As is often the case, United CEO Scott Kirby gave a masterclass on where he sees the airline industry going during his initial comments in the airline’s earnings call last week. There was a lot of talk of comparing revenue to Gross Domestic Product (GDP) and how cost convergence would result in higher profits. This may be tough to follow, so I thought I’d break it down to explain exactly what Scott is seeing.
Revenues as a Higher Percent of GDP
Over the years, comparing airline revenues to GDP or comparing growth in airline revenues vs growth in GDP have been commonly-used metrics. Depending upon the timeframe used, it can be a good but very general judge of if the airline industry is growing too fast or not fast enough to keep up with economic growth. That is where Scott’s story begins, but I should make it clear that his comments were zeroing in specifically on the US industry here and he’s looking at domestic revenues only.
Taking domestic airline revenues and comparing them to the US GDP, we get this slide which United included in its presentation:
Before 9/11, there was a completely different dynamic with GDP. Fares tended to be much higher, leading to higher revenues, and low-fare carriers were a smaller piece of the industry. But there was a permanent shift starting after 9/11 when fares tumbled due to the dramatic drop in demand. Low-fare airlines started to grow significantly and the legacy airlines spent the next decade trying to find the right way to deal with the demand environment, creating low-fare carriers of their own. (Remember Ted, Song, Metrojet… you name it? Now forget them. Those were all terrible ideas.)
This lowering of revenue decoupled the airlines from their GDP reference point. Even as airlines were adding capacity, the fares were not keeping up. In the decade before the pandemic, it settled into a steady place where domestic airline revenues were 0.49 percent of GDP which, as legend has it, is how the San Francisco 49ers got their name.
During the pandemic, all hell broke loose. There’s no reason to even bother thinking about what happened during those dark times, but it should be noted that the estimate for 2022 is that it will climb back to 0.45 percent. If you remember, fares were still pretty low at the beginning of 2022 when Omicron reared its ugly head. They rapidly picked up steam in the second and third quarter, however.
Consider this: according to DOT data, domestic airline revenue in Q1 was $41.2 billion in Q1. It soared to $55.3 billion in Q2 and $53.1 billion in Q3. If we hit 0.45 percent in 2022, then getting back to the now-standard 0.49 percent in 2023 should be possible. And Scott made it clear in his remarks that he sees opportunity to get back up to the mid 50s. Blue skies ahead for everyone (except the 49ers).
The Magic of Cost Convergence
In a sense, this is all backwards looking. The question is… why would the industry be able to get above where it’s been for the last decade? And that’s where this gets much more interesting.
Ultimately, this comes down to what Scott calls “cost convergence.” As he sees it, the low-fare carriers will have their costs get closer to the costs of the legacy airlines. Just doing the math, that will force fares up and that’s good for the legacies. Here’s a crude map with no real numbers just to illustrate the point.
The idea is that the low-fare carriers are the price leaders and the legacies must compete with them. If the ULCCs have higher costs, their fares will have to rise to maintain their current level of profitability. This will allow legacy fares to rise as well (not as fast since not all fares match ULCCs), but since legacy costs are going up more slowly, legacy profits will increase.
Poof… MAGIC. But WHY are ULCC costs increasing faster? Scott had several reasons for that.
Pilots, Pilots, Pilots, and Other Stuff
The biggest impact on costs will be the ongoing pilot shortage. The legacy airlines are setting the bar. We have a very rich contract on the table for Delta. If that gets approved, I think we can all safely assume that American and United will follow with something similar if not identical. So, pilots will get trained up and once they reach the 1,500 hours required to fly for an airline, they’ll roll into the regionals as a de facto training program. The regional flow-through programs are a way to get put right into the high-paying legacies after a short time.
Regionals had to jack up their pay dramatically, but now that combined with the promise of drastically increasing legacy pay within a couple years will keep the pipeline flowing and the legacies as the preferred pilot destination. If the supply remains tight, how do the other airlines get pilots if they aren’t paying as much and the pilot pool remains shallow? The ULCCs seem to believe that with legacy pay jumping so much higher, they’ll have plenty of room to increase and still attract people. Whether that’s true or not, only time will tell.
They will always be able to get some pilots. After all, some people like the bases that one airline may offer over another. Another person may like flying for, say, Allegiant, because they get to sleep in their beds every night. But for these airlines to grow, they need to attract a lot more than just those niche groups. Spirit pilots just approved a new agreement, and that will certainly help. But the way Scott views it, that won’t be enough. If they can’t get closer to legacy pay, they are never going to be able to staff all the growth that’s on the order books. At least, that’s the narrative being put forth, but it’s not a guarantee.
At some point, the pilot shortage has to ease compared to what we have now. The demand for pilots is through the roof, and the economy is booming. Growth will slow, the economy will turn down, and more pilots will probably come into the system. Some things, like the 1,500 hour, aren’t likely to change, so that will keep the squeeze on the pilot pipeline. Ultimately, however, there is going to some a cyclical nature to this and we are at the top of the cycle.
The key here, however, is that Scott isn’t saying the ULCCs won’t get any pilots. He’s saying that for them to be able to get enough pilots to fund all their planned growth, they’ll have to pay relatively closer to what the legacies pay than has been the case historically. That’s cost convergence.
But pilot pay is only part of it. Scott also says that more employees are needed to run every flight because sick calls are higher. At United, the back half of 2022 saw sick calls 19 percent higher than the back half of 2019. United views this as a permanent shift and not one that will go back to where it was. The other airlines will either have to be fine running a worse operation or they’ll have to hire more and increase costs. Even if they want to hire more, it may be tough to do without increasing pay again. Since the ULCCs have already run leaner operations, it might require them increase more than the legacies… cost convergence.
On top of this, there are two things that will keep a lid on capacity. First, the aircraft manufacturers are behind and can’t catch up. So, deliveries are delayed and all those airlines that retired a bunch of aircraft during the pandemic are now just waiting for more capacity. I expect this will eventually ease, but it does keep a lid on things in the near to medium term at least.
Also, and what is presumably a bigger, long-term issue is the country’s air traffic control system. As Scott sees it, we’ve basically hit our limit on capacity in the US, at least in certain parts of it. Sure, there can be more on clear days and you can fly to and from any city in Eastern Montana as much as you’d like, but additional growth where the bulk of the population lives is just going to compound these meltdowns that have become more and more regular. This used to be primarily an issue in the northeast, but we’ve also seen several air traffic slowdowns in Florida as of late. Unless the government does something about that, growth will have no choice but to slow down.
When you combine that all together, what do you get? You have lower capacity than airlines would like along with higher costs. That will push fares higher, and with cost convergence, that will lead to higher legacy airline profits.
Will This Happen?
We do have to keep in mind that this comes from the CEO of a legacy airline. This kind of pressure on his non-legacy competition is music to his and his shareholders’ ears, and it tells a grand story that supports United’s future. That being, said, Scott has a strong track record when it comes to predicting macro changes.
There are still a lot of “ifs” here that are out of anyone’s direct control. Pilot demand will shift, but will ULCCs be able to maintain pay at the levels they are at right now? I’d imagine some will do better than others, but it’s not entirely clear how this will play out. Will sick rates stay where they are? Maybe. Maybe not. Scott is much more confident about the path than I am, but I can certainly see how it could play out this way, and I think it would be a mistake to just write off the idea completely.
I wonder if an even better measure would be a percent based on GDP-minus government spending, to take out things like Iraq War, TARP, and Covid spending.
Interesting point, but it is worth pointing out that the US government is the single largest purchaser of airfare in this country. The GSA business is huge for the airlines, so it’s probably best to keep government spending in there.
Let me break it down a bit more. With cost convergence comes service convergence.
The legacies with become more and more like the ULCCs. You’ll get your 29” seat and extra costs and no service on American and United because they have to compete with Spirit and Frontier. And passengers are too unaware to pay a few more bucks for more space or service.
We’re pretty much there already.
It’s ridiculous, isn’t it?
There is a way out of the “capacity” issue, just as there is with “capacity” in housing, transportation, etc.: you can up-gauge the capacity. This is where Southwest thinking (single-fleet at all costs) eventually will come back to bite low-cost and ULCC airlines: for many places, there are literally geometric limitations on the availability of air and port space. You can try to shove more and more in there, but you’re just making things more and more fragile and susceptible to meltdowns.
Look at Alaska in SEA. There are enough passengers with enough frequency that they could easily throw 787s on some of their domestic routes. But they know the finances of moving away from single fleet are disastrous. Scott Kirby is right: airlines, like Alaska, will (and have been for a few years now!) face *geometrical* limitations in their ability to grow. Alaska *can not grow* in Seattle anymore. Seattle is *full.* And Seattle is basically Alaska’s one and only lower 48 growth market, because it’s the only place they can scale the economics of their business model. But you can only scale up so much with a 737-10 as your biggest aircraft. There are X number of gates/departure slots at SEA, and Y minutes in the day, and Z minutes to turn each aircraft. Z is not getting shorter, Y will never grow, and X is reaching the limits of growth.
Hand-in-hand with the post 9/11 economic restructuring of the industry that CF talks about in this post, here, is the explosive growth of LC and ULCC carriers tied to the “single fleet” strategy. Space-constrained airports in Japan fly 787s around the country on distances that people in America would be stunned at. Every single day ANA flies almost 20 777s and 787s between Haneda and Fukuoka (547 miles! That’s less than PDX-SFO!) They don’t try to fly 50 maxed-out 737s between those cities. (Remember when DC-10s flew routes like GEG-ORD?)
There are limitations to growth when the biggest aircraft you will *ever* fly is a 737. Legacy carriers are positioned better than most to take advantage of that lack of limitation on scale. United’s 100 aircraft order for 787s reflects this reality, too.
Why wont z get any shorter?
There are just plain old elements of geometry and physics that will prevent fully-loaded 737 MAX 10s from unloading and reloading their passengers and cargo much faster than the 40-or-so minutes typically allotted, right now. The 737 cargo is not containerized or container-izable, and I highly doubt we will see dual-door deplaning/emplaning deployed widely for speedier boarding. Turn time was low-hanging fruit that LCC/ULCCs pounced on early. There’s not much more efficiency to wring out of it.
I was thinking about the upgauge as well. However, there are a few things preventing this from happening overnight. First, obviously Boeing and Airbus can’t churn out widebodies fast enough. Second, not all gates in major US airports/population centers can handle widebodies. Actually, most gates are designed for smaller jets. Putting widebodies in there may effectively prevent one or even two neighboring gates out of commission reducing X. Z is also longer widebodies. I guess what I am trying to say is it is more than simply replacing a 737 with 787 and call it a win.
“Alaska *can not grow* in Seattle anymore. Seattle is *full.* ”
But wait – we have all those lovely D Annex Bus Gates to use up again (URGH!)
1. Scott Kirby loves to act as if he invented any concept but multiple airline CEOs and analysts have tracked the disconnect between GDP and air travel expense throughout the pandemic. Delta has multiple slides in its Dec 14, 2022 investor update on the subject. The point airline CEOs are trying to make is that air travel spending will return but it still isn’t at all clear how much high-value business demand will return and how much other types of demand will take its place.
2. Kirby’s message during the earnings call was that other airlines could not adapt to the reality of the new operational reliability but United could – and yet United’s cancellation rate for several months (including regional carriers which Kirby was happy to exclude) is higher than other carriers. He said that other airlines would suffer similar problems as Southwest ‘s December mess and yet there is no evidence of that.
3. Regarding aircraft orders and deliveries, UA has more aircraft on order and is thus more subject to delays because UA has the oldest fleet among US airlines and waited to replace aging widebodies and swap regional jets for mainline aircraft, strategies that other airlines began years ago. Other airlines do have orders and options in place and the expected mega-order from Air India says there is still a lot of slack in the production pipeline for large orders. The MAX 10, a major part of UA’s orderbook is still not certified.
4. Scott Kirby touted a year ago that United would lead the industry with a new pilot contract both in timing and wealth – and yet most of the smaller low cost carriers have already signed new pilot contracts or have tentative agreements while Delta’s contract which is being voted on is worth 2X as much per year as UA’s. AA, UA and WN are the laggards on new pilot contracts while DL shook the industry during the pandemic by adding boarding pay for flight attendants, something no other large US airline does. The pinch point in 2023 will be labor and potential discontent and UA is potentially in worse shape than other airlines.
5. Despite what SK states, ATC capacity is coming back as fast as airlines’ ability to add flights. The pinch points are still UA hubs including EWR, the only non-slot controlled airport in NYC where UA has had to cut capacity in order to improve on-time performance even though the airport is still the most delay-prone as well as DEN, which is exploding with growth and has the usual IROP challenges of any airport that DEN avoided for years.
United and the rest of the industry will be just fine, United won’t have the advantage that Kirby thinks, and UA has a lot of unique challenges in 2023 which other airlines already addressed, some years ago.
Regional aircraft should be up-gauged and insourced. Mainline trunk routes should have additional a330 or 777 service. While ATC may be at capacity, the system is operating much better with less regional aircraft frequency.
Sad day for RAA lobbyists, the ting yang has turn on them and all sin waves point to mainline pilots holding strong on scope. Delta’s added a220 orders and United’s orders will add further regional outsourcing pressure.
Oh and don’t forget that a national seniority list would increase pilot productivity by 6%, but who wants to do more with less when airlines can make competitors suffer.
What pilots are going to go for a national seniority list?
And each airline is different and cockpits are different.
The benefit should be limited only to new hires of DAL, UAL, AA, JetBlue, Alaska & Hawaiian. It could operate more as a mutual bid exchange than a seniority list. No Pilot could change lists without another pilot wanting to swap with them. Both would loose a few hundred seniority number and pay for half their training cost. Tremendous fuel savings could be realized by eliminating commuting while increasing commuter pilot productivity. This would create added value for other pilots and the companies involved.
You may not realize it, but this is the same kind of thinking that people do when advocating an airport in the middle of nowhere.
Bigger planes take mean longer wait times at the counter, security, and for bags.
You start adding that up and making business travelers lose 2-3 hours a flight and the costs start adding up.
I didn’t realize. Figuring out what causes people to possess ideas or to be possessed by them is always an interesting subject.
I assumed I was more projecting many of Delta’s insourcing decisions into the comment.
Kirby shrugs off dramatically increased labor/ operation costs and monetary inflation. Eventually, domestic travelers will put away the credit cards which looks to be happening already.
This explains so much about WestJet.
Of course, Mr. Kirby’s line of thinking assumes that individual airlines will continue doing what they have always done, which is not a very good business assumption, go forward plan or business plan, and a recipe for disaster.
From my perspective, the airline industry’s future is not about pilots or cost convergence, but about Operational Excellence driven by true innovation and real competition.
For example, if just one airline breaks their 1950s, “day of” Operational Dismality, where they unnecessarily accept delivering 30% of their customers late (A0), at a huge annual expense (10% of total revenues), Mr. Kirby’s line of thinking goes out the window.
Consider a 1998, fortune Magazine quote by Adrian J. Slywotzky which sums up the issue nicely, “If you are not rethinking your business model every five years, you are running your business at a very high level of risk.” Mr. Slywotzky is an American consultant and the author of several books on economic theory and management.
Clearly, Mr. Kirby, or any airline today, is not “rethinking” their business model, and hasn’t for decades.
And the result is that a single large airline with 4,000 flights/day unnecessarily wastes upwards of $5 Billion annually, of which 20% is easily recoverable ($1.077 Billion). And I believe that this is a conservative analysis.
Additionally, based on the current block/gate time buffers in a large airline’s schedule, that airline wastes 103 aircraft, 1,333 pilots, 363 million gallons of fuel, generating 24 million tons of CO2. And again, around 20% can be easily recovered/prevented.