Streetwise Professor

March 21, 2016

The Seen and the Unseen, Hedging Edition (With a Bonus Explanation of Why Airlines Hate Speculators)

Filed under: Commodities,Derivatives,Economics,Energy,Politics,Regulation — The Professor @ 8:16 pm

Most media coverage of hedging is appalling. It tends to focus on the accounting, and not the economics. Unfortunately, managements and analysts too often fall into the same trap.

This WSJ article about hedging by airlines is a case in point:

After decades of spending billions of dollars to hedge against rising fuel costs, more airlines, including some of the world’s largest, are backing off after getting burned by low oil prices.

When oil prices were rising, hedging often paid off for the airlines, helping them reduce their exposure to higher fuel costs. But the speed of the 58% plunge in oil prices since mid-2014 caught the industry by surprise and turned some hedges into big money losers.

Last year, Delta Air Lines Inc., the nation’s No. 2 airline by traffic, racked up hedging losses of $2.3 billion, while United Continental Holdings Inc., the No. 3 carrier, lost $960 million on its bets.

Meanwhile, No. 1-ranked American Airlines Group Inc., which abandoned hedging in 2014, enjoyed cheaper fuel costs than many of its rivals as a result. “Hedging is a rigged game that enriches Wall Street,” said Scott Kirby, the airline’s president, said in an interview.

Now, much of the rest of the industry is rethinking the costly strategy of using complex derivatives to lock in fuel costs, airlines’ second-largest expense after labor.

Roughly speaking, hedgers “lose”–that is, their derivatives positions lose money–about half the time. If the hedge is done properly, that “loss” will be offset by a gain somewhere else on the income statement or balance sheet. The problem is, it’s not identified specifically. In the case of airlines, it shows up as a lower cost of goods sold (fuel expense), but it isn’t identified specifically.

The tendency is to evaluate the wisdom of hedging ex post. But you cannot evaluate hedging that way. You hedge because you don’t know which way prices will go, and because a price move in one direction hurts you more than a price move in the opposite direction of the same magnitude helps you. If you knew which way prices were going to go, you wouldn’t need to hedge.

That is, hedging is valuable for an airline if reducing the variability of profits attributable to fuel cost changes raises average profits. How can this happen? One way is bankruptcy costs. If an airline loses $1 billion due to a fuel price spike, it may go bankrupt, and incur the non-trivial costs associated with bankruptcy: this is a deadweight loss. The airline receives no bonus equivalent in magnitude to bankruptcy costs if it gains $1 billion due a fuel price decline. Therefore, reducing the variability of fuel prices reduces the expected deadweight losses (in this case, expected bankruptcy costs), which is beneficial to shareholders and bondholders.

As another example, an airline that becomes more highly leveraged because of an adverse fuel price movement may underinvest (relative to what an unleveraged firm would) due to “debt overhang”: it underinvests because when it is highly leveraged the benefits of investment accrue to bondholders rather than shareholders. Again, there is unlikely to be a symmetric gain when the company becomes unexpectedly less leveraged due to a favorable fuel price movement. Here, reducing variability reduces the expected losses due to underinvestment.

Hedging can also reduce the costs of providing incentives to management through tying pay to performance. Hedging reduces a source of variability in performance that is outside of managers’ control: since they are risk averse they demand compensation for bearing this risk, so hedging it reduces compensation costs, and makes it cheaper to tie pay and performance.

The problem is, none of these things show up on accounting statements with the clarity of a 9 or 10 figure loss on a derivatives position put on as a hedge. The true gains from hedging are often unseen. The true gains are the disasters avoided that would have occurred in the absence of a hedge. There’s no line for that in the financial statements.

The one saving grace of the WSJ article is that it does mention a relevant consideration in passing, but doesn’t understand its full importance:

Another factor in the hedging pullback: a round of megamergers, capacity cuts and more fuel-efficient aircraft have fattened the industry’s profits, leaving carriers in better financial shape—and less vulnerable to a spike in fuel prices.

Two of the factors that make hedging value-enhance that I mentioned before (bankruptcy costs and underinvestment) are more relevant for highly leveraged firms that are at risk of financial distress. Due to the factors mentioned in foregoing quote, airlines have become less financially distressed, and need to hedge less. But that should have been the focus of the article, rather than the losses on previously undertaken hedges.

And that should be what is driving airlines’ decisions to hedge, although the statement of American Airlines’ president Kirby doesn’t provide much confidence that that is the case, at least insofar as AA is concerned.

Airlines are interesting because they have historically been among the biggest long hedgers in the energy market. This is true because they are one major consumer of fuel that (a) cannot pass on (in the short run, anyways) a large fraction of fuel price increases, and (b) are big enough to make justify incurring the non-trivial fixed costs associated with hedging.

Fuel costs are determined by an airline’s routes and schedule, and fuel consumption is therefore fixed in the short to medium term because an airline cannot expand or contract its schedule willy-nilly, or adjust its aircraft fleet in the short run. Thus, fuel is a fixed cost in the short to medium term. Furthermore, the schedule and the existing fleet determine the supply of seats, and hence (given demand) fares. Since supply and hence fares won’t change in the short to medium term if fuel prices rise or fall, airlines can’t pass on fuel price shocks through higher or lower fares, and hence these price shocks go straight to the bottom line. That increases the benefits for financial hedging: airlines have no self-hedges for fuel prices.

This is to be contrasted to, say, oil refiners. Refiners are able to pass on the bulk of oil price changes via product price changes: pass through provides a self-hedge. Yes, crack spreads contract some when oil prices rise (higher prices->lower consumption->lower utilization->lower margins), but refiners are able to shift most of the crude price changes onto downstream consumers. This reduces the need for financial hedges.

Further, many downstream consumers–gasoline consumers like you and me, for instance–don’t consume in a scale sufficient to justify incurring the fixed costs of managing our exposure to gasoline price changes. Therefore, a large fraction of those who are hurt by rises in the flat price of energy don’t benefit from financial hedging.

Conversely, those hurt by falls in flat prices, firms like oil producers and holders of oil inventories, don’t have self-hedges: they are directly exposed to flat prices. Moreover, they are big enough to find it worthwhile to incur the fixed cost of implementing a hedging program.

This leads to an asymmetry between long and short hedging, which is evident in CFTC commitment of traders data for oil. This asymmetry is why long speculators are essential in these markets. Without long speculators, the (predominant) short hedgers would have no one to take the risk they want to get rid of. This would put downward pressure on futures prices, and increase the risk premium embedded in futures prices.

Which is why airlines have been in the forefront of those hating on speculators. Not because speculators distort prices. But because long speculators compete with long hedgers like airlines to take the other side of short hedgers like oil producers and traders holding oil inventories. This competition reduces the risk premium in futures prices.

This makes it costlier for airlines to hedge, but their higher costs are more than offset by lower hedging costs for producers, stockholders, and other short hedgers. This is why speculators are vital to the commodity markets, and thereby raise prices for producers and reduce costs for consumers.

But apparently this is totally lost on Elizabeth Warren and her ilk. But as the WSJ article shows, ignorance about hedging–and hence about the benefits of speculation–is widespread. Unless and until this ignorance is reduced substantially, policy debates will generate much more heat than light.

Print Friendly, PDF & Email


  1. But apparently this is totally lost on Elizabeth Warren and her ilk.

    Perhaps it would be less time consuming to compile a list of subjects that were not?

    Comment by Tim Newman — March 22, 2016 @ 3:48 am

  2. Good piece, prof. The point about asymmetry of big impacts isn’t one I’d thought about before.

    Another bonus to hedging, regardless of its outcome, is that it makes airlines’ forward cashflows more predictable, by fixing another variable. Lease costs are fixable, staff costs are fixable, FX risk is hedgable and fuel costs are hedgable, so pretty much the only variable left is seat sales. This being so, airlines can give their lenders forward indications of what they expect the P and L to look like which should be pretty accurate. This makes them a more reliable credit risk, which shaves a few basis points off the cost of borrowing.

    Most airlines historically weren’t interested in hedging programs that purported to shave a few cents off fuel cost, for this sort of reason. Having an out-there hedging policy compared to other airlines simply introduces a new variable into the mix that’s unrelated to what being a successful airline is about. So they are mostly interesting in doing whatever other airlines are doing, regardless of whether that’s smart.

    Comment by Green As Grass — March 22, 2016 @ 4:52 am

  3. @Tim-LOL. It might be more of a challenge, because I’ll be damned if I can think of a single one.

    The ProfessorComment by The Professor — March 22, 2016 @ 8:33 am

  4. @Green-Thanks. Re asymmetry . . . it’s an implication of Jensen’s Inequality. For a reduction in risk in a particular variable (e.g., fuel prices) to increase average payoff, profit must be a concave function of that variable.

    My list of hedging benefits wasn’t exhaustive. What you mention is definitely one of them. Hedging increases debt capacity. The predictability also plays into performance measurement/compensation. You can budget more accurately if you hedge, and benchmark performance vs. budget.

    Unfortunately, you’d think that managements would figure out that hedging allows them (and investors) to focus on doing what an airline is about. But you put your finger on an important point. There is a game theoretic aspect to hedging here. Doing what everybody else does is a way of avoiding being singled out esp. by the ignorant (who are many, unfortunately). “Don’t blame me! Everybody else does it (or doesn’t)!” Is a classic management fudge.

    The ProfessorComment by The Professor — March 22, 2016 @ 9:13 am

  5. Isn’t much of airlines current profitability driven by low fuel costs, which have yet to flow through to consumers in the form of lower ticket prices?

    In other words, it seems like airlines reduce hedging in “good” states of the world when they have a sizeable equity cushion / low oil prices, but increase hedging in “bad” states of the world, when their equity cushion has eroded. As you say, this might be rational due to high distress costs in “bad” states of the world. However, by pursuing this trading strategy, they are effectively buying high and selling low repeatedly throughout the oil cycle.

    There might be other reasons than oil prices for airline profitability swings (market power issues come to mind). Nonetheless, if oil prices are a key determinant of profitability, it seems wildly irrational to pursue a dynamic hedging strategy based on the likelihood of financial distress.

    Comment by Jeff — March 22, 2016 @ 9:23 am

  6. The current drop in prices has revealed another asymmetry in the airlines’ exposure: the asymmetry of their ability to pass through costs. You might not have noticed but airline tickets are very cheap now (no one yells about that!). This is because some airlines did not hedge and have brought down prices in line with costs. If you were one of the airlines that did hedge, you are not going to sell any tickets if you keep your prices in line with your costs. If all of your competitors had hedged this would not be a problem, but inevitably at least some of them will not hedge.

    It is possible that they are able to capture additional profits on an increase in oil prices, but I suspect not at the same rate at which they lose in a competitive pricing environment on the way down.

    I suspect this observation is accepted as fact at the airlines and makes them much more nervous about hedging lately. Despite the bumper profits at airlines right now, there are plenty of people who are concerned about the revenue per seat-mile.

    Obviously if you knew the distribution of risk you could come up with a tailored options strategy, but the risk premium on those is much higher. Also, the chance that the bank you deal with finds a clever way to make this much more expensive through understanding the transaction better is basically 100%.

    Comment by Brian — March 22, 2016 @ 12:36 pm

  7. Never a dull moment, Prof.

    Firstly, let’s have a look at ‘speculation’ which I agree with you is essential to market liquidity. My definition of a speculator would be an ‘active’ investor who puts his ($ etc) capital at risk in pursuance of a ($ etc) transaction profit.

    To do so the speculator buys and sells, but not necessarily in that order, and I would submit that a speculator’s time horizons tend to be short to medium term. So while long speculation is indeed competing with airlines hedging, this is a net figure and relatively short term.

    IMHO more significant than the ‘speculator’ – but not distinguished from them by regulators or most others – are the ‘passive’ investors who are ‘long only’ and invest through eg ETFs and Index funds. These risk averse investors are not aiming to increase their capital for a dollar profit: they rather aim to preserve their capital by avoiding a dollar loss. Or another way of putting it is that they ‘hedge inflation’ (to use Goldman’s brilliant meme) by off loading dollar risk and taking on oil price risk. Moreover, as I understand it, the funds involved will invest – via intermediaries or otherwise – as far down the curve as possible, and with interesting consequences.

    Second point is that of the ‘physical hedge’ aka market verticals. Here we see the likes of Emirates, Etihad and Qatar Airways off-setting increased jet fuel costs through increased profits from increased crude prices and vice versa. Not dissimilar from the way that the Saudis are protecting oil market share through buying/participating in refineries eg Motiva.

    I think we can expect to see a lot more of this: certainly Iran is studying closely the extension of existing geographic swaps/virtual pipelines (eg Caspian oil into Northern Iran swapped for oil out of the Persian Gulf): or category swaps such as Iran’s supply of gas to Armenia in exchange for power.

    There are plenty more variants and hybrid swaps available.

    Comment by Chris Cook — March 22, 2016 @ 3:25 pm

  8. You all make it seem complicated but it’s not.
    Customers buy airline tickets in advance and absolutely HATE fuel surcharges.
    To satisfy customers airlines buy fuel in advance.
    They are therefore hedged.
    To pay for certainty they must pay a risk premium, i.e. insurance.
    Because the market is deep and liquid this premium is low, but will over multiple hedges incur small losses.
    Airlines buy insurance from Wall Street. OMG!

    I look forward to Liarwatha telling everyone not to insure their houses because insurance companies make a profit.

    Comment by James Harries — March 22, 2016 @ 3:47 pm

  9. @ Chris Cook

    ‘Speculator’ versus ‘passive’ investor is a distinction without a difference.

    If you look at CoT data for say CME WTI you can see that Managed Money longs are 31% of the open interest. Other is 30.0, Swap Dealer 26.2, Commercials 8.7%. So commercial players are the smallest reportable category and ‘speculators’ the largest. Airline hedgers would be somewhere in the Swap Dealer figure.

    Hedgies clearly aren’t investing very far down the curve. They are 31% of the OI and there isn’t enough OI in the back months for all of it or even very much of it to be there.

    Comment by Green As Grass — March 23, 2016 @ 8:32 am

  10. Elizabeth Warren came out of left field in the last paragraph there.

    Comment by Noumenon72 — March 23, 2016 @ 11:07 am

  11. @Nourmenon72-Warren not out of left field at all. Warren has been going after me personally on the speculation issue, and in two letters to the CFTC has quoted from the NYT article as the basis for her claim that I am hopelessly biased.

    The ProfessorComment by The Professor — March 23, 2016 @ 12:44 pm

  12. I imagine the hedging also has to competition from other airlines than savings costs. If you have five identical airliners and all but one hedge, the one that is wrong will go out of business. If the are all wrong at once then they can all pass the costs to the consumer.

    Comment by grey enlightenment — March 23, 2016 @ 5:47 pm

  13. For a more skeptical take on airline fuel hedging, see here:

    The analysis isn’t really different, but the question is really how much default risk a given firm faces. If it is far from bankruptcy it is hard to see any benefit from hedging. OTOH, there is some empirical analysis by Javier Gimeno suggesting that airlines’ pricing behavior is affected by the presence of fuel hedges, even though opportunity cost probably ought to dictate ignoring them.

    Comment by srp — March 31, 2016 @ 8:47 pm

  14. @srp-Hedging is at its root a capital structure choice. It reduces the cost of debt, and so is particularly valuable for highly leveraged firms as you note. Even if a firm is not leveraged now, but it is at risk to underinvestment problems if it becomes much more leveraged due to an adverse shock (e.g., a big increase in the price of an input that it cannot pass on to customers) hedging can also be valuable.

    Airlines have substantial operational leverage because most of their costs are fixed in the short to medium term. Therefore, even a modestly leveraged airline might benefit from hedging. This would be particularly true of a growing airline, like Southwest.

    The connection between pricing behavior and fuel hedging is interesting, and puzzling. Yes, it should be opportunity costs that matter. Further, in the short to medium term, since fuel is largely a fixed cost, fuel prices shouldn’t affect prices much anyways.

    The ProfessorComment by The Professor — April 2, 2016 @ 4:22 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress