Optimal Price Design in the Wholesale Electricity Market.

• Author: Bigerna, Simona

1. INTRODUCTION

   One of the main pillars of the electricity market liberalization at the end of the 1980s has been the creation of organized electricity markets. The three main components are a day-ahead market (DAM) to determine quantity and price for the next day, an adjustment market (AM) to perform intraday operations and an ancillary service market (ASM) to ensure adequate dispatching resources, which are needed to build the reserve margins for the Transmission System Operator (TSO) security management of the transmission network.

   These markets were designed when conventional thermal sources (CTS) were predominant in the market and renewable energy sources (RES) were only a small fraction of the total supply. The actual configuration of the electricity markets is profoundly different from the ideal framework envisioned in those times. There is a growing conflict in the market among CTS and RES that goes beyond the desirable degree of competition and risks to endanger the entire market structure (Henriot and Glachant 2013). Transmission line congestion results in the creation of rents, and price signals for future investments may be distorted. Therefore, we deem it necessary to rethink the entire market design to provide a proposal for comprehensive reform.

   The main aim of this paper is to define an optimal design to determine zonal prices in an organized electricity market. We think that the market design must take into consideration the promotion of competition among all suppliers. This means promoting pro-competitive incentives to all generators to bid the marginal cost (MC) for every hour, which is the short-term variable cost, knowing that the cumulated difference (integral) between the short-term MC and the equilibrium System Marginal Price (SMP) in the long run will recover the fixed costs of the investments (Cardell et al. 1997, Hogan 2014). Specifically, we deem it necessary to take into full account the opportunity cost of RES for society when submitting bids to the market. In addition, the market design must recognize the existence of externalities associated with the transmission security management. This calls for designing correct price signals on the demand side through an optimal Ramsey (1927) pricing scheme.

   According to the uncontroversial theoretical principle of efficient market equilibrium, the regulation of the electricity market entails that each generator submits a bid to supply a certain quantity at a desired price and that each buyer submits a bid to purchase a certain quantity at a desired price. The time span is generally short: the typical electricity market is organized by the hour, e.g., Nordpool in the Scandinavian countries, EEX in Germany, Powernext in France, IPEX in Italy, OMEL in Spain, and PJM in the US, while some other markets are organized by the half-hour, as in Australia, New Zealand, and the UK (Green, 2005).

   Each bid is attributed a merit order; supply bids are cumulated according to an increasing price order to construct a market supply function, and demand bids are cumulated according to a decreasing price order to construct a demand supply function. The efficient price is determined by the intersection of supply and demand every hour, yielding the SMP. It is evident that this outcome is efficient insofar as the following four ideal conditions are satisfied. (1) The supply side is constituted by a large number of perfectly competitive units. The aggregate marginal cost function is non-decreasing, yielding a non-negative sloped supply function. There is no impediment to free trading, and there are no externalities in the market. Unfortunately, none of the four above-mentioned conditions are fully satisfied in the world's electricity markets.

   In practice, regulators have tried to improve efficiency, introducing various peculiar features in the electricity markets. In the UK, a 2013 reform (Energy Act 2013) introduced a Capacity Market (CM), which will help ensure security of electricity supply at the least cost to the consumer and Contracts for Difference (CfD), which will provide long-term revenue stabilization for new low carbon initiatives. In Italy, CFD were envisioned by regulations since their creation in 2004, while CM was effectively introduced with regulatory reform in 2014 and will start operations in 2016. In the EU, there is a coordinated effort under EU regulations to achieve a common model of market coupling among various national markets (Glachant 2010). In the market reform debate, geographical differences in the demand structure have not been adequately considered.

   In this paper, we propose a new and comprehensive set of rules for determining the efficient equilibrium price in the market, accounting simultaneously for the existence of supply market power, line congestion, RES abundance and heterogeneity of buyers' behavior (Cardell et al. 1997). As an empirical example, we elaborate upon the elementary bid data of the Italian IPEX market from the period 2010-2014 (as published by the Italian market operator, GME), to simulate counterfactual market outcomes in the DAM.

   The remainder of the paper is organized as follows. Section 2 presents the theoretical framework for the optimal market design. Section 3 shows the empirical methodology and describes the dataset that was used. Section 4 discusses the empirical results, and Section 5 presents policy implications and concludes the paper.

2. OPTIMAL MARKET DESIGN

   In a perfect theoretical market, the SMP provides an efficient solution insofar as there are only price taking agents (buyers and sellers) and there are no externalities. In reality, electricity markets contain several departures from the ideal model of perfect competition. We analyze four main causes of inefficiencies embedded in the specific functioning of the electricity market and discuss the appropriate solutions (Garber et al. 1994, Glachant 2010).

   The first cause is the existence of market power exercised by the large oligopolistic generators. The resulting equilibrium price can be different from the competitive solution because oligopolists may adopt a markup strategy for setting the price. This is a relevant issue especially when the exercise of market power is intertwined with transmission line congestion (Bigerna et al. 2016). We assume that it can be properly measured and can be contrasted with adequate pro-competitive regulatory actions (Garber et al. 1994, Bigerna et al. 2015b). For this reason, we do not dwell further on this issue and instead focus mainly on the other three causes of inefficiencies, which are related to the externalities arising from the abundance of RES supply and from line congestion.

   The second cause is constituted by the positive externality related to network security management, which has some characteristics of a public good (Tangeras 2012). In fact, the network management must take care of network security, which is an indivisible good. As a virtual thought experiment, consider a bolt of lightning that unexpectedly strikes a transmission line, causing it to trip and creating congestion between two adjacent zones. The resulting increase in the electricity load of the hit area yields a higher zonal price. This raises the issue of who should pay for this increased congestion cost. A pure zonal pricing mechanism would put the burden uniquely on the shoulders of the consumers in the hit area. If these consumers had the opportunity to make a decision in advance, however, they would probably have rescheduled their behavior optimally. Notice that if weather conditions are systematically different among different areas, i.e., lightning occurs more often in one area than in another, the consequences are certainly socially unfair.

   The third cause is constituted by the positive externality of the existence of RES. In other words, the large amount of RES creates a positive externality because RES provides relief from pollution and therefore provides a public good, which is clean air (deLlano-Paz et al. 2015). The localization of RES in a country is largely a consequence of the incentive schemes designed by the regulator. There is a normative issue here as well. If the incentive is equal (as is the case for most countries) in all areas of the country, this yields a profitability advantage to the south for solar technology or to windy coastal and hilly areas for wind technology. In other words, a citizen who invests the same amount of resources garners benefits in a region where RES are relatively more productive with respect to another area. This occurs because the combination of the same investment and different electricity generation output yields a higher profit because the incentive is proportional to the amount of electricity generation obtained (Henriot and Glachant 2013).

   The fourth cause is the negative externality arising from the uneven and specific patterns of the localization of RES in a country, which can be different from the distribution of the demand load in the country. There are two consequences, one in the short term related to the transmission line congestion (and network security), and the other in the long term related to the need for new line investment. In other words, congestion results from the peculiar localization of RES supply, such as solar where there is sunshine and wind where there is wind, which is not necessary coherent with the localization of the bulk of industrial and residential load. This can possibly result in market splitting and zonal price differentials (AEEG, 2014).

   In other words, like weather conditions, RES availability is not under consumers' control. If the south is more endowed with sunshine, it could create congestion in the transmission lines from south to north, or alternatively, a need for additional investment in new lines. This raises an immediate question: why is the cost of investment in new lines to evacuate RES supply borne by the TSO...