Environmentally Significant Individual Behavior
Federal agencies should support a concerted research effort to better understand and inform environmentally significant decisions by individuals. Research in four specific areas could provide usable results in the relatively near term: (1) indicators of environmentally significant consumption, (2) information transmission systems, (3) integration of information with regulatory and market-based policy instruments, and (4) fundamental understanding of consumer choice and constraint. Research in the first and the last of these areas is likely to have the most important and lasting impact.
THE RESEARCH NEED
The activities of individuals have major environmental consequences in the aggregate. Consequently, there can be major environmental effects from change in the behavior of individuals and households. The often-cited estimate that consumer expenditures account for two-thirds of gross domestic product suggests the environmental importance of consumer power, although it is likely to be an overestimate of the direct environmental effect of consumer behavior. Households directly account for slightly under half of U.S. carbon emissions and smaller percentages of some other important effluents (e.g., Stern and Gardner, 1981; Cutter et al., 2002). For several pollutants that have been greatly reduced by effective regulation of the industrial sector, individuals and households have become major sources of the remaining emissions (Vandenburgh, 2004). Dioxins and dioxin-like compounds are a prime example. Between 1987 and 2002, total emissions
of these compounds in the United States were reduced by 92 percent, with the result that backyard barrel burning, which was the source of 4 percent of emissions in 1987, had become the source of almost 60 percent of national emissions 15 years later (Institute of Medicine, 2003).
Individual behaviors have significant direct impact in the aggregate in the areas of transportation, housing, energy-using appliances, solid waste, water, and food. Individuals also influence environmental quality indirectly, in their roles as citizens, investors, and members of organizations that make environmentally important choices. And to an important degree, small businesses and nonprofit organizations have impacts (and make decisions) much like individuals and households. However, these latter areas are not central to the scope of the panel’s consideration of individual environmental choices.
The environmental impact of an aggregation of individual consumer choices is circumscribed because, in many cases, the links from individual behavior to its environmental consequences are indirect and conditioned by a variety of forces and constraints in complex social, economic, institutional, and technological systems (Shove, Lutzenhiser, Guy, Hackett, and Wilhite, 1998; Lutzenhiser, Harris, and Olsen, 2001). For example, individuals who want to make green choices may find options limited and costs prohibitive because of a lack of the relevant products or infrastructure, as is generally the case in the United States with alternatives to petroleum-burning private motor vehicles for personal transportation. When green choices are more readily available, the potential for environmental improvement may be limited because of the difficulty or cost of acquiring trustworthy and timely information about the environmental consequences of decisions. This continues to be the case, even though an increasing number of products and services are marketed with claims that they are environmentally benign or beneficial.
The links from policy to individual behavior are also weaker than sometimes supposed. For example, governments sometimes provide information in their efforts to promote greener individual behavior—increased recycling, more careful use of household chemicals, purchase of energy-efficient appliances, testing for radon in homes, reduced use of motor vehicles during air pollution crises, and so forth. They presume that, with better information, people will act in more environmentally beneficial ways. But the record of environmental information programs is unimpressive (Hirst, Berry, and Soderstrom, 1981; National Research Council, 1984, 2002a). Information can be more effective, however. Research has identified some of the ways in which information, usually in interaction with a variety of other factors, influences environmentally significant choices (e.g., National Research Council, 1984, 1997a, 2002a). It shows that environmental information can be effective in influencing behavior if it is delivered
in appropriate and timely ways. The ways in which information is constructed, conveyed, cognitively processed, and weighed (in interaction with other factors, such as individuals’ values, attitudes, and affective processes; social norms; economic resources and incentives; and technological availability) determine whether, when, and how information affects action.
The actual environmental impact of effectively delivered information on individuals’ choices is not frequently measured, but it is in many cases likely to be modest because of the other factors affecting these choices. Improving information for individual choice is nevertheless important. In some arenas, informing citizens’ decisions is a basic duty of government. Also, environmental policies and programs that rely on regulatory or incentive strategies can fall far short of their potential if their information components are not implemented effectively. For example, identical incentive programs operated by different energy utility companies differed by more than a factor of 10 in their effects as a function of how the programs were marketed (Stern et al., 1986). We give special attention to information in this science priority because this policy strategy has significant untapped potential, because its perceived appropriateness does not fluctuate sharply with events and political shifts, and because, in some policy arenas, improving information is among the few policy strategies available.
AREAS OF RESEARCH
Despite a thin stream of research over several decades (e.g., National Research Council, 1984, 1997b; Gardner and Stern, 2002), we are only beginning understand the ways individuals’ values and preferences, social influences, available choices and constraints, and other factors combine to shape environmentally significant individual behavior. Developing this fundamental understanding and improving measures of the environmental effects of consumer behavior are critical needs for anticipating the aggregate environmental effects of individual behavior and for informing policies affecting this behavior. We have identified four specific research areas as worthy of particular attention.
Indicators of Environmentally Significant Consumption
Research to develop trustworthy indicators of the environmental impacts of individual and household behavior and choice would help people make their environmentally significant choices conform better to their values and preferences. Many people would like to reduce the environmental impact of their personal choices. Evidence of this desire comes from survey research since the 1970s that, despite periodic fluctuations in the numbers, has recorded very high rates of public support for environmental protection
consistently since the 1970s (Dunlap and Scarce, 1991; Dunlap, 2002). For example, 44 percent of U.S. respondents surveyed in 2001 said they thought environmental regulations had not gone far enough, compared with 21 percent who said they had gone too far. The comparable figures were 34 and 13 percent in 1973, when the question was first asked. On the issue of whether government is spending too much, too little, or the right amount on environmental protection, respondents favored “too little” over “too much” by 61 to 7 percent in 1973 and 62 to 7 percent in 2000 (Dunlap, 2002). Evidence of individuals’ support for environmental protection also comes from the growth of support for and participation in recycling programs and tangible support for, and widespread use of, parks and nature reserves, community gardens, conservation organizations, farmers’ markets, “smart growth” policies, environmentally sensitive products and foods, and other green goods and services. Indirect evidence comes from the prevalence of green appeals in advertising, although it is also the case that advertisers often highlight the nonenvironmental attributes of products and services that have large environmental “footprints.”
Against this evidence of strong environmental concern in the United States is apparently contradictory evidence from consumer behavior, such as the continuing growth in sales of fuel-consuming sport utility vehicles, in the size of new homes, and in other trends that increase the environmental impact of Americans’ behavior per capita and per household. This apparent paradox of green attitudes contradicted by behavior can be illuminated by better measurement and presents important questions in its own right, as noted later in this chapter.
Although many of the consequences of individuals’ choices are easy for them to discern, it is very difficult for individuals who want to consider the environmental impacts of their choices to estimate those impacts. Measures of economic consumption do not adequately capture them (National Research Council, 1997b), and the readily available direct measures have proved less than adequate. For example, utility bills report aggregate energy consumption but do not link it to particular appliances or behaviors. It is therefore not surprising that many consumers hold mistaken beliefs about matters as straightforward as the relative energy consumption of household appliances (e.g., Kempton, Harris, Keith, and Weihl, 1985). Thus, it is important to distinguish, both conceptually and methodologically, between consumers’ intent with regard to environmentally significant behavior and the environmental impact of the behavior. Energy-use indicators are provided on labels on major household appliances in the United States, but they do not communicate well with consumers (duPont, 2000; Egan, Payne, and Thorne, 2000; Shorey and Eckman, 2000). Green labels for other consumer products have not developed much in the United States, but there is some evidence from Europe that they can influence individual behavior
and perhaps also the behavior of firms (Thøgerson, 2002). A striking example is the eco-labeling of household chemicals in Sweden beginning in the late 1980s, which was followed within a decade by a 15 percent decrease in national sales of cleaning and personal care chemicals and in the replacement of 60 percent of the chemicals in soap, shampoo, detergents, and cleaners by less harmful substances. The success of such labeling programs depends, among other things, on their accuracy and on the existence of an effective information delivery system (see below).
Green-minded consumers need information not only to compare different brands or models of the same product, but also to compare different behaviors (e.g., commuting by car versus bus), to identify environmentally superior products (e.g., paper versus plastic bags), and to determine which environmentally motivated behaviors would have the greatest beneficial environmental effect. Conceptually, this information would amount to scaling down indicators such as the ecological footprint (Rees and Wackernagel, 1995) or the Toxics Release Inventory to the individual level.
Trustworthy information on the environmental impacts of consumer choices depends on the availability and continual improvement of appropriate impact measures, such as direct energy consumption, embedded energy, material flow measures, and life-cycle environmental impact of both products and activities. The recommended research would make such measures behaviorally relevant by linking them to important consumer choices. Such indicators are not now readily available, although increasingly there are efforts to develop them, building on the work of researchers and environmental groups (e.g., Uusitalo, 1986; Durning, 1992; Vringer and Blok, 1995; Lutzenhiser, 1997; Brower and Leon, 1999; Redefining Progress, 2004). The need is to develop, validate, and gain widespread use of sound, behaviorally relevant measures.
Careful accounting studies of materials and energy flows that combine physical science expertise and an understanding of human behavior patterns can provide useful information about the relative impacts of common behaviors and choices, as well as significant alternatives (driving vehicles with various attributes, using private vehicles versus public transit, adoption of different home heating and cooling technologies and retrofits, eating different sorts of foods, eating at home versus in restaurants, using different sorts of appliances, buying durable versus disposable consumer products of various kinds, buying bulk versus prepackaged products, recycling, conserving water, etc.). It may be possible to aggregate the environmental indicators for various consumer products and services into indicators for the companies that provide these goods and services in ways that could usefully inform individuals’ investment decisions.
A good accounting of the environmental impacts of behaviors, products, and so forth can better inform individuals’ environmentally significant
choices. It is likely to reveal that certain practices that are widely believed to be pro-environmental may turn out to have fewer benefits than other, less obvious, choices that may be much more environmentally significant. As a general rule, for example, the environmental impact of purchases of consumer durables such as motor vehicles and major appliances that lock in environmental consequences for long periods is considerably larger than that of variation in how these items are used (Gardner and Stern, 2002). Efforts to develop indicators that are applicable internationally will also be useful to researchers and government agencies for developing scenarios or projections of future human demand on environmental resources.
Information Transmission Systems
Research on the characteristics and dynamics of environmental information transmission will allow government agencies and others to develop more effective ways to inform and, given appropriate policy decisions, to influence individuals’ environmental choices. Government agencies and firms often rely on consumer information in the form of public awareness campaigns, public service announcements, informational flyers, labeling and rating systems, and so forth, to inform and influence environmentally significant behavior. It is sometimes assumed that lack of information is the main barrier to action and that informing or educating consumers will automatically lead to desired choices and behaviors. However, environmental programs that rely on disseminating information about behaviors and how to perform them have been notoriously ineffective on average for promoting the desired behaviors (Ester and Winett, 1982; Geller, Winett, and Everett, 1982; Hirst, Berry, and Soderstrom, 1982; National Research Council, 1984; Gardner and Stern, 2002). The results have been disappointing in part because information is often not the only significant barrier to action (Lutzenhiser, 2002; Gardner and Stern, 2002). But information has not achieved its potential in part because it has not been disseminated effectively.
More than two decades of research have identified a number of aspects of informational messages that can often be made better. For example, information is more effective when it is understandable to audiences, presented in a way that attracts attention and stays in memory, and delivered at a time and place close to the relevant choices (National Research Council, 1984; McKenzie-Mohr and Smith, 2000; Valente and Schuster, 2002). Important questions for message design include: How can information be made decision-relevant? and How should messages be designed to suit audiences’ attention patterns, cultural understandings, and cognitive capacities?
Information programs can benefit by sponsoring applied research that
addresses such questions in the context of their particular target decisions and audiences, and organizations that operate such programs should sponsor this kind of research. However, a general federally supported research program on message design is not needed.
Greater value can be gained from new research on the design and functioning of information communication and transmission systems. This research would take into account available knowledge about how information use is shaped by characteristics of information sources and their interactions with information producers and potential users. The significant characteristics include trust in information sources (National Research Council, 1984); the presence and behavior of intermediaries between information and its audiences (National Research Council, 1984); the use multiple information sources, including direct personal interactions (e.g., Werner and Adams, 2001; Mileti and Peek, 2002); the multiplicity of audiences that require different kinds of information and that trust different sources (Gardner and Stern, 2002); and the presence in most information environments of multiple, sometimes conflicting, sources and messages that purport to offer useful information (National Research Council, 1989, 1996).
The research would seek to develop and evaluate interventions in information transmission systems that take the above insights into account in order to get accurate and trustworthy information to a variety of audiences from sources they trust. Studies might examine environmental monitoring systems, production standards programs, certification schemes, and so forth to analyze their information transmission systems. They might consider how such systems and programs establish the trustworthiness of information about the environmental characteristics of goods and services (agricultural production practices used, resource depletion involved, effluents and their environmental impacts, habitat transformation, etc.), validate the information, and convey it to audience groups. Researchers might examine the roles of intermediary groups that have tried to provide information in forms that suit audience needs, to make it more readily available, or to increase its credibility to target audiences. An example is the creation of a web site by Environmental Defense to make information from the federal government’s Toxics Release Inventory more readily interpretable by nonexperts (Herb et al., 2002). It is important to better understand the strengths and weaknesses of those approaches, how they are being used, and how their design and delivery can be improved.
It is also important to investigate more systematically other aspects of information transmission systems. One of these is the role of commercial advertising claims in environmentally significant consumer purchase decisions, both directly and in relation to public-sector information transmission. Another is the “reverse flow” of information—from consumers to
firms, governments, and intermediaries regarding the kinds of information they need for their decisions, as well as about their desires in terms of products, environmental performance, and public policy. Yet another is the potential to design informational efforts for use when events make environmentally significant attributes of consumer behavior especially salient, as, for example, during the large spike in gasoline process in the spring of 2004.
The recommended research could fruitfully address general questions such as these:
How effective are existing environmental information delivery systems?
What are their limitations as primary sources of information?
What information is most effectively transmitted by what sorts of intermediaries?
How can information delivery systems (messages, intermediary networks, systems that produce indicators, etc.) be designed to produce and transmit information better suited to their consumers?
How do public-sector information and private-sector marketing combine to shape individuals’ awareness and understanding of the environmental implications of their choices?
Such questions might well be explored in the context of particular complex information transmission systems, such as those that lead to the labeling of produce as organic or as grown in accordance with ecological principles, or those that create green investment funds. The goal would be improved insight into the most appropriate roles of public, private, and nonprofit sector organizations in providing trustworthy information to a diversity of individuals, for example, as part of systems of environmental monitoring, certification, and the like.
Integrating Information with Other Policy Instruments
Research on the combined performance of information and other policy tools can significantly increase the effectiveness of all of these tools. It has long been recognized that information has limited value when there are weak incentives for its recipients to use it; also long recognized but not as universally acknowledged is the insight that incentives can fall far short of their practical potential if they are not accompanied by appropriate information about how best to take advantage of them. Several studies in the areas of consumer policy for energy conservation support this general insight (e.g., Stern et al., 1986; Brown, 2001; Brown, Levine, and Short, 2001). Environmental policy could benefit from improved knowledge
of how information can be effectively combined with other policy instruments.
A number of governments and firms actively promote pro-environmental policies, products, and services—ranging from recycling programs and storm water management initiatives to energy efficiency technologies and alternative transportation modes. Policy strategies that have been developed to accomplish these goals include incentives, rebates, tax benefits, cost sharing, prohibitions, regulations, codes, standards, and the provision of infrastructure. The information component to such policies and programs involves making people aware of program opportunities and requirements, providing general guidance and detailed technical assistance on adoption or compliance and showing how particular patterns of voluntary choice will contribute to desired personal, social, and environmental outcomes. Information has been instrumental in changing individual behavior when coupled with public and market incentives, particularly in times of crisis (for an example in an energy shortage, see Lutzenhiser, Kunkle, Woods, and Lutzenhiser, 2003) and in a few intensive policy initiatives, such as the Hood River Project in Oregon, which attempted to bring all the homes in a community up to high standards of energy efficiency without regulation (Hirst, 1987). And as already mentioned, nonincentive factors associated with information transmission can produce a tenfold difference in the effectiveness of some incentive programs (Stern et al., 1986). Generic knowledge about ways to combine information with other policy instruments comes from studies in environmental and nonenvironmental policy arenas, the latter including public health (Valente and Schuster, 2002) and disaster preparedness (Mileti and Peek, 2002).
Useful applied research can be done in a variety of environmental policy contexts in which information and other influences are often combined. These include programs to encourage recycling (U.S. Environmental Protection Agency, 1994), energy-efficiency investments (e.g., Stern et al., 1986; Brown, 2001; Brown et al., 2001), changing behaviors involving household use of toxic chemicals (e.g., Werner and Adams, 2001), and transportation alternatives to the private motor vehicle (e.g., Brown, Werner, and Kim, 2003; Katzev, 2003). Research questions focus on how information can be supplied in complex policy contexts that allow persons to assess the social significance of their individual actions, effectively consider their policy and market options, alter their choices in concert with the actions of other individuals and organizational actors, and participate effectively in both markets and public-sector decision-making processes.
This applied research often involves the development and evaluation of policy innovations; policies are typically designed by focusing on a single tool, such as regulation or financial incentives, whereas these policies involve the coordination of tools. This research therefore presents two unusual
challenges. One is methodological, involving the evaluation of an intervention that depends on the combined effect of several policy elements. The other is a challenge of dissemination that arises because policy innovations cannot necessarily be transferred in their entirety from one setting to another but may need to be adapted. Researchers and research users must find ways to make adaptations that will work in new situations. Such acts of diffusion of policy innovation can be an important object of research in their own right.
Fundamental Understanding of Consumer Choice and Constraint
A basic understanding of how information, incentives, and various kinds of constraints and opportunities, in combination with individuals’ values, beliefs, and social contexts, shape consumer choice in complex real-world contexts would provide an essential knowledge base for understanding, anticipating, and developing policies for affecting environmentally significant consumer behavior. As already noted, environmental effects of individual and household consumption decisions are limited because consumer choice itself is seriously bounded, situated, and constrained by properties of the physical infrastructure, the range of options available in markets, legal and policy strictures, economic and information costs of behavior change, disposable income, household dynamics, and other factors. It is also influenced by commercial advertising, social comparison processes, and a variety of other social and economic forces.
Which of these contextual factors is most important, and the importance of contextual factors in general relative to such personal factors as values, attitudes, beliefs, skills, and information, depends on the type of behavior, its context, and characteristics of the sample of people studied (e.g., Black, Stern, and Elworth, 1985; Guagnano, Stern, and Dietz, 1995; Tanner, 1999; Gatersleben, Steg, and Vlek, 2002; Brown et al., 2003; Tanner, Kaiser, and Kast, 2004). The effects of information or other policy instruments on individual behavior are also dependent on a variety of factors. Thus, for example, to use information effectively, either alone or in combination with other policy instruments, it is important to understand what the possibilities for change are for target populations and what useful roles information of particular kinds can play. To know this, a more thorough knowledge is needed of the nature of real-world choice in messy cultural contexts.
A small but growing number of empirical studies are beginning to illuminate the determinants of several kinds of environmentally significant consumer choices. Some of them have used aggregate scales of pro-environmental behavior (e.g., Kaiser, 1998; Nordlund and Garvill, 2002; Cottrell, 2003). Others focus on particular types of consumer behaviors,
including household energy use (e.g., Black et al., 1985; Poortinga, Steg, and Vlek, 2004), travel behavior (e.g., Bamberg and Schmidt, 2003; Katzev, 2003; Brown et al., 2003), food purchases (e.g., Tanner, Kaiser, and Kast, 2004), and recycling (e.g., Guagnano et al., 1995; Li, 2003; Oom do Valle, Reis, Manazes, and Rebelo, 2004). Some researchers have compared the determinants of different classes of pro-environmental behavior or have examined similar behaviors in different settings (e.g., different countries, where the contexts for such behaviors as recycling and mass transit use are dramatically different) (see, e.g., Stern, Dietz, Abel, Guagnano, and Kalof, 1999; Aoyagi-Usui, Vinken, and Kuribayashi, 2003; Thøgerson, 2004).
The findings so far suggest a few general principles, such as that the relative importance of contextual factors vis-à-vis personal ones is positively correlated with the environmental impact of the behavior (Black et al., 1985) and that the effect of personal variables, including information, is likely to be greatest when contextual pressures for or against a behavior are weak (Guagnano et al., 1995). Some studies suggest that behaviors that are grouped according to certain dimensions of similarity may have similar determinants (e.g., Stern et al., 1999; Thøgerson, 2004). But such generalizations have yet to be adequately tested across behaviors and settings. Additional research that looks systematically across behaviors, settings, and populations could considerably improve understanding of where the possibilities lie for information or other interventions to influence environmentally significant behavior.
Research also highlights the importance of at least four major classes of influences, attitudinal factors, contextual forces, personal capabilities, and habits (Stern, 2000). Researchers have been exploring finer distinctions within most of these broad categories, with the greatest amount of research attention being paid to the attitudinal class of variables, which includes basic values, environmental attitudes, identification with nature, beliefs about the environmental consequences of behavior, and other factors. Increasingly, researchers are developing coherent theoretical accounts of environmentally significant individual behavior that consider the roles and interactions of various kinds of influences and the possibility that different behaviors may respond to different collections of influences (e.g., Ölander and Thøgerson, 1995; Dahlstrand and Biel, 1997; Stern, 2000; Vlek, 2000; Gardner and Stern, 2002; Bamberg and Schmidt, 2003; Schultz, Shriver, Tabanico, and Khazian, 2004; Thøgerson, 2004).
An important research area is the apparent paradox already mentioned of strong environmental concern expressed in surveys of U.S. public samples combined with trends toward increasingly high environmental impact consumer choices, such as the purchase of larger homes and motor vehicles. Do these findings mean that some people’s expressed environmental concern is only skin deep? That people do not understand the
environmental implications of their purchases? That advertising has drawn consumers’ attention away from the environmental attributes of products? That nonenvironmental product attributes are more important to consumers than environmental ones in their major environmentally significant purchases? That public policy and private business routinely deliver less environmental benefit than people want, in some sort of market failure and democracy failure? That consumer behavior and citizenship behavior are driven by fundamentally different psychological processes? Unraveling the paradox would provide valuable information to public- and private-sector decision makers concerned with the environmental impact of consumer choice.
It will be important for research effort on such questions as the above to build bridges across concepts and insights currently segregated by social science disciplinary boundaries. The dynamics of choice have been extensively studied by psychologists, anthropologists, sociologists, organizational analysts, and, more recently, behavioral economists (Smelser and Swedberg, 1995; Wilk, 1996; Bagozzi, Gurhan-Canli, and Priester, 2002; Camerer, Loewenstein, and Rabin, 2003). Their respective literatures are extensive, yet researchers in each area are, for the most part, uninformed about salient work in the others. The disciplinary literatures are also rendered somewhat alien to one another by differences in interests, paradigms, data, methods, and analytic vocabularies. However, these impediments to synthesis around a unified model of real-world situated choice could be overcome if significant funding were targeted to interdisciplinary inquiries in which just such a synthesis was the required outcome.
We stress the fact that the needed work is inherently interdisciplinary, which is also to say that work from any single perspective, even when aware of findings from cognate areas, will tend to fall victim to biases or blind spots that may result in misleading characterizations. The interdisciplinary work needed would best be done through the exploration of specific choices in particular settings, with an aim of developing policy-relevant knowledge on a number of questions: Which behaviors are most changeable? Which are most changeable directly by individuals? Which are most constrained by the choices of organizations or by societal infrastructure? How does this vary across social groups and technical circumstances? The applicability of research in specific settings to other settings can begin to be tested as bodies of knowledge develop in different settings, allowing a comparative analytic approach.
It will also be important to test generalizations developed in research in North America and Western Europe on consumers in other countries, particularly in developing countries where the most environmentally significant behaviors and the most significant incentives and constraints related to them are quite different from their counterparts in the high-income coun-
tries. Very little is known about environmental consumption in those countries at the individual and household levels. Understanding the forces affecting environmentally significant behavior in those contexts will be critical for projecting the environmental implications of consumer behavior in these countries overall and for designing policies for promoting green choices in those contexts.
Also little researched is the role of information in individuals’ behaviors as citizens or as investors. Although it seems likely that the factors that shape individual environmentally significant choices differ when someone acts in different roles (e.g., consumer, citizen, or investor), little is known about this. Neither has there been much research on the relationships between individuals’ consumer purchase or “demand” choices (e.g., paper versus plastic) and their actions and views regarding public policy. For example, the energy crisis in California in 2000-2001 showed that persons were willing to do their part by conserving electricity, but also that they expected significant visible conservation from businesses and aggressive government response to perceived gouging by energy suppliers (Lutzenhiser et al., 2003).
RATIONALE FOR THE SCIENCE PRIORITY
Likelihood of Scientific Advances
In all four areas identified under this science priority, past research has led to concepts and insights that open the way for further research. The opportunities lie both in fundamental research on consumer choice and in applied research focused on specific problems of measurement or of policy design and evaluation. The recommended research is likely to lead to understanding of broader applicability to issues of policy analysis and to fundamental understanding of individual behavior under complex real-world conditions. The availability of validated measures of environmental consumption would open a major avenue of research on the determinants of change in consumption.
It is worth noting that the geographic center of research on environmentally significant individual behavior, which was in the United States in the 1970s and 1980s when government interest in energy conservation provided research funding, now seems to lie in Western Europe. Even U.S.-based journals that publish in this area are increasingly dominated by non-U.S. researchers. The most likely explanation of this shift is the scarcity of research funds from U.S. sources for this research. An infusion of research funds has a high likelihood of turning the continuing interest in environmental topics among senior and young social scientists into significant contributions to policy-relevant knowledge for U.S. decision makers.
Research on information transmission systems and on developing and assessing mixed policy instruments would have obvious value for policy makers who want to influence individual behavior with information or other policy instruments. Improved basic understanding and measurement of environmental consumption would have significant value for environmental forecasting and modeling by improving on such common proxy measures as income, which do not account for behavioral differences among people with similar resources. It could also aid in anticipating the likely environmental effects of increased consumer spending in developing countries. Fundamental research also has longer term practical value by building a stronger base for applied research and policy-related understanding.
Likelihood of Use
In each of the four research areas identified, a variety of potential users of the results of the recommended research are likely to put the findings to use in policy and program development. It is important to note that state and local governments, which are often more dependent on voluntary citizen or consumer action by itself or in conjunction with regulation, may find this work particularly useful when applied to local circumstances and resource constraints.
Trustworthy indicators of consumption would be used by individuals as consumers and investors; government agencies at various levels that have environmental education mandates or that need to forecast consumer demand on environmental resources; producers and vendors of green products and services; and nongovernmental (e.g., environmental and consumer) organizations, green investment advisers, and firms concerned with environmental accounts. The extent of use by consumers will depend on how the information in indicators is delivered and on concurrent policy and market incentives and constraints on choice. Credibility is also likely to be an issue. Cooperation among researchers, environmental agencies, intermediaries, and likely users is desirable for making indicators useful and credible.
Improved understanding of information transmission systems would be useful for government agencies that require or provide environmental information, particularly to achieve environmental goals that may be difficult to achieve by regulation or economic policy instruments; firms that supply environmental information and have reputation concerns; and trade associations, investment advisers, and nongovernmental organizations that might act as intermediaries or monitors in complex information transmission systems. Organizations concerned with designing effective environ-
mental certification systems or credible green investing instruments would be quite likely to apply findings from such research. Government agencies might also apply these findings when designing information production and delivery systems and when evaluating the likely success of privately developed information programs (such as those presented as alternatives to regulation). Use of this research will depend in part on the degree of pressing need for credible information transmission systems. Strong pressure for effectiveness in information-based policy will also increase the likelihood of use. If credibility scandals arise with existing information systems, such as those based on information voluntarily provided by polluters, many people will wish that research on information transmission systems had been conducted sooner.
Improved understanding of how information can be integrated with other policy instruments would allow governments to apply this knowledge to design more effective policies for behavior change that combine public information, market incentives, infrastructure investments, and other interventions. In California, for example, serious consideration is now being given to promoting large-scale energy usage changes through a combination of public interest advertising, consumer incentives, research and development, and codes and regulations. A number of cities are involved in similarly ambitious efforts—e.g., addressing a variety of air quality, water quality, traffic congestion, solid waste, wastewater, and infrastructural capacity issues through combinations of prohibitions, inducements, public information, and voluntary civic action. They would use the results of this research if some of it is conducted in the context of their programs and policies and if research results are actively disseminated through relevant intermediaries, such as associations of state and local government officials.
Improved fundamental understanding of the dynamics of choice and constraint would be useful to environmental forecasting enterprises and the designers of environmental policy instruments: firms and governments concerned with strategic targeting of messages, incentives, and products. It would provide policy makers with background information regarding the relative potential for informational and other policy instruments both alone and in combination; the limits of informational interventions and the places where barriers would need to be removed if information is to be effective; the kinds of information that should be targeted to different types of audiences (e.g., individuals acting for themselves, intermediaries, people acting on behalf of the organizations). In combination with the development of environmental consumption indicators, this research would help in setting priorities so that attention is given first to behaviors that are both changeable and worth changing. Knowledge of the fundamental dynamics of choice would also have applications in nonenvironmental areas concerned with the effects of choice on economies, polities, health and welfare systems, etc.
The use of the results of the recommended research is likely to depend significantly on the quality of communication between researchers and the potential users of their findings. Particularly because much of the recommended research would be locally based, communications networks will be important for promoting the diffusion of insights from such research to other places where they might be applied. Government agencies concerned with the widespread application of the recommended research should therefore also consider providing support and encouragement to efforts to establish ongoing communication between the producers and potential users of research pursuing this science priority. Such efforts might be organized by existing organizations, such as consumer groups, trade associations, or associations of local governments, which might create networks of researchers and practitioners, invite researchers to meetings of possible research users, or develop other innovative communication mechanisms.