse-drawn society. An 1897 estimate of the economic costs of inundating Canada's salt marshes would have put a much higher price on these impacts than a similar estimate today. Similarly, eastern hardwoods, which were regarded by the professional foresters a century ago as weeds, are now valuable commodities harvested for biomass.

Discounting and Intergenerational Equity

Based upon the assumptions that people are impatient and would rather receive a dollar today than tomorrow (the pure rate of time preference) and that an extra dollar is worth less to a richer person than a poorer person (the declining marginal utility of income), economists often apply a discount rate to future values in order to aggregate these across time. The decision to use discounting and the choice of a discount rate are important because of the sensitivity of calculations of total values and thus policy recommendations that may flow from them. For example, at a 7 percent discount rate (as is commonly used in short-horizon project analysis), impacts of $1 billion 50 years hence have a present value of only $33.9 million; the same impacts 200 years hence have a present value of only $1,300.

The general notion of discounting is based on the idea of a single long-lived agent able to make rational trade-offs between present and future benefits. This assumption does not hold for the long time horizons involved in many global and environmental changes, which involve decisions about the inter-generation allocation of costs and benefits. Just as aggregation across commodities and across individuals within generations cannot be done independently of ethical judgments about distribution, so the choice of a discount rate with which to aggregate values across time is fundamentally an ethical choice about intergenerational equity. On these and other bases a number of authors have argued that discounting is inappropriate to the evaluation of issues involving long-term global and environmental change.

Conclusions

The weakest part of many global environmental assessments, particularly most Integrated Assessment Models, is in how the value of impacts is represented. Much of this stems from philosophical difficulties in attaching values to goods, services, and other aspects that are not normally valued in the same way as marketed commodities. Others are particularly related to issues of scale that present themselves with the need to do aggregations across commodities, individuals, space, and time.

Many of the problems identified here are not amenable to resolution from the perspective of an "objective" science. There will always remain a normative and subjective element to how they are dealt with, thus allowing for a multiple of possible conventions. As Norgaard (1989) argued concerning attempts to incorporate environmental values in the System of National Accounts, an open exploration of multiple approaches is desirable, from the perspectives of both scientific rigor and intellectual honesty.

References

Arrow, K. J., 1950. A Difficulty in the Concept of Social Welfare. Journal of Political Economy, LVIII: 328-346.

Costanza, Robert , R. d'Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R. V. O'Neill, J. Paruelo, R. G. Raskin, P. Sutton, & M. van den Belt, 1997. The value of the world's ecosystem services and natural capital, Nature, 387: 253-260.

Hicks, J. R., 1939. The Foundations of Welfare Economics. The Economic Journal, XLIX: 696-712.

Kaldor, N., 1939. Welfare Propositions of Economics and Interpersonal Comparisons of Utility. The Economic Journal, XLIX : 549-552.

Klaasen, G. A. J. and J. B. Opschoor, 1991. Economics of Sustainability or the Sustainability of Economics: Different Paradigms. Ecological Economics, 4: 93-115.

Mendelsohn, R. and J. Neumann (Editors). forthcoming. The Impacts of Climate Change on the American Economy . Cambridge University Press, Cambridge.

Norgaard, R. B., 1989. Three Dilemmas of Environmental Accounting. Ecological Economics, 1:303-314.

Tol, R. S. J., 1994. The Damage Costs of Climate Change: A Note on Tangibles and Intangibles, Applied to DICE. Energy Policy , 22: 436-438.

Tol, R. S. J., 1996. The Damage Costs of Climate Change Towards a Dynamic Representation. Ecological Economics, 19:67-90.

The weakest part of many global environmental assessments, particularly most Integrated Assessment Models, is in how the value of impacts is represented.

---

Forward to Next Section//Back to Table of Contents//AGCI Homepage//Comments: agcimail@agci.org