Competition in Germany's Minute Reserve Power Market: An Econometric Analysis.

AuthorHaucap, Justus
  1. INTRODUCTION

    The nature of electricity requires that the balance between production and consumption is maintained in the electricity grid at each single point in time. Typically, this task is imposed as a responsibility on the transmission system operator (TSO) to ensure system stability by procuring so-called (electricity) reserve power. (1) For this reason, generation units are obliged to reserve some fraction of their capacity which can be used then by TSOs to restore frequency and load in the electricity grid. (2) As imbalances between supply and demand can be caused, e.g., by incorrect demand predictions, stochastic fluctuations of electricity generation from renewable energy sources, especially wind and solar energy, and/or breakdowns of generation units, the provision of ancillary services such as frequency control is a crucial element for ensuring system stability.

    In Germany, as well as in all other 33 member states of the European Network of Transmission System Operators for Electricity (ENTSO-E), three different qualities of reserve power are distinguished: 1.) primary control power (PCP), 2.) secondary control power (SCP), and 3.) minute reserve power (MRP) (tertiary control power). Two types of SCP and MRP have to be distinguished: incremental (positive) reserve power and decremental (negative) reserve power. While the former is used when demand for electricity exceeds supply, the latter is needed when more electricity is generated than consumed. In Germany, the prices charged for each of the reserve power products are two-part tariffs. The so-called capacity price is paid for the stand-by provision of reserve power, while the operational price is paid in case of delivery. The capacity price reflects the suppliers' opportunity cost of committing not to use the reserved capacity to supply it to electricity wholesale (spot) markets. The focus of our paper is on MRP capacity prices for both incremental and decremental MRP. It should be noted that MRP is only used if both PCP and SCP are insufficient to restore the desired grid frequency of 50 Hz.

    The German market for reserve power has been subject to two important regulatory changes in recent years. The first regulatory change has focused on the synchronization and standardization of the four control areas in Germany. More precisely, a common web-based tendering platform (www.regelleistung.net) was launched as a result of the Energy Industry Act of 7 July 2006 (see Bundesnetzagentur, 2006, 2007a, and 2007b) which synchronized and standardized the four German control areas in time and place on 1 December 2006, for MRP, and one year later, on 1 December 2007, for PCP and SCP (3) According to the German Federal Network Agency (Bundesnetzagentur), the aim of these reforms has been to foster competition and to increase efficiency by eliminating strategic behavior and by facilitating market entry (see Bundesnetzagentur, 2006).

    The second regulatory change obliged the four TSOs (4) in Germany to gradually interconnect and to coordinate their operations (see Bundesnetzagentur, 2010). This reform solely tackled the inefficient use of reserve capacities without directly affecting the competitive process. For illustration purposes, consider the following example: Suppose that one control area exhibits excess electricity supply, while another control area has excess demand for electricity. Further, assume that the resulting frequency deviations are independent and equal in volume. Based on the 'old' regulatory framework, one TSO would have to procure decremental MRP, while the other needed to purchase incremental MRP to eliminate the resulting frequency deviations. With interconnection and cooperation, there is no need to procure any kind of MRP, since the excess production of one control area completely offsets the excess consumption of the other. Hence, such compensating deviations can be managed internally through joint TSO balancing. One implication is that generation units are now automatically able to offer SCP and MRP in all four control areas. That is, there is one merit order including all four control areas.

    The academic literature on competition in electricity reserve markets is rather scarce. Most papers studying the efficiency of the MRP market in Germany analyze the market design from a theoretical perspective. While some papers focus on the possibility of strategic and collusive behavior given the procurement auction design (see, e.g., Muller and Rammerstorfer, 2008), others either study optimal decision rules for network operators and reserve capacity suppliers (see Swider, 2006; and Swider and Weber, 2007) or analyze productive efficiencies (see Swider and Ellersdorfer, 2005).

    Very few papers analyze the effects of the structural reforms on the MRP market empirically. Growitsch et al. (2007) examine the impact of synchronization and interconnection on MRP prices. They use time series methods to test for a structural break when the common web-based tendering platform for MRP was launched in December 2006. In addition to MRP price time series, they use data on electricity spot market prices, and find that the launch of the common web-based tendering platform had no significant effect on MRP prices. Growitsch and Weber (2008) analyze the spread between incremental MRP prices and electricity spot market prices. They apply a mean reversion model to test whether the degree of market integration between the MRP market and the spot market has increased due to the new market design. Growitsch and Weber (2008) show that the MRP market has become more efficient, although the price spread has increased over time. Finally, Riedel and Weigt (2007) provide a correlation analysis to study the interdependence between the four German regional MRP markets and their relationship to the electricity spot market.

    We extend these papers into four directions. First, we have created a unique dataset for the period from 1 January 2006 to 30 September 2010 to apply panel data models accounting for unobserved heterogeneity between the four German control areas. Second, we estimate causal effects using instrumental variable techniques. We control for potential endogeneity of the wholesale electricity (day-ahead) spot market price using German weather data as instruments. Third, we also consider the synchronization of the PCP markets and SCP markets as well as the interconnection of the four TSOs, and analyze whether this had an impact on MRP prices. However, our main focus is on the launch of the common web-based tendering platform for MRP, because it is natural to expect a direct effect on MRP prices. We perform Chow tests to check whether or not each of the reforms led to a significant change of MRP prices, where we take decreases in MRP prices as being indicative of increased competition in the German MRP market. (5) Finally, we quantify the reforms' joint success by comparing the actual MRP prices with those prices that would have been realized without the reforms.

    We find that market synchronization and standardization significantly decreased both incremental MRP prices and decremental MRP prices, while the second regulatory change apparently had no further impact on prices. More precisely, TSO interconnection and cooperation partially led to an increase of MRP prices or did not significantly affect MRP prices at all. Hence, the effect is ambiguous. Nevertheless, the reforms' joint effect on MRP prices is negative and led to considerable savings in each of the four regional markets.

    The remainder of the paper is structured as follows. Section 2 provides a brief overview of the regulatory changes. In Section 3, we present the data and describe the econometric analysis. Section 4 concludes the paper.

  2. REGULATORY CHANGES IN THE GERMAN ELECTRICITY RESERVE POWER MARKETS

    Initially, each of the four TSOs in Germany procured reserve power in its own control area at various times based on bilateral contracts with affiliated generation plants. In 2001 and 2002, respectively, the German Federal Cartel Office required the TSOs to replace these bilateral contracts by procurement auctions. However, the four control areas remained distinct and time-separated, facing different prequalification requirements.

    The synchronization and standardization of the electricity reserve power markets in Germany started on 1 December 2006, when a common web-based tendering platform was launched for MRP. The timing of auctions, the prequalification procedure, (6) and the rates for creating merit orders for actual delivery were specified and standardized. Moreover, the new market design prescribed that the procurement auction had to close before the electricity (day-ahead) spot market at the European Energy Exchange (EEX) opened. The procurement auction after 1 December 2006 can be basically characterized as a i) repeated (daily), ii) day-ahead, iii) multi-unit (incremental and decremental), iv) one-sided (only reserve capacity supplier submit offers), v) multi-part (capacity price and operating price), and vi) pay-as-bid auction (see, e.g., Muller and Rammerstorfer, 2008). (7) The market synchronization was completed on 1 December 2007, when joint web-based tendering platforms have been also launched for both PCP and SCP. In contrast to MRP, the procurement auction was initially held monthly, and the prequalification procedures were more restrictive due to the inherently higher technical requirements of PCP and SCP. (8)

    The four TSOs had already started to interconnect and to coordinate their operations before the German Federal Network Agency made it mandatory as of 16 March 2010. Initially, two alternative concepts were discussed to reduce the inefficient use of reserve power capacity in the SCP market. (9) On the one hand, one central and overriding TSO was proposed to control the frequency in all four control areas. This alternative was favored by Amprion which is the TSO...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT