Equilibrium Analysis of a Tax on Carbon Emissions with Pass-through Restrictions and Side-payment Rules.

• Author: Diaz, Gabriel

1. INTRODUCTION

   Threats of global warming are the main driver behind the implementation of climate and environmental policies that seek to curb carbon emissions. To date, nearly 25% of global carbon (1) emissions come from the burning of fossil fuels to produce electricity, which is why most of the existing climate policies are targeted to this sector of the economy (Field et al., 2014). Some of these policies include carbon taxes and cap-and-trade programs (Chen and Tseng, 2011), renewable targets (Lyon and Yin, 2010), feed-in tariffs (Couture and Gagnon, 2010), and production tax credits (Wiser et al., 2007). The focus of this paper is on the long-term effects of a carbon tax with pass-through restrictions and side-payment rules, inspired by the current carbon emissions policy used in the electricity market in Chile.

   A carbon tax is a market-based regulation that forces agents to internalize the costs that carbon emissions impose on the environment. In theory, if the tax is set to a value that equals the social cost of carbon (SCC) emissions and the market is perfectly competitive, agents will adjust their consumption and production decisions until the marginal benefit that results from an additional unit of carbon emissions equals the SCC (Pigou, 1920; Metcalf and Weisbach, 2009). The result is a reduction of carbon emissions to the socially optimal levels, accounting for all future externalities that will be caused by carbon emissions that result from the current use of fossil fuels. However, practice is much more difficult than theory since estimates of the SCC are rather sensitive to assumptions about key factors such as discount rates, future emissions, and how climate will actually respond to increasing levels of carbon dioxide in the atmosphere (Pearce, 2003; Tol, 2008; Nordhaus, 2017).

   In electricity markets, carbon pricing policies aim at displacing carbon-intensive generation (e.g., coal) for other technologies with lower emissions rates (e.g., natural gas, hydro, wind, solar, etc.) by incorporating the SCC in the operating cost of each generation unit (Cramton et al., 2017). This can be accomplished directly, through a tax on C[O.sub.2] emissions, or indirectly, through a C[O.sub.2] cap-and-trade program (Chen and Tseng, 2011). As shown in Fabra and Reguant (2014), carbon pricing can have an immediate effect in the dispatch of generators in the short term if firms choose to pass through the full costs of emissions regulations. The result is a change in the system's supply curve that leads to an increase of electricity prices. (2) In the long run, carbon pricing changes investment incentives since carbon-intensive technologies are dispatched less often and become less profitable than cleaner technologies (Chen and Tseng, 2011; Nelson et al., 2012). Hereinafter, we refer to this pricing mechanism as a standard carbon tax or as a standard cap-and-trade program.

   Currently, there are 47 jurisdictions that have adopted some form of carbon tax or cap-and-trade program, which cover nearly 15% of global greenhouse gas emissions (WB, 2017). Most of these initiatives have been implemented in relatively developed countries or individual states in Europe, North America, and Australia, with a few exceptions in developing nations. In South America, Chile was the first country to enact a carbon tax, which became active on January 1 2017 (IEA, 2018). The tax was set to 5 $/C[O.sub.2] and applies to all stationary sources with a capacity of at least 50 MW. Although the tax rate is modest compared to carbon taxes in developed countries such as Denmark (27 $/tC[O.sub.2]), France (36 $/tC[O.sub.2]), Switzerland (87 $/tC[O.sub.2]), and Sweden (140 $/C[O.sub.2]) (WB, 2017), this initiative has been described as a positive first step to reduce carbon emissions in the electric power sector. (3,4)

   However, there is one aspect of the carbon tax in Chile that sets it apart from other tax or cap-and-trade programs in the rest of the world. The law has a pass-through restriction that states that carbon charges cannot be reflected in the dispatch and pricing of electricity in the real-time market. The regulation also states that generation firms that face the tax and that cannot cover their full costs (i.e., marginal cost plus carbon charges) from spot prices--that, as mentioned, do not reflect carbon charges--should receive a side payment that is financed by all units operating at a given hour, including inframarginal generators that do not use fossil fuels. Clearly, the current implementation of the carbon tax in Chile has no effect on carbon emissions in the short term due to the existing pass-through restriction. However, the policy does change investment incentives in the long term since firms are forced to absorb an administrative definition of carbon emission costs every year.

   In this paper, we develop an equilibrium model with endogenous investments in generation capacity to study the long-term economic effects of the current emissions policy used in Chile. We benchmark the efficiency of this policy using two additional equilibrium models that assume a standard carbon tax implementation and a cap-and-trade program without pass-through restrictions and side-payment rules. Since we assume perfectly competitive markets, we compute equilibria for the two benchmark models using linear programs. For the Chilean emissions policy we find an equilibrium using an iterative Gauss-Seidel algorithm, which allows us to consider the pass-through restriction and the side-payment rules that determine the annual carbon charges per generator.

   We study the effect of these policies using three different case studies that resemble different hypothetical market conditions in Chile for year 2050 under increasing tax levels. Our results indicate that the current implementation of the carbon tax in Chile is rather inefficient compared to emissions policies without pass-through restrictions and side-payment rules. Furthermore, we find that increasing the tax level under the current implementation in Chile yields, in general, higher average electricity prices and higher emissions levels than under a standard carbon tax. In fact, under the current policy in Chile carbon emissions can even increase as a result of a rise of the tax level. This implementation is also detrimental for the development of carbon-free technologies with low marginal costs, such as wind and solar, which must also absorb some carbon charges to support the side-payments for generators that do emit carbon dioxide in periods when prices are not sufficient to recover their full costs, which is it at odds with the current renewable and environmental goals of the country.

   While there is a large body of literature focused on the impact of standard implementations of carbon taxes and cap-and-trade programs in competitive electricity markets (Nelson et al., 2012; Vera and Sauma, 2015; Eser et al., 2016), there are few studies that have quantified the effects of carbon pricing rules in imperfect markets. The market failure that receives most attention is market power. For instance, Downward (2010); Perez de Arce and Sauma (2016); Limpaitoon et al. (2014); Siddiqui et al. (2016) demonstrate that if electricity markets are not perfectly competitive, environmental policies can have unintended consequences on electricity prices, investments, and carbon emissions. Policy uncertainty and risk aversion can also have an impact on the effectiveness of carbon policies (Bergen and Munoz, 2018). However, to our best knowledge, the existing literature on the impacts of policy exceptions and administrative rules is limited to features such as priority dispatch (5) of renewable generators (Deng et al., 2015) and the potential for carbon leakage or emissions spillover when carbon policies are only applied to subregions of interconnected electric power systems (Chen, 2009; Bushnell and Chen, 2012).

   In this context, we contribute to the existing literature by analyzing the economic effects of the current pass-through restrictions and side-payment rules of the carbon emissions policy used in the Chilean electricity market. Our analyses demonstrate how a regulator's (presumed) predisposition to avoid price increases in the short term as a consequence of a carbon tax--in this case, by implementing pass-through restrictions--can lead to inefficient market outcomes in the long term. We want to highlight that such predisposition to try to protect one side of the market (i.e., customers) through second-best policies instead of implementing first-best ones that focus on overall market efficiency is not unique to this case. There are many examples of regulatory authorities elsewhere that also choose policy instruments or market designs that aim at protecting consumers in the short run at the expense of potential reductions of market efficiency in the long run (Hogan, 2005; Joskow, 2008; Szolgayova et al., 2008; Jenkins, 2014; Newbery, 2016; Cramton, 2017; Munoz et al., 2018). We demonstrate that pass-through restrictions can be detrimental in the long term, even if they do protect consumers from price increases in the very short term as a consequence of the implementation of a carbon tax. Based on previous work by Greenberg and Murphy (1985), we also contribute to the existing literature by developing an assessment framework to compute long-term equilibria in electricity markets subject to complex carbon tax rules that cannot be represented through closed mathematical forms.

   We structure the rest of the paper as follows. In Section 2 we provide a brief overview of the Chilean electricity market and a detailed description of the current policy used in Chile. In Section 3 we describe the equilibrium models used to analyze the long-term effects of such carbon policy. In Section 4 we present some general findings. In Section 5 we present a summary of data assumptions for our case studies and our...