JetBlue's Test of Pilot Fatigue Rule Upsets Regulatory Apple Cart
There's a way for firms to do regulatory research more effectively

23 Oct 2006 in

Wall Street Journal reporters Andy Pasztor and Susan Carey describe (subscription may be required) detailed human factors research JetBlue performed with FAA district office approval to test the longstanding assumption that, under certain conditions, pilots can safely fly longer than Federal Aviation Administration rules permit. The result: an unflattering portrayal on Page One of the Journal, a punitive regulatory response from FAA headquarters, and an FAA threat to further punish JetBlue by indefinitely delaying any review of the scientific merits of its hours of service limits for pilots.

In retrospect, adherence to the procedures and standards of the Paperwork Reduction Act and Information Quality Act might have been a better way for JetBlue to proceed -- even though the PRA doesn't apply to regulated entities.

JetBlue Airways sought and obtained approval from the FAA's New York district office to conduct a test of the assumption that eight hours (the current regulatory limit) is the maximum time pilots can safely fly.They worked with a human factors consultant to collect detailed data on pilot alertness for "as many as 10 or 11 hours a day."

The results of the test haven't yet been made public -- they are expected to be published by the end of the year -- and JetBlue executives say even they don't know the findings. But the experiment has landed JetBlue in hot water while fueling a fierce debate within the airline industry about how long pilots should be allowed to stay at the controls.
Headquarters officials at any federal regulatory agency don't like it when their district or regional counterparts don't consult with them before making decisions that could backfire in Washington. It's officials in Washington who have to respond to the complaints, in this case from union and consumer activists who have their own reasons to oppose longer hours of service rules. From their perspective, district and regional officials perceive themselves as much more familiar with how to make regulations work in the real world. They reject the notion that they are mere automatons and functionaries of Washington without any capacity to exercise discretion sensibly. They chafe under what they perceive as heavy-handed Washington control and second-guessing.

The subhead of Pasztor and Carey's Page One article emphasizes the fact that JetBlue did not tell passengers is was conducting the test. This seems to imply that JetBlue placed passengers at risk, but there's no evidence supporting that inference in their article. Under the experiment JetBlue placed a third pilot onboard -- precisely what FAA rules require for flights that exceed regulatory limits for the normal two-pilot crew.A more plausible explanation is that JetBlue thought that obtaining the agreement of FAA officials whom it knew best was sufficient. Raising the issue to FAA headquarters in Washington could well have killed the idea, as regulators can be highly averse to change they they do not initiate.

But these are normal features of the federal regulatory system. Like markets, which advocates of regulation often criticize as messy and riddled with imperfections from the standard models offered by economists, regulatory systems are messy and riddled with imperfections from the standard models offered by regulators. If perfect competition is the unrealized ideal of economic theory, its analogue in regulatory practice is the errorless effectiveness of central planning. When evaluating the merits of a regulatory approach to remedy an imperfect market, it's essential to take into account the analogous imperfections that arise in implementing regulations. Too often, what's compared is a parody of the market (in which consumers are fools and businesses are venal) within idealized and imaginary regulatory regime (in which central planning works like a frictionless machine).

The existing FAA rule on pilots' hours of service reflects a sensible rule of thumb when it was established, but it is a rule of thumb nonetheless. There is little scientific evidence showing that eight hours is the correct upper bound on pilot hours, or for that matter, that the correct value is constant irrespective of the type of commercial flight involved.Pasztor and Carey write:
JetBlue and some of its pilots argue that longer flight shifts could actually improve the quality of life for pilots and perhaps enhance their alertness. Flying from New York to California and back in the same workday, they say, would allow crews to sleep in their own beds, enjoy better rest and avoid hotel stays at odd hours that tend to disrupt natural sleep rhythms.
Research could help reveal that current FAA rules are too strict. But it also could show that they are too lenient. An hour battling a severe thunderstorm or unexpected wind sheer might be enough for one work day.

What JetBlue might have done instead is follow an established public process for obtaining data, such as the Paperwork Reduction Act. The PRA applies to federal agencies and private entities who collect information on their behalf, but not to private parties such as JetBlue. So if FAA had wanted to conduct an experiment like JetBlue's, it would have been required by law to comply with the PRA.

The fact that private parties are not covered by the PRA shouldn't be the end of the matter. Rather, regulated entities should seriously consider following PRA procedures anyway, and adhering to the PRA's recently enacted information quality guidelines as well. If JetBlue had done this, it would have been assured that the final product met the highest information quality standards and it could have provided a sound scientific basis for for revising the pilot hours of service rule, which it said to have been in place for 40 years. The airline also could have avoided an unnecessary and counterproductive public relations controversy.

As matters stand, JetBlue stands accused by FAA of trying to skirt the hours of service regulations, according to reporters Pasztor and Carey. If JetBlue believes it's been treated unfairly, which seems completely reasonable, then starting over using the PRA process would be a good way to correct the record.