Airline pricing is always something that captivates and confuses people, and I love talking about it. I received this great question with a pricing example from a reader which gives me a good excuse to ramble about the dynamics at play. As you can guess from the dates in here, this was sent a couple months ago, but I’m just getting around to posting it now.
I read your site regularly and with great interest and would greatly appreciate you [sic] view on the following subject regarding hidden city tickets:
I’ll use the following example: Detroit to Cincinnati on Delta, may 3, one way, leaving at 7:10. Prices are from Delta’s own site at time I wrote this.
Nonstop Detroit to Cincinnati – DL4509. $358.
Hidden city: Detroit to Dallas: DL4509 to Cincinnati connect to DL4250 to Dallas $121.
DL4250 on its own Cincinnati to Dallas is $98.
Now I acknowledge that no one would book the connecting flight if they could, because Delta has a non stop to Dallas (DL3748) leaving at 7:10am for just $113.
So here is my question.
The two legs on the hidden city journey priced together is $121. The two legs priced separate ($358 + $98) is $456. The nonstop flight that the two compete with is $113. The non-stop I want is $358.
I think … they are grossly overcharging for the Detroit to Cincinnati flight. Way more then demand would indicate. This to me is taking advantage of having monopoly power. Let’s remember that airlines themselves are not really a free market. There are limits in airports (with public funds) and airspace. So traditional free market does not apply.
Delta says they can take me to Dallas on two flights for $121. If this is profitable, there is no need to upcharge the Detroit to Cincinnati flight. If it’s not profitable then why offer it at all? Why not prohibit the connecting flight so they have space to sell the $358 flight plus $98 flight?
If the $358 is really market price they should have no problem selling it.
Thank you for your time and opinion.
-Jeremy
Alright, there is a whole lot going on here, and I even edited the question, because it got even deeper into the weeds originally. The basic premise of airline pricing is 
straightforward to explain. The airlines try to maximize the amount of revenue they can generate on each route, from origin to destination, regardless of connecting point. This, of course, means that tickets are sold from origin to destination only, and you can’t use a partial ticket.
In general, airlines charge more for nonstops, because that’s one of the few things for which customers have historically been willing to pay a premium. From Detroit to Cincinnati, if you want to fly nonstop, you fly Delta and you pay a premium for it. If you’re willing to connect on another airline, you can pay less.
This is where hidden-city ticketing comes in. Detroit to Cincinnati is expensive, but Detroit to Cincinnati to Dallas is cheap, because there are a ton of airlines that fly that route with a connection. You may want to buy that cheap ticket to get to Cincinnati and just not board the second flight, but the airlines don’t allow that. If they did, it would blow up the entire fare structure, which is fairly fragile anyway.
This is where you probably start wondering… well if airlines can charge more for nonstops, then why is it so cheap to fly from Detroit or Cincinnati to Dallas on a nonstop? Well, this is where the competition starts to warp things. In the former, you have Spirit flying the route (as well as American and Southwest to Love Field). In the latter, you have Frontier (and American). So Delta no longer has the ability to set its prices. It becomes a price follower and has to match, at least on some flights. (Note that Jeremy points out that the flight he wants is significantly more at a different time of day.)
The reason for this is because, whether you want to believe it or not, a ton of people choose their flight based on price. And time and time again we’ve seen that when legacy airlines don’t match, the low cost carriers grab a toe-hold and grow from there as people flock to them. Now the legacies aren’t willing to give an inch.
This makes for a perverted fare structure where the old-school high fares exist until competition comes in to knock them down. It’s hit or miss in every market, but this isn’t a new phenomenon. Southwest started knocking down legacy fares 40 years ago. The practice has just expanded greatly since that time as many low cost carriers have continued to expand their reach.
And this is why there’s a new focus on the value of smaller connecting markets. I wrote about how United was really looking at domestic connections after its President Scott Kirby took his knowledge from American and its Charlotte hub. Small cities have far fewer passengers, but the fares are a lot higher because they’re too small for most low cost carriers to enter.
That means that these markets are the last bastion of the old fare structure. Detroit-Cincinnati is one of the dwindling number of larger markets that don’t have low cost carrier service. That time will come eventually, and we’ll be left with one fare structure for big markets and another for little ones. That will not only confuse people but it will further hurt small city demand.
So if you wonder why you see so many different kinds of fares in similar markets, it’s because there is no coherent pricing strategy in place. There are the remains of an old strategy, dented and dinged from years of the airlines losing pricing power to their low cost competitors.
I’ve had a lot of discussions with people about how different structures could work. I continue to be obsessed with the concert ticket model (I even did a podcast about it), but I’ve yet to figure out how to make it work when connections are in play. But some day, there is going to be a better solution. It’ll just require someone willing to take a risk. The problem is that when legacy carriers make a bold pricing move, they’re rarely rewarded for it.
32 comments on “How Do Airlines Come Up With These Ridiculous Fares? (Ask Cranky)”
Quoting your comments: “You may want to buy that cheap ticket to get to Cincinnati and just not board the second flight, but the airlines don’t allow that.”
I get that they won’t like it, but how can they not allow it? If I don’t have checked baggage, I get off the first flight and, instead of heading to the gate for the second flight I go out the door. I’ll be a no-show on the second flight, but do they have a recourse? Can they hit my credit card, flag my name in their reservation system for the next time I fly Delta, or what?
They can’t force you to board the second flight, but if they see you no-show repeatedly, they can suspend your frequent flier account or ban you from the airline.
Last year I wanted to book Bozeman to Denver. The fare was $403. Bozeman to Colorado Springs was $199, connecting in Denver. I booked the Colorado Springs ticket and had my wife pick me up in Denver. Obviously I only had carry-on luggage.
There were no consequences that I could tell.
jspencer – The airlines make the rules and they do have recourse if you break them. I received an email earlier this month from someone who got caught. I’ll just paste it here.
“I received a letter from United demanding I pay them over $5000 for booking hidden cites flights over the last 3 years ( 18 flights ). In the meantime they shut down my frequent flier account and told me in the letter that they reserve the right to sue me and keep me off future flights.”
also keep in mind, their responsibility is orgin-destination, not specific routing. So the other potential is if your flight is cancelled and they choose another way to get you there.
I.e on Delta, maybe they route you through a different hub, such as MSP vs. DTW. If you were hoping to get off in DTW, then you are out of luck. Best to book without FF# (though they can still track names) if you even attempt to do this.
oh, they can also cancel future legs, so if you book it round trip and get off after flight 1, flights 2-4 will cancel as with any no show
Back when CVG was a true Delta hub, many people in the Cincinnati area (especially in the wealthier surburbs north of Cincinnati which in terms of drive time are almost as close to DAY as they are to CVG) would drive to DAY, have an easier and quicker time going from car to gate, and then take a puddle jumper down to CVG for a connection. The additional drive time to the airport was minimal, if any, and the difference in fares from DAY to wherever (with a stop in CVG in between) compared to a nonstop from CVG to wherever was HUGE.
I’m sure similar things still occur today at airports pairs like, say, LAN vs DTW, or Chattanooga vs ATL.
I used to do this with Augusta, rather than Columbia for Delta. AGS was a 15 minute flight, but surprisingly the fares were significantly cheaper than flying out of Columbia, and much shorter of a drive than driving to Atlanta (at the time my work took me closer to AGS than Columbia, and so I would fly after work).
Thanks for the interesting post, Brett.
As for ending your journey in a connecting city (hidden city ticketing), you say “but the airlines don’t allow that.” I know its in the fine print of every ticket that you’re not supposed to do it, but I’m curious the extent to which they enforce these rules / punish violators. Obviously you couldn’t check bags and it’d have to be the last segment of your trip that you’re skipping out on (because the rest of the itinerary gets canceled). Beyond that, would an airline really turn down my money next time I want to fly?
Full disclosure: I’ve done this 2 or 3 times out of 100+ trips over the last 10 years (generally only when the savings is $100+). Never had any negative repercussions and I even got frequent flyer miles for the segments I flew (I figured withholding FF miles might be a fair punishment, but less and less so now that they’re constantly being devalued).
Thoughts?
“I’m curious the extent to which they enforce these rules / punish violators.” Should add to that sentence if/how airlines even find people who are doing this.
Well, it’s pretty easy for them to *find* people. They just look for people who boarded the first let but not the second leg.
The two main drawbacks are first, that it only works on one-way tickets. An airline will cancel the remainder of a round-trip if you do this. And second, you put your frequent flier account in jeopardy. Practically, if neither of these things applies, there’s not much an airline is likely to do. You are still in violation of your agreement with them, though, so you take a theoretical risk by doing this.
I agree. What I meant by “finding” them was more about the operations of it. Do they have a centralized process where data analysts search millions of itineraries each week to find these people? I doubt it.
Seems like leaving it up to the gate agents though would be inconsistent and error prone (not to mention they have more important things to do). Plus there could be perfectly legitimate reasons for not boarding a connecting flight (i.e. family emergency, last minute change of plans, etc.).
I guess what I’m saying is that if the Big 3 are gonna nickel and dime me for a window seat and a carry-on bag, I have no problem nickel and dining them when I can. Plus it seems like they don’t have a good way to stop it at the moment.
An army of analysts isn’t necessary. Algorithms can be designed to find these kinds of “people” with less than 15 minutes of code writing effort. For example, you can tell your computer systems to find all passengers who have been a no-show on a connecting flight, without airline-related misconnect issues like weather, etc., more than 3 times in their last 10 flights, or something. Done.
It’s not difficult to find the habitual hidden city ticketers. It’s how the airline goes after them that’s the issue.
American’s “Special Services” reps in suits met hidden city ticketers at DFW to personally escort them to the onward flight they didn’t want to take.
They may not come after you but if you use a third party they can sting the agent with an Agrncy Debit Memo.
British Airways (who have a huge issue with cheaper ex-EU business class tickets and HCT) slammed a £68,000 ($100,000 in pre-Brexit dollars) ADM on a travel agent (more than) suspected of encouraging hidden city ticket walk offs at London by booking airport change connections (arrive LHR, connect from LGW or LCY but just walk away) – http://www.flyertalk.com/forum/british-airways-executive-club/1731641-our-ex-eu-horror-story.html
How exactly can anyone “escort” you to an onward flight you didn’t want to take? That sounds like kidnapping to me. If you pick up your carry-on bag and walk out of the terminal, what are they going to do about it?
Couldn’t they threaten legal action if you don’t, since you’re breaking a contract?
a lot of carriers export data to a warehouse which is a large relational database used for all kinds of analytics and reports. Most likely, these are ad hoc reports, not a constant analysis. One airline I am quite familiar with can see vast amounts of data – for each PNR they can see routing, pricing, if boarded, purchase date and time, and changes, etc. For each check in (by PNR) they can see if you did kiosk, web, or mobile, if a kiosk the specific serial number, and get timestamps for how long you spent on each screen. All of this data might be spread in a few systems, it might get deleted or stored (and varying amount by time, i.e kiosk data may expire after a week). So to write a SQL query that asks for boarding flight 1 + no show flight 2 and filter out 2 or fewer times per year or something or filter out date spans when there was weather / IROP really isn’t that complex. It’s probably 1 person <1 week to formulate the plan, run the query, and filter/clean the results. But if they can find 100 people, each who saved $100 it probably is worth it in a pure cost-benefit.
Agreed – it’s not that hard to write a search query, can it, and repeat for reporting on a regular basis.
And even if they aren’t using that sort of information to enforce something against those folks, it’s more grist for the mill in terms of figuring out holes in their revenue management (pricing) strategy.
Hidden-city ticketing makes sense economically, but somewhat falls apart when looked closer.
Airlines can be assumed to be in Bertrand competition for coach customers, and flyers tend to book the cheapest ticket available. Under Bertrand competition, consumers only care about price; companies respond by setting prices below their competitors, who respond in turn until price=marginal cost.
While airlines themselves have large start-up cost, there aren’t really any start-up costs to fly between two cities that an airline has already leased gates in (there are opportunity costs to consider though). What this means is that another airline that already has gates in both cities, United for example, could fly between these two cities with a price of $350, steal Delta’s customers, and capture profits until Delta cuts prices in turn. This doesn’t really happen in the real world (the first barrier that comes to mind would be crew and plane positioning, but that’s easily fixed by routing planes in a circle on existing routes, such as ORD->DTW->CVG->ORD. If anything, so long as they can fill planes marginal costs are lower for United as the distance is shorter and direct passengers don’t tie up scarce ORD resources.
With the exception of Southwest and the ULCC’s, we don’t see airlines going after each other’s profits as assertively as would logically expect.
What I think you’re missing is that you sell individual tickets, but you fly larger planes, AND since most people fly round trip, you have to have flights in both directions.
Delta can look at DTW-CVG and know that it doesn’t have to fill 76 seats of people doing CVG-DTW, it’s got to fill 76 seats of people in Cincinnati either flying to Detroit or willing to connect there. United doesn’t have that option.
Circle routes can help, but imagine that the jet only has enough time to do one of those loops in a day (that’s not the case with this distance, but it is in some, but even if you do multiple loops, the issue’s the same if you assume that equal numbers of people want to fly in and out of a city on a given day). What happens to people that need to go from CVG-DTW? They have to connect. Now you’re selling a direct flight one way, but a less attractive connection on the way back, or you’ve got to have planes do the loop in both directions (either by doing a loop one way first, followed by a loop in the other direction, or doing it in reverse).
One of the hidden risks of hidden city ticketing is when flights are delayed or cancelled. If you buy the DTW/CVG/DFW ticket planning on getting of in CVG, what happens when the DTW/CVG is cancelled or delayed and they automatically re-route you either on the non-stop DTW/DFW flight or via MSP or ATL? This could technically happen to any hidden city ticket unless you’re flying out of a city that only has service to one Hub.
This is actually the best (and basically only) practical rationale I’ve heard against doing it.
What I have noticed is that the connecting verses non-stop fare difference is related to peak travel seasons. In the off-peak times, the airlines are pricing the tickets based on the competition for connecting fares between two markets. In the peak-travel times, the connecting tickets are basically priced as the sum of two non-stop tickets between the origin to the connecting airport and then from the connecting airport to the destination. During peak-travel times, airlines want your bottom in their seat for as little time as possible. During off-peak travel times, they are happy to send you all over the place before getting you to your destination. It makes sense, I’m just surprised how much the change is. The non-stop goes up a little during peak travel times while the connecting ticket skyrockets from being below the non-stop to priced well above the non-stop.
I guess it the same craziness as I found NYC to Pittsburgh
Ah yes, a day without fare a question is like a day without…well, whatever. Aren’t we about due for someone like UA to have a system meltdown and again start selling everything domestically for $0 + a security fee. I used my $0 to go RT DC to LAS..I just felt too morally honest to try to find a seat to Hawaii, but…! Same with the hidden city mess. I await the day when some computer goes haywire and decides to refund overcharges for some couple hours for every PAX who couldn’t have used a hidden city, but didn’t. Again, I’ll feel guilty but just take the money and keep complaining about the complexity of our airlines’ fare structure…an argument, Cranky, will just go nuts over!
“who COULD have used,” I mean. I return, arguing with myself, as whether or not I should call those things “Mistake fares,” or as DOT seems to want to, “Mistaken fares.” It’s the heat, I guess!
My friend is flying DFW to YOW. It costs much more to Ottawa in money or miles than to go to Europe. Crazy, what?
Regarding that hidden-city thing: Airlines are not allowed to forbid that under German law. Basically a consumer rights issue. But good luck with that in the US ;)
Part of this reason airlines charge more for nonstops is because they want to protect connecting traffic. Take DFW-LAX. AA, DL, and UA all have nonstops in the market, each for different reasons. Airlines don’t want nonstop traffic, they want the connecting traffic. DL and UA use DFW-LAX traffic to feed their connections at LAX. AA has DFW-LAX originally for DFW connections, but now with their build up of LAX, AA wants both connecting traffic at DFW and LAX. Even Southwest, with DAL, they have build up connections at both ends DAL and LAX. The connecting traffic, over the long term, is more value to the airline because it expands the number of city pairs to offer increasing market share. This allows the ULCCs, like Virgin (DAL-LAX, and I know not a true ULCC, but niche carrier) and Spirit (DFW-LAX) to pick off “local” DAL/DFW-LAX traffic virtually unnoticed.
The legacies will match some low fares, but usually only at the times that match the flights of the ULCC, if they do at all. Basically, the ULCCs ‘bottom feed’ local traffic from the legacies. And the legacies usually let the ULCCs have their “token” local bottom feeding traffic to keep yield high.
Nice explanation. When I lived in DEN I wondered why fares were so high, except to LAS. They wanted connections not DEN residents.
That’s the disadvantage of living in a hub city like you and I do (I live in DFW). I imagine Southwest has built up a connecting network at DEN.
I used to fly Frontier all the time, connecting via DEN. However, now that they are a ULCC, it appears they given up on connections and are going for local traffic. As you can see, UA and WN will pretty much let F9 and NK have all the local traffic, especially since both carriers generally don’t operate more than 1 or 2 flights per city.
UA and WN have to ask, do they want the $98OW leisure traffic DEN-LAX, or the $800RT businessman from MKE-DEN-LAX, which pro-rated, would be about $200 each segment. They will match the $98, but limit the number of seats to protect the higher connecting revenue.
And to add a further twist, there maybe times the MKE-DEN-LAX fare would be low enough (say $350RT), the pro-rated DEN-LAX segment maybe lower than the leisure DEN-LAX fare. In this case, the pro-rated segment would by $87.50. UA may open that fare up over a DEN-LAX $98OW fare because they know MKE-LAX tends to have a majority of business travellers. And while this fare isn’t more, in the long run, this is a better decision for the airline.
Hello Cranky and Jeremy,
You dismissed, Cranky, the option of Jeremy buying the Detroit – Cincinnati – Dallas ticket and just ditching the Cincinnati – Dallas sector. Why dismiss that? I have done that same thing quite a number of times without problem. However it has substantial limitations. You cannot do it as part of a reservation as other travel, otherwise the airline will cancel all the remaining sectors after the one for which you are a no-show. Secondly, it does not work if you must use checked luggage. They will require to check it thru to your supposed end destination, in this example Dallas, not part-way to Cincinnati. The third problem is that when you later want to fly from Cincinnati back to Detroit, you will have to be creative all over again. You will need to look for a destination beyond Detroit that you can connect to via Detroit, and repeat the same process.
It is all do-able if you can travel with carry-on luggage only and don’t mind not using your frequent flyer number when making your bookings. It is not as if they can do anything about your out playing them using their own crooked fare structures to your advantage. If ever questioned, just say that when you got to Cincinnati, that you felt ill, and that later when you were well enough to travel again, that you had missed your meeting in Dallas and therefore flew home to Detroit.
SSmith3104
_____