American Loves a Fortress Network, So Why Is It Growing New York City?


When I wrote recently about American’s efforts to create a fortress network, there was one obvious outlier. While American has tried to grow its position in the places where it is most dominant, the airline has behaved in the exact opposite way in the Northeast US, especially in New York.

If we look in the Northeast, American’s strong point was the Philadelphia hub it inherited in the US Airways merger. Pre-merger American’s strongest presence in the region was in New York, but that was nothing to write home about. As you can see below, American’s share had been on the decline in New York for many years.

American NYC Airports Departing Seat Share

Includes America West and US Airways before the merger… Data via Cirium

Looking to stop the bleeding, American Chief Commercial Officer Vasu Raja and his team hatched the plan to create a partially-immunized collaboration with JetBlue called the Northeast Alliance (NEA). This would allow the airlines to work together in order to try to compete with United’s Newark hub as well as Delta’s LaGuardia/JFK split hub operation. You can see on the above chart exactly when that happened post-pandemic.

Seats jumped. The first big driver in American’s New York offerings was the introduction of several new long-haul widebody flights that added a lot of seats. (This, the airline was able to do by using some of JetBlue’s slots at the right times.) The second big driver was American’s decision to upgauge all those 50-seaters buzzing around with 2-cabin 76-seat regionals. Altogether, those moves added up to growth.

At the same time, JetBlue took over some of American’s LaGuardia slots and both airlines launched service to a variety of new destinations. This should be considered a good news story for travelers. It meant more flights on more airlines to more places, and that is what people want to see. Yes, we are still waiting for the lawsuit to get resolved in which the Department of Justice (DOJ) challenged the deal as anti-competitive, but I will be shocked if that goes DOJ’s way.

The reality is that all signs point to this being a winner for travelers in New York, but what about the airlines? More specifically, is it a good deal for American? It’s hard to make that case.

To tell this story, we have to think about Philadelphia as well. Philly remains American’s only true Northeast hub that flexes up in the summer with significant Transatlantic flying. Philly may not be New York in terms of market size and all that, but it can provide American with one thing New York does not… a captive audience.

American Philadelphia Departing Seat Share

Includes America West and US Airways before the merger… Data via Cirium

At the time of the merger, Philly was owned by the combined American/US Airways with a seat share of over 75 percent. The airline has seen that dominant position erode ever since with the most recent months hovering just over 60 percent.

Clearly some of this is due to the dynamics of the Transatlantic market — especially how long it took for borders to open compared to more domestic and Latin-focused hubs — as well as the widebody shortage that American has faced due to delayed deliveries. But make no mistake, a big part of this is also the airline’s focus on building New York which takes resources away that could have gone toward Philly.

Most notably, New York has taken a number of those 76-seat regional jets that are so hard to staff with pilots these days. It’s no secret that American’s regionals have not been able to live up to their promises due to the pilot shortage, so every decision to put airplanes in New York is a decision to take flying away from Philly and other hubs.

Of course, if New York is outperforming, this is a smart move. The thing is… it’s not. It’s performing quite poorly, at least on the short-haul end of the spectrum.

To start, I decided to focus on LaGuardia domestic flying under 1,000 miles since in Q3 2022 alone, that was good for 75 percent of American’s LaGuardia departures. It also makes it easy to compare to Philly which saw 71 percent of departures in that range. These are the bread-and-butter markets.

I took a look at Cirium data and started with load factors for the first three quarters of 2022.

American Load Factor on Routes < 1,000 Miles

Data via Cirium

Those are a lot of empty seats on those airplanes in New York. I also pulled regional-only flying which makes New York look even worse. Remember, there were a lot of upgauged airplanes from 50 to 76-seats at LaGuardia, but it doesn’t seem like those seats were needed at all.

The argument for the 76-seater, however, isn’t about the number of seats but the quality of them. They have extra legroom and First Class, and that’s — as the narrative goes — what will make business travelers start flying American. So, I took a look at the local fare on nonstop flights only (still under 1,000 miles).

American Average Fare on Routes < 1,000 Miles

Data via Cirium

The fare out of Philly is significantly higher across the board. Now, there is the issue here that Philly is a smaller local market with a real hub structure and more of the onboards are connections which will presumably generate less revenue. On short routes, Philly is about half local while LaGuardia is closer to two-thirds. So, let’s look at that.

In Q3 2022, the peak quarter of last year, Philly flights under 1,000 miles had stage-length adjusted passenger revenue per available seat mile (unit revenue) of 18.1 cents. LaGuardia? That was sitting down at 12.7 cents.

The PRASM number is pretty damning, but I get it. It’s not easy to determine how exactly to allocate connecting fares on to a single segment. So instead of looking at the number itself, let’s just compare across American’s hubs using stage length adjusment (SLA). Cirium is using the same allocation method throughout, so the comparison should at least tell us something directionally.

American Q3 2022 SLA PRASM by Hub on Routes < 1,000 Miles

Data via Cirium

The story is the same. New York sucks, at least on short-haul.

I also took a look at the opposite end of the spectrum showing JFK vs Philly to Europe, and it is more of a mixed bag. On average local fare, JFK lagged behind Philly throughout the year. Using Cirium’s ARC/BSP data, Q2 saw an average fare of $574 from JFK and $640 from Philly. Q3 saw $599 from JFK and $639 from Philly. But local fare isn’t everything.

If we look at peak Q3, JFK had an 86 percent load while Philly was slightly behind at 83 percent, so there was much more parity than in the short-haul domestic flying. Also, remember that Philly is more focused on connecting traffic than JFK. JFK’s Europe service generates more revenue on a segment because of that shift. Then again, it costs a fair bit more to operate from JFK, so that is a sizeable offset.

But does a decent Transatlantic performance justify the whole operation? With in-demand 76-seat aircraft going out with tons of empty seats and fares in the toilet on short-haul, it’s a tough decision to allocate all those shells. When asked, American provided this statement:

American Airlines is proud to call New York and Philadelphia hubs. As we navigate the current demand and supportability environment, we are crafting our network to fly to the destinations our customers want to visit most. Propelled by the Northeast Alliance with JetBlue, American has withdrawn all single-class regional jet operations and now offers premium cabins on all departures from New York.  This summer, American will increase domestic capacity in Philadelphia by over 9% year-over-year, including new service for 2023 to three cities: Charlottesville, VA; Panama City, FL; and Portland, Oregon. Amid the regional pilot shortage, American has taken bold steps to ensure we’re able to offer a reliable schedule for our customers, including by introducing motor coach service with Landline and pay increases for pilots at our wholly owned regional carriers. We continuously evaluate our network and will adjust our capacity based on demand and supportability

Over the last few years, American has really tried to lean on partners in places like LA and New York, but there is the argument to be made that it’s not even worth it because of the skin American has to put in the game to make the partnerships work. That has not stopped American from cutting back in Chicago and LA, but in New York, it hasn’t been willing to make a move.

Part of this is undoubtedly due to the fact that New York is slot-controlled, and if you give up your slots, you won’t get them back. Slots make airlines do crazy things. But in a world where American is developing its fortress network, this focus on New York is an outlier. At the very least, this whole plan deserves to be revisited.

Get Cranky in Your Inbox!

The airline industry moves fast. Sign up and get every Cranky post in your inbox for free.

66 comments on “American Loves a Fortress Network, So Why Is It Growing New York City?

  1. @cranky is it to early to do a deal dive into the performance of the NEA? It seams to me that with all the slots JetBlue was giving access in JFK and LGA that they are the one who truly benefited from it. You could almost call the AA long haul flight portion of it as regional (far away regions) feed for JetBlue’s hub.

    1. American Loves a Fortress Network, So Why Is It Growing New York City?

      I think the answer lies within the NEA in conjunction with the merger between Spirit & JetBlue. American sees an opportunity as the DOJ’s case is weak on both fronts & is likely to lose barring a settlement. Keep in mind JetBlue was bringing a lot to the table to get the Spirit merger completed & American is recognizing this & is moving accordingly.

    2. Rob – It’s not too early and this is one part of that. But yes, there’s obviously more to explore over time.

  2. AA should think of NYC and LAX as either “destination hubs” or One World Hubs. LAX and NYC are horrible connecting hubs (even for DL and UA) but I would focus on them as destination or international/Oneworld (same for DCA) while PHL(EastCoast and Atlantic); CLT (Southern); MIA (Latin America); Chicago (Midwest); PHX (West Coast); Boston (yes it should be a hub, AA’s operation is bigger then DL – Northeast). DFW can be the “Gobal” Hub and SEA / SFO as One World Hubs with AK and others.

    I would rather have a smaller profitable operation in NYC and LAX and focus on major cities in/out of these two rather then trying to compete as a hub.

  3. Great analysis, CF.
    One addition fact for JFK is that AA and BA have anti trust immunity JFK-LHR, where BA is clearly the 800 lb gorilla. Connections to BA through their LHR fortress greatly benefits AA’s transatlantic performance.

  4. Is there a world in which AA should look at NYC (all airports) as merely a destination and beef up PHL (and other hubs) instead?

    1. Does AA “owe” it to oneworld to make sure passengers can connect from LHR on British Airways to Cleveland or Richmond or any of the other places they are flying those half full RJs? Or could they essentially say “too bad, so sad” by removing many of those connections?

    2. If so, the 76 seaters could be sent back to PHL, Boston or wherever they could fetch a better load.

    3. Seems like it wouldn’t take too much to ramp Philly back into a fortress hub and push fares and PRASM even higher.

    4. Hadn’t thought about BOS before sunviking’s comment but, if AA didn’t need to focus on JFK as a half-assed connecting hub, they really don’t need the NEA which could lead them to trying to push ahead in BOS because they wouldn’t care about ticking off B6 anymore. BOS appears to be a pretty tight 3 horse race as opposed to NYC where they are miles (furlongs?) behind.

    5. For all the conversation and litigation it has spurred, the NEA is looking like a half measure at best. And you Breaking Bad fans know how Mike Ehrmantrout felt about half measures! Question is whether Vasu Raja wants to go full measure by essentially telling OneWorld and JetBlue to pound sand.

    1. 1) No, AA doesn’t ‘owe’ connections in JFK. In fact, since the fare for LHR-CLE is going to be much the same whether connecting through JFK or PHL (subject to availability), it makes more sense to route the conx through PHL than JFK, and save the JFK seat for someone who actually wants NYC (and is willing to pay for it). AA/BA share revenue and data on the routes anyway, due to the JBA, so they know how to incentivize/dis-incentivize specific routings if they want.

      1. Makes perfect sense so to follow your logic a large reduction of flights at JFK would not be seen as a negative for 1W as long as they can flow connecting traffic thru Philly, Charlotte, etc.

  5. I’m not going to suggest it’s the right or wrong thing to do, but it’s important to note that the B6 and AS alliances have one thing in common: they impact Delta the most and put them on the defense in competitive markets Delta thought they had won or come close to winning since their competition struggled to provide international competition, were primarily domestic, and on the backfoot with corporate customer contracts.

    Pre pandemic, Delta was on the offense re: American. Delta took Latam, declared Miami a mini hub, stated intention to make austin a focus city, and was taking frequent fliers in historical major aa markets like Boston, austin, RDU, NYC (though, as noted…. That had been going on a while in NYC).

    AA’s moves, while they came at the expense of places like PHL, are low cost ways for AA to force delta to prioritize resources away from AA high yield markets. The moves put delta on the defense and allowed aa to worry less about losing valuable share in key markets like austin, Miami, RDU, etc, and forced Delta to prioritize its equally-limited resources in places like SEA, Boston, and NYC to compete against revitalized competitors that now had an international option to sell to customers via AA and a substantially increased network via aa to sell to corporate customers. Arguably, aa/b6 and aa/as have a total overall better network in terms of dots served in total, not necessarily nonstops, to sell in those key markets than delta.

    Delta has largely given up on a true focus city in austin (upgauging to an a321 to Atlanta just isn’t indicative of a focus city by delta’s own standards), largely retreated from a the pre pandemic size of the focus city in RDU, the mini hub in Miami is a shell, if that, of what delta stated in their LATAM announcement, and Atlanta didn’t grow to pre pandemic capacity levels anywhere near as quickly as dfw and CLT giving aa the ability to take valuable market share in SE markets from delta.
    In LAX, while delta has grown to the largest metal operator there, the AAdvantage program is, without doubt, the most valuable program in LAX and SoCal due to the AA domestic partnerships while aa has been able to prioritize the pre pandemic resources at lax elsewhere in their network without sacrificing network breadth to their loyalty base.

    While PHL has suffered and a somewhat myopic view of those two stations in isolation may suggest an incorrect decision, my personal opinion is that the AA network has benefited overall from the NYC small prioritizations relative to PHL and AA successfully stopped delta’s offensive assault against much of the aa network pre pandemic. Promises were made about the NEA that force prioritization in NYC and those cuts had to be made somewhere in a resource constrained environment.
    Nobody likes the Eagles anyway ;)

    1. Delta has not given up on AUS being a focus city. More cities have been added and additional international partners have begun to fly there

      1. Which cities have been added by delta, specifically, since the pandemic? CVG and RDU aren’t new adds. CVG is going sub daily soon. Increased frequency to hubs from austin is indicative of a good spoke, not a focus city.

        Lufthansa flies to Austin too, Copa is a new entrant (both Star). that KL and VS fly there has little to do with delta and a focus city.

        1. DAL and LGA comes to mind. Plus people in the Sky Club have said that AF, LATAM and Korean are planning on adding a flight to Austin.

          1. I look forward to your trip review on Delta flying austin-LaGuardia and Austin-Love field.
            Neither route exists. LGA-AUS is outside the LGA perimeter while Delta has infinitely better uses for their one love field gate than repeating VX’s mistake on the route.
            Per sky club agent chatter… really?

            There are simply zero new Delta routes to suggest they still view Austin as anything more than a profitable spoke to their hubs.

            1. The international routes were from the Sky Club. The other routes are from my neighbor who is a reserve pilot for Delta. Who crowned you the route expert?

            2. I’m not a route expert, just noting and speaking in reality and actual routes. The new route examples you cited don’t exist and, in one case, can’t (except Saturday only) exist for the lga-Aus example.

              Citing sky club agent gossip just isn’t a basis for reality. They’re usually the last people to know about new routes.

              You responded to my initial comment saying Delta has added new destinations since AA’s AS and B6 alliances that would suggest Delta still treats Austin as a focus city as opposed to a (very) profitable spoke. They simply have not.

    2. Those are very interesting points. I agree with what you are saying in that I think that AA took a pure game theory approach to it’s competitive situation and eschewed market level profitability to determine its strategy.

      However, this game theory approach has produced a company with low-single digit profit margins and the worst profit margin in the US airline industry. That is quite remarkable given AA’s domestic and short-haul international focus, where yields have been higher than long-haul international for the past several years, and focus in high growth sunbelt states that remained open during the pandemic.

      In an era of limited pilot availability, use of large regional jets will be sharply limited. Given that focus cities and hubs with limited connectivity rely highly on RJ’s for non-hub flying, the opportunity cost of operations like AA@NYC, AA@AUS, DL@MIA, DL@RDU, etc. is less fortress hub feed from smaller cities. This game theory move takes arguably profitable flying from fortress hubs and moves it to focus city capacity that seems to provide seemingly less passenger utility.

      I agree with your analysis of AA’s intentions and approach (though I think that there are other causes for DL’s relative retreat, namely regional crew availability). However, I don’t see a path for this strategy to lead to long-term profitability for AA. The most profitable companies produce a product that no one else can compete against, hence why the fortress hub strategy has historically been used (produce the best schedules for customers at the hub and its regional catchment area). Simply waging battle against competitors while not generating a meaningful return on investment is terrible for shareholders. Maybe AA is nothing better than a bondholder investment.

      1. AA is now generating margins close to DL and United to a lesser extent
        Old paradigms die hard but the current AA is being run for profitability.
        They just have to figure out NYC. They’ve got pretty good strategies for the rest of their network and that is refreshing

  6. AA and DL were head to head 5 N/S a day for years during the week. Now AA is 2 or 3/day while DL stays at 5. Meanwhile, DL has the A220 on some of those flights vs AA’s 175s. Guess who I’m flying most of the time?

  7. I’m with Julie in that AA’s weak position in NYC left them open to attack from Delta on the NYC-based domestic corporate customer front. The Northeast Alliance really helps here, since most corporate travelers have to fly economy on domestic flights anyway.

  8. The simple explanation for AA’s uncompetitive position in NYC even though they were once headquartered there and were the largest carrier at JFK for years is that their decision not to file for bankruptcy in the five years post 9/11 cost them dearly as their costs became unsustainable compared to other carriers. JBLU took advantage of AA’s high costs by growing into AA’s core NYC markets including in the Caribbean and transcon markets. When Delta decided to close its DFW hub – where it was a distant #2 to AA – actually still is the #2 airline at DFW in revenue – and shift all of the connecting capacity to build NYC, AA’s historic position in NYC was finished.
    While AA argued post 9/11 to abolish a percentage of slots at LGA and JFK when the FAA briefly removed slot controls, DL chose to add flights to both airports and those flights were grandfathered in when slot controls were reinstated. The DL-US slot swap sealed DL’s position as the largest carrier at LGA and JFK was solidified.
    The NEA was an attempt by the DOT not to take AA’s slots away from them even though they were underutilized and they couldn’t remove slot controls at LGA and JFK; high value passengers recognize that B6 does not offer service anywhere close to what the big 3 offer and even what they do offer is far less reliable than any of the big 3. AA cannot fix its problem of lack of corporate travel as long as B6 is its partner and B6 has the business plan it has.

    As for DL, they managed to become the #1 carrier at LAX during the pandemic and also became the #1 carrier by revenue in BOS and neither position seems threatened. Developing an AUS focus city right now is simply not on table esp. as long as AA and WN slug it out w/ each other.

    DL execs just said on an investor conference that their partnerships with Latam and Korean Airlines are both ripe for development and Delta is just getting started on both including in MIA. Delta does at least one thing very well and that is that it is able to develop long-term growth strategies and methodically develop them one at a time.

    1. I gotta say, if the decision was to trade DFW for JFK, then hat’s off! Anecdotally I remember that was the case from our spoke, TLH. My father enjoyed using DFW to fly west on Delta and was disappointed when that was traded for an extremely short lived TLH-JFK daily.

      However, as I’m sure Tim and others remember, the loss of the DFW hub was significantly mitigated by the partnership with CO such that Dad could go west via IAH and still get his sky miles and visit crown rooms. It’s ancient history at this point and I don’t know how the money was carved up but DL travelers certainly didn’t lose much by dehubbing DFW and instead gained a powerful position in NYC.

  9. There is a lot of discussion regarding JetBlue, AA and the NEA and them winning against the DOJ opposition. I am not sure this is a foregone conclusion. AA has been very vocal about their new strategy of dominating their markets (Fortress Network) and no longer valuing corporate business. They laid off half their sales force and are reducing/eliminating corporate contracts. They don’t believe they need corporate business any longer because they are so big and dominant that corporations cannot shift off of them. I think that alone helps to make the DOJ case against this alliance being uncompetitive. Why would the DOJ want a carrier that is clearly touting their dominance any more power? Unfortunately JetBlue gets caught up in AA’s new strategy and arrogance. This was not the right time for AA to take this position in my opinion as they are helping to make the DOJ argument of being anticompetitive. I also find it amusing that AA states corporate business is not important compared to leisure and bleisure business. I think they forget that the B in bleisure is business and that it is the business travel that drives this segment and not the leisure. I think AA’s arrogance will hurt them with both corporate business and the DOJ complaint.

  10. Interesting.

    I’m guessing (and only guessing) that slots aren’t the only reason American hasn’t cut back at LGA and JFK. The other is the Atlantic joint venture with British Airways. That joint venture is a backbone of American’s overall international network and is important to oneworld. Los Angeles and Chicago probably don’t have the same importance to American’s overall network as New York does, especially since the airline is largest in the eastern part of the U.S. I remember a poster named “commavia” who once predicted that JFK would be more focused on O&D, while Philadelphia would focus on connections. That made sense to me then, and still does now. Even though American has shrunk in Philadelphia, a 60% market share is still pretty strong, and may be an ideal size moving forward. Philadelphia and Newark are the only fully functional hubs in the greater New York/Philadelphia metropolitan area. LaGuardia and JFK aren’t, no matter how big Delta gets there.

    American’s internal accounting data are far more complete than publicly available figures, and they’re real time numbers not a snapshot of the past. As the warnings in investment prospectuses tend to read, “Past performance is no guarantee of future results.” Just because American has struggled in New York in the past doesn’t automatically mean that it’s doomed to do so forever. The airline industry isn’t set in stone. And neither are passenger habits. Based on what I’ve read over the years, it took Delta (the world’s only perfect airline according to Tim Dunn) years to become profitable in New York. But it hung in and is now reaping the benefits. To reiterate: It’s quite possible that LGA/JFK are vital to American’s **overall** network profitability in spite of the data in the article. Maybe American should submit this to the judge in the NEA case as evidence that fares haven’t risen because of “collusion.”

    1. it’s funny that you argue that only AA knows how profitable AA is in NYC and that is true – but then you argue that it took years for DL to become profitable there; just curious what data you or anyone else used to come to the conclusion about DL’s profitability since the lowest level of profitability data that is publicly available or can be calculated using public data (including by competitors) is at the global regional level.

      AA argued for years that it was strategically necessary to serve LAX to China and yet they finally admitted that both routes lost money for years after I said the same thing to the objections of many.

      DL operates hubs at both JFK and LGA; just because you want to think they are not hubs doesn’t mean they aren’t… LGA for DL acts and quacks a whole lot like AA at DCA and unless you want to tell us that DCA is not a hub for AA but AA sure seems to think it is. DL moves more revenue through JFK than AA does at PHL so that argument pretty much falls apart.

      The point is that AA is NOT of a large enough size to compete in NYC and they have had an on and off approach to NYC that spans 2 decades – which explains why they can’t get traction. Now, AA’s strategies are essentially being held hostage by the DOJ’s lawsuit and B6′ desires to secure its future which might be worth a lot more than holding onto the relatively small benefits they get from the NEA compared to the NK merger.

      As for the commentary about AUS above, how many people think that DL put showers in its new 20,000 square foot Sky Club in Chicago for the domestic passengers? Delta put those showers there and built such a large facility for the use of its partners that also use Terminal 5. Many forget that DL made a fortune leasing out gates and its then Crown Room while Terminal 1 was being built for United and the international terminal operated out of a parking garage. DL had real gates, a real club, and global airlines flocked to DL’s facilities. The same – sans construction applies not just to ORD now but to more and more non-hub cities including AUS where DL is creating a focus city for its partners. MCI and BNA are both on the same track, the latter of which opens a new international terminal this fall.

      SkyClub agent gossip like that on will ramp up as Delta takes delivery of dozens of new widebodies per year and very likely confirms a new widebody order in the near future. Its domestic system is being restored and that is not great news for AA which has offered the most connectivity among US airlines for the past 3 years, something AA and DL both did well pre-pandemic. Unlike AA, DL and UA also see international as a huge growth potential and are doing it from proven hubs on both sides of the Atlantic and Pacific.

      1. Tim,

        If you would read what I actually write instead of automatically getting defensive about your perfect airline, it might help.

        I wrote, “Based on what I’ve read over the years…” I never claimed that I had any hard data to back up what I wrote about Delta’s profitability in New York, now did I? If you have hard evidence to the contrary, please provide it.

        Again, if you would have read a bit more carefully, you’d note that I used the term “fully functional hubs” when describing Philadelphia and Newark. Last time I checked, there’s a perimeter rule at LaGuardia that tends to preclude it from being used as a trans-Atlantic or long haul domestic hub.

        What do you mean by “large enough to compete”? The DOJ seems to think the NEA is large enough to reduce meaningful competition in New York, an opinion I believe you’ve echoed a number of times. Even as a stand alone, American effectively competes with Delta between New York and London, doesn’t it? Why does every airline have to compete on every route or be everything to everybody? And to repeat myself for the umpteenth time, when did it become a crime for airlines to have different business models? I would argue that none of the U.S. airlines will ever compete on every route, nor can they ever – unless they all merge to form “AmAir” – which would probably turn out to be the functional equivalent of Alitalia or Aerolineas Argentinas. Yikes!

        I never mentioned anything about trans-Pacific flying, so why even bring it up?

        1. Nobody is getting defensive. I am challenging your notion that LGA and JFK aren’t full hubs for DL. Feel free to tell us the criteria for a hub and good luck convincing anyone that your definition rules.
          DCA is also slot controlled. DL moves more revenue through both LGA and JFK than other airlines move through other hubs. Refrain from imposing your criteria on what other airlines do. AA says DCA is a hub. DL says LGA and JFK are hubs and there is plenty of evidence that both are right. Leave it there. Different strategies and all that stuff…

          The Pacific matters because AA execs like UA’s admitted they flew routes which didn’t make money. Even when DL execs said just a few years into building JFK as a hub – over 15 years ago – they didn’t talk about specific routes but the general profitability of the whole hub which is part of the process of starting any new line if business

          CF’s whole analysis here is solid. There is solid evidence of AA’s strategies in its other hubs but a lot of that evidence just isn’t there using the same criteria for AA’s 2 NYC hubs

          1. Tim,

            Please re-read what I’ve written about New York hubs. I used the words “fully FUNCTIONAL”, not “full.” I didn’t write that LGA or JFK aren’t “hubs” – just not that they aren’t fully functional. Newark and Philadelphia are fully functional hubs, as are Atlanta and Detroit for Delta – one airport handles all the traffic because it has no restrictions on the type of flights that can be operated there. CF puts it thus in the above piece, “This would allow the airlines AA & B6) to work together in order to try to compete with United’s Newark hub as well as Delta’s LaGuardia/JFK split hub operation. “Split” hubs aren’t needed if one airport is fully capable of handling all kinds of traffic. To repeat, I didn’t write that LGA and JFK weren’t hubs for Delta. You inferred that. You read more than what was written. That’s what people do when they get defensive.

            This piece and my comments were about New York, not the Pacific. I usually try to stick to the topic. I don’t always succeed.

            I didn’t write that CF’s analysis wasn’t solid. You inferred that. The numbers are the numbers. That’s why I facetiously wrote that CF should submit the piece to the judge in the NEA case. It’s pretty solid evidence that the NEA hasn’t caused fares to rise in New York.

            As both you and CF have observed, AA’s situation in New York is long standing. Those things usually take time to reverse. It could turn out to be foolish for American to try to compete on an equal basis with DL and UA in New York. In a more analogous situation than the Pacific, Delta apparently chooses to be relatively small in Chicago. That hasn’t hurt its profitability at all. American may ultimately make the same choice in New York. Only time will tell.

        2. In fairness, if Tim listened to many investor calls (not necessarily the 8k) from delta over the last decade, delta has talked about their slow road to profitability in NYC many many times.

          1. I listen to investor calls and read the transcripts of each of those from the big 4 on a regular basis
            Feel free to let us know the quarters in which Delta expressed concern about JFK profitability
            You won’t post them because Delta doesn’t talk about profitability of its own hubs, doesn’t compare financial performance of its hubs and doesn’t try to convince its investors of the financial performance of other carrier hubs esp unlike Scott Kirby.

            Let’s be clear that the root issue is that you and ghost want to believe that DL is afflicted by the same challenges that AA is BBC and the evidence simply says otherwise

            MYC and DCA are weak for everyone due to slower rates of return to travel but DL’s performance pre-COVID for NYC was similar to its system and CF could show that but that is not true for AA in NYC

            1. Tim.

              Julie didn’t write that Delta “expressed concern” about JFK profitability. She wrote about a “slow road” to profitability in NYC. News flash: Those aren’t the same.

              You tend to infer things that aren’t there. That’s what people do when they get defensive. That’s also what they do when they want companies to be liquidated.

            2. Tim
              It’s tough to take you seriously sometimes when you start from a place of “delta has always been profitable in New York. Prove me wrong”.

              Everyone knows that isn’t true. Even you. Delta has said it many many times. Asking others to go back into decade old transcripts to prove your ignorance doesn’t make you look good.

              Listen to some investor calls. You’ll learn a lot. Delta was quite transparent about NYC profitability and was asked about it quite a bit ~2013-2018 on investor calls in the Q&A.

      2. Gee, Tim, thank you for finally admitting something that I’ve been saying for quite a while now: that the SkyClub in Terminal 5 at ORD is a SkyTeam lounge, not a Delta lounge. I’ve received a lot of flak for saying that the SkyClub would be for SkyTeam use since before Delta moved to Terminal 5. Your fellow Widget junkies kept promoting that lounge as a sign of how great and wonderful Delta was, when the intent of the SkyClub was always for SkyTeam use. That’s the whole purpose behind the ORD21 project: advance the needs of airline alliances. Terminal 5 was always earmarked for SkyTeam and “others” and has limited space for lounges (the curse of being the last international terminal built before the modern era of lounges and alliances).

        So it’s good that you, the ultimate Delta fanboy, finally admitted what the purpose of the SkyClub in Terminal 5 was. Now you cannot pimp this up as a mammoth accomplishment for the Cult Of The Widget.

        1. I think you just proved I am not really a Delta fanboy
          If you have been watching ORD as long as I have, you know like me that Delta knows its role in Chicago
          Supporting its global partners is one of its key jobs
          None of the big 3 build assets for their own use anymore. They all are connected to their partners.
          Delta just knows that it’s future in Chicago is as much about serving its partners as it is about getting premium revenue in the markets it does serve

          Maybe that is the model American should be pursuing in NYC

          1. Honestly, Tim, you saying that you’re not a Delta fanboy is like Tom Cruise saying that Scientology isn’t a cult. Massive amounts of evidence suggest otherwise.

            I cannot charge the Widget with hypocrisy in labeling what is clearly a SkyTeam lounge as a Delta SkyClub, because my beloved UA does the same thing over in Terminal 1 at ORD. Realistically, the United Club at Gates B16-18 should be called a Star Alliance club, since B16 and B17 are where LH and NH normally fly out of. Ditto LH; the Lufthansa Lounge at DTW is not only a de facto Star Alliance lounge, last time I was there, it had signage at the door saying “United customers welcome”.

            But the question remains: if the partnership between SkyTeam is so important to Delta and it’s one of DL’s “key jobs”, as you put it, why not label the Terminal 5 SkyClub as SkyTeam since that’s what it is?

            1. Ask Lufthansa why they don’t call their Senator lounges in Frankfurt as Star alliance lounges and you will have your answer.
              Just as for LH in FRA the majority of DL’s Sky Club visitors in Chicago are Delta customers. Not sure why it is hard for some to admit that Delta succeeds at its own strategies. Different is good, right?

  11. You post about how AA is not doing well in NYC, and how Delta is slot squatting. This begs the question – who is actually doing well in New York? What is the right strategy for the market?

    1. Anthony – Delta has only been temporarily slot-squatting and it has started to move those away. For United and Delta, the hub operations they are running both make sense. They are #1 on their respective sides of Manhattan and they can compete within Manhattan.

      1. Brett, would you say the UA EWR and DL JFK/LGA hubs are comparable? Both seem to work well, but with DL limited on LGA routes they can fly due to perimeter rule and the need to transfer connections between airports, it seems they would rely more on local traffic without benefitting from connections to help fill the flights.

        1. Not Brett but I’d say United has a significant advantage by having everything under one roof instead of an international hub at one airport and a domestic at the other. Delta makes it work but with a lot of duplication (flights, personnel, infrastructure) that makes it much less efficient than United having everything in one place at EWR.

        2. Mark – I don’t think they’re really comparable, but in terms of being able to generate enough traffic to serve the needs of New York City travelers, they both do their jobs well. Still, the makeup is fairly different. For example, looking at this month, 16.5% of Delta’s seats are on flights under 300 miles while United is down at 11.4%. Meanwhile, United has 27.2% of flight between 1500 and 3000 miles while Delta is at 19.2%. And United has 14% over 300 miles while Delta is at 10.4%. So Delta skews shorter haul and that makes sense because they have to duplicate a lot of the shorter haul flying across two airports. At the same time, LaGuardia being restricted means that it can’t have some longer haul flights it might like to have.

          1. Delta has 5 effective runways between LaGuardia and JFK for roughly the same number of flights that United operates out of Newark on 2 runways. I am not sure why it is hard for some to see that a split operation in NYC might be the best operational option in a very operationally challenging city

            Being the largest carrier at 2 of 3 airports in a region isn’t bad either.

            Delta right now is operating more flights at all 3 NYC airports than any other airline

            1. Who gives a damn about runways? You could have 50 runways but it wouldn’t matter since everything is led by the same TRACON.

              Until going from LGA to JFK is as easy as going from terminal A to terminal C, DL’s split baby approach will forever be inferior to single location operation.

              That being said, they have definitely made the best this situation but it will continue to be inferior until they get Elon Musk to build one of his supersonic hamster tunnels between the two airports.

        3. The two approaches have different trade-offs.

          United at EWR benefits from running a consolidated hub, which provides better economies of scale than splitting traffic between two airports. United is also able to offer more frequency to most destinations, because it is operated as a “true” connecting hub in the United network, with lots of connecting traffic to fill those additional planes.

          United’s biggest disadvantage at EWR is the perception that it is less convenient for NYC-bound travelers, especially those traveling into Manhattan. For travelers who plan to take transit, this is definitely true – the EWR transit options are significantly worse than those from LGA or JFK. The EWR AirTrain is cramped, slow, and doesn’t even go to the new Terminal A (you need to take a shuttle bus). NJ Transit is infrequent (especially on weekends) and comparatively expensive, and Penn Station isn’t particularly close to the most common travel destinations in Midtown. At the same time, if you’re willing to pay for a (pretty expensive) Uber, then travel time from the airport is comparable to LGA, and a bit shorter than JFK in most cases. United has spent a lot of money marketing this message in New York (particularly on top of cabs), but I think most people still assume that the airports that are “in NYC” are quicker to get to.

          By contrast, LGA commands a premium from NYC O&D passengers because of an (often correct) perception that it’s the quickest airport to get to, and will enable the passenger to spend less time traveling to/from the airport. But the perimeter rule and general congestion prevent airlines from building a true hub operation there. As you mention, airlines at LGA do rely more on local traffic, with relatively few connections. But there is an enormous amount of local traffic to serve, so this isn’t too limiting.

          JFK benefits from the substantial number of international routes, including on many international carriers. This provides the critical mass of connecting passengers to feed hub operations from AA and DL. It also retains some advantages for NYC O&D traffic: It has a much better transit connection (AirTrain to 2 different subway lines) and better general name recognition from those outside the NYC area.

          Those advantages and disadvantages are all most relevant for travelers to and from NYC itself, and especially Manhattan. For travelers with origins and destinations in the broader metro area, the various airports enjoy essentially local monopolies in their direction from the city, because traffic makes it miserable to cross the metro area to go to another airport. EWR captures essentially all passengers with origins or destinations west of the Hudson. Passengers traveling to/from Brooklyn, Queens, the Bronx, Westchester County or Long Island will choose LGA or JFK, with a bias towards whichever one is closer to them. Staten Island ends up split between JFK and EWR.

  12. You can’t be everything to everyone. The sooner AA comes to this conclusion, the sooner it can fix the problem. AA has two problems as it relates to NYC:
    1.) Their hub in NYC is too small relative to their competitors, so they can never win this battle in slot controlled airports
    2.) AA has WAY too many hubs in the northeast: PHL, DCA, LGA, JFK. That is 4 hubs within 225 miles of each other.

    Hubs are extremely costly to operate and there are too many redundant traffic flows across the 4 airports. I think the only smart move financially for AA is to acknowledge they cannot win the NYC battle and close some hubs in NYC.

    It may be an ego blow and loss of some corporate contracts, but they should come out financially stronger.

    1. ” 2.) AA has WAY too many hubs in the northeast: PHL, DCA, LGA, JFK. That is 4 hubs within 225 miles of each other.”

      If we were talking about an area of the US like DFW to OKC, sure (for geographic comparison purposes)… four hubs would be too many. But you’re talking about the most densely populated part of the US, the NE Megaregion . An area (including BOS and the Tidewater region) that is ~58M people. DCA and the NYC airports can largely survive on local O&D alone, PHL probably requires connections but that point is made by Cranky. That PHL does a good job utilizing its limited O&D combined with a better (no slot constraints) airport to flow connections than JFK.

      This comment about “too many NE airports” has been said before but it largely ignores the enormous population density surrounding the area and the very large connecting population within ~400 miles of that megaregion. The greatest O&D demand in 3 of the 4 airports in the country.

      1. Oh it is too many hubs. Perhaps you’ve missed the fact that the population center covered by those 4 AA hubs also includes hubs at JFK/LGA for DL, hubs at EWR and IAD for UA, hubs at BOS and JFK for B6. And this doesn’t even address the gains made by Spirit and Frontier in this geographic area. Now re-run your math and get back to us.

        1. you are right, Eric.
          the problem is that hubs in the same region compete for the same FLOW traffic. It doesn’t matter how large the local market for a hub is; it is all of the flights to medium and small cities that it takes to fill longhaul flights including international is what duplicated hubs in the same region are challenging from a network standpoint.
          Add in that AA is adding longhaul international flights where it either didn’t do well before or are using inefficient aircraft – the 787 is not based at JFK or so I have heard – which means that AA is competing from a position of weakness even before considering their size relative to DL and UA. International fare data is much more protected (because many foreign carriers don’t have to provide the same data) but AA has underperformed DL and UA in many NYC longhaul markets outside of its core partner hubs and to GRU and EZE for a long time.

        2. Eric, I’ve done the math many times. Delta has 3 hubs already in the mega region, jfk/lga/Bos with dtw a rather short distance away and serving much of the surrounding region.

          Delta and AA both have two connecting hubs in the mega region, BOS and PHL.
          They both have what they call hubs at lga and jfk, though the size of each is obviously different.

          The only difference between the two airlines in that megaregion is dca. Are you seriously suggesting that delta would say no if dca suddenly opened up massively and delta could make it a hub like the size aa has there? Of course delta would jump at the chance. Then delta would also have four hubs in the area. Too many by you and Tim’s strange standards.

          Go back and work on your math and logic. It’s a very very large population in the NE megaregion with enormous local demand.

          1. Thanks for the quiet tim and Eric. Nice to know when commenters know they can’t compete. If only tim could do this more often

      2. Julie,

        To your point, about 25% of the U.S. population lives in the Northeast corridor between Boston and Washington.

    2. True hubs are primarily driven by connecting traffic. While it’s possible to connect over DCA, pull up any schedule on AA in this part of the country. If you see a dozen connections, you’ll see half of those via CLT, some over PHL and one over DCA. DCA is primarily an O&D airport with some connections. Anecdotally I’d guess 20% or so are connections. Therefore, not really a hub.

  13. American’s diversion of regional jets out of Philadelphia isn’t just ‘we need them in New York’ it’s that with less Philadelphia transatlantic flying – a lot of those domestic flights were geared towards feeding transatlantics – the domestic flying no longer makes as much sense. American’s Philadelphia hub was a transatlantic connecting hub, without the same O/D traffic as New York. Philadelphia also faces increased low cost carrier competition compared to the old US Airways hub.

    Now, you point out that New York is a bastion of low fares and PHL fares are higher. But those higher fares are with reduced capacity! That’s not necessarily an argument for bringing on more capacity and lowering fares!

    The NEA may wind up not making New York as profitable as other hubs, but remember they can’t grow more than they already have there (other than upgauging). And to your point about slots making you do funny things, under the DOT agreement that got government signoff on the deal (before the government sued to dismantle it! Despite no new information!) if AA/B6 didn’t increase seat capacity in the New York market *they would lose more slots than they’ve already divested*.

    The goal of the NEA involves selling more business together to corporate buyers but that hasn’t fully materialized yet because the New York business travel market has come back even more slowly than the country’s as a whole. Seems too early to tell whether this gets them where they need to go or not.

    1. AA is not trying to sell more business to the business market, they are pulling away from corporate. They have made a strategic decision to devalue corporate business.

      1. They pulled away from small corporate but the NEA was very much about selling managed travel together. And managed travel hasn’t fully recovered.

    2. Gary, it seems like you’ve also pointed to NYC as about capturing cc spend as well. I wonder what the analysis looks like including potential upside from the changes in the AAdvantage program and the effort to feel more present in the NYC market?

  14. Cranky,

    Based on your SLA PRASM chart, are you suggesting that Chicago O’Hare is no longer an American Hub? You’ve got every hub listed –except Chicago.

    I realize that United is beating their brains out in Chicago and can only imagine what would happen to local fares if American pulled out of Chicago.

    1. Dave – Nah, of course Chicago is still a hub, though a shrinking one. I just left it off by accident. Chicago is at 16.2 cents.

  15. All of this begs the question: What’s better – an unprofitable 30% market share of a profitable 20% share? Pre-consolidation, the “answer” was often the former. Now, more and more, it’s becoming the latter. At least I think so! LOL

  16. ultimately EVERY US airline exists as for-profit enterprises.
    Sure, an airline operates networks and can justify a loss leader – as grocery stores do on some items – for the larger good of the company but AA for too long took the approach that there were routes which DL and UA flew which AA also had to serve and AA got its clock cleaned; there seems plenty of evidence that DL and UA would reinstate all of their China authorities if they could – but they can’t.

    CF’s analysis is fascinating here. I’d love to see the same for other carriers but I suspect he highlighted what he did because NYC sticks out for AA and I doubt if you would see the same thing for other airlines.

    We get that AA needs to found its niche in NYC and we all, at this point, understand why. CF’s data – which is as real as some of the stuff I talk about from the DOT – tells a story – maybe not all of it – but a compelling story that is why CF does the nitty gritty work that other aviation bloggers wish they could do.

    And using depreciated, high operational cost assets in a market where you are trying to turn a profit seems a little counter to the notion of doing all to ensure success. Anyone notice that DL’s A330-900s see more of NYC and BOS than they do ATL and yet ATL is further? I always found it fascinating that, at least last year, DL flew the A330-300 on one of its longest transatlantic flights – ATH to ATL while BOS got the new generation aircraft. But maybe DL recognizes they could make money in ATL on alot of routes even if they resurrected the L1011.

    Maybe worrying about depreciated assets in NYC is the least of AA’s problems. I mean, they could probably make money flying DC-10s from DFW to Asia (ok MD-11s) if they had to. Maybe it just comes down to route and hub accounting – all of which rolls up to ultimately hit the bottom line which investors see.

  17. Every legacy airlines NYC strategy revolves around business travelers, so 2022 numbers aren’t necessarily a good predictor of future success. Business travel was still really weak in 2022, and still hasn’t fully bounced back. Airlines are building their networks for the expected steady state, with more business travel than 2022, though maybe less than 2019. If business travel rebounds, the bet on NYC could end up looking great. It is a bet, though.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Cranky Flier