Imperfect Competition in Electricity Markets with Renewable Generation: The Role of Renewable Compensation Policies.

• Author: Brown, David P.

1. INTRODUCTION

   Reducing greenhouse gas emissions has become an increasingly important economic issue worldwide. Governments often employ policies to motivate the deployment of renewable generation to reduce the reliance on fossil fuels in electricity markets. There has been an increased use of competitive auctions for contracts to deploy renewable resources. As of July 2017, forty-eight countries have adopted renewable auctions to subsidize renewable energy (Attia et al., 2017). (1) There is substantial heterogeneity in the details of these procurement auctions. One important detail is the form of compensation that firms receive when they win a contract to build renewable capacity.

   In this paper, we analyze the impacts of different compensation policies in renewable procurement auctions in an oligopolistic setting where firms have market power and can invest in potentially heterogeneous renewable resources. We focus on the two most widely used compensation policies: a fixed-priced and premium-priced Feed-in Tariff (FIT). (2) Under a fixed-priced FIT, firms that win a contract receive a fixed-price per unit of output that is independent of the wholesale market price. Alternatively, under a premium-priced FIT firms receive the wholesale price plus a fixed mark-up (premium) per unit of output. (3) We demonstrate that the renewable compensation policy has important effects on both the nature of competition in wholesale markets and the characteristics of the renewable resources that win the renewable procurement auction.

   We employ a model where the regulator specifies a fixed quantity of renewable capacity that it aims to procure. Then, firms compete in a renewable procurement auction to deploy renewable capacity before competing via Cournot competition in a wholesale spot market. An important additional feature that needs to be considered when modeling firm behavior in electricity markets is the presence of forward markets in which firms trade in advance of the wholesale market to supply output at fixed prices. Forward contracts have been found to reduce market power incentives in the wholesale spot market (Wolak, 2000, 2007). Consequently, after the renewable auction, we allow firms to choose forward contracts before competing in wholesale spot markets.

   We find that firms have a stronger incentive to exercise market power in the wholesale spot market when their renewable generation is compensated under a premium-priced compared to a fixed-priced FIT. This arises because renewable output is compensated at the prevailing spot price. As renewable output expands, firms have a stronger incentive to withhold production from conventional generation in order to mitigate the reduction in spot market prices that is paid both to their conventional and renewable generation output.

   In addition, we find that increases in renewable output have an ambiguous impact on firms' incentives to sign forward contracts. Firms expand their forward output as renewable output expands under a premium-priced FIT, while they reduce their forward quantities when a sufficiently large proportion of their renewable output is compensated at a fixed-priced FIT. This arises because of the larger strategic incentive to forward contact in the premium-priced setting. This result is in contrast to the previous literature which has found that forward contracting unambiguously declines as renewable output expands (Ritz, 2016; Acemoglu et al., 2017). The increased incentive to forward contract under the premium-priced FIT mitigates some of the elevated incentives to exercise market power in the spot market, but prices remain unambiguously higher under the premium-priced setting for a fixed quantity of renewable capacity.

   In our model, firms compete in a renewable auction where the identity of the winner impacts firms' subsequent wholesale profits. Consequently, firms internalize the impact of the allocation of renewable capacity on their subsequent profits when making bidding decisions. At this stage of the model, we primarily focus on a setting with symmetric firms and no existing renewable capacity. In this setting, we demonstrate that the renewable compensation policy affects the types of resources that win the renewable auction. Under a premium-priced FIT, we identify conditions under which the "more valuable" renewable resource wins the auction. (4) The opposite result arises under the fixed-priced FIT. If the cost-reductions from the more efficient renewable resource being adopted are sufficiently large, they can dominate the elevated market power observed in the spot market under a premium-priced policy resulting in an overall increase in welfare.

   We extend the baseline model to allow a potential entrant to compete for the renewable capacity. Under a fixed-priced policy, the less valuable renewable resource continues to win the auction regardless of ownership. However, under the premium-priced policy, an incumbent wins the auction if the cost of production from conventional resources are sufficiently large or the entrant's potential renewable resource is not substantially more productive in high demand hours. Importantly, we identify conditions under which an incumbent with a less valuable renewable resource wins the auction to prevent entry. However, we also demonstrate via numerical simulations that an entrant with a "more valuable" renewable resource is more likely to win the renewable auction as the number of incumbents increase.

   These findings emphasize the various trade-offs under the premium and fixed-priced policies in renewable procurement auctions, and stress the importance of accounting for both the endogenous adoption of heterogeneous renewable resources and the nature of market competition. Our findings are consistent with the movement towards premium-priced policies to motivate investment in renewable resources whose output is more positively correlated with market demand. However, we demonstrate that firms have an elevated incentive to exercise market power in this setting, offsetting some (or all) of the gains associated with procuring the "more valuable" resource.

   Section 2 discusses our contribution to the literature. The model is detailed in Section 3. Section 4 presents the equilibrium results for the forward and wholesale (spot) markets. We detail the solution to the renewable auction in Section 5. Section 6 introduces a competitive fringe into our model. Section 7 provides extensions to our baseline analysis to demonstrate the robustness of our findings. Section 8 concludes. The proofs of all formal results are presented in the online Technical Appendix (see Brown and Eckert, 2019).

2. RELATED LITERATURE

   There exists a large literature that investigates the trade-offs associated with different renewable compensation policies. These studies often assume markets are perfectly competitive or that output decisions are made by a central planner or system operator, focusing on the incentives for renewable investment created by different compensation mechanisms (e.g., Lesser and Su, 2008; Garcia et al., 2012; Ambec and Crampes, 2017; and Antweiler, 2017).

   Several recent articles consider the implications of wholesale market power on the effectiveness of different forms of renewable compensation. In a closely related study, Oliveira (2015) considers a two-stage model in which firms with exogenous conventional generation capacity first choose investments in renewable capacity, and then act as Cournot competitors in the spot market. (5) The author compares outcomes when renewable generation is compensated by a fixed-priced versus a premium-priced FIT. The author highlights results that also emerge in our analysis; while premium-priced FITs can result in greater exercise of market power, premiums also provide a stronger incentive to invest in renewable capacity with production more highly correlated with demand.

   Our paper differs from Oliveira's (2015) analysis in several ways. We introduce forward markets, which play an important role on firm behavior. Our model allows firms to own a combination of fixed-priced and premium-priced FIT renewable contracts. We model renewable procurement auctions that are increasingly employed in practice. This form of procurement has important impacts on strategic behavior as firms consider the impact of their rivals' winning the auction on their subsequent wholesale profits when making their bidding decisions. In addition, we consider the relative incentives of incumbents and new entrants to invest in renewable generation. (6)

   Our study is also related to a recent literature that considers the interaction of increased penetration of renewables with forward markets. It has been established in prior studies that firms' incentives to exercise market power depend critically on the quantities committed to through forward contracts signed in advance at fixed prices (e.g., Allaz and Vila, 1993; Wolak, 2000, 2007; Bushnell et al., 2008). Ritz (2016) extends the strategic forward-contracting model of Allaz and Vila (1993) to consider the effect of intermittent renewables production owned solely by a competitive fringe, and finds that renewable output reduces firms' forward quantities. (7)

   In contrast, Acemoglu et al. (2017) examine the effect of shifting ownership of renewable capacity from non-strategic third parties to oligopolists who also operate conventional generation in a Cournot setting with forward markets. (8) The authors demonstrate that the "merit order" effect through which increased renewable generation reduces market prices is weakened as the proportion of renewable capacity owned by the Cournot producers increases. (9) Conventional generators who also own renewable capacity have an incentive to withhold production. While instructive, certain assumptions limit the generality of their results. In particular, the authors assume that conventional generators hold an equal amount of...