Elsevier

Energy Policy

Volume 86, November 2015, Pages 812-832
Energy Policy

The economic growth enigma revisited: The EU-15 since the 1970s

https://doi.org/10.1016/j.enpol.2015.04.027Get rights and content

Highlights

  • One-size-fits-all macroeconomic policy is a travesty of reality.

  • The need to coordinate capital–labour–energy policies is shown

  • The importance of capital, labour and useful energy is country specific.

Abstract

Current macro-econometric models mostly incorporate just two factors of production, labour and capital (with a time-dependent multiplier representing technological change or total factor productivity). These models assume that energy is an intermediate product of some combination of human labour and capital. These models also assume that the supply of energy is driven by economic demand. We assume the contrary, i.e. that useful energy is a primary input, derived (mostly) from natural capital. This failure to capture the impact of primary resources (as useful energy) on economic growth leads to inappropriate formulation of economic growth theories. To understand that impact better we need explicit evidence of marginal products of capital, labour and useful energy or useful work. As applied to the explanation of the past half century of economic growth of the EU-15 countries, the new results demonstrate the use of non-parametric relationships between capital, labour and useful energy to explain economic growth. They also indicate that marginal products of capital, labour and useful energy are variable – the marginal product depends on the levels of capital stock, labour input and useful energy flows. The proposed semi-parametric production function suggests country-specific policy implications for the EU (and other countries).

Introduction

It is too often forgotten that economic growth has been a relatively episodic phenomenon in human history. Modest periods of real growth have occurred at various times in the more distant past, but growth that took a century in the late middle ages is now sometimes compressed into a decade, or even a single year. What has changed?

Ultimately, the debate about future of economic growth is an empirical one, and resolving the debate requires examining fundamental structural parameters of the economy. One key question is: How big are the marginal products of capital, labour and useful energy?

We estimate a semi-parametric production function econometrically using a unique dataset. We then analyze the properties of the production function in order to understand the dynamic behaviour of the three factors of production (capital, labour and useful energy) and of overall economic growth in the EU-15 countries since the 1970s.

The key motivation of the work presented here is to enable the formulation of economic growth theories that fit the historical facts. Methodologically, the work presented here is consistent with Haavelmo's Probability Approach in Economics (Haavelmo, 1944) in the sense of constituting a probabilistic formulation of economic theory using the generalised production function of Eq. (2). Standard macroeconomic analysis then assumes the ‘correct’ economic model and then uses econometrics merely to estimate its parameters from the best available data.

In contrast to this prevailing orthodoxy, we formulate the growth model as a probabilistic production function, but we allow the form of the function to be estimated from data without assuming any particular parametric form. To quote Haavelmo (1989): “The basis of econometrics, the economic theories that we had been led to believe in by our forefathers, were perhaps not good enough. It is quite obvious that if the theories we build to simulate actual economic life are not sufficiently realistic, that is, if the data we get to work on in practice are not produced the way that economic theories suggest, then it is rather meaningless to confront actual observations with relations that describe something else” (Section 4, paragraph 2).

An important advantage of the proposed generalised production function is that it is sufficiently realistic (see Section 3) to permit meaningful links between policy measures and actual growth outcomes.

The familiar Cobb–Douglas production function (of capital and labour) takes the following mathematical form (Cobb and Douglas, 1928):Yt=AtKtαLtβwhere Yt is the GDP, Kt is the stock of capital and Lt is the stock of labour, while parameters α and β are constants. Following Solow, 1956, Solow, 1957, At has been interpreted as the effect of ‘technical progress’ (now called total factor productivity TFP). This multiplier presumably captures the unexplained fraction of total economic output that is not caused by capital and labour.

We take the view that useful energy is an important factor of production. We reject the standard assumption that the output elasticity of each factor must be equal to the cost share of that factor in the GDP (Kümmel et al., 2010). We also reject the usual assumption that the functional relationship between GDP and capital, labour and useful energy is (log) linear (as is the case with the Cobb–Douglas production function). Instead, we explore empirically whether the marginal products of capital, labour and useful energy are constant, or not. If they are not constant, important policy implications can follow.

The rest of the paper is organised as follows. Section 2 provides a discussion of the key concepts of capital, labour and useful energy. This is followed by the mathematical description of the proposed generalised production function. Section 4 presents the unique dataset for the EU-15 countries used in this study. It also explains how useful energy consumption is estimated. It also presents the historical evolution of the useful energy categories: mechanical work, light, heat, muscle work and other electrical uses. Section 5 presents the key insights of the estimated production function of the EU-15 countries. Concluding remarks and key policy implications are given in Section 6.

Section snippets

Key concepts: capital, labour and useful energy

How can we account for capital? The so-called Cambridge controversies in the 1960s highlighted key problems, notably the difficulty of aggregating heterogeneous capital comprising machines, buildings, inventories, infrastructure, money and even natural resource stocks (Harcourt, 1972). In practice, we adopt Maddison's perpetual inventory method (PIM) to measure capital in monetary terms, accumulating capital from new investment less depreciation (Maddison, 1982).

Another issue is how to define

The generalised (semi-parametric) production function

Our generalised semi-parametric production function should not be understood as a tool for making point-estimates, or deterministic forecasts, as standard parametric (linear or non-linear) production functions are typically used. When (or if) one has prior knowledge of the parametric relationships between the GDP and the factors of production, then it is possible to use non-linear parametric models. However, when, or if, those relationships are not known in advance, our non-parametric approach

EU-15: data and empirical developments

It is hard to see how even the best of theoretical models could do more than frame the questions for empirical studies to address. Thus in the balance of this study, we examine insights that we have gathered from exploring the history of economic growth of the EU-15 countries.

The empirical production function of the EU-15 countries

Because it is not possible to provide a detailed discussion of all the EU-15 countries in a single paper, we provide below some thought-provoking observations to stimulate the thinking of readers.

It is important to note that the marginal product of capital, labour and useful energy should be interpreted with respect to the changes in the (log) level of the factor of production such as percentage changes in the stock of capital or useful energy flows. This is because the proposed production

Conclusions and policy implications

As a way of addressing the economic growth enigma for the EU-15 countries, we re-examined the importance of useful energy by postulating a generalised production function with three factors of production, namely capital, labour and useful energy. Effectively, we argue for a macroeconomic theory that fits the facts, not the contrary.

It is difficult to dispute the argument that abundant useful energy at low cost is fundamental to highly industrialised economies like the EU-15. As argued by

Acknowledgements

We would like to thank the four anonymous reviewers for their suggestions and comments. Furthermore, we want to thank the participants of the 2015 GCEP ‘Net Energy Analysis’ workshop at Stanford University.

References (30)

  • Haavelmo, T., 1989. Econometrics and the Welfare State. Technical report, Nobel Prize...
  • C. Hall et al.

    Energy and the Wealth of Nations: Understanding the Biophysical Economy

    (2011)
  • D.J. Hamilton

    Oil and the macroeconomy since World War II

    J. Polit. Econ.

    (1983)
  • Harcourt, G., 1972. Some Cambridge Controversies in the Theory of Capital. Modern Revivals in Economics. Gregg...
  • IEA, 2014. Energy Balances of OECD Countries, Documentation for Beyond 2020 Files....
  • Cited by (24)

    • An energy-based macroeconomic model validated by global historical series since 1820

      2022, Ecological Economics
      Citation Excerpt :

      In this paper we present a simple, energy-based macroeconomic model to study the world economy, as an essential step to unravel the economy-environment nexus. Inspired by ecological economics(Cleveland et al., 1984; Ayres and Miller, 1980; Kümmel, 1982; Costanza, 1980; Stern, 1993; Ayres and Warr, 2010; Keen et al., 2019; Ayres and Voudouris, 2014; Voudouris et al., 2015) and neoclassical macroeconomics(Aghion and Howitt, 2008; Jones and Vollrath, 2013; Romer, 2012), this model aims to reconcile both schools. Since the subject of an energy-GDP relation has been treated by many in the last decades, we will start in Section 2 with a review of the literature from which a few teachings and issues emerge.

    • Technical limits for energy conversion efficiency

      2020, Energy
      Citation Excerpt :

      It enabled the understanding of the historical role of energy efficiency improvement in overall economic development firstly proposed by Aryes [5,6] through and analysis of the US economy. The model was also successfully applied to other developed regions such as the EU and Japan [23,24] highlighting that Useful exergy should be considered as a factor of production that explains economic growth. Societal exergy analysis has also enabled a framework to prioritise energy efficiency action in different countries and sectors [25].

    • Growth imperatives: Substantiating a contested concept

      2019, Structural Change and Economic Dynamics
    • Real capital investments and sustainability - The case of Sweden

      2019, Ecological Economics
      Citation Excerpt :

      They have primary criticized the fact that energy and especially fossil fuels have been disregarded as a key factor behind the economic growth during the 20th century. Cleveland et al. (1984), Ayres and Voudouris (2014) and Voudouris et al. (2015) have shown that access to (useful) energy is a fundamental requirement and a potentially limiting factor for economic growth. The importance of energy and other resources are not questioned by conventional economists but are considered less of a constraint.

    View all citing articles on Scopus
    View full text