Fare Determination in Airline Hub-and-Spoke Networks

This article provides the first evidence linking airfares to the structure of airline hub-and-spoke networks. The hypothesis tested is that any force that increases traffic volume on the spokes of a network will reduce fares in the markets it serves. This effect arises because of economies of density on the spokes. For example, since a large network (as measured by the number of city pairs that it connects) is expected to have low costs per passenger as a result of high traffic densities, fares in the individual markets served should be low, other things equal. Similarly, holding size fixed, a network that connects large cities should have higher traffic densities on its spokes (and thus lower fares in individual markets) than one serving small cities. Our empirical analysis supports these predictions. We find that network characteristics are important determinants of fares in 4-segment city-pair markets (these are markets requiring a connection at the hub). Furthermore, our empirical model predicts that the TWA-Ozark and Northwest-Republic mergers should have reduced fares in the 4-segment markets served by the hubs at St. Louis and Minneapolis.