Portraits of imperfectly competitive markets.

Abstract

The dissertation sketches three applications in the field of industrial organization. Chapter 1 examines the significance of multiple airports in a neighborhood in affecting airfares. Chapter 2 compares the profitability of airlines under expiring and non-expiring frequent-flyer programs. Chapter 3 explores the determinants of percentage rent for various tenant categories in shopping center leases. Previous studies on air traffic demand have focused on differentiation in carrier and itinerary. But the choice of airport has rarely been considered. Chapter 1 employs a definition of neighborhood not restricted by single city boundaries to group the domestic travel market into 84 single and 22 multiple airports neighborhoods. Results from estimating the demand schedule and first-order condition indicate that fares are lower across all airlines and airports in multiple airport neighborhoods. At multiple airport hubs, however, this fare discount is empirically mitigated by the hub fare premium effect. In the late 1980s, many frequent-flyer programs shifted from awarding non-expiring to expiring coupons. Chapter 2 extends the Caminal and Matutes (1990) endogenous switching cost model to a three-period context to analyze airlines' profitability under both regimes. Solution in the subgame perfect equilibrium favors non-expiring awards, with the intuition that airlines can lock consumers in for longer. While this does not directly explain the current expiring coupons, it is consistent with the recent move towards expiration from the date of miles earned rather than award issuance. That is, airlines that better induce loyalty enjoy higher profits. Moral hazard is an often cited incentive for percentage sharing in mall leases. Ability to generate externality is the main argument for differing percentage rents among tenants. Nevertheless, this argument can only address the low percentages paid by anchors but not the high percentages paid by traffic-generating businesses like movie theaters. Chapter 3 proposes profit margin as an additional positive influence on percentage rates. While the lack of good data results in an insignificant coefficient for profit margin, the negative relationship between demand variance and percentage provision is consistent with predictions from a double-sided moral hazard model.