71st International Atlantic Economic Conference

March 16 - 19, 2011 | Athens, Greece

Price-cap and Non-concave Profit Functions in Electricity Markets using Italian Data

Friday, 18 March 2011: 09:00
Bruno Bosco, Ph.D , of Legal and Economic Systems, University of Milan-Bicocca, Milan, Italy
Matteo Pelagatti, Ph.D , Department of Statistics, University of Milan-Bicocca, Milan, Italy
Lucia Parisio, Ph.D , of Legal and Economic Systems, University of Milan-Bicocca, Milan, Italy
In recent years electricity markets have been reorganized as competitive (i.e. uniform price) auction markets where several firms compete for the supply of bulk energy. The behavior  of profit maximizing firms in these markets is described in the literature by bidding functions that continuously map  a correspondence from the state domain of generators’ costs and market demand volumes into the asked prices codomain. Recent empirical investigations show, however, that in many cases the bids posted by large producers cannot be seen as the result of a profit maximizing strategy, particularly when their residual demand is  high and competitors are on the verge of their production capacity. Hence, an apparent paradox emerges: why do firms refrain from exploiting fully the market power they posses? In this paper we investigate the terms of this paradox and show that under reasonable demand conditions the profit function of  an “indispensable” producer may become non concave at the price where competitors offer their entire capacity. This makes: i)  his residual demand constant; ii) his theoretically optimal bid equal to infinity; iii) his profit function continually increasing in the price space.  Actual behavior largely coincide with this prevision and prices are well above marginal costs. However, asked prices of indispensable bidders are not only below infinity but also below the regulation caps, and this means that firms obtain large rents but do not exploit fully their market power. We explore the extent of this divergence of actual from optimal bids at theoretical level and use a large sample of Italian data to test the sensitivity of this behavior to demand, capacity and costs conditions.