On June 10, 2008 Robert
Crandall gave a speech to the
Wings Club, which advertises itself as "the premier global society of aviation professionals and the leading forum on aviation issues." The event was widely advertised and as I understand it, covered by major media outlets, including National Public Radio.
A
video of the speech and the text of his speech can be found at The Wings Club web site.
It's an interesting read. What's even more interesting is how Mr.
Crandall is billed. Note this screen shot of the The Wings Club home page.

Mr
Crandall is advertised as the "Former Chairman, President, and CEO of American Airlines." This will become important later on. But note for now that every single major media outlet coverage of the Mr.
Crandall's speech describes him this way.
Mr.
Crandall is long know for his opposition to deregulation of the airlines industry. According to
Wikipedia, Mr.
Crandall was a vehement opponent of airline deregulation and actively opposed the 1978 Airline Deregulation Act. In his most recent speech Mr.
Crandall is still opposed to deregulation and is now calling for re-regulation.
Read the speech for yourself. His remarks are full of hand-wringing about the financial state of the airline industry, citing how many airlines have come and gone since deregulation and how much money airline investors have lost. He calls for us to think of airlines not as companies, but as public utilities that need to be managed by the federal government. And he speaks very negatively about the United State's capabilities, "The United States used to be good at solving problems. These days, we don't seem up to the job."
I give Mr.
Crandall credit for not blaming the airlines' problems on rising oil prices. He correctly notes, IMHO, that the airlines were suffering their current malaise well before the recent spike in oil prices.
Interestingly, he also is quick to argue against the nationalization of airlines.
"If consolidation were really the answer, it is conceivable that the system could be run by a single efficient operator. However, consumers clearly benefit from the existence of multiple airlines; the absence of alternatives does not encourage good customer service. Thus, our goal should be to harness competition and regulation to create a system responsive to both the imperative of efficiency and the desirability of decent service."
So Mr.
Crandall wants a combination of competition and regulation. Which parts should be competitive and which parts should be regulated? To quote from his speech, he outlines what he believes are the three broad fundamental problems in the airline industry:
- "First and foremost, we have failed to confront the reality that unfettered competition just doesn't work very well in certain industries, as amply demonstrated by our airline experience and by the adverse outcomes associated with various state efforts to deregulate electricity rates. In my view, it is time to acknowledge that airlines look and are more like utilities than ordinary businesses.
- Second, our government has failed to develop a national transportation plan of any kind and has thus been indifferent to the continuing decline of our highways, our railroads and our airlines.
- And third, the government has failed to invest in the capabilities and resources which only it can provide, most notably by failing to implement the new air traffic control system that everyone agrees we desperately need."
Lets' look at these in reverse order.
He says that the government has failed to invest in the capabilities and resources which only it can provide, notably the air traffic control system that "everyone agrees we desperately need."
Fair enough. For better or worse, when the airline industry was in its infancy, the FAA was created and one of its responsibilities is to create, manage, and run the air traffic control system. That's the way it's been since the beginning and it's not likely to change in the foreseeable future. The air traffic control system effectively limits the capacity of the industry. It governs the flight routes, and how many planes can take off at any given time. Ultimately this creates a limit on the number of airplane seats there are for any given time period. Mr.
Crandall doesn't specify what types of investment are needed in the air traffic control system. Does the air traffic control system have an availability problem? In other words, is the industry suffering because the air traffic control system goes down too often and prevents airlines from making their flights, effectively taking airplane seats out of the air? While there have been some highly publicized cases of such events in the past few years, I find it difficult to believe that all of the airline's financial woes can be blamed on that. Does the FAA have a safety problem? Has there been mistakes in the air traffic control system that have made air flight significantly more risky? Not that I've heard. The only other issue that I can think to lay at the feet of the FAA is _capacity_. And I do recall seeing news reports about lack of capacity at airports and lack of capacity in the routes. A quick
internet search can yield lots of stories about lack of capacity in the air traffic control system and assessments of the air traffic control system like this
one show big increases on air traffic capacity while keeping costs and safety risks constant. So absent any other indicators from Mr.
Crandall, I have to assume his statement about failing to invest in the air traffic control system is about problems with capacity.
Now let's look at the second fundamental problem Mr.
Crandall outlines. "failed to develop a national transportation plan of any kind and has thus been indifferent to the continuing decline of our highways, our railroads and our airlines." I give Mr.
Crandall high marks for rhetorical flair. He dutifully lists highways, railroads, and airlines as things that have been neglected. Now let's play that Sesame Street game of "one of these things doesn't belong here." If you guessed airlines, you guess correctly. Why? Because the railroads, at least from a passenger service perspective, are already run by the Federal government. Highways, are run through a combination of federal programs and state programs. So their infrastructure is completely run by government. But airlines aren't. While it makes sense for the government to have a "plan" for infrastructure that it is responsible for. It doesn't make sense for the government to have a "plan" for an industry that it doesn't run. So by including it in the list, he implies that the decline of the airline industries are the responsibility of the government. As noted above, the government does run parts of the airline infrastructure. But it doesn't run the airline industry as a whole and therefore is not responsible for having a plan for the airline industry as a whole.
What does Mr.
Crandall think a "national transportation plan" should include?
- "First, to strengthen our national economy by encouraging the creation of a cost effective, energy sensitive transportation network which will permit people to move easily from one place to another
- "second, to assure safe, courteous and on time service for consumers
- "and finally, to improve the financial performance and international competitiveness of America's airlines.
Gee. How can you be opposed to any of that? I especially love the "cost effective, energy sensitive" part. Could we get any more vague? Can we get any more motherhood and apple pie? The question is what the feds should be doing to achieve these noble and lofty goals.
And now to take his first "problem". He states that we need to be treating airlines as "utilities" not as businesses. OK. What's the difference? Well, as far as I can see the two big differences are that with utilities, the governing authority says who can and can't operate. It sets standards for service. It sets prices. It controls how much profit or not the utility is allowed to make.
And indeed, these are the main issues that Mr.
Crandall wants to have addressed.
He calls for the government to reduce it's labor costs by giving it more negotiating power with labor unions. He says that "airplane seats" cannot be stockpiled and an airline has huge fixed capital costs. So strikes have a huge impact on airlines. He calls for the railway labor act to be amended to include airline labor. This would force the airline workers into binding arbitration rather so that organized labor would be "deprived of its ultimate weapon." We'll get back to this later, but for now I'd just point out that there are other industries that are in similar situations. Hotels for example. You can't stockpile hotel nights. So when their labor strikes, they are severely affected and they have huge capital costs that can't be quickly reduced in the event of a strike. So would Mr.
Crandall support amending the railway labor act to include hotel workers into binding arbitration and taking away their right to strike?
Mr.
Crandall also calls for the end of congestion at major airports. How. By outlawing it. He says,
"In the short term, the only solution is a regulatory mandate that limits the number of flights scheduled to what the runways, terminals and air traffic control facilities at a given airport can handle."
So every one has to give up an equal share of their traffic at congested airports. he does NOT call for the reduction of traffic at all airports. He just calls for the reduction of traffic at congested airports. For the sake of argument, I am going to make the claim that "congested" airports are synonymous with "major airports" that have a large number or flights, and in particular, I'm going to make a claim without supporting evidence that most, if not all airports that are major airports with congestion are also airports which are a major hub for one of the big airlines. This is an important claim because Mr.
Crandall has a lot to say about hub and spoke systems which I'll get to in a minute.
Next he calls for stricter financial standards for start up airlines. The reason for this is that
"In the years since deregulation, nearly 200 airlines have come and gone. These inadequately financed carriers, whose principal goal has often seemed limited to either lasting long enough to reap the rewards of an initial public offering or satisfying the ego of yet another would be airline mogul, have consistently cut prices to attract passengers. Such short-term antics have destabilized the pricing structure required by a healthy industry, and have offered no lasting benefit to anyone."
Apparently Mr.
Crandall doesn't consider price reductions for airline customers to be of any benefit to the public good. If you turn this around, the "stricter financial standards must therefore prevent start up carriers from offering prices that are lower in order to not jeopardize the health of the industry. So apparently the health of an airline is more important than the prices consumers pay. Who says? Sounds like the protectionist rhetoric of an industry insider to me. We'll have more to say about this later.
But Mr.
Crandall's most radical recommendations are that he wants price controls for the airline industry.
As he puts it:
It must now be clear to all that one of the industry's fundamental problems is the way in which it prices its product. As you all know, airlines work with a very distorted supply-demand equation. The instant perishability of empty seats, the impossibility of quickly reducing fixed and semi-variable costs when demand falters, the public's view that all airline seats are interchangeable commodities, the plethora of competitors and the desire to protect the reach of networks all create a great temptation to sustain volume by selling seats too cheaply."
So the problem, in his view, is that airlines are selling seats too cheaply. So guess whether or not he wants airline seats to be more or less expensive? He wants them to cost more, of course. And he wants government regulation to create more expensive airline seats. This is what anyone else would call a government subsidy. And the subsidy comes directly out of the pockets of air travelers.
Mr.
Crandall does a lot of panic mongering, implying that the drive to cost reduction has led to a myriad of safety and convenience issues. The conclusions he'd have you draw is that by forcing air travelers to pay more to subsidize the airline industry, they'd get more convenience and safety. But as the government air safety assessment linked to above clearly shows, there hasn't been a decrease in safety even during the recent years. As to inconvenience, I don't know. I'll say that every air traveler I have personal experience with has consistently chosen price over convenience. Maybe my friends are not typical.
Mr.
Crandall paints himself as a moderate by eschewing complete government controls of prices and then offers this relatively moderate proposal which he calls the "sum of local fares" model.
"Suppose for a moment that in a world where every airline set its own prices, a regulatory agency required that any passenger traveling between two points via a connecting point pay the sum of the local fares on his or her itinerary."
"Ask yourself this: if the non-stop fare from Detroit to Los Angeles is $450, why should a passenger be able to travel via a connection for the same amount or, as is the case today, for even less?"
Why indeed. Especially if you buy into Mr.
Crandall's assertion that
"As we have all known for many years, the cheapest way to carry a passenger from point a to point b is non stop, and the most efficient way to do it is by using the largest airplane compatible with demand."
I don't buy it. Sure. there are some costs that are linear in relation to the number of miles traveled. For example, the amount of jet fuel used is in direct proportion to the number of miles traveled. And if the only thing you look at is the costs that are directly related to the miles traveled then of course the direct routes are going to cost less than routes that require a going through a hub.
But surprise! There are in fact costs to airlines that are not directly related to miles travelled. In fact, the airline industries have done a lot of analysis of operating models and there is a large body of academic research that supports the idea that the hub and spoke model is the most cost efficient model for a transportation networks and that the savings in other costs more than offset the increased costs related to miles traveled. To get a feel for this start with the
Wikipedia article on "
Spoke-hub distribution paradigm." The benefits of the hub and spoke model include:
- For a network of n nodes, only n - 1 routes are necessary to connect all nodes; that is, the upper bound is n - 1, and the complexity is O(n). This compares favorably to the (n(n - 1))/2 routes, or O(n2), that would be required to connect each node to every other node in a point-to-point network.
- The small number of routes generally leads to more efficient use of transportation resources. For example, aircraft are more likely to fly at full capacity, and can often fly routes more than once a day.
- Complicated operations, such as package sorting and accounting, can be carried out at the hub, rather than at every node.
- Spokes are simple, and new ones can be created easily.
- Customers may find the network more intuitive. Scheduling is convenient for them since there are few routes, with frequent service.
The
Wikipedia article also notes some drawbacks to hub and spoke, but they do not relate to efficiency.
So when Mr.
Crandall says
"As we have all known for many years, the cheapest way to carry a passenger from point a to point b is non stop, and the most efficient way to do it is by using the largest airplane compatible with demand."
it is simply not true. The airline industry has known for decades that the most efficient way, when you factor in ALL of the relevant costs, of getting a passenger from A to B is NOT ALWAYS a direct flight and in fact more often than not it is actually cheaper, OVERALL, to send a passenger through a hub.
Mr.
Crandall ran American Airlines for many years and was a key player in major airlines before that. He should know the benefits of the hub and spoke model. Scratch that. I'll go so far to say that Mr.
Crandall MUST know the benefits of the hub and spoke model. It's inconceivable IMHO for a CEO of American Airlines to so completely misunderstand the economics of running a major airline.
At very least, he owes us a more complete explanation about why the benefits of the hub and spoke model don't apply to the airline industry when they apply to other forms of transportation and the principles of hub and spoke apply to other domains of study such as distributed computing.
So when he says "As we have all known for many years, the cheapest way to carry a passenger from point a to point b is non stop." I have to conclude that his is deliberately misleading his audience.
But why? Why would be make such an easily refuted claim in a public forum?
Why is Mr.
Crandall so against the hub and spoke model? Why does he want to use government regulation to destroy the economic advantage of the hub and spoke model? If the hub and spoke model is more efficient and the airlines can offer airplane seats through a hub that are cheaper than direct flights (when you factor in all the costs) why should the government use its regulatory power to eliminate the advantages of this efficiency and force passengers to pay more?
So let's recap. When you factor in all the costs of running an airline, the hub and spoke model is more efficient than having direct flights. In fact, it's common today for a passenger to spend less on a route through a hub than a direct flight. It's to the airline's economic advantage to do so, even factoring in the inconvenience to the passenger for having to travel longer. So the hub and spoke model _helps_ airlines stay economically viable.
Why would Mr.
Crandall be against it? Why would Mr.
Crandall arguer for the government to artificially raise prices of passengers going through the hubs in order to eliminate the cost efficiencies of the hub and spoke model? Why would Mr.
Crandall have an interest in making the direct routes more competitive to the hub and spoke routes? Why would Mr.
Crandall advocate that passengers should pay more, while at the same time eliminating the efficiencies of the hub and spoke model in favor of less efficient direct routes?
In five minutes of
internet searching, I could find at least one credible reason that Mr. Crandall would make these counter-intuitive recommendations. I found at least one credible reason why we should not trust Mr. Crandall's recommendations.
Mr. Crandall has a financial interest in destroying the hub and spoke model. In fact basically every single recommendation Mr. Crandall made has more to do with his financial self-interest than any interest in the welfare of the big airlines and the airline industry. In fact, Mr. Crandall has a financial interest in ensuring that the big airlines are less competitive.
In his speech, Mr. Crandall said, "As you all know, I am no longer active in airline management, and thus lack the sophisticated analytical tools I so enjoyed in years past." And remember the web site advertising Mr. Crandall as "Former Chairman, President, and CEO of American Airlines." This would imply that Mr. Crandall is no longer working in the airline industry.
That speech was delivered on June 10th, 2008. But as I write this article on July 10th, a month later. Mr. Crandall is in fact the Chairman and CEO of
POGO airlines. Check it out for yourself on the web site's
leadership page. And there is a
press release currently on the POGO web dated February 28th 2008. The press release announces the appointment of two new people to POGO's Board Of Directors and at the end of the press release the press release clearly states that "POGO’s management team is led by Robert Crandall, the former president, CEO and chairman of American Airlines. I don't know about you, but I am unable to reconcile this statement with Mr. Crandall's statement that "As you all know, I am no longer active in airline management.
So Mr. Crandall is at the most charitable interpretation, misleading the audience when he says he is "no longer active in airline management." Given the gravity with which his opinion is taken in the industry I'll go so far as to say it's unethical of him to not disclose his obvious financial interests in the airline industry in his speech.
But wait, it gets worse.
POGO airlines is not just any startup airlines. They have a unique strategy.
From the "
What is POGO?" page of the web site:
POGO is a development-stage company, not yet certified or offering service. We plan to become operational as a FAA-certified Part 135 operator by the first quarter of 2009.
Our idea is simple: to make on-demand, private air travel more affordable, easier to use, and more convenient.
Our plan is to use the low purchase price and operating costs of today’s Very Light Jets with intensive use of the planes to significantly lower the cost of private aviation. Our on-line booking system will permit customers to quickly arrange a flight that most closely fits their needs. And, by serving a vast network of underutilized small airports, POGO customers will be able to bypass congested hubs. In fact, our customers will be able to choose from over 400 airports in our initial service area."
Oh really? So by using the smaller, lighter airplanes and underutilized small airports, POGO's customers will be able to arrange a flight that more closely fits their needs. Is the web site implying that maybe customers are more likely to find a direct flight that bypasses the congested hubs? Cool! Customers will find it much more convenient to fly a POGO flight because its routes are more likely to be direct.
But what if, due to the hub and spoke model, the big airlines are able to actually offer flights between the small airports at a lower cost that POGO can due to it's reliance on direct routes? As the Wikipedia article shows, there are clear efficiency benefits to using the hub and spoke model. So what if the big airlines really can get a passenger from a to b cheaper than POGO can? Now the customer has a choice between the lower cost or more convenience. I don't have access to industry survey data, but I'd be willing to bet that, other things being equal, my friends would choose the lower price every time.
So what's a CEO to do? Here's an idea. Why not convince the government to force the big airlines at those congested hubs to peg their prices to the number of air miles traveled. That way there's NO WAY, enforced by government regulation, that big airlines can compete with POGO on price.
So it turns out the Mr. Crandall's "sum of the local fares" model has little to do with the health of the airline industry, little to do with global warming or the price of oil and all about making his new shiny airline more competitive at passenger expense.
And in fact, when you look at Mr. Crandall's speech in light of the knowledge that he was in fact the Chairman and CEO of an airline at the time he made his speech, every single recommendation he made makes much more sense in the light of his self interest in the success of his shiny new airline.
Let's go through some of Mr. Crandall's recommendations again.
As we've noted. The "sum of local fares" model of price controls he advocates will prevent the big airlines for getting passengers from a to be cheaper than his airlines can.
He calls for more investment in infrastructure, especially in the air traffic control system. Note that because his airline is not following the hub and spoke model, he'll have far more flights and routes. In fact they brag about using more and smaller planes. As Wikipedia notes, if he doesn't use the hub and spokemodel, the number of routes needed for his airline in crease with the square of the number of airports he serves. So you bet he's concerned with the capacity of the air traffic control system!
Mr. Crandall calls for the railway labor act to be amended to include airline labor so that airline labor can't strike and is forced into binding arbitration. Hmm, as CEO of POGO, do you think Mr. Crandall could foresee himself in labor negotiations in the future? Wouldn't it be nice if the government made the labor unions weaker for him?
Mr. Crandall calls for the government to use its power to eliminate the number of flights going into and out of "congested" airports. By the words of the POGO website, the congested airports are the hub airports. So if the government forcibly reduces the number of flights going into and out of hubs, and assuming the number or travelers remains more or less the same, won't that increase demand for POGOs routes, which bypass the hubs?
He argues against the nationalization of airlines, maybe because he's trying to start one himself.
I could go on, but that's enough for you to get the idea.
Maybe some of the recommendations make sense for the pubic and maybe they don't. But in my opinion, Mr. Crandall's speech to the Wings Club was unethical, misleading, and entirely self-serving and cannot be taken seriously as a honest policy analysis.