The impact of economic growth on environmental performance

The main subject of interest in this research is the impact of economic growth on environmental performance. Meadows et al. (1992) argues that, although a threat to the environment in the long-run, economic growth is essential to maintaining environmental quality (PAPER 1).

This concern has resulted in a large stream of research based on the notion of environmentally sustainable economic development that investigates the trade-off between economic growth and environmental quality (Anderson, 1992). The principal view is that a conventional trade-off between economic growth and environmental quality is preventable. In actual fact, there is high possibility of reducing or reversing trade-off with environmental management through appropriate policy interventions (Antle and Heidebrink, 1995; Grossman and Krueger, 1995; Selden and Song, 1994; Shafik, 1994). This is of major significance especially to developing countries like Brazil who, under pressure to attain rapid economic growth face the risk of implementing economic policies that run contrary to the objective of their long-term environmental sustainability (Serageldin and Steer, 1994) (PAPER 1).

This paper aims to investigate the environmental and economic relationship based on the EKC analysis in the case of Brazil which will essentially enable the economy minimise trade-off cost. In this chapter, the theoretical framework of the relationship between economic growth and environmental degradation will be reviewed. Also considering various literatures that have attempted to empirically analyse the relationship.

CONCEPTUAL ISSUES

THE ENVIRONMENTAL KUZNETS CURVE:

The EKC became an autonomous and essentially empirical research issue, following the studies of Grossman and Krueger (1991), Shafik and Bandyopadhyay (1992) World Development Report 1992 and Panayotou (1993) (PAPER 3). Subsequent literature on EKC investigated the existence of significant statistical correlation between the level of economic activity and environmental degradation without profoundly assessing the nature of causation between the variables.

In the EKC analysis, environmental degradation is the dependent variable. This suggests that during an empirical examination, a unidirectional causal relationship is implicitly assumed. Implying that a change in the level of economic activity results in subsequent change in environmental quality. Although this postulate may apparently seem rational, it may not apply to all situations. However, the assumption of the direction of causality may hinder a profound understanding of the environment-economic relationship and this may have significant policy implications such as erroneous policy conclusions (paper 5).

ENVIRONMENTAL POLLUTANTS:

The empirical results for the net impact of economic development on environmental degradation seem to depend on the nature of different pollutants (Shafik and Bandyopadhyay, 1992; Hettige et al., 1992; Birdsall and Wheeler, 1993; Diwan and Shafik, 1992) such as carbon monoxide(CO), carbon dioxide (CO2) the principle greenhouse gas emitted as a result of human activities such as the burning of coal, oil and natural gases, chlorofluorocarbons, sulphur dioxide(SO2) gas produced from burning coal and nitrogen oxide(NO) which causes smog and acid rain.

Selden and Song (1994) investigated considering different air pollutants like SO2, NOx and CO and found akin results related to EKC. However, the inflection points are substantially different across studies. Holtz-Eakin and Selden (1995) found that CO2 emissions did not show the same EKC pattern. Rather, Shafik and Bandyopadhyay (1992) show that the CO2 emissions and per capita GDP are monotonic in direction (PAPER 1).

THEORETICAL REVIEW OF ECONOMIC GROWTH AND ENVIRONMENTAL DEGRADATION

Grossman and Krueger (1991) were the first to articulate the EKC concept whilst analysing the relationship between economic growth and environmental degradation. In their test, they included greater openness to trade implying that it may result in reduced environmental standard in order to preserve international competitiveness. Consequently, a dynamic link between the environment, resource use and economic activity was established in the work of Kolstad and Krautkraemer (1993), where they found a long-run negative impact of resource use on the environment despite its economic benefits (PAPER 2).

The case of energy use was further examined by Suri and Chapman (1998) who emphasised commercial energy consumption, considering it as the root cause of severe environmental problems. They discovered in their study that energy requirements increased in both industrialising and industrialised countries due to exportation of manufactured products and thus causing the upward sloping portion of the EKC. Although, this reduced in industrialised countries with migration to importation of manufactured products resulting in the downward slope (PAPER 2).

In a study centred on energy consumption, Kadnar (2004) patterned a model for Central Asian economies aimed at predicting the future short-term energy needs. He employed the connection between population growth, consumption and real gross domestic product (GDP) for two scenarios (no growth and 5% sustained economic growth), and obtained mixed results (PAPER 2). However, using an index of 11 environmental indicators for 21 countries MacGillivray (1993) obtained minor evidence of a correlation between income and an index of environmental degradation (PAPER 4).

Indeed, for the full span of EKC to exist there should be tendency to reduce energy consumption, waste and material use, or all three elements. Ultimately, the existence of an EKC for an individual country does not imply a universal application of the relationship. Since a decrease in environmental degradation for a developed economy may simply be a result of heavily polluting industries migrating to less-developed countries that engage in environmental mismanagement (PAPER 4).

In terms of environmental management, Goldemberg (1998) maintains that catastrophic environmental incidents may be avoided by jumping the process followed by developed countries and integrating environmentally efficient technologies early in the development process. Nevertheless, the key finding of Panayotou (1997) is based on quality of policies implying that the country’s institutions can significantly reduce environmental degradation at low-income levels and accelerate improvements at higher-income levels. Efficient Policies such as more secure property rights under a rule of law and effective environmental regulations can facilitate flattening the EKC and lessen environmental cost of higher economic growth (PAPER 1).

EMPIRICAL REVIEW OF ECONOMIC GROWTH AND ENVIRONMENTAL DEGRADATION

There are a considerable number of studies that have examined the reciprocal link between environmental degradation and economic growth using numerous variables. Following Kraft and Kraft (1978), these studies resulted in diverse results (Akarca and Long, 1980; Yu and Hwang, 1984; Yu and Choi, 1985; Erol and Yu, 1987; Nachane, Nadkarni and Karnik, 1988; Abosedra and Baghestani, 1989; Hwang and Gum, 1992 and Bentzen and Engsted, 1993) and suffer from omitted variables bias. However, Stern (1993) was the foremost advocate of multivariate setting to comprehend the relationship between energy consumption and income growth. Prior to Stern’s work, most authors like Grossman and Krueger (1991), Lucas et al. (1992), Shafik and Bandyopadhyay (1992) used large scale sample with cross section time series data (PAPER 2). Subsequent to the work of Stern, numerous authors have also used non-multivariate settings and found significant correlation for economic-environmental relationship.

Tucker (1995) analysed changes in CO2 and income using yearly cross-sectional analyses for the period 1971-1991. The research discovered that changes in CO2 emissions are evidently correlated to changes in oil prices, although did not incorporate them into the analysis. While the Agras and Chapman (1998) research highlights the importance of energy prices and includes it in the econometric EKC framework assessing energy-income and CO2-income relationships. The long-run models find that income is no longer significant indicator of environmental quality or energy demand price-income (PAPER 2).

The linkage between GDP and energy consumption was investigated by Erdal et al. (2008) and Lise and Montfort (2006 (PAPER 8). Using co-integration analysis, they studied environmental pollutants and output and energy consumption and output nexuses in the case of Turkey. Lise’s (2007) study of 1970-2003 with the Engle- Granger co-integration approach rejected the EKC hypothesis and discovered a uni-direction from GDP to energy consumption. While Erdal et al. (2008) used Johansen co-integration approach and established a bi-directional causality between energy consumption and national income in the period 1970-2006

Soytas et al.(2001) investigated the causal relationship between energy consumption and gross domestic product applying Engle-Granger co-integration analysis over the period 1960-1995.They construe that energy consumption uni- directionally Granger-causes GDP. Furthermore, Soytas and Sari (2007) for the period of 1960-2000 applied impulse response analysis with unit root test and also researched the Granger causality relationship between economic growth, energy consumption and CO2 emissions in Turkey. The gross fixed capital formation and labour were controlled for but gave interesting results whereby, carbon emissions seem to uni-directionally Granger cause energy consumption, but not vice versa (paper 5).

The period 1970-2002 was examined using simple regression analysis by Say and Yucel (2006) and the study showed a positive relationship between national income and total CO2 emissions. While Tunc et al.(2007) utilized an input-output approach to define the link between production, fuel use and CO2 emissions in 43 industries. Altinay and Karagol (2004) used unit root tests and causality tests for the period of1950-2000 to validate the existence of causality between GDP and energy consumption. The research found no causality between the variables (PAPER 8).

Similarly, a causality test between energy consumption and GDP was performed by Yang (2000), Glasure and Lee (1997), Cheng (1996, 1999) and Yu and Choi (1985). The Granger causality test(GCT) (Granger 1969) was employed by Yang to examine causal relationship between energy consumption and GDP for Taiwan (1954-1997) using time series data. Yang’s result suggested a bi-directional causality between both variables. Furthermore, an investigation was carried out for South Korea and Singapore by Glasure and Lee to identify the causality between energy consumption and GDP. Using the techniques of co-integration and error-correction model, the result indicated a bi-directional causality for the countries in question. Nevertheless, applying the GCT suggested the absence of any causal relationship for South Korea and presence of uni-directional causal relationship from energy consumption to GDP for Singapore.

While the Johansen’s technique of co-integration for India suggests energy consumption, economic growth, capital and labour to be co-integrated in Cheng’s study. Moreover, applying the Hsiao’s version of GCT (Hsiao, 1981) in addition to co-integration and error-correction model analysis, Cheng achieved unidirectional causality from economic growth to energy consumption in both short and long-run. In Yu and Choi’s framework, they discovered that GNP and energy consumption has no causal relationship in USA, UK and Poland, while causality existed from GNP to total energy consumption in South Korea and causal linkage from total energy consumption to GNP in Philippines.

The above findings appear to emphasize the necessity to further investigate the causality nature of the income-environment relationship (paper 5).

With regard to the EKC hypothesis, Focacci (2005) carried out an empirical analysis regarding the environmental and energy policies in Brazil, China and India. The study includes ratio analysis using two key ratios namely, emission intensity ratio and energy-intensity ratio to relate to EKC model. The results show mixed results with respect to application of EKC model for these three economies. It suggests that resulting trends in these three countries are different from the other developing countries (PAPER 2).

Akbostanci et al. (2009) also peformed an EKC study for Turkey (1968-2003) with CO2, SO2 and PM10 emissions in terms of traditional parametric regression and the results did not support the EKC hypothesis. The authors dispute that pollution would not vanish automatically in the presence economic growth. List and Gallet (1999) discovered different values for the EKC income turning points across the United States of America, for NO2 and SO2 emission. In other words, the American States face strong heterogeneities concerning its pollutant emissions. However, the USA is regularly believed to be more homogenous than other countries, this research supports the benefit of time series estimations over cross-country studies (PAPER 3).

Previous researches have used a cross-section or panel data technique to assess the relationship between per capita income and diverse environmental indicators. Although some of these researches establish arguments complimenting the EKC hypothesis, from a strictly econometric standpoint, it is not totally evident that any individual country in the sample will follow the same degradation path estimated for the panel of countries.

Considering this framework, the empirical analyses of Grossman and Krueger (1994), Bradford et al.(2000) and Cole (2004) for the EKC using cross-section and panel data techniques, predicted income turning point altering between 4000 and 8000 USD (in PPP, constant 1985 prices). These are values that highly exceed the levels of per capita income attained by most developing countries. Basically, most developing countries are presently below their EKC turning point and for economic growth to have a positive effect on environment, emission would have to have to reach a significantly high level as predicted by the EKC. Moreover, it should be noted that growth induced environmental damage is not reversible and that decrease in environmental degradation differs from restoration to previous environmental standard.

More importantly, every country has its peculiar environmental stock, income generating capacity and various levels of impact of both variables according to the EKC pattern. However, generalising these factors across developed countries and other category of countries is not only unrealistic but also dangerous. Stern et al. (1996) argues that the only benefit from cross-country EKC analysis is a descriptive statistic. According to this author, it is obvious ”that a more fruitful approach to the analysis of the relationship between economic growth and environmental impact would be the examination of the historical experience of individual countries, using econometric and also qualitative historical analysis” (p. 1195). Therefore, the nonexistence of a standard EKC for all the countries necessitates the research on individual countries and demonstrates optimism for developing countries to identify their own path of sustainable development (PAPER 3).

Few studies using time series data based on the EKC literature for single countries have been done. Among these studies are Lindmark (2002) for Sweden (1870-1997), Day and Grafton (2003) for Canada (1958-1995), Egli (2004) for Germany (1966-1999), Ang (2007) for France (1960-2000) and Fodha and Zaghdoud (2009) for Tunisia (1961-2004). The core conclusion from these studies is based on the importance of each country’s distinct features, such as technological advancement and the evolution of its structural disposition for pollution and income. Akbostanci et al. (2009) postulates that the existence of EKC for different pollutant emissions can only be thoroughly analysed using a single country (PAPER 3).

All the research literatures propose that the constant rise in global CO2 emission appears to be increasing the crisis of environment degradation causing global warming, which was also highlighted by the Intergovernmental Panel on Climate

Change (IPCC, 2007). The main cause of emission is energy consumption, implying an emission reduction strategy to be reduced energy consumption. However, reducing the energy consumption of a growing economy like Brazil led by industrialisation and investment will mean to trade-in rapid economic growth for good environmental quality (PAPER 2).