Stop Triggers and Quiet Defaults

(Bloomberg View) -- FX trading.

A fun thing to do, if you are a trader at a bank, is to trigger stops. If a customer has an order to sell a thing when it gets to 80 or below, and it is currently at 82, then you might consider selling the thing short to push down its price to 80. At 80, you know someone else will be selling, because you have a customer order to sell at 80. So you sell short at 82 and 81, and the stop triggers, and your customer pukes out her position at 80, and you go ahead and buy in your short at 80 or 79 or whatever, making a nice little profit.

This is fun but also obviously very much frowned upon. It is not good customer service: Your customer put in the stop order to limit her losses, and you went ahead and handed her exactly the losses she was worried about. It is probably -- using even a reasonably narrow definition of the term -- "front-running." In the foreign exchange markets, it is forbidden by the FX Global Code (Principle 10). It's probably illegal most places. But at least in theory, it might be hard to prove: Maybe you sold at 81 to lay off risk, or to accommodate customers, or because you foresaw the market moving against you. How can anyone prove that you did it to trigger stops?

The answer is usually "because you sent your colleagues dumb chat messages high-fiving about how you made so much money triggering stop orders," but here is a novel variation on that answer:

From at least April 2010 to June 2013, Credit Suisse employed an algorithm designed to trade ahead of clients’ limit and stop-loss orders. Credit Suisse programmers designed the algorithm to predict the probability that a client’s limit or stop-loss order would be triggered. Credit Suisse programmers and traders had ongoing and significant discussions about how the algorithm could be used to “front run” customer orders. Credit Suisse predicted that this particular algorithm would generate approximately $2 million in profits for the calendar year 2013.

Huh. That ... is .... terrible ... and yet ... kind of cool? I said the other day:

A useful way to think about modern electronic market structure is that in the olden days humans traded stocks and options, making markets based on gut instinct, and then those humans were replaced by computers that used algorithms that largely replicated the humans' gut instincts but more efficiently. But also, in the olden days, those humans did various shady things, and over time the computers have started to replicate the humans' shady-thing-doing abilities, because, you know, the shady-thing-doing tradition runs deep.

But this story is both of those things. On the one hand, yes, obviously, front-running customers by triggering stops is shady, and automating it is automatically shady. But on the other hand, triggering stops is an imprecise art: If the thing is at 82 and you sell it short, you can't be sure that you'll trigger the stop, and if there's some big buyer on the other side then you could lose money. You have to carefully evaluate the risks and rewards, and consider the probability of actually triggering. Usually that is an instinctual call. But Credit Suisse AG found a way to more efficiently replicate its traders' gut instincts -- about front-running clients -- with an algorithm. 

Anyway the New York Department of Financial Services fined Credit Suisse $135 million for this and other bad stuff in its foreign exchange business. The other bad stuff is enjoyable in its way, but it is mostly stuff we have seen before. Credit Suisse's human FX traders allegedly shared information with traders at other banks to manipulate fixing windows; that conduct has been fined many times before. There was also just random casual manipulation among friends:

On September 7, 2012, a Credit Suisse customer (“Customer 1”) enlisted the assistance of a Credit Suisse trader, Trader 18, in seeking to push down the price of the U.S. dollar/Turkish lira pairing. Customer 1 asked Trader 18, “can you walk down usdtry for me pls.” Trader 18 replied, “Yeah, no problem.” Customer 1 then stated, “just offer 1 at like 72 . . . just walk it brotha,” to which Trader 18 replied, “No sweat.” Customer 1 cheered on Trader 18, saying “come on . . . just walk it,” to which Trader 18 replied, “Collapsado.” Apparently upon achieving success, Customer 1 stated, “thks [Trader 18] for walking it down . . . great job . . . you really shellacked it.” Trader 18 quickly replied, “pleasure.”

Credit Suisse's electronic system also allegedly used its "last look" capabilities in customer-unfriendly and poorly disclosed ways: Credit Suisse advertised that it used last look to protect itself from "toxic order flow," but ended up using it widely to increase its profits. (This is something that we have also talked about before, when Barclays Plc did it and was also fined by the New York DFS.) Its Head of Strategy wrote in an email:

We should think about whether we might want a small amount of deal acceptance (last look) for all FIX clients by default. . . . Right now we typically only use deal acceptance when there’s a problem but I think it would be worth doing it as standard and having it turned on from day one. We should . . . instead use our rejects to improve P&L. I’m not suggesting going last look crazy I just think there’s no need to have a 100% fill rate for API clients. With most there’s an ‘acceptable’ reject rate and by not making use of that – or wasting it by rejecting trades for non-P&L-impacting reasons – we’re leaving money on the table (or at least giving up a free option.)

Others disagreed, and the Head of eCommerce sensibly suggested they take it offline: "[T]ightening P&L pressure should never be a cause for reversing matters of principle. If you disagree with this we should talk rather than exchange further emails." Those are two fantastic sentences: We will never compromise our principles for money, but let's discuss further in person. Guess how those in-person conversations went! "Despite these concerns, Credit Suisse expanded the application of Last Look by applying a default setting to almost all eFX customers, without regard to whether the customer posed a trading risk to the Bank." 


Venezuela's default on its debt continues to be curiously non-event-y. "Venezuela and its state oil company are now officially in default," write Bloomberg News's Ben Bartenstein, Patricia Laya , and Katia Porzecanski about the big ratings agencies' declarations of default. "It changes nothing for bondholders":

What really matters to bondholders is that officials continue to say they’ll honor obligations and in fact seem to be making determined efforts to do so, with the government information minister affirming Tuesday that delayed sovereign payments were being made. Speaking Monday before a gathering of creditors summoned to downtown Caracas for restructuring talks, Venezuelan Vice President Tareck El Aissami pledged that the oil-rich nation would work to find new ways to deliver bondholders’ money.

Casting blame on international financial institutions for holding up the funds, he pointed out that PDVSA delivered $2 billion of payments within the past couple of weeks, part of a total $70 billion that President Nicolas Maduro says the government has shelled out since he took office in 2013.

Venezuela might be running out of money, but it is not yet out of money. It might actually be in a situation a little like that of Argentina in 2014 through 2016: It has the money, and wants to give it to bondholders, and the bondholders want to take it, but the workings of the international financial system are holding it up. (A U.S. court injunction in its pari passu litigation was blocking Argentina; U.S. sanctions on the Maduro regime might be delaying Venezuela.) If you're a bondholder, it's time to start thinking hard about the plumbing. 

Elsewhere in plumbing:

Many traders were waiting on Tuesday for a recommendation on whether Venezuelan bonds should be traded without accrued interest, said Siobhan Morden, managing director at Nomura. Such a determination is made by the Emerging Markets Traders Association when a country isn’t expected to pay interest on its bonds going forward. The EMTA said it was unable to reach a consensus on the matter Tuesday.

All the ratings agencies have said that Venezuela is in default, but the more practical question of whether Venezuela's bonds should trade like they're in default is still unresolved. If it's going to keep paying interest -- a bit late, a bit annoyingly, but pretty consistently -- then they should probably trade with accrued interest; if it's going to go into real default and restructuring talks then they probably shouldn't. Nobody quite knows what the answer is yet.

And here is a consequence of sanctions that I hadn't really thought about:

Because of sanctions, it has been unable to hire a team of top bankers and lawyers who might help reach a favorable agreement with creditors. ... Unusually, the government has asked bondholders to come up with a plan for restructuring the debt. In most cases when a sovereign nation runs out of cash, a debt proposal is imposed on investors.

I have to say that if your debtor tells you "hey can you propose a plan for how much less I should pay you, but meanwhile I am just going to do my best to keep paying you the full amount," you should probably not be in too much of a rush to come up with a plan.


If a group of investors led by SoftBank Group Corp. buys some shares from Uber Technologies Inc. at a $68 billion valuation, and it buys some other shares from existing investors in a tender offer at a $50 billion valuation, then how much is Uber really worth? There is an objectively correct answer to this question, which is, "who cares": A willing buyer paid one price for some shares and another price for some other shares and all the willing sellers got the amount of money they agreed on, and putting a headline number on the whole thing is just meaningless scorekeeping. But of course people love scorekeeping, particularly in the Enchanted Forest of the Unicorns: Since Uber's stock doesn't trade on an exchange, there is no public record of its value changing from day to day, and it can point to its last fundraising round as its still-current "official" valuation, whatever has happened since. But if it does a new round at a lower price, it won't be able to play that game any more. And as Uber gets closer to an initial public offering, that game becomes more important: It's harder to argue for an $80 billion IPO after a $50 billion private down-round.

There are two separate things going on here. One is that the shares SoftBank is buying from Uber are actually worth more than the shares it is buying from existing investors. If Uber buys common stock from early-round investors, it should pay less than it would pay for late-series preferred stock with a high liquidation preference. In theory, if Uber is sold or liquidated for less than $68 billion, that late-series preferred will be paid off at a $68 billion valuation before the common stock gets anything. (In practice, the only viable exit for Uber is in an initial public offering that would collapse those share classes, so don't take that preference too literally.) Two academics recently did a study of unicorn valuation terms and argued that most unicorns' headline valuations were significantly higher than the actual implied value of their stocks; they calculated a "real" value for Uber of $60.6 billion, versus its $68 billion headline valuation.

The other thing that is going on here is that Uber has probably become less valuable since its last fundraising round at a $68 billion valuation, which occurred before a lot of scandals that undermined Uber's story and caused it to get rid of its founder-chief-executive-officer, Travis Kalanick. So Uber might "really" be worth only $50 billion (roughly SoftBank's target area for its tender offer valuation), and SoftBank might be overpaying for the shares from the company in order to let it save face. This is a little awkward: Presumably SoftBank has some blended valuation that it is willing to pay, and the more money it spends on overpaying Uber, the more it has to underpay the selling shareholders in order to get an acceptable blended price. Arguably the fairer approach would be to buy all the shares from everyone at a market-determined price, but I guess Uber is within its rights to demand more for itself and less for the shareholders who are, after all, cashing out early.

Blockchain blockchain blockchain.

I guess the classic advice is that if you are at a gold rush, sell shovels, and if you are at a cryptocurrency rush, sell ... I don't know, graphics cards and small scraps of paper on which people can write their private keys ("Cryptopaper: The most secure way to store your bitcoins!") and of course legal services, so many legal services. Here's a story about domain names: " is listed for sale on a marketplace called Uniregistry Market for about $10 million." That's like 30,000 ether, though I guess if you are a real cryptocurrency enthusiast why would you trade your ether for a domain name?

Novogratz said he expects major financial firms will soon start to offer bitcoin or similar products as an investment option, one that could be easily purchased over the phone.

A turning-point product from a big financial firm could arrive within six months, he said, though he declined to name a specific company.

“When it’s that easy, the price of bitcoin or ethereum is going to go much higher. And that is a lot closer than people think.”

Look, on the one hand: Sure, yes, when everyone can buy bitcoins over the phone from their banks, the price will go a lot higher. On the other hand, soon you'll be able to buy bitcoins, the disruptive alternative currency that was meant to disintermediate the banks and rebuild the financial system based on computer-science principles, by picking up the phone and calling your bank! Because that's what people want! It does seem simultaneously bullish for bitcoin -- more buyers! -- and yet somehow deeply bearish: People will be buying bitcoin while rejecting its premise.

People are worried that people aren't worried enough.

"David Swensen, Yale University’s longtime chief investment officer, said the lack of market volatility in the current geopolitical environment is a major concern and warned that another crash is possible." We talk sometimes about the argument that current low volatility is a sign of the market's increasing rationality and efficiency, and there is a soft spot in my heart for that argument, but on the other hand Swensen is not obviously wrong. Elsewhere: "President Donald Trump’s power to launch nuclear weapons is under scrutiny by both Republicans and Democrats in Congress concerned over his comments about striking North Korea." Even if markets are showing the correct amount of concern about the present value of future expected dividends, they still just seem somehow not worried enough.

People are worried about unicorns.

There is an unfair stereotype about the tech industry that it is a bunch of privileged young men solving the problems of privileged young men. Don't like doing laundry? A million Ubers for laundry bloom. But one problem for privileged men is that they seem to harass women a lot, and so one tech entrepreneur is solving that problem by teaching bro-culture classes at Duke, sure why not:

Caldbeck said he is setting out to educate young men about the dangers of “bro culture” in the workplace. He plans to release a website on the topic soon, and on Thursday, he delivered a 51-slide presentation to 50 students in a finance class at Duke University, his alma mater. He said he wasn’t paid and that he hopes to speak at other colleges. His goal, he said, is to “create positive change for women by educating young men about how to be better in the workforce.”

That's venture capitalist Justin Caldbeck, who "resigned in disgrace from his job running Binary Capital in June after six women accused him of unwanted sexual advances." In June! It is all so tech-y. When being an alleged sexual harasser turns out not to be a winning strategy, you quickly pivot to a new strategy of being a sexual-harassment-preventer, you work nonstop to roll out a minimum viable product (talking at Duke) as quickly as possible, and then you move rapidly to scale (the website). I assume he'll be raising a Series B by January.

Elsewhere: "What Hollywood Could Learn from Wall Street's Sexual-Harassment Reckoning."

Things happen.

White House Considers Former Pimco CEO [and my Bloomberg View colleague] Mohamed El-Erian for Fed Vice Chairman Post. HSBC agrees to pay €300m to settle probe into tax evasion. The Final Days of a Tax Haven. The Only Traders Seen Scoring Pay Bumps Run Electronic Platforms. At Last, Deutsche Bank’s John Cryan and Big Shareholder HNA Meet. AT&T Engages Its Washington Firepower With Megadeal on the Line. Muni-Bond Market Braces for Borrowing Rush Ahead of Tax Changes. SEC Rejects Exchanges’ Request to Delay Surveillance Database. High-Yield Canary Isn’t Singing About Markets Doom. Passive investing is a 'chaotic system' that could be dangerous, warns Robert Shiller. "Launch ten funds, with each one designed to go to cash or net short the stock market during a different year out into the future."  Why Mobile Payments Are Still So Hard for J.P. Morgan. Will Cash Disappear? Senate Plan Guts Obamacare, Sunsets Many Cuts. Gary Cohn's job sure sounds awkward. The 'Taxi Prince' Fighting to Make Sure Uber Won't Win in Japan. Cheese and Bourbon Face Risk of Backlash From U.S. Solar Tariff. Famous Liquor Store Says French Bandits Stole $800,000 Worth of Its Finest Whiskeys

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

For more columns from Bloomberg View, visit

©2017 Bloomberg L.P.