Delta And United Go To War Over Tokyo Haneda Airport

There are a limited number of takeoff and landing slots at Tokyo’s convenient Haneda airport that have been made available to U.S. airlines. These were agreed to between the governments of Japan and the United States, and so the U.S. Department of Transportation had to decide which airlines to let use them.

  • The standard procedure is that airlines submit proposals to the Department of Transportation
  • DOT makes a judgment about which proposals are best for consumers
  • If an airline doesn’t want to fly the route anymore, they can stop doing so and return the slots to DOT – who then makes a judgment about what’s best for customers

American has a joint venture partner in Japan, Japan Airlines. United has a joint venture partner in Japan, ANA. Delta doesn’t have a Japanese airline partner, and DOT gave Delta the most Haneda slots.

These airlines are now allowed to serve Tokyo Haneda airport from:

  • American Airlines: Dallas-Fort Worth; Los Angeles
  • Delta Air Lines: Seattle; Detroit; Atlanta; Portland; Honolulu; Los Angeles; Minneapolis
  • Hawaiian Airlines: Honolulu
  • United Airlines: Newark; Chicago O’Hare; Washington-Dulles; Los Angeles; San Francisco

Delta got most of what it wanted, though it didn’t actually grow service in Tokyo. They pulled down Tokyo Narita flying and stopped their connecting flights into Asia. They connect passengers over their joint venture partner Korean Air’s hub in Seoul instead.

They no longer want to fly Portland – Tokyo Haneda, and they’ve asked DOT to allow them to fly something else with those slots instead. What they’re requesting isn’t just flexibility for them, but for all airlines. American Airlines would like to move LA flying to Dallas, and Hawaiian Airlines wants flexibility too, so they’re supporting Delta’s position.

United is opposed. And there’s a great battle of rhetoric in DOT submissions going on between the two airlines. Here’s United’s objection to Delta’s plan, and Delta’s reply. (HT: @crankyflier)

Delta lays out how United’s response is self-serving. And of course it is! Just like Delta’s request is self-serving. The question isn’t the motives of each airline, it’s how should DOT handle it when an airline no longer wants to serve the route that was awarded to them?

United’s Answer has it backward. It is United who is prioritizing profits over consumers and communities by objecting to a proposed solution that would afford all U.S.-Haneda slot holders – not just Delta – the ability to offer enhanced and more convenient flying options between the United States and Tokyo’s prized airport. It is United who is prioritizing its parochial interests over competition by suggesting that the Department should strip Delta of two Haneda slot pairs and redistribute them to United and ANA – the ATI-enabled transpacific joint venture that has the largest slot portfolio at Haneda and enjoys gateway flexibility on the 10 daily Haneda-U.S. flights operated by ANA8 – so that they can serve Haneda from two additional U.S. points, Houston and Guam. It is United’s opposition that is self-serving, as demonstrated by the fact that United is the only U.S.-flag Haneda competitor of the four who opposes Delta’s request.

The Atlanta-based carrier notes that of those airlines which have responded, a majority want the flexibility that Delta is after. But this misses the point: the goal isn’t to do what the airlines want (an airline subsidy). Delta never once addresses why the standard for what to do with these limited resources should no longer be judged on the basis of consumer benefit. Why should an airline’s parochial interest trump consumer interest in the judgment of the government?

Ultimately Delta’s position that a DOT process, where government bureaucrats decide who flies where, isn’t a market-based approach is correct. But a process where airlines ask DOT for slots based on a commitment of specific service, gain a government handout, and turn that into a property right that they can do with as they wish isn’t market-based either.

Maybe consumers are best off with Delta having the most slots, and doing with them as they please, but that’s precisely the argument to make – how this benefits consumers. Under current rules, the slots should be returned to DOT and DOT should then consider proposals – including Delta’s. Delta wants to skip this step.

Leaving what Haneda routes make the most sense up to airlines when the number of routes are limited in law doesn’t really work on its own.

  • What’s the justification for handing them out to a specific airline over another?

  • At most DOT could take the total available slots and divide them equally among those airlines that want them, on a use it or lose it basis. But that method, which is far more market-based, would mean Delta loses out since the previous government method benefited Delta the most (with the most slots).

Delta wants airlines to be able to ‘tailor capacity to demand’ and fly where they wish, arguing that is necessarily doing best for consumers. But it pre-supposes a world where Delta gets to keep the largest number of slots awarded to them on the basis of commitments for where they’d fly.

So DOT should have a new proceeding for any slot that an airline wishes to convert from a route-specific award to a flexible one, considering consumer benefit (including consideration of the question of whether Delta needs more slots and than other airlines to balance out the strength of the joint ventures of competitors) in order to determine which airline to award it to – and then allow the airline to decide where to fly.

About Gary Leff

Gary Leff is one of the foremost experts in the field of miles, points, and frequent business travel - a topic he has covered since 2002. Co-founder of frequent flyer community, emcee of the Freddie Awards, and named one of the "World's Top Travel Experts" by Conde' Nast Traveler (2010-Present) Gary has been a guest on most major news media, profiled in several top print publications, and published broadly on the topic of consumer loyalty. More About Gary »

More articles by Gary Leff »


  1. I’d love to see a New York-Tokyo-Singapore flight on Singapore Airlines or Delta. Or a Hawaiian flight from Tokyo to San Francisco or Las Vegas connecting to JetBlue.

  2. Doesn’t HA only support this not because they want flexibility for themselves but because they know DL will move their own HNL slot elsewhere, decreasing competition for their own HNL-HND flight?

  3. The public goal should be to move the most possible people for the given number of slots. Every year, the HND route/slot that moves the fewest people should get pulled for new proposals and re-allocation. The incentive for each airline is to put the largest planes with the most capacity and the lowest ticket prices to maximize the number of people they move on the route and don’t get it taken away – exactly the public’s interest.

  4. Deltas flaw in their argument is that they do not say what their relief specifically will be. What routes do they want to try instead? Will more consumers benefit?

    If they say they can provide more connections options (say via JFK)that too would strengthen their argument. United specifically outlined their desires, why does delta not?

  5. This has been well-discussed on other sites but the basic issue is that the Japanese airlines from the beginning have been able to use their HND frequencies where they want.

    AA and UA which have joint ventures with JL and NH resp. gain gateway flexibility.

    Although the US has Open Skies with Japan, HND is still a limited access market which is frequency limited, unlike nearly every other Open Skies agreement with the US.

    The DOT chose to grant HND frequencies on a route allocation basis and to not grant gateway flexibility.

    US and foreign airlines are free to add and drop routes in Open Skies markets so arguments that they aren’t in the best interests of consumers is counter to what the US has done with more than 100 other countries.

    Delta is right that United is trying to protect its dominant market position and the 3 other US airlines that serve Tokyo want that market flexibility.

    Delta clearly wants to start at least JFK-HND service which is a huge shot at UA’s EWR-East Asia network since DL is also likely to start JFK-Seoul/ICN. Given that UA cannot fly EWR-China or HKG service, DL could end up w/ a larger NYC to E. Asia network than United. Same is true for DTW and MSP for DL vs. ORD for UA.

    The DOT will rule but chances are high that they will grant DL’s request

  6. I blame Delta for being so shortsighted and asking for a *Portland* gateway.

  7. Consumer benefit should be the key here. Take underutilized slots back and redistribute.

  8. Could DOT make a split decision, allowing DL and the others to move one of the two slots?

    Which one would they move? Would the other just be flown at the bare minimum required, or would that get their wrist slapped again?

  9. @Tim, why do you set so many qualifiers to create circumstance where you can say Delta is the best. No airline that truly is the best has to twist themselves in knots to put themselves on top.

    Why not include India, Israel, Jordan, Dubai when factoring routes in? Why not mention that UA has two flights to east Asia, which would match watch DL might have one day?

    If NYC is so strong, why did DL previously cancel the route and not even bother applying for it back in 2019? Not even as their last choice.

    I don’t agree with your assertion, but you’ve said split hubs don’t work. You were talking about a split HND/NRT hub without realizing you were also criticizing DL’s split hub in NYC. With the split hub, DL will have a harder time feeding connection to the JFK flight on top of the difficulty with downline connections in HND.

    UA has a single NYC hub to feed connections to the NRT and HND flights, along with connection opportunities downline at both airports. This is why UA has enjoyed the type of success from NYC to Asia that DL would love to see.

    Of course, as you’ve pointed out, JVs make UA and NH one airline on paper, so UA makes money off of their JFK flights too.

  10. @Mark

    Slow your roll. The only measure of best should be profitability. DL was the most profitable over the Pacific AND Latin America in 2022, and across the entire network for all US airlines in the second half of 2022. DLs top 10 performing Asia routes from Seattle were Tokyo, Seoul, and Shanghai. The fact that DL entered COVID in a position of strength, and is now is even in a stronger position relative to its peers shows how unmatched DL’s strategy is. Vasu once said DL was yesterday’s genius, but that can’t be farther from the truth. Investors know it too, which is why DL has the most recovered market cap among the US majors relative to the end of 2019.

    DL has stated offline in some discussions that it wants to fly to Hong Kong but not from Seattle or LA where it was previously unsuccessful. As the saying goes, if you can’t make Asia work from the west coast, then pivot to the Midwest or East Coast gateways and then up-gauge capacity. If you go all in, your competitors have no choice but to fold. So having A350-1000 flying from the east coast where there previously was no existing service is a winning strategy that neither AA or UA will successfully replicate yet alone attempt.

  11. Mark. Unfortunately you are wrong. NY HND has been tried by US carriers many times. See AA’s failed route back in 2016 time frame. As Tim pointed out correctly, the GOJ puts very tight slot restrictions on all non-Japanese access to HND. Simply put, the take off and arrival times for JFK HND were horrible (lose money). To say that any US airline had/has flexibility into and out of HND is not correct. Imagine you own a coffee shop…in this case the only time the GOJ says you can be open is 6pm til midnight.

  12. AA and only AA has flown JFK-HND and then ONLY with nighttime slots which have been proven to not work from anywhere in the continental US.
    UA received none of them because they were too large so is happy to criticize any of the other 3 airlines that had problems w/ the nighttime HND flights.

    DL last flew NARITA to JFK as part of the DL NRT hub which did had been pulled down before covid and before the latest round of HND routes were granted.

    The finances don’t lie. UA does not generate much more revenue than DL on its international network and certainly not more profits over a number of years. UA burns a whole lot more jet fuel and pays more for it in order to add a few more dots to its route map. And UA consistently generates profits lower than DL even for transportation services – before DL’s industry leading revenue from its loyalty program, Tech Ops and the refinery.

    LGA is a DOMESTIC airport. LGA and JFK are not split hubs but complimentary hubs. ORY/CDG, LHR/LGW and NRT/HND are all international hubs.

    And, again, it is clear that UA wants to prevent Delta from flying – FOR THE FIRST TIME – JFK-HND. There is a very real possibility that with the right planes and being able to fly to Tokyo’s preferred airport and Seoul, DL’s JV hub, as well as anything else on top of that, Delta will overtake United as the primary US carrier from NYC to E. Asia.

    DL is simply asking for itself AND EVERY OTHER US airline what NH and JL already have – gateway flexibility for its slots. UA has every reason as the largest carrier through its JV to handicap anyone that might compete on as close to as equal basis

  13. @Mitt Nud: It’s sad that Delta pulled out of Hong Kong long before Hong Kong’s collapse, the closure of Delta’s Narita hub, and the pandemic. However, I don’t see any real demand for Hong Kong besides existing service from other airlines. Hong Kong is not the Hong Kong of 20 or 30 years ago, let alone 5 years ago. Manila, Singapore, Dubai, Kuala Lumpur — those are the places Delta should be expanding. Delta would be better returning to Taiwan before Hong Kong.

  14. FNT,
    it is clear from statements by AA and UA that they lost money on HKG and mainland China flights which DL didn’t operate.
    DL said the economics of HKG were deteriorating when it pulled out of HKG – from SEA.
    It is more telling that AA is not returning to HKG even though it has an alliance partner there.
    Maybe JFK-HKG will work and maybe it won’t – but I don’t think it is on the short list of new routes for DL. JFK-HND and -ICN certainly are.

    The Japanese government determined the future of DL’s NRT hub. The Japanese carriers were always humiliated by it and the Japanese government communicated its intention to turn HND into the primary Tokyo airport for longhaul premium carrier flights while NRT would become a low cost airport. The Japanese government’s decision to not allow DL to operate any beyond Tokyo flights from HND sealed the fate of the NW/DL Tokyo hub.

    The latest versions of the A350-900 and A350-1000 could easily penetrate a number of NYC and East coast to Asia routes in ways that UA’s fleet cannot do. What DL chooses to do remains to be seen but, because of the Russian airspace restrictions, DL has the potential to displace UA as the largest carrier from NYC to E. Asia.

  15. @FNT

    There will be a bit of lead time before the A350-1000 comes in service. However by that time, the 777-300ER will likely all be retired, and HKG will certainly be back on the table.

    The A350 has a proven track record of serving the Northeast-East Asia market. It has enabled a market advantage for carriers serving those routes such as NH, JL, KE, MU, and CA. The early bird gets the worm and when the early bird eats, it feasts.

    Had UA chosen to own up and take delivery of the A350 rather than procrastinating, then it would have easily been the first US carrier to restart service to markets impacted by Russian airspace restrictions such as India.

    UA may not realize it now, but in a few years it will show that having the 787-9 establish ULH routes so much earlier than the competition was a catastrophic and strategic failure. This is a marathon not a re-election, incumbents have no advantage.

  16. I’m disappointed that Delta has cut their direct service from Asia into Portland (Oregon). It leaves those who want to go there having to transfer via Seattle or SF. The immigration transfer time in Seattle was terrible the last time I went through there because they had too many people going through with such a small capacity. I have heard it has gotten better now. Still going through Seattle I would figure you’d need a minimum of 2 hours to transfer.

Comments are closed.