Cyclical dynamics of airline industry earnings

Abstract

Aggregate airline industry earnings have exhibited large-amplitude cyclical behavior since deregulation in 1978. To explore the causes of these cycles we develop a behavioral dynamic model of the airline industry with endogenous capacity expansion, demand, pricing, and other feedbacks; and model several strategies industry actors have employed in efforts to mitigate the cycle. We estimate model parameters by maximum likelihood methods during both partial model tests and full model estimation using Markov chain Monte Carlo methods to establish confidence intervals. Contrary to prior work we find that the delay in aircraft acquisition (the supply line of capacity on order) is not a very influential determinant of the profit cycle. Instead we find that aggressive use of yield management—varying prices to ensure high load factors (capacity utilization)—may have the unintended effect of increasing earnings variance by increasing the sensitivity of profit to changes in demand.