Strategic bidding in vertically integrated power markets with an application to the Italian electricity auctions ☆ Bruno Bosco a , Lucia Parisio a , Matteo Pelagatti b, ⁎ a Department of Legal and Economic Systems, Università degli Studi di Milano-Bicocca, Piazza Ateneo Nuovo 1, 20126 Milano, Italy b Department of Statistics, Università degli Studi di Milano-Bicocca, Via Bicocca degli Arcimboldi 8, 20126 Milano, Italy abstract article info Article history: Received 29 June 2011 Received in revised form 12 November 2011 Accepted 12 November 2011 Available online 19 November 2011 JEL classification: D44 L13 L41 L94 Keywords: Electricity markets Optimal bid functions Market power Vertical integration In this paper we apply a model of optimal bidding behavior to the Italian wholesale electricity market under three hypotheses: i) costs of generation are private knowledge, ii) firms can be vertically integrated, and iii) firms can sell part of their production in advance with bilateral contracts. We first use optimal bid functions and market data to retrieve time-varying marginal cost functions, price–cost margins and Lerner Indexes of market power for a sample of Italian companies. Then, we use estimated costs and actual equilibrium prices to evaluate the elasticity of these series to fuel price variations and estimate a possible differential impact of the dynamics of input expenditures (fuel price above all) on generation costs and final electricity prices. Our es- timates suggest that the elasticities of costs and equilibrium prices with respect to oil price are virtually the same and, therefore, that the auction mechanism per se does not limit the extent to which cost increases are transferred to prices. © 2011 Elsevier B.V. All rights reserved. 1. Introduction A major concern of regulators and governments when they evaluate the performances of the recently restructured electricity supply industries is the exercise of market power by firms competing in the wholesale market. These markets work as single price competitive auctions and operate on hourly (or semi-hourly) fre- quency. The market operator collects demand/supply bids and orders them in descending/ascending order to determine an equilibrium price and a total quantity exchanged. The creation of short term electricity markets was inspired by the idea that introducing compe- tition at the wholesale level would result in lower electricity prices for final consumers. Unfortunately, starting from the first experiences of privatization and deregulation, it was clear that such markets may be prone to collusion or other forms of anticompetitive behavior. Electricity producers can increase substantially their profits if they are able to influence market price through their bidding behavior. It is therefore very important for regulators to understand what forces drive the bidding behavior of generators and how the market mecha- nism translates bidding strategies into market clearing prices. Economic theory alone gives little guidance to researchers when they have to evaluate the performance of multi-unit auctions. For that reason the empirical analysis becomes very important to evaluate the performance of these mechanisms. Electricity markets are in this respect a privileged field since all the relevant market data are made available from the market operators. Given the data and a theoretical model that describes the optimal bidding behavior of producers, the researcher is able to measure unilateral market power of the firm and the way in which it influences the final equilib- rium price. We use data published by the Italian market operator to evaluate the performance of the Italian electricity market. The issue is particularly interesting since this market has registered the highest average price level among the other European markets since the very beginning of its operation (April 2004). The reasons explaining this price level have been strongly debated. Some stress the role of the technology mix of the generation park; others claim that the Italian market is too much isolated from the other continental markets. Another possible reason may be that the Italian Electricity generation industry is concentrated and prone to the exercise of Energy Economics 34 (2012) 2046–2057 ☆ This research was financed by the Ministerial Grant PRIN 20074PFL7C. We thank Pia Saraceno and researchers at Ref (www.ref-online.it) for comments on an early version of this paper and for sharing with us their dataset on the Italian produc- tion units. We also thank two anonymous referees and the editor for useful comments and suggestions. ⁎ Corresponding author. Tel.: + 39 02 64485834; fax: + 39 02 64485878. E-mail addresses: bruno.bosco@unimib.it (B. Bosco), lucia.parisio@unimib.it (L. Parisio), matteo.pelagatti@unimib.it (M. Pelagatti). 0140-9883/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.eneco.2011.11.005 Contents lists available at SciVerse ScienceDirect Energy Economics journal homepage: www.elsevier.com/locate/eneco