Money Stuff

Blockchain ICOs and Uber CEOs

Also online lending, LinkedIn nudes, email hoaxes, factor crowding and vampirism.


Is there a good way to think about initial coin offerings? The idea of an initial coin offering is that a company builds, or promises to build, some blockchain-based platform that allows people to buy or sell some product or service. (It seems like it's usually cloud storage.) And then it creates a digital token -- its own "coin" -- that you have to use to buy or sell the product. And then it sells the coins, all at once, to the public, often before the product is fully up and running; the buyers can use the coins to buy the product, or try to sell them to someone else for a profit. And then the company goes to conferences to gloat about how it has disrupt ed the venture-capital funding model and shown the way to the future of capitalism and so forth:

More and more startups are offering tokens as a way to raise money upfront in so-called initial coin offerings (ICOs), a nod to traditional initial public offerings of securities. So far this year, 63 sales have raised $521 million, according to blockchain research firm Smith + Crown. That has already far surpassed the $260 million raised in 2016, says Emma Channing, general counsel at Argon Group, a year-old digital finance investment bank in Los Angeles.

One thing you could think about is what this tells us about how businesses should be funded. Most companies want to sell a product, don't have enough money to build it, and so go to investors to get the money. Then they build the product, sell it to customers, get money from the customers, and give some of the money back to the investors. Raising money from customers rather than investors, as ICOs notionally do, has some obvious appeal: It cuts out the middleman, in a sense, and it also lets the company share some of its profits with early-adopter customers rather than investors. The company and the customers are necessary components to the system; the investor, in the new model of blockchain capitalism, is just an interloper, and can be dispensed with.

On the other hand, if a company has raised tens of millions of dollars by pre-selling its product to customers before ever building the product, why would it build the product? This is a problem for investors too -- if a company has raised tens of millions of dollars by selling shares to investors before ever building a product, why not run off to Belize rather than build the product? -- but there is a well-developed set of legal protections for investors, for exactly that reason. The legal status of coin-buying customers is a little vaguer.

Another thing you could think is: Well, is it vaguer? One obvious worry about ICOs is that some of them might "really" be securities offerings, that "coins" are in many cases just quasi-equity interests in blockchain-based businesses and should be subject to all of the standard securities regulations. Here is a December white paper on the subject, but one overly simplistic and not-at-all-legal-advice way of thinking about it is: If customers are buying coins as a way to fund their eventual purchase of cloud storage or whatever, then the coins are not securities, but if speculators are buying coins as a way to profit from the growth of the business (or the speculative frenzy around ICOs), then they probably are, and are probably subject to securities regulation. Which do you think it is?

“It seems as if there’s this enormous financial pressure of money that wants to move into this space, but there’s really limited places for it to go,” says Brian Lio, chief executive officer of Smith + Crown. But an offering that sells out in seconds may work against some of the bigger goals of a blockchain enterprise. Tokens are supposed to be used by customers. “If your token instantaneously sells out to a few institutional investors, you’re missing a big point of tokens, which is to get them to users and build a community,” says Lio.

One way to tell the story is that bitcoin demonstrated definitively that a privately created digital currency can become worth tens of billions of dollars based solely on ... I don't know, based solely on some sort of psycho-sociological thing that you can give your own name to: "trust" would be a good word to fill in the blank, or "widespread acceptance," or "speculative frenzy" would work almost as well. Everyone vaguely knew that -- it's how, like, dollars work -- but bitcoin demonstrated it in a particularly pure and salient way. 

But it's hard to repeat that pure feat: "Bitcoin will replace fiat currency" is a good pitch for a new cryptocurrency, but "mattcoin will replace ether which will replace bitcoin which will replace fiat currency" is less compelling. But you can tap into it obliquely. If you are a company with a digital product, or an idea for a product, you can yoke that product to the bitcoin phenomenon: "Buy one unit of cloud storage and we'll throw in five free units of speculative frenzy!" And people who want cloud storage might buy your coins, or not. (There are other cloud storage companies.) But people who want speculative frenzy will definitely buy your coins, because initial coin offerings are the main way to get in-on-the-ground-floor bitcoin-style speculative frenzy these days.

By the way, I say "it's usually cloud storage," but in fact there are ICOs for all sorts of internet businesses, and some not-quite internet businesses:

Which may (sort of) explain LGD, the digital token for the Legends Room, which describes itself as “a world class Las Vegas gentlemen’s cabaret ­r­­e-imagined using blockchain technology,” according to the website. It sold about $1 million in LGD in an ICO that closed in May, says Peter Klamka, a consultant who set up the token strategy for the business.

Here's the website, which is perhaps not entirely safe for work. I was vaguely hoping that the club would exist entirely on the blockchain -- that would be innovative! -- but no, there seems to be a physical location. 


I like that Uber Technologies Inc. is applying the principles of the "sharing economy" to the job of its chief executive officer:

The decision by Uber Technologies Inc.’s chief executive to take an indefinite leave of absence will put management of the world’s most valuable startup in the collective hands of more than a dozen managers—a challenging structure for a company that has favored a strong central leader and pitted executives against each other.

It's the future of work! No one has a full-time job; you just set your own schedule based on supply and demand. Need some extra money? Be CEO of Uber for a couple of hours in the morning, after dropping your kids off at school. You're a night owl? Run Uber between midnight and 4 a.m. "Surge pricing" means that you'll get paid more for being the Uber CEO at high-demand times, like during board meetings or in the wake of public-relations disasters. Really they should build an app, and if an Uber employee needs a chief executive officer, she can open up the app and it will tell her like "your CEO will arrive in four minutes," and then nine minutes later one of the "14 senior executives" who are holding down CEO Travis Kalanick's spot in his absence will show up at her office and make an executive decision, and she will rate him from one to five stars. 

Whoever deals with this will presumably get surge CEO pay:

One of the U.S. government’s most powerful consumer protection watchdogs appears to be quietly probing Uber and the company’s privacy practices.

The inquiry is under way at the Federal Trade Commission, according to four sources familiar with the matter, where the agency’s investigative staff appears to have focused its attention on some of the data-handling mishaps that have plagued the company in recent years — perhaps including employees’ misuse of “god view,” a tool that had previously allowed some at Uber to spy on the whereabouts of politicians, celebrities and others using the ride-hailing app.

And Naked Capitalism has Part Ten (!) of Hubert Horan's series predicting Uber's doom: "Uber’s narrative that its powerful business model will inevitably achieve global dominance is no longer credible and can no longer divert public attention from Uber’s dismal economics."

Bad loans.

Here's a story about how Prosper Marketplace Inc. and LendingClub Corp. "don’t always check whether borrowers are lying to them, and if they find errors in an application, they may still approve the loan": 

Alia Dudum, a LendingClub representative, said the company uses “machine learning and other techniques to build robust models that segment which borrower applications need verification and which do not.”

Doesn't that seem ... fine? Like, in theory, I mean. Lending, as a business, is essentially statistical; the point is not to have all your loans pay off, but to have enough of them pay off that your losses are covered by the interest that you charge. Even if you verify income on every loan, some of those loans will default anyway. And while verifying income, or whatever, on every loan might help your statistics, you don't have to believe that a priori. You could just test it out: Your machine learning could look at the statistics and tell you, you know what, if you skip verification on half your loans, your default numbers won't go up materially.

Of course you could still be wrong:

Loans made on platforms including Prosper and LendingClub have gone bad faster than security underwriters had expected, according to data from Morgan Stanley, and most of the startups have seen their funding costs rise over the last year.

But never mind that. One basic tension in modern finance is that the financial industry largely thinks of itself in statistical terms, while regulators and the public tend to think about individual cases. When Wells Fargo fired 5,300 employees for opening fake accounts, the public was aghast, but Wells Fargo executives "received the figure positively, believing it proved that a vast majority of individuals were behaving appropriately." Banks and securitizers and online loan marketplaces know that there will be defaults, and want to optimize the level of defaults for maximum profit. The public wants lenders to take appropriate care with each loan. 

Bad LinkedIn.

"It’s Probably Not Okay to Send Naked Pictures on LinkedIn," is the headline here, but that is actually a novel and interesting legal question. I mean, it is a straightforward and obvious question of aesthetics and internet etiquette -- it is not okay to send naked pictures on LinkedIn! -- but the legal question is, if you send naked pictures on LinkedIn, is your company responsible?

In a lawsuit filed Tuesday night in Los Angeles, a mid-level financial industry professional identified only as Jane Doe alleged that a recruitment conversation on LinkedIn took a turn for the inappropriate when she received sexual messages from a banker—using his corporate account—who had been trying to recruit her. One of the messages included a photograph of his genitals.

Women have long complained of unsavory conduct on LinkedIn, but in the California state court complaint (filed by the law firm of celebrity lawyer Mark Geragos), Doe argues that an employer is responsible for employee behavior on the platform. LinkedIn is an extension of the workplace, similar to going into the office or attending a corporate networking event, the theory of the case goes.

I don't know about that theory, but the good news is that if it works then employers will have to be much more careful about letting their employees use LinkedIn unsupervised. If this lawsuit leads to widespread workplace bans on LinkedIn, we might all be better off for it.

Also here is how this allegedly went down:

The two met when Doe, at a previous employer, worked on a deal that involved Eichler’s company, Doe said in an interview. Eichler initially messaged Doe about potential job opportunities. When Doe expressed interest in hearing more, the messages shifted from professional, she alleged. “So what are you doing up so late?! Here’s my number if you wanna play,” wrote Eichler, according to court filings. He later added that it could be a “late night secret” before sending a graphic photograph, she alleged. After Doe didn’t reply, he wrote “Ugh, I guess I screwed up :( bummer dude.” 

While I cannot recommend sending nudes on LinkedIn, I will admit that ":( bummer dude" really is the perfect way to respond to rejection, whether romantic or professional, on LinkedIn.

Bad email.

There is this prankster who keeps emailing bank chief executive officers pretending to be their colleagues and trying to trick them into saying impolitic things, and they reply with bland politeness, and he gloats "got 'em again," and articles are written about how he tricked the CEOs and exposed flaws in the banks' email security, and I don't know, man. The latest is an exchange with James Gorman of Morgan Stanley, in which the prankster impersonates former U.K. Chancellor of the Exchequer (and Morgan Stanley board member) Alistair Darling, and Gorman is polite and vague. "Security experts warned that the episodes have exposed weak spots in the big banks’ defences, at a time when they are under near-constant bombardment from criminals, 'hacktivists' and disaffected insiders," writes the Financial Times. I tend to take the other side of it: After five attempts, no bank CEO has actually been tricked into saying something inappropriate or giving away any information. Eventually a bank CEO will reply "hello, I see you are emailing me from a fake email address, so I will not engage any further," and the hoaxer will post that exchange online and be like "yes, another flawless victory!"

Rules-based trading.

Here is Robin Wigglesworth on last Friday's sell-off in tech stocks, which he attributes to selling by factor-driven systematic funds that sold off when big tech companies weakened on their momentum and low-volatility factors:

Marko Kolanovic, a senior JPMorgan strategist, estimates that passive and quantitative investors now account for about 60 per cent of the US equity asset management industry, up from under 30 per cent a decade ago, and reckons that only roughly 10 per cent of trading is done by traditional, “discretionary” traders, as opposed to systematic rules-based ones.

We have talked occasionally about the possibility that the modern low-volatility environment might be not so much a sign of complacency as a real evolution in the markets: Stock prices famously overreact to news, but as markets are increasingly run by coldly logical machines, those overreactions will diminish, and stock prices will become less volatile. I guess that is one way to interpret the fact that only 10 percent of trading is done by humans making their own decisions? The other is that volatility is being smoothed day to day but concentrated into the tails: The systematic funds will all flock together, mostly keeping prices smooth and stable, but when they do panic, they will all panic at the same time.

Share-count investing.

Yesterday I asked:

To what extent is there a viable investing strategy of (1) actually reading corporate filings, (2) buying companies that have fewer shares than the market thinks they have, and (3) shorting companies that have more shares than the market thinks they have? 

I was half-kidding, but a reader pointed out that in 2015 Muddy Waters Research put out a note arguing that Bolloré SA was undervalued because it had fewer shares than the market thought it had:

BOL effectively owns a significant portion of itself. Most of the market seems to take at face value BOL’s reported shares outstanding of 2.5 billion. We estimate that the effective shares outstanding are only 1.1 billion (57.2% lower than reported).

Muddy Waters forecasted a 95 percent upside, and the stock is actually down almost 10 percent since then, but still, share-count investing: could be a thing.

People are worried about vampires.

"No, Peter Thiel is not harvesting the blood of the young," is the headline here, and next they will be telling me that he doesn't sleep hanging upside-down from his ankles in a crypt. I ask you: Would you rather live in a world in which Peter Thiel is harvesting the blood of the young, or one in which he isn't? Do you want to be disabused of your belief that Peter Thiel might, just might, get himself injected with the blood of young people to become immortal? What fun is that? 

Things happen.

Fed Raises Rates, Sets Out Plan to Shrink Asset Holdings Beginning This Year. FOMC issues addendum to the Policy Normalization Principles and Plans. Wells Fargo Is Accused of Making Improper Changes to Mortgages. Dana Gas Plans Profit Rate Cut on Sukuk in Restructuring Twist. Saudi Aramco IPO Plans Slowed Over Where to List. Anbang chairman’s detention underlines China business risks. Venezuela Turns to Its Secretive Bag of Tricks to Avoid Default. Large investors attack Dutch plans to curtail foreign takeovers. Fed Considers Easing Capital Rule Seen as Hampering Swaps Market. Morgan Stanley to let financial advisers text clients. Ex-Visium Analyst Gets 18 Months for Inflating Bond Values. Goldman Sachs raises $7 billion for new private equity fund. Inside Formula One’s race for global domination. Israeli Watchdog Head: “Binary Options are Like Blood Diamonds.” Phil Falcone Bets on a Casino (Where the Locals Can’t Gamble). York Capital Names Two Eventual Successors for Founder Jamie Dinan. "Since taking office in 2011, Cuomo has consistently steered money intended for the subways toward road projects instead." Fox News Is Dropping Its ‘Fair & Balanced’ Slogan. Meet the Goopies. Microsoft A.I. achieved maximum 'Ms. Pac-Man' score. Very good dog crashes every shot of this South Korean island’s street view map. 

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.

    To contact the author of this story:
    Matt Levine at

    To contact the editor responsible for this story:
    James Greiff at

    Before it's here, it's on the Bloomberg Terminal.