| 英文摘要: | State implementation of new Environmental Protection Agency climate regulation may shift behavioural strategies from sidelines to forefront of US climate policy. In a rare move, the US Environmental Protection Agency (EPA), in a new draft rule known as 'The Clean Power Plan', has signalled that it will allow states and utilities to meet emissions standards by reducing electricity demand. The details of this regulation will have a substantial impact on its effectiveness1, creating a tremendous opportunity to put integrated, multidisciplinary science to the practical end of mitigating climate change. Huge untapped potential exists for using knowledge about how the public responds to new technology, financial incentives and regulations2. Financial incentives for home weatherproofing, for example, have varied tenfold in their impact on rates of adoption, depending on a range of features of programme implementation beyond the financial incentives offered3. Incorporating insights generated from such integrated science could reduce compliance costs and achieve emissions reductions with minimal intrusion. But achieving these objectives will challenge the thinking of regulators who have more experience with mandating best available technologies than with programmes that target decision-making and voluntary behaviour. It will also challenge behavioural scientists to engage seriously with the practical issues of energy policy development and implementation. Finally, it will require a coordinated research agenda and greater collaboration between industry and academia.
Programmes and policies that draw on social and behavioural science to support reductions in energy demand have gained widespread attention in the United States and around the world for their potential to contribute to climate change mitigation2, 4, 5. A recent estimate suggests that energy efficiency improvements in the United States could reduce end-use demand by 23% by 2020, avoiding 1.1 gigatonnes of greenhouse-gas emissions per year6. Others have estimated that, given historical responses to evidence-based programmes and policies, a 7% reduction in US emissions is possible if targeted efforts to reach the household sector are expanded2. Behavioural programmes — those that promote both behaviour change in how energy is used and the adoption of products that reduce carbon footprints — have been successfully used by national and subnational governments7, 8, non-governmental organizations9 and energy providers10. But they are often viewed as minor adjuncts to more traditional regulatory actions, not as a core part of the response to energy and climate challenges11. The implementation of the EPA's proposed Section 111(d) 'Clean Power Plan' rule could change this. By enabling states to meet federal emissions targets with programmes for electricity demand reduction, including those integrating behavioural knowledge, the regulation has the potential to lower compliance costs by stimulating the design of innovative programmes, improved assessment methods for behavioural interventions, and new investments in basic and applied behavioural research.
The EPA's draft 111(d) rule12 seeks to reduce 2030 power-sector carbon dioxide (CO2) emissions by 30% from a 2005 baseline. The rule, which is scheduled to be finalized in mid-2015, will create incentives for states to submit compliance plans that include switching from coal-fired electric generating units to natural gas. But the draft rule also enables states to meet their targets through beyond-the-plant demand-reduction programmes, including actions involving consumers. The draft rule proposes an overall state goal of 1.5% annual electricity savings from demand-side energy efficiency efforts between 2020 and 2029. Given the results of well-designed behavioural programmes, this goal is very modest; more could be achieved. States will be required to submit the first phase of their compliance plans a year after the rule is finalized and will have a first compliance deadline of 2020, so substantial time remains to ensure that states exploit the full potential of demand-side reduction initiatives. The proposed carbon rule recommends the inclusion of both technological and behavioural programmes, including programmes that “accelerate the deployment of both energy-efficient technologies and behaviours by addressing market and cultural barriers”12. The social and behavioural sciences have documented a range of strategies and programme design features that can achieve these goals8, 10, such as providing enhanced energy-related information to the user, social communication, real-time energy feedback, and targeted information that simplifies the task of weighing complicated and sometimes expensive alternatives when making choices about household technologies. State decisions about what demand-reduction programmes to include in compliance plans will be critical to realize the potential of behavioural strategies.
Although legal challenges will affect the role of demand reduction, much will also depend on the EPA's decisions about acceptable evaluation, measurement and verification (EM&V) plans. States will be required to develop EM&V plans that explain how -demand reduction will be measured12. The draft rule states that these plans must be “rigorous, complete, and consistent” with forthcoming EPA requirements and guidance. These requirements may determine the fate of some behavioural programmes. The EPA has expressed concern that behavioural programmes may pose quantification and verification challenges, specifically citing information-based and targeted behavioural programmes as potentially problematic12. For a variety of reasons, the effects of behavioural approaches require evaluation methods different from what has historically been used to capture the effects of policies and programmes13, 14. Accurate estimates can, however, be achieved with rigorous approaches common within the behavioural sciences, such as randomized controlled trials13, 14 and 'state of the science' analysis of data over time15. These evaluation methods are relatively new to utilities, and partnerships between researchers and industry may help to ensure success as EM&V standards are set and adopted13. These approaches also have the potential to improve the accuracy of evaluations of more traditional programmes that incentivize specific technological improvements, for example consumer rebate programmes to promote heating, ventilating and air-conditioning retrofits. Traditional estimates of annual energy savings are often based on the estimated demand reduction achieved by an equipment upgrade multiplied by the number of customers who upgrade. But actual demand reduction can differ substantially from these estimates because of usage changes (that is, rebound effects, spillover effects)16. In addition, some programmes may feed consumers into other programmes and that may amplify the effectiveness of the other programmes, leading to the possibility of either double counting or extra benefits from spillover. Improved EM&V techniques can more accurately estimate these results and also provide insights into how to design programmes that complement one another to maximize results. As behavioural programmes become more common and standardized approaches are developed, the challenges associated with EM&V will be substantially reduced. New research can accelerate the development and implementation of effective and easy-to-evaluate behavioural programmes. Improved communication between industry, government and the scientific community will be critical to ensuring that future research is designed in a way that will generate results but is also sensitive to the regulatory environment within which states and utilities operate. To address EM&V concerns, the EPA is considering including a pre-defined list of well-understood programmes in the final rule, and requiring greater documentation and explanation when state plans include programmes with less well-understood features12. If a pre-defined list is adopted, the breadth of the list will be important, as will opportunities to modify the list as research in this area advances. The EPA is also considering limiting eligible demand reduction to this pre-set list12. This is unfortunate, as a zealous effort to eliminate phantom reductions from the list can also limit states' opportunities to innovate. It is critical for decision-makers to be aware of cost-effective, scalable and verifiable behaviour-based programmes when these decisions are made. Behavioural scientists can assist in filling knowledge gaps by synthesizing results from rigorous evaluations of behavioural programmes that are candidates for inclusion in compliance plans. Research by industry and academia is rarely integrated into more robust meta-analyses, but substantial insights could be gained if it were.
If the final rule is friendly towards the inclusion of programmes engaging behavioural science, and if states respond by including robust demand-reduction efforts in compliance plans, much research will be needed to expand the current knowledge base. States should be encouraged to follow the lead of the Northeast Energy Efficiency Partnership, which has provided a forum for public utility commissioners and air-quality regulators to act collectively to share research results and reduce the costs of EM&V and other 111(d) compliance issues. Behavioural scientists should include as research targets metrics of success that are relevant to EM&V requirements, once such metrics are defined. Collaborations with industry, a coordinated research agenda within the scientific community, and the establishment of 'best practices' guidelines for researchers can ensure a more streamlined transition from research into practice. Much of the success of the rule will be determined by the decisions of federal and state regulators in the months and years to come. If the rule is too restrictive in its requirements for states to demonstrate effects, it could discourage the use of cost-effective approaches, inhibit innovation and result in a rule that fails to capitalize on the immense opportunity to reduce greenhouse gas emissions through behaviour change2. The response of the EPA and the states on issues such as the scope of acceptable behavioural programmes, and the level and type of documentation required, may determine whether demand-reduction approaches achieve their full potential or whether concern about phantom demand reductions induces the EPA and the states to throw the baby out with the bath water.
- Fowlie, M. et al. Science 346, 815–816 (2014).
- Dietz, T., Gardner, G. T., Gilligan, J. M., Stern, P. C. & Vandenbergh, M. P. Proc. Natl Acad. Sci. USA 106, 18452–18456 (2009).
- Stern, P. C. et al. Eval. Rev. 10, 147–176 (1986).
- Ansolabehere, S. & Fri, R. W. Daedalus 142, 162–171 (2013).
- Sovacool, B. K. Nature 511, 4–5 (2014).
- Granade, H. C. et al. Unlocking the Energy Efficiency in the US Economy (McKinsey, 2009); http://go.nature.com/TvICkU
- US National Research Council Limiting the Magnitude of Future Climate Change (National Academy Press, 2010).
- Vandenbergh, M. P., Stern, P. C., Gardner, G. T., Dietz, T. & Gilligan, J. M. Environ. Law Rep. 40, 10547–10554 (2010).
- Fleet Drivers and Fuel-Smart Driving (Environmental Defense Fund, 2009).
- Mazur-Stommen, S. & Farley, K. ACEEE Field Guide to Utility-Run Behavior Programs (ACEEE, 2013).
- Vandenbergh, M. P., Carrico, A. R. & Bressman, L. S. Minn. Law Rev. 95, 715–781 (2011).
- Environment Protection Agency Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units 79 FR 34829 (Federal Register, 2014); https://federalregister.gov/a/2014-13726.
- Allcott, H. & Mullainathan, S. Science 327, 1204–1205 (2010).
- Todd, A., Stuart, E. & Goldman, C. 2012 ACEEE Summer Study on Energy Efficiency in Buildings (ACEEE, 2012).
- Frank, K. A. & Maroulis, S. J. Educ. Eval. Policy Anal. 35, 437–460 (2013).
- Truelove, H. B., Carrico, A. R., Weber, E. U., Raimi, K. T. & Vandenbergh, M. P. Glob. Environ. Change 29, 127–138 (2015).
Download references
M. McLean and B. Tenbarge provided research assistance. T.D.'s contributions were supported in part by Michigan AgBioR |