Southwest Moves Toward a Hub Model With May Schedule


Southwest has been busy cutting its schedule like everyone else, and for May it’s down by more than half to about 2,000 flights a day. That’s still a lot, and it’s more than demand would require, but the airline has some unique circumstances that make it willing to operate more flights than it should. One of those circumstances is the lack of a hub-and-spoke model, a strategy it is temporarily giving up on during the May schedule.

To be clear, Southwest is not all-of-a-sudden building mega-hubs. It is just focusing on its largest cities as ways to maintain connectivity throughout the network. Though it is stopping service to all international destinations until the first creeps back into the network in late May, all domestic cities will keep Southwest flights, and that puts Southwest into a tough situation.

For the big three US airlines, maintaining service is a fairly easy process. Just serve the smallest cities from the nearest hub and then connect those hubs to each other. You see that in American’s network, which I analyzed last week. Southwest, however, has more of a mesh network. It has always had some cities which receive a greater focus, like Chicago, Denver, Baltimore, etc. But its real power lies in being able to connect the mid-size cities like Austin or Kansas City with a variety of destinations both nonstop and via single-stop, same plane service. That mesh makes it harder for Southwest to pare its network without impacting these cities.

With this new schedule, Southwest has thinned out those mid-size and smaller markets, mostly linking them with the biggest focus cities that look even more like hubs now than they did before.

Let’s break that down further.

The Hubs

There used to be 15 cities in the Southwest network that served more than 30 destinations. Now there are 10.

These are effectively Southwest’s hubs for getting people around the country. What’s interesting is that some have taken a bigger hit than others. Look at Phoenix and Dallas. They’ve lost the most destinations in this group. That’s primarily because of their geographic position at the extremes.

Take a look at Dallas in greater detail. The green lines are routes that stay while the red lines are routes that go.

Maps generated by the Great Circle Mapper – copyright © Karl L. Swartz.

The red lines are to the furthest destinations, especially secondary ones. Those are all cities that can be better served through other “hubs” while demand is down. That being said, Dallas remains an important point for the south-central part of the country.

Mid-Sized Cities Get Whacked

The next tier down are cities that had their own sizable operations before, but they’ve seen massive cutting in this round. These are the cities I talked about up top.

The first one at left is the most fascinating. Houston used to have 50 destinations in the Southwest domestic system, but now it is down to a mere 28. What the heck happened? Well, if Dallas was geographically undesirable, Houston is worse. Not only is the international system gone, but nearly every other city in the US is better served from another hub. It lost a ton of short-haul flights to places like Amarillo, Birmingham, El Paso, and Lubbock — because those can be better served through Dallas — in addition to the far-flung destinations.

The rest of these cities were strong in the Southwest network previously, but in a hub-and-spoke type of network, they don’t serve much of a purpose so they’ve seen destinations slashed.

There’s one other thing I want to highlight here, because you won’t see it addressed on any other chart. San Jose got hit hard, but that’s mostly because Oakland lost very little. Oakland didn’t have more than 30 destinations so it didn’t make the top chart, but it still behaves like a regional hub.

Bigger Spokes Connect to Fewer Hubs

All those cuts at the top mean that the rest of the destinations have far fewer points to connect into the Southwest network. Here are some of the larger Southwest “spokes” that used to have between 10 and 20 destinations. Now they all have fewer than 10.

The idea with Southwest’s new schedule is that breadth isn’t as good, but every destination that matters still has multiple points to connect into the rest of the network.

Smaller Spokes Hang On By a Thread

At the bottom end of the spectrum are the smaller Southwest cities that had fewer than 10 destinations previously. They’ve now lost at least half of those.

In these cities, the name of the game is to just keep these cities connected to some piece of the network. Demand is nothing, but at least it keeps some semblance of a presence.

Thinking It Through

This kind of strategy benefits Southwest in a few ways. In a world where Spirit and Frontier are asking the feds to abandon two or three dozen destinations, Southwest isn’t looking to abandon any. You know those airports won’t forget this, and it should allow Southwest to benefit down the line.

Further, having all this capacity in the market that really isn’t necessary may seem like a waste, but it only helps to bleed airlines with worse financials. Southwest can outlast them all, so for the short-term, it might like to turn the screws.

I don’t know how much of this is intentional, but Southwest looks like a hero to its airports, its few remaining customers, and once again, to the government. It can afford to play this game for a couple of months easily. If things get worse, then it might have to shift. But for now, it’s winning the long game by flying empty airplanes all over.

Get Cranky in Your Inbox!

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

36 comments on “Southwest Moves Toward a Hub Model With May Schedule

  1. It just seems to me that WN is flying way more than they should be flying. heard some rumour that WN might cut another 500 flights to around 1500. Honestly, they should be closer to 1000 flights daily based on the current demand level. It seems like the big4 are still quite concerned about market share battle at this point. WN has more cash on hand than any of the other major carriers, so it can afford do this. But I don’t think this is a wise move. Does it really make sense to fly 10% full planes all over the country to sustain your market share? A year from now, WN will be smaller than 2019 like everyone else.

  2. To the point about airports not forgetting that WN didn’t abandon them… I’m not so sure that the airports will remember it 12 months from now, especially if WN eventually abandons airports a few months down the road, or if/when other carriers restore service.

    That said, have there been any reports of airlines cutting deals with airports, in the sense of promising to maintain service for the next X months in return for reduced fees (now and/or in the future)? I’d be surprised if the airlines were not playing hardball a little, even given the uncertainties of the times, especially with the more borderline airports they serve.

    1. There’s a limit to what airports can do in this regard, because if airports have taken FAA money they are obligated to treat airlines equally.

      That doesn’t mean that every airline has to be treated the same regardless of how it chooses to serve the airport, but it does mean that if two airlines are serving the airport in the same way, they can’t be treated unequally. The airport can have a menu of charges (one menu item might work better for a carrier that flies seldom, another might work better for a carrier that flies often, etc), but all carriers have to be free to choose whatever menu item works best for them.

    1. That is exactly the southwest model…milk runs, i was on a MSY-ATL flight that if you tracked the full flight number that started in BWI and ended in OAK with 6 stops along the way

  3. In the midst of the dizzying and continual amount of flight cancellations, your schedule analyses and conclusions are certainly helpful, CF.

    WN does have more cash to sustain its operations but if they, like perhaps other airlines, decide the fed grant money isn’t worth it, they might have to tighten their belts a little more than airlines that accept the grant money.

    Many of the trends you note are based on WN’s market share in each of those markets as well as how much business and/or leisure demand each of those markets represents. WN dominates markets like HOU and DAL and can afford to cut deeply because no one else is going to take any share, either temporarily or longer term. Markets like SJC are heavily dependent on the tech industry and travel will likely be slower in recovering than in other markets. Heavily leisure markets such as in Florida will take more time to recover because a large portion of the market is incremental travel that will be impacted by economic contraction as much or more than medium to long-term concern about the virus.

    Looking forward to more airline specific analysis not just for May schedules but when guidance is relaxed to the point where people begin to venture back on planes.

  4. Where is LAX? I would be interested to see what that station looks like.

    Also, I am somewhat surprised to see Vegas retain he number of destinations it did given that tourism is their #1 industry.

    1. Good point about Vegas. I haven’t been paying particular attention to that part of the country, but from what I’ve read and heard, the Strip and other “touristy” parts of Vegas are absolute ghost towns right now.

      Given the time that it takes to organize, promote, and arrange conventions, one would have to imagine that we could be well into the fall as a best case before the convention traffic really starts to return… Non-convention traffic may return a little earlier, but we’ll see. Definitely going to be tough for the economy out there.

    2. Eric – Yeah a few stations were left out because they didn’t fit entirely into one category and LAX is one. That saw very little cut. It had 24 destinations and is now down to 18. Service was cut to Austin, El Paso, New Orleans, Portland, San Antonio, and Tucson.

      As for Vegas, remember it’s still a big city with a lot of people who live there. It’s also a useful connecting point for much of the west.

      1. The point of Vegas being a decent connecting hub has merit.

        Per the Las Vegas Convention and Visitors Authority, 41% of jobs in Southern Nevada depend on tourism, directly or indirectly. Perhaps once things start to improve we’ll see people from California look to drive to Vegas, as opposed to flying to there or taking longer distance trips to other destinations by plane. Until then, however, I would imagine that like many tourism destinations Vegas will have a very tough time in the next few months.

        1. Oh, no doubt about that. If you go to YouTube right now, you’ll find dozens of videos involving people driving up & down the strip & all you see is shear emptiness with the exception of the one off curious pedestrian or a few homeless. Keeping the theme of the blog alive, one of these videos was done by a local realtor named Angela O’haere.

    1. W – DCA and LaGuardia haven’t been cut much. In fact, National lost no destinations at all, maintaining all 17. I assume this is a political strategy. LaGuardia also kept all destinations, but it only had 10 in the first place.

      1. DCA is very interesting indeed, especially when you consider the fact that AA whacked their DCA hub. I believe AA is only flying to their hubs from DCA. Meanwhile, Southwest decided to retain all their flights.

  5. There are a couple of interesting dynamics at work here. But first, a disclaimer or two. The main reason my observations could be dead wrong is that I don’t have access to the data in Cirium, and have no compelling reason to subscribe to services like it. It’s also good to remember Mark Twain’s thought that “There are three kinds of lies – lies, damned lies, and statistics.” But based on my reading, which could be incomplete, a couple of thoughts come to mind. The first is geographical. Baltimore is in a geographically extreme position. But it’s also in the highly-populated northeast corridor which makes it an ideal spot to connect north/south traffic. Not to mention that the other two airports into and out of the area are being dramatically reduced in scope because of the decline of government-related travel, both domestically and internationally. Which brings up the next thought. Southwest has far less international exposure than the other three major airlines. Measures like “available seat-miles” are skewed to some extent by large aircraft flying long distances. A large international network seems to be more of a disadvantage in this environment than an advantage. It should be an interesting few weeks, if not months. I saw a Helane Becker forecast this morning predicting that the major airlines will emerge from this about 30% smaller than before the pandemic. We’ll see.

    1. I guess the 30% smaller depends on the point in time you measure them. However, I would not be surprised to see the weak hubs at all major airlines reduced to spokes. You can support money losing hubs during good times via the rest of the network, but definitely cannot keep them during bad times.

    2. This is from the e-mail I received this morning via Seeking Alpha:

      Cowen expects U.S. airline industry to shrink
      Apr. 14, 2020 7:21 AM ET|About: American Airlines Group Inc. (AAL)|By: Clark Schultz, SA News Editor

      Cowen’s Helane Becker forecasts what the airline industry will look like on the other side of the pandemic.

      “Demand is 5% of what it was in February, and we continue to believe it will take 3 to 5 years for domestic demand to return to 2019 levels and 4 to 6 years for international demand to get back to those levels,” she notes.

      “We believe airlines will be 30% smaller at the end of his year than they were at the start of the year. Also, we expect there will be between 100K and 200K fewer employees at the end of the year than there were at the beginning of the year,” she adds.

      Becker thinks U.S. airlines will retire between 800 and 1000 aircraft this year.

  6. A friend’s son just flew nonstop PHX – BWI on WN and there were only 4 people on the flight. He has flown that route over 20x and said every single time other than this flight, it was 95%+ full. I am on the BWI flight path and can see WN planes continuing to land. They have a very large presence here (2nd after MDW in their system), and I believe were planning on putting a maintenance facility at the airport. Let’s see if that is still in the cards.

  7. Nice analysis, thank you for putting it together.

    One point: I’d say Southwest is fairly unique in that it hasn’t been cutting flights like all of its competitors – every other mainline passenger airline in the US has reduced the number of flights they’re flying by 70%+, Southwest has only cut theirs 50%.

    1. They kind of have to. I’d imagine they have been burning cash at unnecessary rate at the moment from their reluctance so far. The actions that the airlines are taking right now is going to really affect where they are coming out of this.

    2. Looks like they finally realized that they were wrong in attempting to cut so few flights. What took them so long is my question…

    1. The post actually mentions OAK:

      “ There’s one other thing I want to highlight here, because you won’t see it addressed on any other chart. San Jose got hit hard, but that’s mostly because Oakland lost very little. Oakland didn’t have more than 30 destinations so it didn’t make the top chart, but it still behaves like a regional hub.”

      Hope this helps!

  8. From March to May Southwest at Memphis went from 7 destinations/9 flights to 4 destinations/6 flights.
    I am surprised BWI was cut and ATL stayed.

  9. One thing I wonder about as I read all these posts talking about how airlines aren’t cutting enough based on the fact there’s virtually no demand. That’s all taking about passenger demand. What about cargo? We know the passenger airlines are getting into the cargo charter market, offering flights (especially on widebodies) for moving cargo without passengers. Perhaps keeping these scheduled passenger flights plays into that. There may not be many people in the cabin, but what’s going in the belly? We’re all shopping online, and not everything goes in an Amazon 767.

  10. I have a trip in early June on WN from MHT to PHX. Got three schedule changes in three days last week. All pretty minor but will be interesting to see how many more. I already know I am going to cancel since my conference was cancelled, but want to see if they cancel first.

  11. “This kind of strategy benefits Southwest in a few ways. In a world where Spirit and Frontier are asking the feds to abandon two or three dozen destinations, Southwest isn’t looking to abandon any. You know those airports won’t forget this, and it should allow Southwest to benefit down the line.”

    another one of those meaningless comments seeing that by virtual of 737 only, other than small town Texas and others in close proximity to Love Field because of the traffic pattern build-up over the previous Wright Amendment years, Southwest has much higher floor of a minimum market size it was a serving compared to others. Heck, even the airport without perimeter restrictions in the NYC area wasn’t sustainable for Southwest.

    And particularly against the ULCCs, which specializes in your truly tertiary airport to sun beach and gambling destinations.

  12. Southwest is playing games with availability. I have a flight in two weeks that is showing as “unavailable”, while all other flights that day are selling for $45. I’m curious why Southwest hasn’t canceled the flight yet, maybe they just want people to cancel first so that they don’t have to give a refund. They already canceled my return flight and moved me 4 hours back, but I can’t get a refund without calling and I haven’t done that yet. I understand that they want to hold on to their cash, but the rules say that if they cancel a flight they need to give a refund, and the games they are playing are not appreciated.

    1. Update: this morning, a week before departure, I finally got a text that my flight was canceled. The flight was still showing on my reservation, and now in the evening it’s still showing as “unavailable” on, but the text was reason enough for me to call Southwest and ask for a refund. Wait times were short, and they promptly agreed to refund my ticket (apparently the agent could see that there’s no flight).

      The agent explained that I would get an email stating that I am receiving travel funds, and that in a few days they would follow with an email that says I’m actually getting a refund; the money itself would take two weeks or more. I guess that’s the famous Southwest IT systems. I’m now curious about the reason there’s no web-based option to get a refund for a canceled flight: Are they deliberately trying to push people towards getting travel funds, or is it just something their IT can’t handle? Perhaps it’s a bit of both, functionality that doesn’t currently exist and that they’re not in a hurry to implement.

      Now that I expect to get my money back, my next question to Southwest is what happens with expiring LUV vouchers: they’ve been very vocal about their policies on extending travel funds, but there’s complete radio silence on LUV vouchers (just a few questions on Southwest forums). My guess is that they haven’t gotten around to this yet. I have a LUV voucher expiring in August; hopefully they’ll sort out the voucher issues by then.

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