Customs Wait Against Hall Volume, Hour by Hour

Summary

In 20 of 20 hours of the day, the passport-control halls that handled more non-US passengers also had longer median waits for them; the median rank correlation across hours is +0.6. The busiest third of halls waited longer than the quietest third in 20 of those hours.

Customs and Border Protection reports its passport-control waits for each terminal and hour. This report holds the hour of the day fixed and compares halls: in each hour, whether the halls that process more arriving passengers have longer waits.That removes the time-of-day pattern that every hall shares and leaves the difference between big halls and small ones. The non-US passport queue and the US passport queue are kept apart, because they are separate lines with separate staffing and each is compared with its own passengers. Across 20 hours of the day with at least 9 halls, the median rank correlation for the non-US queue is +0.6.

Each hall’s figures are its median wait and its average passengers per reported hour for that hour of the day, all dates pooled. Thirds are the busiest and quietest third of the halls compared in that hour, by passengers; the wait shown is the median of their median waits. Rank correlation uses average ranks for ties.

The same measure for the queue US passport holders join. Each hall’s figures are its median wait and its average passengers per reported hour for that hour of the day, all dates pooled. Thirds are the busiest and quietest third of the halls compared in that hour, by passengers; the wait shown is the median of their median waits. Rank correlation uses average ranks for ties.

Staffing follows volume. CBP opens booths in response to arriving passengers, so a big hall with a short wait may simply be well staffed for its load, and a link between volume and wait is not evidence that volume causes the wait. The unit is a CBP terminal, which is a hall inside an airport and not the airport itself, so an airport with two terminals contributes two halls. Halls differ in who arrives: preclearance flights, the mix of countries, and the share of Global Entry members all move the wait without changing the passenger count. The hour is the hour passengers were processed, not the hour the aircraft landed. A hall-hour is included when at least 300 dates of data stand behind it, and an hour of the day is compared only when at least 9 halls qualify; 610 hall-hours are compared for the non-US queue and 610 for the US queue.