65

A quick google search returns several articles such as this one claiming that airlines make less than $10 in profit from each customer.

I feel there are a few reasons to doubt this premise:

You can connect through some destinations more cheaply than landing there

Skiplagged is a service helps users find cheaper airfare by booking flights that connect through their actual intended destinations to ones that are less popular and therefore are being sold more cheaply. Users are then expected to simply intentionally miss their final connection.

This means that a travel agency or airline can book seats on more planes and sell that connecting flight for less money. If the airlines have very small profit margins, then that proposition wouldn't just return less profit; it would lose money.

Ticket prices are higher in recent years

After 30 years of gradual decline due to competition, prices have been increasing since 2010 and have returned to levels they were at before the 'great recession'.

Rising prices are contrasted by apparent decreasing costs

  • Fuel prices have come down since spiking in 2010
  • Pilots work long hours and in some cases for surprisingly low pay
  • There's less leg room than ever due to cramming more seats on each plane
  • Meals and other niceties common in the 90's have disappeared from most flights
  • Many flights now don't have screens to watch movies on or ports to connect earphones to; customers connect their devices to the entertainment system via wifi

So, is the claim that airlines turn $10 profit per customer accurate?

Oddthinking
  • 140,378
  • 46
  • 548
  • 638
Kevin Laity
  • 829
  • 1
  • 7
  • 9
  • Also, the North America reference in the title can go, the IATA article linked to addresses other flights and companies as well. –  Apr 13 '17 at 15:01
  • 6
    Note that overbooking has more to do with the fact that quite often, not everyone who buys a ticket shows up, and airlines want to make as much money as they can regardless of the actual per-passenger profit. http://www.post-gazette.com/news/transportation/2017/04/13/united-airlines-incident-doctor-passenger-dragged-off-plane-overbooked-flights-fly-rights/stories/201704130011?pgpageversion=pgevoke – JAB Apr 13 '17 at 15:25
  • 2
    @JAB You're of course correct, but the question is, is that merely a 'cash grab' or is it truly a necessity to avoid running a loss? – Kevin Laity Apr 13 '17 at 15:27
  • 2
    @Jan Doggen I've removed the mention of the video, but not all of my cited information applies worldwide, and the picture of 'what does the average airline make from each customer' could vary widely if asked about the whole world. – Kevin Laity Apr 13 '17 at 15:36
  • 8
    I don't understand the first point. The $10 per customer claim is clearly an average, and it doesn't contradict the possibility that some tickets are much more profitable (e.g. those where the airline charges more for a nonstop than for an onward connection). – Nate Eldredge Apr 13 '17 at 16:59
  • 2
    @JAB: If making money was the goal, it would seem more reasonable to have an "all ticket sales are final" policy, then if the passenger doesn't show up, make more money by selling the seat again on standby. – jamesqf Apr 13 '17 at 17:37
  • @jamesqf By my understanding, most lower-cost ticket sales _are_ final (nonrefundable, nontransferable, etc.). Refundable tickets tend to be more expensive. – JAB Apr 13 '17 at 17:45
  • @NateEldredge Exactly. Obviously, an airline would prefer to sell every seat for top dollar. But it certainly is plausible that they might take a loss by selling some seats cheaply so long as it's less of a loss than they'd take if they didn't sell those seats at all. If the business was perfectly competitive, we'd expect the airline to make pretty close to zero on the average seat which means they must be taking a loss on some of them given the wide variation in prices. – David Schwartz Apr 13 '17 at 22:14
  • 5
    @KevinLaity: I would dismiss your premise - that there is a qualitative difference between a "cash grab" and "necessity to avoid running a loss" for a business motivated to make a profit for its shareholders. – Oddthinking Apr 14 '17 at 02:40
  • Maybe this is inference, but the question seems to suppose that $10 profit per customer isn't much profit. It sounds like a reasonable profit to me in a mature and evolved market. – Ian Newson Apr 16 '17 at 06:34
  • @jamesqf The problem is your most profitable customers demand flexibility (and are happy to pay through the nose for it if you want to offer it). When you look at major finance houses or whatever who have people commuting twice a week between London and New York in business class on unpredictable schedules, any airline that stuck with such a hardline policy would lose a lot of very lucrative business. That said, the airlines' biggest customers get significant (often upwards of 40%) discounts on the public fares. – Calchas Apr 17 '17 at 00:22
  • The reading of the claim as given by Kevin Laity in the question, as indicated by the questions about flights being cheaper suggesting airlines taking a loss at $10 profit per customer, deserves an answer. Unfortunately, answers to that (entirely valid) interpretation are being deleted by the mods. Since moderators seem to be dismissing premises left, right, and centre, I figured I'd go for a comment: if you read "profit per customer" as "profit from each customer", then you're after Unit Contribution Margin, and its value will be larger than $10. – Glen O Apr 19 '17 at 15:41
  • 10 bucks of pure profit by customer is a really good deal, after you consider how well paid airliner staff is usually paid and how expensive the stuff they operate is. We are talking about pure profit, after operational costs - so, surplus money! – T. Sar Apr 19 '17 at 18:54

4 Answers4

102

TL:DR

United makes about $9.50 per passenger.

Detailed:

United Airlines Q4 and full year 2016 performance results offer a great starting point for these calculations.

We can see that United recorded a Q4 net operating income of $397 million. In the same time frame, United recorded having moved 36.023 million passengers. This gives us a baseline number of $11.02 net income per passenger.

This number is only a good starting point though, as it assumes all of United's revenue is from passenger traffic, but they also recorded revenue of $1,291 million from "Cargo" and "Other Operating Revenue". To get a more accurate picture of the profit/passenger, we'll have to dig a bit deeper.

United registered a net revenue from "Total passenger revenue" of $7,761 million for Q4. With their given pre-tax margin of 9.8%, we can see that they had a pretax revenue per passenger of $21.11. Their "Income tax expense" of $487 million on "Income before income taxes" of $884 million represents 55.09%. Applying that rate to their per passenger income drops it to $9.48 net income per passenger.

Their yearly result using the same method is $9.22 net income per passenger

I've been unable to find Southwest's passenger counts, but they have a pretax operating margin of 18% with a net income margin of 11%, so I would expect that they make a bit more per passenger than United.

Same story with American Airlines. No passenger counts, but a pretax operating margin of 11% with a net income margin of 7%.

A Bailey
  • 2,101
  • 1
  • 13
  • 13
  • 40
    So if someone buys one of those silly bottles of "artisan water" that cost $15, they provide about twice the profit of a traveler who doesn't? – Nat Apr 13 '17 at 19:02
  • 62
    Yes. Much like if you buy a TV and an overpriced HDMI cable at an electronics retailer, they'll make more profit off the cable than the TV. – Aric TenEyck Apr 13 '17 at 19:06
  • 8
    I feel I should point out that, if you're restricting the revenue to money spent by passengers, you also have to restrict the expenses to those relevant to passengers. – Glen O Apr 14 '17 at 05:51
  • 1
    @GlenO, That was my goal when calculating expenses as a percentage of the passenger based revenue. I think it would be cool to open up their books and really calculate the revenue and costs associate with particular flights, but I don't feel that contacting United to ask for detailed financial information will get me anywhere. – A Bailey Apr 14 '17 at 12:37
  • @Nat indeed. And those horrific numbers are a prime reason many airlines prefer cargo over passengers, and will treat their passengers worse than their cargo. – jwenting Apr 14 '17 at 12:57
  • 2
    Interesting to note is that although airlines may make $10/p who shows, they make en extra entire ticket cost for the price of nothing depending on their refund policy. Also to note is baggage fees aren't taken into consideration for this, and that $25+ a bag in a lot of cases. – Anoplexian Apr 14 '17 at 14:13
  • 1
    @Anoplexian, take a look at the comment section to PoloHoleSet's answer. There's a good discussion about the profitability of no-shows vs the cost of bumping passengers there. The bag fees are a great point. Although Southwest has show a much higher margin without bag fees than the other airline who utilize them. I'd love to have a few days with everyones books just to see what's happening and why – A Bailey Apr 14 '17 at 14:45
  • @ABailey Indeed, just thought it important to note. – Anoplexian Apr 14 '17 at 14:58
  • 10
    @Nat this answer uses calculations based on the _total_ profit United makes and averages it out per passenger. It's likely that if you actually broke it down per passenger, those who buy no items or check no bags actually make the airline less than $9.50/person by a fair bit, but it's pretty difficult to fully determine the actual costs per person - maybe they paid more for their ticket, etc. – enderland Apr 14 '17 at 18:25
  • I think this answer mixes up `,` and `.` in numbers. You may want to double-check that. – user Apr 14 '17 at 18:35
  • 1
    @MichaelKjörling, https://ux.stackexchange.com/questions/23667/is-using-a-comma-as-a-number-separator-a-cultural-thing – A Bailey Apr 14 '17 at 19:59
  • @MichaelKjörling I think you may be mixing up net with gross. – Matt Apr 14 '17 at 21:35
  • 5
    "After you _tip_ your flight attendant, please return them to the upright and locked position." – Dennis Williamson Apr 14 '17 at 22:21
  • The 2017 results are going to be interesting... – papirtiger Apr 15 '17 at 18:06
  • 1
    "pretax revenue per passenger of $21.11" — shouldn't that read "pretax income"? – vartec Apr 24 '17 at 20:35
  • 2
    Keep in mind that those are _profits_, money that they get after paying for everything, including staff. Those ten bucks is _extra money_. – T. Sar Apr 26 '17 at 11:40
12

Generally, yes, but that's not necessarily an argument for overbooking as aggressively as they do (the context in which the claim was made). They don't make a huge amount of profit per customer, as per the numbers in A Bailey's answer. By the same token, since they move such a large volume of customers, they don't incur huge per customer costs, either.

So, the argument that they must overbook because they don't make a large amount of profit per customer is offset by the fact that reducing the overbooking, slightly, would also mean they're only losing a small amount of profit per customer, as well.

You can't claim one and ignore the other. Airlines are hugely profitable. 2016 was a record year for them, as a whole ($35.6 B in profits), and 2015 was a record before that ($25.6 B in profits). Fuel costs are 1/3 what they were two years ago, yet fares are still about the same.

NYT: Airlines reap record profits, passengers get peanuts

IATA: Another Strong Year Expected For Airlines

CNN Money: Airlines posted record profits last year (2015)

PoloHoleSet
  • 9,608
  • 3
  • 34
  • 41
  • 16
    "they're only losing a small amount of profit per customer". Not true, since these customers have paid full price, but not utilized any of the services. A no-show to an overbooked flight represents mostly profit to the airline. $194/no-show for United by my calculations. To put it another way, United has to sell about 20 less tickets for each non-show to an overbooked flight. – A Bailey Apr 13 '17 at 18:15
  • 3
    There are several wyas you can calculate the value of a no-show, but however you choose to do so, they are much more valuable to an airline than $10 – A Bailey Apr 13 '17 at 18:16
  • 5
    @ABailey - I didn't say that it was the same. Your calculation ignores the costs associated for bumping a paying customer, which disappear if you don't overbook flights AS MUCH. I'm not sure why you are fixating on no-shows, we're talking about overbookings and people who do show up. Flush with revenue and profits, airlines have recently upgraded much of their fleets to newer planes, which also skews any per passenger calculation for years when they go ahead and make that investment. What needs to be figured to be truly accurate is the non-capital operating costs of a "average" flight. – PoloHoleSet Apr 13 '17 at 18:22
  • 20
    This is flawed. Flying a plane costs almost the same amount irrespective of the number of passengers. Which means each person who flies makes the airline $10 while each person who doesn't fly costs them the ticket price (e.g. $300). [E.g. a full 300 seat plane at $300/ticket makes the airline $90k revenue. With $10pp profit, that's means there were $87k in costs. The same plane with 280 people makes $84k in revenue with almost the same $87k in costs, meaning the airline *lost* $3k]. It's this asymmetry that drives overbooking. – JoshG79 Apr 13 '17 at 19:14
  • @JoshG79 Is that assuming they purchased a refundable ticket (which would be more than $300)? – Michael Apr 13 '17 at 19:18
  • @JoshG79 - See my comment above. How much does it cost to bump people from an overbooked flight? Your analysis is also flawed. A full flight of 300 makes 90K revenue. Underbooking by 1 % costs the airlines $900 in revenue. Overbooking and bumping 1% also costs the airlines an equivalent amount - they have to give additional free vouchers, house and pay for expenses for the bumped passengers, the free vouchers mean that other flights have less capacity for paying customers. – PoloHoleSet Apr 13 '17 at 19:19
  • 3
    @PoloHoleSet, I agree with you that bumping passengers is not a free endeavor for the airlines. Every airline company knows that too. They don't want to bump people. But they have access to better data then we do. Years of records on who shows up to which flights and who doesn't. If overbooking flights cost the airlines more money than it makes them, I guarantee you, they would stop doing it. – A Bailey Apr 13 '17 at 19:30
  • 7
    @PoloHoleSet they're not overbooking with the *intention* of bumping. They're overbooking to fill slots that invariably end up empty because people don't show. You're focusing on the wrong outcome - bumping. The airlines know that for an *average* flight, X% of the people who reserved seats for the flight won't show up. Without overbooking, planes would fly X% empty on average which costs them X% in passenger revenue. By overbooking they can close that gap at the cost of occasionally bumping someone when – JoshG79 Apr 13 '17 at 19:38
  • 1
    @JoshG79 - yes, but the cost of bumping a single passenger is many times the amount of leaving that single seat open, for reasons already stated. And you're claiming that I'm saying overbooking should not happen, at all. As I said before, it's the degree of overbooking and bumping that can't be justified. – PoloHoleSet Apr 13 '17 at 19:46
  • 4
    Your logic is sound only if the two outcomes are equally likely. What if 99% of the time an overbook fills an otherwise empty seat and only 1% of the time it results in a bump? In such a case overbooking makes sense by a long shot. Two things I know for sure are 1) airlines know the ratio of fills:bumps a lot better than you do; and 2) they're really greedy and if overbooking was costing them money, they wouldn't do it. – JoshG79 Apr 13 '17 at 19:54
  • 2
    @JoshG79 - look at their profits. The claim that they are on such tight margins that they HAVE to overbook to this degree is nonsense. They are also juggling with the dueling priorities of incurring more overhead by offering more flights while making the customer experience less of a nightmare, or squeezing out more profits and screwing over the customer. Yes, they want to make MORE money, but the idea that they have to do this to the degree that they in order to make ANY money (aka, "We HAVE to overbook this much" instead of "We CHOOSE to overbook this much") is the phony proposition. – PoloHoleSet Apr 13 '17 at 20:05
  • 1
    @PoloHoleSet Their profits are not outsized relative to other industries. But that's besides the point - a corporation's job is to make as much profit as possible for their shareholders. Their overbook policy is designed with that goal and that goal only. They would undoubtedly be more customer-friendly with less overbooking. But if they're going to trade better customer experience for lower profits, it's not even clear less overbooking is the best way. Maybe bigger seats? Or no bag fees? Or simply lower ticket prices? Less overbooking is way down the list. – JoshG79 Apr 13 '17 at 20:11
  • 1
    @JoshG79 - they also have an obligation, in a service industry, to their customers, as well. Again, the argument they are advancing is that they have to do this in order to be profitable. If they want to say "we do this because we value eclipsing record shareholder profits over delivering service" I'd say "fair enough." The contention that this is necessary, because their margins are so slim is simply a lie. I'm not saying they are not allowed to make profits, I'm saying that the context of the claim is a phony one. – PoloHoleSet Apr 13 '17 at 20:13
  • 3
    @PoloHoleSet we got on a tangent; back to the real points. 1) This claim of yours is simply false: "they don't make a large amount of profit per customer ... [therefore] they're only losing a small amount of profit per customer". The profit per person and the loss from not overbooking a seat are unrelated. 2) Your other claim is simply a value judgement about whether a specific company makes too much profit in your opinion. That's not the point of this site and not answering the current question. – JoshG79 Apr 13 '17 at 20:30
  • @JoshG79 - I disagree. Their profits are made from a huge volume of passengers. The need to overbook flights to the degree they do vs. leaving some capacity for needs will not take them from record profits to insolvency. Any individual or couple of passengers on any given flight one way or the other will not make or break, because any individual customer does not represent that much impact vs the overall picture. – PoloHoleSet Apr 13 '17 at 20:41
  • @ABailey - I think most no-shows are full-fare flexible tickets, not non-refundable tickets. In decades of flying, I've never once been a no-show on a non-refundable ticket (other than weather delays that made me miss a connection), but have made probably hundreds of last minute flight changes on full-fare tickets (and the company paid that higher fare exactly because of this flexibility) – Johnny Apr 13 '17 at 21:03
  • 29
    I did an analysis of this about 15 years ago for a university project. Based on publicly available data, I determined that about 15% of tickets sold were neither used nor refunded, and that the value to the airline of an unused ticket was so high that, considering only first-order effects (eg. compensation to bumped passengers), profit was maximized by overbooking to the point that nearly every flight had more passengers show up than could fit in the airplane. (Second-order effects such as loss of goodwill were outside the scope of the analysis.) – Mark Apr 13 '17 at 22:19
  • @Johnny That's your personal experience, not necessarily the experience for everyone (though it is likely that such flight changes are more common for people who have made the decision to spend more money on such tickets). – JAB Apr 13 '17 at 23:09
  • @JoshG79 your calculation is overly simplistic. I've seen airlines (even United did that) offer certificates ($300 - $500) to people to get off over-booked flight. So overbooking is not safe money. – Akavall Apr 13 '17 at 23:27
  • @Akavall And do they offer that to the majority of those they boot off? Is the certificate worth more than the cost of the ticket those people spent? Airlines are concerned with optimizing their overall income, not individual expenditures. – JAB Apr 14 '17 at 00:12
  • @JAB, My understanding was that, in most cases, airlines did not just throw people of the planes (this might be incorrect unfortunately), people who took certificates, would just have to wait for the next flight, so they would not have to buy a new ticket. Airlines still benefited, because the gain from no shows was greater than the certificates, but it was not price of the entire ticket on average. – Akavall Apr 14 '17 at 00:48
  • 1
    @PoloHoleSet "their profits are made from a huge volume of passengers". Yes, which is why making a change that reduced revenue per passenger by $1 would cost them, in aggregate, a billion dollars per year. You're arguing against yourself. – hobbs Apr 14 '17 at 05:13
  • @JAB according to this: http://www.nbcnews.com/news/us-news/david-dao-doctor-dragged-plane-files-court-papers-demanding-united-n745721, In the infamous United incident passengers were offered $800, but nobody was willing to take it. (I am in no way justifying United actions, I am just pointing out that ignoring vouchers that airlines pay, leads to inaccurate calculations). – Akavall Apr 14 '17 at 05:23
  • @Akavall: Was it cash or just credit for a future flight? You'd think if they really cared they'd offer cash. These problems have solutions, they're just not willing to exercise them... – user541686 Apr 14 '17 at 07:19
  • 5
    This answer repeats the claim in the question without any actual evidence. – Sklivvz Apr 14 '17 at 11:48
  • @Mehrdad FWIW, the times I have been bumped the compensation has always been travel vouchers. Because the vouchers were less than the price of a ticket and on an airline that wasn't the one I normally used for personal travel, I ended up letting them expire unused. That's just one person's anecdote, of course, but it suggests that at least some of the compensation for bumping passengers never actually gets paid out. – Nobody Apr 14 '17 at 13:30
  • 3
    It doesn't really matter whether the no-shows are on refundable tickets or not, overbooking let's an airline put a butt in that seat for whatever the ticket price is. And at the absolute best, vouchers I've seen have been worth maybe two non-refundable tickets, usually closer to one, and often less. So you only need, say, two seats filled for every bump to make it worthwhile, for example. I obviously am not privy to the exact numbers, but I'm sure airlines watch the bump-to-filled ratio vary closely. – PGnome Apr 14 '17 at 13:56
  • @pwcnorthrop It's also worth noting that the compensation may well be less than the price of a full-fare ticket, so it can be worthwhile for the airline to sell a last-minute Y fare on a full flight, even if it did mean bumping someone. And, quite frankly, having received a significant amount of free travel as a result of this, I'm personally fine with it. It's a win-win-win in most cases: the business traveler gets his expensive seat, the airline gets more money, and I get a free trip. – reirab Apr 16 '17 at 07:20
  • Regarding the complaint about ticket prices vs. fuel prices, fuel is only maybe a quarter of the costs for airlines. Also, [Jet-A is almost exactly the same price now as it was 2 years ago.](http://www.iata.org/publications/economics/fuel-monitor/Pages/price-development.aspx) It dropped a lot last year, but it's recovered those losses over the past year. It is, however, about half of what it was 4 years ago, but fares did indeed drop around the same time. – reirab Apr 16 '17 at 07:47
  • 3
    "The claim that they are on such tight margins that they HAVE to overbook to this degree is nonsense." You're confusing total profit with profit margins. The margins are tiny. IATA [recently issued a press release](http://www.iata.org/pressroom/pr/Pages/2016-12-08-01.aspx) talking about how they expect 2017 to be a 'strong year' for airlines with a profit margin of a whooping 4.1%... – reirab Apr 16 '17 at 07:55
  • @Akavall: Assume a flight with 200 seats at $300 per ticket, no refunds for no shows. If you sell 200 tickets and 190 show up, you made $60,000. If you sell 210 tickets and 200 show up, you made $63,000. If you sell 210 tickets and 202 show up, you made $62,400 from passengers and eight no-shows, so even if you pay $1,200 to the two bumped persons, you are still no worse off then by selling 200 tickets. It's not safe money, but it pays on average. There will be an optimum - selling 300 tickets and compensating 80 is not clever. Selling 210 tickets and sometimes compensating a few is. – gnasher729 Apr 17 '17 at 08:42
  • @mark "about 15% of tickets sold were neither used nor refunded" — any idea how many of these were return tickets or second leg of multi legged itinerary? – vartec Apr 24 '17 at 20:40
8

According to American Airlines Group Reports Fourth Quarter and Full Year Profit AA had 199M passenger enplaments resulting in 2676M Net income in the 12 months ending December 31, 2016. This means 13.47 USD or so per passenger.

Southwest in 2016 had $2.24 billion net income from enplaning 151,740,357 passengers. We are looking at 14.76 USD per passenger.

Of course the above figure presumes all income comes from passengers. There are cargo operations etc which further lower this figure. So below 10 USD is perhaps not true for the more profitable airlines but it's not too far from the truth either.

chx
  • 721
  • 4
  • 13
  • 2
    In addition to cargo, airlines also get [tons of revenue from their loyalty programs](https://thepointsguy.com/2017/03/do-loyalty-programs-make-money/) (by selling miles) and also make a lot from their co-branded credit cards. The latter is pretty much pure profit from the airline's perspective. In all likelihood, the profit on actual tickets is indeed below $10 across all of the major U.S. airlines. – reirab Apr 16 '17 at 07:36
2

False. According to The Economist (April 22, 2017 print edition, online article),

Airlines in North America posted a profit of $22.40 per passenger last year; in Europe the figure was $7.84.

Graph:

enter image description here

According to the above graph, the claim would've been true in 2012 and 2013. But not in the past few years.

Another source is the IATA's Dec 2016 forecast for 2017 per passenger profits:

  • North American carriers: $19.58.
  • European: $5.65.
  • Asia-Pacific: $4.44.
  • Middle-Eastern: $1.56.
  • Latin American: $0.76.
  • African: −$9.97 (that's a loss of $9.97).