The economics and geography of regional airline services in six countries

• We find interesting differences between trunk markets and regional markets.
• We find strikingly large market enhancement effects on regional air routes.
• We find significant roles for cost, competition and demand-side factors.
• Pricing model reveals puzzlingly large differences across countries in airfares.

Do the determinants of service and pricing on “regional” routes – linking towns and smaller cities to main trunk routes and/or to each other – differ from the established results from the literature? We study all flights (about 3000) on all regional routes (about 250) with scheduled airline service from one of about 130 regional towns or cities, in regional airline markets in six countries: Australia, Canada, New Zealand, Norway, Sweden, and a sample of three U.S. states which closely resemble the other regions studied. For each flight we have observations on up to five prices offered at different times before flight date. We also have equipment type and social-economic data. Overall, our results give qualified support to the standard gravity model of the extent of service between city pairs, though with two interesting differences: operators on regional routes have greater flexibility in the size of aircraft they can deploy, which results in a finer-grained variability of service offerings and, the presence of competition on regional routes has a large effect on the total supply of seats. We are able to successfully estimate a well-specified airfare model, which shows strong effects of competition on prices, quite substantial intertemporal price discrimination, and interesting differences between regional and main trunk route pricing.