Elsevier

Energy Policy

Volume 95, August 2016, Pages 32-41
Energy Policy

Revisiting Environmental Kuznets Curves through the energy price lens

https://doi.org/10.1016/j.enpol.2016.04.038Get rights and content

Highlights

  • EKC literature usually dismiss the major influence of energy prices.

  • Relative energy prices invalidate the evidence in favour of the EKC hypothesis.

  • Results may explain some contradictory observations found in the EKC literature.

  • Nowadays reduction in energy prices may break decarbonisation trends.

  • Direct action by policymakers is required to break the positive GDP-CO2 link.

Abstract

The goal of this paper is to provide new insights to elucidate the inconclusive results from the Environmental Kuznets Curve (EKC) empirical literature. For the first time in empirical literature, an econometric analysis includes the relative prices for several energy sources. The paper provides strong evidence on the relevance of energy prices to CO2 emissions. Accordingly, one reason for the lack of agreement in the EKC literature may be the absence of energy prices in empirical exercises. The presence of relative energy price changes in the econometric specification confirms a monotonic and positive relationship between CO2 and gross domestic product (GDP). Therefore, we may conclude that there is a decoupling process but without reaching any turning point on that relationship. The policy implications are straightforward. Direct climate action by policy makers is required to break the positive relationship between CO2 and GDP. That conclusion has been reinforced by the reduction of energy prices since the middle of 2014. Otherwise, the trend in energy prices may reverse the relative decarbonisation processes accounted for in recent years in major developed countries.

Introduction

An important strand of literature searches for relationships between CO2 emissions and gross domestic product (GDP), population growth, or both variables simultaneously. The underlying idea behind most papers in the literature is the Environmental Kuznets Curve (EKC) hypothesis: a sort of inverted U-shaped curve that relates pollution to economic development. It suggests that per capita CO2 emissions may rise as a response to per capita income growth in the early stages of economic development up to a turning point, after which emissions begin to decline.

The EKC hypothesis seems to beg the question of whether income growth beyond a turning point might serve as a solution to environmental degradation, rather than representing the source of the problem (Kaika and Zervas, 2013a, includes a very interesting literature review on this issue). Thus, the search for a turning point is central in the EKC literature as highlighted, for instance, in Galeotti and Lanza (1999). Their survey includes some references in favour of the EKC hypothesis but the estimated turning points are beyond the sample income levels (e.g., Holtz-Eakin and Selden, 1995, Cole et al., 1997). Strictly speaking, these papers only provide evidence of a decreasing marginal propensity to emit greenhouse gases. There are also papers that estimate turning points within sample income levels. Some recent examples are provided in the work of Ang, 2007, Coondoo and Dinda, 2008, Apergis and Payne, 2009, Dutt, 2009, Pao and Tsai, 2010.

An important policy implication derives from those papers providing evidence in favour of the EKC for CO2 with turning points within sample income levels; further increases of incomes beyond the turning points lead to unambiguous reductions in CO2 emissions. Consequently, favourable evidence in the EKC literature might legitimize lax political responses to climate change (or delays in adopting environmental policies [e.g., Beckerman, 1992]), as long as it provides a medium- and long-term solutions. That kind of statement is explicit or implicit in some papers’ messages. In some cases, authors seem to agree with this view; others just mention that it may be the conclusion behind any eventual support of the EKC hypothesis and/or within-sample turning points (for recent examples, see, for instance, Richmond and Kaufmann, 2006b, Apergis and Payne, 2009, Pao and Tsai, 2010, Niu et al., 2011)). Dutt (2009) makes an important point with regard to the view that the EKC hypothesis can legitimize lax political responses: “Based on all the findings so far, this [increase in income automatically resulting in lower emissions] does not seem to be the case. […] Policies, combined with better institutions and public awareness could have helped reduce emissions, and in this study these happen to be associated with high-income countries”.

Many papers have questioned whether the empirical data accommodates the EKC hypothesis. The main insight from the empirical literature is highlighted by Kaika and Zervas (2013a) as follows: “the EKC literature is quite large, and results are at best mixed. No clear conclusion can be drawn”. As Agras and Chapman (1999) point out, the EKC hypothesis arises from a set of relationships that need to be fully identified. One important relationship that might influence the EKC hypothesis is the interaction between pollution and energy prices. However, energy prices have been omitted from most of the empirical specifications.

The main objective of this particular research is to provide additional evidence of the important role of energy prices in the EKC debate. Our main hypothesis is that substitution and income effects in response to energy price changes exert a major influence on the relationship between GDP and CO2 emissions.

This paper contributes to the literature by including, for the first time in an empirical EKC model, relative energy prices (for coal, oil products and natural gas). The focus of this paper is on OECD countries because they are more likely than other countries to display the turning point predicted by the EKC hypothesis. Our results emphasise the need to account for relative energy prices in order to avoid biased conclusions. The paper develops in the following manner: Section 2 reviews the literature on the EKC hypothesis; Section 3 presents the econometric methodology; Section 4 introduces the data employed in this study; Section 5 provides the econometric results and some further discussion; finally, Section 6 summarises the paper's conclusions and its main policy implications.

Section snippets

A survey of the literature

The literature on the EKC hypothesis began, to our knowledge, with the research agenda developed by the World Bank as part of the World Development Report (see, for instance, the series of working papers published in 1992: Lucas et al., 1992, Radetzki, 1992, Shafik and Bandyopadhyay, 1992)). As a result of this line of research, Beckerman (1992) concludes that the “strong correlation between incomes and the extent to which environmental protection measures are adopted demonstrates that, in the

Methodology

In this section, we outline the econometric methodology that we use to validate the EKC hypothesis for CO2 emissions. As previously mentioned, there is not a consensus in the empirical literature on the relationship between economic growth and CO2 emissions. This absence of consensus may be related to the identification dilemma researchers faced when looking for a proper reduced form of estimation for the inverted U-shaped curves, as long as “it is crucial to proper inference to impose

Data

The focus of this paper is on 15 OECD countries: Austria, Belgium, the Czech Republic, Denmark, Finland, France, Hungary, Italy, Japan, Poland, the Slovak Republic, Switzerland, Turkey, the United Kingdom and the United States. According to the EKC hypothesis, these countries may be rich enough (per GDPpc) to be on the right side of the inverted-U relationship (for a similar approach, see the in-depth survey in Kaika and Zervas (2013a, 2013b). The lack of data for other OECD countries (mainly

Results and discussion

This section presents the results, and it will highlight the important role of energy prices to disentangle the effect of GDP on CO2 emissions from unobserved time-related effects.

Conclusions and policy implications

This piece of research provides important evidence on the role of final energy prices in the EKC debate. In particular, the inclusion of changes in relative energy prices invalidates the evidence that we might otherwise find in favour of the EKC hypothesis in our econometric exercise. Our focus on final energy prices relates to the fact that they may provide valuable information related to unobserved, country-specific time factors. We may conclude then that the widespread omission of relative

Acknowledgements

M. Rodriguez gratefully acknowledge the financial support from Spanish Ministry for Science and Education (Projects ECO2013-45706-R), and the Galician Government (Projects GRC2014/021 and EM2014/044). Y. Pena-Boquete wishes to acknowledge the financial support of the Galician government, through project GRC2014/021 and the fellowship programme Ánxeles Alvariño. J. C. Pardo-Fernández has been financially supported by grant MTM2014-55966-P (ERDF support included) of the Spanish Ministry of

References (55)

  • S. Dinda

    Environmental Kuznets Curve hypothesis: a survey

    Ecol. Econ.

    (2004)
  • S. Farhani et al.

    The environmental Kuznets curve and sustainability: a panel data analysis

    Energy Policy

    (2014)
  • M. Galeotti et al.

    Richer and cleaner? A study on carbon dioxide emissions in developing countries

    Energy Policy

    (1999)
  • D. Holtz-Eakin et al.

    Stoking the fires? CO2 emissions and economic growth

    J. Public Econ.

    (1995)
  • K.S. Im et al.

    Testing for unit roots in heterogeneous panels

    J. Econom.

    (2003)
  • V.C. Jaunky

    The CO2 emissions-income nexus: evidence from rich countries

    Energy Policy

    (2011)
  • D. Kaika et al.

    The Environmental Kuznets Curve (EKC) theory — Part A: concept, causes and the CO2 emissions case

    Energy Policy

    (2013)
  • D. Kaika et al.

    The Environmental Kuznets Curve (EKC) theory — Part B: critical issues

    Energy Policy

    (2013)
  • K. Kanjilal et al.

    Environmental Kuznet's curve for India: Evidence from tests for cointegration with unknown structural breaks

    Energy Policy

    (2013)
  • M. Kijima et al.

    Economic models for the environmental Kuznets curve: a survey

    J. Econ. Dyn. Control

    (2010)
  • L. Lau et al.

    Investigation of the Environmental Kuznets Curve for carbon emissions in Malaysia: do foreign direct investment and trade matter?

    Energy Policy

    (2014)
  • A. Levin et al.

    Unit root tests in panel data: asymptotic and finite-sample properties

    J. Econom.

    (2002)
  • G. Marrero

    Greenhouse gases emissions, growth and the energy mix in Europe

    Energy Econ.

    (2010)
  • S. Niu et al.

    Economic growth, energy conservation and emissions reduction: a comparative analysis based on panel data for 8 Asian Pacific countries

    Energy Policy

    (2011)
  • B. Ozcan

    The nexus between carbon emissions, energy consumption and economic growth in Middle East countries: a panel data analysis

    Energy Policy

    (2013)
  • H.T. Pao et al.

    CO2 emissions, energy consumption and economic growth in BRIC countries

    Energy Policy

    (2010)
  • A. Richmond et al.

    Is there a turning point in the relationship between income and energy use and/or carbon emissions?

    Ecol. Econ.

    (2006)
  • Cited by (0)

    View full text