Impact of Permit Allocation on Cap-and-trade System Performance under Market Power.

• Author: Wang, Mei
• Position: Report

1. INTRODUCTION

   Climate change, caused mainly by increasing greenhouse gas emissions, has been recognized as a serious challenge to the survival and development of human beings. It has been universally agreed that all countries need to take actions to mitigate global C[O.sub.2] emissions (Zhao et al., 2016; Yang et al., 2017; Zhang et al., 2017b). In this context, emissions trading has become an important policy instrument for controlling C[O.sub.2] emissions (Labandeira et al., 2009; Lanz and Rausch, 2016; Wu et al., 2016). C[O.sub.2] emission permit allocation is the foundation of an emissions trading system (ETS); popular C[O.sub.2] emission permit allocation methods for firms in the existing carbon markets include grandfathering, benchmarking, and auctioning (Wang and Zhou, 2017). Under grandfathering, firms freely receive C[O.sub.2] emission permits according to their historical C[O.sub.2] emissions. By benchmarking, firms are freely allocated C[O.sub.2] emission permits based on their outputs. Auctioning requires firms to buy C[O.sub.2] emission permits from auctions organized by governments. Despite the existence of different allocation methods, debates about how to allocate C[O.sub.2] emission permits in different industries still exist.

   According to the Coase theorem, if there is no transaction cost in a competitive carbon market, the market equilibrium will be cost-effective and independent of the C[O.sub.2] emission permit allocation method. Nevertheless, Hintermann (2011) and Hintermann (2017) show that the existing carbon markets, such as the European Union (EU) ETS, are unlikely to be competitive, and power firms holding a large amount of C[O.sub.2] emission permits may have market power to steer the C[O.sub.2] price away from the cost-efficient level. Although market power may not exist when the number of firms covered is very large, it is likely to be a serious problem for some regional carbon markets (Maeda, 2003; Antelo and Bru, 2009; Montero, 2009; Alvarez and Andre, 2015).

   Some previous studies have been devoted to examining the impact of C[O.sub.2] emission permit allocation on the cost-efficiency of ETS in the presence of market power. The efficiency principle refers to achieving C[O.sub.2] emission reductions at the lowest cost (Eshel, 2005; Hagem and Westskog, 2009; Montero, 2009; Hahn and Stavins, 2011; Meunier, 2011). Hahn (1984) first points out that the presence of market power in the carbon market can lead to an efficiency loss, which is dependent on free initial allocation of permits. Hahn finds that, if the dominant firm is a seller of C[O.sub.2] emission permits, it tends to act like a monopolist and hold back C[O.sub.2] emission permits to drive up the C[O.sub.2] price. On the other hand, if it is a buyer of C[O.sub.2] emission permits, it has an incentive to act as a monopsonist and buy fewer C[O.sub.2] emission permits to keep the price lower. The efficient solution is to allocate the dominant firm with the exact amount of permits that it requires, thereby removing it from the carbon market. These results still hold in the case of multiple dominant firms acting as Cournot players in the carbon market (Westskog, 1996). Maeda (2003) derives formulae for estimating the degree of market distortion and shows the existence of a threshold for effective market power. Andre and Arguedas (2018) show theoretically that the distortion of a firm's technology adoption is dependent on a dominant firm's initial allocation of permits, which seems to be aligned with the conclusion drawn by Hahn (1984). Several other studies, e.g., Antelo and Bru (2009), Alvarez and Andre (2015), and Jiang et al. (2016), find that auctioning also distorts the C[O.sub.2] price when market power exists in the carbon market.

   Past studies have also contributed to comparing C[O.sub.2] emission permit allocation methods from the perspective of fairness. The fairness principle is often related to more general concepts of distributive justice (Rose, 1990). Of the alternative allocation methods, grandfathering is often regarded as a fair allocation rule, as each firm has the right to emit (Zhou and Wang, 2016). However, empirical evidence shows that some energy-intensive firms covered in the first two phases of the EU ETS passed the C[O.sub.2] costs on to consumers, even though they received the C[O.sub.2] emission permits freely from grandfathering, resulting in windfall profits (Sijm et al., 2006; Zachmann and Hirschhausen, 2008; Alexeeva-Talebi, 2011; Nelson et al., 2012; Jouvet and Solier, 2013; Fabra and Reguant, 2014; Hintermann, 2016; Meleo et al., 2016; Woo et al., 2017; Woo et al., 2018). Besides, several scholars argue that C[O.sub.2]-efficient firms should be either compensated or rewarded for previous emission reduction efforts, and, thus, that benchmarking is fairer than is grandfathering (Groenenberg and Blok, 2002; Zetterberg, 2014).

   The choice of an appropriate allocation rule needs to consider both efficiency and fairness principles (Zhou and Wang, 2016). As the efficiency loss resulting from market power depends on the initial allocation of emission permits, it is worth analyzing the impact of the C[O.sub.2] emission permit allocation method on the cost-effectiveness of ETS in the presence of market power, which is the main purpose of this paper. Meanwhile, we examine the C[O.sub.2] cost pass-through under different C[O.sub.2] emission permit allocation methods, which is regarded as a fairness principle in ETS (Wang and Zhou, 2017). By considering the specific characteristics of each industry, this paper finally yields some policy implications as to how to choose a proper emission permit allocation method for different industries.

   In this paper, we first present a basic model in which the dominant firm is assumed to have market power in both the carbon market and the product market. This assumption has been made in earlier studies, such as Hintermann (2011), and is likely to be realistic due to the following considerations. The carbon market covers C[O.sub.2]-intensive firms from different industries. Power firms tend to emit more C[O.sub.2] emissions than do other firms and, thus, probably have market power in the carbon market. Power firms having market power in the carbon market are generally dominant firms in power markets, as regional power markets are almost geographically isolated (Meunier, 2011). The assumption has also been relaxed in Section 4 by assuming that multiple market power firms exist in the carbon market and that all firms in their product markets are entirely competitive.

   This paper contributes to the literature by analyzing the effect of the C[O.sub.2] emission permit allocation method on a dominant firm's price manipulation in ETS in the presence of market power. Furthermore, this paper examines three popular C[O.sub.2] emission permit allocation methods from the perspective of both fairness and efficiency principles. In the field of regulatory economics, this paper identifies different industry characteristics and provides policy makers with suggestions for choosing a proper emission permit allocation method for different industries.

   The remainder of this paper is organized as follows. In the next section, we describe briefly the three main C[O.sub.2] emission permit allocation methods: grandfathering, benchmarking, and auctioning. Section 3 first provides a basic model for the case in which one firm has market power, and the other firms in the market are price takers. The main results derived from the model are reported, including the efficiency loss and the C[O.sub.2] cost pass-through under different C[O.sub.2] emission permit allocation methods. Based on the modelling results, we present the policy implications regarding how to choose a C[O.sub.2] emission permit allocation method for different industries. Section 4 gives a more general model, with multiple market power firms, and provides some extended discussions and implications. Section 5 concludes this study.

2. C[O.sub.2] EMISSION PERMIT ALLOCATION METHODS

   Theoretically speaking, C[O.sub.2] emission permit allocation methods may be classified broadly into an indicator approach, an optimization approach, a game theoretic approach, and a hybrid approach (Zhou and Wang, 2016). In the existing ETSs, grandfathering, benchmarking, and auctioning are used widely (Zhang et al., 2015; Wang and Zhou, 2017).

   Grandfathering means that the allocation of C[O.sub.2] emission permits is based on firms' historical C[O.sub.2] emissions (Zetterberg et al., 2012). It is a free allocation method, and firms with higher C[O.sub.2] emissions in past periods will receive more C[O.sub.2] emission permits in later periods. As grandfathering has the advantages of simplicity and maintaining the competitiveness of international firms and the potential for reducing C[O.sub.2] emission leakage (Schmidt and Heitzig, 2014; Hintermann, 2016), it is the most widely used permit allocation method in the early stage of ETSs, including the first period of the EU ETS and most China pilot ETSs (Zhang et al., 2017). Though grandfathering ensures the wide acceptability of firms covered in ETSs, C[O.sub.2]-efficient firms regard this as an unfair permit allocation method, based on the reasoning that firms with higher historical C[O.sub.2] emissions caused more damage in the past and, thus, have a larger responsibility to reduce C[O.sub.2] emissions in the future.

   Suppose that [e.sub.i] is the amount of free allocation of C[O.sub.2] emission permits for firm i, f is the C[O.sub.2] emission permit allocation coefficient, and [e.sup.0.sub.i] is the amount of historical C[O.sub.2] emissions of firm i in the base year(s). The C[O.sub.2] emission permit allocation coefficient and the base year(s) are set by policy makers. The initial C[O.sub.2] emission permits of firm i under grandfathering are equal to the product of the C[O.sub.2] emission permit...