David Einhorn Has Rough Quarter Despite Gains On Wirecard Short (Q2 2020 Letter)Jacob Wolinsky
David Einhorn's Q2 2020 letter to investors below.
The Greenlight Capital funds (the “Partnerships”) returned 1.0%1 in the second quarter compared to 20.5% for the S&P 500 index.
We ended our last quarterly letter with a promise to discuss how we have been positioning the portfolio in anticipation of rising inflation. An inflation bias was already embedded within the portfolio. We have owned gold for a long time in our macro book. During the quarter, we adjusted our position by reducing our direct exposure and adding the VanEck Vectors Gold Miners ETF (GDX) and a small position in a speculative gold miner. A number of our existing long equities have exposure to rising prices, including Green Brick Partners (GRBK) for house prices, Brighthouse Financial (BHF) for equity markets and eventual higher long-term interest rates, CNX Resources (CNX) for natural gas prices, The Chemours Company (CC) for titanium dioxide prices, and Buzzi Unicem (Italy: BZU) for cement prices. We added a new medium-sized long in Teck Resources (TECK), which should benefit from base metal price increases.
All that said, as we studied the various ways to profit from higher inflation, we settled on the most direct method – betting on unexpected increases in the U.S. Consumer Price Index (CPI). Inflation swaps are a highly liquid derivative of Treasury Inflation-Protected Securities (TIPS), the value of which are based on the official CPI at a future date. In May, we observed that the 2-, 5- and 10-year inflation swaps implied future annual inflation of approximately 0.1%, 0.8% and 1.3%, respectively. As annual inflation has averaged 1.7% over the last 10 years, we recognized that we could make a substantial return if actual inflation merely reaches the long-term average. If inflation turns out to be even higher, so much the better. Accordingly, we created a new, large macro position in 2-, 5- and 10-year inflation swaps. At quarter-end, inflation expectations had already begun to rebound to 1.3%, 1.4% and 1.6%, respectively.
In theory, higher inflation creates higher long-term interest rates and lower Treasury prices. However, we do not believe that shorting Treasuries is nearly as attractive as inflation swaps, as Treasury prices no longer reflect a free-market price. Not only has the Federal Reserve (the Fed) directly entered the Treasury market with large purchases, it has further announced it will “study” yield curve control (YCC). Speculating on materially lower bond prices is now a tougher call, as the market knows that the Fed is considering YCC. To wit, since the March lows, 10-year inflation expectations have risen an entire 1% without any move in 10-year Treasuries.
We believe there are two important implications of this. First, we have negative real interest rates, which is triggering a weaker U.S. dollar and a rally in gold and gold stocks. Second, any calculation of long-term asset values that is tied to an assumption about long-term interest rates appears flawed, as the market is not setting the long-term interest rate. The Fed’s intervention is analogous to when a company agrees to be taken over – once the take- over is announced, the share price becomes pinned near the deal price, and unless there is risk of the deal being derailed, the share price ceases to be impacted by changes in the company’s fundamentals, the macro economy, or the prices of other stocks. Similarly, once the Fed indicated it is likely to control the yield curve, barring a large change in fundamentals, Treasuries are effectively pinned in a narrow range.
We believe the market groupthink that profitless growth stocks that trade at astronomical valuations, in part on the basis that interest rates are low, will be disrupted by rising inflation expectations even in the absence of a corresponding increase in Treasury yields. Other markets like Japan and Europe have long recognized that artificially-controlled long-term interest rates are no justification for stratospheric equity valuations.
We also added a new large equity position in Atlas Air Worldwide Holdings (AAWW) at an average price of $36.28. AAWW operates the world’s largest fleet of Boeing 747 freighters and is a sizable owner, operator and lessor of 767, 777 and 737 freighters.
Prior to COVID-19, approximately 50% of global airfreight was carried in the belly of passenger planes, mostly on long-haul international flights. With long-haul international passenger traffic down more than 90% year-over-year (and likely to be the last segment of passenger travel to recover), there is a historic shortage of airfreight capacity. After an initial surge in demand to ship Personal Protective Equipment (“PPE”), the market is transitioning back towards more traditional airfreight products such as electronics, capital goods, perishables and pharmaceuticals. Market shipping rates increased by over 100% year-over-year in the second quarter and are expected to remain strong. As a result, we expect AAWW to see significant growth in earnings per share in 2020 (from the $5.24 it earned in 2019).
In response to the capacity shortage, some passenger widebodies are temporarily operating as freighters (nicknamed “preighters”), particularly to fulfill urgent PPE demand. However, due to lower cargo capacity, more cumbersome loading and unloading and similar overall trip costs, preighters cost roughly 2.5x as much per ton shipped compared to dedicated freighters. Preighter activity departing from China and Hong Kong has already declined by more than 50% since May as shipping rates have partially normalized. Over the next three years, we don’t expect many large freighters to be either produced or converted from passenger service given the cost and lead-times involved.
While most of the increase in earnings will occur in AAWW’s charter segment, AAWW also has attractive and substantial long-term contractual relationships serving DHL and Amazon, which stand to benefit from the growth in e-commerce and relatively steady business supporting the U.S. military. We acquired our shares at 0.54x Q1 2020 tangible book value and approximately 7x 2019 earnings that were achieved during much more competitive conditions. AAWW ended the quarter at $43.03.
In 2010, we were pitched the idea that Wirecard (Germany: WDI) was a fraud. At the time, the shares traded at about €8 per share. We spent some time analyzing the situation but didn’t get involved. In the following six years the stock went up 5x. After that, it went parabolic and went up another 5x in just 18 months. The shares peaked as the company was added into Germany’s prestigious DAX 30 index.
On June 18 of this year, WDI’s auditors disclosed that they couldn’t find about €2 billion of cash. In response to the apparent fraud, the shares lost about 99% of their value in a week. (We had a small short position.)
When a fraud like this is exposed, it’s customary to ask, what were the signs?
In WDI’s case, they were hiding in plain sight – which is to say, they weren’t hiding at all. The German publication Handelsblatt wrote about the anomalies in March 2010. Other financial press, notably the FT, followed with an array of stories. There were many voices yelling “FRAUD!” at the top of their lungs.
Rather than investigate the fraud allegations, the auditors continued signing the annual financial statements. The German authorities launched a criminal investigation into the relationship between short-sellers and the press and, for a time, they restricted short-selling of the stock. The sell-side analysts following the stock refused to engage in any real analysis of the controversy and instead weighed in with mostly “Buy” recommendations and astronomical price targets.
To paraphrase one observer, “The bears believe there is fraud and the bulls believe there will be a short squeeze.”
The current regulatory regime has a policy of not doing much about fraud, because taking strong action will hurt the shareholders, who would become the real victims. The problem with this thinking is that not policing deception allows small frauds like WDI in 2010 to become national embarrassment-sized frauds like WDI in 2020. The shareholder losses aren’t just deferred, they grow exponentially (in WDI’s case, by 25x).
Of course, this brings us to Tesla (TSLA).2 A couple years ago, Elon Musk engaged in the most blatant securities manipulation in recent memory. In response, the SEC slapped him on the wrist with a trivial fine and SEC Chairman Clayton declared:
It often is the case that the interests of ordinary shareholders – who had no involvement in the misconduct – are intertwined with the interests of offending officials and the company. For example, corporate fines often are financed with funds that could otherwise benefit shareholders, and the skills and support of certain individuals may be important to the future success of a company [emphasis added].
The potential shareholder loss due to TSLA unravelling is substantially larger today than when Clayton said this.
There are so many examples of TSLA abusing its stakeholders that we could write about them for pages every quarter. For the sake of space, we will talk about just one that has not received a lot of attention.
There have been more than a hundred documented incidents where drivers have reported that Teslas have sudden unintended acceleration (SUA). It is generally believed that SUA results from driver error for most cars. However, according to a study by Dr. Ronald A. Belt,3 there is a problem with how a Tesla’s braking system interacts with its battery regeneration system. The result is that, in certain instances, pressing the brake pedal causes the vehicle to accelerate, and the harder the driver presses the brake, the faster it accelerates. In January, the National Highway Traffic Safety Administration (NHTSA) announced it would investigate the SUA complaints. TSLA, as per custom, denies there is a problem. The reality is that the more technologically sophisticated products become, with complex, intertwined hardware and software systems, the more difficult it is to design them for universal adoption in real-world large scale deployments. The issue may be that the combined software/hardware problem may not be easily fixable. In a way, it’s a similar problem to what Boeing encountered with its new 737 MAX. And as Boeing showed, doing the right thing and recalling a defective product can be financially ruinous to the company and career-ending for management. TSLA’s management cannot be trusted to do this on their own. Rather, it is up to the NHTSA to perform its statutory mandate and order a recall of any Teslas that have safety-related defects.
The latest driver of TSLA’s parabolic move is speculation about its inclusion in the S&P 500 index. By rule, until a company is profitable for a year, including in its most recent quarter, it isn’t eligible to be in the S&P 500. Through what appears to be sheer abuse of the accounting rules, TSLA has now contrived reported profits to make it technically eligible. In addition to its routinely questionable accounting maneuvers,4 Tesla appeared to defer employee compensation, depreciation expense on its new plant in China, and research and development spending.
However, the most notable development in the last two quarters came from the enormous increase in sales of regulatory credits. Historically, TSLA’s accounting for regulatory credits included the following sentence:
We recognize revenue on the sale of automotive regulatory credits at the time control of the regulatory credits is transferred to the purchasing party [emphasis added] as automotive revenue in the consolidated statement of operations.
However, in the Q1 ’20 10-Q, Tesla removed this language.5
In both the March and June quarters, TSLA sharply increased its recognition of regulatory credits. The company reported regulatory credit sales of $782 million in the first half of 2020, compared to $594 million in all of 2019. Historically, TSLA received cash for regulatory credit sales, but this year they have piled up in accounts receivable. TSLA’s primary customer for these credits (and exclusive customer in Europe, where most of the credits are generated) is Fiat Chrysler (FCA), yet TSLA’s recognition of regulatory credits seems inconsistent with FCA’s recognition of expenses for regulatory credits – FCA recognized €327 million (around $370 million) of regulatory credits in the first half of 2020, or less than half of TSLA’s total. Historically, FCA’s accruals closely matched TSLA’s revenue recognition. Moreover, the increase was in the face of TSLA delivering far fewer cars in Europe in the first half of 2020. TSLA reported total GAAP earnings of $120 million so far in 2020, which would have surely been negative without the regulatory credit over-accrual.
We question whether TSLA’s accounting, which does not appear to correspond to the creation of regulatory credits through auto sales, transfers of those credits to a counterparty nor payment for those credits, conforms to GAAP accounting.
We suspect that TSLA changed its accounting policy during a non-audited quarter to manipulate eligibility in the S&P 500 index. The consensus is that S&P will add TSLA to the S&P 500 index at the next opportunity with a large weighting, forcing millions of passive investors to sell the other 499 stocks to make room for TSLA at whatever the price du jour. We think the S&P 500 Index Committee has a tough decision to make as to how to respond to being gamed like this.
As with WDI and the DAX, we expect the TSLA parabola to end around the speculated inclusion in the prestigious S&P 500 index.
While we take a value-oriented approach, our investing style is not a closet index of long value and short growth. We look for security-specific differences of opinion and hope to capitalize on being right and the market eventually seeing it our way.
Take BHF, for example. In 2017, we bought the company at a $6 billion valuation. At the time, the consensus view was the company would have trouble transferring “adjusted” earnings into GAAP earnings and wouldn’t be able to return capital to shareholders until 2020. Fast forward to today: the company has since generated $6 billion of retained earnings, while repurchasing 22% of its shares. In response, the market now values the company at about $2.5 billion.
In May, the company demonstrated that it had effectively hedged the first quarter’s market turbulence. Due to a large hedging gain, the company reported $47 per share in quarterly earnings. Yes, BHF earned more than the entire price of the company in a single quarter. Moreover, it bought back 12% of the outstanding stock at depressed prices. Despite all of this and a rapidly rebounding S&P 500, BHF stock only advanced from $24.17 to $27.82 during the quarter, leaving it down 29% for the year. During the quarter, the shares barely outperformed the company’s bonds maturing in 2027, which now yield 3.3% – although the equity languished, the credit markets have reversed any pandemic-induced concerns. Perhaps we just need Dave Portnoy to select “B”, “H”, “F” out of his bag of Scrabble tiles.6
We understand that book value and GAAP earnings are problematic metrics for BHF. There is a change of accounting pending next year that we expect will take out a chunk of the book value, but at a price to book ratio of 0.16x, this should not be a risk to the stock price. In addition, there is a mismatch between the GAAP accounting for hedges and the risks being hedged. As a result, we expect a large loss in the second quarter, as some of the hedge gains will be reversed due to improving market conditions – just as the first quarter resulted in record GAAP income.
Like many companies, BHF provides “adjusted” earnings that take into account the problems with GAAP accounting. On that basis, its P/E ratio is 3.3x. Management has targeted a goal of repurchasing $1.5 billion of stock by the end of 2021. It has $636 million to go. At current prices, this implies the company will retire an additional 23% of the stock over the next 18 months.
We exited a position in Adient with a 50% loss over two years. The management was well on its way to executing a successful turnaround when COVID-19 hit. The unexpected macro event has clearly impaired the near-term opportunity, so we exited. We also exited Altice USA with a 36% gain over two years. The company executed on cost-cutting and growth strategies and repurchased a significant amount of shares at attractive prices. We sold over concern about the 2020 advertising environment and potential broadband regulations.
Bryan Nowicki left in the second quarter to pursue an opportunity at a start-up fund. We thank him for his contributions and wish him well.
We added a couple of new partners the old-fashioned way. While we have historically avoided pair trades, we celebrated as Jason and his wife Kelli welcomed twins Rylie Dawn and Sloane Bari Lewis on May 6. The girls are named for their late grandfather, Don Barry Lewis. If you twist Jason’s arm, he just might be willing to share photos of the identical cuties with you. Congratulations to the Lewis family!
At quarter-end, the largest disclosed long positions in the Partnerships were AerCap Holdings, Atlas Air Worldwide Holdings, Brighthouse Financial, Green Brick Partners and VanEck Vectors Gold Miners ETF. The Partnerships had an average exposure of 123% long and 68% short.
“I am involved in the stock market, which is fun and, sometimes, very painful.” – Regis Philbin (former Greenlight intern)
Greenlight Capital, Inc.
1 Source: Greenlight Capital. Please refer to information contained in the disclosures at the end of the letter.
2 We are using a disproportionate amount of this letter to discuss TSLA relative to its current weighting in our portfolio, just because we find the situation fascinating.
4 TSLA’s accounting for accounts receivable, warranty reserves, residual value guarantees, deferred revenue recognition and currency translation are all questionable (at best) practices for which TSLA provides no answers and bullish analysts refuse to seriously discuss.
5 Further confusing the situation, the language reappeared in the Q2 ’20 10-Q.
6 This is a fun video: https://twitter.com/stoolpresidente/status/1274076383615090688.