Elsevier

Energy Strategy Reviews

Volume 2, Issue 2, September 2013, Pages 146-152
Energy Strategy Reviews

Analysis
Economic development, energy market integration and energy demand: Implications for East Asia

https://doi.org/10.1016/j.esr.2012.12.011Get rights and content

Abstract

This paper uses a general method of moment regression technique to estimate an energy demand function with a dataset covering 71 countries between 1965 and 2010. The estimated results show that countries undergoing rapid economic growth may show relatively higher income and price elasticities in the long run. The higher income elasticities and lower price elasticity in the short run of rapid growing countries may impose pressure on energy demand in the domestic and international markets. Energy market integration can help to reduce such pressure by smoothing energy demand through lowering its income elasticity and creating a flexible energy market through increasing its price elasticity. These findings have important implications for forecasting energy demand and promoting international cooperation in East Asia.

Highlights

▶ Higher income elasticity of rapid growing countries creates a shock to demand. ▶ Price elasticity of rapid growing countries will increase over time. ▶ EMI smoothes energy demand through lowering its income elasticity. ▶ EMI creates a flexible energy market through increasing its price elasticity. ▶ These findings have implications on energy outlooks and EMI.

Introduction

Price and income are two primary factors shaping the energy consumption of a country, and estimation of the elasticities of these two factors is thus essential for defining the energy demand function. In previous empirical studies [1], [2], [3], [4], [5], the income and price elasticities of energy products have been widely estimated with either single-country time-series data or cross-country data. However, there is no general agreement on representative values of the income and price elasticities, and in particular it is still not clear why the magnitude of these elasticities may differ across countries with disparate economic development levels and institutional arrangements.

Using a comprehensive survey of quantitative studies on country-specific energy consumption, Dahl [6] shows that the demand for energy is price inelastic and slightly income elastic at the national level but there is no clear evidence that the developing world should be less price elastic or more income elastic than the industrialized world. In contrast, Brenton [5] and Ferguson et al. [2], when using cross-country energy consumption data to estimate different energy demand functions, find that the price elasticity for energy is usually higher in the poor countries than in the rich countries, and that the income elasticity for energy declines with rising income.

To explain the above inconsistent findings on the estimated energy demand elasticities, many studies including Maddala et al. [7], Garcia-Cerrutti [8], Lowe [9], Bernstein & Griffin [10], and Yoo [11] attempt to incorporate regional characteristics, such as country-specific energy consumption preferences and different energy-usage technology in production, into the estimation of the energy consumption function. These studies provided some interesting results with respect to the relationship between economic growth, policy making and energy consumption by providing new estimates of the cross-country income and price elasticities for energy. However, they could not explain two important phenomena [10]: (1) estimated energy consumptions using the cross-country data generally lack price elasticity, and they are significantly different from those in country-specific studies; (2) estimated energy consumptions using the single-country data usually show different trends over different time periods.

These two phenomena raise the interesting question of whether economic development and the institutional arrangements associated with the energy market, which are the two important features associated with the different phases of economic development and income levels of a country, can be identified as affecting the income- or price-energy consumption relationships.

This paper attempts to measure the income and price elasticities of energy consumption in a country and to link these elasticities to the country's economic development and institutional arrangements related to Energy Market Integration (EMI). Contributing to the previous literature examining the relationship between energy consumption and income, such as Asafu-Adjaye [12] and Lee and Chang [13], our study is the first to highlight that the economic development stage and its related industrialization and urbanization may reshape the energy consumption behavior of a country by affecting its income and price elasticities. In addition, we also demonstrate the role that EMI may play in changing a country's energy demand when the country passes through different economic development stages.

Implications from this study shed light on two policy issues in the East Asian Summit (EAS) region. The first policy issue is that since many EAS countries are less developed and will industrialize in the future, a projection of the relationship between energy demand and industrialization is critical in choosing how to deal with the potential energy supply challenge. The second policy issue is that the study provides the method for valuing the costs of, and benefits from, participating in EMI. An incentive for EAS countries to participate in EMI is that regional integration may help to secure the energy supply for sustainable economic growth whilst reducing income disparity in the region. However, to what extent this goal can be achieved and how much benefits each country can obtain from participating in EMI depend on the impact of regional integration on the income and price elasticities of energy demand.

The paper is organized as follows. Section 2 summarizes the structural change in energy demand of some major countries. The analysis shows that there are often significant structural changes in energy demand when a country moves from a lower economic-development stage to a higher one, and different institutional arrangements associated with the energy market may impose different impacts on such structural changes in energy demand. Section 3 develops a dynamic panel data model, which incorporates economic growth stages and EMI into the estimation of the energy demand function. Section 4 presents the estimated results which show that countries in different stages of economic development, and with different involvement in EMI, would demonstrate different levels of demand for energy consumption. Section 5 presents conclusions and policy implications for East Asia.

Section snippets

Changing world energy demand and its determinants

The world's demand for energy has experienced rapid growth over the past five decades, despite a slight drop due to two supply shocks in 1973 and 1979. Up to 2010, the total world energy demand had reached 12.0 billion tonnes of oil equivalent (toe) which is 3.2 times the level in 1965 (3.8 billion toe) [14]. Behind the steadily increasing trend of world energy demand, countries with different levels of economic development have demonstrated different energy demand patterns. Three

Methodology and data

As shown in Section 2, it is necessary to incorporate economic development stages and EMI into the estimation of the energy consumption function. Based on a standard consumption function, we assume that the energy demand is determined not only by changes in income and price but that it also varies with different economic development stages and institutional arrangements with which a particular country is associated. The energy demand function in double-log form for panel data regression can

Empirical results

Following Roodman [25], we specify suitable instrumental variables from the lagged or differentiated dependent and independent variables and use the difference GMM methods to investigate the relationship between integrated series with dynamic panel data. The Johansen (1991) order test is employed for this purpose.

Before estimating any relationship between energy consumption and its explanatory variables, one may need some identification strategy either from an economic or a statistical

Conclusions and implications

This study uses a GMM regression technique to estimate a cross-country demand function for energy products, using 45-year long data from 71 countries, and examines the income and price elasticities of energy consumption between 1965 and 2010.

The results show that countries with different economic development stages demonstrate different levels of energy demand and thus the associated price and income elasticities. In particular, we found that countries undergoing rapid economic growth may show

Acknowledgments

This paper is one of the outcomes of the Economic Research Institute for ASEAN and East Asia (ERIA) project on ‘Energy Market Integration in East Asia: Theories, Electricity Sector and Subsidies’ in the financial year 2011. We thank ERIA for the financial support and the participants in the two project meetings as well as Fukunari Kimura for the very helpful comments.

References (29)

  • R. Blundell et al.

    Initial conditions and moment restrictions in dynamic panel data models

    Journal of Econometrics

    (1998)
  • D.W. Jones

    How urbannization affects energy-use in developing countries

    Energy Policy

    (1991)
  • D.R. Bohi et al.

    An updated econometric study of energy demand behaviour

    Annual Review of Energy

    (1984)
  • L.D. Taylor

    The demand for electricity: a survey

    Bell Journal of Economics

    (1975)
  • Cited by (25)

    • Analysis of the robustness of energy supply in Japan: Role of renewable energy

      2020, Energy Reports
      Citation Excerpt :

      In the year of 2010 and before the earthquake disaster nuclear energy, renewable energies and natural gas and oil accounted for 75.9%, 20.2% and 3.9% of the domestic energy production, respectively. However after since 2010, solar energy technologies, biofuels and waste have had a noticeable growth of 207.8% and 19.7%, respectively (IEA, 0000; Sheng et al., 2013). The situation of energy supply in Japan after the nuclear disaster in Fukushima on March 11, 2011 changed and Japanese government tried to substitute the loss of nuclear energy in power sector which was facing the –62.9% reduction in production rate with additional low Sulfur crude oil, natural gas and fuel oil (Fig. 3) (Energy White Paper, 2012).

    View all citing articles on Scopus
    View full text