Chen and Rey (2019, Rand Journal of Economics) argue that multiproduct firms in a competitive market adopt a pricing strategy, so-called competitive cross-subsidization: the loss from the below-cost pricing of weak products is entirely offset by the profit from the above-cost pricing of strong products. This paper studies whether or not competitive cross-subsidization is robust when one firm discontinues its weak product lines which generate losses, instead of cross-subsidizing it. We show that there exists an equilibrium in which one firm discontinues its weak product and the opponent uses the monopoly profit from that market to cross-subsidize the other product. Moreover, this equilibrium Pareto dominates the competitive cross-subsidization equilibrium with no discontinuation. Cross-subsidization is robust but it no longer arises in the competitive market: it is practiced by the firm that monopolizes the market from which its opponent withdraws.
This paper investigates the allocative efficiency of spectrum auctions, particularly focusing on Auctions 110, which adopt the ascending clock auction format. Using a revealed preference approach, we estimate the private valuations of bidders across 406 markets and explore both standalone values and geographic complementarities. A key objective is to assess whether the FCC’s auction design, particularly the block cap of four per bidder, effectively promotes participation from smaller bidders. We estimate the complementarities between markets and simulate optimal allocations to measure potential efficiency losses. Our counterfactual analysis reveals a welfare loss of approximately $3.6 billion under the current allocation, highlighting significant room for improvement. Additionally, we explore geographic substitution effects and bidder strategies, contributing new insights into how the auction design affects participation, especially for smaller entities.