The estimated airline fixed effects roughly correlate with average on-time performance for the
airlines as reported by Bowen and Headley (2001), and the estimated airport fixed effects are also as
expected. Dummy variables were also included for each month’s data – delays were much greater
in July than in January or April. This may indicate that delays are getting worse over time, but with
only one year’s data some other seasonal variation cannot be ruled out. Day of the week controls
indicated that flights were more likely to be on-time on light travel days (Tuesday and Saturday),
while Friday flights had significantly longer delays. These results are not reported in Tables II and
III, but are available from the author on request.