# 17-079/VIII (2017-09-05)

Author(s)
Daniel Rijken, VU Amsterdam, The Netherlands; Vincent (V.A.C.) van den Berg, VU Amsterdam, The Netherlands; Tinbergen Institute, The Netherlands
Keywords:
Mergers, quality, airlines, schedule delay, frequency
JEL codes:
D22, L13, L93, R40

We investigate the impacts of five airline mergers on one quality dimension, namely route frequency. We use monthly data on routes between the largest 64 US cities from 1999 to 2016. On average, the mergers decrease the frequency, but there are large differences between the five mergers. We hypothesize that these differences resulted from differences in the market and network structures. Our estimations indicate that, if a destination has more connecting flights of the merging airlines, the merger is less detrimental to the frequency, possibly because the merger removes serial marginalization in the quality and price setting. For the market structure effect, we use two distinct set-ups. In the first set-up, the effects of mergers depend on a lagged variable measuring the current market structure. On routes with stronger competition, mergers decrease the frequency more, possibly due to a larger effect on the market structure. When the merging airlines control all the flights, mergers have almost no impact on the frequency. The second set-up uses the market structure before the merger. When one of the merging partners controlled all the flights between two airports, the merger does not directly affect the market structure and seems to have little to no impact on the frequency. Surprisingly, if both partners were flying between two airports before the merger, this merger does not seem to be more harmful to the frequency than when only one partner was operating on the route.