Defining ‘total energy use’ in economic studies: does the aggregation approach matter?

doi:10.1016/S0140-9883(99)00014-6

Energy Economics

Volume 21, Issue 5, 1 October 1999, Pages 485-492

https://doi.org/10.1016/S0140-9883(99)00014-6 Get rights and content

Abstract

In economic models, the aggregate consumption of energy resources is normally expressed either in terms of its total heating value (e.g. Btus or barrels of oil equivalent) or in terms of its economic value (e.g. Divisia indices or total expenditures). For most major OECD countries, we find that it matters little whether various energy resources consumed in the industrial sector are aggregated in terms of their heating value or their economic value — similar trends emerge regardless of which measure is adopted. Consequently, as a practical matter, macroeconomic and energy modeling efforts need not be too concerned about which measure of total energy usage is adopted. The prominent exceptions are the United States and Mexico, where measures of total industrial sector energy use based on heating values and economic values diverge greatly due to some unique trends in electrification and energy prices. Consequently, attention to aggregation issues may be an important consideration in efforts to model the energy consumed in the US or Mexico.

Introduction

It is often debated how the total quantity of energy resources produced or consumed in a nation or region should be measured for use as a variable in economic models. Most modeling efforts rely on measures of total energy resources in which various types of energy resources (e.g. electricity, natural gas, coal, and oil) are aggregated based on their heating value, in such units as British thermal units (Btu) or barrels of oil equivalent (boe). In this paper, we shall refer to these measures as heating value aggregates. Such measures are readily available from government sources.

Beginning with Berndt (1985), many economists have pointed out the theoretical flaws inherent in aggregating different types of energy resources based on a heating value index and have suggested alternatives, such as Divisia energy aggregates which combine different resources based on their economic value. Zarnikau et al. (1996) trace the microeconomic foundations of the Divisia energy aggregate and discuss how the form-value attributes of different energy resources are reflected in Divisia aggregation approaches.

Empirical studies may be very sensitive to the means used to aggregate different energy resources. Hong (1983) demonstrated how the Divisia and Btu indices lead to very different conclusions regarding trends in energy-output ratios for the US economy. Nguyen and Andrews (1989) demonstrated the superiority of a Divisia aggregate over a Btu aggregate in estimating elasticities of demand for factor inputs in manufacturing in the US. Zarnikau (1996) reported that findings regarding the direction of causality between energy consumption and GNP in the US economy are quite sensitive to the manner in which various energy resources are aggregated. Zarnikau and Gupta (1997) showed that heating value energy aggregates tend to overstate the gains in energy efficiency achieved in various states of the US, relative to Divisia energy aggregates. Zarnikau (1999) demonstrated that the energy efficiency indices recently proposed by the US Department of Energy will tend to overstate energy efficiency achievements if they are based on heating value energy aggregates.

Each of the empirical studies cited above relied on data pertaining to national or regional energy resource usage in the US. In recent decades, Divisia energy aggregates for US industrial and residential energy consumption have grown much faster than heating value energy aggregates. This divergence is due to increasing electrification and price-related factors. Consequently, the sensitivity of econometric results to the choice of aggregation approach is not surprising.

Paralleling the debate in energy economics over the appropriate means of aggregating energy resources are similar debates in other branches of economics. In the field of monetary theory, economists have extensively studied alternative means of aggregating currency, savings deposits, money market accounts, and other financial assets into a single monetary aggregate (for a survey of this literature, see Barnett et al., 1992). Agricultural economists have debated whether to add different types of food by calories, weight, or economic value to derive an estimate of total food consumption or production. The Commission recently established to examine estimation error in the US consumer price index devoted considerable attention to appropriate means of aggregating all of the consumer products offered for sale in the country.

This paper constructs Divisia energy aggregates to measure industrial energy consumption in various OECD countries. We explore whether the divergence between heating value energy aggregates and Divisia energy aggregates that has been found for the US has also been experienced in other major OECD countries. We come to the interesting conclusion that this divergence seems to be unique to the US and Mexico. For other nations studied, Divisia energy aggregates and heating value energy aggregates follow very similar patterns. Some explanation is offered for this finding.

Section snippets

Deficiencies in heating value energy aggregates

To measure the total consumption of energy resources in a nation, region, process, firm, or household requires a means of adding up different types of energy resources (e.g. electricity, coal, oil, and natural gas). The heating content of various resources, usually expressed in Btu or boe, is often used for this purpose. There are a number of well-known limitations to aggregating different energy resources based upon their heating potential. A heating value energy aggregate ignores form-value

The Divisia energy aggregate

Recognition of the problems inherent with heating value energy aggregates has led to the introduction of Divisia energy aggregates. These indices combine different energy resources based on their market value, thus indirectly adjusting for differences in the form-value attributes of different energy resources. The resulting index provides a measure of the trend in the aggregate quantity of energy resources over time. In constructing the aggregate, the component energy resources are weighted

Measures of national energy consumption

While the Divisia aggregation approach enjoys a firm theoretical foundation, Divisia energy aggregates do require considerably more data and effort to construct than heating value aggregates. Here we explore whether the extra effort is really necessary if an analyst’s goal is to econometrically estimate relationships between the total demand for energy resources and other economic and demographic factors.

For major OECD countries, Divisia energy aggregates were constructed to measure total

Implications

Theory should guide the selection of an appropriate aggregation method. However, as a practical matter, the results of heating value and Divisia aggregation approaches may be so similar that the approach matters little. The comparison of heating value and Divisia energy aggregates for industrial energy consumption in Australia, Canada, Japan, Italy, the United Kingdom, France, and Germany suggests that econometric results are not likely to be greatly affected by the selection of aggregation

References (11)

W.E. Diewert
Exact and superlative index numbers
J. Econom.
(1976)
J. Zarnikau et al.
Can different energy resources be added or compared
Energy Int. J.
(1996)
W. Barnett et al.
Consumer theory and the demand for money
J. Econ. Lit.
(1992)
E. Berndt
Aggregate energy, efficiency, and productivity measurement
Annu. Rev. Energy.
(1985)
N. Hong
Two measures of aggregate energy production elasticities
Energy J.
(1983)

There are more references available in the full text version of this article.

Cited by (12)

Has energy efficiency performance improved in China?—non-energy sectors evidence from sequenced hybrid energy use tables
2017, Energy Economics
Citation Excerpt :
The bigger this value, the more pronounced the restriction. Zarnikau (1999a, 1999b) pointed out that, when economic units were used to measure energy consumption, the price of energy was taken into consideration. MEI represents a clear economic criterion to economic entities, which can directly trigger energy conservation, energy substitutes, and energy efficiency within the context of a market economy.
We conduct a comparative analysis of two energy efficiency indicators for China: heating value energy intensity (HEI) and economic value energy intensity (MEI). We formulate 1997–2002–2007–2012 hybrid energy comparable sequence use tables in an input-output accounting framework, and compare the two indicators using a randomized block ANOVA. The results show that MEIs and HEIs have significantly different variability patterns among sectors and are evolutionarily divergent over time. The directional changes in MEI and HEI are found to be inconsistent at both the sectoral and national levels. A further analysis with a LMDI index decomposition model shows that the difference between HEI LMDI and HEI LMDI is principally caused by energy prices. Based on the evidence from the two indicators and their relationship to energy prices, we are unconvinced about China's purported improvements in energy efficiency in recent years.
Explaining the (non-) causality between energy and economic growth in the U.S. -A multivariate sectoral analysis
2012, Energy Economics
Citation Excerpt :
The problem is that the weight of each energy source critically depends on the ‘correct’ price, which may not correctly reflect the marginal product of each energy source (Kaufmann, 1994). Hong (1983) and Zarnikau (1996, 1999) demonstrated that the application of the Divisia index and thermal equivalents leads to different conclusions regarding trends in energy-output ratios for the U.S. economy. Divisia energy indices for U.S. industrial and residential energy consumption have grown much faster than heating value energy aggregates.
The rapidly growing literature on the relationship between energy consumption and economic growth has not univocally identified the causal relationship yet. We argue that bivariate models, which analyze the causality only at the macro level, are eventually misleading, especially in cases where both variables do not cover the same scope of economic activity. After discussing appropriate pairs of variables, we investigate Granger causality between energy and growth in the U.S. for the period from 1970 to 2007 for three sectors, industry, commercial sector, transport, as well as on the macro level. Using the recently developed ARDL bounds testing approach by Pesaran and Shin (1999) and Pesaran et al. (2001), we find evidence for unidirectional long-run Granger causality in the commercial sector from growth to energy, as well as evidence for bi-directional long-run Granger causality in the transport sector. The dependence of causality on the level of aggregation is interpreted as evidence for ‘Simpsons' Paradox’. The choice of control variables is based on findings from the Environmental Kuznets Curve literature: we find that controlling for the increasing energy productivity of production as well as trade significantly improves the fit of the bivariate model.
Getting to Zero: Green Building and Net Zero Energy Homes
2011, Energy, Sustainability and the Environment
This chapter outlines some serious limitations to zero net energy (ZNE) homes entering the market, but a review of leading green building programs, zero home energy use initiatives, and deep green communities. It is possible to reduce net direct energy purchases to negligible levels in new homes and communities. Lifestyles will need to change so that homeowners downscale their energy use habits and redefine “living comfortably.” There is evidence of a mindset shift occurring in the developed world as it becomes aware of ballooning share of the world's consumption of natural resources. Homebuilders are realizing that future generations may want housing that does not consume finite resources for heating, cooling, and lighting. Many homes are energy inefficient today, the result of decades-old building practices being followed. Because residential homes are generally built to last, the worst-case scenario is that housing will be built today that is both energy inefficient and built to last—meaning that future owners will be stuck paying the bill for years to come. Reducing energy use without sacrificing lifestyle or comfort has always been the goal of the energy efficiency industry, but it is difficult to get to negligible energy consumption without making some sacrifices.
Getting to Zero
2011, Energy, Sustainability and the Environment: Technology, Incentives, Behavior
This chapter outlines some serious limitations to zero net energy (ZNE) homes entering the market, but a review of leading green building programs, zero home energy use initiatives, and deep green communities. It is possible to reduce net direct energy purchases to negligible levels in new homes and communities. Lifestyles will need to change so that homeowners downscale their energy use habits and redefine “living comfortably.” There is evidence of a mindset shift occurring in the developed world as it becomes aware of ballooning share of the world's consumption of natural resources. Homebuilders are realizing that future generations may want housing that does not consume finite resources for heating, cooling, and lighting. Many homes are energy inefficient today, the result of decades-old building practices being followed. Because residential homes are generally built to last, the worst-case scenario is that housing will be built today that is both energy inefficient and built to last—meaning that future owners will be stuck paying the bill for years to come. Reducing energy use without sacrificing lifestyle or comfort has always been the goal of the energy efficiency industry, but it is difficult to get to negligible energy consumption without making some sacrifices.
China's energy consumption: A perspective from Divisia aggregation approach
2010, Energy
Citation Excerpt :
Thereafter, some empirical studies appeared. For example, Bernard and Cauchon [20], Bernard and Côté [21], Berndt [22], Choi and Ang [23], Cleveland [24], Cleveland et al.[25,26], Hong [27], Nguyen [28], Nguyen and Andrews [29], Patterson [30] Zarnikau [31,32]. Most of the empirical studies focused on developed countries, especially the United States.
China's total energy consumption, according to the official data, decreased impressively during 1997–1998 and increased sharply during 2003–2007, which in turn resulted in energy intensity fluctuation. Many literatures explained this “unusual phenomenon” from the perspectives of technical change, economic structure shifting and statistical data quality. They measured aggregate energy in thermal units by using linear summation approaches. In this paper, from the perspectives of heterogeneity and imperfect substitutability among diverse energy types, we further examine China's aggregate energy consumption by using Divisia (Sato-Vartia) approach. The results show that China's aggregate energy consumption and intensity fluctuated slightly less than values calculated by using conventional linear approaches, and the “unusual phenomenon” is partly explained. It also implies that China's energy intensity changes in 2006–2007 are slightly more optimistic than those officially reported, and the official communiqué of provincial energy intensity reduction achievements are partly bias. Some provincial achievement are underestimated or overestimated on some provinces. Our empirical results are also helpful to further research, such as energy–economic modeling, energy price elasticity, and elasticity of substitution among capital–labor–energy–material (KLEM). The difficulties or defects when using Divisia approach are also discussed in this paper.
Energy prices, energy conservation, and economic growth: Evidence from the postwar United States
2009, International Review of Economics and Finance
The effects of energy prices and energy conservation on economic growth have been examined empirically for the postwar U.S. economy. A vector autoregressive model includes real GDP, real capital, labor, real energy prices, and the Divisia energy index. A key feature of our finding is that some damaging effects of energy conservation on the macroeconomy are statistically insignificant in the short run, and the insignificant short-run effects are quickly enervated over time. Alternative measures of energy use also suggest that energy conservation has no significant impact on real output growth. The findings are generally consistent with the neoclassical position that real economic growth of the United States is neutral with respect to changes in energy use. One exception is the case that energy prices are omitted from the model.

View all citing articles on Scopus

¹: Tel.: +1-512-327-6830, fax: +1-512-327-2553.

View full text

Defining ‘total energy use’ in economic studies: does the aggregation approach matter?

Abstract

Introduction

Section snippets

Deficiencies in heating value energy aggregates

The Divisia energy aggregate

Measures of national energy consumption

Implications

J. Econom.

Energy Int. J.

Consumer theory and the demand for money

J. Econ. Lit.

Aggregate energy, efficiency, and productivity measurement

Annu. Rev. Energy.

Two measures of aggregate energy production elasticities

Energy J.