Working Paper: CEPR ID: DP5000
Authors: Marc Ivaldi; Gerard McCullough
Abstract: Since the publication by Williamson (1968) of his seminal paper on antitrust there has been a growing recognition by regulators of the need to assess trade-offs between merger-related efficiency gains and merger-induced increases in market power. This paper addresses that need by presenting a structural econometric model of recent mergers in the US rail industry. The paper extends the structural methodology by evaluating actual (as opposed to simulated) merger effects and by incorporating parametric estimates of merger efficiencies. Our empirical finding is that consumer surplus in US rail freight markets increased by about 30% between 1986 and 2001 despite dramatic industry consolidation, suggesting that to date the Williamson trade-off has favoured rail customers. We find that behaviour in these markets is consistent with the Kreps-Scheinkman (1983) model of a two-stage game where capacities are chosen first and then prices are set to give the Cournot outcome.
Keywords: differentiated product markets; logit models; merger analysis; railroads
JEL Codes: L11; L13; L41; L92
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| mergers (G34) | consumer surplus (D46) |
| increased market power (D43) | consumer surplus (D46) |
| efficiencies from mergers (G34) | consumer surplus (D46) |
| consolidation (G34) | increased market power (D43) |
| mergers (G34) | increased market power (D43) |