ANALYSISThe rebound effect: Microeconomic definitions, limitations and extensions
Introduction
The rebound effect is the focus of a long-running dispute with energy economics. The question is whether economically worthwhile improvements in the technical efficiency of energy use can be expected to reduce aggregate energy consumption by the amount predicted by simple engineering calculations. For example, will a 20% improvement in the thermal efficiency of a heating system lead to a corresponding 20% reduction in energy consumption? Economic theory suggests that it will not. Three separate mechanisms may reduce the aggregate energy savings achieved (Greening et al., 2000):
- •
Direct rebound effects: Improved energy efficiency for a particular energy service will decrease the effective price of that service and should therefore lead to an increase in consumption of that service. This will tend to offset the reduction in energy consumption provided by the efficiency improvement
- •
Indirect effects: The lower effective price of the energy service may lead to changes in the demand for other goods, services and factors of production that also require energy for their provision. For example, the cost savings obtained from a more efficient central heating system may be put towards an overseas holiday.
- •
Economy wide effects: A fall in the real price of energy services may reduce the price of intermediate and final goods throughout the economy, leading to a series of price and quantity adjustments, with energy-intensive goods and sectors likely to gain at the expense of less energy-intensive ones.
Numerous empirical studies, principally from the US, suggest that these rebound effects are real and can be significant (Greening et al., 2000). However, while their basic mechanisms are widely accepted, their magnitude and importance are disputed. Some analysts argue that rebound effects are of minor importance for most energy services (Schipper and Grubb, 2000), while others argue that the economy-wide effects can be sufficiently important to completely offset the energy savings from improved energy efficiency (Brookes, 1990, Saunders, 1992). The policy implication is that non-price regulations to improve energy efficiency may neither reduce energy demand nor help to mitigate climate change.
Indirect and economy-wide rebound effects involve general equilibrium adjustments that are difficult to analyse empirically.1 In contrast, direct rebound effects can be investigated more directly through quasi-experimental studies or the econometric analysis of secondary data. However, such studies raise a number of definitional and methodological issues that are inadequately discussed in the literature. The disagreement over the magnitude and importance of the rebound effect may result in part from lack of clarity over these basic definitions and issues. Moreover, since many empirical studies overlook key methodological issues, their estimates of the rebound effect could potentially be biased.
This paper examines the definition and measurement of the direct rebound effect for individual energy services. Indirect and economy-wide effects are not discussed. The focus throughout is on energy efficiency improvements in consumer goods such as cars and central heating systems, since this is where the bulk of the empirical evidence lies. While analogous arguments apply to energy efficiency improvements by producers, the evidence here is weaker and harder to interpret (Greening and Greene, 1998).
The paper is structured as follows. First, we present a general ‘household production’ framework for characterising the demand for energy services that helps to illustrate the different trade-offs involved. Second, we show how the direct rebound effect can be represented as an efficiency elasticity of energy demand and how it may be decomposed into the sum of elasticities for the number, capacity and utilisation of energy conversion devices. Third, we show the relationship between the rebound effect and the price elasticity of the demand for ‘useful work’, as well as the price elasticity of the demand for energy, and show why empirical studies using these definitions provide a primary source of evidence for the direct rebound effect. We then expose the limitations of these definitions, focusing on: a) the potential correlation between various input costs and improvements in energy efficiency; b) the endogeneity of energy efficiency and the implied need for simultaneous equation estimation; and c) the role of time costs and time efficiency in the production and consumption of energy services. We identify some of the factors that need to be controlled for to obtain accurate estimates of the rebound effect and argue that the neglect of these factors by several existing studies may lead the rebound effect to be overestimated.
Section snippets
The demand for energy services
The demand for energy (E) derives from the demand for energy services (ES) such as thermal comfort, refrigeration and motive power. These services, in turn, are delivered through a combination of energy commodities and the associated energy systems, including energy conversion devices. Consumers are assumed to derive utility from consuming these services, rather than from consuming energy commodities and other market goods directly. In practice, nearly all services require energy in some form,
The rebound effect as an efficiency elasticity
The energy efficiency (ɛ) of an energy system may be defined as ɛ = S / E, where E represents the energy input required for a unit output of useful work (however measured).2
The rebound effect as a price elasticity
Since PS = PE / ɛ, raising (lowering) energy efficiency (ɛ) when energy prices (PE) are constant should have the same effect on the energy cost of useful work (PS) as falling (rising) energy prices when energy efficiency is constant. Under the ceteris-paribus assumptions given above, the effect on the total cost and hence the demand (S) for useful work should be symmetrical. If other inputs are held constant, we can write the demand for useful work solely as a function of energy prices and energy
Correlation between energy efficiency and other input costs
For an individual energy service, changes in energy commodity prices are unlikely to be correlated with changes in other input costs or with changes in the broader attributes of the energy service. But the same cannot be said about changes in energy efficiency. In many (although by no means all) cases, energy efficient conversion devices will have a higher capital cost than inefficient models (i.e. ɛ and K will be positively correlated). For example, UK building regulations now require high
Endogenous energy efficiency
Definitions 1 and 3 assume that energy efficiency is independent of the values of other independent variables — in other words, that it is exogenous. This follows naturally from Khazzoom's original focus on the effect of mandatory energy efficiency standards for household appliances. In practice, however, the level of energy efficiency is likely to be influenced by one or more of the other dependent variables — in other words, energy efficiency must be considered partly endogenous. In
Energy efficiency and time costs
The model summarised in Annex A is based upon Becker's work on the allocation of time within household production (Becker, 1965). As Binswanger (2001) has argued, time costs and the efficiency of time use have important implications for energy use in general and the rebound effect in particular. However, empirical work in this area remains in its infancy (Jalas, 2002).
For consumers, time is a necessary input to the production and enjoyment of energy services. For example, it takes time to drive
Summary
This paper has sought to clarify and bring together a number of definitions of the direct rebound effect and identify their underlying assumptions. It has clarified the relationship between the ‘engineering’ definition of the direct rebound effect as an efficiency elasticity and the more common definition in the economic literature as a price elasticity. It has discussed a number of factors that need to be taken into account when developing such empirical estimates and emphasised the trade-offs
Acknowledgements
This research was funded by the UK Research Councils as part of the Technology and Policy Assessment (TPA) function of the UK Energy Research Centre (UKERC). An earlier version of this paper was presented at the 29th IAEE International Conference ‘Securing energy in insecure times’, held in Potsdam, Germany on the 7–10th June 2006. The authors gratefully acknowledge comments from Harry Saunders, John Feather, Blake Alcott, Brenda Boardman, Mark Barrett and Ed Steinmuller as well as from
References (58)
Estimating the rebound effect in US manufacturing energy consumption
Energy Economics.
(2004)- et al.
Defining the rebound effect
Energy Policy
(2000) Technological progress and sustainable development: what about the rebound effect?
Ecological Economics
(2001)The greenhouse effect: the fallacies in the energy efficiency solution
Energy Policy
(1990)- et al.
Analyzing gasoline demand elasticities — a survey
Energy Economics
(1991) - et al.
The imperfect price irreversibility of non-transportation of all demand in the OECD
Energy Economics
(1995) Gasoline demand revisited: an international meta-analysis of elasticities
Energy Economics
(1998)- et al.
A general equilibrium assessment of rebound effects
Energy Economics.
(2004) - et al.
Residential energy demand in OECD-countries and the role of irreversible efficiency improvements
Energy Economics
(1998) - et al.
The impact of consumer behavior on residential energy demand for space heating
Energy and Buildings
(1998)