Renewables, Allowances Markets, and Capacity Expansion in Energy-Only Markets.

• Author: Falbo, Paolo

1. INTRODUCTION

   The past five years have witnessed a systematic decrease in the electricity sector operational profitability and a consequent decline in capacity investments in market-based systems (Financial Times, 2015; The Economist, 2015). Overcapacity of fossil fuel electricity generation and a larger share of renewable electricity generation (1) are among the main causes (Koch et al., 2014). Particularly in Europe, renewables (2) have not just put pressure on margins; they have also transformed the established business model of utilities (The Economist, 2013). Electricity from renewables has favorable access to the grid, squeezing the earnings of polluting generation. However, this preferential grid access is not solely the result of policies favouring renewables. It is also logical in electricity markets that operate based on the merit order: since the marginal cost of renewables is virtually zero, grids would take their electricity first anyway. In this paper, we explore some of the implications of this changing business model on the long-term capacity expansion decision of the electricity sector when polluting emissions from fossil fuel generation are regulated by an Emissions Trading System (ETS).

   The long-term capacity expansion decision in the electricity sector (the so-called energy-mix decision) is key to the goal of a low-carbon economy. Electricity and heat generated by fuel combustion are responsible for approximately 42% of C[O.sub.2] emissions worldwide (International Energy Agency, 2017). Consequently, electricity sectors in many countries have been brought under ETS regulations. This is the case in Europe (EU ETS, European Union Emissions Trading System), the U.S. (Californian cap-and-trade program), South Korea (KETS, Korea Emissions Trading Scheme), and China (pilot ETS in the Province of Guangdong). In particular, the EU ETS is a cornerstone of the European policy to reduce C[O.sub.2] (80% by 2050, see European Commission, 2011) and, ultimately, to combat climate change. Research studying the energy-mix decision in the presence of an ETS is slowly emerging and has primarily focused on the short-term effects, e.g. the fuel switching effect (see Kirat and Ahamada, 2011, and references therein). We contribute to this literature by investigating the impact of emission caps on long-term capacity expansion decisions in energy-only markets. The policy implications of our work directly speak to the debate about the role of capacity markets and equivalent administrative interventions.

   Central to this problem is the tradeoff on profits associated with increased renewable generation. Investments in renewables drive fossil fuel plants out of the market, resulting in costly idle capacity. This is the so-called merit order effect. Yet, investments in renewables generate higher rents because green generation can be sold at the marginal cost of fossil fuel plants. Which of the two effects dominates depends on which generation is at the margin. This means that electricity producers might have an incentive to withhold investments in renewable capacity. The market structure and the regulatory framework are crucial for determining the extent to which the electricity sector withholds capacity investments. In recent contributions, Murphy and Smeers (2005), Zottl (2011), Murphy and Smeers (2012), and Grimm and Zoettl (2013) investigate capacity investment incentives when markets are not competitive. These authors show that withholding capacity investments can in fact increase profits, ultimately hampering adequate capacity installation. These papers develop their analysis either abstracting from the presence of emission regulations, or treating emission regulations in the form of an ETS as given. In particular, the price of allowances associated with an ETS is either omitted or treated as a given parameter. (3) However, in the energy-mix decision, which is long-term, the price of allowances heavily depends on the energy-mix itself. Consequently, such a price should be treated (as we do here) as an endogenous variable and should be part of the long-term capacity expansion decision.

   In this paper, we contribute to this literature by investigating how an ETS affects the profits resulting from long-term capacity expansion decisions. This line of analysis adds to the growing literature that stresses the need to account for the full effects of coexisting emission constraints and renewable energy source policies due to the sometimes conflicting incentives of the stakeholders involved. (4) Acknowledging these effects is critically important for the design of long-term electricity markets since conflicting incentives can often lead to suboptimal outcomes, or even outcomes in contrast to the compelling goals of the environmental policy. For example, Acemoglu et al. (2017) investigates the incentives of fossil fuel energy producers to increase the share of renewables in their energy portfolios. They demonstrate, in an oligopolistic setup, that the installation of renewable capacity can actually decrease net welfare as a result of reductions in energy production when the supply of renewables is high. We also assume a cooperative oligopoly in capacity investments and adopt an equivalent monopolist set up to examine the interplay between incentives to the electricity sector and prices of electricity and allowances.

   The monopolist set up allows us to transparently investigate the potential distortion of market prices in energy-only markets via an ETS. (5) We do not undertake a full-fledged welfare analysis. Rather, we focus our attention on key socio-economic variables, such as the percentages of renewable generation and fossil fuel generation, the level of electricity prices, and the level of profits accruing to the electricity sector, and compare the solutions under a Business-As-Usual (BAU) scenario and under an ETS scenario.

   Our theoretical results show that the monopolist has an incentive to drive the allowances price to the level of the penalty for non-compliance. (6) This occurs when allowances to cover emissions associated with fossil fuel generation are insufficient. Thus, the decarbonizing potential of an ETS is weakened. We illustrate this result with a quantitative example and present under which limiting market setting renewable capacity is increased. These results provide insights into the observed decline of new investments in renewables in Europe, (7) where the flagship ETS is entering its fourth (more stringent) Phase and the existing share of renewable capacity is already significant. As such, our analysis also contributes to the current discussion on the reform of energy and environmental policies.

   The remainder of the paper is set out as follows. In Section 2, we develop an analytical model of the long-term capacity expansion decision of a monopolist that can supply electricity both from fossil fuel and renewables plants. In Section 3, we solve the long-term capacity expansion problem of the monopolist and describe how the installation of renewable capacity impacts on profits. In Section 4, we illustrate the combined effects of increased renewables penetration and more stringent emission regulations by numerically solving the long-term capacity expansion problem under two scenarios and for three distinct market settings. Section 5 discusses the implications of relaxing the key model assumptions of monopoly and perfectly inelastic electricity demand, as well as relaxing various modelling choices (adjustments of capacity expansion; inter-temporal allowance banking; and regulatory adjustments). Section 6 concludes. The Appendix reports several details of the solution to the long-term capacity expansion problem, with a particular focus on a sensitivity analysis with respect to two key model parameters.

2. MODEL

   2.1 Market Structure

   We consider an electricity sector endowed with polluting and green technologies. We also assume that plants are owned by few large electricity producers, so that the electricity sector can be seen as an oligopoly. Around the world, electricity markets remain highly concentrated, with few major national producers controlling a substantial share of the domestic capacity. For example, according to EIA, (8) the first 108 U.S. electricity producers (over a total of 4,138) accounted for 60% of the national nominal capacity in 2017. In addition, in the nine largest European national markets, (9) approximately 3,208 electricity producers covered 95% of net domestic demand in 2016, but only 30 of them had a domestic market share larger than 5%. (10) Yet, cross-border electricity transmissions are increasing, contributing to the development of a more competitive electricity market.

   In order to simplify our analytic treatment and concentrate on the price distortion of an ETS, we assume that electricity production is sold through the standard system of uniform auction. This price fixing mechanism and the presence of a share of smaller competitive electricity producers can exert a significant pressure on major producers to make them bid at the marginal cost (as if they were acting under perfect competition). Notice that assuming collusion on the electricity market by producers that, consequently, would charge a constant markup on top of their marginal costs would leave our main conclusions unaffected. Indeed, such a markup would not eliminate the economic incentive to profit from exceeding the emissions cap. This will be discussed in more detail later. Moreover, we assume that electricity producers cooperatively act to decide the energy-mix. By adopting a cooperative oligopoly (or, equivalently, a monopolist set up), we obtain a tractable model of the aggregated energy-mix decision. This allows us to obtain an analytic solution of the problem and to transparently describe the distortion mechanism that an ETS can spawn. Besides, the adoption of a cooperative oligopoly is not an...