The use of the Environmental Kuznets Curve in the Management of Sustainable Development

The relationship of economic growth and environmental impact has spurred fierce debates between growth optimists referring to the phenomenon of the environmental Kuznets curve, and pessimists referring to the limits to growth. The article draws some hints from a critical assessment of the literature on the environmental Kuznets curve. In particular, it is argued that the optimistic implications of this literature on the sustainability management are ungranted. However, analysis of environmental Kuznets curves al/ows a clarification of the few basic conditions for the management of sustainable development, including the sustainability of globalisation. These conditions can be met by implementing a systematic policy strategy aimed at shifting the Kuznets relations downward.


INTRODUCTION
General theoretical urgency_ "Sustainable growth" has recently become an issue of international debate at the EU. However, these debates are not based on the criteria that have to be fulfilled for the growth to be sustainable. But it is more uniquely accepted that the dominating development models contain serious deficiencies, and that the current development trajectory is clearly unsustainable. Discussions of the recent years on the dissociation of the previously positive relationship between economic growth and resource utilisation (environmental degradation) are clearly related to the research of environmental Kuurets curves, where resource degradation will increase initially with per capita income growth and then eventually decline, thus exhibiting the characteristic inverted V-shaped relationship between affluence and emissions known as the "environmental Kuznets curve", where emissions initially worsen but ultimately improve with income.
Problem. In order to incorporate environmental issues into the macro-economic analysis, a totally new approach to the economic theory is needed, which could allow to review the standard technique of calculating national income, as ecological factors have not acquired an acceptable expression in the indicator system of modem economic development.
Research objects. The attention is focused on the analysis of the indicators of sustainable development. The environmental indicators that must be used in the decision making about sustainability are presented.
Objectives. The colltent oJ environmental indicators, including concepts oJ ecological space and ecological footprint, and the problems from the perspective of their suitability for decisions on economic del/e!opment sustainability are critically investigated.
Tasks. In order to fulfil these objectives, the following tasks had to be accomplished: to analyse the need of the indicators in the management of the sustainable development; to analyse the potential of macro-economic indicators for evaluating welfare and sustain ability; to review environmental indicators used in economic development sustainability decisions; to discuss the ecological footprint and the environmental space concepts from the viewpoint of their sustainability evaluation potential; to formulate theoretical principles for calculating the environmental space for certain resources. Methods of research. In the article were used the methods of logic abstraction, which encompasses generalisations on economic and management theories and thoughts, theoretical systems analysis of the problems of sustainable development according to the conclusions and reasoning of scientists from other countries, a comparison and research of the processes ofthe development of economic systems (industries). The main scientific works related to the problem in question have been reviewed and thoroughly analysed.

THE ESSENCE OF THE ENVIRONMENTAL KUZNETS CURVE CONCEPT
During the most part of the course of industrial development, economic growth entailed a parallel growth in resource consumption and 56 environmental degradation. Though this relationship still holds, experience of the last decades indicates that economic growth and increase in resources consumption and environmental degradation can be de-linked to a considerable extent. The path to environmental sustainability lies in maximising this delinking prosess. So, the theoretical explanation of the environmental Kuznets curves plays a central role in the theoretical foundation of sustainable growth.
Some relative data on environmental quality and the utilisation of natural resources and income per person allows us to make a presumption that environmental quality worsens with a low income level. But the situation improves with an increase of the income level, which reflects "the pressure of dissociating environment with economic growth" (Simonis, 1989). This relationship -as the income of an economy grows over time, emission level grows first, reaches a peak and then starts declining after a threshold level of income has been crossed -was first suggested in the early 1990s and has thereafter been subject to intensive research. The inverse relationship between pollution and per capita income has been explored for a variety of pollutants, such as nitrogen oxide, sulfur dioxide, suspended particulate matter, carbon monoxide, lead, and for deforestation, biological oxygen demand and others (List, Gallet, 1999;Selden, Song, 1994;Stern, et al. 1996;World Bank, 1992;Panayotou, 1997, Grossman, Krueger, 1995. The empirical literature about environmental Kuznets curves, which studies the empirical relationship between per capita income, generally interpreted as a proxy of the stage of development measured on the horizontal axis, and environmental deterioration measured on the vertical axis by different indexes: total environmental deterioration, or more often its per capita value or its value per unit of income; extensive critical survey is given in (Stern, et aL, 1996;Borghesi, 2001).
This interrelation between the national income per person and the concentration level of industrial waste by P. Dasgupta and K.-G. Maler (1995) is called the environmental Kuznets curve, analogous to a traditional curve proposed by Simon Kuznets (1955), which demonstrates a similar relationship between actual income per person and income inequality ( Figure 1). (As is well known, S. Kuznets observed that inequality tends to increase during the early stages of growth to decrease later on, describing an inverted-U shaped relationship between per capita income (on the horizontal axis) and income inequality (on the vertical axis). This relationship, called the Kuznets curve after the name of the author, was very popular during the 1970s when it was taken as an empirical regularity of the economy).
Most commonly, the studies of environmental Kuznets curves have taken econometric Resource degradation or pollution B approaches using data based on cross-sections of countries and sometimes combining this with time series data. Historical approaches to the environmental Kuzllels curve or other emission patterns, such as studies of individual country's historical emission trajectories, have been relatively rare. Environmental Kuznets curve studies for single countries most often address developing countries (e. g., Patel et aI., 1995;Vincent, 1997). Rare exceptions addressing industrialized countries include (De Bruyn et aL, 1998;Friedl, Getzner, 2003). But, as shown by M. Lindmark (2002), historical studies of individual countries offer an advantage over cross-section approaches in bringing the analyses closer to the dynamics that modity the environmental Kuznels curve pattern. An investigation of the time-series data of a single country may be able to account for historic experience such as environmental policy, development of trade relations, and exogenous shocks such as the oil crisis (Stem et aI., 1996). It has also been suggested that the environ-  A number of survey articles (Ekins, 1997; Stern, 1998) prompted several clarifications concerning both the actual development of various pollutants and the methodology used to explain the environmental Kuznets cUIVe patterns. The literature has mostly considered the environmental Kuznets cUlve as an empirical phenomenon and examined the presence or otherwise a significant statistical association between the level of economic activity and environmental degradation without explicitly discussing the nature of causation between these variables. The principal explanatory factor is income based on assumptions of the initially high but falling marginal utility of consumption and the initially low but increasing disutility of emissions as the incomes rise. Thus, it is presumed that the relationship between the income and pollution is a relationship of unidirectional causality with income causing environmental changes -viz. a change in the level of economic activity/per capita income causes a consequent change in the environmental quality and not vice versa. Additionally, technological and structural changes, including trade patterns, may also influence an the environmental Kuznets cww pattern. These changes may in turn interact with price changes.
As given in the paper of S. Borghesi and A. Vercelli (2003), in the case of air-quality indicators the existence of the environmental Kuznets cUIVe found good support for local airpollutants (e. g., Grossman, 1995) but not for global pollutants (such as CO 2 ), which have a limited direct impact on population (Cole et al., 1997). CO 2 emissions cause problems on a global scale, and the social costs of global warming accrue both across time and nations. Therefore, free-rider behaviour might lead to a close relationship between carbon emissions and income at all levels of per capita income (Arrow et al., 1995). In line with this argument, a linear relationship for CO 2 emissions and GDP per capita was confinned in early studies (Shafik, 1994). But the international nature of global warming is not the only reason that prevents de-linking greenhouse gas emissions from economic growth. The intergenerational nature of the negative impact of greenhouse gas emissions may have also been an important factor preventing the implementation of greenhouse gas abatement measures in the past. For water quality the evidence is more mixed, with studies giving conflieting results on the shape, position and peak of the curve according to the different indicators used. As for the other indicators of environmental degradation, the environmental Kuznets cUIVe hypothesis receives very little corroboration. Environmental problems that have a direct and strong impact on the population (such as access to urban sanitation and clean water) tend to improve steadily with the process of development, while environmental problems that can be transferred elsewhere (such as municipal solid wastes) do not exhibit any clear tendency to diminish with development (see Rothman, de Bruyn, 1998). Whatever degree of corroboration seems reasonable to attribute to the environmental Ku.znets cUIVe hypothesis on the basis of cross-country studies, single country studies reach very questionable results even in the cases best supported by cross-country studies (see, for instance, Vincent, 1997).
So, it should be noted that environmental Kuznets curves do not explicate on the systems' consequences of environment utilisation, therefore they should not be further used as a proof or a critical argument in grounding the statement that economic growth is sufficient to achieve environmental improvement. Thus, considering all arguments, the environmental Kuznets curve should be viewed as the hypothesis on the inte"elation between economic growth and environmental quality.
It must be mentioned that recent studies have also tested for a possible third order polynomial relationship between emissions and income (Moomaw, Unruh, 1997). However, they conclude that neither the inverted "U" nor an A cubic (i. e. "N" -shaped) relationship between CO 2 emissions and income provide a reliable indication of future behaviour. Hence the use of environmentaL Kuznets curve models to forecast future emissions may not be appropriate. On the other hand, Jones and Manuelli (1995) using an overlapping generations model show how the interaction of individual optimal decision making and collective regulation may lead to an environmental Kuznets curve, but also a N-shaped curvature is possible. An N-shaped relationship between Austria GDP and CO 2 emissions is found to fit the data most appropriately for the period 196(}""1999 in the research done by Friedl B., Getzner M., 2003 too.

INDIVIDUAL PREFERENCES AND ENVIRONMENTAL KUZNETS CURVE
M. Pasche (2002) showed that the sources of the environmentaL Kuznets curve can be summarized into two groups: (a) the structuraL change to service and information-based economic activities which are less pollutionintensive than physical production and (b) the growing ecologicaL efficiency of production and consumption by means of a "greening" technical progress. The driving forces behind these two determinants may be a change of preferences favoring environmental goods or at least a sufficiently high income elasticity of demand for environmental goods on the one hand, and regulating activities like, e. g. technical standards, legal restraints, environmental taxation on the other hand.
However, within the extensive body of literature that has been published in the recent years conccrning the environmentaL Kuznets curve, two main theoretical arguments have been formulated to account for the fact that beyond a particular level of per capita income the relationship between economic growth and environmental quality becomes a "virtuous" circle (Roca, 2003). Both arguments concern the changes in levels of relative demand that occur as per capita income varies.
The first argument suggests an endogenous change in the demand structure for goods and services. According to this first argument, the sectors that become increasingly important as per capita income increases are those whose environmental impact is less. The evidence that generally underlies this position is the increasing demand directed at the service sectors at the expense of demand directed at the industrial sector. However, much more empirical research needs to be done on the assumptions this argument is based on: some activities that are regarded as services may have as much or more environmental impact (direct and/or indirect) as others involving the industrial sector (consider, for example, long-distance tourism). In any case, this argument would only explain a reduction in environmental pressures per unit of GDP as income increases; it would not explain a reduction of these pressures in absolute terms unless we suppose that the sectors that are most environmentally problematic produce inferior goods. In fact, this is not at all likely (Torras and Boycc, 1998). In other words, if we apply the distinction made by de Bruyn and Opschoor (1997), the change in demand structure could account for a "delinking" of economic growth and (some) environmental pressures in the "weak (or relative) sense" but not in the "strong (or absolute) sense".
The second argument, as mentioned by J. Roca (2003), is also based on individual preferences and changes in relative demand that occur as income increases. In this case, however, it is not the changes in relative demand for different goods and services acquired in the market that are crucial, but those between the consumption of marketable goods and services on the one hand and environmental quality on the other. According to Lopez (1994), the relation between the level of pollution and the income level then depends on the elasticities of substitution of goods and the risk preference of the households. And a high "income elasticity of demand for environmental quality" could potentially explain the delinking of economic growth and environmental pressures in the "strong sense".
In other words, under the environmental Kuznets curve hypothesis, with growth of income the status of emission as an item of consumption gradually changes from a necessary to an inferior good (thUS reflecting a clear preference for a cleaner environment at higher levels of living).
An interesting idea is given in the K. A. Brekke, R. B. Howarth and K. Nyborg (2003) paper about the status-seeking and material affluence, evaluating the Hirsch hypothesis. When individuals hold a preference for high relative consumption, competition to achieve social status can lead to inefficiently high levels of production and consumption, contributing to natural resource depletion and environmental degradation. In the 1970s, F. Hirsch (1976) argued that an increasing portion of expenditure is allocated to status-seeking as average income rises. If Hirsch's hypoth-esis were correct, then status-seeking would contribute substantially to the problem of overconsumption in the world's richest nations (see Sachs et aI., 1998), and consumption levels could be reduced significantly without accompanying reductions in the quality of life. However, as shown by the authors of the paper, Hirsch's hypothesis depends critically on the empirical assumptions that mayor may not be satisfied in the real world. When the social status is defined in terms of the algebraic difference between an individual's consumption of the status goods and the average consumption level in society, Hirsch's hypothesis holds true, and growth in the level of productivity and output can lead to declines in human welfare. If, on the other hand, the status is linked to the ratio of individual and average consumption, 'Hirsch's hypothesis is valid only if social status and non-status goods are poor substitutes.

GLOBALISATION AND ENVIRONMENTAL KUZNETS CURVES
In the original definition of sustainable development, suggested by the Brundtland Commission (WCED, 1987); "sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs", inequality and environmental deterioration are conceived as equally important and interdependent conditions of sustain ability. The recent process of globalisation of international markets has managed to sustain the economic growth of the countries that have actively participated in this process. The available empirical evidence suggests, however, that it has been accompanied by a worldwide increase in environmental degradation and economic inequality. Therefore, there is growing concern that these features of the globalisation process may jeopardise its social and environmental sustainability. On the other side, we must take into account that no operational set of simple criteria has been developed so far to assess the sustain ability of a given growth pattern. This is all the more true for measures combining environmental and social criteria.
However, one thing is obvious: the causal relationship between globalisation and global environmental degradation is quite complex and ambiguous. As pointed out in the paper of S. Borghesi and A. Vercelli (2003), environmentally sustainable globalisation requires a policy strategy directed to shifting the relationship between environmental Kuznets curve relations downwards. Within the process of globalisation it is possible to modify the shape and position of the environmental Kuzllets curves, thus improving social and ecological conditions. In the case of the environment, public opinion can influence environmental quality not only through the voting system, but also through the market: "greener" consumer demand contributes to a shift in production and technologies towards less polluting activities. Globalisation may increase competition and thus strengthen public opinion pressure for environmental quality. In a more competitive market consumers are likely to have more alternatives to polluting products and thus more chances to express their environmental demand. This positive impact of globalisation on the environment, however, crucially depends on the actual capacity of globalisation to increase competition. If a greater market concentration comes together with globalisation (as occurs in some sectors), then the opposite might be true and environmentalfriendly consumers might end up with fewer opportunities to express their preferences. S. Borghesi and A. Vercelli (2003) conclude, therefore, that globalisation might contribute to a more sustainable development by enhancing th« impact of public opinion pressure on government and market decisions and thus shifting the environmental Kuznets curve relations downwards.
We must take into account one argument that may play a role for the explanation of the environmental Kuzllets curve in industrialized countries: the relocation of pollution-intensive industries from developed to less developed countries and the re-import of the products. This may be an empirically valid argument (see (Suri, Chapman, 1998», but this practice is obviously not a basis for global sustain ability. As show by (Muradian et ai, 2002), who estimated embodied pollution in trade of 18 industrialized countries with (a) the rest of the world and (b) developing countries from 1976 to 1994, the balance of embodied emissions in trade seems to follow an inverted-U shape across time in Japan and Western Europe, and an N-shape in the US. In the period of analysis, the Japanese and European environmental terms of trade with developing countries "improved" (from the Japanese and European point of view), whereas the American environmental terms of trade with developing countries "deteriorated" over time. Although there is no statistical trend between income and embodied emissions in imports in a cross-section analysis, there does seem to be a positive relationship between both variables at a national level. The results suggest that, despite many shortcomings, this type of assessment may provide useful insights on the international aspects of sustainable development.
The empirical magnitude of future technological change may well swamp the effect of depletion of natural resources and environmental pollution (Weitzman, 1997). In these circumstances, future generations will enjoy higher rates of well-being than today, even if the physical availability of different types of capital declines. Technological progress reduces the long-run costs of policies to meet sustainability goals, as higher initial costs elicit costreducing innovations at later stages. Second, it increases the benefits from early actions, as higher initial costs may translate into technology leadership at a later stage (Grubb et aI., 2001). Technology may also help to de-link economic growth from environmental degradation and to respond to the needs of the world's poorest. This impact is not automatic, though.
However, M. Pasche (2002) showed that under reasonable assumptions about technical progress a positive sustainable growth rate fails to exist. The argument is that a growing part of income has to be spent for continuing technical progress in order to compensate the pollution effects of growth. Hence, in the long run either the sustainability condition will be violated or the growth rate must decline to zero. Furthermore, in finite time the level of wealth will decrease despite growing income and a constant pollution level, thus further growth is no longer reasonable. Similar arguments also hold for the environmental effects of structural change favoring less pollutionintensive economic activities, when the part of pollution-intensive production can decrease while less pollution-intensive services or information-based production increases. Hence, the level of output can rise with constant or reduced emissions. An evolutionary change of goods and production technology may shift the limits of growth and is hence a prerequisite for a long-run environmental KlIznets curve. But the possibilities of a rational sustainability policy seem to be limited.
So, per capita income, in turn, affects inequality and environmental degradation through several channels, as suggested by the literature on the environmental Kuznets curve. Although the findings regarding the "environmental Kuznets curve" are not conclusive (the mentioned survey gives only limited support to the environmental Kuznets cun'e hypothesis), most empirical studies have generated very high income turning points beyond the maxi-mum income level of the data they used in their analysis, and beyond the level of affluence to which most developing countries might realistically achieve in the foreseeable future. For example, Holtz-Eakin and Selden (1995) generate an out-of-sample turning point on 35000 per capita (1986 US$) that indicates that substantial economic growth would be required before CO 2 emissions begin to decline. de Bruyn (1997) provides a survey of the empirical studies.
Several empirical studies of the 1980s and 1990s give an optimistic view: in industrialized countries many pollution indicators decreased despite a growing per capita income, while in less developed countries growth yields increasing pollution. Therefore, economic progress with less impact on the environment seems to be possible. The idea that economic growth is ultimately beneficial for the environment has caused some authors to maintain that only economic growth is necessary, because the surest way to improve the environment is to become rich (Beckerman, 1992). This viewpoint implies that environmental problems are a temporary phenomenon, since economic growth and technological innovation will resolve these problems in due time.
But the fact that the nations that formerly had or currently have a low per capita income are experiencing increasing pollution while industrialized countries are successful in abating emissions does not imply that economic development will solve environmental problems quasi automatically. It is possible to make only one statement: the research results have proved the presumption that economic growth can be conformed to environmental improvement, if accordingly a specific policy is worked out. The key policy conclusion is that even if such a curve characterized past growth, there is no reason for developing countries passively to accept "historical determinism" along their future development path. In effect, lower-income countries could learn from the experience of wealthier nations and adopt policies that permitted them to "tunnel" DE (Figurc 1) through the curvc (Munasinghc, 1999). However, in no case can it be expected that public environmental problems will automatically be solved as a result of economic growth, without the need for environmental policy (Arrow et aI., 1995). If we deny the need for environmental policy, we are renouncing the mechanism through which higher income could lead in some cases to a reduced environmental impact. And some recent studies show that economic and social policy may have a very important role in determining the emergence of the downward sloping part of the environmental Kuznets curves (Panayotou, 1995;Grossman, 1995;Torras, Boyce, 1998).
So, as mentioned by Spangenbcrg ct aI, 2002, while it is unrealistic to expect that economic growth per se (if only high enough) would improve distributional justice and reduce environmental pressures, it has been demonstrated that a delinking of growth and environmental pressures (emission levels and -increasingly -resource input) is possible in both the short and the medium term. This, however, does not happen automatically but needs deliberate policies. Green taxes, energy saving measures, employment subsidies and the like have all been repeatedly suggested (for an overview on international eco-tax studies, see Bosquet, 2000).

CONCLUSIONS
1. Discussions on dissociation of growth and environmental degradation are based on environmental Kuznets curves. Applying critical scientific argumentation, it can be stated that environmental Kuznets curves must be viewed as a hypothesis on the interface between economic growth and environmental quality.
2. A principal explanatory factor of environmental Kuznets curve is income. Additionally, technological and structural changes, including trade patterns, may also influence the environmental Kuznets curves pattern. These changes may in turn interact with price changes.
3. Some economists maintain the optimisic view that individual preferences of rich people eventually lead to a virtuous-circle relationship between rising income and environmental degradation. Several critical comments could be made concerning this point of view, but the most important point is related to the various manners in which environmental costs are displaced.
4. Hirsch's hypothesis that the share of income devoted to the purchase of status goods should rise in the face of economic growth holds true only under particular assumptions regarding the structure of individual preferences, even if the marginal utility of consumption is decreasing and the supply of status is fixed.
5. The process of globalisation may render the world's development more sustainable simply by pushing the world economy towards the decreasing part of the bell-shaped environmental Kuznets curves.