Code Sharing


No, don't worry, this isn't another post about open source, but about the practice of "code sharing" between airlines, the topic of a recent New York Times op-ed by former American Airlines honcho Bob Crandall. The way it works, as I understand it, is that you book a flight with some airline, but some portion of the journey actually occurs on a plane run by a different airline. That flight will be co-branded (hence the "code sharing") by both—Lufthansa 210 and United 428, in the listings. Mr. Crandall offers some perplexing logic, according to which this is bad news for the U.S. airline industry.

The one intelligible argument we get is that this amounts to something approaching fraud: you think you're flying on Chic Air and get the equivalent of Budget Rent-a-Plane instead, or some such thing. But I don't see how there's a special code sharing problem here. If an airline promises certain amenities and then doesn't deliver them, that's a scam regardless of who's running the flight. That aside, I'm having trouble distinguishing this from the case in which, instead of "code sharing," the airline just leases a plane for a few months. In other words, Crandall seems to be arguing that it's bad to have this co-branding arrangement, but would be just dandy to buy the precise same plane and do the run yourself.

It sounds as though all code sharing does is eliminate the need to have redundant flights with half-empty planes (or numerous smaller planes), and I can't for the life of me figure out why that's bad. If there's a competitive problem created here, it's not that the government allows this, but that it picks and chooses which particular arrangements it's going to allow. The solution to that, though, would be more leeway for co-branding, not less.

Crandall points out that one effect of these alliances is to reduce the proportion of trans-Atlantic flights flown by American carriers. But… why is this a problem? If it's cheaper to outsource some leg of a flight to Lufthansa, or whomever, why exactly are we supposed to think there's some intrinsic virtue to having a plane owned by a domestic company do it? These arrangements are supposed to create less "competition," which I suppose is true in the sense that any time a firm outsources some aspect of its production to another company, they're not running their own (by definition competing) subsidiary. So if a car manufacturer buys radios from an electronics company, instead of making their own, that necessarily means less competition in stereo manufacturing. And anything short of yeoman farmer self-sufficiency limits agricultural competition, too… Anyway, the incentive to "do it yourself" when there's actually some prospect of being competitive (i.e. running the route at a lower cost) remains. So, am I missing something, or is this just warmed over mercantilism?

NEXT: Second World: Globalize Us, Please!

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. “Skeptics, yes,” said my friend the Gnome of Zurich. “We stand for disbelief. We are basically cynical about the ability of men to manage their affairs rationally for very long. Particularly politicians.”

    The Money Game – by Adam Smith

    Anyhoo, am I correct in thinking the country is on a trend of growing international enmity and protectionism? I’ve always heard the whole “buy american” line, as if there was just something wrong with supporting our brothers of differing nationalities (“Thanks, asshole, but I’ll just buy from the person who offers me the best deal”), but it sure seems louder. Then again, that could just be my perceptions – perhaps it’s just about the same as it always has been, but I hear about it more often now because I pay attention to it more.

    On the direct topic of the thread, it would seem as if the author of the article is against co-operation itself, of mutually beneficial arrangements between competitors. Well, I can see the problem if this turns into collusion…but is that it?

    Competition is a nice word sort of word to many people, but when divorced of pre-judgements and judging things by results, I fail to see the problem with sometimes-competitors cooperating.

    The only problem I can see is in companies claiming that something is “theirs”, when it isn’t really…but I fail to see how that is different than every other company on the planet, as no company outside Atlas Shrugged produces _everything_ it uses. Crest, for instance, sells toothpaste as if it was all it’s own, but obviously it buys a whole lot of stuff from other companies, possibly ranging from boxes to employees to raw materials to tubes to use of facilities. People simply have to adjust to the idea, or reject it entirely; so long as information is not hidden, such that interested parties may know of it and choose accordingly, I see no problem with any such arrangements.

    After all, there is a long history of companies buying products from other people and just sticking their own branding onto it; it’s an entire industry, for goodness sakes. Of course if one has the reputation for security or dependability of manufacture, one might do well to avoid behaving quite like that, as it could damage the brand, but I would think that such decisions are up to those with the proper right to make them – the companies.

    I didn’t know that the industry could be that fucked up by governmental regulation, that you actually needed special permission for this kind of thing…but perhaps I simply do not understand the system well enough – though, naturally, that wouldn’t neccessarily make me incorrect.

  2. It’s also interesting to consider the effects of this on the Fly American Act, which stipulates that flights paid for with federal money must be flown on US carriers, when available. Thankfully, code sharing agreements are ok, so long as the US carrier appears on the ticket or E-ticket documentation.

  3. “The main beneficiaries of these arrangements are the airlines, which can mislead passengers into believing that they will be flying a complete itinerary on one carrier, when in fact, they will be using several airlines.”

    The problem with this claim is that passengers do know when they will be flying mutliple airlines due to codeshare agreements. Look at any travel site such as or When one looks at the details of an itinerary that shows up, one may see that many a United flight from Philadelphia to Madrid is operated by U.S. Airways. When customers actually make their decisions to purchase, they have all necessary information to know what airline will actually be operating any segment of a flight.

  4. Everyone seems to be missing the issue here.
    Code-shares are substitutes for mergers that
    would not be allowed by the competition
    authorities. They reduce the number of
    competing airlines on particular routes.
    There is lots of empirical evidence to show
    that prices decline with the number of
    airlines flying a route (albeit non-linearly).
    On the other hand, as noted here, there are
    potential efficiency gains on the other side.
    Much of the cost of a flight is fixed, so
    having half-empty planes is a bad idea.
    In a Chicago world, the competition sorts
    this stuff out, hopefully based on economic
    theory and evidence. In a George Mason world,
    the competition authority does not exist and
    the firms just merge instead of code-sharing.
    In a world with no economists, you have
    Robert Crandall.


  5. Change “competition” to “competition authority”
    towards the end of the previous post.

    Goin’ too fast …


  6. Jeff says codeshares “reduce the number of
    competing airlines on particular routes.” But he doesn’t address my explanation about that codeshares increase the number of competing airlines selling tickets on the same plane. Prices differ!

    Jeff says “In a George Mason world, the competition authority does not exist and
    the firms just merge” but fails to recognize that if you dislike codeshares, you at least want to open up the domestic market to foreign competition — since KLM or Virgin or whomever could offer domestic US flights instead of slapping their code on someone else’s plane. Certainly that happens in a world without government restrictions on trade.

    Codeshares aren’t really the issue, either. The question of a particular airline code on a plane is only one piece of the competitive puzzle. More important is frequent flyer alliances in driving purchasing decisions. That’s why Sweden’s competition authority so dislikes loyalty programs. Since they give the consumer something for sticking with a carrier, they are therefore anti-competitive. Or so that European argument goes. Ahem.

  7. Code sharing is OK so long as air travel is truly fungible – if it makes no difference whatsoever what airline you fly and price and schedule are the only differentiators.

    However, that’s not true. Airlines advertise differences in seat size, meals differ, on-time arrival performance is a useful metric; heck, some airlines have prettier stews with shorter skirts.

    If you go to the trouble to establish a brand then your company had better deliver.

  8. anon @ 4:29-
    True, but the dual forces of market pressure to preserve a brand and recourse to suit in the case of genuine false advertising seem to cover that one well enough, whether it’s a matter of one’s own planes or “shared” ones.

  9. I wonder how passengers will take it once they figure out what’s going on? My girlfriend only flies Delta, because she’s convinced they’re safer and better somehow. She was terrified the time I made her fly Southwest. (which is why I had to spend 18 hours driving to Florida last trip we took)

  10. The Crandall piece contains numerous half-truths.

    He identifies 1983 as the start of INTERNATIONAL codesharing. But the practice of codesharing dates back to the 1930s when Western Airlines flew from Los Angeles to Salt Lake City, and United flew from Salt Lake City to the East Coast. Neither could fly coast to coast on their own because the government granted all route authorities back then. But Western, for its financial survival, was given permission to coordinate with United and they operated each others planes and shared crews.

    For 70 years codesharing has both been about the economics of leveraging large capital investments (planes) and about navigating regulatory waters. It still is! As Crandall observes, KLM wanted to offer lots of US destinations, and their partnership with Northwest allows them to. Crandall thinks this is disingenuous — but of course KLM can’t offer domestic US flights on its own because the US government will not permit it. So what is KLM to do?

    Plus, codesharing benefits consumers because you have more than one airline offering pricing on the same route. Prices are often different. You actually get competition not just through different planes but through different sellers of seats on the same plane!

    Another half-truth is the Pan Am example. Pan Am flew international routes because of its political muscle. They were content to allow other airlines to fly domestically and deliver customers to Pan Am for international flights. Permission from the US government for more airlines to fly internationally meant that Pan Am was dead without a domestic route network. So they tried codesharing, and they tried acquiring National Airlines but that merger failed.

    What Crandall doesn’t mention is that his former airline, American, is getting a bit left out in the cold as other carriers are leveraging international codeshares. Part of this is American’s fault, and part of this is the government’s. But his argument is rather convenient because it cuts against American’s competitors more than against AMR.

    American codeshares with British, but they aren’t permitted to codeshare across the Atlantic because of competition concerns about flights to Heathrow. BA would have to give up slots at LHR before the AA/BA combination can move forward. Meanwhile, United and Lufthansa have even entered into revenue sharing for their transatlantic flights, and the KLM-Northwest-Continental combo looks to get stronger with the Delta-NW-Continental combo, that may even bring Air France into the mix.

    Although KLM is insisting on Air France privatization before this all happens. So codesharing may lead to less government!

    But Crandall is clearly arguing from the self-interest of the company he ran for 13 years.


  11. Just to clarify, I agree that laws prohibiting
    foreign carriers from flying point to point in
    the US (or via hubs in other countries – though
    this is not so relevant in the US it is very
    relevant in Canada) are silly and should be
    gotten rid of. Everyone who has ever flown
    Air Canada – the world’s only bankrupt de
    facto monopoly – will agree.


  12. John Thacker: there’s an interesting public choice explanation for the government’s requirement that international travel with federal funds be on US carriers.

    There’s not much a constituency for USAID funding or in many cases State Department activities. International affairs and foreign aid just don’t have as big a domestic political constituency as many other interests.

    This act makes the domestic airline industry an important political constituency for more government funding in international affairs…

  13. So whom do we sue when trouble occurs (like a plane going down with my wife and kid on board, for example.) Do we sue Leg A or Leg B or both?

  14. Jeff – I agree. The domestic point-to-point airline business may be the only industry in America today where I have absolutely no alternative to the Unionized, American-made product. Foreign competition was a good thing for the electronics industry, it was a good thing for the automotive industry, and it would be a good thing for the airline industry.

  15. I think it’s worth mentioning here that not all codeshares are created equal.

    Some are well executed, with clear disclosure of the carrier actually operating the flight and any differences in product clearly identified (“xyz offering not available on flights operated by AirWhatever.”) Solid, harmonized procedures have been crafted and communicated to front line staff. As a result, they offer a seamless, hassle-free product to their customers.

    Others have poorly aligned schedules (marketing airline’s departure is 0920 while the operating airline’s departure is 0845), don’t offer through-check-in on connections to flights operated by the other carrier, don’t offer pre-assigned seats on the codeshared flights, or in some cases, can’t even conduct ticketing transactions (changes, re-issues, etc.) on behalf of the partner carrier at the codeshared destination. Arrangements like these often lead to lost reservations, second-class treatment of codeshare passengers, lost baggage (when the codeshare flight numbers or city codes aren’t correct on the bag tags), and a whole host of problems.

    Some of the critical points airlines should be watching are highlighted on a consulting firm that deals with these problems. Truly some airlines have the expertise and technical capability to offer a truly seamless product with their codeshare partners while others do not. Codeshares which aren’t well executed do little more than scare off repeat business for the airlines operating them.

  16. EMAIL:
    URL: http://www.TRY-PENIS-PILLS.NET

    DATE: 12/11/2003 04:04:06
    It’s not all lies – not all of it. That’s the age-old dilemma.

  17. EMAIL:
    DATE: 12/21/2003 04:39:42
    It is dangerous to confuse children with angels.

  18. EMAIL:
    DATE: 01/19/2004 03:04:21
    I like it very much, i should get one site like this too

  19. EMAIL:
    DATE: 05/19/2004 10:49:31
    I dont know what to say, but i likeed it.

Please to post comments

Comments are closed.