Working Paper: CEPR ID: DP9015
Authors: Federico Ciliberto; Jonathan W. Williams
Abstract: We show that multimarket contact facilitates tacit collusion in the US airline industry using two complementary approaches. First, we show that the more extensive is the overlap in the markets that the two firms serve, i) the more firms internalize the effect of their pricing decisions on the profit of their competitors by reducing the discrepancy in their prices, and ii) the greater the rigidity of prices over time.Next, we develop a flexible model of oligopolistic behavior, where conduct parameters are modeled as functions of multimarket contact. We find i) carriers with little multimarket contact do not cooperate in setting fares, while carriers serving many markets simultaneously sustain almost perfect coordination; ii) cross-price elasticities play a crucial role in determining the impact of multimarket contact on collusive behavior and equilibrium fares; iii) marginal changes in multimarket contact matter only at low or moderate levels of contact; iv) assuming that firms behave as Bertrand-Nash competitors leads to biased estimates of marginal costs.
Keywords: airline industry; airport facilities; collusion; differentiated products; multimarket contact; price rigidity
JEL Codes: L13
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Bertrand-Nash assumption (C72) | biased estimates of marginal costs (C51) |
| increased multimarket contact (L19) | higher equilibrium fares (D50) |
| increased multimarket contact (L19) | reduced price discrepancies (L42) |
| increased multimarket contact (L19) | increased price rigidity (E31) |
| low or moderate levels of multimarket contact (L19) | significant collusive behavior (L12) |
| high levels of multimarket contact (L19) | diminishing return effect on collusion (D43) |
| cross-price elasticities (D11) | extent of collusion (L12) |