When the Department of Transportation announced in December 2009 that they were setting a hard limit on the amount of time flights could be delayed (the so-called "3-hour rule") before the operating carrier would be subject to significant fines there was quite the debate as to its potential impact. Many lauded the government for helping protect the passengers while others (self included) raised the point that more cancelled flights is worse for passengers than being stuck on the airplane but actually getting to the destination. Nearly a year into the new program the data available suggest that, so far, customers are losing far worse than expected from this policy.
With nine months of data available it is pretty clear that the number of 3-hour or longer tarmac delays is down. The DoT is making sure to trumpet that news and many media outlets are picking it up and running with it. They even have a pretty graph showing the impact of the rule:
Seems great, right? It is not.
Yes, there are fewer folks spending entirely too long in the plane on the way to wherever they are going, but that does not mean they are actually getting where they are going. It turns out that the number of flight cancelations has increased dramatically during this period as well, both as an absolute number and as a percentage of total scheduled flights. Actually, the latter is even worse as the total number of scheduled flights has decreased since the rule went into effect.
Here’s another look at the data, compiled based on the statistics published by the same federal agency that the DoT is using for the above numbers (click the images for the source data):
Yes, the variation in the size of the lines is much more dramatic in the bottom graph, but Marshall McLuhan would not be amused. It may look like the decrease in 3-hour delayed flights is more dramatic than the rise in cancelations, and in terms of percentages it is. But most significant is that the number of passengers displaced is MUCH higher because of the increased cancelations.
Oh, and the total number of flights operating each month was lower in 2010 than in 2009 so the fact that the cancelation numbers are higher in 6/10 samples and essentially the same in 3/10 is not a particularly comforting thought at all. That’s 90% of the sample since the rule went into effect where the percentage of flights canceled was higher than the previous year.
Load factors are running at or near all-time highs. This means that once a flight is canceled it is even harder to accommodate the affected passengers as there simply are no empty seats to place them in. So increasing the number of canceled flights certainly doesn’t seem to be the way to solve the problem of inconvenienced customers.
It should be pointed out that correlation certainly does not equal causation. Maybe there is another reason that there were so many flights canceled. But there is weather every year and the impact of the Eyjafjallajökull eruption was mostly confined to April 2010. Ironically that was the one month of the past 10 that had a lower cancelation rate.
Penalizing lots of customers for the sake of a select few is a bad way to operate. The Department of Transportation needs to take a hard look at these numbers and reconsider this policy. Delays and cancelations suck for everyone. The government should not be in the position of inducing them.
A bit more coverage of this phenomena can be found here, too