سیاست تغییرات آب و هوا، ساختار بازار و نشت کربن

The 1997 Kyoto Protocol on climate change obliges the industrialized countries to initiate the international effort of abating anthropogenic greenhouse gas (GHG) emissions. If such an initiative is to be taken, the associated competitive effects may lead to significant relocation of developed countries' energy-intensive production. This paper examines this issue. I adopt an oligopolistic structure combined with increasing returns to scale production technologies to represent the strategic interaction among the firms producing energy-intensive products. This representation is then embedded within a multi-regional computable general equilibrium model, which in turn is used for quantifying these relocational effects. The results suggest that significant relocation of energy-intensive industries away from the OECD may occur, depending on the type of market structure, with leakage rates as high as 130%, in which case GHG control policies in the industrialized countries actually lead to higher global emissions.

The 1997 Kyoto Protocol obliges developed industrial countries to initiate an international effort to abate anthropogenic greenhouse gas (GHG) emissions (Conference of the Parties to the United Nations Framework Convention on Climate Change, 1998).1 As only one part of the world would constrain its emissions under this agreement, many concerns are raised about the agreement's effects on trade and the location of industrial production. One particular effect is that the associated competitive effects may lead to significant offsetting emissions increases in the countries without controls—a phenomenon typically referred to in the literature as carbon leakage.2 The leakage process may occur through two different but related channels. First, the lower world prices of energy (reduced demand in the constrained economies exerts a downward pressure on energy prices) encourage substitution towards energy in countries without a carbon constraint. Second, energy-intensive industry may relocate from countries with carbon controls to those without, and export its energy-intensive products back to the constrained countries. The existing results on production effects and leakage from a carbon policy in industrialized countries are mainly derived from computable general equilibrium (CGE) models.3 Though these models treat the first channel relatively well, they fall short of adequately representing the industry relocation channel. For example, some of these models do not even have a separate representation of production and trade in energy-intensive products, whereas others, which include such a representation, suffer from at least three maintained assumptions: constant returns to scale (CRTS) technologies, perfect competition, and an Armington structure characterizing trade in energy-intensive products.4 Given the restrictive nature of the modeling assumptions, the scope for leakage tends to be quite limited in these models.5 This paper addresses these shortcomings by exploring more fully the effects of the market structure on the geographic distribution of energy-intensive production, trade, and leakage. In doing so, the paper captures two sources that may contribute significantly to the offshore production and leakage effect: the pro-competitive effect of carbon abatement policies in an initially monopolized industry, and the effect of entry and exit of firms in such an industry. These effects are captured by explicitly modelling the strategic interaction among firms producing energy-intensive products through spatial Cournot oligopolies with free entry and exit6 in a seven-commodity seven-region applied general equilibrium model of the world economy. With this richer representation of the leakage process, we find cases where significant relocation of energy intensive industries away from OECD occurs with leakage rates greater than 100%. Whether the Kyoto Protocol will go into force (or not) remains an open question, given the recent deadlock in international climate negotiations (Jacoby and Reiner, 2001) and the latter developments in Bonn and Marrakech (Babiker et al., 2002). Nevertheless, the general results of this paper are readily applicable to any attempt to implement climate policies in which a significant portion of the world is not included. The rest of the paper is organized as follows. Section 2 describes the data. Section 3 presents the numerical model. Section 4 describes calibration and baseline. Section 5 presents the results, and Section 6 provides some concluding remarks.

Most studies have found that the carbon leakage rate resulting from a policy of the size of the Kyoto Protocol to be in the range 5% to 25%. Yet, all these studies have adopted similar modeling structures that do not capture the phenomenon of industry relocation, particularly over the long run. By extending the structure of these models to include features such as economies of scales, market power, and richer representations of international trade, this paper demonstrates that there exists a wider scope of leakage than previously thought. The competitive effects induced by OECD abatement of GHGs can lead to significant increases in offshore energy-intensive production associated with relocation of firms. The extent of these locational effects is found to depend crucially on the representation of the market structure and the substitutability among the traded energy-intensive products. When energy intensive products are modeled as Heckscher–Ohlin goods, the global carbon leakage rate is found to range between 50% and 130%, in which case a policy to limit carbon emissions in the OECD has the perverse effect of increasing global emissions. The regional effects of such a policy vary widely, with most of the leakage occurring in China, India, and the dynamic Asian economies. These results demonstrate that implementing a policy of limiting carbon emissions that fails to include many regions of the world may, by ignoring the role of the global trading system, fail to achieve its stated ends. The striking results in terms of carbon leakage alone in the Heckscher–Ohlin cases are reasons for the world to pause in implementing subglobal carbon policies if there is a reasonable chance that these model features are realistic. I argue that increasing returns to scale and product homogeneity may not be implausible characteristics of the energy-intensive industry in the long run. First, energy intensive industries such as chemical and petroleum products are recognized to have large economies of scale. Second, technology diffusion through multinational corporations, harmonization of international trade, and the extending international web of communication are all forces that tend to gradually diminish price differences among regions. Hence, a Heckscher–Ohlin assumption is a reasonable or at least not an inconceivable approximation in the long run.