FOSSIL ENERGY IN ECONOMIC GROWTH: A STUDY OF THE ENERGY DIRECTION OF TECHNICAL CHANGE, 1950-2012

摘要

OverviewClimate change mitigation challenges national economies to increase productivity while reducing fossil energy consumption. Fossil energy-saving technical change has been assumed to accomplish this, yet empirical evidence is scarce (Bowen & Hepburn 2014). Empirical studies attempting to determine the relationship between output and fossil energy use econometrically without an underlying theory of production and technical change report contradictory findings (eg Ozturk 2010, Wagner 2008). Results tend to be sensitive to method, sample size and enpoints chosen. Empirical studies starting from economic models of growth and technical change find plausible results about corrrelations between price movements, and rate and direction of technical change, but tend to report results only for single or few, high income countries due to lack of data such as energy prices for longer time ranges or many countries (Hassler et al. 2012). Because of the tight integration of national economies into global trade networks with resulting varying national economic structures that differ in their energy intensities, national patterns of energy used in production cannot be used to infer the worldwide relationship.This paper investigates the long-run relationship between the rate and direction of technical change with respect to fossil energy and labor for the world economy. Growth rates of labor productivity and the fossil energy–labor ratio are examined for more than 95 percent of world output between 1950 and 2012. The average elasticity of the energy–labor ratio with respect to labor productivity is close to one, implying highly energy-using technical change, but no trade-off between factor productivity growth rates. This stylized fact suggests the importance of a cheap, abundant energy supply for robust global growth, and a crucial role for renewable energy. Integrated assessment models in the current Intergovernmental Panel on Climate Change report do not incorporate this restriction in their baseline scenario, which raises questions about the empirical basis of their transition pathway projections.Section 2 reviews empirical literature on past energy intensity trends, both with and without reference to economic production, and explains the simple production model with energy on which this analysis is based. Section 3 describes how the dataset was constructed and details the method of analysis. Section 4 presents results of an analysis of the correlation of rate and direction of technical change both in cross sections of national growth rates, and in time series of regional growth rates. The influence on the fossil energy direction of renewable energy sources in the mix is also analyzed. Section 5 discusses the implications of the results both for current policy and for scenarios of future economic growth in integrated assessment models. Section 6 concludes by summarizing the results and identifying promising research avenues.MethodsData for output and employment are from the Total Economy Database based on estimates by Angus Maddison (Conference Board 2015). Fossil energy data is from from the International Energy Agency (IEA) for non-residential primary fossil energy consumption after 1970, and from Darmstadter et al. (1971) for primary fossil energy consumption for the period 1950-1968. Methods of analysis are qualitative and quantitative. Visual analysis of scatter plots of compound annual growth rates of cross sections and time series identify long-term patterns; outliers are analysed against the historical economic context of their period. Linear and locally weighted polynomial regression (loess) yield elasticity estimates and the domain over which a linear relationship between productivity growth and the direction of technical change prevails, and a panel estimate tests robustness of cross-sectional correlations to unobserved heterogeneity.Results1) In compound annual growth rate cross sections, the national elasticity of the fossil energy labor ratio with respect to labor productivity (η) has been close to one over a large domain of labor productivity rates, and for the greatest part of the sample; in other words, different labor productivity rates have been achieved at a similar rate of change of energy intensity. Neither an ‘innovation possibilities frontier’ trade-off between several single-factor productivities nor fast ‘decoupled’ growth occurred.2) This pattern holds with remarkable constancy both in the 1950-70s and after 2000; different technical change regimes in the 1980s and, to a lesser extent, in the 1990s are explained partly by an adjustment to higher energy prices due to two OPEC crises, and the transition of formerly socialist economies. A panel estimate with country and time fixed effects confirms this relationship.3) Recent substitution of fossil energy with low carbon energy sources has lowered the elasticity, η, without having a significant impact on labor productivity growth.ConclusionsThe results show that historically, faster labor productivity growth from scale economies has coincided with more fossil energy-using technical change at a similar rate for the world economy. The robustness of the relationship over the last sixty years suggests a ‘stylized fact’ of the relationship between productivity growh and the fossil energy-labor ratio. Due to its centrality for ‘decoupled’ growth, aggregate models of economic growth with greenhouse gas should incorporate it.By singling out the 1980s and 90s as exceptional due to shocks from OPEC crises and the transition from plan to market, it goes some way to explaining the difficulty of identifying long-run relationships in datasets that start in 1971 (IEA data) or 1980 (US Energy Information Administration data), used by most extant econometric analyses. This also raises questions about extrapolations in integrated assessment models that typically base their historial trends on the period 1970-2010, of which two decades appear as exceptions due to ‘shocks’ that cannot be replicated by policy; moreover, their projections of global future rates of change of labor productivity and energy intensity do not reflect the restriction identified in the historical data.At the national level, episodes of successful national development have been highly energy-using, suggesting that future economic development and growth will need to rely on cheap and abundant energy supplies. This points to the importance of low-cost low carbon energy sources, which the data suggest permit fossil energy-saving technical change without hampering productivity growth.