A Modest Proposal on Tarmac Delays

Government Regulation

It looks like my wish for a more thorough review of tarmac delays has been answered. Instead of the off-the-cuff claim of victory that the Department of Transportation (DOT) has been espousing, Darryl Jenkins and Josh Marks went back after their preliminary study and have now come out with a deeper analysis of the impacts of the tarmac delay rule over the entire summer. What’s more, they aren’t just suggesting to eliminate the rule. They’ve come up with ways to alter the rule to help blunt the impact on cancellations. Hopefully the DOT doesn’t just dismiss them as it did last time and instead actually listens to this truly modest proposal (which unlike the original, is not satire.)

Want to guess what they found? Here’s the summary:

Tarmac Delay Summer Summary

But the summary is only the tip of the iceberg on this study. As we all know and expected, long ground delays went down and cancellations went up. Year-over-year, there were 534 fewer long ground delays. Now, it’s tempting to say those were all reduced because of the rule, but that’s not going to be true. The weather was much better this year, and the airlines were also more conscious about the issue. So I imagine that only some of these reductions are directly attributable to the rule, but that’s fine. The rule did reduce long ground delays as advertised.

On the other side, there were more than 5,000 additional cancellations this year versus last. These also cannot all be attributed to the rule. But with weather being better, you would think that cancellations would be on the lower side. I know that Delta admitted to running a fairly poor operation for part of the summer, so that could account for some of the cancellations, but these numbers were consistently higher across all the airlines. This isn’t an airline operation-specific problem. The rule has to be responsible for some of the increase.

Oh, and about that weather . . . the study shows that conditions were far better this summer than last. In fact, it was one of the best weather summers in years. The study looked at the type of weather, the time of day, and the length of the weather event and the result was the same all around. Weather was better, around 30 percent better in fact.

From this information, I think it’s safe to say that the rule has caused significantly more people to be inconvenienced this year versus last. Yes, this year the inconvenience was due to more cancellations while last year it was due to long ground delays, but the total number of people hurt was higher. So should we scrap the rule? Nah, that’s what I like about this study. It actually suggests some fairly minor tweaks to the rule to make it more effective and reduce the impact on cancellations. Here’s the plan.

  1. Instead of having a nebulous fine structure that simply says it could be up to $27,500 per passenger, change it so that there are very clear fines published for everyone to see. In addition, have a graduated fine structure so that it starts out relatively low if it’s between 3 and 3.5 hours and then climbs higher the longer the delay.
  2. Clarify the current language in the rule and fix inconsistencies so that airlines really know how this is supposed to work. There all several pieces of the rule that are currently left to interpretation and it should never be that way. Airlines should clearly know what they’re dealing with.
  3. Do not punish flights that attempt to return to the gate at the 2h30m mark. Today, airlines start canceling earlier and earlier to make sure they don’t cross 3 hours. If the gate return is started at 30 minutes before the time limit, then that would encourage airlines to at least keep their planes out there until that point instead of canceling earlier just to be safe. This will get more flights in the air.
  4. Require all airlines to report this data (not just the bigger ones) and make them code cancellations that were caused specifically by the tarmac delay rule so even more accurate measurement of the impact can be made. Right now, it requires a lot of guess work because the data isn’t fully broken out.
  5. Do not expand the rule to small airports and international flights until changes can be made to reduce the number of cancellations caused by this rule. The impact on travelers on international flights will be greater because there are fewer options for getting people to their destination.

Sounds entirely sensible, so the DOT will probably ignore it. Even worse, I bet Secretary LaHood will once again come out beating his chest and attacking these guys for actually doing real statistical analysis. I really hope I’m proven wrong on that point, but I don’t have much faith.

What do you think?

Get Cranky in Your Inbox!

The airline industry moves fast. Sign up and get every Cranky post in your inbox for free.

32 comments on “A Modest Proposal on Tarmac Delays

  1. I think the most important thing about this is the defined fees. The airlines want to know their costs of keeping a flight on the ramp for 3.25 hours and getting it in the air but right now they don’t.

    I’d also think of adding two things to the chart:
    1. Scale the fines by flight length, a 1 hour flight shouldn’t have the same fine as a 4.5 hour flight. I’m much more willing to wait if its a longer flight. Still publish it in table where airlines can know their exact costs.
    2. Provide a credit on the fine, if that flight is the last one to that destination of the operational day. It’s one thing if the flight cancels and I can get there later in the day, it’s another if I’m gonna get there tomorrow.

  2. I’m not thrilled about the “We really, really, tried after 2 1/2 hours to get back to the gate” idea. That doesn’t solve the airlines not having any gates or tarmac available to actually accommodate the returned flight.

    1. Take a look at the study to see more info on where this is coming from. (I couldn’t get into too much detail without making everyone fall asleep here.) The rationale here is that if airlines don’t start returning to the gate until 2h30m, then a lot more flights will end up getting out without having to cancel. For the ones that return at 2h30m, most of them are back by 3h with nearly every single one back by 3h15m. I don’t recall the exact numbers right now, but it’s in there somewhere.

  3. On the other side, there were more than 5,000 additional cancellations this year versus last. These also cannot all be attributed to the rule. But with weather being better, you would think that cancellations would be on the lower side.

    This goes to show up that FLIGHT CREWS are being worked to maximum hours or long days. 10 to 12 hours is an average day for most crews. Now add in a tarmac delay and crew legalities will attribute to most cancellations. We can NOT fly past 15 HOURS nor should we. I’ve been put into several situations this summer on how to commute home after my trip had ended. Weather causing delays for departures. One of my first decisions was to try and track down the crew and ask them, “How long have you guys been on duty and when do you go illegal?” That ANSWER has gotten me home EVERYTIME.

  4. How about that the airlines overscheduled flights? In the past they could hold people hostage for several hours, all while pretending that they had these time slots for departures. Now they can’t get away with this.

    I only take the claims of the new rules hurting when someone can show that there was not overscheduling in the first place, but sadly none of the reports ever seem to look at that.

    1. So for the airlines not to over schedule flights they’d have to agree how many slots each of them get at any given airport. Once they do that, they’re running afoul of anti-trust laws that prevent them from dividing up a market. (Which is what dividing slots up is.)

      I think there have been a few cases of airlines asking for limited anti-trust exemptions to do this, but I forgot what happened with that.

      1. You do know they already agree on slots? Slots can be bought, sold or traded.

        The problem is when the airport capacity is 60 departure slots an hour under perfect weather conditions and the airlines schedule 65 for that hour (or even 60 for that matter). All it takes is one problem to push departures further and further out. I’ve read articles about how airlines “have” to offer peak time departures because competitors do when there isn’t the actual airport capacity to back them.

        An example article about slots and capacity:


        And from CF himself, although the conclusion about hurt is silly. If something is more desirable you pay more – this is by far the best solution to things we don’t have infinite supplies of like slots, housing, cars etc.


          1. If this rule is going to be blamed on extra cancellations then what reason other than insufficient capacity would there be for those extra cancellations? (Yes weather, airspace etc locally and elsewhere could affect the capacity.)

            Without the rule it was up to airline discretion what to do when they (and their competitors) were trying to funnel too much traffic into too little capacity. Now at least they have a hard deadline and so do their competitors.

            An alternate rule would be to allow any passengers to get off the plane on request after some amount of time has passed without takeoff. That would create significantly worse problems although it is fairer in leaving choices up to the individual passengers.

          2. I know that people are looking into airport capacity as we speak. Hopefully we’ll have more on that in the near future. But the reality is that if airlines are overscheduling, there’s an easy fix. The feds just need to cap departures. Period.

            That happens today at the three big New York airports and those are really the only ones that have capacity-based slots. (Places like Long Beach and Washington/National have slots but not because airlines were overscheduling.)

            Outside of New York, the problem doesn’t appear to be an issue of overscheduling during good weather. The problem is the variability when weather goes south. That’s when the vast majority of long delays take place. Bad weather brings capacity down. That’s the issue. But you can’t schedule for worst case bad weather. I mean, you can, but then fares will skyrocket and you’ll have plenty of empty runway on most days of the year.

          3. Airlines are publicly-traded corporations and as such have a primary duty to maximize the return on investment of the shareholders. They can be sued by shareholders for failing to do so. Similarly, any appearance of collusion between companies can be prosecuted under anti-trust law. One of the few exceptions is when the company is forced by local, state, or Federal law to do something to the contrary.

            Scheduling is a great example. Differences of mere minutes can drive sales, particularly with online travel agencies. YYZ is case in point: there is a strictly enforced noise abatement rule that restricts departures before 0630 local. Yet, some airlines there schedule departures before 0630 and allow in the block time to have the aircraft sit at the end of the runway for 15 minutes. That 0615 published departure time is apparently worth the extra revenue.

            Slots are not considered contrary to anti-trust law because they are imposed by the government. The only way to get airlines to schedule flights in a non-competitive manner is by forcing slot arrangements on every airport in the country. I just don’t see the political will to do it.

    2. Written by Roger on November 22, 2010.
      Reply How about that the airlines overscheduled flights?

      Wasnt it, YOU, the flying public that wanted choices? Arent the airlines meeting demand? That said, these tarmac delays (over 3 hours) were very, very small compared to how many flights operated on-time or with a delay, even in bad weather.

  5. Hey, Cranky, I agree, except for point number 5. If the DOT really believes this rule is working so well, they should immediately expand it to small airports and international flights. You will see an additional increase in domestic flight cancellations as the airlines attempt to provide slots for their internationals, which are impossible to protect.
    PS. Let’s replace LaHood with Jim Oberstar.

    1. Right, the DOT may think it’s working well, but the point of this study is to show that it’s not. The biggest problem with international and small city is that there’s less frequent service and therefore it will be harder to reaccommodate people when a flight cancels. Remember that the issue isn’t just US airlines but international carriers. They don’t have the ability to cancel domestic flights to make room for international. And for small cities, those are the first flights that US airlines will cancel.

  6. The Feds have their hands full right now dealing with the back lash in the media on pat-downs, so this report will go on the back burner and fade away. It may get attention later when the winter snow storms start causing problems and some news outlet picks up on it.

  7. I’m don’t know much about the inner workings of this issue (i.e. slot management, etc.), or what the best solution is, but short of an exceptional event, I just cannot wrap my arms around how it is acceptable to keep passengers hostage in their seats for anything past 1 hour without providing comfort and movement that one would have available in flight. Maybe I’ve missed something on this, but exactly what passenger’s ‘rights’ apply to this condition?

    1. Being able to get up on the ground is something the feds could fix if they want. It’s not the airlines trying to hold people hostage. The rule is that you have to be seated and belted when on an active taxiway. And when airplanes are sitting in a conga line toward the runway, they don’t have the option of getting out of line without losing their spot. I agree that it sucks to keep people seated for that long and I would bet the airlines feel that way as well.

  8. Pardon my skepticism! I’ve never trusted the airlines giving me the real reasons why a flight may have been delayed, so I am not willing to trust the airlines with “reasons” for cancellations.

    Be that as it may, have complaints to airlines and to DOT about cancellations gone up since the rules went into effect? I would hope DOT has data to show this one way or the other. If complaints have gone up, I would hope somebody independent, like GAO, would evaluate the data and report on the situation. The above plan recommendations may be perfectly valid, but I would like to see more analysis.

    1. I think there’s a lot of noise in the complaint numbers. Correct me if I’m wrong, but didn’t the DOT recently begin taking complaints electronically? I seem to remember there being something that happened that made complaints easier to make so numbers spiked. It also doesn’t help that there have been two mergers which also will make year over year comparisons tough.

      But I’m with you on the GAO. That’s one governmental body that I think does great work, and I would assume that they will tackle this eventually. Not sure what other data they can analyze though, because this is really all the data that’s out there.

  9. I know I’m gonna lose this argument, but I seem to have a love for banging my head against brick walls these days.

    By your argument, the total number of people “hurt” was higher under this rule. I’ll stipulate to that fact for the moment. But how much were they “hurt”? Was their trip delayed perhaps by only a few hours, during which they could move around in a terminal rather than being confined to an aircraft seat? Or did the delay cause a loss of at least one travel day? Statistics along those lines need to be collected as well.

    Even if you can quantify that, how do you compare that “hurt” against the horror stories of being confined to an aircraft for hours on end, without any effective recourse? How do you compare the loss of a day’s travel against six hours in an aircraft with no food or water, no toilet privileges, and no effective climate control? Is it more fair to “hurt” a larger number of people, or to “hurt” a smaller number of people but create the potential for horrors like we’ve seen? I don’t know how to make the comparison; it’s a very “apples-and-oranges” sort of comparison.

    Having said all of that … sure, tweaking the rule to make it sure it achieves the ends it really wants makes sense to me.

    1. You bring up a good point in my mind, and that is how the “hurt” differs for each person that’s affected. Take a hypothetical example of a flight from DFW-LAX with a return 3 days later, 115 passengers on board each time. Let’s now pretend that there’s bad weather at DFW at departure. The airline pre-cancels the flight and says nothing’s available until tomorrow. Let’s say 59 passengers live in the area and are originating at DFW, 21 live in LA and are heading back home, and 35 are connecting from somewhere else on their way to LA. For those originating at DFW, you may actually feel less “hurt” by a pre-cancel than sitting on the runway for 6 hours. Yeah, it sucks to have a trip shortened by a day, but you’re probably just going to go sit at home or go back to the office. Not a bad situation if this were me. On the other hand, those trying to go back to LA or connecting in DFW to get there are probably really hacked off right now, because they’re now stuck in DFW at their own expense (airlines don’t compensate you for weather cancellations). I can’t speak for everyone, but I would MUCH rather sit on the plane for 4 or 5 hours than either be out a couple hundred bucks for a hotel room and meals, or scrimp and sleep on the terminal floor – and have to take an extra unplanned day off work to boot – in this situation.

      The problem is, I’m making (and for that matter, anyone who studies the rule is doing the same) an assumption that one group is “hurt” whereas the other isn’t, but without actually surveying each passenger to see how annoyed they really are, it’s impossible to quantify the real effect. I tend to agree that this rule is leading to increased cancellations, but is it a good thing or bad thing? Not sure we’ll ever be able to tell for sure.

    2. It would be nice to have that info available, but it’s just not there. So, it’s hard to win or lose that argument, because we just don’t know. By the same token, it would be nice to know how many of those people who had a long tarmac delay were happier having that than going back to the gate and canceling. Again, it’s info we’ll never know.

      But the comparison you make with 6 hours on an airplane with no water or toilets isn’t really a fair one. There are other parts of the rule that already prohibit that, and I haven’t seen anyone really arguing that part. It’s just the time spent on the tarmac, not the amenities onboard that are up for debate. Also, the instances of people being stuck on an airplane for 6 hours are incredibly rare. Those are terrible situations and should never happen, but they happen so rarely.

      1. The rarity of those events isn’t a valid argument against the policy. Acts of terrorism involving commercial air flights are infinitesimally rare — yet that doesn’t stop the government from demanding the right to see every passenger virtually naked in the name of “homeland security”.

  10. Looks like the DOT has now responded and, as expected, has completely dismissed the claim using bogus data interpretation. Here’s the blog post:

    The most relevant piece is this:

    Recently, a pair of airline consultants cooked up a report alleging that our rule limiting tarmac delays was responsible for thousands of cancelled flights between May and September. That is simply not true. In fact, the same airline consultants’ report admitted that 4,793 of the 5,068 additional flight cancellations they counted in May–September 2010 were cancelled before the plane even left the gate, which means they have nothing to do with limits on tarmac delays.

    The notion that any flight that was canceled before the plane left the gate is irrelevant to tarmac delays shows that Sec LaHood has absolutely no understanding of how airline operations work. Sad to see such an oddly dismissive piece.

    1. Argh! This rule gets me more and more incensed every time it comes up. Not to go too political, but I’m nominally a democrat, although this sort of crap makes me agree with the republicans.

      I’d be perfectly fine with a tiered system that starts out low with fines for say delays between 3 hours and 3 hour 30 minutes, then ramps quickly to the maximum but this stuff is insanity.

    2. You really expected a different response, Cranky? As a cabinet secretary, he’s little more than a water boy for his bosses, and right now, his bosses are completely controlled by the flyer rights crowd. You can throw out any statistics you want, but you’ll be dismissed as a shill for the airlines and “big corporations”. That’s basically how DOT views anybody who disagrees with the tarmac delay rule, unfortunately.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Cranky Flier