Flash Crashes and Trump Trading

Also antitrust and index funds, unicorn swaps, Wells Fargo, Liberland and perpetual motion.

Programming note: This will be my last Money Stuff for the next few weeks, because I will be on paternity leave. Thanks for reading, and I'll see you in January.

Flash crash.

Aww, poor guy:

The Bank of England has said publicly that the October 7 crash is “set apart by the lack of a clear fundamental trigger” but its investigation of the event focused on a single incident, according to a person briefed on the probe.

People with knowledge of events at Citi that day said one of the US bank’s traders placed multiple sell orders when the currency slumped in unusually fragile market conditions. One of the people said the trader “panicked”.

That's about the October 7 "flash crash" of the pound in thin overnight trading. There's a common explanation of flash crashes in which a falling price triggers a bunch of stop orders, forcing automatic selling at a time when there aren't any buyers, pushing the price down further and triggering another wave of stops. In this story, the stop order -- an automatic order to sell if the price gets too low, meant to limit an investor's risk in a falling market -- is the culprit. It's a sort of crystallized form of pre-panic: The investor surveys the possible future states of the world, decides that if the pound falls below $1.25 she'll panic, and gives her broker instructions to panic for her if she's not around to do it herself.

But the stop order is just a tool to implement the panic. You don't need the stop orders to have the panic. You can just be there, and panic. (Or: You can not be there, because it's the middle of the night, and your less experienced trader in Tokyo can panic for you.) The reason stop orders exist is because they correspond to a popular human behavior: When the price of your position falls, you want to get out of it, so you sell as quickly as you can. This is not always a smart behavior -- sometimes it is! -- but it is common enough that an order type exists to automate it. But you can have the behavior without the order type. That seems to be what happened here.

There's another common explanation of flash crashes in which they are caused or exacerbated by some sort of "computer glitch." The intuition is that humans are good and normal at trading, but computers add weird randomness to the market in ways that are beyond human comprehension. There's some suggestion of that here; the Financial Times reports that "the sell instructions from Citi’s desk in Tokyo kept coming and started tripping over each other in a pattern known as 'looping' that is normally constrained by safety nets embedded in bank trading tools."

But it's also possible to read this story as a very simple one: The price of the pound went down a bit, and a trader panicked and tried to sell a lot of pounds before the price went down any more, and there was no one around to buy, so his selling pushed the price down much more. Also there were computers around. The computers implemented the human decisions -- perhaps incorrectly, and certainly quickly -- but the human decisions by themselves would be perfectly sufficient to explain why the pound went down.

Trump trading.

Most U.S. presidents do not spend a lot of their time giving investment advice, though there are some precedents. Barack Obama put out a long-term buy recommendation on stocks in March 2009; the S&P 500 hit its bottom a week later, and has grown by 227 percent since. But Donald Trump is taking a different, more activist approach, tweeting his plans to reward and punish individual companies, and while he has yet to include tickers or price targets in his tweets, investors have started to notice:

Efrem Hoffman, founder of a market analysis firm called Running Alpha, said Trump's tweets represent a new source of market information for those willing to study them and identify patterns.


"An algorithm could have easily gotten the direction of the trade wrong today. However, given Trump's penchants for simple language and negative statements, it wouldn't surprise me if someone were able to develop an accurate predictive algorithm."

Yeah it does not seem all that difficult? Like, if you find a company name in a tweet, and the tweet ends with "Sad!" or "Bad!" or frankly just an exclamation point, you can probably go ahead and sell it for a quick profit. We live in fascinating times for markets, and technology, and politics.

What if Trump shorted (or bought) a stock, and then tweeted his plans to punish (or reward) that company to move markets and make himself a quick profit? Would that be insider trading? This is a hypothetical that I refuse to address, because I am trying to cling to a little bit of dignity around here. But I suppose we'll have to talk about it if it ever actually happens, which will probably be like next week.

Trump's latest Twitter attack was on Boeing Co., where he was at some point a shareholder. But "the president-elect sold all his stock holdings in June, according to Jason Miller, a spokesman," though "is not possible to independently verify Mr. Trump’s stock sales." Meanwhile, here is some speculation from Steven Davidoff Solomon and Michael de la Merced about how Trump could sell off the rest of his business interests. The private real estate holdings seem relatively straightforward, but I am puzzled by how anyone thinks Trump could get rid of the parts of his business that consist of branding and licensing his name. They write:

In some ways, these are comparable to Martha Stewart Living Omnimedia, which Sequential Brands Group acquired last year for $353 million. It’s unclear how any company that bought Ms. Stewart’s business could continue to profit from the brand without her participation. The same goes with Mr. Trump’s branding businesses.

It would be fun to try, though, right? Like what if he sold off 100 percent of the business -- to a private buyer, or in an amazing IPO -- and it became an independent company with branded Trump Hotels? And foreign diplomats paid extra anyway to stay at Trump Hotels because, you know, his name's still on the door, and he'd probably find that flattering? And then one day he took to Twitter to say "Trump Hotels are overcharging diplomats and using immigrant labor. We're going to crack down! Bad!"

Will Donald Trump make index funds illegal?

You know how, last month, I mentioned that Donald Trump's presidential campaign was a blank canvas for the projection of everyone's wishes, from white nationalists to Fannie Mae privatizers? And you know how, a couple of weeks later, I mentioned a new paper from Eric Posner, Fiona Scott Morton and E. Glen Weyl, arguing that the government should use its antitrust powers to restrict large institutional investors from owning shares of more than one company in the same industry? Can you guess where this is going? Here are Posner, Morton and Weyl on Trump:

He channeled populist anger against elite corporations by, for example, calling the proposed merger between AT&T and Time Warner “too much concentration of power in the hands of too few.”

But the real challenge to competitive markets today does not come from mergers like this one. The great, but mostly unknown, antitrust story of our time is the astonishing rise of the institutional investor ...

Oh yes: They argue that Trump, as the champion of the working class, should implement their plan to restrict cross-holdings by institutional investors, because those cross-holdings are bad for workers and consumers:

Vanguard and BlackRock are the largest owners of Apple and Microsoft, and among the top three owners of CVS, Walgreens and Rite Aid. If you zoom down to, say, the market in cooking stoves, you will see that the largest owners of two of the three major competitors — GE, Whirlpool and Electrolux — are Vanguard, BlackRock and State Street. The same patterns appear in airlines, soft drinks, you name it.

Economic theory tells us that when a single investor owns large stakes in competing firms, the investor will want firms to keep prices high and wages low. Price and wage competition lowers profits and stock values.

I will not hold my breath for the Trump administration to crack down on index funds, or to be a vigorous antitrust enforcer generally. But Trump's apparent policy of picking winners and losers on Twitter does suggest a certain philosophical difference with diversified investors who passively own all the companies in every industry. So maybe he'll go for this!

Elsewhere in Trump and regulation: "Top US bankers look to Trump administration to help big finance." And: "Ten Key Implications of Donald Trump’s Electoral Victory for Financial and Securities Regulation."

Illegal unicorn swaps.

The thing about private tech companies is that there is no public market for their stock. If there were a public market for their stock, they would be public tech companies. Historically, this was a pretty straightforward distinction: Private companies were small, and owned by their employees and a few committed venture capitalists, none of whom needed to sell their stock. And they were weird, risky, and unknown, so no one wanted to buy their stock. Eventually the companies and employees and venture capitalists would decide they wanted to sell, and so they'd file the paperwork to become public companies, and then go on a roadshow to explain the company and find investors to buy the shares.

But now private tech companies are, like, Uber, and Airbnb, and Snap. They are massive. They have lots of investors, including big public mutual funds and employees (and former employees) with tons of stock that they'd like to diversify. And they are household names, with lots of people clamoring to invest even without any financial information. Gains from trade are possible. And yet: They are private. There is no public market for their stock. Many companies have lost interest in going public. They can raise money and retain employees without it, and they can keep more control -- and provide less disclosure -- if they stay private.

Markets abhor a vacuum, and businesses have sprung up to solve this problem. But it is legally tricky. Trading is restricted by the securities laws, which mostly limit trades of private stock to "accredited investors" (rich people and institutions). It is also sometimes restricted by the companies themselves, which can put more limits on trading by their employees and investors than public companies can. To get around those restrictions, some people have set up substitute markets for trading swaps on private stock. If I'm an Uber employee, I keep my Uber stock, but I write a contract promising to give you money if the stock's value goes up, and you promise to give me money if the value goes down. For instance:

Equidate Inc. sought to provide liquidity for employees of private, growth-stage companies in the Silicon Valley and others holding restricted shares of their stock, and its platform essentially matched these shareholders with investors seeking to invest in the potential economic return on those shares. Equidate conducted transactions through contracts that its subsidiary entered into with the shareholders and investors, and payment provisions were triggered by such events as a merger, acquisition, or IPO at the underlying company.

But that is from a Securities and Exchange Commission settlement with Equidate, because this is really no less legally tricky than trading the shares is. Equidate's contracts were "security-based swaps," and swaps need to either be traded on exchanges or limited to "eligible contract participants" (even richer people and institutions). Equidate couldn't put its contracts on public exchanges for a number of reasons, the main one being: There is no public market for the stock. And while it asked investors if they were accredited investors, it allegedly "did not take any steps to confirm whether any shareholders or investors were eligible contract participants." So it got in trouble and had to pay $80,000 to the SEC, which tsked:

“Market participants are free to capitalize on the growth of private technology companies in the Silicon Valley or elsewhere, but laws must be followed to ensure security-based swaps are registered and sold through platforms where investors have full disclosure and protections,” said Jina Choi, Director of the SEC’s San Francisco Regional Office. 

The second part of that sentence sort of kills the first part. You're free to build a market for private technology stocks, but you have to follow the disclosure and investor-protection rules, which means that you can't really build a market for private technology stocks. Because those stocks don't have the full disclosure and investor protections of public stocks. If they did, they'd be public. There is an intense and widespread yearning for public markets for these private companies' stocks. But it doesn't work that way.

Oh, Wells Fargo.

We have talked before about Wells Fargo & Co.'s efforts to arbitrate its disputes with customers over the millions of fake accounts that it opened in their names, and I expressed some sympathy for Wells's efforts to avoid class actions. But I must confess that I missed a pretty obvious problem with Wells's approach:

Ms. Zeleny, a lawyer who lives outside Salt Lake City and opened a Wells Fargo account when she started a new law practice, said it would be impossible for her to agree to arbitrate her dispute over an account that she had never signed up for in the first place.

Umm. Right? If you set up a real account, you agree to arbitrate it in your account agreement. If you set up a fake account, the account agreement is fake, and the agreement to arbitrate is also fake; it seems rough to enforce it against you.

Oh, Liberland.

Well here is a delightful story about Liberland, a libertarian utopia on three square miles of swampland on the Danube between Serbia and Croatia. (I mean, a supposed utopia supposedly on that swampland; the Liberlandians are not actually allowed into their territory.) It is everything you'd expect: disorganized, mostly male, and very into bitcoin:

Liberland receives $10,000 monthly from Bitcoin's angel investor Roger Ver, and the President has repeatedly vowed his commitment to keeping the cryptocurrency as the basis of the nation's economy. At the conference, however, Herve Lacorne, founder and CEO of Trade Solutions Group, introduces a proposal for a central bank that will also deal in fiat currencies. A few attendees seem caught unawares. George, a Hungarian wearing an "I (Heart) Bitcoin" t-shirt who later identifies himself to me as an independently wealthy trader, wants to know whether a central bank dealing with fiat will decrease the large percentage of donations to the country that now come in via Bitcoin.

Contrarian indicators?

What does it mean when the Wall Street Journal runs an article with the headline "The Economy's Hidden Problem: We're Out of Big Ideas"? I feel like now might be a good time to go long perpetual-motion-machine stocks. I guess those are hard to find. Maybe human-immortality stocks, like Alphabet Inc.?

People are worried about unicorns.

I'm gonna say that the Equidate thing counts. Also: "Will More Unicorns Take the IPO Plunge in 2017?"

People are worried about bond market liquidity.

Here is a presentation from market structure expert Larry Harris about "Transaction Costs, Trade Throughs, and Riskless Principal Trading in Corporate Bond Markets," and I suppose if you are a market structure expert, the corporate bond market must look a little horrifying. Harris notes that "The average customer roundtrip transaction cost was 125 bp, or about 4 months interest for a 4% bond," which is "Equivalent to 50¢/share for a $40 stock!" And "47% of all trades trade through a standing quote." The stock market is a quote-driven market, where bids and offers to buy and sell are publicly available, and you can generally expect to be able to sell at the bid or buy at the offer. The bond market is ... different.

Me yesterday.

I wrote about the Supreme Court's insider trading decision in Salman v. United States. 

Things happen.

JPMorgan, HSBC, Credit Agricole Fined $521 Million Over Euribor. Credit Suisse Steps Up Cost Cuts as Revenue Eludes CEO Thiam. J.P. Morgan, Bank of America Trading Activity Continues Strong After Election. Banks’ growing capital surplus since the crisis. MetWest Surpasses Pimco as Largest Active Bond Fund Manager. China to help 1MDB settle multibillion-dollar legal dispute. Burned Carlyle Investor Backs Suit Against Private-Equity Firm. "After saddling investors with a 99.99 percent loss over the past five years, the Hong Kong-listed property firm is asking shareholders to plow fresh capital into the business for the eighth time in five years." Bain & Company is the best place to work. CEOs Enjoy Richer Perks. Pfizer fined £84m for charging ‘unfair’ price for epilepsy drug. Ad Agencies Probed Over Contracts to Produce Commercials. Supreme Court Gives Samsung a Reprieve in Apple Patent Case. Deciding Who Should Get First Dibs in Bankruptcy Liquidations. More from Elaine Ou on smart contracts and legal enforcement. Apple’s Top Assembler Foxconn Confirms Plans for U.S. Investment. After Meeting Trump, Japanese Mogul Pledges $50 Billion Investment in the U.S. Poll: Republicans Are Now More ‘Anti–Free Market’ Than Democrats. "Unless you’re building the Taj Mahal and using hand tools, it shouldn’t take that long for facade work." Non-traditional dog names. Therapy pig.

If you'd like to get Money Stuff in handy e-mail 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.