Money Stuff

Uber Secrets and Trading Insights

Also a compilation of worries and bitcoin futures.

Oh, Uber.

Look, I get it. I can see why you would do this stuff:

Uber used encrypted, ephemeral messaging “to protect sensitive information and ensure we didn’t create a paper trail that would come back to haunt the company in any potential criminal or civil litigation.” It also used “non-attributable devices” to communicate with third parties providing on-the-ground intelligence about threats to Uber, and to research Uber protestors.

Here's the problem: That's from a 37-page letter that Uber Technologies Inc.'s "former manager of global intelligence, Richard Jacobs," sent to Uber's deputy general counsel. That letter is now in the hands of U.S. prosecutors, who shared a copy of it with Waymo, Alphabet Inc.'s self-driving car unit that is suing Uber for theft of trade secrets. Using ephemeral messaging to avoid creating a paper trail that will haunt you in potential criminal or civil litigation? Sure, yes, great idea. (Not legal advice!) Writing a 37-page memo about it and then providing that memo to the other side in criminal and civil litigation? Terrible idea! 

To be fair, Jacobs's letter was sent as part of a dispute with the company, and Uber's lawyers "suggested it was written to get a payment from Uber." (It worked.) But still it seems like a failing in the system. If you are taking the time to get employees to do their plotting in disappearing messages, you should also make sure they're not airing their grievances by certified letter with a copy to the U.S. Attorney.

We have talked a few times recently about Uber as a regulatory-evasion company, and I have noted that its form of "regulatory entrepreneurship" is different from the "regulatory arbitrage" that I am familiar with from the financial industry. In regulatory arbitrage, you follow the rules, but in creative ways, pushing them to the limit in order to achieve your aims. In regulatory entrepreneurship, you just break the rules and then sneeringly tell the rulemakers to keep up.

It is tempting to take Jacobs's letter -- which also says that Uber has a unit that "exists expressly for the purpose of acquiring trade secrets, code-based and competitive intelligence," and that the company "offered training on how to 'impede, obstruct or influence' ongoing legal investigations," including through the "use of attorney-client privilege on written documents, and encrypted, ephemeral communications" -- as more evidence of Uber's penchant for rule-breaking. But I am not sure that it actually broke any rules here. Using attorney-client privilege and ephemeral messaging to avoid a paper trail, trolling Github for competitors' code: Generically, those things might be fine. (Not legal advice!) "While he didn’t believe that work was illegal, 'I had questions about the ethics of it,' Jacobs testified" at a hearing in the Uber/Waymo case yesterday.

This is a different sort of story about Uber. This is a regulatory-arbitrage-ish story, but at a company -- in an industry -- whose culture is very different from that of the financial industry. None of this stuff is surprising, exactly, if you have worked at a bank, or if you have followed the last decade of scandals at banks. Lots of bankers and traders are dimly and dangerously aware of the magic incantation of "privileged and confidential." Everyone knows about the emailed conversations that end with "LTL," "let's talk live," just as they are getting to the good bits. But it's not, like, in the employee manual. There's no formal training in how to be sneaky. When David Viniar told a Senate panel "I think that's very  unfortunate to have on email," about something that was very unfortunate to have on email, there were audible gasps. Everyone knows that there are some things that you don't email about, but you keep up the polite fiction that it's because you don't do those things, not because you do them on Confide or Signal.

But the innovation at Uber -- and in Silicon Valley more broadly -- is to make explicit what traditional companies leave implicit. It turns fuzzy human intuitions into explicit algorithms. 

Lawyer: We don't have to give that email to the other side in this lawsuit; it is marked "Privileged and confidential" and was sent to a lawyer.
Engineer: Wait that's how it works?
Lawyer: Well there are many factors --
Engineer: So if we mark all our emails "privileged and confidential" and always cc a lawyer, no one will ever be able to see them?
Lawyer: I mean not --
Engineer: I've put that in the training manual and written code to automate it.

It feels weird if you are used to subtler and more genteel forms of legal optimization. But of course this is kind of Uber's schtick. Raising prices when demand is high is totally normal human behavior, the sort of thing that is blandly explained early on in economics textbooks. Writing an algorithm to do it automatically, and bragging about that, and calling it "surge pricing," is ... a little ... unseemly? One thing that Uber is doing is a sort of norms arbitrage, a bet that it can get away with doing openly and always things that other people do quietly and occasionally. (It's probably right: We once talked about a survey finding that Americans think Uber's surge pricing is fairer than "florists raising prices on holidays." The florists are fatally sheepish about it; Uber's blatancy works.) 

And it's not just Uber. So much of the tech industry is about formalizing informal human behavior, making it machine-legible, automating it, and then pushing it to weird limits. Facebook is, always, the best example: Do you sometimes wish your friends a happy birthday? What if you could wish all your friends happy birthday, every time one of them has a birthday, just by pushing a button? Wouldn't that be better? How could it not be? If humans do something sometimes, why not build algorithms to do it constantly? Why not scale it up?

Outsider trading.

"Insider trading is not about fairness," I often say; "it's about theft." It's not a crime to know things that other people do not know, and to trade on your knowledge. In fact it is good. In fact it is the point of securities markets, or point of them anyway: Society wants capital to flow to good projects and not to bad ones, and so we encourage people to work hard to get information about which projects are good and which are bad, and to buy the good ones and sell the bad ones. 

On the other hand, we don't want people in positions of trust at companies to just steal from the shareholders. If you are the chief executive officer of a company and you know it will have bad earnings, you could dump all your stock, but that seems like a breach of your duties to your shareholders. When outsiders try to get good information about companies they make markets more efficient and socially useful; when insiders use their positions to enrich themselves and their friends, there are some downsides. 

Here is Stuart Kirk of Deutsche Bank worrying about a dystopia in which specialized professionals use advanced technology to make capital-allocation decisions that are almost as sophisticated as the decisions Inc. makes in deciding how to price soap:

The latest trend, however, is the pursuit of observational truths. A satellite supported with clever software can potentially see what high street shops people are going into. Are they leaving with one bag or three? How many cars is that locomotive pulling to the port in Karratha, Western Australia? What blend of iron ore does its colour indicate?

Soon it is likely that investors will know more about the fortunes of a business than a line manager relying on weekly or monthly reports. This changes everything. For a start, the definition of insiders and outsiders becomes redundant. So, therefore, does the regulation.

People find this worrying! I don't, and don't even really understand why people find it worrying. "How should regulators respond to a world where some investors listening to a quarterly earnings call already know the exact number of trucks that left the company’s warehouse over the summer" asks Kirk, "with shadow length indicating how weighed down they were?" Why should they have to respond? Why is learning about the world bad? Why is it a matter for regulation? It is not generally in other fields: If a television studio observes what people like and then tries to make shows that fit their tastes, no regulator steps in to punish them, or to demand that they share their research with everyone.

"The nuclear option would be for regulators to make a clear distinction between data and analysis, then insisting that every scrap of information about a company is published as soon as it comes in — from insiders and outsiders alike," says Kirk. But why "every scrap of information about a company"? Why not every scrap of information about everything? Why is the understanding of capital markets a uniquely public good where everyone needs equal access, while the understanding of every other aspect of the world is a matter left to individual effort and insight and technology? 

The endgame here is that sophisticated professionals will seek out investing edge using data and computing power that is not available to "the average mum and pop investor," which will drive those ordinary investors away from trying to beat the market on their own. If they want decent performance, they will be forced to choose between passive indexing or hiring those sophisticated professionals to manage their money. People, again, find this worrying, and again I do not. (Again I think it is obviously an improvement!) We do not generally expect the average amateur to outperform the most sophisticated professionals in dentistry or brain surgery or plumbing or basketball or any other sphere of human activity in which there are amateurs and professionals. But for some reason that expectation is built into many people's thinking about stock markets, with constant bizarre results.

People are worried about crowded quant trades, "cyber," China, disruptions at the close, crowded index trades, a bitcoin bubble, recessions, "a failure of confidence somewhere," Japan, oil, BoNY Mellon's repo dominance, low volatility risk premia and geopolitical disruptions.

Some days it feels like this newsletter is just a compilation of things that people are worried about, but good news, yesterday Bloomberg actually published a compilation of things that people are worried about, so you can go read that. If I had to pick two broad themes from these worries -- which come from interviews with prominent markets people -- they might be:

  1. people worrying about the things everyone else is excited about; and
  2. plumbing.

Worry 1 is of course the most classic of all market worries: If a thing goes up a lot, you might worry that it will come down; if it's popular, you might worry it's a fad; if there's a rally, you might worry it's a bubble. It is a constant of human psychology: We get overenthusiastic about something, and turn out to be wrong. Worry 2 -- what if the stock market breaks at the close? what if putting all repo transactions into one bank increases failure risk? -- is a bit more interesting; it is not a contrarian take on other people's enthusiasms, but something orthogonal to them. You might buy all the right stocks, but if the stock market shuts down because of the cyber then you can still end up in the soup.

Bitcoin futures.

wrote last week:

If you are buying bitcoin futures, it is probably because you think that bitcoin will go up. If you think that bitcoin will go up, it is probably because you think it will be more widely adopted as a currency and a store of value and an alternative to the current financial system. But if you really thought that, you'd just buy bitcoins. You're buying the futures because, deep down, you prefer the existing system. "The need for a bitcoin ETF," I once wrote, and it is just as true of a bitcoin future, "is an argument against buying it."

A number of readers emailed to object. The purpose of bitcoin futures isn't just to avoid buying bitcoins, after all. A major purpose -- though a controversial one, given bitcoin's volatility -- of bitcoin futures is to provide leverage. (Another purpose is of course to allow convenient short selling, but shorting bitcoin through futures is perfectly consistent with distrusting bitcoin's alternate financial system.) If you are really confident that bitcoin is going to go up, then you should buy all the bitcoins you can, and doing that with leverage through futures may be the most efficient way to do so.

That's fair. My point was really about the fact that bitcoin futures are cash settled: When your bitcoin futures expire, you get paid any profit or loss in dollars, rather than getting delivery of bitcoins. This is unlike many commodity futures markets, where when your oil futures expire you get oil. The implicit message is that it's more of a pain to store bitcoins than to store oil, though you could interpret it other ways. (For instance: All futures are financial bets and should be cash settled, and physical settlement of oil futures is an inconvenient anachronism.)

Also: If your view of bitcoin is that it will replace the existing financial system, then shouldn't it provide a way to get leverage? Bitcoin exchanges generally offer shorting, margin, etc., but the bitcoin blockchain is rather literal. There is no fractional-reserve banking on the blockchain, no way to sell more bitcoins than you own or buy more than you can afford. Many bitcoin evangelists think that this is a feature, not a bug, but I suppose the people buying on margin disagree.

One reader also emailed to ask how bitcoin futures will handle forks. CME Group, one of the exchanges that will offer bitcoin futures, answers on its website: "A hard fork of the bitcoin blockchain" is one listed example "of unusual and extreme circumstances," in which case the administrator of the futures contract "shall be responsible for recommending the necessary actions and responses to ensure the relevance and integrity of the Bitcoin Pricing Products." Elsewhere, it says that it "is developing a hard fork policy for capturing cash market exposures in response to viable forks. The policy may involve cash adjustments to position holders or listing related futures that are also issued to position holders." This will involve bitcoin futures exchanges, as it has involved bitcoin exchanges, in the determination of which forks are "forky" (CME says "viable") enough to require adjustment -- so that a holder of bitcoin futures is entitled to (the cash value of) both original and forked bitcoins -- and which ones can be ignored.

Elsewhere in bitcoin futures, here is Izabella Kaminska arguing that cash-settled bitcoin derivatives could lead "to the sort of anomalies that distort price discovery until the mother of all corrections can occur." Here is pseudonymous blogger Kid Dynamite arguing that "excess demand for BTC-F [bitcoin futures] will translate into upward BTC pressure more readily than excess supply of BTC-F would translate into downward BTC pressure." And here is Gillian Tett arguing that futures could domesticate bitcoin:

Until now, bitcoin has essentially been perceived as a ringfenced product. The unstated assumption among investors has been that it occupied a distinctive space where normal investing rules did not apply. But as bitcoin becomes integrated into the derivatives market, that boundary may break down and investors apply for “normal” criteria in their valuations. Given the unresolved questions about how bitcoin works, that may lead to more scepticism; indeed, it is precisely what has often happened in the evolution of financial markets before. 

And elsewhere in bitcoin: "Bitcoin Mining in Electric Vehicles Raises Other Questions." It sure does. It sure does. (Like: Are you driving for Uber while mining bitcoins in your Tesla?) This is a fun dumb arbitrage -- "Teslas and other EVs have free access to power at many charging stations, so it was probably only a matter of time until somebody decided to plug their mining computers in" -- and also a good reminder of the surprising interconnectedness of all things, or at least all buzzy modern things. It feels like there is a natural law lurking here somewhere, an equation describing the universe, Bitcoin equals Uber Tesla squared, that sort of thing.

Also bitcoin hit $10,000, yay.

Things happen.

UK bows to EU demands with breakthrough offer on Brexit bill. Judge Backs Trump Administration in Fight to Control CFPB. Vanguard Is Finally Bringing Its Active Smart Beta ETFs to the U.S. London Stock Exchange CEO Xavier Rolet to Resign Immediately. The Odyssey of a Turkish Trader Now Spilling His Secrets in U.S. 20 New Residential Towers Are Being Built in Frankfurt Because of Brexit Demand. Supreme Court Questions Whether Dodd-Frank Protects All Whistleblowers. Hackers Now Have Incredibly Sophisticated Ways to Breach Banks' Defenses. Inside Valeant’s Efforts to Be a Normal Drug Company. Potential EU leaving names. The Complicated Legacy Of A Panda Who Was Really Good At Sex.

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.