ارزیابی تعادل عمومی از اثرات ریباند
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|28588||2004||22 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Energy Economics, Volume 26, Issue 2, March 2004, Pages 261–282
In this paper we use a general equilibrium model applied to a national economy (Norway) to explore the potential for energy efficiency improvements to trigger economic forces that offset potential savings from using more efficient technologies (rebound effects). Two types of energy efficiency improvements (electricity and oil) are introduced into various sectors of the economy. Our results suggest significant and surprising differences across sectors concerning both energy use and consequences for the build-up of greenhouse gases. Rebound effects are found to be quite significant for manufacturing sectors since long-term consumption of energy undergoes minor reductions or increases in response to efficiency improvements. In other sectors, rebound effects appear to be weak or almost absent.
The application of efficiency improvements, in order to achieve reductions in pollution and resource consumption, has been on the political agenda since the early 1970s and is now frequently suggested as a measure towards the realisation of a sustainable development (see e.g. World Commission on Environment and Development, 1986, United Nations, 1995, Organisation for Economic Co-operation and Development, 1995 and Organisation for Economic Co-operation and Development, 1998). Recent advocates of efficiency improvements have also introduced new concepts. One example is Eco-efficiency, proposed by the World Business Council for Sustainable Development (1999), used about measures that reduce ecological impacts and resource intensity throughout the life cycle of goods and services. A second example is Factor 10, which appeared in ‘The Carnoules Declaration’ and called for an increase in current resource productivity by an average factor of 10 during the next 30–50 years, in order to reach sustainability (Wuppertal Institute, 1994). While emphasising the importance of efficiency improvements, the above literature ignores the possibility of ‘take-back’ or rebound effects. Rebound effects can be defined as economic forces (demand side effects) that over time weaken the potential (technical) savings associated with efficiency improvements. One important cause of such effects is that higher efficiency reduces energy costs, which again increases demand. Khazzoom, 1980, Khazzoom, 1986, Khazzoom, 1987, Khazzoom, 1989 and Khazzoom et al., 1990 all discuss the significance of such effects. Khazzoom questions the adequacy of energy savings programs since greater efficiency could lead to increased, rather than decreased energy demand.1Khazzoom (1987) also presents a critique of Lovins (1985) for ignoring rebound effects when savings from more efficient mandated appliances were assessed. This again triggered a debate on the importance of rebound-effects (see e.g. Lovins, 1988, Henly et al., 1988 and Khazzoom, 1989). The controversy reappeared a few years later—now in the context of fossil fuel consumption and greenhouse gas abatement. A forerunner to this debate was a work by Manne and Richels (1990), which analysed the economic costs arising from CO2-emission limits. This study showed that the autonomous energy efficiency index (AEEI) had a dramatic impact on the economic cost of reducing CO2-emissions.2 The same study found that a higher value on the AEEI (higher energy efficiency) would reduce both energy use and greenhouse gas emissions. Brookes, 1990 and Greenhalgh, 1990 nonetheless believe that widespread improvements in energy efficiency will not by themselves do anything to halt the build-up of greenhouse gases globally. Reductions in the energy intensity of output are associated with increases, rather than decreases in energy demand. Consequently, Brookes (1990) considers efficiency improvements to be an inappropriate way of combating the greenhouse effect. Grubb (1990), however, has challenged his arguments. Saunders (1992) brought the debate a step further by applying Neo-classical theory in the analysis of the Khazzoom–Brookes’ postulate which suggests that ‘energy efficiency improvements might increase rather than decrease energy consumption’.3 Calculations were undertaken in a one-sector growth model with three factors of production (labour, energy and capital) using Cobb-Douglas and nested CES production functions. Saunders (1992) conducted simulations under the assumption of a fixed real energy price and found that efficiency improvements increased the consumption of the resource that was exposed to higher efficiency rates as well as the two additional factors of production. The only exception concerned a nested CES technology given the elasticity of substitution being equal to 0.5. In this case, capital and labour combine in a Cobb-Douglas fashion, while the two together combines in a CES fashion with energy. As a consequence, Saunders (1992) concludes that the Khazzoom–Brookes’ postulate holds regardless with Cobb-Douglas, and with nested CES it holds under most conditions.4 Demand studies using micro-data are also present in the literature. Studies on household electricity demand identify rebound effects ranging from 2 to 13% (e.g. Dubin et al., 1986). In a study using US data from a residential conservation program (electric household appliances) it was found that approximately 60–70% of the initial savings were eroded by the rebound effect (Khazzoom 1986). Similarly, Haugland (1996) refers to Norwegian survey data from 1990 describing the outcomes of programs launched to reduce oil consumption, which suggest rebound effects amounting to approximately 40% for households and 10% for commerce. Econometric analyses on the US transport sector have also been undertaken. Greene (1992) studies light-duty vehicle miles during the period of 1966–1989. His findings suggest a rather small rebound effect, approximately 5–15% or less. The short-run (1-year) adjustments accounted for essentially all of the change in journeys due to fuel price and fuel economy changes. This result is supported by previous studies by Mayo and Mathis, 1988 and Gately, 1990. Jones (1993) undertook an analysis using the same data as Greene (1992), but chose model specifications better able to capture long-term effects. The results showed somewhat more significant rebound effects (∼30%). The above discussion illuminates strong differences between the neo-classical growth approach and econometric analyses at sector level with respect to the magnitude of rebound effects. In this study an attempt is made to fill in some gaps between the two approaches by applying a macro-economic analysis of rebound effects at industry level. This is done by using an applied general equilibrium model that enables us to conduct analyses at sector level, while at the same time avoiding drawbacks usually associated with single-equation estimations. The consequences of efficiency improvements for various sectors can now be studied with complete sets of equations and without violating consistency demands for example with respect to budget constraints. A general equilibrium framework also allows for an examination of the overall long-term effect of efficiency improvements. As pointed out in the literature, rebound effects do not only arise from substitution and income effects generated by changes in the effective price (implicit price) of a resource, but may follow from consumption ramifications extending economy-wide. Input–output relationships, cross-price effects and income elasticities are all important for sectoral complementarities and backward–forward linkages, and such effects may well be important. The analysis is undertaken in a multi-sectoral model describing the Norwegian economy, thus we go beyond aggregate data and a stronger empirical foundation is provided than is the case for Saunders (1992). Our approach to CGE modelling is econometric, which constitutes a sounder empirical basis than for example calibrated approaches. A general problem with analyses applying rich (complex) general equilibrium models is to provide precise explanations for the results arrived at.5 However, by analysing efficiency improvements in one sector at a time it becomes easier to track down important causal factors. We follow the literature on rebound effects by interpreting efficiency improvements as exogenous factor productivity changes. Consequently, efficiency improvements are not policy-driven, but can be interpreted as following general economic trends. Policy-driven efficiency improvements, however, should include an explicit modelling of relationships between public funding and/or public-induced price changes and innovation processes. Such relationships are indeed complex, but if introduced, would generate effects beyond those present in our analysis. Furthermore, we ignore any change in costs that are normally associated with changes in efficiency improvements, such as installation and operating costs.6 Our main focus is on the impacts of energy efficiency improvements. We study such effects for six different sectors of the Norwegian economy, and two different energy sources (electricity and oil). Our main concern is about changes in the demand for energy, however, as a reflection of the concern in the literature, we also report on CO2-emissions. In contrast to Saunders (1992) we allow for efficiency improvements to influence all prices in the economy including energy prices. The first section of this paper contains a brief outline of the general equilibrium model. Thereafter, descriptions of the baseline scenario (BS) together with an explanation of how energy efficiency improvements are interpreted in our model are given. Finally the findings arrived at are presented and compared.
نتیجه گیری انگلیسی
This paper examines the role of rebound effects for the Norwegian economy. The tool of analysis is a multisector computable general equilibrium model. Efficiency improvements at sectoral level are found to reduce the long-term consumption of the energy source that is being exposed to the improvements in five out of six sectors considered. Our findings contrast with those of Saunders (1992), where the demand for energy under most conditions was found to increase in response to energy improvements. Saunders (1992) conducted his analysis by undertaking numerical simulations in a one-sector Neo-classical growth model given fixed energy prices and a nested CES technology with substitution elasticities higher than one. One may argue that the explanation for the opposing results lies with the difference in aggregation level across the two studies. In MSG-6, substitution elasticities at the industry level are in general between 0 and 1, while Saunders (1992) derives his results from a one-sector model of the economy that makes assumptions about high substitution possibilities adequate. However, the above explanation is a too simple one. Our analysis has shown that significant rebound effects are at work in industries with limited substitution possibilities (metals) and that rebound effects can be weak in spite of a flexible technology (chemical and mineral products). The degree of substitutability among inputs (or aggregation level) is important but not essential for the results arrived at. An equally important factor is the degree to which the activity level in a sector is affected by efficiency improvements. The rebound that originates from this activity effect is not given sufficient attention in the literature. This ‘effect’ is a result of changes in costs which occur in response to efficiency improvements which again depend on factor-intensities, cost shares, and market structure—all factors which determine a sector's ability to take full advantage of a given efficiency improvement. Another important causal factor to our results is the treatment of energy as two substitutable sources of energy instead of as one aggregate production factor. The results show that there are substantial variations across sectors with respect to the energy source that is not exposed to energy efficiency improvements. In addition, a CGE model approach introduces other effects beyond those appearing in a one-sector model. The significance of such factors, however, is quite difficult both to trace and to quantify in rich CGE models. The results show that efficiency gains have interesting, non-intuitive, and maybe provocative impacts on energy consumption and carbon emissions. Although the analysis is related to Norway which have an unusual character of energy supply, since electricity is virtually all produced by (non-carbon-emitting) hydropower, the analysis have wider relevance. Many economies base their energy consumption on several energy sources with different carbon content, examples are Sweden (hydropower, nuclear energy and oils), France (nuclear energy, oil, natural gas), Germany and Denmark (oil, gas and coal). The results arrived at are partly determined by the role of inter-sectoral shifts in demand due to improvements in energy efficiency in one particular sector. An extension of our analysis would be to introduce efficiency improvements simultaneously in all sectors of the economy. As a consequence, consumption would not be shifted so much from one sector to the other—an effect that, ceteris paribus, could weaken rebound effects in each sector analysed. However, it is not obvious that such a policy would produce a long-term decline in total energy consumption. First, because an economy-wide electricity improvement will have a stronger effect on energy prices. This is an effect that pulls in favour of stronger rebound effects. In addition, we need to know how the remaining sectors will respond to efficiency improvements. The overall effect will depend on the importance of rebound effects in all sectors of the economy, as well as of their relative size of the national economy. A sector of particular interest is the household sector due to its considerable size. A Norwegian demand study suggests that rebound effects are stronger for households than for commerce (Haugland, 1996), if this is the case, economy-wide efficiency improvements in Norway may induce an increase in the long-term consumption of electricity. It is important to be aware of the limitations associated with the approach chosen in this analysis. First, our study does not meet all of the criticism of macroeconomic approaches presented by Lovins, 1988 and Grubb, 1990. Our predictions are still based upon observed historical relationships. Nonetheless, our results do show that rebound effects may be absent or weak even if macroeconomic considerations are introduced. Second, we ignore any costs that may arise from efficiency improvements. In reality, efficiency improvements, either generated by institutional, organisational or technological changes, are outcomes of processes with cost implications. This applies also for innovations occurring in response to governmental programs. Funding of such initiatives must be raised either by taxation of the private sector or at the expense of other public activities. Third, efficiency improvements may themselves have an impact on future substitution possibilities. Fourth, technology improvements are not likely to be energy-specific. Fifth, innovation structure may be more important than the sector in which they occur. The absence of rebound effects does not necessarily provide a rationale for governments to promote efficiency improvements. Whether policy induced innovations is desirable from a societal point of view will in general depend on the presence of market imperfections. If, however, such policies are implemented, it does not necessarily follow that they should be directed at sectors with weak rebound effects. In addition it is important to consider the costs associated with encouraging such policies. Cost–benefit analysis is a tool that can aid in identifying sectors with the highest net social returns. However, an understanding of the causal factors behind rebound effects is important and should be an integrated part of any cost–benefit analysis where resource savings is the policy goal.