Hard hats on: buy AER, short BOC Aviation

Every now and then, when markets go haywire and in particular airline/travel stocks get crushed and my beloved Aercap (NYSE: AER) gets obliterated, I pull out my favorite chart. It shows the long-term trend in global air travel, and it looks like this:

Screenshot 2020-02-27 22.12.10

Doesn’t that chart look great? No matter what crisis has occurred – be it terrorism, economic calamities, pandemics – we all seem to continue to want to hop on a plane and go somewhere. Global air travel demand has basically doubled every 15 years, like clockwork, and even though it has slowed down or stopped growing for the odd year here and there (after 9/11, and post the GFC), this chart is basically inviolate. People will return to the skies – it is simply a matter of when, not if.

This is not going to be a Pollyannish expose of why you should rush in to buy airline and/or aircraft leasing stocks. If you think Covid-19 turns out to be the pandemic to end all pandemics – if this really is ‘the big one’ – then it will necessarily change human behavior permanently; the above chart will fail; and huge swathes of the global economy (namely most every levered airline, hotel, and restaurant chain) will go out of business. Airline lessors will of course be no exception.

But if you believe that – like most calamities to have befallen us – we will both find a solution; persevere; and then, eventually, if not immediately, return to normalcy – it is time to at least consider what to own for the rebound. After all, the Spanish Flu in 1918 killed an estimated 20-50mm people…but here we are today, still traveling more than we ever have, year after year.

Aercap today trades at ~63% of book value (probably <60% of book value once we open on Monday). That is, shareholder’s equity is ~$9.6bn right now yet it is available in the market at ~$6bn. Since this business does around $1.1bn of net income a year, the market seems to be saying both that a large portion of the existing book will be impaired – through credit losses as airline customers default – AND that go-forward earnings will be nowhere near $1.1bn a year (around a 12% RoE on book today).  I am not going to spend any time discussing theoretical forward returns and how this may compare with a theoretical cost of equity looking out 5 years. These questions are important, but they are not, to my mind, the reason why the stock is trading at a discount today. Instead the discount is being driven entirely by credit risk – that is, the market thinks many of their customers will default, impairing assets in the book today or near-term.

Stress-testing book value

OK, so let’s talk about aircraft defaults. A few things worth mentioning straight off the bat:

  • there were 23 airline bankruptcies in 2019 yet AER generated record profits. This is the airline industry – there will be always be failures – but the good lessors have devised ways to manage the risks inherent in the customer base;
  • If the lessor is any good they will maintain security and maintenance deposits covering a large portion if not all of the expected loss on default for troubled lessees. This is why – for its entire listed history – AER has not taken credit losses in excess of the maintenance deposits and security deposits that get swept into revenue when you do have a default/repossession event;
  • There are many cases where airlines default on their debt and yet lessors still get paid (see Alitalia, for example). This is intuitive: if you don’t pay your lessors they take the plane. For many carriers (especially flag carriers), maintaining routes (and thus employee jobs) can be political and thus keeping the planes is paramount (especially in a very tight market where being able to release a plane soon thereafter as a bankrupt airline may be tricky). Recall that the narrow body market was – until corona – incredibly tight and the OEMs had sold out backlogs for many years.

The key points really are that even in bankruptcies, it is no guarantee that leases will be rejected en masse; nor is it that losses will be taken (that has certainly not been the example over the recent past). This is the key difference between lessors and the airlines: the lessors are essentially senior secured creditors – they are the last to not get paid. Even employees will be asked to take haircuts before lessors (cf Cathay Pacific at the moment).

That is all very well and good, you say – but in the past you had isolated failures so planes could be recycled quickly into a firm global market. But coronavirus is a different beast: we are seeing a global meltdown in demand for air travel – there is no safe space here. Surely the risk of a system-wide collapse in multiple airlines is much higher now than in previous isolated cases?

The short answer is yes. The longer answer is, in a force majeure event such as this – and no one would really deny that Coronavirus is that – it seems quite likely that many of the airlines will be supported by governments (certainly this seems quite likely for all the flag carriers). It hardly makes sense for governments to allow their national carriers to become insolvent – and thus potentially lose their fleets – because of a virus that was impossible to predict or prepare for. My bet is – with interest rates near zero globally – governments just print money to fund fiscal stimulus programs – part of which will clearly devolve to support state-owned airlines.

So, I think its reasonable to assume most all the state-owned airlines are not at specific credit risk from this force majeure event. Right there that removes a large portion of AER’s fleet – perhaps 25% of gross aircraft assets (Air France/KLM, 5% of the book, is joint owned by two governments; another 2/3 of the China-exposed book is state-owned, that’s about 15% of total book; and the rest should contribute another 5%+).

OK, now let’s make some very generic – and punitive – assumptions about what happens to the rest (75%) of the non state-owned customer book. Let’s simply assume:

  • half (so 37% of gross aircraft assets) of the remaining customers default within a year;
  • half of this half rejects all their leases (call it 19% of gross assets);
  • loss given default on this half is 50% (above the top end of realized LGD rates that I can observe, looking back the last 10 yrs) – call it 10% then of gross aircraft assets.

Since AER gross aircraft assets are around $36bn, this would equate to a $3.6bn hit (all to equity clearly) – which is basically the discount to book implied in the stock today.

Of course, even in this crazily-punitive scenario, I think the actual hit would be far less, because:

  • before these customers default they are – theoretically – still paying rents, so AER would get some portion of its $1.1bn in annual profits accreted during the year (maybe half?)
  • the loss given default would likely only be realized if AER was forced to turn around and write new leases, immediately, at much lower rates than the returned aircraft (thereby forcing the impairment to be taken through equity).

This second point seems quite unreasonable to me. Here’s why. This is what the business has done in terms of PnL over the last 5 years:

Screenshot 2020-03-09 02.19.57

You can see I have X’d a fair few lines. I am trying to establish the cash costs of running AER as it stands on an annual basis – it looks like cash open is around $1.5-1.6bn (interest + SG&A), maybe a bit higher in reality. Revenues – from the leasing business only – run around $4.7bn, covering cash opex by roughly 3x.

In our punitive downside scenario, we speculated that ~20% of the customer base would default. Lets just assume its EVEN WORSE than this at the revenue line (some customers may ask for deferrals even if they intend to pay) – so lets bump this up to 40% of the customer base, AND assume again that this situation lasts for a whole year. Then, revenue falls to $2.8bn – still well in excess of cash opex. So no problem, right?

Not exactly. AER still has chunky plane delivery commitments, of which ~$3.5bn fall in 2020:

Screenshot 2020-03-09 03.04.51

If we assume there would be no deferral/extension of these deliveries (again, wildly pessimistic), AER would theoretically burn ~$2.3bn in cash in 2020 (haircut revenues less cash costs less cash out for new plane deliveries). And that’s before maturing debt of $3.5bn is taken into account (even though all AER’s debt trades above par and doesn’t seem to present any refinancing risk). This sounds extreme – until you realize that AER has ~$8bn+ in available liquidity – kept just for situations like this:

Screenshot 2020-03-09 03.10.19

In summary: even in an armageddon-type scenario where a huge portion of their customer base becomes highly distressed, and this situation persists for a whole year, it seems reasonable to suggest AER would not be a forced seller/lessor of assets at temporarily-distressed levels, as they could survive this situation AND be shut out of the debt markets completely for over a year. Meanwhile, I have trouble seeing how absent the end of travel as we know it, they take more than a ~35% haircut to current book. To me, that makes AER at current – with the best management and capital allocators in the business – an incredible value at current levels.

How do you hedge? BOC Aviation

But what if we’re wrong? What if no one ever hops on a plane again? Well, thankfully, Mr Market is still somehow pricing a fairly similar (though, in my view, inferior) business at a crazy relative price – BOC Aviation (2588.HK). This business has more leverage; more Asian and direct China exposure (>50% of book); a much larger order book relative to its own fleet (meaning, more new lease placement risk); and is much smaller (only ~300 owned planes), with chunky and risky exposures to lessees like Cathay Pacific, Norwegian, and Thai Air:

Screenshot 2020-03-09 03.17.18

While there are some other differences in fleet composition, age, and structure of the funding, and BOC is 70% owned by Bank of China, in the kind of general market meltdown that AER stock now discounts, it is impossible to see how BOC could protect its book value any better than the market leader. The fact that it still trades at 1.5x book value (!!) despite facing all of the same risks as a larger entity with more diversified funding and client base that currently trades at 0.6x book for a similar asset pool, still boggles my mind (and is why I’m short). If worst comes to worst, this spread must but narrow, significantly.

Disclosure: Long AER, short 2588.HK

25 thoughts on “Hard hats on: buy AER, short BOC Aviation

    • i think its a combination of things: 1) smallish float (30% of the co floats); 2) perceived BOC backing; 3) div yield (even if this is illusory in a crash); 3) asian investors asleep at the wheel

      • Good strategy to me, if I were Robert Martin, I would ask BOC for funding and buy AER to control it.

        From Asia
        PS. As least I m not asleep.

      • they don’t have the scale to do so even if AER was a willing seller (which they’re not). but they should be dumping their own shares and planes to buy AER shares in the open market, yes.

      • Bocavi has to raise 6-7bn to meet its commitment and roll their debt this year. Even so, bond yield of AER and Al is nearly double of bocavi. The boc tag will play a huge part in lessor competition looking forward.

  1. Great write up! Couple questions:
    1. AER traded as low as .3x in 2008. While fundamentals may not support it, sentiment may drive it down. Is there a reason to believe that sentiment on AER may not sour as bad as it did in 2008? Was there something unique about that liquidity shock that with certainty may not take place this time around?
    2. Seems AL and AER trade at similar PB, but AL has less leverage, slightly better returns, newer fleet. Why is AER the preferred lessor? Seems can make an argument both are well postioned if not AL being better?
    3. Smallish float and BOC backing will not ever change, nor will asian investors wake up until the dividend yield is cut. Barring a global recession (where the yield gets cut), any catalysts for why BOC u/p AER?


    • Hi there – thanks.
      1. Prices can always trade at any level – I am not forecasting that prices stabilize here (that is impossible to call) but simply that the discount to intrinsic value in my view is now extreme. However – there are many reasons to think this shouldn’t trade like it did in the GFC. Remember that then the global financial markets basically shut entirely – there was no financing available for any asset, let alone aircraft. Now, not only AER but the entire world is awash in liquidity. There is not a liquidity crisis or seizing up of the financial system – rather, just a massive demand shock to air travel/etc. AER’s specific position now is also MUCH stronger than then (much bigger, better rating, more diversified funding, less secured assets, infinitely more liquidity today, etc). So, I would think it shouldn’t trade like it did in 08/09.

      2. AL is a fine business, but it is smaller (ie less diversified); it is much more growthy (much larger order book), so there is a huge amount of risk in terms of being able to place the planes if there is a more secular decline for a longer period. I don’t dislike it but I prefer the biggest platform with the best record and most financial resources in a time like this.

      3. I think you are assuming the div yield won’t get cut. If the current environment continues for a quarter or two it will definitely get cut (at least by a prudent management team). There are plenty of Asian investors who lazily owned this for the yield and will likely start dumping when they see the plethora of disastrous headlines from Asian airlines, etc that comprise most of BOC’s book. Anyway I like it as a disaster hedge for this position given the valuation disparity.

  2. Hi thanks for this idea. How do you look at the risk that the Chinese government will support BOC and/or BOC aviation? You cite government support in favour of the long but could actually hit your short harder/faster …

    • its a good question. no doubt they will support them if they get into deep distress but look at how it’s priced – it is still above book value and paying a dividend…clearly a long way to fall before they go knocking on the Chinese door for support.

  3. Hi, thanks for the write up. Like the idea.
    How comfortable are you with the various covenants not being tripped on the revolver? I haven’t combed through the filings, but if they use the revolver on capex or to refinance debt, couldn’t the covenants around interest coverage, debt/EBITDA, very easily get tested?

    • the main risk is tripping the EBITDA covenant on the revolver (1.5x on an LTM basis). since this is an LTM test, earnings would have to crater for multiple quarters continually to come close to breaching. ie, they can’t just have one horrible 2Q – which they will have – and violate it. of course, at the end of the day I believe banks would waive this covenant if it come to that since they have nothing to gain by tipping AER and forcing them into a position where they are forced to sell planes into the market at the worst possible time.

      by that point, of course, I believe we will be almost back to normal (in terms of demand) and thus the crisis will have been ameliorated.

      • Where did you find the covenant ratios? While the filings state that AerCap’s revolvers contain covenants that require the company to maintain compliance with a maximum ratio of debt/equity, a minimum fixed charge coverage ratio, and a maximum ratio of unencumbered assets to certain financial indebtedness, I didn’t see anything in the financials saying what these minimums and maximums were.

      • you need to dig through the credit agreement on the revolver. there are a few other covs in there too which i neglected to mention, but frankly i think are less relevant as not at risk of being tripped (in my view) for a very long time

  4. Hi P,

    I appreciate your ideas. After the recent restrictions and draconian measures in several countries, did you have the time to adjst your investment case?
    many thanks

    • yes. obviously its not great for AER but the stock now (at 18) is being priced for imminent bankruptcy, in my view. that is simply insane given the liquidity picture here.

  5. Pingback: Will AerCap be a Multi-Bagger? - Stox

  6. Thanks for the write up. Have a question that is a bit irrelevant to this analysis. How did you size this trade? Is it beta neutral or dollar neutral?

    • hi Jared – I like the equity so by definition I like everything above it too 🙂 in all seriousness I don’t have enough information to analyze the specific structured debt products; but I clearly think the unsecured bonds trading in the teens yield are also an insane bargain and money good.

  7. Pingback: AerCap (AER) – Mike Sandrik

  8. Hi Jeremy,

    Thanks for the write-up.

    1. I was wondering if you changed your mind re potential scale of impairment now that we have a little bit more information. I’m sure you read Avolon’s statement that c.80% of their customers asked for a deferral representing c.90% of their annualised contracted rental cashflow (https://www.avolon.aero/newsroom-and-thoughts/avolon-q1-business-and-covid-19-update). I was wondering to what extent can deferrals (which, and please correct me if I’m wrong, on their own would not lead to impairment under US GAAP) morph into lease renegotiations that would result in asset impairment and lock Aercap into less favourable leases for many years to come. I understand that lease agreements are watertight from a legal perspective, but surely if a lot of Aercap’s customers have most of their fleet on the ground, are furloughing workers (who are also likely taking a pay cut) and slashing any discretionary opex/capex, they will attempt to renegotiate leases as well. If all stakeholders (likely including creditors) take some pain to save airlines, I’m not sure Aercap will come out of this unscathed.

    2. To get the current market cap and assuming 0.8x P/B you require c.14% impairment on Aercap’s total assets (15% on aircraft assets, no impairment on cash and 5-15% on the rest). And I understand that the picture is more nuanced (lower risk for state-backed + China + Korea part of the book and you need bankruptcy + lease rejection + high LGD), but this scale of impairment does not seem to be outside of the realm of possibility in the current environment. How do you get comfort (a high bar in this market!) on this point when historical data is to a large extent useless? (looking forward to ‘0% revenue for x months’ downside cases in the future though)

    • Many interesting points. 1) lease deferrals: most all their (solvent) customers will get 3mo deferrals, then (I believe) they will have to start paying or AER will probably move to take the planes. I say 3mos because for most clients (not all) AER has 3mos worth of rents held as security/maintenance deposits, so they are ‘covered’ to an extent. I don’t believe AER (or many other lessors) will take much pain before all the equity value is impaired, however (remember the lessors are senior secured, hell or high water, and have significant leverage over whether the airline can operate or not).

      2) renegotiations: this will of course happen but probably not out of deferrals and not in a way that threatens book value massively. They will likely allow renegotiations for lease extensions, perhaps at a lower rate – but for longer and more $$ (eg turning 2-3yr leases into 8-9yr leases, perhaps at a lower monthly rate). So, yes, the monthly lease factor goes down but for planes with 5-10yrs left of live that were to come off lease in a year or two, they lock up that time period with a – presumably – health counterparty. On the other hand, I expect AER to play EXTREME hardball with clients who look like they may file anyway (Norwegian; Virgin Australia; etc) and simply take the planes and warehouse them for a stronger market environment.

      3) book impairment: you are forgetting that AER has been selling planes in the market at 30% above their stated book value, for years. So, they would need to see more like a 25% haircut across all their planes – with no offsets, no gains on sale, no maintenance/security deposits, and no earnings in between – to justify the current price. Thats simply not happening. Remember that AER will only need to impair an asset when a lease is rejected (after bankruptcy or through mutual agreement) AND when it cannot be placed in the market at a level that protects the book value, in management’s judgement. I highly doubt management thinks the lease rate today is representative of true value and so I believe they will live with a lower utilization (that is, warehousing any planes that they are forced to repossess) for say 6-12mos before finding them a home, on reasonable terms. They are able to do this since they have ample liquidity, and because they are quite likely to get deferrals from all the OEMs given their importance as a customer to BA/AIR.

      hope that helps.

      • Thanks Jeremy, that’s helpful and I agree that there is more cushion than most people likely assume and Management is capable. In terms of book impairment per p.41 of the 2019 20-F, it seems that AER also performs quarterly impairment tests based on potential aircraft sale transactions or repossessions so we might get a very early indication of what they are seeing in the market at Q1.

        Airbus & Boeing slashing production rates should also help balance the market and preserve the value of aircraft in the longer term (helpful alignment of interests!). On top of that airlines are retiring planes early which sounds positive, but some of the older aircraft in AER’s fleet might become a lot less desirable (mostly older planes from owned fleet, e.g. see Lufthansa https://www.businesstraveller.com/business-travel/2020/04/07/lufthansa-to-permanently-retire-selected-a380-a340-and-b747-aircraft/).

        Further data point – older aircraft (15 years and older) are the most vulnerable, but there appears to be a decent cushion as the estimated sum of undiscounted cash flows was 82% above the aggregate carrying value as at 31 Dec 2019 (threshold for impairment testing; p.42 of the 20-F). At $2bn these 175 aircraft represent 6% of the book. Good buffer, but the demand shift towards more modern aircraft makes me a little bit nervous about lease rates/residual value for the older models.

        Again thanks for the write-up and good luck!

Leave a Reply