Elsevier

Energy Policy

Volume 89, February 2016, Pages 114-124
Energy Policy

Analysis of the importance of structural change in non-energy intensive industry for prospective modelling: The French case

https://doi.org/10.1016/j.enpol.2015.11.014Get rights and content

Highlights

  • Defining NEI industries with a quantitative approach from relevant indicators in France.

  • Developing new decomposition method given in additive form with no residual in NEI.

  • Structural change is the overwhelming factor in improving energy performance within NEI.

  • Revealed consistent trend with level of sector disaggregation if homogeneous industries.

Abstract

A large number of studies have been conducted on the contribution of technological progress and structural change to the evolution of aggregate energy intensity in the industrial sector. However, no analyses have been done to examine theses changes in the non-energy intensive industry in France. We analyzed their importance in French industry with respect to their energy intensity, energy costs, value added, labour and the diffusion of production sites by using data at the 3-digit level with 236 sectors. Using a new decomposition method that gives no residual, this paper attempted to examine, over 10 years from 1996 to 2005, the changes that occurred in an area that has been neglected in energy analysis. We found that structural change had an overwhelming effect on the decline of aggregate energy intensity. Furthermore, we found that the higher the level of sector disaggregation, the more significant the changes that can be attributed to structural change, due to the homogeneity of this industrial group. The results of our study show that it is important to take into account the effects of structural change in “bottom-up” modelling exercises so as to improve the accuracy of energy demand forecasting for policy-makers and scientists.

Introduction

It is well known that change in aggregate energy intensity is caused by structural and technological effects. In recent years, a considerable number of analyses have studied their impacts on industrial energy intensity over time (Cortuk and Singh, 2011, Chen et al., 2011, Sauramo and Maliranta, 2011, Al-Salman, 2008, Ma and Stern, 2008, Chòliz and Duarte, 2006, Ramirez et al., 2005, Enevoldsen et al., 2007; Ussanarassamee and Bhattacharryya, 2005; Zhang, 2003; Fagerberg, 2000; Farla et al., 1998; Gardner and Elkhafif, 1998; Sinton and Levine, 1994; Gardner, 1993; Huang, 1993; Li et al., 1990; Ang, 1987; Boyd et al., 1987; Motamen and Schaller, 1985; Sterner, 1985; Jenne and Cattell, 1983). In France, the non-energy intensive industry (NEI) group represented almost 40% of total energy consumption in industry in 2005, compared to 30% for 1993. For both electricity and natural gas, we observe a respective increase in consumption from 45% to 55% and from 40% to 45% (Seck, 2012). Thus, NEIs possess a large share of the technical pool for energy savings.

Under the ESC mechanism (Energy Savings Certificate), energy operators are required to achieve a certain amount of energy savings per 3-year period. A failure to achieve these objectives incurs penalties. To avoid double counting with the European ETS (Emission Trading Scheme, or mechanism for trading emissions of CO2), the ESC cannot be applied to sites that come under the national quota allocation plan (PNAQ, which are generally the energy intensive industries (EIs)). NEIs are thus a priority target for energy operators. They are expected to play an important role in reducing the aggregate energy intensity of industry in the long term because of their economic importance (80% of value added in French industry) and have relatively high growth rate. However, these sectors have been neglected in energy analyses, despite the continuing policy focus on energy efficiency and the many reports and books written on the topic. This oversight encouraged us to do long-term modelling exercises with a bottom-up model like MARKAL/TIMES (The Integrated MARKAL-EFOM System developed by ETSAP (Energy Technology Systems Analysis Programme) under the aegis of the International Energy Agency (IEA)) (Seck, 2012; Seck et al., 2013, Seck et al., 2015; Djemaa, 2009; Gargiulo, 2009; Loulou and Labriet, 2007; Loulou et al., 2004, Loulou et al., 2005; Fishbone and Abilock, 1981, Fishbone et al., 1983).

These models will aim to inform energy companies on how much to invest in new technology so as to facilitate its incorporation and achieve the expected energy savings. Thus, it would be wise to analyze the importance of structural effects in energy efficiency studies in order to take them into account in modelling exercises and avoid skewed results. A deeper understanding of the sources of aggregate intensity change can help to improve the quality of energy demand forecasting. Energy demand forecasts that can explicitly take into account important effects of structural change will be more useful in policy making than those that cannot (Sinton and Levine, 1994). Past trends in aggregate energy intensity can help us better understand the current situation and have an influence on future decisions aimed at reducing energy use. This paper examines the relative contributions of technological and structural effects on change in aggregate energy intensity in the NEI group in France from 1996 to 2005. The paper is broken down as follows: Section 1 shows the level of disaggregation detail considered, the data and sources used in the analysis, along with a presentation of the decomposition method, which allows us to analyze the influence of structural and technological effects on changes in aggregate energy intensity; Section 2 deals with a way to distinguish between non-energy intensive and energy intensive industries in the industrial perimeter by using criteria which could explain their difference through a data mining approach; Section 3 presents the results of the analysis of the influence played by structural change on the evolution of the aggregate energy intensity.

Section snippets

Data and methodology

This article uses economic data published by EUROSTAT on the industrial sector (manufacturing industry) for France (value added, production value, energy costs). The data on energy consumption, employment and number of production sites came from the internal database ENERVISION of EDF R&D departments designed by EPI (Eco-Efficiency and Industrial Processes) and ICAME (Business Innovation, Market Analysis and their Environment) at the 4-digit level of NACE rev 1.1 (Statistical Classification of

Segmentation of the French industry

Each industry has its own structure and characteristics, which explains why in several articles; studies were conducted to distinguish segmentation into several families of an industrial perimeter. Most analyses distinguish them by considering some major energy-intensive industries and combining the rest into another group, or by basing their study on economic characteristics and/or energy in each sector. In their study of the Taiwanese manufacturing sector, Li et al. (1990) disaggregated the

The relative importance of structural and technological change

Fig. 6 below shows that the non-energy intensive industries have an important weight in the final energy consumption and value added of the industry in major European countries. They represent approximately 40% on average of the total final energy consumption for almost 75% of the total value added in the EU-27 industry. This weight of NEIs in final energy consumption is rather heterogeneous with values oscillating between 15% (as in Luxembourg with an industry which is dominated by iron and

Conclusion and policy implications

We can conclude that the structural effect was the overwhelming factor (75%) in improving energy performance in the period 1996–2005 within the non-energy intensive industry in France. Energy prices also played a stimulus role in enhancing that same energy performance, either through technical progress (energy efficiency) or through a structural effect with slower growth for some major sectors as we have seen. Note that technical progress was only a dominant factor in the NEI industry during

Acknowledgement

This research was supported by the Chair Modelling for sustainable development, driven by MINES ParisTech, Ecole des Ponts ParisTech, AgroParisTech, ParisTech, the FI3M and the Fondation de l'Ecole des Ponts ParisTech supported by ADEME, EDF, Renault, Schneider electric and Total.

References (81)

  • B.W. Ang et al.

    Handling zero values in the logarithmic mean Divisia index decomposition approach

    Energy Policy

    (2007)
  • B.W. Ang et al.

    Sector disaggregation and the effect of structural change on industrial energy consumption

    Energy

    (1992)
  • B.W. Ang et al.

    Factorizing changes in energy and environmental indicators through decomposition

    Energy

    (1998)
  • B.W. Ang et al.

    Perfect decomposition technique in energy and environmental analysis

    Energy Policy

    (2003)
  • B.W. Ang et al.

    A generalized Fisher index approach to energy decomposition analysis

    Energy Econ.

    (2004)
  • G.A. Boyd et al.

    Decomposition of changes in energy intensity – a comparison of the Divisia index and other methods

    Energy Econ.

    (1988)
  • S. Chen et al.

    Structural change, productivity growth and industrial transformation

    China Econ. Rev.

    (2011)
  • K.-H. Choi et al.

    Decomposition of aggregate energy intensity changes in two measures: ratio and difference

    Energy Econ.

    (2003)
  • K.-H. Choi et al.

    Attribution of changes in Divisia real energy intensity index – an extension to index decomposition analysis

    Energy Econ.

    (2012)
  • O. Cortuk et al.

    Structural change and growth in India

    Econ. Lett.

    (2011)
  • M. Enevoldsen et al.

    Decoupling of industrial energy consumption and CO2-emissions in energy-intensive industries in Scandinavia

    Energy Econ.

    (2007)
  • J. Fagerberg

    Technological progress, structural change and productivity growth: a comparative study

    Struct. Change Econ. Dyn.

    (2000)
  • J. Farla et al.

    Energy efficiency and structural change in the Netherlands 1980–1990

    Energy Econ.

    (1998)
  • S.L. Freeman et al.

    Measuring industrial energy intensity: practical issues and problems

    Energy Policy

    (1997)
  • D. Gardner

    Industrial energy use in Ontario from 1962 to 1984

    Energy Econ.

    (1993)
  • D.T. Gardner et al.

    Understanding industrial energy use: structural and energy intensity changes in Ontario industry

    Energy Econ.

    (1998)
  • R. Hoekstra et al.

    Comparing structural and index decomposition analysis

    Energy Econ.

    (2003)
  • J.-P. Huang

    Industry energy use and structural change: a case study of The People's Republic of China

    Energy Econ.

    (1993)
  • C.A. Jenne et al.

    Structural change and energy efficiency in industry

    Energy Econ.

    (1983)
  • S.S. Khan et al.

    Cluster center initialization algorithm for K-means clustering

    Pattern Recognit. Lett.

    (2004)
  • J.Z.C. Lai et al.

    A fast k-means clustering algorithm using cluster center displacement

    Pattern Recognit.

    (2009)
  • J.-W. Li et al.

    Structural change and energy use: the case of the manufacturing sector in Taïwan

    Energy Econ.

    (1990)
  • F.L. Liu et al.

    Eight decomposing the aggregate energy-intensity of industry

    Appl. Energy

    (2003)
  • N. Liu et al.

    Factors shaping aggregate energy intensity trend for industry: energy intensity versus product mix

    Energy Econ.

    (2007)
  • C. Ma et al.

    China's changing energy intensity trend: a decomposition analysis

    Energy Econ.

    (2008)
  • H. Motamen et al.

    Structural changes in the UK energy market

    Energy Policy

    (1985)
  • J.M. Peña et al.

    An empirical comparison of four initialization methods for the k-means algorithm

    Pattern Recognit. Lett.

    (1999)
  • C.A. Ramirez et al.

    The non-energy intensive manufacturing sector, an energy analysis relating to the Netherlands

    Energy

    (2005)
  • S.J. Redmond et al.

    A method for initialising the K-means clustering algorithm using Kd-trees

    Pattern Recognit. Lett.

    (2007)
  • P. Sauramo et al.

    Structural change in Finnish manufacturing: the theory of the aggregation of production functions and an empirical analysis with a plant-level panel

    Struct. Change Econ. Dyn.

    (2011)
  • Cited by (5)

    • Decomposing industrial CO<inf>2</inf> emissions of Southern European countries into production- and consumption-based driving factors

      2017, Journal of Cleaner Production
      Citation Excerpt :

      Another factor that contributed significantly to the reduction of CO2 emissions was the change in fuel mix, while structural changes in industry had a positive but marginal effect on CO2 emission reduction in most countries. On the other hand, Seck et al. (2016), who studied only the case of industry in France, showed that the effect of structural change at a 3-digit level of disaggregation of the industrial sectors was more important than the energy intensity effect on improving energy performance in the period 1996–2005. Still, structural changes and technological improvements in the above mentioned studies refer to the production side of an economic system and not to the consumption side.

    • Urban long term electricity demand forecast method based on system dynamics of the new economic normal: The case of Tianjin

      2017, Energy
      Citation Excerpt :

      The economic growth does not match the electricity demand growth, and this is associated with the entire urban industrial structure and profit levels [26]. The Solow model divides the changes in the electricity demand into three parts: economic growth, industrial structure and power consumption per unit output; the model assumes that considering the power consumption per unit output can fully embody the energy-saving effect [27]. In terms of the industrial structure, scholars have selected the non-energy-intensive industries as their study object [28] and explained the impact of structural changes and technological advances on energy consumption using the index decomposition method [29].

    View full text