Theory and practice of price formation in the German wholesale electricity market
Analysis of bidding behavior and the economic efficiency of power plants on the day-ahead market in 2025
The day-ahead (DA) market is the central short-term price signal for European electricity trading. Prices are determined using the pay-as-cleared method, in which all successful bids receive the same price set by the marginal power plant.
Since the energy price crisis of 2022 in particular, there has been intense debate as to whether this method could lead to potential excess profits.
This project systematically examines how bidding behavior and the economic efficiency of power plants will look in 2025. It analyzes the historical basis of price formation, the actual bid curves, and the economic efficiency of various power plant technologies.
Historical development of the European day-ahead market and the price formation mechanism
Today’s market structures have developed over decades:
- 1990/1991: The first competitive electricity pool with a uniform clearing price is established in England/Wales, followed a year later by Norway with a similar market model [1].
- 1996: An EU directive establishes the framework for electricity market liberalization – without specifying the pricing model [2].
- 2000/2001: The first spot market transactions take place in Germany (2000) and France (2001) – both with a pay-as-cleared mechanism [3], [4].
- 2008: The electricity exchanges of both countries merge [4].
- 2021: Much of Europe is linked via a uniform auction procedure known as Single Day-Ahead Coupling (SDAC). SDAC ensures efficient cross-border pricing and optimizes the flow of electricity across national borders. The SDAC members are shown in Fig. 1 [5].
Analysis of bidding behavior on the day-ahead market
According to economic theory, power plants in the pay-as-cleared mechanism offer electricity at their marginal costs in order to maximize their profits on the day-ahead market. Whether this theoretically predicted bidding behavior corresponds to the bids actually observed on the market was analyzed in the second part of the project.
Methodology
For the year 2025, three different market situations were evaluated by comparing bid curves, generation structure, load, and day-ahead prices:
- Low renewable energy generation with low load
- Low renewable energy generation with medium to high load
- High renewable energy generation with medium to high load
In addition, relevant characteristics of bidding behavior were identified. Finally, the bid curves were compared with a modeled merit order [7] to compare marginal costs with actual bids.
Results
Three key findings emerge from the analysis of bidding behavior on the German day-ahead market in 2025.
Particularly striking is the high proportion of very low bids: around two-thirds of all successful bids were at the technically specified minimum price of –500 €/MWh. This behavior is not only found among renewable energy plants, which are secured by EEG remuneration or market premiums and therefore generate stable revenues regardless of the exchange price. A large number of thermal power plants also bid at these minimum prices. There are many reasons for this: regulatory incentives dominate in the renewable energy sector, for example through support mechanisms such as market premiums for older EEG plants, while high startup costs or the goal of remaining in the market and serving followup hours may play a role for conventional plants.‑up costs or the goal of remaining in the market and serving follow‑up hours may play a role for conventional plants.
Another key finding is that only around two-thirds of Germany’s net electricity load is actually traded on the EPEX Day-Ahead Market. A significant proportion of electricity continues to be procured through bilateral contracts or through energy suppliers’ own generation capacities. At the same time, however, a clear trend can be observed: The share of exchange trading has risen steadily in recent years, indicating increasing market transparency and the growing importance of short-term marketing strategies.
The evaluation also shows that in the majority of the hours examined, only a small group of players determine prices. The supply curves of these players are often very steep, much steeper than would be expected based on pure marginal cost logic. However, marginal costs and final auction prices are close to each other on an annual average.
An example bid curve for April 28 at 4 a.m. is shown in Fig. This is an hour with low renewable energy generation and low load. Here you can see that many plants, including non-renewable ones, are offering the minimum price of -500 €/MWh. In addition, the total trading volume is very low in relation to the average German load of 45 GW during this hour, and the bid curve is significantly steeper than the modeled merit order, albeit with a very similar resulting price.
Annuity-based assessment of power plant profitability
In the third part of the project, the total costs of the power plants were offset against the power plant-specific revenues in order to determine the economic efficiency of operations in 2025.
Methodology
To evaluate the economic efficiency of different types of power plants, the annual revenues from electricity sales were compared with their total annuity costs. The calculation of costs included investment costs, annual operating costs, and additional costs, for example for the start-up of assets: The former were converted to annual values using annuity factors so that both capital commitment and a typical depreciation period could be represented. The ongoing annual operating costs consisted of fuel costs, CO₂ certificates, and fixed and variable operating and maintenance costs. Finally, the start-up costs for fossil fuel power plants were also included in the analysis.
The revenues of the individual power plant types were calculated on the basis of the amount of energy generated annually, which was multiplied by the technology-specific market value of the electricity price in 2025. This market value reflects the fact that different technologies produce electricity at different times and thus benefit from different price levels. For example, the high simultaneous feed-in from PV systems leads to lower prices.
Various sources were used to ensure that the economic assessment was based on sound data. The technical efficiencies of the power plants considered were based on the FfE power plant list (based on the market master data register) [9] and supplementary research on specific plants. The investment and operating costs were taken from the BDI/BCG study “Climate Pathways 2.0” [10]. For renewable energies, the full-load hours and electricity generation costs from current publications by Fraunhofer ISE [11] were also taken into account. The start-up costs of thermal power plants were based on data from the empirical study by Öberg et al. (2022) [12].
Results
The analysis shows that in 2025, only wind turbines (onshore and offshore), combined cycle gas turbines (CCGT), and run-of-river power plants were economically viable. These technologies benefited either from low operating costs (renewable energy plants), high lifespan (run-of-river power plants) or high market values at feed-in times (CCGT). Coal-fired power plants, gas turbines and Photovoltaic systems were not economically viable in 2025. The continuing rise in CO₂ costs placed such a heavy burden on coal-fired power plants that even when fully depreciated, they were unable to cover their operating costs in full.
Photovoltaic systems continued to prove highly dependent on subsidy instruments in 2025. Without EEG subsidies or market premiums, operation was not economically viable on average. The decisive factor here was the very low market value of photovoltaics, which was only around €42/MWh in the year under review, 53% less than the annual average price of €89.3/MWh in Germany. Since PV systems deliver most of their generation simultaneously and thus at times of high feed-in, the achievable market revenues could not cover the annuity investment costs. In contrast, the market value of electricity prices at times of CCGT and gas turbine feed-in is 41% and 65% higher than the average, respectively.
The analysis also shows that the economic efficiency of conventional and renewable technologies is determined by different factors. While operating costs, especially fuel and CO₂ costs, are the main determining factors for conventional power plants, the main factors for renewable plants are the level of investment costs and the underlying depreciation period. As a result, the results are very sensitive to assumptions about the efficiency of conventional power plants, investment costs, or depreciation periods. Even moderate changes in these parameters can significantly shift the economic efficiency of individual technologies and lead to different assessments.
Based on the assumptions made, no systematic excess profits could be identified. However, the high sensitivity of the results with regard to the input parameters must be taken into account. After all, 2025 was a very windless year, which meant that gas-fired power plants were more active in the market and prices tended to rise. This may also explain the relatively high market values for wind energy compared to previous years.
Conclusion
The study shows that supply curves often deviate significantly from pure marginal cost logic when it comes to pricing, mainly due to regulatory conditions (such as EEG) and different marketing strategies. Nevertheless, auction prices and marginal costs are close to each other on average.
No systematic excess profits could be proven. Furthermore, the analysis shows that fossil fuel power plants (exemption is CCGT), and PV systems (without subsidies) did not produce economically in 2025.
Literature
[1] von der Fehr, N.-H. M., & Harbord, D. (1998). Competition in electricity spot markets: Economic theory and international experience (Memorandum No. 05/1998). University of Oslo, Department of Economics. https://www.econstor.eu/bitstream/10419/90803/1/Memo-05-1998.pdf
[2] European Parliament & Council. (1997). Directive 96/92/EC concerning common rules for the internal market in electricity (OJ L 27, 30.1.1997, pp. 20–29). Publications Office of the European Union. https://eur-lex.europa.eu/legal-content/DE/TXT/PDF/?uri=CELEX:31996L0092
[3] Müsgens, F. (2004). Market power in the German wholesale electricity market (EWI Working Paper No. 04.03). Institute of Energy Economics at the University of Cologne (EWI). https://www.econstor.eu/bitstream/10419/23154/1/Ewiwp043.pdf
[4] EPEX SPOT. (2020). EPEX SPOT celebrates 20 years since the first day-ahead auction. https://www.epexspot.com/en/news/epex-spot-celebrates-20-years-first-day-ahead-auction
[5] ENTSO-E. (2026). Single day-ahead coupling (SDAC). European Network of Transmission System Operators for Electricity. https://www.entsoe.eu/network_codes/cacm/implementation/sdac/
[6] Deutscher Bundestag, Wissenschaftliche Dienste. (2022). Merit Order: Alternativen zum Preisbildungsmechanismus an der Strombörse (WD 5‑111/22). https://www.bundestag.de/resource/blob/922150/ef7b04eda9b6b5034876248539891467/WD-5-111-22-pdf-data.pdf
[7] Forschungsgesellschaft für Energiewirtschaft (FfE). (2025). Dynamische Darstellung der aktuellen Merit-Order deutscher Kraftwerke. https://ffe.de/veroeffentlichungen/dynamische-darstellung-der-aktuellen-merit-order-deutscher-kraftwerke/[8] EPEX SPOT. 2024. “Market data“, https://www.epexspot.com/en/market-data
[9] Bundesnetzagentur. (2025). Marktstammdatenregister [Online-Datenbank]. Abgerufen am 1. Juni 2025 von https://www.marktstammdatenregister.de/MaStR/Einheit/Einheiten/OeffentlicheEinheitenuebersicht
[10] Boston Consulting Group. (2021). Klimapfade 2.0: Ein Wirtschaftsprogramm für Klima und Zukunft. https://web-assets.bcg.com/58/57/2042392542079ff8c9ee2cb74278/klimapfade-study-german.pdf
[11] Fraunhofer Institute for Solar Energy Systems ISE. (2024). Stromgestehungskosten erneuerbarer Energien. https://www.ise.fraunhofer.de/content/dam/ise/de/documents/publications/studies/DE2024_ISE_Studie_Stromgestehungskosten_Erneuerbare_Energien.pdf
[12] Öberg, S., Odenberger, M., & Johnsson, F. (2022). Exploring the competitiveness of hydrogen-fueled gas turbines in future energy systems. International Journal of Hydrogen Energy, 47(1), 624-644. https://www.sciencedirect.com/science/article/pii/S0360319921039768
Further Information
