Beyond Technology: Strengthening Energy Policy through Social Science

Chapter 2: Workshop Summary

Back to table of contents
The Alternative Energy Future

Leslie Berlowitz, President of the American Academy, opened the workshop with the observation that interdisciplinary study is critically needed in the areas of energy production and use, saying, “This project is trying to apply social science expertise to better understand how public attitudes, economic trends, and government regulations affect the development and adoption of clean energy.” Steven Knapp, a Fellow of the American Academy and President of The George Washington University, welcomed participants, and the program featured keynote speeches by Steven Koonin, Under Secretary for Science at DOE, and Myron Gutmann, Assistant Director for Social, Behavioral, and Economic Sciences at NSF.

Koonin spoke on the realities and challenges of the energy system, laying out the administration’s clean energy goals, which include reductions in oil imports and greenhouse gas emissions and increases in energy efficiency and electric vehicles. Noting that “the challenges of policy and human behavior have become even more critical,” he urged participants to look at how social science research can further the transition to a clean energy future, citing specific areas that need more study, including incentives, discount rates, energy awareness, and the acceptance and adoption of new technologies. Regarding discount rates, he said, “Consumers have a two- or three-year discount rate in their head when they buy automobiles. How do we get people to think more long term and understand the difference between capital and operating expenses?” On energy awareness, he noted, “One of the great triumphs of modern society is that we’ve hidden the infrastructure. Nobody really understands where electricity, gas, or water come from.” Because 95 percent of this energy system is constructed, owned, and operated by the private sector, Koonin said, “nothing is going to happen in energy of any consequence unless the private sector is engaged.”

Gutmann explained that while NSF is investing significantly and thinking seriously about energy issues, “It’s very much up to people like you in the research and broader scientific community to help us define the research problems and to help us make clear what are the best strategies to getting the critical answers in this area. What is clear is that every approach we take will cross traditional disciplinary boundaries.” Citing the need for more research, he said, “We don’t want to just know whether people will adopt a new technology; we need to understand fundamental questions; for example, about how markets work. This is what engages the economic community and the decision community, and we’re not going to advance the science unless we do that.”

Gutmann emphasized that “the critical questions are fundamental questions about behavior: how are people thinking about and reacting to new energy sources? So instead of figuring out where to put the outlets for plug-in hybrid cars, we should be theorizing about where to look for the next innovation behaviorally beyond the plug-in, or how to think about optimizing commuting and residential patterns to enhance conservation, but also to enhance how the technology community thinks about their innovations.”

Nicholas M. Donofrio, a Fellow of the American Academy, Senior Fellow at the Ewing and Marion Kauffman Foundation, and former Executive Vice President of Innovation and Technology at IBM, tied the workshop presentations to the need for both technological and policy innovation, saying that at IBM innovation is “all about creating value by understanding the problem.” He advised participants to strive to understand the problem and then “apply your technology, your knowledge, your invention, your creation, and your discovery in a unique and facile way to unlock that hidden value.” Donofrio described innovation cultures as being collaborative, open, multidisciplinary, and global.

Workshop attendees brought a diverse range of experience and expertise to the meeting. The workshop included representatives from DOE, NSF, and the National Renewable Energy Laboratory, as well as senior staff from public utility commissions, universities, industry, and nongovernmental organizations. A series of six panel discussions focused on institutional and consumer behavior, policy analysis, and energy regulation.

Session A: Behavior and decision making related to energy efficiency

Thomas Dietz, Professor of Sociology and Environmental Science and Policy; and Assistant Vice President for Environmental Research, Michigan State University

Paul C. Stern, Study Director, National Research Council
Charlie Wilson, Lecturer, Tyndall Centre for Climate Change Research
Marsha L. Walton, Senior Project Manager, New York State Energy Research and Development Authority

Although the first panel discussion centered on lessons from the behavioral sciences related to energy efficiency, participants noted at the outset that much of the behavioral research on energy efficiency also applies to other energy technologies. The panel focused on three agents that can contribute to greater energy efficiency in residential buildings: households, regulators, and third-party intermediaries.

Households account for 30–40 percent of the energy use in the United States, which points to the potential benefits of greater energy efficiency in the household sector. By focusing only on available no-or-low-cost behavioral interventions that do not require a major lifestyle change, such as weatherizing houses or properly maintaining vehicles and HVAC equipment, energy consumption in the residential sector U.S. energy consumption.2 However, achieving these savings will require combining behavioral interventions with policies aimed at facilitating their adoption. Such policies are not common in the United States but could be implemented at either the state or the national level.

The reduction estimates associated with available no-or-low-cost interventions exclude the savings that could be achieved through federally mandated high-efficiency lighting technologies, and in fact an equal reduction in energy use is achievable through more-efficient household technologies. Up-front cost is a major barrier to the adoption of more-efficient technologies, and this consideration often outweighs the potential for long-term savings. Other barriers to household actions include existing regulations, infrastructure issues, limited consumer choice, and a lack of information about energy savings.

The most productive strategy will be to identify and promote the behaviors and technologies that can have the greatest impact on energy consumption and simultaneously to address the many barriers to these choices through major outreach campaigns. The failure to address even a single barrier can block progress. Energy policies often address only individual barriers and thus do not gain much headway at the consumer level. Similarly, simply providing consumers with pertinent information on energy savings, while important, is not sufficient to effect change.

How should energy efficiency programs be designed? First, outreach programs should focus on those actions and technologies that are likely to have the greatest impact; that is, those with the most technical potential and the greatest potential to change behaviors and attitudes among the largest number of individuals. Second, where applicable the available financial incentives must be sufficient to get people’s attention. Third, an effective marketing campaign must be put into action. Fourth, credible and accessible information must be made available to the consumer. Fifth, participation in the program must be simple and easy. Finally, a trustworthy quality control mechanism must be in place to ensure that products and services meet expectations. The 2009 “Cash for Clunkers” program met these criteria, whereas existing home retrofit programs often violate these principles with poor marketing, delayed incentives, burdensome paperwork, and uncertain product quality.

A major regulatory effort is now under way in the United Kingdom to substantially increase the energy efficiency of the residential sector. Introduced in Parliament in December 2010 with an anticipated launch in 2012, the “Green Deal” would allow the up-front costs of energy efficiency retrofits to be paid by third parties, who would then share the long-term savings in energy costs with the property owners. This incentive program would complement existing regulations requiring the use of certain energy efficiency technologies. With a long lead time, such regulations can have a major effect on technology adoption with relatively little resistance from industry: a U.K. regulation requiring the use of high-efficiency boilers resulted in the market share of these boilers increasing from 20–30 percent to around 90 percent in approximately one year.

European regulators have been especially proactive in requiring efficient building technologies: all houses being sold or rented must now have an energy performance certificate, and this information must be provided to prospective buyers and tenants. In Germany this requirement has already been extended to encompass existing houses.

A developing area of research is the potential role of third-party intermediaries in promoting energy efficiency technologies at the household level. Third parties such as realtors, mortgage providers, and service engineers are well positioned to deliver information about potential energy efficiency upgrades to current and prospective property owners. A particularly promising opportunity to educate households about energy efficiency arises when those households undertake non-energy-related home renovations. Research is needed on the potential effectiveness of this approach and possible negative consequences.

Ultimately, effective policy design requires a combination of applied behavioral research and learning through experience. Behavioral barriers that need to be addressed include loss aversion, status quo bias, and entrenched social norms. The New york State Energy Research and Development Authority (NYSERDA) behavior research program is one initiative that is performing field experiments and employing community-based social marketing to increase participation in energy conservation programs. Program evaluation is a critical aspect of these approaches, and NYSERDA is actively involved in designing more-effective evaluation protocols.

How can successes at the local level be scaled up to the state or national level? To scale up local programs, geographic differences in behavior and environment must be taken into account.

The NYSERDA Behavioral Research Program

The workshop highlighted two successes of the NYSERDA behavioral research program. For an Albany-area vanpooling program, a personal marketing approach employing images, employer-mediated communication, and personal testimonials resulted in a large increase in participation. In a second project, a pilot study at Ithaca College examined the effect of descriptive norms (i.e., observed behaviors) and injunctive norms (i.e., descriptions of desirable behavior) to convince students to turn off computer lab equipment when their sessions were finished. Consistent with existing behavioral research literature, the descriptive norm of computers being turned off when students arrived was effective only when attention was called to this norm through an injunctive norm, such as a sign exhorting students to turn off the computer when done.

On which issues would regulatory mandates be more effective than voluntary actions? One area where mandates might be particularly effective is the problem of split incentives in rental properties, where the purchaser and the end user of appliances and energy are not the same.

Does providing consumers with increased information actually allow people to make more-informed decisions? Here, the main issues are informational overload (too much information can lead to analytic paralysis) and the structural obstacles consumers face in adopting new technologies, such as the complexities of hiring contractors and overseeing their work. More research is needed on how to design information to be easily understood, how to disseminate this information through trusted sources, and where well-designed information has the highest impact.

Can financial discounts persuade people to participate in energy conservation efforts? Some research suggests that providing financial incentives can have the opposite effect, leading people to feel that, because they are willing to pay extra, they are justified in using more energy. For this reason an approach based on intrinsic values is often more effective at encouraging participation.

More research is also needed on how to reduce the “rebound” effect, where consumers offset the financial savings from energy conservation by using more energy for other activities.

Session B: Public acceptance of new energy technology

Douglas Arent, Executive Director, Joint Institute for Strategic Energy Analysis, National Renewable Energy Laboratory

Juliana Birkhoff, Vice President of Programs and Practice, RESOLVE
Jeanne Fox, Commissioner, New Jersey Board of Public Utilities?
Jennifer Layke, Director, Institute for Building Efficiency, Johnson Controls Inc.
Eugene Rosa, Professor of Sociology, Washington State University

This session addressed the challenges related to acceptance of new energy technologies that introduce new factors into collective decision making, both within communities and among institutions. Panelists presented industry, academic, and public policy perspectives highlighting the complexities of the social dimensions of adopting new energy technology solutions and addressed such issues as privacy, equity, and individual rights.

The panel discussion examined broader issues relating to civil society, including factors and strategies that strengthen public acceptance of energy efficiency and new generation technologies. Panelists noted that a substantial body of research is available on the impact of effective public engagement on policy development and on the reasons for the dearth of public participation in government decision making. Speakers also highlighted many examples of successful dialogues between government, industry, and the public.

Trust is critical to any dialogue. But citizens tend not to trust government; indeed, the level of trust in government agencies has been in decline for several decades. The reasons for this lack of trust include the perception that governments do not tell the truth and are incompetent to carry out programs effectively. Trust is easy to lose; gaining it back is difficult. Another barrier to building trust is the fact that perceptions of costs and benefits differ for individuals and groups, and these perceptions also differ from those held by scientific authorities.

Public participation in decision processes nearly always builds trust and improves the outcome of those processes. Public participation is particularly valuable in building support for science-based decisions but requires clear and common goals, ample planning and resources, broad representation of interests, and transparency about how models are developed.

Public involvement in policy development may fail in several ways to meet the standards suggested by current research. First, the collection of data relevant to the decision may not be coordinated with the dialogue, with the result that when dialogue does happen the necessary data are unavailable or are out of date. Second, the public is often involved at too late a stage in the process, when it is too late to have a conversation about the overall goals of the policy or program. Third, the dialogue may not include all stakeholders, an outcome that is often the result of following a narrowly crafted model for public participation that fails to include all interests. Finally, many government agencies possess insufficient capacity to plan and execute a productive public outreach program, a problem that can be compounded by institutional skepticism regarding the usefulness of such programs.

Public Engagement on Offshore Wind: A Success Story from New Jersey

A large body of research explores the impact of effective public engagement on technology adoption, but application of this research within technology deployment programs has been limited. Public outreach programs often fail to include a broad representation from all sectors, and agencies lack the capacity to adequately involve the public in decision making on issues such as the siting of new generation facilities.

In an example of a successful public engagement effort, the state of New Jersey held public meetings in the spring of 2005 in the four counties bordering the Atlantic Ocean, to receive input on pending proposals for offshore wind farms. Much of the initial reaction was negative, even from interests as diverse as the fishing industry and the New Jersey Audubon Society. The state then embarked on a successful outreach program and commissioned studies on predicted economic impacts and the risks to migrating birds. This outreach effort generated widespread public support for offshore wind; Cape May fishermen even formed a group called Fisherman’s Energy to bid for offshore wind leases. In April 2011 the state granted permits to Fisherman’s Energy to build New Jersey’s first demonstration-scale offshore wind farm, to be located in the waters off Atlantic City.

Nevertheless, third-party intermediaries have facilitated successful and productive dialogues between the public and policy makers. Among numerous examples are the National Wind Coordinating Collaborative, the Nuclear Power Joint Fact-Finding Dialogue, and the National Conversation on Public Health and Chemical Exposures.

Public acceptance of new practices or technologies can be increased in several ways: by emphasizing usefulness, by imposing government mandates, by offering financial incentives, and by touting social image. Additional factors considered by public utility commissioners and other regulators include political considerations, special interest groups, and the impact on ratepayers. Utilities, by contrast, focus on the bottom line and are thus hesitant to adopt technologies that do not increase profits even if they are seen to serve the public good. To get utilities on board often requires the provision of additional incentives or offsetting revenue. One mechanism is to decouple revenues from sales; for example, by financially rewarding utilities for investing in renewable energy.

The leading factor in technology adoption by companies is the potential cost reductions achievable through those technologies. This is particularly true for building technologies. Other factors include reputation gains, greenhouse gas reductions, government policies, and government incentives. A less important factor that is nonetheless growing in importance is employee retention: greener buildings are seen as fostering more attractive workplaces and organizations.

Given these drivers, what factors increase the likelihood that building technologies will be adopted? In the government sector, standardized contracts and General Services Administration and Federal Emergency Management Program procurement guidelines play a large role in achieving energy efficiency goals and provide a venue for collecting feedback on the effectiveness of technologies and regulations.

Energy performance contracting provides an additional means to bundle many technologies together under a single program while financing improvements through third-party mechanisms. Energy performance contracting, whereby third-party service providers install efficiency or other carbon reduction technologies with a guaranteed return on investment, is increasingly common in both government and academic sectors. Colleges and universities are especially attractive locations for large-scale efficiency improvements because of the high level of technology interest and expertise on campuses, a prevailing sense of progress and social good, and the ability to do the long-term planning necessary for deep retrofits with long payback periods.

In the commercial sector, where the general perception is that low-carbon technologies—particularly renewable energy—are not cost-effective, technology acceptance dramatically increases when companies in a peer group engage in facilitated dialogue, which allows them to learn from one another’s experiences how to overcome behavioral and technical barriers. Major barriers include financing hurdles, lack of capacity to evaluate technologies, and uncertainty in project design and evaluation. Other needs include providing data in a format that is useful to decision makers and research on how best to accomplish this goal.

The likelihood that any given technology will be accepted in the marketplace is small, and common psychological and social factors determine the acceptance of both large, complex technologies and smaller, individually matched technologies. Such factors include psychological overload, framing effects, interpersonal influence, social status, and trust. These influences on consumer decision making run counter to the commonly used “rational actor” model, which holds that individuals make rational decisions on technology use based on in-depth analysis of all relevant information and the costs and benefits of all available options. In reality, individuals often make decisions on the basis of incomplete information or the advice of trusted but nonexpert acquaintances.

Particularly for large, complex technologies such as power plants, the research literature demonstrates that experts and laypeople often have divergent perceptions of risk. One reason is that experts emphasize quantitative considerations while laypeople emphasize qualitative features. The gap between lay and expert understanding of complex systems is growing, and thus an ever-increasing level of trust in experts is required of the public. One study found that the French and American publics have similar perceptions of the risks of nuclear power yet exhibit vastly different levels of support because of differing trust levels. Survey data show that public trust in almost every major American institution has declined since the 1960s. A key concept that has developed in recent years is that experts and the general public should collaborate in an analytic deliberative process to assess risk in technology and policy development.

Given the paucity of social science expertise in government agencies, participants asked how the social science community could assist policy makers in identifying high-quality social science research. Professional facilitators can help but only if policy makers know what goal they are trying to achieve. Participants concluded that the research on public participation provides clear direction on how to design successful participatory processes that educate both the public and the experts, incorporate local knowledge, reduce misinformation, and build trust.

Session C: Incorporating behavior in policy analytic tools

James Sweeney, Professor of Management Science and Engineering, Stanford University

Alan Krupnick, Research Director, Senior Fellow and Director, Center for Energy Economics and Policy, Resources for the Future
John A. “Skip” Laitner, Director of Economic and Social Analysis, American Council for an Energy-Efficient Economy
Holmes Hummel, Senior Policy Advisor for Policy and International Affairs, U.S. Department of Energy

This panel built on the previous two discussions to examine how individual and institutional attitudes and behavior could be more effectively integrated into available tools for developing policy, with special attention to how energy-economic modeling could incorporate actual behavior patterns. Panel chair Jim Sweeney noted the need to improve both the existing mathematical models and also our mental models of causality. In both cases these models are currently dominated by the role of technology, engineering, and economics, with behavioral science being underrepresented. Although economists focus on price, other factors also influence decision making.

Modeling has two general approaches, both of which have strengths and deficiencies. Top-down models embody the principle that economic actors seek maximum economic benefits, but in treating the economy in an aggregative manner these models miss many details about individual technologies. Bottom-up models contain a wealth of information about individual technologies but do not always fit with actual economic data. In both cases economists tend to focus not on energy quantities but on overall welfare—how well-off are people economically? However, several commonly cited metrics for economic prosperity—including gross domestic product (GDP), job growth, and energy quantities—may not be the best measures of social welfare.

Although economics is fundamentally a behavioral science, several aspects of behavior are difficult to incorporate into economic modeling. One issue is the paucity of data for new technologies and the often poor quality of data on old technologies. For example, the supply curve for extracting natural gas is poorly understood. A second problematic aspect of modeling behavior relates to the difficulty of modeling capital investments and innovation. A third problem is the wide variation in data on phenomena such as how consumers respond to energy price increases.

One concern that is often expressed during debates about energy policy is the potential for a large future gap between energy supply and energy demand. However, participants noted that a tenet of economic theory is that rising energy prices will stimulate innovation on both the supply side and the demand side.

Thus, energy modeling could potentially be improved in several areas. Much research needs to be done on improving both top-down and bottom-up economic models: top-down models need to incorporate more detail, whereas bottom-up models need more calibration with real-world data. More research is also needed on how to model imperfect compliance with and enforcement of regulations and on how to incorporate lessons from other behavioral sciences into economic models.

A specific case of how energy models do not predict real-world behavior is the “efficiency paradox” or “efficiency gap,” which describes the failure of individuals and institutions to adopt energy efficiency practices that are financially beneficial. Although the efficiency paradox is commonly described as a market failure, it exists at least in part because of hidden costs associated with energy efficiency; for example, the poorer-quality light emitted by high-efficiency lightbulbs. Economic theory holds that government intervention is justified in the case of market failure but not in the case of hidden costs.

Another problem posed by the use of highly detailed models is false precision—how should modelers sort out meaningful results from background noise? A solution might be to rely more on conceptual models that focus on fundamental aspects of the energy system, although these, too, will pose problems that require a more detailed analysis, such as comparing various proposals for a clean energy standard to determine which would be most effective. A single model should be used to compare different policies to provide the most useful conclusions regarding the relative predicted efficacy of those policies. At the same time, comparison of results from multiple models can help reduce errors or biases incorporated into a single model. Finally and critically, models must be kept up to date. This is a particularly difficult problem for models developed outside of government.

One speaker asserted that significantly improving energy efficiency through informed attitudes and behaviors would have a profound impact on American prosperity. This statement is based on a finding that from 1950 to 1980 the efficiency of converting energy production to work increased by 1.4 percent per year, with the economy growing an average of 2.25 percent per year. Since 1980, by contrast, this efficiency has declined by 1 percent per year, and the economy has grown much more slowly than in the previous thirty years. However, other participants questioned both the data and the implied direction of causality.

Desirable behavioral changes will result from changes in attitudes and motivations, not vice versa. For example, technology adoption is affected not only by price but also by payback time. From the consumer standpoint, payback time can be measured as a discount rate: what annual return in energy savings do consumers require before they will use a technology? Because this rate is profoundly affected by behavioral considerations, social science research can provide guidelines on how to reduce it so the up-front cost of energy efficiency becomes less of a deterrent.

A reduction in the discount rate would lower the carbon price required to drive technological or behavioral change, illustrating the importance of including consumer behavior and preferences in policy analyses. Behavioral parameters can and are being integrated into economic models; however, although ample data exist on behavior, these data are not readily available to economic modelers. Panelists noted the critical need for more coordination on data collection and data assessment in order to organize data so they can be readily inserted into existing models.

Modelers may respond to the problem of insufficient behavioral data by omitting behavioral considerations entirely while nevertheless incorporating overly optimistic estimates of future technological innovation. In such cases the output of the models will likely overweight the potential of unproven technologies such as carbon capture and storage (CCS) or hydrogen fuel cells while underweighting the potential of energy efficiency technologies such as controls for building lighting. In general, most economic models, including the widely used National Energy Modeling System, are relatively insensitive to behavioral changes, resulting in a bias in selecting which policies receive further consideration by policy makers. This problem is compounded by an emphasis among policy makers on using technical improvement metrics as a measure of policy success.

Two general cases where policy makers could benefit from more input from social scientists are in understanding and managing society’s tolerance for risk to human health and welfare from new technologies such as natural gas hydraulic fracturing and in developing tools to calculate and demonstrate to the public the societal benefits of these technologies. A specific area of concern is technology commercialization, and participants noted the key role that the social sciences could play in solving the so-called valley of death between technology development and technology deployment, a persistent problem that is not due solely to market failures and can not be entirely explained by standard neoliberal economic theories. A difficulty in studying any of these cases is that they concern the policies of many government agencies, each of which is primarily concerned with evaluating its own policies rather than the interaction of policies across government.

A common theme throughout this session was the paucity of economic analysis on the costs and benefits of various behavioral interventions in the energy system. One reason for the lack of data may be the difficulty of applying economic theories that are based on a certain understanding of the relationship between prices and costs to situations where that relationship does not apply. Also, many factors pertaining to the commercial sector have yet to be integrated into economic modeling, including how manufacturers determine the best timing for capital improvements and the impact of innovative financing mechanisms such as third-party financing on institutional behavior and decision making.

The session closed with a discussion of whether the research community should focus on developing new economic models to account for behavior or should instead concentrate on modifying existing models. Although a large body of social science research could be applied to economic modeling, little funding is available for model development. Participants noted that integrated policy assessment models require millions of dollars and several years to develop, and only a few successful models have been developed. These observations suggest that the best approach may be to reengineer existing models to be more sensitive to consumer choice and behavior.

Session D: Policy durability and adaptability

Kelly Sims Gallagher, Associate Professor of Energy and Environmental Policy, Tufts University

Kevin Carroll, Chief of the Energy Branch, Office of Management and Budget
Margo T. Oge, Director, Office of Transportation and Air Quality, U.S. Environmental Protection Agency
Philip R. Sharp, President, Resources for the Future

This session examined the extent to which policy durability and adaptability will be necessary to achieve an alternative energy future. Government officials and experts discussed the tension between the provision of consistent and long-term signals and the need to make policy responsive to new information. Participants also explored the complications that stem from relying on quick fixes for enduring energy problems.

The session opened with the observation that other countries are being more innovative than the United States in experimenting with how to construct durable and adaptable energy policies. The United States could learn from these efforts. Panelists also underscored the difference between durable policies and the indefinite provision of subsidies. For example, predictability in energy policy can be achieved through the planned phase-out of subsidies.

Speakers described several attributes that contribute to policy sustainability. These attributes group into two general categories: either the policy is affordable and effective, with broad consensus; or the policy is driven by a group of motivated stakeholders with little vocal opposition and infrequent review or oversight. Sustainability, moreover, implies general agreement about the nature of the problem being addressed. Agreement on energy issues, however, is often difficult to realize. Finally, policy durability can be negatively affected by unforeseen negative consequences if those consequences outweigh the benefits of the policy. Liquid biofuels were cited as a relevant area of concern.

Available policy tools include discretionary tools, such as government research and development funding or loan guarantees, and mandatory programs such as taxation, tax credits, and regulations. An important quality to consider when evaluating policies is whether their scale can change as the magnitude of the problem changes. Such policies are self-extinguishing: as the problem is overcome, the program ends without intervention from policy makers. This approach reduces unnecessary intervention in the market and also provides regulatory certainty to technology investors.

A major challenge for any policy is that the energy supply in the United States has been cheap and abundant for much of recent history, and any alternative fuel must be similarly inexpensive to be considered a valid alternative by a broad section of the populace. This is particularly true for automobiles, where petroleum-based gasoline has been the dominant fuel for over a century and automobile use is an ingrained aspect of American culture. An important question is thus, how will the public respond to advanced energy technologies and how willing is it to pay a premium for those technologies?

One institutional barrier is the number of agencies that exert regulatory influence on the transportation sector, including DOE, the Department of Transportation, EPA, and the state of California. An important recent development was the exercising of presidential authority to direct federal agencies to collaborate in crafting a revised Corporate Average Fuel Economy (CAFE) standard and to direct EPA to allow California to move ahead with more-progressive standards for its own vehicle fleet.

Why do more consumers not demand greater fuel efficiency, given that for an additional up-front cost of $900 or less they could realize $3,000 in fuel savings over the life of the car? This reluctance is particularly surprising in the commercial freight sector. Complicating factors include the difficulty of estimating fuel savings, the inherent complexity of the vehicle purchasing process, and the many competing attributes that consumers look for in a vehicle.

Panelists discussed three major realities that impact energy policy. First, energy markets are huge and global in nature, with oil being an especially fungible commodity. As a result, the United States has little ability to affect the price of energy through policy development. The U.S. government also exerts little direct control over energy markets within the United States. Most government policies are aimed at influencing private investments; for example, through loan guarantees or tax incentives. Finally, where direct government authority does exist, it is distributed among the three branches of government and among the federal, state, and local levels. Even within a given government institution the goals are constantly changing to reflect political considerations, thereby complicating the creation of durable energy policies.

The discussion period explored more deeply the question of what constitutes a durable policy. Participants discussed whether it is inconsistent to say that regulations are inherently durable when organized opposition to them is often substantial; these two conditions may be mutually incompatible. Speakers suggested that organized opposition can be countered by general public support, as is the case for the Clean Air Act.

A related issue is whether one can demonstrate to those opposed to a given policy that the policy is in their best interest. Participants discussed the specific example of production tax credits for wind power. Economists generally feel that a structured phase-out of production tax credits would be beneficial for the wind industry, but the industry generally opposes this policy. One panelist suggested the problem is that rapid political turnover means an emphasis on short-term goals: short election cycles foster an attitude of “take what you can get when you can get it.”

A critical aspect of establishing a durable and effective policy is to ensure an organized, persistent third-party evaluation of its effectiveness in achieving the stated goals in a cost-effective manner. Policy evaluation has been inconsistent, and one area of research might be to explore how to create a comprehensive framework that could be applied to all policies.

Regulations tend to endure while policies with budget implications—taxes and incentives, for instance—do not. Examples of enduring regulations include CAFE standards and the 1978 Public Utility Regulatory Policies Act (PURPA), although in both of these cases effectiveness has waned over time as technology has caught up with the standards. While regulations can be strengthened through subsequent legislation, a much easier approach is to ensure that regulators are given the power to bolster standards over time.

The panelists were pressed on how adaptability can be built into policy, especially in cases such as CCS where the technical (or, in the case of CCS, geologic) constraints are poorly understood and will vary among the individual projects covered by the policy. Citing the Clean Air Act as an example, panelists described how regulators could update the definitions of terms such as pollutant as new scientific or technical information becomes available. Policies can also include a requirement that regulators periodically update relevant standards to reflect technological progress, or policies can be updated through subsequent legislation, as with the 1990 Clean Air Act amendments. Policies ought to be based on sound scientific information so they can adapt to new scientific knowledge.

The private sector should be asked to identify potential risks associated with a policy and to present proposed solutions to regulators, as opposed to relying solely on regulators to identify problems and impose solutions on a resistant industry. The former approach is more common in European countries (e.g., Norway and the United Kingdom), but some American examples exist as well. The auto industry improved the efficiency of catalytic converters from 30 percent to 99 percent in the span of three decades, and this improvement was largely the result of soliciting industry input on how government policies could stimulate technical improvements. Industry input identified the sulfur content in fuels as being a major barrier, and subsequent policies requiring low-sulfur fuels permitted the development of higher-efficiency converters.

A major problem in policy design is how to avoid unintended consequences, a primary example being the use of methyl tertiary butyl ether (MTBE) to raise the oxygen content in gasoline and thus reduce pollutant emissions, as was required by the 1990 Clean Air Act amendments. MTBE was used despite general knowledge that this chemical presents a severe risk of groundwater contamination. Because groundwater contamination is addressed through the Clean Water Act, the Clean Air Act is not required to address this concern. The example of MTBE demonstrates a problem of accountability: if policy designers are not held accountable for unintended consequences, those consequences will not be considered during policy development.

Session E: Federalism

Ann Carlson, Professor of Law, University of California, Los Angeles

Barry Rabe, Professor of Public Policy, University of Michigan
Marilyn Brown,
Professor of Public Policy, Georgia Institute of Technology
Paul Centolella, Commissioner, Public Utilities Commission of Ohio

This panel addressed questions of how federalism relates to energy policy. Key issues included legal and political obstacles to the effective implementation of energy policy; the division of responsibility among the federal government, states, and localities; and possible alterations to the allocation of power among these levels of government that would facilitate the transition to an alternative energy future. Addressing energy challenges requires input from technologists, physical scientists, social scientists, and policy analysts, yet the presence of so many stakeholders can result in a diffusion of responsibility among these communities that impedes the creation of technological and policy solutions.

One approach to understanding how to deal with diffusion of responsibility is to examine the problem of scale. Generally the proper scale of an intervention or policy is analyzed less than the type of intervention. Environmental and energy policy is often dominated by the principle of subsidiarity: problems should be addressed at the lowest possible level; that is, state or local as opposed to regional, national, or international. For example, water pollution issues are often best addressed by local policy action, whereas climate change may be most effectively addressed at the international level.

Political realities often prevent action at a given level. In the absence of effective federal policy governing, for example, building codes, individual states have created their own laws. Thus the establishment of building codes does not require a unified national policy, but the resulting patchwork of building construction standards creates confusion among builders and necessitates a cadre of consultants to advise the industry on the practices of each state. (Another example, renewable portfolio standards [RPSs], is discussed below.)

Polycentric governance offers a mechanism with which to realize the benefits of multiple levels of policy action. The dilemma in blending actions at different levels is that each scale presents different benefits. The diversity of local actions fosters innovation, flexibility, and efficiency, whereas state and, especially, federal actions offer economies of scale and discourage polluters from simply moving to the state with the fewest regulations. Polycentric governance involves the simultaneous operation of energy and climate policies at many scales, while engaging many stakeholders. This approach provides backup policy mechanisms that offset imperfections that arise from intervention at a single level. Challenges include the potential for policy redundancy or conflict, and these issues must be dealt with on a case-by-case basis. A larger challenge to polycentric governance is how to establish such a system. Coordinating policies among states and localities requires federal governments to develop ways to compensate jurisdictions that suffer adverse consequences.

On global issues such as climate change, state governments, driven by anticipation of federal policy action, are often the first to develop policies. For instance, many states developed RPSs in anticipation of international treaties regulating carbon emissions. In a polycentric governance system these standards would combine with a federal carbon pricing regime to lower carbon emissions. To date, no such federal policy is in place. One reason for this lack of action is that states, through their representatives in Congress, have different bargaining positions and seek different objectives, including maximizing the benefits of their existing policies and procuring federal assistance for infrastructure improvements that may be necessitated by federal policy.

If U.S. energy policy is to remain state-dominated in the near term, what types of state policies will prove most effective? As twenty years of policy experimentation among the fifty states demonstrate, an inverse relationship exists between the economic desirability of a given type of carbon-mitigation policy and its political viability (Figure 2-1). The consensus in the economics literature is that carbon taxes are the most cost-effective, followed by cap-and-trade regimes. RPSs are the least cost-effective but have gained the greatest acceptance among state legislatures. Cap-and-trade systems have been adopted less frequently. No state has yet instituted a blanket carbon tax, although twenty-eight states have instituted a surcharge on electricity bills that funds renewable energy programs. Although carbon pricing schemes do not enjoy universal political or public support, applying revenues generated from such schemes to renewable energy research and development is broadly supported.

Figure 2-1

Figure 2-1. Relationship between economic desirability and political feasibility of three carbon reduction policies. Numbers in parentheses indicate the number of states that had adopted each regulatory approach as of May 2011.

Examples of polycentric governance
  • Denmark’s Electricity System
    From 1980 to 2004 Denmark lowered its per-GDP carbon dioxide emissions by nearly 50 percent while simultaneously building the world’s largest wind turbine export industry. These achievements were made possible by a combination of taxes on emissions, fuels, and electricity; investment subsidies and long-term financing guarantees; and regulatory support for small-scale renewable energy generation and combined heat and power facilities. Denmark’s polycentric approach blends small-scale decentralized community control with national standards and policies.
  • Germany’s Feed-in Tariff
    Germany’s feed-in tariff is often cited as a model for similar policies around the world. In addition to providing a financial incentive for small-scale renewable energy production, German regulations provide for the participation of utilities while also requiring them to provide grid access to small generation facilities. The feed-in tariffs differ by energy source—they are determined by each source’s associated generation costs—and are programmed to decrease over time in concert with expected cost reductions. Germany’s polycentric approach integrates residential and community producers of wind and solar energy with federal policy concerning tariffs and digression rates.
  • Brazil’s Proálcool Program and Promotion of Flex-Fuel Vehicles
    Brazil launched its National Alcohol Program (Proálcool) in 1975. As a result of this program ethanol production is now cost-competitive with gasoline production in the absence of subsidies, and the country has recouped its investment through large reductions in oil imports. The program included mandates for blending ethanol with gasoline, partnerships with automobile manufacturers to produce and promote flex-fuel vehicles, the use of gasoline taxes to subsidize the price of ethanol, and partnerships with environmental groups to design regulations to protect rainforests and other environmentally sensitive land.
  • Singapore’s Urban Transport Policy
    Singapore’s urban transport policy employs a combination of approaches to reduce road congestion and improve air quality in this island city-state, including the world’s first congestion pricing scheme (subsequently expanded to cover a larger area), certificate of entitlement auctions for vehicle ownership, and the integration of land-use and transportation planning. Singapore’s polycentric approach harnesses public-private partnerships to operate mass transit systems and works with automobile manufacturers to equip vehicles with electronic road pricing devices.
  • Bangladesh’s Grameen Shakti
    Grameen Shakti is a nonprofit company that provides microcredit-based financing and technical assistance for rural renewable energy projects, including solar photovoltaic and biogas installations and improved cook stoves. A critical aspect of this initiative is the enrollment of local communities, both in project financing and in maintaining the installations, as well as the engagement of district and national policy makers, international donors, and lending firms. The program has resulted in a large reduction in deforestation in Bangladesh, which relies on trees and bamboo for nearly half of its energy.
  • The EPA Toxics Release Inventory
    The Toxics Release Inventory is a publicly available database containing annually updated information from industry groups and the U.S. government regarding releases of toxic chemicals in the United States. A number of nongovernmental organizations were instrumental in establishing the database and improving public access to it. The project is managed by the national government, but facilities compile and report information on their own releases.

How would a federal RPS be instituted in the face of so many state RPS policies? A large body of social science literature describes how the framing of a policy affects its public acceptance. Across the political spectrum, state policy makers tend to frame state RPS policies not in terms of energy or climate concerns but in terms of economic development and the potential to benefit from anticipated federal policies. Although this framing could be useful for promoting a federal RPS, it might be counterproductive if costs rise more than anticipated or if the RPS fails to deliver economic benefits to every state.

State public utility commissioners face challenges related to economic security, energy security, and cyber security. A major economic challenge is the $1.5–2 trillion investment that will be needed over the next twenty years to replace aging infrastructure and to build new facilities to meet the anticipated growth in electricity demand. Energy security will require greater diversity in energy sources, particularly transportation fuels, and mechanisms must be developed to defend the energy system against cyber attacks. Short-term concerns of maintaining an adequate energy supply and reasonable prices must be balanced against long-term planning to improve the system over time and to adapt to changing environmental and market conditions.

Achieving the proper balance between short-term and long-term planning requires expanding the range of regulatory options and thinking about investment in a different way. How can utilities foster innovation, address market failures related to energy efficiency, engage consumers on the question of dynamic electricity pricing, and pursue smart grid technologies as a platform for innovation? Least-cost planning—the historical model for investments by public utilities—may not be the best approach in the face of uncertainty over future economic, environmental, and regulatory changes. Rather, planners should value the development of an adaptable electricity grid, incremental and modular infrastructure investment rather than large monolithic capital investments, and the ability to delay making irreversible investment decisions when outcomes are uncertain. A critical issue for the social sciences to consider is how planners should value options and risks.

The role of regulators also needs to be restructured. The historical function of regulators has been to decide whether projects proposed by utilities are prudent, reasonable, and necessary. Regulators must learn to be more proactive in clarifying the objectives of existing statutes and aligning stakeholder incentives with those objectives. One means for achieving this goal is to sponsor policy workshops aimed at defining new terminology, facilitating dialogue, and disseminating new information.

Participants asked whether utilities have incentives to sell less electricity and identified two types of disincentives. One is the reliance on volumetric charges to recover fixed costs. Solutions include straight-fixed-variable rate design and direct compensation for revenues lost as a result of energy efficiency programs. A less tractable disincentive, especially for utilities that sell electricity entirely on the wholesale market, is the fact that reducing demand for electricity also reduces its price. A counteracting factor is the expense of replacing retiring generation facilities. In many cases the more cost-effective option is to support demand-reduction programs.

The Smart Grid Interoperability Panel

This session highlighted several examples of successful collaboration between public utility commissions and other stakeholders. The Smart Grid Interoperability Panel is a public-private partnership created by the National Institute for Standards and Technology to accelerate the development of technical standards for the smart grid. Through this initiative, standards that would normally require four or five years to develop are being completed in twelve to eighteen months. This success is due in part to strong early support from the Departments of Energy and Commerce and from about seventy leaders in industry and public policy. Critical social science questions include: How should the development of technical standards for the electric grid be organized and funded in the long term? And what is the most appropriate way to support research and development within a regulated industry?

Combined heat and power (CHP) plants are a more efficient alternative to large, centralized generation facilities. Although common in Europe, CHP has struggled to gain acceptance in the United States. CHP can be facilitated by grid interoperability standards, by available efficiency credits, and by providing higher offer prices for electricity generated by CHP plants located in areas of marginal grid stability. In many locations the major barrier that must be addressed is the existence of exclusive service territory statutes, which prevent microgeneration facilities from selling electricity to nearby consumers.

Session F: How do regulations need to change?

Granger Morgan, Professor and Head, Department of Engineering and Public Policy, Carnegie Mellon University

Edward A. (Ted) Parson, Professor of Law and Professor of Natural Resources and Environment, University of Michigan
Robert R. Nordhaus, Member, Van Ness Feldman
Jonathan Cannon, Professor of Environmental Law, University of Virginia

Changing the energy system will require new regulations as well as alterations to existing policies that inhibit this response. For example, implementation of CCS technologies will require a regulatory regime that does not yet exist. In addition, a variety of existing tax policies must be modified so as not to discourage investments in alternative energy technologies and energy efficiency. The panel examined how energy regulations could be altered to promote the spread of new technologies.

Innovation in the energy system requires either the opportunity for financial gain or regulation that induces innovation. Regulations can also improve the financial outlook for a technology: putting a price on carbon, for instance, would increase the cost-effectiveness of CCS as a carbon-mitigation strategy. Performance standards are another useful regulatory tool.

How can regulation or other public policy measures more effectively promote socially beneficial technological change to reduce environmental burdens? Despite much academic research over the past four decades, few answers are available, and many fallacies persist. New insights into this long-standing question can be gained by more closely examining the problems posed by uncertainty over future technological capabilities and costs. This uncertainty contributes to a “policy double bind” in which policy makers are expected to design regulations that will drive new technologies without demanding technology improvements that are impossible to achieve. Despite decades of industry success in meeting new technical standards for vehicle and appliance efficiency, policy makers are often faced with the difficult, if not impossible, task of proving that newly proposed standards are technically achievable. Compounding this problem, regulators often do not have full access to industry data regarding the capabilities of technologies that are under development. One solution is to involve a third party from the private sector that has a financial interest in convincing the target industry to disclose critical information. For example, new technical information on the potential for reducing motor vehicle emissions came to light because of efforts to resolve the competing claims of auto companies and oil companies.

In exploring how to design regulations that foster innovative research and cost-effective technologies, a useful step is to study historical cases where policy development was complicated by competing claims over technological capabilities, such as motor vehicle emissions, ozone-depleting chemicals, and dioxin discharge from pulp mills. One can ask what technological breakthroughs were anticipated, how that understanding was advanced by new breakthroughs, and how developing knowledge impacted the policy debate. Such case studies have revealed that regulators have many opportunities to align incentives for the private sector with the goals of government regulations. Conversely, policy makers must be careful not to specify unrealistic penalties for a failure to meet a regulatory requirement, because such penalties are impossible to enforce and hence do not influence industry behavior. For example, the motor vehicle standards in the 1970 Clean Air Act specified damages for failing to meet emissions standards that were twice the average vehicle cost at the time.

The practice of using federal regulations to enable specific technologies dates at least to the 1920 Federal Water Power Act, which established a licensing system for nonfederal hydroelectric projects. The act, which still governs approximately half of all hydroelectric generation in the United States, covers all siting issues in one regulatory package, including environmental considerations, land use, and eminent domain. The 1938 Natural Gas Act took a similar approach to establishing a national system of gas pipelines. Other examples of federal legislation that has fostered technological innovation include the 1954 Atomic Energy Act, the 1970 Clean Air Act, and the 1978 PURPA (Figure 2-2).

  • Federal Water Power Act (1920): Hydroelectric licensing; established a comprehensive facility-siting authority for nonfederal hydroelectric development.
  • Natural Gas Act (1938): Natural gas pipeline regulation; comprehensive regulation of siting and construction of interstate natural gas pipelines and rates, terms, and conditions of service.
  • Atomic Energy Act (1954): Licensing and safety regulation of commercial nuclear reactors, including limitation of liability for nuclear accidents under the Price-Anderson Act (1957) and a framework for permanent disposal of spent nuclear fuel under the Nuclear Waste Policy Act (1982/1987).
  • Clean Air Act (1970): Created statutory standards for light-duty motor vehicle emissions that drove the deployment and use of catalytic converters.
  • PURPA (1978): Opened up electricity markets to nonutility renewable generators and cogeneration facilities.

Figure 2-2. Federal regulatory interventions that have accelerated the deployment of new energy technologies.

Efforts are under way to develop a similarly comprehensive federal regulatory regime for the widespread deployment of CCS. Legal issues that remain to be resolved include who owns the right to make use of deep geological formations, how permitting should be managed, how to resolve long-term liability issues such as compensation for personal injury or property damage, and how eminent domain laws should be applied or modified to develop an extensive carbon dioxide pipeline infrastructure. Regulations on liability should provide both incentives for safe operation and mechanisms for compensating injured parties and should also clearly establish the parties responsible for long-term site remediation. One proposal is to pay for the long-term costs of storage, remediation, and liability through a per-ton fee on sequestered carbon.

Participants noted that carbon dioxide pipelines are already in operation in the United States, primarily for enhanced oil recovery, and asked whether existing regulations would be sufficient to cover an expanded pipeline system. A complicating factor is that carbon dioxide pipelines are currently regulated at the state level, and only a few states have created a regulatory regime for this infrastructure. Furthermore, existing state laws and regulations, including laws pertaining to eminent domain, vary widely, and this variation would make the construction of interstate pipelines difficult under current state regulations. One panelist recommended an opt-in national program to supplement state-by-state regulation.

Although the legislation listed in Figure 2-2 has been generally effective in meeting energy and environmental goals, instituting a legally effective regulatory regime does not ensure that a technology will gain public acceptance. Hydroelectric power and nuclear waste management are two areas where public opposition has halted projects despite the presence of effective regulatory schemes. Useful mechanisms for increasing public acceptance include community outreach, transparency, and independent technical analyses. Regulations mandating these strategies ensure that policy makers address the societal dimension of technology deployment but also risk imposing burdensome administration and lengthy litigation. For this reason, incorporating decision-related concerns into a workable regulatory regime requires careful attention to how additional administrative requirements will work in practice.

The panel also discussed the value of regulating total greenhouse gas emissions as a driver of technological innovation. Economists generally accept that placing a price on greenhouse gas emissions through a carbon tax or cap-and-trade scheme is more effective at driving change than relying on mandates or other prescriptive regulations. This is because carbon pricing regimes give the private sector the flexibility to experiment with cost-effective ways to reduce emissions without requiring that governments pick winners and losers. Carbon pricing regulations may also have lower administrative costs than do more-prescriptive policies, because developing separate policies for each industry and each sector of society could be time-consuming and expensive.

Pricing carbon emissions can also foster innovation and long-term planning, although companies need to have confidence that the pricing arrangement will be in effect for a long period of time before they will commit to long-term investments and new research and development programs. Where carbon pricing mechanisms are insufficient to achieve ambitious, short-term carbon reduction goals and to drive innovation, they can be complemented by prescriptive regulations such as building, appliance, and automobile efficiency standards and RPSs. As in previous sessions, participants observed that lessons from other countries could be applied to designing regulations in the United States.

A final social science issue concerns the influence of social discourse on the adoption of regulations. The 1970s and the early 1980s saw widespread, bipartisan support for environmental regulations. More recently, market-based solutions such as cap-and-trade policies are increasingly seen as representing an unacceptable level of government intrusion. Although this shift in the social discourse may affect the political feasibility of particular regulatory changes, participants cautioned against basing long-term planning on recent political trends and public opinion. Now is the time to develop an understanding of how to design and implement useful regulatory regimes so that constructive intellectual frameworks are immediately available to policy makers when public support does materialize.


2 Dietz, T., G. T. Gardner, J. Gilligan, P. C. Stern, and M. P. Vandenbergh, “Household Actions Can Provide a Behavioral Wedge to Rapidly Reduce U.S. Carbon Emissions,” Proceedings of the National Academy of Sciences 106 (2009): 18452–18456. doi:10.1073/pnas.0908738106.