Economic analysis of energy efficiency - Work package 1

What are the market failures that hamper the financing of energy efficiency? What policy interventions can address the identified market failures?

Lead partner: CIRED

Energy efficiency policies are often justified with different arguments by economists and policy makers. Incidentally, economists and policy makers do not focus on the same policy evaluation criteria. In this work package, we will review the evaluation criteria and disentangle the normative views (e.g. welfare considerations) and the positive views (e.g. energy considerations) they encompass. We will provide guidelines as to their appropriate integration in the subsequent work packages.

Objective

Public support to energy efficiency is justified in both economists’ and policy makers’ discourse by the argument that it is the most cost-effective carbon dioxide abatement technology. From an economic point of view, the optimal policy to internalise the carbon dioxide externality is to put a price on emissions (through a tax or tradable permits). However, such a policy often faces acceptability constraints. If it cannot be implemented, then subsidising carbon dioxide abatement technologies can be a desirable second-best option. Since energy efficiency is widely considered as the cheapest of these technologies, cost-effective subsidies should target energy efficiency first. Next to the carbon dioxide abatement argument, the “energy efficiency gap” provides other economic justifications for public interventions targeting energy efficiency.

Considering the building sector, several market failures in energy efficiency markets can explain the energy efficiency gap:

• Incomplete and asymmetric information: The energy efficiency of buildings depends on many idiosyncratic factors such as climate or architectural characteristics, as well as human factors such as the quality of installed technologies or the occupant’s behaviour. As a result, energy efficiency technologies can be seen as credence goods, the performance of which is never completely known to the buyer. This characteristic gives rise to information asymmetries (adverse selection and moral hazard) between installation contractors and building managers, between building owner and occupants or between prospective buyers and the seller of a building. Information provision or innovative incentives can help guarantee energy efficiency in building management contracts, in building rental contracts and in building sale
• Credit constraints: Some energy efficiency measures, like home energy retrofits cannot be secured by a collateral, hence entail a high borrower’s default risk. Some credit enhancement measures can be justified to address this problem
• Technology adoption spill-overs: Some early adopters are needed for a mature technology to start This positive adoption externality is also known as the neighbour or peer effect. It can be internalised by temporary subsidies.
• Imperfect competition: If there is some market power in energy efficiency industries, then the supply of energy efficient technologies is inefficiently low and their price is inefficiently high. Some anti-trust policy is needed to address this
• Inattention and other “behavioural anomalies”: Besides imperfect information or competition issues, there is empirical evidence that consumers depart from purely rational behaviours in their energy efficiency-related 2

Policy implementation does not necessarily follow the normative economic principles mentioned above. In political discourses, in particular, energy efficiency policies are mainly motivated by the internalisation of carbon dioxide externalities, regardless of other potential market failures. Accordingly, policy makers set targets in energy saving or energy efficiency terms. For instance, in France the building sector is bound to achieve a 38% energy use reduction and 500,000 annual home energy retrofits by 2020. Yet, there is no assurance that such targets correspond to welfare improvements, a metric most relevant for economic evaluation.

The disconnection between economists and policy makers about evaluation metrics has profound implications when it comes to assessing energy efficiency policy. In this work package, we will discuss several evaluation criteria to ease the dialogue between economists and policy makers.

Methodology

For each policy instrument considered in subsequent work packages, we will proceed first with a normative analysis based on a comprehensive review of the literature.

The literature review will systematically examine the following questions:

• What market failure motivates implementation of the policy under scrutiny? Some policies can serve different goals; their overall efficiency should thus be examined in relation to each of these goals. For instance, energy efficiency subsidies can be justified in two different ways. They can be seen as a means of internalising carbon dioxide emissions. Regarding that objective, however, they are only a second-best solution: they are not the cheapest way to save one unit of carbon dioxide emission. This is because unlike energy taxes (the first-best solution), they induce a rebound effect, that is, a higher use of energy efficient durables after investment (Giraudet and Quirion, 2008). Alternatively, energy efficiency subsidies can be seen as a first-best solution to internalize technology adoption spillovers, if any.
• How to assess the welfare improvements due to the policy? Social welfare is defined as the sum of the consumer’s surplus and the producers’ surplus. The consumer’s surplus can be thought of as the margin between the price paid by the consumer and the utility he or she derives from energy services (e.g. the willingness to pay for a higher indoor temperature, or for better lighting). Unlike price data, the consumer’s utility is often In such situations, some sufficient statistics have to be rigorously established to approximate the consumer’s surplus (e.g. Giraudet and Houde, 2014). The producer’s surplus can be thought of as the margin between the price charged and the production cost. Just like the consumer’s utility, the producer’s cost is rarely observable. The degree of market power can be helpful to approximate the price-cost margin.
• What are the costs induced by the policy? Policy implementation induces costs of at least three types which must be weighed against welfare improvements. First, one key challenge with energy efficiency policies is the measurement and verification of energy Ex post measurements of energy savings can diverge from ex ante predictions for many reasons, including the rebound effect or some installation defects. This challenge has been widely documented with white certificate obligations (see Giraudet and Finon, 2014). Second, some policies like subsidies are likely to attract infra-marginal participants, that is, consumers who would have purchased the good even in the absence of the programme. This can be seen as an adverse selection problem from the regulator’s perspective. It is an increasing concern in the evaluation of energy efficiency subsidies (Grösche and Vance, 2009; Boomhower and Davis, 2014). Third, public policies have general equilibrium effects. Yet for practical reasons, they are mostly evaluated in partial equilibrium frameworks. The opportunity cost of public funds can be factored in the analysis to bridge the gap between the two approaches (as done by the Cour des Comptes for many public policies in France).

The output of the literature review will help draw an evaluation framework applicable to evaluate energy efficiency policies. For example, we will consider the level and cost of carbon dioxide abatement as a cost-effectiveness criterion. Some policies may be implemented with the explicit purpose of meeting a quantitative carbon dioxide abatement target, even though a normative economic analysis would find other justifications for implementation (e.g. energy efficiency market failures). In any case, carbon dioxide abatement cost-effectiveness can be computed, according to which policies can be ranked. One should bear in mind, however, that such a criterion only offers a partial view of the overall policy performance.