THE IMPACT OF INCREASED ENERGY EFFICIENCY: A COMPUTABLE GENERAL
EQUILIBRIUM ANALYSIS FOR SPAIN
Pablo Arocena, Universidad Pública de Navarra (Spain), +34948169684, pablo@unavarra.es
Antonio Gómez, Universidad Pública de Navarra (Spain), +34948169348, agomezgp@unavarra.es
Sofía Peña, Universidad Pública de Navarra (Spain), +34948169333, sofia.pena@unavarra.es
Overview
The improvement of energy efficiency is one of the pillars of the energy policies in most countries, and constitutes
one of the cornerstones of the European Union’s 2020 strategy (EC, 2010). Environmental considerations are not the
only force pushing towards more efficient use of energy. In many advanced economies there is widespread concern
about energy dependence, while the dramatic increase of energy prices over the last decade has led energy costs to
occupy a prominent place in the debate on the drivers of industrial competitiveness (e.g. IEA 2013). In Spain, with
energy intensity rates and energy prices among the highest in the European Union (e.g., Mendiluce et al 2010), the
need to reduce energy consumption has come to occupy a central role in the political agenda. This is the objective of
the national strategy for energy conservation and energy efficiency promoted by the Spanish government (MITYC
2011).
The impact of the increase of energy efficiency on the economy is not limited to energy consumption but also to
other variables (e.g. employment, GDP). Further, the expected energy savings resulting from the improvement in the
efficiency in the use of energy may not correspond to actual energy savings, because of the rebound effect (e.g.
Sorrell, 2009). That is, the possibility that improving energy efficiency could lead to reductions in energy
consumption lower than expected, or even increases in consumption. To estimate the economy-wide effect of an
increase of energy efficiency a computable general equilibrium (CGE) analysis is required (e.g. Guerra & Sancho,
2010; Wei, 2010). This paper analyzes these issues in the Spanish economy through a CGE model using the InputOutput Framework of the Spanish economy for the year 2005. Our model allows quatifying in which sectors and
what types of energy is more likely to produce the desired effect, or what consequences might arise from an
improvement in energy efficiency over other variables such as employment, prices or GDP.
Methods
The model we use is a static CGE model (Shoven and Whalley, 1984) that describes an open economy,
disaggregated into 27 production sectors, with 27 consumer goods, a representative consumer, the public sector and
a simplified rest of the world. It allows performing a general equilibrium comparative static analysis.
Previous esearch has shown that six CGE model features are important for the resulting estimates of the impact of
efficiency improvements on energy use (Allan et al, 2007). (1) Treatment of energy in the production function: we
base the nestings on MIT-EPPA, which is backed with econometric estimates (see Paltsev et al., 2005). (2)
Sensitivity of results to the elasticities of substitution with energy in production: we perform a deep sensitivity
analysis on them in section 6. (3) Capital closure: we perform a sensitivity analysis on different capital closure rules
in section 6. (4) Treatment of the labour market: Unlike similar models, this model has the particular feature of
including unemployment as a specification derived from the literature of trade unions models, given the high
unemployment rate of the Spanish economy. Additionally, we perform the sensitivity analysis of different
assumptions on wage flexibility and substitutability in section 6. (5) The role of increased government revenue from
increased economic activity: We apply a revenue neutral rule, isolating the role of the public sector in the model. (6)
The modelling of the energy efficiency improvement: we apply an autonomous energy efficiency improvement at
different quantitative levels, to check the robustness of the results.
The model is calibrated using data for the Spanish economy. The calibration of benchmark equilibrium is
represented by the National Accounts data, and is reflected in the Social Accounting Matrix (SAM) with a set of
elasticities taken from the available empirical evidence. A detailed explanation of the calibration technique used can
be found, e.g., in Mansur and Whalley (1984) or in Dawkins et al. (2001).
The simulations consist in improving the productivity of energy-related inputs. Specifically, it is simulated a
reduction of the use of 5 energy inputs: electricity, natural gas, coal, petroleum and oil fuels (simulteously and
individually) per unit of output produced.
Results
(i) Macroeconomic results
(ii) Microeconomic results
Base case: 5% energy improvement in all energy inputs
Coal
Oil
Refining
Energy consumption
-5.8%
-0.1%
4.9
Rebound effect
-0.16%
98.1%
>100%
Electricity
-1.8%
64.1%
Gas
-2.8%
43.7%
(iii) Sensitivity analysis
- Results at sectoral level are sensible to functional forms, specifically for production functions.
However, this is not observed at macroeconomic level.
- Different elasticities of substitution between capital and labour do not affect the results.
- Specific capital instead of perfect capital mobility does not seem to affect the results.
- Changes in the flexibility/rigidity of wages do not affect the results.
Conclusions
An improvement in efficiency of 5% in the use of all energy inputs would have the following effects:
- A decrease in the total consumption of coal, oil, electricity and gas (positive rebound effect in the case of
oil, electricity and natural gas and negative for coal).
- An increase in the consumption of refining (petroleum products), precluding any saving (backfire effect).
- An increase in the welfare of about 0.8% and a reduction in the unemployment rate of around 6%.
- It also produces an increase in both real wage and rents of capital, although the increase of real wages is
lower than the increase of rents of capital.
References
Allan G., Hanley N., McGregor P., Swales K., Turner K. (2007) The impact of increased efficiency in the industrial
use of energy: a computable general equilibrium analysis for the United Kingdom. Energy Economics 29, 779-98
Dawkins, C., Srinivasan, T.N. and Whalley, J. (2001) Calibration, in J.J. Heckman and E.E. Leamer (ed.) Handbook
of Econometrics, volume 5, chapter 58, 3653-3703, Elsevier, Amsterdam.
EC (2010) Europe 2020. A strategy for smart, sustainable and inclusive growth. European Commission.
Guerra, A. and Sancho, F. (2010), Rethinking economy-wide rebound measures: an unbiased proposal, Energy
Policy, 38, 6684-6694.
Mansur, A. and Whalley, J. (1984) Numerical specification of applied general equilibrium models: Estimation,
calibration, and data, in H. E. Scarf y J. B. Shoven (eds.), Applied general equilibrium analysis, Cambridge:
Cambridge University Press: 69-127.
MITYC (2011) Plan de Ahorro y Eficiencia Energética en España 2011-2020. Instituto para la Diversificación y
Ahorro de la Energía. Ministerio de Industria, Turismo y Comercio, Madrid.
Paltsev, S., J.M. Reilly, H.D. Jacoby, R.S. Eckaus, J. McFarland, M. Sarofim, M. Asadoorian and M. Babiker (2005)
The MIT Emissions Prediction and Policy Analysis (EPPA) Model:Version 4, MIT Joint Program Report 125.
Shoven, J.B. and Whalley, J. (1984) Applied General-Equilibrium models of taxation and international trade: an
introduction and survey. Journal of Economic Literature 22 (3), 1007-1051.
Sorrell, S. (2009) The rebound effect: definition and estimation, in Evans J. and L. Hunt (eds.) International
handbook on the economics of energy, chapter 9, 199-233.
Wei, T. (2010) A general equilibrium view of global rebound effects. Energy Economics 32, 661-672.
IEA (2013) World Energy Outlook 2013. Paris: International Energy Agency.