Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
art Understanding Voluntary Measures Thomas Dietz Several important lessons about voluntary arrangements involving the pri- vate sector emerge from the previous chapters. First, it is clear that far more attention must be paid to evaluation of voluntary efforts. This task will be facilitated by the longstanding tradition of evaluation research in the social sciences. Second, voluntary agreements present fundamental challenges to social theory, and a better understanding of them will require more integrative concepts of organizational behavior than are currently available. In particular, understanding voluntary agreements will require tentative solutions to the aggre- gation problem of linking the behavior of individuals to that of organizations. Third, voluntary agreements may provide important opportunities for meeting new environmental policy challenges, but for the most part, the voluntary agree- ments in place today are dealing with problems of the sort that traditionally have been handled by command-and-control regulation. Some creative institutional design will be required to engage voluntary agreements as tools for dealing with as yet uncontrolled forms of pollution, such as that from nonpoint sources. Fi- nally, more attention must be paid to the comparative analysis of voluntary agree- ments. Although it can be difficult to generalize across political economies and cultures, other industrial nations are engaged in interesting experiments with voluntary agreements, and much could be learned from this experience.) Further- more, some of the most important voluntary agreements, such as the International Standards Organization (IS O) environmental protocols, are international in char- acter. In this chapter, I will try to address each of these lessons, suggesting where research must move in the future if we are to have the knowledge base necessary to support innovative environmental policy. 319
320 UNDERSTANDING VOLUNTARY MEASURES POLICY EXPERIMENTS AND EVALUATION RESEARCH As Mazurek (this volume, Chapter 13), Nash (this volume, Chapter 14), and especially Harrison (this volume, Chapter 16) note, the initial claims of success for some innovative voluntary agreements should be viewed with skepticism. When care is taken to learn what changes in environmental impact can legiti- mately be attributed to a voluntary program rather than to other factors, the estimated effects may be less impressive than those contained in first reports that are less rigorous in methodology. Thus in developing and assessing the merits of voluntary agreements in the future, it would be wise to think carefully about methodological and conceptual issues of evaluation. Environmental policy evaluation is still in its infancy. The amount of re- search in the peer-reviewed literature is still small, though some evaluations also are found in the "gray" literature of internal agency documents and consulting reports. A major effort is underway to assess the effectiveness of international environmental treaties (Young, 2001). But if systematic evaluation is a new issue for voluntary agreements and other environmental programs, evaluation has been routine practice for many social programs.2 A great deal can be learned from this experience that will enhance our understanding of all environmental programs, especially programs involving the private sector. Avoiding Hubris More than thirty years ago, Campbell (1969) argued that nearly all policies and programs should be considered experiments. Clearly environmental policies are social experiments. We never have enough information about the particular circumstances of implementation nor a complete enough set of models and theo- ry to predict with great accuracy the outcomes of an innovation. As a result, we would be wise to treat changes in policy as opportunities to learn rather than assuming we know the outcomes in advance. We need informed humility, not hubris, in approaching new policies and programs. Informed humility requires that we build into the design of a policy evaluation research that can inform us about what has happened and why. In that way we will build a cumulative knowledge base that will allow better designed reforms in the future. This same general point has been made in the context of adaptive ecosystem management and social learning (Gunderson et al., 1995). Our knowledge is imperfect, things change, and the only wise way to proceed is one that is reflexive, allowing for trial, analysis, and revision. Our reflexivity can be improved greatly if attention is paid to evaluation while a program is being designed rather than attempting to construct an evaluation after crucial design decisions have been made and oppor- tunities to collect baseline data have been lost.
THOMAS DIETZ 32 Estimating Program Effects Is Difficult Only rarely in social or environmental programs can we conduct true exper- iments with random assignment of subjects (individuals, communities, or firms) to treatment and control groups. Such assignment typically would be cumber- some and perhaps illegal and/or unethical. In the absence of such an experimen- tal design, it can be difficult or impossible to attribute causal influence to a program because many other factors also are influencing the behavior of interest. A substantial part of the literature on evaluation research deals with the prob- lem of assessing causality and estimating true program effects in the absence of an experimental design. Researchers refer to "quasi-experiments" in which there is some variation in who participated and did not participate in programs, but there is not random assignment to experimental and control groups of the sort that facili- tates strong causal inference.3 As Harrison notes (this volume, Chapter 16), selec- tion bias can be a particular problem firms who join voluntary programs may be very different from those who do not. Appropriate statistical models can help in dealing with selection bias, but such bias is still a major issue in most evaluations of voluntary agreements (Heckman, 1976, 1979~. Although it may not always be possible to implement the ideal design, or even have sufficient data for the best possible statistical analysis, careful attention to the factors that interfere with esti- mating true program effects can help in avoiding gross errors and allow for proper caveats in reporting the estimates that are possible. Nor is the design question a purely statistical matter. Policy and normative considerations enter into determining the proper basis for making a comparison. The design of evaluation research must address the issue of what constitutes a "fair" comparison to use as a basis for judging the success or failure of a volun- tary program. The overall success of a voluntary agreement in reducing environ- mental impact needs to consider growth in the amount of production and con- sumption activity that can be attributed to an industry. A seemingly successful voluntary agreement could reduce impact per unit consumption or production even as overall impact increases as a result of growing population, affluence, and consumption per unit affluence. Evaluations should consider the improvement in environmental performance that is likely to have occurred over time without any intervention. Evaluations should also consider the reasonable upper bound of improvement set by factors such as the rate of replacement of capital equipment. They also should attend to possible tradeoffs in impacts for example, a reduction in emissions of a target- ed pollutant, at the cost of an increase in emissions of one not targeted. Integrat- ed measures of environmental impact, such as the "ecological footprint," may be useful in assessing changes in overall impact (Rees and Wackernagel, 1994; Rees, 1992; Wackernagel and Rees, 1996; Wackernagel et al., 1999~. Early evaluations that did not consider these factors were perhaps too generous in their assessments. In the future we can anticipate an important debate about what
322 UNDERSTANDING VOLUNTARY MEASURES comparison standards are appropriate. Some of that debate is foreshadowed in the chapters of this section, where the issue of what to measure in evaluating voluntary programs is discussed. Thinking About Program Goals The most obvious goal of voluntary agreements is to reduce environmental impact. But in evaluating voluntary programs, the obvious goal may not be obtained quickly or directly. As Herb and colleagues (this volume, Chapter 15), Harrison (this volume, Chapter 16), Furger (this volume, Chapter 17), and Prakash (this volume, Chapter 18) emphasize (see also Dietz and Stern, this volume, Chapter 1), a benefit of voluntary agreements may be a general change in organi- zational culture among participating firms that in turn leads to environmental protection receiving a much higher priority than otherwise would be the case. Changes in culture that reduce environmental impact may act more slowly than the implementation of new command-and-control regulations. But such cultural changes have large and broad effects in the long run. The implication for pro- gram evaluation is that there may be program outcomes that are worth monitor- ing in addition to short-term reductions in pollutant generation or resources used. Of course, measuring and evaluating such alternative or interim goals should not be a substitute for assessing success or failure with regard to the primary goal reduction in environmental impact. Indeed, one of the standard criticisms of voluntary agreements is that they produce good feelings, good publicity, in- creased awareness, and little if any change in environmental impact. In some circumstances it may be appropriate to design programs whose goal is to change organizational culture in order to provide the substrate for changes that reduce impact. If that is the goal, the program should be clear about what will and will not be achieved. It also may be useful to assess interim goals such as changes in organizational culture, establishment of environmental accounting systems, and so forth because they may be good indicators that a voluntary agreement is on the way to producing the desired environmental benefits. Interim indicators that are known to be good predictors of environmental outcomes may be useful as evaluation tools and as secondary program goals.4 Of course, even when a voluntary agreement program does not reduce envi- ronmental impact over what might be achieved by command-and-control regula- tion, it may be able to achieve the same change in impact with lower costs. Such savings can prevent increases in consumer prices that might be associated with environmental protection, provide both more funds for further investment in environmental quality, and reduce future resistance to addressing environmental problems. From the viewpoint of society as a whole, these spinoffs of cost- effective environmental regulation are beneficial. So one criterion for evaluat- ing voluntary agreements is not just their overall environmental impact, but the cost per unit of pollution reduction.
THOMAS DIETZ 323 The possible importance of secondary and interim goals such as changed organizational culture and practices leads us to the distinction between summa- tive and formative evaluation. Summative evaluation looks at the overall suc- cess or failure of a program with regard to its most important goals. It provides information intended to influence decisions about expanding or contracting or even eliminating the program based on what has been achieved. In contrast, formative evaluation is intended to provide guidance to program administrators and participants while the program is in process. The goal is to suggest mid- course corrections. Formative evaluation usually focuses more on program pro- cess and interim indicators of progress than on ultimate outcomes. It intends to provide advice rather than a judgment. Developing successful voluntary agree- ments will require both formative and summative evaluation. Participants and managers can benefit from research that helps them do better, but policymakers and society as a whole must be able to assess how well a voluntary agreement approach is working in a given domain. To the extent that voluntary agreements are intended to produce changes in organizational culture and routines, it may be necessary to rethink not only the variables that should be considered as outcome criteria, but also the statistics used to summarize the behavior of participants in a program or agreement. The mean and its analogs (conventional regression analysis, analysis of variance, and other versions of the general linear model) are appropriate if the intent of an evaluation is to focus on the sum of benefits to society. This follows because the mean times the size of the population is the sum of a variable across all members of the population. Thus analyses focused on the mean give a sense of the total benefits from a program. But in some circumstances, such as when a program is consid- ered a demonstration to point the way toward best practices, it may make more sense to focus on the tails of the distribution of organizational performance. Those firms that have done exceptional jobs in reducing environmental impact and those that have made the least progress may provide more information than firms whose performance under the program is "typical" (at the center of the distribution). The methods available for statistical analysis of "outliers" are not as well developed as that for understanding the center of a distribution, but increases in computational power are making more tools available for such analyses.5 A particularly helpful evaluation design might include a summative evaluation using procedures based on the mean performance to assess the overall effectiveness of a program with an analysis of the outliers. The outliers then could become the subject of case studies to determine why those organizations have done so well or so poorly compared with typical performance a formative evaluation goal. WHY PARTICIPATE? The problem of altruism versus self-interest has been one of the most vital themes in the social sciences over the last quarter of the twentieth century and is
324 UNDERSTANDING VOLUNTARY MEASURES central to any analysis of voluntary agreements.6 As Harrison (this volume, Chap- ter 16), Furger (this volume, Chapter 17), Prakash (this volume, Chapter 18), and Randall (this volume, Chapter 19) detail, the interplay between altruism and self- interest is also central to the motivations for participating in voluntary agree- ments. Given a voluntary agreement, profit-maximizing firms have substantial incentive to "free-ride" by signing on to the agreement, but not taking any costly steps to reduce environmental impact. Such free-riders receive most of the ben- efits of the agreement good public image, freedom from intrusive regulation- without bearing the costs. If many firms follow this narrowly rational strategy little reduction in environmental impact will occur. As Olson (1965) noted more than 35 years ago, there will be an undersupply of public goods, in this case environmental protection, compared to what would be optimal. Although the free-rider model is compelling in its logic, decades of research also have shown that it is not always a realistic model of human behavior. Codes of silence are observed by prisoners, commons are managed sustainably, and people make sacrifices for the collective goody In the case at hand, firms some- times change their behavior as a result of participation in voluntary agreements even when they could free-ride. To design better voluntary agreement programs, we need a better understanding of why firms act so as to reduce environmental impacts. Achieving that understanding will require careful theoretical and meth- odological investigations of the behavior of organizations. Although research on participation in voluntary agreements is in its early stages, several key issues already have emerged. Voluntary Participation Is Not Wholly Voluntary As Randall (this volume, Chapter 19) and Nash (this volume, Chapter 14) emphasize, when a voluntary agreement is developed for a serious environmen- tal problem, the threat of command-and-control regulation often is on the hori- zon. Participation in the voluntary agreement may be the result of comparing the costs and benefits of the agreement requirements with those that would flow from a regulation that can be anticipated if the voluntary agreement is not suc- cessfully implemented. Even when the threat of government regulation is not present, other firms who are customers may require products that meet environ- mental requirements. As Rejeski and Salzman note (this volume, Chapter 2), some corporate buyers dwarf their suppliers and can bring about substantial changes in products and production processes. Nor are corporate customers the only source of pressure. Even the largest retail firms may fear boycotts motivated by environmental concern, while, as Nash notes, environmentally friendly products may allow for niche markets that are advantageous to those who produce them. Finally, it is increasingly common for environmental standards, such as those incorporated in ISO 14001, to influ-
THOMAS DIETZ 325 ence access to export markets.8 Firms that hope to penetrate markets where such standards have substantial sway will act to meet the requirements and may alter products and production processes beyond what is required in their other markets simply because such uniformity is simpler. So voluntary action may be "co- erced" by concern with future government regulation, or by concern with present or anticipated actions by either corporate customers or consumers. A number of studies document such effects on firms' environmental policies (Henriques and Sadorsky, 1996; Nakamura et al., 2001~. Voluntary action may also be induced by regulatory or other environmental policy tools. As Herb and colleagues note (this volume, Chapter 15), one out- come claimed for the U.S. Toxics Release Inventory, a regulatory program re- quiring only the production of information, is voluntary efforts by firms to re- duce their output of listed pollutants. "Green" labeling programs (see Th0gersen, this volume, Chapter 5), whether developed by government or nongovernment groups, can provide a similar stimulus for voluntary action. Such possibilities for synergism between voluntary measures and other policy instruments are im- portant considerations for both program design and evaluation. Networks Matter At least since C. Wright Mills's The Power Elite it has been clear that major corporations typically do not make decisions in isolation from one another, even though direct collusion is prohibited by U.S. law.9 Trade or industry associa- tions may be especially important links in these networks, as Furger (this vol- ume, Chapter 17) and Nash (this volume, Chapter 14) note. To some extent, their influence comes through acting as the representatives of a business sector in policy debates.~° In that role, they may convey information about future government regulations that may encourage preemptive voluntary action. But they may also play a role of sharing information across firms, mediating and encouraging voluntary action. Linkages across firms that do business with each other also can be very consequential. Such transactions, even when governed by contracts, require trust, a point noted by Furger and Nash. Trust may be damaged not only by actions that directly affect the interests of a firm, but also by observing that a trading partner exhibits public behavior that is out of line with industry norms. Like any other social actor, firms often compare their performance to that of their peers and those with whom they have frequent contact, and modify their behavior to conform to the norm, or even to be perceived as a leader and an innovator. Thus effective participations in voluntary agreements may be encour- aged by other members of a firm's network directly and may be rewarded by a good reputation within the network. Conversely, free-riding may bring informal sanctions from trade associations and other firms.
326 UNDERSTANDING VOLUNTARY MEASURES Individuals Matter Decisions within firms are made by individuals who bring their personal beliefs, values, and norms to bear as well as the corporate interests. At least since Galbraith (1967), there has been debate about the degree to which manag- ers have autonomy from stockholders in steering the corporation. As noted in this volume, Chapter 1, many senior corporate managers are members of cohorts that have been exposed to environmental arguments since college or before. Nakamura et al. (2001) and Henriques and Sadorsky (1996) have found that the environmental concerns of corporate managers influence at least some aspects of firms' decisions. Protecting the Environment May Protect the Bottom Line Analyses from industrial ecology often argue that much environmental impact generated by the production process is a result of poor design and represents wasted resources (Socolow et al., 1994~. Better engineering could both reduce environmen- tal impact and save money. If these views are internalized by a firm, then voluntary agreements become a mechanism for receiving public approval for implementing reduced impact, finding flexibility within government regulations to innovate, and sharing information with peer firms, while the core motivation remains cost reduc- tion. Andrews (1994) suggests that adoption of industrial ecology strategies by firms can be greatly facilitated by voluntary agreements. We agree, but suggest that the reverse also may be true that the insights from industrial ecological analysis may be a major motivation to participate in voluntary agreements. Organizational Structure Matters Standard models from microeconomics begin with the assumption that the firm is a unitary entity that makes decisions to maximize profits or some other measure of economic performance. If this is the case, then explaining why firms would participate in voluntary agreements reduces to explaining how voluntary agreements might enhance economic performance. This is the essence of the rational actor model that underpins the analysis in several of the preceding chap- ters. But at least since Simon (1947), the assumption that organizations, includ- ing private-sector firms, can be seen as purely rational utility maximizers has been dubious. Two problems can be raised about the rational utility maximizer view of the firm. One is the assumption that organizations act to maximize profits. Other decision rules may be more plausible. The second, related problem is viewing the organization as a unified decision making entity, whatever the decision rule applied might be. i2 Who makes decisions is, of course, related to the problem of what criteria are used in making decisions. If Galbraith is correct in suggesting
THOMAS DIETZ 327 that in modern corporations managers have considerable independence from stockholders, then executives may be free to pursue goals they consider desirable, perhaps for ethical or career reasons, without strict adherence to profit maximi- zation. Nakamura et al. (2001) contrast models of profit maximization with those of utility maximization that take into account managers' preferences and concerns, and find that both are relevant to the environmental protection deci- sions of Japanese firms. DeCanio & Watkins (1998) found that patterns of stock ownership had some influence on decisions to participate in the U.S. Environ- mental Protection Agency' s (EPA' s) Green Lights energy conservation program. NEW TOOLS FOR OLD PROBLEMS As the literature reviewed in previous chapters makes clear, voluntary agree- ments are being used primarily to deal with environmental problems that previ- ously have been addressed with either command-and-control regulation or with market-based strategies. In the United States, experimental programs are at- tempting to reduce the emission of targeted pollutants and increase energy effi- ciency of the private sector. To date, there seems to be little experience with using voluntary agreements to deal with nonpoint pollutants or with the special problems of the service sector noted by Rejeski and Salzman (this volume, Chap- ter 2~. The exception seems to be with ISO 14001 and other "green" labeling and certification efforts that are encouraged by consumer demand for environ- mentally friendly products, demands from large purchasers on their suppliers, or demands of import regulations in important markets. Although targeted at man- ufacturers, some of these have the potential of reducing nonpoint sources when products used by consumers are a source of environmental problems. One of the central problems in dealing with nonpoint sources is that they are often composed of many tiny point sources. Households, small firms, or local governments may generate pollution almost inadvertently as a result of routine activities. Those carrying out these activities often have little awareness of the environmental consequences and little control over these impacts. The impacts may be embedded in the products or processes used, and the users may have little time, money, or technical capacity to develop for themselves or find in the market lower impact alternative products or practices. Thus policies targeting manufacturers and suppliers may be an appropriate strategy for dealing with nonpoint sources. Voluntary agreements may be an effective tool in changing the behavior of these corporate actors as an indirect way of addressing the non- point sources. ATTENDING TO THE GLOBAL Most work in this volume has focused on the U.S. experience. As noted at the beginning of Part Three, this is appropriate as a first step in developing an
328 UNDERSTANDING VOLUNTARY MEASURES understanding of education, information, and voluntary measures that is useful for U.S. decision makers. But a continued neglect of experiences outside the United States would unnecessarily and inappropriately limit understanding of New Tools, for two reasons. First, although there are many obstacles to generalizing across political and economic institutions and cultures, it is just such comparisons that clarify what is general and what is unique. Comparative analyses can lend substantial analyti- cal power to research on voluntary agreements. As many industrialized nations experiment with new forms of policy, it will become even more important to understand what approaches may transfer to the U.S. context, what approaches will not work, and why. In addition, as consumption and production increase in nations that still are only beginning to evolve their environmental policy sys- tems, comparative analyses are essential to provide useful guidance about the best ways to approach the challenges they will face. For example, approaches that are found effective in the U.S. context may not work well in a country with a different regulatory context or where a target industry has a different structure.~3 Second, we must recognize that environmental policy is now inherently international. This holds for at least two reasons. First, a large portion of the firms regulated in any one nation will have trade with and often have production facilities in other nations and so must deal with a web of national regulations. As a result, voluntary agreements nearly always will operate in the context of multiple national laws and regulations. Second, a growing fraction of national environmental laws and regulations are developed as a result of participation in international regulatory regimes.~4 Randall's point (this volume, Chapter 19) that firms may act in anticipation of regulation can apply to expected internation- al treaties as well as to uniquely national policies. So voluntary agreements are acting not just in the context of multiple national regulations and multiple na- tional markets for products and supplies, but also in the context of global envi- ronmental treaties and the national laws and regulations that respond to them. CONCLUSIONS Voluntary agreements may have substantial potential to reduce the environ- mental impacts of production and consumption activities both directly and by changing organizational culture and capabilities. But as the chapters in Part 3 demonstrate, voluntary agreements come in many forms, and we do not under- stand many of them very well. In the worst case, it is not clear what changes in environmental impact have flowed from the voluntary agreements that have been implemented, in part because of limits in the design of the existing evaluations and in part because we have few evaluations of the less tangible impacts of voluntary agreements such as organizational changes. The participation of firms in voluntary agreements, including the effective implementation of environmen- tal policies and procedures, depends on many factors that require further investi-
THOMAS DIETZ 329 gation. Some factors are intraorganizational: organizational structure and lines of control, culture and routines, available expertise, and available capital. Others depend on external factors: existing and anticipated regulations, the structure of the firm's industry, and the perceived interests and strategies of suppliers, cus- tomers, and competitors. To understand voluntary agreements, to evaluate their impacts, and especially to offer guidance on the design of future programs, sub- stantial efforts at conceptualization, modeling, and empirical analysis are essen- tial. The chapters here review the first efforts along these lines and point toward future research, but it is clear that the interest in voluntary agreements has run well ahead of our understanding of them. Even less well researched is the potential of voluntary agreements to deal with the sources of environmental impacts that flow from nonpoint sources. In addition, we are only beginning to think through the implications of international environmental regulatory regimes for national and international voluntary agree- ments. It is almost certain that voluntary agreements will be an important part of environmental policy over the next decade. What is less certain is the benefits that will flow from them and the best strategies for developing and implementing them. NOTES 1 Analyses of the international experience are beginning to appear (tenBrink, 2002). 2 One of the first efforts to apply program evaluation methods to an environmental policy is Poppitti and Dietz (1983). In examining the Resource Conservation and Recovery Act, we were surprised by how little evaluation research was available on environmental programs. This has been less true of energy and water conservation programs, where there is a long history of evaluation research using both experimental and quasi-experimental designs. See, for example, Dietz and Vine (1982), Kowalczyk et al. (1983), Harris and Blumstein (1984), Stern et al. (1986), and Vine and Crawley (1991). 3 The classic Campbell and Stanley (1963) volume documents the threats to valid causal infer- ence in various research designs. Cook and Campbell (1979) update this discussion, while Achen (1986) discusses statistical issues in the analysis of quasi-experiments. Issues of evaluation research design also are discussed in the ecological literature (Manly, 2001: Chapter 6; Underwood, 1997). 4 This suggests that an important goal of research should be developing an understanding of the effects of organizational change on the environmental performance of firms. 5 Examples of methods for modeling the "spread" of a distribution include those described by Gould (1992, 1997). "Extreme values" or the tails of a distribution are of interest in many applica- tions of statistics to environmental problems, where methods are steadily improving (Manly, 2001). 6 Axelrod (1997) considers the altruism problem the "Eshrecia coli" of the social sciences, while Dietz et al. (2002) refer to it as the "Drosophila melanogaster." Whichever analogy to biology is chosen, the argument remains the same: The problem of altruism, posed as Prisoner's Dilemma, the Tragedy of the Commons, or the Logic of Collective Action, presents a marvelous test bed for the theory and methods of the social sciences. 7 In addition to empirical studies of altruism, formal models of cultural evolution and of repeat- ed Prisoner's Dilemma games give analytical insights as to why cooperation may not be as rare as earlier, simpler models suggested (Boyd and Richerson, 1985; Richerson and Boyd, in press; Richer- son et al., 2002; Sober and Wilson, 1998).
330 UNDERSTANDING VOLUNTARY MEASURES 8 ISO 14001 does not regulate production or products per se, but simply requires that environ- mental management systems are in place. 9 Such prohibitions differ radically across nations. For example, in Japan the keiretsu corpo- rate groups are an important part of the national economic structure, and many firms have links through banks that hold large portions of their stock equity (Aoki, 1988). However, Nakamura et al. (2001) found that these links have little impact on firms' participation in the ISO 14001 certification process. 10 Laumann and Knoke (1987) and Dietz and Rycroft (1987) show that professional and trade associations are important participants in the energy and environmental risk policy networks. 11 In a series of papers, Wernick, Ausubel, and their collaborators document the reductions in environmental impact that have been achieved in a number of industries and suggest that better engineering could substantially alleviate environmental problems (Ausubel, 1996; Wernick, 1997; Wernick, 1994; Wernick et al., 2000). Mol and Sonnenfeld (2000) present arguments that this is a general trend, driven by societal concerns with the environment rather than just cost reduction. The literature on the environmental Kuznets curve makes a similar argument increased affluence leads to reduced environmental impact per unit affluence (see Nordstrom and Vaughan, 1999, for a review). York et al., (2001) and Roberts and Grimes (1996) present skeptical views of the trend toward reduced environmental impact. At best, such reductions may occur with regard to local impacts, but there is little evidence of such a dynamic for global impacts such as greenhouse gas emissions. Voluntary agreements have been directed at both local and global pollutants. 12 For example, organizational ecologists in sociology do not necessarily assume strict profit maximization, but do treat organizations as unified wholes with regard to their strategies (Hannan and Freeman, 1989). This view is critiqued in McLaughlin (1996). 13 Because of possible differences among industries in their likelihood of developing effective voluntary agreements such as those noted by Nash (this volume, Chapter 14) and Furger (this vol- ume, Chapter 17), it is important that comparative research be conducted both across countries and across industries. 14 These agreements are themselves voluntary agreements in that there is no mechanism to impose them on individual states. In addition to the importance of studying voluntary agreements in the context of international treaties on the environment, it seems likely that the literature on volun- tary agreements and that on international regulatory regimes might provide theoretical and method- ological guidance to each other. REFERENCES Achen, C.H. 1986 The Statistical Analysis of Quasi-Experiments. Berkeley: University of California. Andrews, C. 1994 Policies to encourage clean technology. In Industrial Ecology and Global Change, R. Socolow, C. Andrews, F. Berkhout, and V. Thomas, eds. Cambridge, Eng.: Cambridge University Press. Aoki, M. 1988 Information, Incentives and Bargaining in the Japanese Economy. New York: Cam- bridge University Press. Ausubel, J.H. 1996 Can technology spare the earth?American Scientist 84:166-178. Axelrod, R. 1997 The Complexity of Cooperation. Princeton, NJ: Princeton University Press. Boyd, R., and P.J. Richerson 1985 Culture and the Evolutionary Process. Chicago: University of Chicago Press.
THOMAS DIETZ Campbell, D.T. 1969 Reforms as experiments. American Psychologist 24:409-429. Campbell, D.T., and J.C. Stanley 331 1963 Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNally. Cook, T.D., and D.T. Campbell 1979 Quasi-Experimentation. Chicago: Rand McNally. DeCanio, S.J., and W.E. Watkins 1998 Investment in energy efficiency: Do the characteristics of firms matter? Review of Economics and Statistics 80:96-107. Dietz, T., E.Ostrom, N. Dolsak, and P.C. Stern 2002 The drama of the commons. Pp. 3-36 in National Research Council, The Drama of the Commons, Committee on the Human Dimensions of Global Change. E. Ostrom T. Dietz, N. Dolsak, P.C. Stern, S. Stonich, and E.U. Weber, eds. Washington, DC: Na- tional Academy Press. Dietz, T., and R.W. Rycroft 1987 The Risk Professionals. New York: Russell Sage Foundation. Dietz, T., and E. Vine 1982 Energy impacts of a municipal conservation policy. Energy 7:755-758. Galbraith, J.K. 1967 The New Industrial State. Boston: Houghton Mifflin. Gould, W.W. 1992 Quantile regression with bootstrapped standard errors. Stata Technical Bulletin Re- prints 2:137-139. 1997 Interquartile and simultaneous-quartile regression. Stata Technical Bulletin Reprints 7: 167-176. Gunderson, L.H., C.S. Holling, and S.S. Light, eds. 1995 Barriers and Bridges to the Renewal of Ecosystems and Institutions. New York: Co- lumbia University Press. Hannan, M.T., and J. Freeman 1989 Organizational Ecology. Cambridge, MA: Harvard University Press. Harris, J., and C. Blumstein, eds. 1984 What Works: Documenting Energy Conservation in Buildings. Washington, DC: Amer- ican Council for an Energy Efficient Economy. Heckman, J. 1976 The common structure of statistical models of truncation, sample selection, and limited dependent variables. The Annals of Economic and Social Measurement 5:475-492. 1979 Sample selection bias as a specification error. Econometrica 47:153-161. Henriques, I., and P. Sadorsky 1996 The determinants of an environmentally responsive firm: An empirical approach. Jour- nal of Environmental Economics and Management 30:381-395. Kowalczyk, D., J.C. Cramer, B. Hackett, P. Craig, T. Dietz, M. Levine, and E. Vine 1983 Evaluation of a community-based electricity load management program. Energy 8:253- 243. Laumann, E.O., and D. Knoke 1987 The Organization State: Social Choice in National Policy Domains. Madison: Univer- sity of Wisconsin Press. Manly, B.F.J. 2001 Statistics for Environmental Science and Management. Boca Raton, FL: Chapman and Hall. McLaughlin, P. 1996 Towards an Ecology of Social Action: Merging the Ecological and Constructivist Tra- ditions. Fairfax, VA: Human Ecology Research Group, George Mason University.
332 UNDERSTANDING VOLUNTARY MEASURES Mol. A.P., and D.A. Sonnenfeld, eds. 2000 Ecological Modernisation Around the World. London: Frank Cass Publishers. Nakamura, M., T. Takhashi, and I. Vertinsky 2001 Why Japanese firms choose to certify: A study of managerial responses to environmen- tal issues. Journal of Environmental Economics and Management 42: 23-52. Nordstrom, H., and S. Vaughan 1999 Trade and Environment. Geneva, Switz.: World Trade Organization. Olson, M. 1965 The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. Poppitti, J., and T. Dietz 1983 Social impact evaluation of the U.S. Resource Conservation and Recovery Act. Envi- ronmental Management 7:501-504. Rees, W.E., and M. Wackernagel 1994 Ecological footprints and appropriated carrying capacity: Measuring the natural capital requirements of the human capacity. Pp. 362-390 in Investing in Natural Capital: The Ecological Economics Approach to Sustainability, A.C. Jannson, M. Hammer, C. Folke, and R. Costanza, eds. Washington, DC: Island Press. Rees, W.R. 1992 Ecological footprints and appropriated carrying capacity: What urban economics leaves out. Environment and Urbanization 4: 121 - 130. Richerson, P.J., and R. Boyd in The evolution of subjective commitment to groups: A tribal social instincts hypothesis. press In The Evolution of Subjective Commitment, R.M. Neese, eds. New York: Russell Sage Foundation. Richerson, P.J., R. Boyd, and B. Paciotti 2002 An evolutionary theory of commons management. Pp. 403-442 in National Research Council, The Drama of the Commons, Committee on Human Dimensions of Global Change. E. Ostrom, T. Dietz, N. Dolsak, P.C. Stern, S. Stonich, and E.U. Weber, eds. Washington, DC: National Academy Press. Roberts, J.T., and P.E. Grimes 1996 Social Roots of Global Environmental Change: A World-Systems Analysis of Carbon Dioxide Emissions. New Orleans, LA: Department of Sociology, Tulane University. Simon, H.A. 1947 Administrative Behavior. New York: Macmillan. Sober, E., and D.S. Wilson 1998 Unto Others: The Evolution and Psychology of Unselfish Behavior. Cambridge, MA: Harvard University Press. Socolow, R., C. Andrews, F. Berkhout, and V. Thomas 1994 Industrial Ecology and Global Change. New York: Cambridge University Press. Stern, P.C., E. Aronson, J. Darley, D.H. Hill, E. Hirst, W. Kempton, and T.J. Wilbanks 1986 The effectiveness of incentives for residential energy conservation. Evaluation Review 10: 147- 176. Underwood, A.J. 1997 Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variance. Cambridge, Eng.: Cambridge University Press. Vine, E., and D. Crawley 1991 State of the Art Energy Efficiency: Future Directions. Washington, DC: American Coun- cil for an Energy Efficient Economy.
THOMAS DIETZ 333 Wackernagel, M., L. Onisto, P. Bello, A.C. Linares, I.S. Lopez Falfan, J.M. Garcia, A. Isabel, S. Guerrero, and M.G. Suarez Guerrero 1999 National natural capital accounting with the ecological footprint concept. Ecological Economics 29:375-390. Wackernagel, M., and W.O. Rees 1996 Our Ecological Footprint: Reducing the Human Impact on the Earth. Gabriola Island, British Columbia, Can.: New Society Publishers. Wernick, I.K. 1994 Demarterialization and secondary materials recovery: A long run perspective. Journal of the Minerals, Metals, and Materials Society 46(4):39-42. 1997 Consuming materials: The American way. Pp. 29-39 in National Research Council, Environmentally Significant Consumption: Research Directions, P.C. Stern, T. Dietz, V.W. Ruttan, R.H. Socolow, and J.L. Sweeney eds. Washington, DC: National Acade- my Press. Wernick, I.K., R. Herman, S. Govind, and J.H. Ausubel 1997 Materialization and dematerialization: Measures and trends. Pp. 135-156 in Technolog- ical Trajectories and the Human Environment, J.H. Ausubel, and H.D. Langford, eds. Washington, DC: National Academy Press. Wernick, I.K., P.E. Waggoner, and J.H. Ausubel 2000 The forester's lever: Industrial ecology and wood products. Journal of Forestry 98(10):8-14. York, R., E. Rosa, and T. Dietz 2001 Social Theories of Modernization and the Environment: An Empirical Analysis of the Human-Environment Relationship. Unpublished paper presented to the 2001 Sympo- sium on the Political Economy of the World System, Anaheim, CA, August 17. Young, O. 2001 The effectiveness of international environmental regimes: A midterm report. In Studies in the Human Dimensions of Global Change, A. Diekemann, T. Dietz, C. Jaeger, and E. Rosa, eds. Cambridge, MA: MIT Press.