A Comparison of National Energy Efficiency Policy Evaluation Methods: Models versus Indexes

Panel: Panel 3. Monitoring & evaluation: understanding change and how to deliver energy efficiency

Author:
Marvin J. Horowitz, Ph.D., President, Demand Research LLC, USA

Abstract

Energy efficiency and sustainable environmental policies are expanding in scope and purpose throughout the world. To ensure the success of these initiatives in meeting their goals, and to monitor the effects of policies from year to year and from country to country, there is a growing need for improved policy monitoring and evaluation research. Towards this end, this paper focuses on understanding the advantages and disadvantages of two distinct statistical approaches to evaluating the annual impacts of national energy efficiency policies. One of these is an econometric modeling approach in which times series, or cross section time series, regression models are used to compare actual energy use to counterfactual (business-as-usual or baseline) energy use, while the other is an index approach in which the energy efficiency portion of a multi-sector energy intensity index is isolated. Using empirical data on energy consumption within the United States, specific comparisons are made between the two approaches to show how each are constructed. This paper should help guide future energy efficiency evaluation planning efforts, such as the EMEEES project, in determining which approaches should be applied to policy analysis.