WeWork - A Case Study in the Lollapalooza Effect
Neil O’Donnell and I have discussed the rise and fall of WeWork constantly over the past few months, and this article is our attempt to synthesize the disparate causes of WeWork’s recent woes within the context of current macroeconomic policy, human psychology, and the real estate industry. We welcome all feedback.
Interest Rates and Macroeconomic Policy
It is a wondrous phenomenon that negative interest rates can be found in our modern day global capitalist economy where, for the better part of human history, most people have assumed a positive relationship between the value of time and money. The foundational idea of saving money today and receiving more tomorrow has been challenged in a sizeable way with over $17 trillion in negative yielding securities outstanding . Consider Howard Marks’ comments in Mysterious, one of his latest memos :
"In the financial world, most of our actions are based on the assumption that the future will be a lot like the past. Positive interest rates and the desirability of compounding have been among the most fundamental historical building blocks. If negative rates become more widespread across the globe, then the financial system needs to be rebuilt on a new set of assumptions. The problem is that we do not yet know what those should be or how they would work.”
Why are rates turning negative? Marks lists the possibilities...
“Are today’s negative rates in Europe and Japan telling us deflation lies ahead? Or have lenders changed their views regarding the time value of money? Or are rates negative simply because governments and central banks want them to be?”
…and potential reasons behind negative rates:
- Because central banks wish to stimulate their respective economies
- Because investors are fearful of long term prospects at the moment, leading to a glut of capital in the system needing storage and an increased demand for safe havens for capital and increased pricing to store said capital (in the form of having to ‘pay’ to store it in an institution!)
- Because of sovereign regulations forcing institutions to buy their own debt, artificially increasing the demand for said debt and driving up the prices (and driving down the yields)
From an institutional perspective, imagine you are a pension or a life insurance company and you must get a return on your assets to match your upcoming pension or insurance claim liabilities… and instead of receiving your principal with interest, you receive less principal than you invested at the outset. Short of storing physical cash (which is not an option for institutions with hundreds of billions in assets), the only rational response is to look, to the best of your ability, for alternatives.
As central banks across the world have continued to ease monetary conditions over the past 10 years, institutional investors flush with capital have been on an historic hunt for higher returns, which, on the margin, has caused massive flows of capital to riskier asset classes dubbed the ‘alternatives’. Consider these comments from the Financial Times on August 5 :
“There is clear evidence that this is happening among institutional investors. The flow of pension fund money into any asset that promises to beat zero-rate bonds has been so dramatic that equities, junk bonds, property, private equity and a host of other more abstruse areas of investment have spiraled in value - and to such an extent that they look highly vulnerable to any shock…”
In a world of negative yields, private equity and venture capital have experienced historic levels of fundraising, driven by large institutions’ requirements for ever higher returns to maintain portfolio hurdle rates [9, 10]. This phenomenon is illustrated well in the two following tables:
Let’s drill down further to one specific sub-alternative class: venture capital. We would hazard a guess that much of the growth in venture capital fundraising has been caused by one man: Masayoshi Son. After reading about SoftBank’s $100 billion Vision Fund, we were astounded at the size of the fund relative to the venture industry and wrote a piece describing how difficult it would be to generate outsized returns with a fund that size:
“Making some simple assumptions, I calculated that Son will have to return something on a magnitude of over $300B in capital to investors to achieve around a 10-12% IRR over a 12 year period (please note these are very rough estimations!). Not only will he have to distribute this much capital, but given that the fund likes to take a meaningful minority equity position at 15-30% of the equity (see TechCrunch article), this could represent somewhere in the $500-750B range of created equity value.”
So, in a world of negative rates where institutions are clamoring for higher returns, is it any wonder that Masa Son could raise a $100 billion fund? And with such firepower, is it surprising that other VC’s followed suit to raise larger funds to ensure they could compete with SoftBank’s offering? And with the increased fundraising momentum and requirements to deploy the capital over a short time frame, is it at all surprising that Masa Son was looking for entrepreneurs that were highly ambitious and full of vision to fund with massive rounds of capital?
Hubris and The Problem of Agency
Pride goeth before destruction, and an haughty spirit before a fall. (Proverbs 16:18)
When Masayoshi Son knocks on your door multiple times with over $10 billion in funding destined for you (or your competitors), you don’t say no. And at the terms offered, why would you? It could be argued that the real driver behind the WeWork rise and fall was Masa Son’s rapid deployment of capital, not Adam Neumann’s missteps. As an executive with fiduciary responsibilities, Neumann’s behavior while running the company was deplorable , but these personal foibles were only magnified and accelerated by the size of SoftBank’s investments. We don’t know Adam Neumann. But it is generally true that great chefs eat their own cooking and great ship captains do not abandon their ship  before going into battle. It is difficult to rationalize how WeWork’s founder was able to cash out nearly $750 million prior to the IPO, create a fire-sale situation by running the company in such a fashion worthy of the highest journalistic scrutiny, and manage to negotiate a pay-out worth over $1 billion to step down as CEO when thousands of employee’s stock options were underwater [17,18]. We can make no judgements because we have never been in such a situation, but the facts speak volumes. How did all of this happen?
We can only paint a likely story at best when dealing with human psychology because it is difficult to know for sure what goes on in the minds of men. But a little creative imagination goes a long way. When an entrepreneur begins to successively raise more rounds of capital, their ownership stake naturally declines unless they invest a commensurate amount into the business to keep their ownership stake at the same level. Adam Neumann owned roughly 18% of WeWork prior to the IPO. Assuming the business valuation on a per-share basis rises after each round of funding, the founder (on paper) becomes wealthier and wealthier. Neumann’s was worth billions on paper prior to the IPO. Both reasons can lead to operator incentives that become increasingly misaligned with their investors’. The founder has less actual skin in the game in terms of personal capital invested and is thus playing with ‘house money.’ Of course there is the risk of personal embarrassment in the case of business failure, but in the world of startups where failure is celebrated, this is of little consequence. And when the lead investor continues to pile on the capital, perhaps you begin to believe that you will live forever, that your company is a new ‘state of consciousness’, and that your descendants will be running the operation 3 centuries from now. Or perhaps Neumann operated under these assumptions all along, and the investment solidified his wild beliefs. We will never know.
Thus, Neumann found himself in an enviable position where his lead investor was pregnant to the tune of over $10 billion worth of equity in his company while he held tight control of the entity. From Neumann’s perspective, he was at the helm of a company that had never been profitable in its entire existence (save for maybe the first few leases) and his stake was worth billions - when given the opportunity, can you blame him for wanting to take some chips off the table pre-IPO to de-risk his personal fortune and lock in his lifestyle? From his perspective it was perfectly rational. Alignment be damned, his payday had arrived.
And after the IPO-turned-fire-sale? Well, a better example of sunk cost fallacy can’t be found. Throwing good money after bad, SoftBank rescued the company and negotiated a large buyout of Neumann’s CEO position along the way. SoftBank’s position was written down by over $8 billion while Neumann walked away a billionaire. Alignment be damned.
The problem of agency is one that plagues all investments, all companies, and all human organizations in general. It’s just plain hard to find situations where incentives align perfectly. At the heart of the problem of agency is the disproportionate leverage that can occasionally be given to operators by investors - for operators, the situation is the best of both worlds: tails you don’t lose too much except for your reputation and heads you win big, all while risking very little. For investors, depending on the incentive structure, the opposite very well may be true. Indeed, an entire volume could be written on alignment of incentives for operators and investors, but we’ll save that for another day.
Who was the patsy in this situation? Masa Son or Adam Neumann? We’ll never know whether Neumann truly believed the stuff he was peddling or if he was merely a highly talented snake oil salesmen. And it remains a mystery whether Masa Son pushed his vision on Neumann or whether Neumann’s magnetic charisma drew Son further into the business.
Regardless of who duped who, when a tech mogul and a startup operator enter the real estate industry, things are bound to get interesting.
WeWork, the Real Estate Industry, and the Always Inverted Yield Curve
The Real Estate Yield Curve vs. Financial Debt Yield Curve
The yield curve for real estate assets is naturally inverted. Short-term leases provide higher monthly rents than those of long-term leases because they are preferred by renters and disfavored by property owners. Short-term leases provide greater optionality as renters have the discretion to find a new property at the end of the lease term. Conversely, property owners need to negotiate new leases and must risk their property sitting vacant at the end of the lease. As a result, a property with a longer leasehold interest has a more dependable rent role which is more valuable and can accommodate greater leverage.
Conversely, yield curves for loans and other financial instruments only become inverted during times of economic distress. Generally, borrowers prefer longer loans to shorter loans as longer loans amortize more slowly (reducing monthly debt service payments) and mitigate cash flow risk because borrowers will not need to rollover loans at higher interest costs. This preference by borrowers for longer loans causes interest rates to increase with longer maturities during normal economic periods. During times of economic distress, short-term rates rise as a result of greater perceived risk in the economy in the short term, causing the yield curve to flatten or invert.
SIVs and WeWork
Structured Investment Vehicles (SIVs) were first created in 1988 by Nicholas Sossidis and Stephen Patridge at Citibank . SIVs issued commercial paper (with maturities of at most nine months) to invest in higher-yielding, longer-term debt securities—typically asset backed securities. We will refer to this business model from here on as the Yield Curve Arbitrage (YCA) model. Before 2000, there were only seven SIVs but this number mushroomed to 36 by 2007 with SIVs holding an aggregate $400 billion in assets . SIVs along with a select group of hedge funds, including Long Term Capital Management, were profiting by harvesting the liquidity premium by funding their long-term credit products with short-term debt and rolling over their debt obligations at the end of the term. The YCA model has, in fact, existed for centuries as banks similarly receive demand deposits and short-term certificates of deposit from depositors and make long-term loans to businesses and homebuyers. SIVs and other hedge funds, therefore, represented a new form of banking that was largely unregulated financing for trillions of dollars of assets.
SIVs, however, were unique as they had virtually no equity that supported the short-term loans that they received. While Citibank’s first SIVs had leverage ratios of 5x and 10x assets, by 2004 SIVs were operating with leverage ratios of 20x to 50x . At 50x leverage, a 2% change in the value of a fund’s assets would wipeout the fund’s net worth. Without an equity cushion, SIVs were far more exposed to disruption in the Financial Crisis. In fact, no SIVs survived in 2008 . Citibank closed the last SIV on November 19, 2008 .
WeWork’s business model represents the flipside of the SIV YCA model tailormade to profit from the inverted yield curve inherent to the real estate industry. Ironically, WeWork began just as SIVs were imploding in 2008 after Adam Neumann’s startup manufacturing baby clothes failed. In response to closing their startup’s doors, Neumann and Miguel McKelvey divided the office that Neumann had leased for his company and began subleasing space to other entrepreneurs . From its beginning, WeWork relied on long term leases (typically 10 years) while permitting everyone from entrepreneurs to small businesses to big corporations to rent month-to-month at a higher monthly rate.
This inverted YCA model harvests the optionality premium by deriving profit from the spread between higher-rate, short-term leases and lower-rate, long-term leases much like SIVs harvested the liquidity premium by investing in higher-rate, long-term loans. Also like SIVs, WeWork had just a sliver of equity financing it’s long-term leases. As of June 30, Wework had $22.0 billion in long term liabilities (including capital leases) against just $2.4 billion in cash  and was burning nearly $2 billion annually. As Aswath Damodaran noted when valuing WeWork, the company was operating with a “super-charged leverage model,” and he found that WeWork’s equity was likely worthless in “a whole host of plausible scenarios” .
In short, we’ve seen the WeWork story before, only in other forms and other industries. SIVs were funds that used massive leverage to harvest profits from investors’ preference for short-term debt. SIVs argued that an arbitrage opportunity existed between higher-yielding, long-term debt and lower-yielding, short-term debt that could be profitably and consistently harvested. In the last crisis, the steady profits of SIVs transformed into huge losses when short term interest rates rose. These losses destroyed all 36 SIVs that were operating at the time and the SIV model was shelved due to its inherent risks. In the case of WeWork, it did not take a macroeconomic crisis to expose the business, but rather a look under the hood at the financials which led to a swift re-rating. Unfortunately for Softbank and other WeWork shareholders, Mark Twain’s dictum reads too true: “history doesn't repeat itself, but it does rhyme.” Staying private longer most certainly would not have masked the risks inherent in the financial leverage applied to the YCA business model.
Synthesis - Lessons Learned
The most efficient way to describe the WeWork debacle is in Lollapalooza equation fashion:
Risk-Seeking Capital + Hubris + The Problem of Agency + Leverage + YCA Business Model
= Permanent Loss of Capital
The perfect storm always seems to be so, well, perfect in hindsight. Neumann’s management style, Masa Son’s ‘vision’, misaligned incentives, and an abundance of capital seeking returns. When these elements were combined with a perfectly viable business model and leveraged to the hilt, it became more and more difficult to see a scenario in which WeWork could succeed.
To wrap up the main piece, here are our lessons learned (and relearned):
1. Patience is the way from rags to riches. Impatience is the way from riches to rags.
2. Great chefs eat their own cooking. Great captains don’t abandon ship before battle.
3. Humility over hubris.
4. Emotion rules the hearts of men, but incentives rule their emotions.
Postscript: Macro Trends and Business Model Discussion
Just because leverage has proven once again to be the thorn in the side of overly ambitious, impatient executives and investors, should the baby should be thrown out with the bathwater? In other words, is the YCA business model (matching short term assets with long term liabilities) actually viable? More specifically, is the flexible office space model viable?
We can think of several cases in which this has proven to be a viable business model when prudent amounts of leverage are employed:
- Hotels (includes the Airbnb model - both the parent company and individual homeowners/investors)
- Banking (borrows at short term rates from depositors, creates long-term loans for businesses)
- And… flexible office space (Regus)
How then are these businesses and business models successful when WeWork wasn’t? To start, increasing the leverage through financial methods on top of an inherently-leveraged business model has almost always proven to be a fatal decision - one that Adam Neumann and Masa Son are relearning all too well.
Consider WeWork’s competitor, Regus, in comparison. Regus employs far less leverage than WeWork ever did. Going through all available annual reports, we’ve spread their financials over the past 20 years  and one thing is certain - the leverage ratios remain consistently low after their 2003 bankruptcy restructuring. It seems they learned their lesson on applying leverage to flexible office space 15+ years prior to the WeWork catastrophe.
Despite going through bankruptcy proceedings in 2003 after the dot-com bust exposed its over-leveraged balance sheet, Regus has limited long term debt to a maximum of ~40% of equity and turned a profit every year (including the downturn of ‘08 - ‘10) since 2005. Consider this chart from the 2017 annual report detailing strong returns on capital employed by vintage year despite a low leverage ratio (.8x net debt:EBITDA):
In addition to Regus’ success, the broader trend towards flexible workspaces seems to be gaining strength. According to a recent CBRE report, flexible office space accounts for only 2% of total leased office space currently in the US, but is projected to be 10-15% by 2030 . Unfortunately for the movement in the short term, WeWork is the dominant player in most markets and represents a big question mark in terms of lease performance. Despite WeWork’s missteps however, the flexible use trend appears to be larger than WeWork and represents the evolution of modern companies’ office needs.
Indeed, Masa Son and Adam Neumann were not wrong in identifying flexible work space as a growing trend in the real estate industry. They merely learned the same lesson that Regus did in 2003 - this business model doesn’t mix well with massive leverage.