Texto para Discussão, Instituto de Pesquisa Econômica Aplicada (IPEA) 1703
This study investigates the determinants of the great difference between winning bids and reserve prices in the transmission electricity sector in Brazil between 1999- 2010, through an econometric approach based on Heckman (1979) that considers the heterogeneous among the winner's and loser's bids, and the endogeneity of that selection. Given the whole data of winner's and loser's bids, the heterogeneity of the groups is modeled and endogeneity bias classification is corrected. The object of this research is justified since studies that focus just on winning bids or ignore the endogeneity of the winning group selection could lead to misleading conclusions. Results can be summarize as follows: i) bids made by state public companies have 50% probability to win the auction; ii) the majority of winning bids come from state public companies in partnership to private national groups; iii) winning probability is related to previous investments made in the bid area due to economies of scale; iv) the great difference between reserve price and winning bids are partly explained by the improvements in Brazil country risk and the profitability of the project, being more important for winners than losers probably due to the better set of information and effectiveness of the winning group; v) the number of competitors decreases the bids but with non linear effects; and vi) outlier bids made by public companies represent 57% of total outliers with an average bid of 40% less than the reserve price, indicating the winner's curse characteristic.