How Airlines Generate “Float”

Recently, I’ve been spending quite a bit of time pondering airlines. Perhaps out of concern that someone could violently drag me down United’s tiny aisles! Good thing I don’t fly United…but I kid, of course.

What has caught my attention is Buffett. He now owns approximately $10 billion of airline stocks. And for anyone that has followed Buffett over the years, this is unusual not least because he’s never minced words with respect to airline industry economics:

“The worst sort of business is one that grows rapidly, requires significant capital to engender the growth, then earns little or no money. Think airlines. Here, a durable competitive advantage has proven elusive since the days of the Wright brothers. Indeed, if a farsighted capitalist had been present at Kitty Hawk, he would have done his successors a huge favor by shooting Orville down.” -Buffett

Following the disclosure of his airline investments at the end of 2016, here’s what Buffett had to say:

“It’s true that the airlines had a bad 20th century. They’re like the Chicago Cubs. And they got that bad century out of the way, I hope…The hope is they will keep orders in reasonable relationship to potential demand.” –Buffett on CNBC

Buffett’s thesis appears to be industry rationality (finally!) now that the industry has consolidated down to 4 major airlines.

Although that could very well be the thesis, it just didn’t feel like the full story to me. After all, memory chips is a consolidating industry, why not own memory chips? Oil and gas services is a consolidating industry, why not own oil and gas service providers? This last comparison is even more compelling when you consider that a company like Schlumberger has much more differentiated technology than peers, whereas every airline offers essentially the same product.

I believe the critical insight is in recognizing that the airline industry now generates a significant amount of “float” and evolution of the airline business model has made it much more similar to an insurance operation. 

Modern Airline Business Model

To understand why, we need to understand how airlines make money today. By chance, I came across the excellent Bloomberg article, Airlines Now Make More Money Selling Miles Than Seatswhich makes the following points:

  • Airlines now make more money selling miles than seats.
  • Many customers carry airline branded cards that earn miles. Banks are responsible for rewarding those miles and buy those miles from airlines at 1.5-2.5 cents each.
  • Revenue from mile sales is estimated to be up to 3x the final cost to an airline (e.g. margins of 66%).
  • Miles are earned and paid for far in advance before redemption. 

“Float”-ing Miles in the Air – Similarities to Insurance

The last point in the list above is key – by virtue of earning the vast majority of income from selling miles that won’t be redeemed for months or even years, airlines have essentially created float.

In insurance, an insurance company receives payment upfront with the promise to pay in the future upon a contingency. In the airline industry, airlines now receive the vast majority of income upfront with the promise to dispense seats/services in the future. That is the essence of float.

In insurance, an insurance company can generate an underwriting profit if premiums received exceed payouts. Similarly, airlines can generate an “underwriting profit” if miles expire, if customers forget to redeem miles, or even if customers develop the habit of accruing miles far faster than they redeem (which is very likely the case).

Airline Float Better than Insurance Float?

Risking the potential that I’ve taken this idea to far, I do wonder if airline float could even better than insurance float:

  • Insurance float is not costless on average since the industry tends to generate underwriting losses with the intention of making it up via investments. Airlines do not appear to suffer from this underwriting tragedy.
  • Insurance contingent liabilities are fairly fixed (e.g. you know the value of the house insured or car insured or life insured), but the value of a mile can change over time (e.g. if necessary, an airline can significantly devalue miles).
  • Monoline insurance leads to exposure to a very specific sector/area (e.g. housing, healthcare, autos), while airline miles are generated via card spending, which is diversified across a number of industries (but most likely concentrated on high frequency, day-to-day spend).

The points above are likely oversimplified, but I believe there are merits to the argument. Taking the miles business model into mind and similarities to insurance “float”, airlines as an investment would appear to be much more Buffett-like. 

 

Berkshire’s Performance Edge that No One Talks About

Over the last half century, Buffett has achieved a marvelous long-term performance record. Many have studied and written about his investments to try to learn how to replicate what he did.

I’ve read a lot of those, and you probably have, too. But I’ve always felt that something was missing. 

Consensus has generally coalesced around a couple of observations on how he’s done it:

  • Buffett focuses on high-quality businesses (but will settle with a partial interest via public equity if he can’t own the whole thing)…
  • Purchased at fair (or cheap) valuations
  • And never letting cash burn a hole in his pocket as he waits patiently for these opportunities

He also appears to have a preference, either by choice or happenstance, for stocks that generally exhibit lower beta.

All of these elements are widely known and well understood (whether it is practice or not is a different story).

More recently, there’s been a greater focus on Buffett’s use of leverage. Since the core of Berkshire is composed of insurance companies, Buffett has benefited from the embedded leverage of those operations. This leverage is generated through the float that Buffett oft sings the praises of.

One of the most detailed (academic) treatments is Buffett’s Alpha by Frazzini, et al. where the authors estimate an embedded leverage ratio of 1.6:1. Through this leverage, Berkshire has been able to take low-beta, high-quality compounding businesses with good returns and generate great returns.

But this has always struck me as too simplistic. After all, many have tried leverage before to terrible ends. Many hedge funds today also have (re)insurance businesses to generate float, but these operations only share conceptual kinship rather than the true underlying essence of what makes Berkshire’s insurance operations great.

The Missing Ingredient?

“History tells us that leverage all too often produces zeroes, even when it is employed by very smart people.” – Berkshire Letter to Shareholders, 2011

“If you don’t have leverage, you don’t get into trouble. That’s the only way a smart person can go broke. I’ve always said if you’re smart you don’t need it and if you’re dumb you shouldn’t be using it.” – Buffett remarks to Financial Crisis Inquiry Commission, 2010

Seems to me the key for Buffett has been taking on leverage while avoiding the downside of leverage. And Berkshire’s insurance core has been the perfect structure to mitigate leverage risks.

There are 2 major problems with traditional leverage: 1) Capital calls at an inopportune time, and 2) Magnification of losses forcing investors to exit positions prematurely.

At Berkshire, float is generated through insurance and reinsurance premiums tied to autos and, generally, “catastrophes”. It’s clear that leverage generated through these avenues are likely to exhibit lower correlation to the general business cycle. People are generally more reluctant to default on autos than on homes, and the timing of major natural disasters are independent of recessions. Not only do these qualities limit the potential of inopportune capital calls, it likely ensures the availability of capital at an opportune time – the bottom of a cycle / recession. 

One can argue that the risk of leverage has not been eliminated, merely offset from the business cycle. After all, he will eventually need to pay out claims. Although this is true, Berkshire’s insurance businesses not only earn underwriting profits, they also generate more float/premiums than they pay out in claims in a majority of years (i.e. cost-free float and negative working capital). With growing float/premiums, losses can be backed with future premiums. The value of this advantage cannot be understated – imagine what you could do with leverage if you knew that it would never have to be paid back or know with high assurance that you can pay it back with future cashflows. This increases what looks like limited duration capital (say 3-5 years for auto float) to far-longer (dare I say permanent?) duration capital.

How about magnification of losses?  At face value, the growing mix of private / full-ownership businesses at Berkshire limit the need to mark-to-market. This materially lowers the risk of having a noxious combo of leverage and paper losses. But, this is a trickier thing, which I am inclined to think there are other secrets that I have yet to understand.