WeWork Accounts for Consciousness

(Bloomberg View) -- WeWork.

The business model of real-estate companies is pretty much, you borrow a lot of money, you use it to buy buildings, and you rent out space in the buildings to tenants. If the tenants pay more in rent than you pay in interest on your loans, you make money; if they don’t, then you don’t.

WeWork Cos. is a real-estate company with a couple of innovative twists on the model. First, rather than owning its buildings, it rents them: It leases office space from regular real-estate companies, adds … beer? … or whatever, and then subleases the space to tenants at higher rates. And second, rather than being valued like a real-estate company, it gets valued like a hot tech startup — “the sharing economy,” ping-pong tables, etc. — so it can raise gobs of money from SoftBank Group Corp. at a $20 billion valuation without ever getting particularly close to profitability. And look at all these words:

Indeed, to assess WeWork by conventional metrics is to miss the point, according to [Chief Executive Officer Adam] Neumann. WeWork isn’t really a real estate company. It’s a state of consciousness, he argues, a generation of interconnected emotionally intelligent entrepreneurs.

Really, what sort of multiple would you put on a state of consciousness?

That first innovation seems a little questionable. Like, there is an established competitive business of office rental in which real-estate companies own office buildings and rent them to companies; I am not sure why there would be a ton of room to compete with that business by interposing yourself as an expensive middleman. Why would a tenant want to pay a profit margin both to WeWork and to its underlying landlord, when it could just rent from the underlying landlord and pay only one profit margin? There is some room for a value-added middleman — and WeWork can add value not only by providing beer but also by splitting office rental into smaller space and time chunks than a big commercial landlord — but, still, it does not seem easy.

But the second innovation is great. For one thing, it is great for the obvious reason: If you can get into a traditional mature highly competitive business, call yourself a tech startup, and get a multibillion-dollar valuation based on potential rather than cash flow, then you have achieved a profound arbitrage and really ought to be rewarded for it. But it also helps solve the first problem: WeWork’s tenants don’t have to pay two profit margins, because WeWork’s investors give it tons of money which it can then spend on giving tenants free rent. In a loose sense, WeWork’s business model is getting SoftBank to buy beer for software workers. Which is fine!

Anyway WeWork issued $702 million of new seven-year bonds this week, priced at a 7.875 percent yield, and people lost their minds a bit. The particular catalyst for much of the outrage is that the bond offering documents included a financial metric called “community adjusted Ebitda.” From FT Alphaville:

The investor presentation focusses on a non-GAAP income measure it calls “community-adjusted EBITDA”. That's essentially EBITDA excluding the cost of sales: Employees' cash compensation, advertising/event expenses, and new-location costs. That figure is positive, which means it's profitable if you exclude its cost to grow profits.

It called the fully adjusted number “community adjusted Ebitda,” by which it subtracted not only interest, taxes, depreciation and amortization, but also basic expenses like marketing, general and administrative, and development and design costs. Those earnings were $233 million, WeWork said.

“I’ve never seen the phrase ‘community adjusted Ebitda’ in my life,” said Adam Cohen, founder of Covenant Review, a bond research company.

Well, sure, Mr. Covenant Review, but I bet you’ve never reviewed the covenants of a state of consciousness either. 

I am tempted to defend “community adjusted Ebitda” a little bit. Loosely speaking it is the amount of rent that WeWork gets paid minus the amount of rent (and other facility operating expenses) that it pays. Dan Primack at Axios notes that it “includes costs and expenses specific to existing WeWork facilities,” including “all tenant fees, rent expense, staffing expense, facilities management expense, etc. for active WeWork buildings,” but excluding “company-wide expenditures,” much of which “relates to growth efforts.”

Why would you want to know that? Well, Alphaville’s shorthand of “profitable if you exclude its cost to grow profits” is helpful. If you are an equity investor like SoftBank, you want to have some sense of WeWork’s endgame. It’s a familiar idea for fast-growing tech companies: You sacrifice profitability to grow rapidly, but at some point in the future you will flip the profitability switch, and when you do that you want it to actually work. “Community adjusted Ebitda” is one measure of whether the profitability switch will work when you flip it. 

If you are a bond investor, though, your interests are different. You want to get your money back; you don't care so much about growth. Obviously if WeWork was super profitable right now, that would be a good sign that you will get your money back in seven years, but you can’t have everything. (In particular, if WeWork was super profitable right now it wouldn’t be issuing bonds at 7.875 percent.) The fact that WeWork is not profitable right now is not great, but it’s not terrible either. If these bonds came due tomorrow they’d get paid off. They’d get paid off because WeWork has $2 billion of cash, but also because WeWork has $20 billion of equity market capitalization. That $20 billion isn’t, like, a real sack of cash, but it is some indication of how highly its deep-pocketed equity investors think of it, and how unlikely they are to lose it by letting it default on its bonds anytime soon.

But the bonds don’t come due tomorrow, and what bondholders really want to know is whether they’ll get paid off in seven years, if that is no longer true. If WeWork is not a fast-growing tech darling, if SoftBank is no longer willing to put money into it, if its equity has lost its luster — will it have enough money to pay back the bonds? In the downside, no-growth, no-hype case, will it generate enough cash to pay its debts? And “community adjusted Ebitda” — look, it’s not perfect, and obviously it’s shaded in a way to favor the company — but it’s arguably a way to measure that. Strip out the growth and the events and the corporate-level compensation, focus on the basic economics of rent coming in and rent going out, and see if it’s enough to cover the debt. In an alternate world where the bondholders seize the company, cut all the nonsense, and focus on paying themselves — will there be enough there for them? (Again, it is a debatable measure: Even a cut-to-the-bone WeWork would have to pay some executive salaries, and would have some sales expenses just to maintain its occupancy rates.)

Even the name is defensible, really. I think they’re just using “community" as a synonym for “building”: It’s the adjusted Ebitda of WeWork’s locations, before taking out corporate costs. But you can also think of “community adjusted Ebitda” as WeWork’s pro forma Ebitda in the alternate reality where it was run as a real-estate company. Just, you know, get rent on one side, pay rent and operating expenses on the other side, not a lot of guff about beer and events and consciousness and rapid startup-style growth and a generation of interconnected emotional blah blah blah. It’s Ebitda adjusting out the community. It makes sense.

WeWork’s bond deal looks like a tech-company deal in another respect, which is that everyone talked about how much they hated it —

“We cannot get comfortable with the company’s financial and operating position, which includes a massive asset/liability mismatch that is usually a recipe for disaster, significant cash burn, cyclically untested real estate business model, and uncertain path to profitability,” Rosenthal said in a report Wednesday entitled “WePass.”

— and then bought it anyway:

The company had initially planned to sell $500 million, and then received orders of about five times that amount, the person said, asking not to be named because the deal is private.

It fell a bit in the aftermarket.

Should index funds be illegal?

Japan’s SoftBank is planning to swap more than $20bn of investments in some of the world’s largest ride-hailing groups, including Uber, Ola, Grab and Didi Chuxing, into its huge Saudi-backed Vision technology fund.

If finalised, the move would allow the fast-growing tech groups access to the fund’s resources, according to people familiar with the talks, with the hope that the ride-hailing providers would work more closely together.

We talk a lot around here about the idea that common ownership of multiple companies in the same industry by the same large diversified institutional investors might be bad for competition among those companies. No one who works in finance, or at companies, really seems to believe this, in part because they have met those large diversified institutional investors, who don’t seem like the type to cook up ways for all the airlines to collude with each other to keep prices up. But as common ownership becomes more common, it is not obvious that all of the common owners will be traditional passive-ish institutional investors. Once you own all the ridesharing companies, it might occur to you to take steps to combine them.

Are banks tech companies?

At a very high level of abstraction, a bank is a set of algorithms to match people who have money with people who need money. Sometimes these algorithms are literally computer algorithms: A bank will build, and let clients use, algorithms to allow customers to buy and sell stocks with each other on the stock exchange. Sometimes they are human analogues of that: A bank will employ bond traders who take phone calls from customers who want to trade bonds with each other, with the bank matching the customers in trades. And sometimes they are opaque traditional banky processes, where it feels a bit forced to call them “algorithms”: A bank will open up branches, take deposits from savers, throw those deposits together into a big pile of money, and lend out bits of that pile to people who need money to buy a house or open a business. Money from savers comes in at one end, and it goes out at the other end in investments, but it’d be hard to reduce what happens in between to a set of explicit instructions.

One trend in banking over the past few decades has been that things tend to be migrating away from the opaque-traditional-banky-process end of that spectrum and toward the explicit-computer-algorithm end. Stock trading used to be done by humans on the phone; now it is done mostly by algorithms. Bond trading used to be done by banks buying and selling the bonds with their own money; now it is increasingly done by matching buyers and sellers directly. Lending used to be done with a meeting and a handshake; now it is increasingly done on the internet, and banks often package loans and sell them to investors rather than keeping them themselves.

A good thing about that migration is that it tends to increase the efficiency of the financial sector. A worrying thing about it, for the banks, is that banks tend to have a lot of competitive advantage in doing opaque traditional banky things, and relatively less competitive advantage in building computer algorithms. If stock trading is just about writing algorithms, upstart electronic trading firms can compete with the big banks. If lending can be done by matching depositors and borrowers algorithmically, without all the branches and handshakes, then internet startups can easily get into the business. 

You know who is really good at writing search and matching algorithms? Google and Amazon. Should that worry the banks? Not yet!

The reason: Banks are rapidly emerging as big potential customers for the fast-growing, cloud-computing businesses of Amazon.com Inc., Alphabet Inc., and Microsoft Corp.  That is making technology companies think twice about alienating them by becoming direct competitors.

“We are very intentionally approaching banking customers with an opportunity to empower them rather than to be in their space,” said Judson Althoff,  executive vice president of world-wide commercial business at Microsoft. ”Banks see other cloud providers investing in mobile payment capabilities, and there is a concern about disintermediation.”

I wonder if that can last forever. Surely there is more revenue in being a bank than in providing cloud services to the bank?  On the other hand, the margins in the cloud business might be better.

No one reads 10-Ks.

There is a trope in financial markets about virtuous stock investors who show up every day and plow through a stack of annual reports on Form 10-K, learning all they can about the companies so they can make informed investment decisions. This trope annoys me a little, given my efficient-market sympathies. If something is in the 10-K, that means it is (1) backwards-looking and (2) public. Markets are forward-looking. The way to beat the market is to have information and analysis that no one else has. The point of the market is to add information to prices. It is nice that everyone has access to the same basic shared set of information in the 10-K, but don’t pat yourself on the back too hard just for reading it, you know? 

So I have mixed feelings about this paper by Lauren Cohen, Christopher Malloy and Quoc Nguyen with the delightful title “Lazy Prices,” which finds that when companies make significant changes to what they say in their quarterly and annual filings with the Securities and Exchange Commission, those changes have predictive value, usually bad:

Changes to the language and construction of financial reports also have strong implications for firms’ future returns: a portfolio that shorts “changers” and buys “non-changers” earns up to 188 basis points in monthly alphas (over 22% per year) in the future. 

On the one hand, the virtuous 10-K readers are right, and I am wrong, about the value of reading the 10-K: It apparently contains information that will allow you to beat the market! On the other hand, the virtuous 10-K readers are either less numerous, or less careful readers, than you’d think:

Investors are missing these subtle but important signals from annual reports at the time of the releases, perhaps due to their increased complexity and length. When you isolate changes to corporate reports using our approach, one can see that document changes do impact stock prices in a large and significant way, but this happens with a lag: investors only gradually realize the implications of the news hinted at by document changes, but this news eventually does get impounded into future stock prices and future firm operations. Thus the message from our paper is quite different, in that our results point to a large amount of rich information that is being hidden in the 10-Ks, and that investors are missing (and continue to miss, even today), rather than the conclusion that corporate documents are becoming less informative and less useful to investors in today’s capital markets.

“Rich information that is being hidden in the 10-Ks”! The 10-K is where you disclose the information! 

We talked the other day about Yahoo! Inc.’s pretty bad disclosure after its huge security breach: The 10-K continued to warn investors that Yahoo might get hacked, and that the consequences might be dire, but forgot to mention that it had been hacked and the consequences were dire. I called it “a perfect example of a certain kind of lawyerly thinking,” disclosing bad things hypothetically so you can say you’d disclosed them without giving away any actual information.

There’s some indication that that might be happening here. The paper mentions the example of a medical-products company that in 2010 suffered a recall of a key device; its annual report for 2009, filed a few months before that recall, showed significant changes in the wording about that device. But those changes weren’t, like, “it’s gonna be recalled.” Nor, to be fair, were they just nondisclosures like Yahoo’s. They were changes that look helpful in retrospect: changing “additional charges … may be required” to “substantial additional charges, including significant asset impairments … may be required”; adding a sentence about how various regulators “have each increased their enforcement efforts” relating to the device. If you focused on those changes, you might have thought, “huh, something is up, there's more risk relating to this device than I had thought.” But you might just as reasonably have thought “ugh, lawyers, always adding more words to be more cautious.” If annual reports are filled with enough hypothetical warnings, it is easy to miss the real ones.


I wrote a little last week about the high-tech and yet weirdly personal surveillance at big financial firms, which are constantly screening their employees’ emails for compliance purposes, but where that screening often involves an actual human reading the emails. I put out a little request:

If you are a compliance officer who is reading this email not because you are a subscriber but because you are surveilling a bank employee who is a subscriber: Hi! This is a little awkward, but it's nice to meet you. If this describes you please shoot me an email; I'd love to hear more about you and whoever's email you're reading. 

Guys: It worked. Here’s an email I got yesterday from a reader at an asset-management firm who would prefer not to be named:

I work in my firm’s compliance department … and sometimes read employee emails (as part of surveillance). I also read your columns, as do some of the people whose emails I read. They also discuss your columns with other people at different firms in instant chat and by email. 

Look obviously it would be better if he only read my columns through his surveillance, but still, this made my day. “I think people tend to enjoy your columns,” he added. I am going to launch a special edition of Money Stuff that you can only read by surveilling the email of a regular Money Stuff subscriber.

Things happen.

The Crypto Crime Wave Is Here. Deutsche Bank has a 60bn problem - but doesn't want to talk about it. Deutsche Bank loses €35m from bad block trade. Barclays Answers Critics With Investment-Banking Growth. Fearing the Fed, Credit Investors Are Buying More Junk Instead. Wells Fargo’s 401(k) Practices Probed by Labor Department. DocuSign Gets $629 Million in Above-Range U.S. Tech Offering. Matthew Klein on the jobs guarantee. “A quick rule of thumb is that when someone seems to be acting like a jerk, an economist will defend the behavior as being the essence of morality, but when someone seems to be doing something nice, an economist will raise the bar and argue that he’s not being nice at all.” Oh No, It’s Me, a Guy You Just Met, and I’m Already Talking About Elon Musk

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.

To contact the author of this story: Matt Levine at mlevine51@bloomberg.net.

For more columns from Bloomberg View, visit http://www.bloomberg.com/view.

©2018 Bloomberg L.P.