The desirability of airport congestion pricing largely depends on whether dominant airlines otherwise fail to internalize their self-imposed congestion delays. Brueckner (2002) and Mayer and Sinai (2003) find (weak) statistically significant evidence of internalization. We replicate and extend these models by refining their measures of delay and controlling for fixed and random airport effects. For twenty-seven large US airports, we estimate every flight’s congestion delay attributable to its operating time. These time-dependent queuing delays result from traffic rates temporarily exceeding airport capacity, and are precisely the delays susceptible to peak-load congestion pricing. As modified, the models reject the internalization hypothesis.

(JEL H2, L5, L9, D6)

Daniel: Department of Economics, Purnell Hall, University of Delaware, Newark, DE 19716,
(e-mail: Joseph.Daniel@verizon.net);

Harback: Department of Economics, Purnell Hall, University of Delaware, Newark, DE 19716.