The problem you’re going to face with your system is that the people who make multiple bookings do so not just out of malice (“Hey, with my fare it’s free to rebook, so why not book a dozen flights just to piss off the airline?”) but for a good reason: They want to keep the flexibility of choosing between several different flights. If your cancellation fee is low, then it won’t deter them from doing that - your $20 cancellation fee in your example is, to them, well worth the flexibility. If you hike up the cancellation fee to a level where it really hurts (say, 80-100 % of the fare), then in essence your system comes down to selling non-rebookable tickets that are limited to a particular flight. But that is precisely what this sort of customer does not want - these customers want to be able to rebook, and that’s why the airlines are willing to sell them such tickets (at higher prices, of course).
The few passengers who habitually multiple reservations are probably buying fully-refundable tickets, though, which means they’re not really at risk of being bumped in an overbooking situation. I’m pretty sure that when passengers are chosen for involuntary denial, the computers take the fare class into account. Meaning they go after the cheap passengers first.
Various posts debated whether/how many tickets of people who don’t board the flight they booked are paid for or reusable, but that only affects the answer to the OP question in degree. No matter what you assume about the revenue from tickets sold to people who don’t end up on that particular flight, selling more tickets means a higher probability of a plane completely full of paying passengers. And a plane full of paying passengers is more remunerative, since the cost of the flight is in fact almost constant whether it’s full or not*.
The only way that gets undone is, to get back into the weeds, if the revenue from unused tickets has a particularly unfavorable relationship to compensation the airline has to pay to bumped passengers**. But there’s no reason to think that’s the case, or the practice wouldn’t exist.
The remaining objection is likely to be ‘well just because it makes the flight more remunerative for the airline doesn’t mean it cuts (consumer) cost’, but that’s just a matter of competition (or lack of enough) and applies just as much to every other aspect of cost. If there’s not enough competition then cheaper airplanes, cheaper labor costs, cheaper fuel, lower airport fees, etc won’t necessarily result in lower ticket prices either. But assuming competition they all will, and likewise airline marketing practices with optimize revenue per flight.
*the capital cost is the same, labor cost the same, operating cost of empty plane the same: the variable fuel cost according to the weight of one more passenger and luggage is a small fraction of each passenger’s pro rata share of the total.
**say some regulation said you had to pay $10k (not max in an auction for volunteers that will never get to $10k in reality, but everybody) to bump, and other regulations said that people who booked but didn’t board a flight, for any reason, always got 100% refund to begin with.
You said “economical.” What you probably meant was “cheapest sticker price”. Those are very different ideas. Which speaks to your confusion / cynicism. Not picking on you, that’s ordinary consumer behavior in this silly messed up business.
If you just go to (e.g.) Expedia and pick an itinerary from ABC to DEF on date X you’ll find that Expedia makes an implicit assumption that the only thing you care about is “cheapest sticker price.” They’ll sort and filter on that basis.
If you dig beneath the surface a bit you’ll find a lot more options. In many cases you need to go directly to the airline’s website or even their 800# customer service to learn about all of them. Just like Home Depot doesn’t stock every flavor of wall paint by every manufacturer, Expedia “stocks” just the most popular stuff.
There’s a circularity here that the customers can only pick from what’s on offer at the store. So that artificially drives “most popular” to be “whatever they stock”. And vice versa. Many, many retailers have blown themselves up by slowly shrinking their stock to just their fastest sellers then some invisible threshold is crossed and suddenly people completely stop shopping there since the selection’s now too small.
These people are already paying extra for the refundable tickets. These are mainly business travelers who need the flexibility (i.e. whose time is very valuable to their employers). Charging penalty fees for cancellation probably won’t change their behavior.
I really don’t see what’s “broken” with the current system. It discriminates against people who pay less, but that’s just capitalism.
The math problem is the trade-off between the cost of bumping and the revenue from no-shows.
The added quirk is the higher-paying passengers who can no-show with no penalty. I was in a meeting once where there were participants from a distant city. the big boss droned on and on and asked too many questions, but fortunately there were plenty of flights - the participants from away snuck out to rebook over and over - 12:30 became 2:00 then 4:00 and finally 5:30.
Now these were top of the line tickets (but not first class, we peons didn’t rate first class). This is the airlines’ nightmare. Someone pays the “free re-booking” ticket and does no show up. The aircraft flies with an empty (unpaid) seat, or you overbook based on history and statistics and gamble you can squeeze a few bucks out of the Gods of Math.
You and I, the bump-able peons, buy the cheap tickets where the cost of rebooking or cancellation typically is close to the total cost of the ticket. If we non-show, the airline still - eventually - gets their $150 change fee or else the ticket is non-refundable or something and it’s cash in their pocket.
For the no-fee rebook tickets, it’s revenue down the drain unless overbooked. A passenger like my compatriots who rebooks over and over with no cost means several empty seats and lost revenue. This, of course, only if the flight could be filled. But, every flight I’ve been on in the last decade seems to be packed solid, whether to Toronto or Vancouver or London or Dubai. Which suggests that when a person with a free-reschedule ticket rebooks, they are in effect bumping a paying passenger out of the cheap seats - if they show up for that booking.
I did a quick Expedia search for the Chicago-Louisville flight. A few weeks in advance, the ticket - one way - is $100. So if the bottom-feeder customer is bumped from that flight, the most the airline would have to legally pay is $400 (4 times the ticket). Add in $25 for a bag, $15 for a seat selection - may be almost $600 max. United were being generous at $800. (To clarify a post objected to in another thread, one news item confirmed Delta, for example - gate agents were only authorized to offer up to $800 and only supervisors could offer $1350 - the absolute legal maximum that the law required them to pay if sued.
The 1350-dollar question, then, is what proportion of the higher paying passengers who have free no-show privileges typically tend to no-show? What does the ticket cost? If, say, there were 5 no-shows per flight, then a good gamble would be to sell 3 to 5 el-cheapo tickets and hope you bump as few as possible. If the open ticket cost $800 and you have to pay your bumped passenger $800 then you broke even.
Again that’s a trade off which would have to be heavily negative (for overbooking) to make overbooking unattractive from airline’s POV. Assuming that trade off is flat, ie. revenue from no-shows just covers compensation for bumping, then overbooking raises profit all else equal. Because it means planes more nearly full of paying passengers, and the cost of the flight is mainly fixed (incremental fuel charge per weight of passenger/luggage is a small % of the cost).
IMO to answer the question asked on the simplest basis we don’t really need to get into the weeds of the great complexity of fares and booking. It was asked ‘How does overbooking flights lower costs’. If we’re not sure everyone understands the basic answer: ‘by raising load factors’, I think it risks confusion to then dive into the ifs and buts of all the details of booking and rebooking.
And IMO it also should be obvious that unlimited compensation for voluntary bumping, via competitive auction among passengers to generate enough voluntary bumpees, is not a major part of the equation. As of now it’s few $100 max per passenger (say a voucher avgs 50% the value of real money) on a quite small % of passengers. Then lift any comp limit and it would still be just a few $100 in the great majority of those cases (per stats that voluntary heavily outnumber involuntary bumps now), then every once in a while it might take $2k or something per person to clear the excess. That would be a rounding error in total airline cost structure.
The mysterious question IMO is why there are comp limits in the regs for bumping and why airlines didn’t override them. There was a skewed outcome of saving a tiny, tiny fraction of their cost base very occasionally having to bully people off airplanes when a few $100 wasn’t enough to generate the necessary volunteers, vs. a not totally unforeseeable possibility of an eventual blow up. UAL might not have saved as much on involuntary bumpings (v the compensation it would have taken to generate volunteers) over many years* as they will now pay out to one passenger, even besides the PR cost. But market participants are often not good at dealing with highly skewed outcomes.
*not the same as the revenue increase by overbooking and voluntary bumping which would have been vastly larger, since the great majority of overbookings don’t result in bumps and the great majority of bumps are voluntary. And limit it to the savings just from kicking people off planes (not denying boarding) involuntarily rather than raising the offer to generate volunteers, and it’s even less likely that unusual maneuver ever generated as much as they’ll now pay one person.
IMO it’s important to keep what happened in Chicago separate from overbooking.
- The now-famous pax was not asked originally to deplane due to overbooking. He was asked to deplane for the convenience of the airline to accommodate their unplanned need to transport crewmembers for other flight(s).
- The now-famous pax was eventually removed (excessively) forcefully for being a security risk. Not for overbooking.
So almost nothing useful can be said about overbooking if one is starting from the Chicago incident. The incident *might *be a reason to bring the idea of overbooking to top of mind as a thread topic. But after that it’s a 100% distraction from what “overbooking” means, how it works, and what it’s consequences typically are (and worst case might be) for all participants on all sides.
Well said.
And, as we have seen, the sudden focus will attract many “obvious” and “fair to all concerned” solutions from people who have thought about the issue for all of 15 minutes and have not bothered to look at any data.
Ok, so what does this mean? ‘by raising load factors’ I’m just trying to understand an admittedly (needlessly?) complex and complicated type of commerce. And now I have another question - LSLGuy: I think you’re a pilot, yes? How does the “pax” go from simply refusing the airline’s request to a security risk? Is that automatic? Did he show annoyance or anger as a result of being asked to give up his seat? Is being pissed at the carrier all that it took to turn him into a security risk? Do you, as a pilot, have final say?
One news report says the flight crew called the police, and to get the police to remove the guy, had to declare that he was disruptive and belligerent - which is why the CEO parroted the party line in his original non-apology the next day. presumably filing a false report with police is a crime.
Again, the problem to get back to the OP, is that the no-shows are predominantly those who have paid extra for flexible booking and will fly (occupy a seat) on another flight. So an empty seat is a non-paying seat that could generate revenue if it was embutted. Whether that guess as to the number of no-charge no-shows is accurate, well, that’s the whole business side of overbooking.
This thread explains in excruciating detail why airlines over book flights. That is not the question of the thread. If the answer is in the posts, I missed it, but there must be a simple explanation. How does selling more tickets than there are seats lower the price per seat? It can’t simply be that the ticket price is calculated according to the cost of the flight divided by the number of tickets sold. I don’t believe that the airlines would do that. If a plane isn’t full when it takes off, the price of a ticket doesn’t suddenly go up. What am I missing?
I’m really not sure what you’re missing.
If airlines do not overbook, every no-show costs them money, because the average ticket is at least partially refundable. If the airlines do overbook, then on some occasions there will be more passengers than seats, and compensation will cost them money. The degree of overbooking on each route is therefore a carefully optimized calculation based on no-show statistics for the route to minimize these expected costs. In the long run, the lower the airlines’ costs, the cheaper they can sell tickets.
As I mentioned about the meeting I was in that went overtime… The people who rebooked and rebooked and rebooked - each only paid once, but tied up 4 different seats on 4 different flights. It’s not that the costs go up - it’s that if the airline could count on (and they do) a certain number of those no-penalty-to-rebook seats going empty, they could re-sell them to paying passengers; but it seems nobody does standby anymore. (Remember flying standby? I asked about it in 1990 and was told it no longer existed after deregulation). Plus someone who is willing to pay a penalty to rebook last minute is additional revenue, or misses they flight and loses it all - if you can also re-sell that seat, even more money.) Without a line-up at the gate waiting to pay, the only solution is to promise someone they could have the seat, and then yank the rug out from under them if too many people showed up to board.
It’s not “the price of the flight goes up if the seat is empty”. It’s “We can make more money this way”. In the long run, making more money does mean the price goes down, as multiple airlines compete to get your business.
I believe I gave the simple explanation but again.
Assume for whatever reasons, things having to do with the excruciating details of modern airline ticket marketing or not* that materially less than 100% of passengers who have booked a flight show up for it on average. In that case selling only 100% of the seats will gtee empty seats on the plane. Selling a well calibrated excess over 100% of seats will result in closer to 100% utilization of the plane. A 100% utilized plane has a lower cost per passenger than an underutilized plane. Simple as that.
It lowers the cost per seat per flight. Whether it lowers the price per ticket depends on the competitive environment. But the same is true for a price cut by Boeing/Airbus in selling the airline the plane, or a fuel price decrease, and nobody is asking for an explanation why those things would lower costs.
*getting sick at the last minute, having car trouble on the way to airport etc.
This is the main point. Airline consumers (at least the “back of the plane” ones) are very price-sensitive. The tend to go to online booking sites and pick the smallest number on the screen. Most of the reason why economy-class travel sucks so bad is because so many people pick an airline solely based on price.
So an airline that finds a way to save money by making travel on their airline just a little worse has a good incentive to use that savings to lower ticket prices (instead of just pocketing it, say).
In some alternate universe where economy-level airline passengers pick airlines based on the softness of the seat cushions, or the nutritional value of the snacks, or the impossibility of being getting bumped, then airlines with reallocate funds to that. But their experience is that price is the only thing people respond to is price, so they aggressively lower prices.
If we assume the overbooking probabilistic math works and makes the flight more profitable because of no-shows (flight sold on average at 108% capacity for example), then either the flight is more profitable or (for the same profit margin) each seat is sold at a lower cost. Market forces should dictate which.
These last two or three posts are it in a nutshell.
Given that the US airline industry has collectively lost money since deregulation, Pedro’s closing question is answered. The benefit goes to the customers in the form of reduced ticket prices.
Which, *pace *leahcim, is exactly what the consumers have told us they collectively value most.
And, I believe, this is why most airlines have resorted to nickel-and-dime fees – for checking a bag, for an in-flight snack, for wifi service, for a seat with a little more legroom, etc. None of those fees are part of the “ticket price”, and thus price-sensitive travelers who are comparing prices on an aggregator site like Expedia, won’t be able to easily incorporate that information into their shopping decisions.
If you believe that the law of supply and demand applies to the airline industry…
Overbooking increases supply and therefore (absent other confounding factors) should lower prices.
Example: Let’s say there are 5 flights with 100 seats from Chicago to Louisville on a given day. If we allow, say, 10% overbooking, the supply of available tickets is 550. If we ban overbooking, the supply is 500.