DEPENDENCE OF ECONOMIC GROWTH ON CO2 EMISSIONS

Abstract

This paper examines the effects of CO^sub 2^ emissions on GDP by using a dynamic model for panel data from 19 OECD countries. The results indicate a significant decline in the dependence of economic growth on pollution, suggesting technological progress toward economic growth with less pollution, and providing indirect empirical support for the environmental Kuznets curve hypothesis.Keywords: Growth, Pollution, CO2, Dynamic Panel Data, OECDJEL classification: C23, Q43(ProQuest: ... denotes formulae omitted.)1. INTRODUCTIONPrevious studies of the relationship between economic growth and pollution have focused mainly on the effects of economic growth on pollution. One of the most widely debated observations in this regard is Grossman and Krueger's (1991) environmental Kuznets curve (EKC), which states that at low levels of income, environmental degradation increases as the economy grows and environmental pollution decreases when the economy develops further enough to exceed a critical threshold. Important studies of the EKC include Shafik (1994), Holtz-Eakin and Selden (1995), Moomaw and Unruh (1997), Schmalensee et al. (1998), Grossman and Krueger (1995), Seiden and Song (1994), Shafik and Bandyopadhyay (1992), and Panayotou (1993, 1997), among others. Stern (2004) provides a chronological and critical review of previous research on the EKC hypothesis. Dasgupta et al. (2002), Lieb (2003), Dinda (2004), He (2007), Aslanidis (2009), Kijima et al. (2010) and Bo (2011) also review the literature on empirical and theoretical discussions on the EKC. Wagner (2008) discusses that the seemingly strong evidence for an inverted U-shaped relationship vanishes when the issues of nonlinear transformation of integrated regressors and cross-sectional dependence in the panel data are taken into account.Previous studies have also examined the directional relationship between energy consumption and growth, including the causality from energy consumption to economic growth. Kraft and Kraft (1978) find causality from GNP to energy consumption in the U.S. by using data for the 1947-1974 period. Coondoo and Dinda (2002) conduct Granger causality tests and report that the causal relationship between CO2 emissions and income varies according to region. Dinda and Coondoo (2006) inve2stigate the causality issue in the income-emission relationship using a cross-country panel data set by employing panel unit root tests. Halicioglu (2009) finds that in Turkey, CO2 emissions are determined by energy consumption, income, and foreign trade, and tha2t income is also determined by CO2 emissions, energy consumptions and foreign trade. Lee (2006) conducts a thoroug2h exploration of the directional relationship by considering the G-11 countries and finds different directional evidences for different countries. A good summary of the evidences in empirical studies up to 2004 is also available in Lee (2006). Lee and Chang (2008) analyze panel data from 16 Asian countries for the 1971-2002 period, and find long-run unidirectional causality from energy consumption to economic growth.In the present paper, we focus on the directional causality from pollution to economic growth. While the extant works (e.g., Lee, 2006; Lee and Chang, 2008) assess the existence of the directional causality and measure the overall magnitude over the whole sample period, we investigate how this directional causality has evolved over time, that is, how the dependence of economic growth on the emission of pollutants has changed over time. The results of this study can be interpreted as the technological development with regard to how much pollution is required for economic growth. A decrease in tins causality over time would suggest 'environment-friendly' technological progress, which is one of the classical arguments for the EKC hypothesis (Panayotou, 1993).To estimate the directional causal relationship from pollution to economic growth, we use panel data from 19 OECD countries that have signed and ratified the Kyoto Protocol. …