Evaluating energy efficiency policies

To reach future climate targets and goals of reduced energy consumption, the recent decades have seen initiatives of various policy instruments to promote energy efficiency. It is important to understand the behavioral responses to the various policy instruments and the effects on energy consumption, to evaluate whether the policy instruments have worked according to the intention. This report provides a mapping of both theoretical and empirical analyses of effects of different policy instruments, by giving an overview of the economic literature in the field. We discuss strengths and weaknesses of various methods for evaluation of measures, and put special emphasis on literature that analyzes how behavioral changes affect the effectiveness of policy instruments (known as rebound effects). The main focus for the discussion of the empirical literature is on Norwegian studies. If there are no adequate Norwegian studies, we use examples from studies in other countries and discuss the transfer value to Norwegian conditions.

The report defines important concepts, including various sources of rebound effects based on economic theory, and discusses differences between energy efficiency, energy efficiency potential and energy conservation. The report gives a brief overview of various methods to evaluate impacts of policy measures and distinguishes between methods based on empirical observations and numerical model analyzes. We have structured the literature by method (empirical observations versus model analyzes), type of policy instrument, and by which sector the policy instrument is geared toward. We draw the following main conclusions:

• The bulk of regulations linked to energy efficiency and energy use in households and heating in businesses, are building regulations and energy labeling of appliances. There are no Norwegian studies of the effects of energy labeling of dwellings in Norway. A Danish study of energy labeling in residences finds no general efficiency of the energy labeling scheme on energy consumption in Denmark.
• The analyzes of heat pump ownership in Norway indicate that in some cases the potential for reduced energy use is not necessarily realized, and in some cases one can observe significant rebound-impacts of efficiency measures.
• The main reasons why pellet stoves have failed is the positive perceptions about other heating options, especially wood stoves and heat pumps. This illustrates why investment support is not a warranty for increased use of a specific energy source.
• Norwegian studies of experiments with automatic meter reading and demand side management show that there is potential to reduce energy consumption and streamline the entire energy system by moving load to less strained energy periods.
• Voluntary agreements that replace energy-fee duties have proven to trigger energy efficiency measures which would be profitable for firms even in the absence of a scheme (both in Norway and in Sweden).
• Evaluation of schemes with white certificates in some countries in Europa finds that the measures prove to be both private and social profitable.
• Computable, general equilibrium models (CGE-models) take into account many interaction effects and is a well suited tool to evaluate impacts of policy instruments. International studies find that expected collected energy savings (when regard is taken to rebound-effects) as a result of an energy efficiency measure will constitute about 40-80% of the initial energy efficiency potential.
• CGE-analyses show that energy efficiency measures are less effective than carbon pricing in reducing carbon emissions, because reduction in emissions happens via reduction in energy consumption and not via substitution away from carbon-intensive energy technologies that will be the case with carbon pricing.