A Structural Estimation Approach to an Asymmetric Auction Model for the Japanese Retail Power Market

In this paper, we develop a structural auction model and quantify the effects of policy measures aiming to enhance competition in the Japanese retail power market. We employ a theoretical model that incorporates asymmetries between the incumbent and entrants in terms of both the cost and information structures, where the costs of the former are assumed common knowledge, and empirically estimate the structural parameters characterizing their cost distributions using public power procurement data. We then conduct counterfactual simulations to quantify two competition-promoting policy measures: a bid preference program for entrants, and an increase in the number of potential bidders. We take a parametric approach to estimate the structural model successfully in contrast to a nonparametric approach that previous studies took. Our simulation results show that these procompetitive measures would barely increase participation by potential entrants but would elicit more aggressive incumbent bidding behavior. Further, a modest bid-preferential rate would improve welfare and reduce the probability of realizing inefficient allocations associated with a costly winning bidder.