Two-settlement system for the Brazilian electricity market
Introduction
Electricity markets have been put in place in many regions worldwide as an alternative to the traditional vertical structure in which the same company is responsible for producing, transmitting and distributing electricity to the end consumers (Joskow and Schmalensee, 1983; Hunt, 2002). The objective of the electricity market is to bring efficiency to electricity production and trading, resulting in system adequacy, suitable payback to investors, and fair price to consumers.
However, the electricity market is not organic, naturally emerging among people or companies willing to sell and purchase goods or services. It is a market that must be designed (Cramton, 2017; Wolak, 2019a). The Regulator, or the policymaker (called the Regulator in this paper), is responsible for designing the market.
When the Regulator designs the market, it has to manage two conflicting targets: supply reliability and low prices to consumers. They are conflicting targets because the more the investment made, the more reliable the system is. In theory, absolute reliability leads to infinite investment. If there is too much investment, the price of electricity rises. It is necessary to raise funds to match the investment. However, low electricity prices may not be sufficient to provide reliability.
In pursuit of the best market design, the Regulator must satisfy both the short- and long-term efficiencies. They make the best use of existing energy resources, and foster the economic signals and incentives for efficient investments in new sources by expanding the installed generation capacity or replacing the existing generation plants with more efficient ones with low-cost electricity supply.
To address both conditions, the Regulator must ensure that the competitive activity is effective, and that the rules of the proposed design eliminate, or at least mitigate, any anti-competitive market participants’ strategy that may artificially raise the electricity prices for extraordinary profits using unilateral market power.
In Brazil, the electricity market was designed considering the huge integrated power system that is responsible for delivering more than 99% of the country's electricity consumption; the hydro-dominated supply, that provides about 70% of the electricity consumption depending on the hydrological conditions of the year; and the electricity demand that increases by about 3% per year (Munhoz, 2017).
An Independent System Operator (ISO) is responsible for scheduling and operating the integrated power system in the Brazilian market. The ISO uses a cost-based dispatch model based on a stochastic dual dynamic algorithm (Pereira and Pinto, 1991) to support its decisions and schedule, for each period, all thermal power plants with a marginal cost lower than the marginal operation cost (MOC) of the system calculated by the model. Hydropower plants and other renewables supply the remaining demand.
Hence, in the Brazilian market, the market participants do not define the short-term price by their willingness to purchase and sell electricity. Computational algorithms set the short-term price, formally called Differences Settlement Price (PLD, in Portuguese), based on data of demand, supply, and system constraints. Therefore, it is a cost-based centralized market design.
Normally, the PLD is equal to MOC. However, the PLD and MOC may have different values for the same period because (i) the PLD has a regulatory cap and a floor and (ii) the PLD calculus does not consider internal transmission constraints of each market in the four Brazilian zones, since each zone must have the same short-term price in all-electric nodes. Nonetheless, when the MOC is between the PLD cap and floor, and there are no internal transmission constraints inside the zones, the PLD value equals the MOC value.
The short-term electricity price in Brazil was set week-ahead in three load levels (light, intermediate and heavy) for each of the four electricity zones in the Brazilian market until the end of 2020. Each week, the Brazilian market had three prices in each zone (Munhoz, 2017). However, from January 1st, 2021, the market changed the temporal price granularity. The electricity price is set day-ahead on an hourly basis (MME, 2019).
The Brazilian cost-based market employs the single-settlement system to clear the electricity. This system adopts a hybrid mechanism in which the amount of electricity measured during production and consumption in the real-time market is cleared with the electricity price set during the operation's schedule in the day-ahead market. The result is that the intersection between demand and supply does not set the electricity price neither in the day-ahead market nor in the real-time market.
This paper proposes a regulatory improvement in the Brazilian electricity market design by modifying the electricity settlement system. The objective is to improve the market efficiency given by the relationship between independent power suppliers and large consumers who may trade in the wholesale market, which, for simplicity, are called generators and demand. The retail market, in which small consumers participate, is not addressed in this work.
The paper demonstrates the flaws of the Brazilian settlement system and how the use of the two-settlement system enhances the competitiveness of the Brazilian market, prices electricity correctly at the moment of production and consumption, encourages market participants to furnish accurate data for the ISO, and mitigates anti-competitive practices like the use of unilateral market power.
A concrete case of the possible use of unilateral market power in the Brazilian electricity market and how the two-settlement system can avoid these problems is presented at the end of the paper. It uses real data that measures the frequency that the thermal-plant companies modified their availability information to the Brazilian ISO.
The organization of the paper is as follows. Section 2 describes the main concepts regarding short and long-term efficiency. Section 3 explains the consequences for the electricity prices when market participants exercise unilateral market power. Section 4 describes the two-settlement system and the Brazilian electricity market settlement system. Section 5 presents the proposal of the paper. Section 6 provides an analysis of the possible use of market power in the Brazilian market, and section 7 concludes the paper.
Section snippets
Short and long-term efficiency
The market design must use the available supply resources following the economic merit order to achieve short-term efficiency. The electricity consumption must be supplied, beginning with the plants with the lowest production costs and ending with the plants with the highest production costs until the demand is met. Fig. 1 (a) shows the ideal supply allocation to meet demand following the economic merit order. The marginal source of supply, which meets the last unit of electricity demanded,
Market power in the electricity market
Market participants exercise power when they are large enough to modify the market price by altering their electricity supply (Borenstein, 2000; David and Wen, 2001). For example, the market participant may increase the market price by offering only part of its power generation capacity or declaring its unavailable equipment to the ISO. Certain studies (Wolak, 2003; Sweeting, 2007; McRae and Wolak, 2017) demonstrate the use of market power in several electricity markets worldwide.
The exercise
Electricity settlement system
Electricity is produced and consumed continuously. However, it is clustered in time intervals of five, ten, or 30 min, or 1 h, depending on the market design for trading. For instance, if a market adopts a cluster of 1 h, the electricity is cleared 24 times per day, one for each hour.
When the Regulator designs the market, it can divide the electricity clearing process into settlement periods. The electricity can be cleared once in a single-settlement system, twice in the two-settlement system,
Proposal
Fig. 5 illustrates that G6 informed maintenance in the operation's schedule but produced electricity in the real-time operation. Table 2 shows that G6's revenue was $ 4000.00 in the two-settlement system and $ 5600.00 in the Brazilian settlement system.
If G6 had informed its real availability in the operation's schedule, its revenue in the two-settlement system would be $ 5600.00 (Eqn 1), a revenue higher than in the assumption that G6 informed maintenance in the two-settlement system. However,
Possible use of market-power
Based on the temporal granularity in effect in 2019, week-ahead in three levels of load, the Brazilian ISO measured the frequency that the thermal-plant companies modified their availability information from the week-ahead to the day-ahead schedule (ONS, 2019b). Fig. 7 computes the data from January to August 2019. The measured frequency computes only the cases where the companies informed the ISO that they would operate their thermal plants inflexibly to generate a certain amount of
Conclusions and policy implications
This work shows how the two-settlement system is superior to the single-settlement system currently employed in the Brazilian electricity short-term market.
In the Brazilian hybrid single-settlement system, the market participants have incentives to hide the plant's real availability in the operation's schedule to raise the electricity price for better profits. Then, these plants operate in the real-time market, clearing the electricity at the price of the operation's schedule that they helped
Disclaimers
This paper is the Author's entire responsibility and it does not necessarily represent the opinion of the Brazilian Electricity Regulatory Agency (ANEEL), the institution where the author is a staff member.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The first version of this study was one of the winners of the SECAP Energy 2019 Award, an initiative of the Brazilian Ministry of Economy.
CRediT authorship contribution statement
Fernando Colli Munhoz: Conceptualization, Methodology, Writing – original draft, Visualization, Investigation, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
I thank Frank A. Wolak for comments that improved the manuscript, and Larissa M. Brandini and John Byers for reviewing the paper.
References (26)
- et al.
Cournot competition, forward markets and efficiency
J. Econ. Theor.
(1993) Understanding competitive pricing and market power in wholesale electricity markets
Electr. J.
(2000)The necessity of more temporal granularity in the Brazilian short-term electricity market
Sustain. Energy Grids Netw.
(2017)Brazilian Electricity Regulatory Agency. Resolution nº 290, August 3rd, 2000. Wholesale electricity market rules
Brazilian Electricity Regulatory Agency. Normative resolution nº 843, April 2nd, 2019. Establishes criteria and procedures for the electricity monthly schedule operation and for the differences settlement price calculations
Electricity market design
Oxf. Rev. Econ. Pol.
(2017)- et al.
Market power in electricity supply
IEEE Trans. Energy Convers.
(2001) Brazilian Presidential Decree nº 5,163, July 31st, 2004. Regulate the Electricity Trading
(2004)Competitive Electricity Market Design: a Wholesale Primer
(1998)Making Competition Work in Electricity
(2002)
Market Design Executive Summary. Report
Markets for Power. An Analysis of Electric Utility Deregulation
Two-settlement systems for electricity markets under network uncertainty and market power
J. Regul. Econ.
Cited by (5)
The electricity market in Brazil: A multilevel perspective of sector agents for liberalization to residential consumers
2023, Energy for Sustainable DevelopmentThe impact of hourly pricing for renewable generation projects in Brazil
2022, Renewable EnergyCitation Excerpt :However, from January 1st, 2021, the market changed the temporal price granularity to a day-ahead model on an hourly basis [30]. In Brazil, mathematical algorithms define the short-term price, or spot market price, based on supply, demand, and constraints data, officially called Differences Settlement Price (PLD, in the Portuguese acronym) [31]. The information's source used to calculate the price is all foreseen and prior to the actual operation of the system, considering the declared availability of generation and the expected consumption of each submarket [1].
Application of a data-driven DTSF and benchmark models for the prediction of electricity prices in Brazil: A time-series case
2023, Journal of Renewable and Sustainable Energy