Money Stuff

Order Types and Ineligible Heists

Also satellites, social media, universal life, bad unicorns, buyback math, etc.

Market structure.

It is helpful to think about U.S. stock market structure by imagining that it makes a little more sense than it actually does. So the way markets kind of work is that there are customers, who want to buy stocks because they believe in their fundamental value and want to own them for a while, and dealers, who make markets in stock for their own account and hope to earn a spread by providing liquidity to the customers. The dealers are market professionals in ways that the customers aren't, and so you should expect them to have some advantages over the customers. You should expect them to have better information, and to invest more in trading technology, and to react faster to news, and to adjust their quotes constantly to reflect new information, and just generally to be better at trading than the customers, because trading is their job in a way that it isn't the customers' job.

On the other hand, the customers should get good service. Ultimately they are the point of the market: If the dealers are just sitting around trading with each other, with the customers as an afterthought, then something has gone wrong. In some loose approximate sense, if a customer shows up and wants to buy stock, then she should get priority over a dealer who also wants to buy that stock, because, you know, she actually wants to buy the stock. The dealer is just churning inventory for a few milliseconds. 

None of this is exactly right, and nothing quite works this way, and there is no clean obvious way to separate dealers from customers. Still, it can be a helpful framework. The big innovation of IEX, the "Flash Boys" exchange, was to provide some order types that were intended to give customer-type traders an advantage over fast dealer-type traders. That advantage can seem unfair to the high-frequency traders, if you think that everyone should be treated equally, but it makes more sense if you think that the point of a stock market is to serve end users.

People obviously liked what IEX was selling, and so IEX's competitors are now rushing to sell it too. Now there is this:

Nasdaq plans to seek SEC approval for a new function that may appeal to institutional investors such as mutual funds. Investors who use the function, known as an “extended life” order type, would move ahead of other similarly priced orders if they agree not to cancel their orders for approximately one second.

If you are a dealer in stocks, in this electronic age, you like to preserve the flexibility to update your orders constantly, because your advantage is in responding rapidly to information. But if you are just a big institution buying stocks, that flexibility is not so important: You just want to buy the stocks. So you would be happy to give up some flexibility -- to mark yourself as a customer rather than a dealer -- in exchange for having a better chance to actually buy the stock you want. Makes sense! And it's pleasing to see Nasdaq come up with a different idea to help institutional investors, instead of just blindly copying IEX's "speed bump" approach. (The Toronto Stock Exchange introduced an order type like this last year.)

I have nothing but nice things to say about this, really. (So does IEX spokesman Gerald Lam: "We’re encouraged to see that IEX has finally inspired other exchanges to better serve investors and we look forward to seeing if any new products can neutralize the advantages they currently sell.") But of course there is a second level to any of these issues, which is: This new order type, like IEX's speed bump, adds a new kind of complexity to the market, and dealers (that is, high-frequency traders) are professionals in understanding and exploiting exchange complexity, in a way that the customers aren't. Adding complexity increases the returns to sophistication; if you think that high-frequency traders are evil, then every new order type is a new opportunity for them to do evil stuff. In theory. So you tell me: If you were a high-frequency trader, how would you exploit Nasdaq's new order type to take advantage of less sophisticated traders?

The Bangladesh heist.

I have been on the internet for too long, and back in the olden days of the internet, there was a famous post called the "Credit Card Prank." It now seems to have disappeared, but it was about a guy who would sign credit card slips with weird drawings or, in one case, by writing "I STOLE THIS CARD," just to see if anyone was paying attention. Mostly, they weren't. I thought of it again when reading about the hackers who "brazenly stole $81 million from Bangladesh’s account at the New York Fed by sending dozens of orders filled with errors of formatting, spelling and sense":

One order seeking money from Bangladesh’s account at the Federal Reserve Bank of New York wanted $20 million for vaguely defined consulting fees. A second sought $30 million for “ineligible expenses.”

No see "ineligible" means ... never mind. I don't think any of the orders were marked "for supervillainy ha ha ha we are stealing this money no really," but "ineligible expenses" does get pretty close. 

It all looks kind of dumb but, you know, the New York Fed is not your mom. You can't really have a functioning international payment system in which someone in New York reviews every transaction for infelicities of language. In fact, the New York Fed "has since tightened its handling of Bangladesh’s account and is reviewing payment orders more closely," and it's a mess:

The scrutiny has delayed some legitimate transfers. On April 11, the Fed canceled an $8.7 million payment from Bangladesh to China Geo-Engineering Corp. for a water supply project, saying it wanted more information about the project.

The checks appear to be creating some friction. In a June 9 letter reviewed by the Journal, the New York Fed complained to two officials in Bangladesh that the new security arrangement “is not sustainable.”

One thing that is satisfying about bitcoin/blockchain as a payment mechanism is that it properly sets expectations. No one expects there to be a person in a room in New York making sure that all the bitcoin trades are kosher. Everyone knows that many of them aren't. There might as well be a check box on every bitcoin transaction to indicate "for supervillainy ha ha ha we are stealing this money no really," but if you check the box the transaction still goes through. The world of international money transfers is necessarily cold and cruel; at least bitcoin wears its cruelty on its sleeve.

Elsewhere: "Swiss probe ex-Abu Dhabi official over alleged Malaysia fund scam." 

Tiny satellites.

One thing that I talk about a lot is the fact that financial markets can never be a level playing field, and that worrying that insider trading is "cheating" or an "unfair advantage" is sort of absurd. So I love articles about how hedge funds are paying to launch satellites to spy on parking lots:

A company called Planet Labs Inc. has launched a small constellation of what it calls “cubesats” that can deliver much more frequent imagery of economically sensitive spots than traditional satellites. Those spots include retailers’ parking lots, oil-storage tanks or farmland.

Sure, you can go to a Wal-Mart and count cars in the parking lot and decide if business is good. But those hedge fund guys: They can count all the cars in all the parking lots all at once. They don't even have to do the counting: A company called Orbital Insight Inc. buys the data from Planet Labs and sells signals to funds "based on its automated analysis of satellite imagery." You'll never have all the advantages they have. Though on the other hand those advantages may not be worth much:

However, such data can be difficult to use without an in-house data scientist and it can sometimes be unreliable. Data like satellite-image analysis has only recently become available to hedge funds, so there isn’t much historical data to use to check its value. Some critics think the benefits of such complicated efforts to analyze companies and the economies are overblown.

There is an arms-race effect: If other hedge funds have satellite data, you need to have satellite data, or your analysis will be just a little bit less informed than theirs is. If you are doing the same sort of analysis. Most of my money is in index funds, so in a sense I get to free-ride off the satellites.


I mean, this sounds silly, but not wrong:

A leading online lender in China has revealed the unorthodox extent to which the country’s booming fintech sector uses personal information to assess credit risk — including examining social media contact lists.

“If we find you are [social media] friends with celebrities from the entertainment or finance industries, we think you must be to some extent trustworthy … since otherwise you wouldn’t have such friends,” said Jesse Chen, co-founder of Jubao Internet Technology, one of China’s top three peer-to-peer lending firms.

You just go where the correlations take you, you know? It seems inevitable that eventually algorithms will make all of our credit decisions. Lots of people worry that the algorithms will be unfair, biased, difficult to correct. But there is also a sort of aesthetico-philosophical worry, about the absurdity of a life where robots decide if you're a good credit risk by scouring your Facebook pictures for signals meaningful only to robots. There's nothing you can do about it: While today's simple credit-scoring algorithms reward you for comprehensible things like paying your bills on time, future algorithms will be built by machine learning and be completely opaque to humans. You'll add a new Facebook friend because he is responsible and always pays his bills on time, and the computer will decide that anyone with such boring friends will never get anywhere in life and shouldn't be trusted with a mortgage. "Lol nothing matters," you'll Snapchat, as your mortgage is denied.

Elsewhere, here are some reviews of "No Man's Sky": "What does it mean to be alive in a world where everything is driven by algorithms, and your existence is solitary?" 

Rates are low.

Low rates are bad for holders of universal life insurance policies, who are seeing their premiums go up, and who are getting bad financial advice like this:

When the Cooks’ daughter, Jo Ann Sparks, asked an expert to explain her options, she recalls: “He said to me, ‘Please don’t take this the wrong way and, not to be morbid, but your mother needs to die.’” 

What is the right way to take that? What sort of an expert was he, anyway? I try not to give investing advice around here, so I will keep my thoughts about universal life insurance to myself, but one piece of financial advice that I am comfortable repeating is that it is better to die later than sooner.

Low rates are also bad for international bond funds, which have been buying more emerging-market bonds as "Government bonds with negative yields have grown to $11.4 trillion, mostly in developed markets." And here is a fun roundup of bonds that are trading at more than 200 cents on the dollar. Robin Wigglesworth says: "Essentially, people are buying bonds hoping for capital appreciation, and stocks for yield."

People are worried about unicorns.

Man, there are a lot of worrying unicorns, or quasi-unicorns. Here's the story of Hyperloop One, which "is trying to make a pod levitate in a tunnel," but which has gotten distracted by a lawsuit between its founders over some governance stuff and also a noose, or quasi-noose:

BamBrogan entered, wheeling his desk chair. On it rested a rope, looped at the end. To BamBrogan, it was a noose and a threat. The company insists it was a lasso, saying in its countersuit that Afshin Pishevar intended it “for someone acting like a cowboy.” Through Willingham, Afshin Pishevar said designating the rope a noose amounts to “an ill-fated attempt to bolster a meritless lawsuit.”

Don't ... loop a rope ... and leave it ... on someone's chair. You don't get to decide what sort of symbolism they put on that. Come on people.

And then there is Gravity4 Inc., Gurbaksh Chahal's self-proclaimed unicorn ("Chahal recruited employees by telling them—and his Twitter followers—that the company was worth $1 billion"), which he will be leaving for a while because he was just sentenced to a year in jail (pending appeal) for kicking his girlfriend while he was on probation for kicking another girlfriend. "Silicon Valley’s Most Self-Destructive Founder" is the headline on that one, and I feel like kicking multiple human beings is perhaps not strictly self-destructive, but there is something very Silicon Valley about treating repeated violent crime as essentially a problem of managerial philosophy:

Chahal’s earlier achievements enabled him to run his businesses unchecked and use vague promises of startup riches to recruit talent. “He has a brilliant mind and a very flawed personality,” said Sam Singer, a crisis communications consultant who worked for Chahal in 2014. “He has become a poster child for everything the public thinks is wrong with Silicon Valley: wealth that comes too fast and too easily, arrogant behavior, the belief that the rules don’t apply to them and they are somehow above the law.”

In nicer news: "How This Woman Went From Homelessness to Running a Multimillion-Dollar Venture Fund." And in just regular unicorn news, Airbnb is not happy about brokers who offer unauthorized secondary-market transactions in its shares, or in forward contracts referencing its shares:

Laurence Tosi, Airbnb chief financial officer, describes them as “bottom feeders trying to steal claims and pass them off as legitimate claims in a gold rush”.

The dream of the unicorn is to run a massive company with easy access to funding but with total control over the trading in your stock. It kind of works. But you have to expect some leakage.

People are worried about stock buybacks.

There is a style of argumentation that consists of doing some algebraic manipulations and then asserting an unrelated claim to your now-mystified audience. The method was probably invented by Pythagoras, who used the thing about the hypotenuse to convince people not to eat beans, but the most famous example comes from Euler, who supposedly won a theological debate with Diderot by saying: "Monsieur, (a + b^n)/n = x, donc Dieu existe, répondez!"

Anyway here's an article about buybacks:

Mr. Colby has developed an illuminating analysis that identifies a crucial difference between many truly successful companies and their underperforming counterparts. The exercise highlights the growth mirage that buybacks have on earnings-per-share measures. In addition, it shows that returns on investment need not be that large for a company to generate growth rates exceeding the evanescent earnings-per-share gains associated with buybacks.

In his test, Mr. Colby compared net profit growth and earnings-per-share gains at pairs of companies in the same industries from 2008 through 2015. In each case, he contrasted a company that bought back loads of shares during the period with another that did not.

It is tempting rewrite the analysis as an algebraic proof:

  1. (y'/x' - y/x)/(y/x) > (y' - y)/y
  2. Therefore buybacks are bad.

Here's an example:

Jack in the Box made quite a contrast. Its annual earnings per share rose by 6 percent over the period, but its net profit declined by 0.5 percent a year.

To bring its net profit to the level of growth it showed in per-share earnings, Mr. Colby said, Jack in the Box would have had to generate after-tax returns of only 4.8 percent on the $1.2 billion it spent buying back shares. That doesn’t seem insurmountable.

Another way to say that is:

  1. Jack in the Box had a business whose returns declined at a rate of 0.5 percent a year from 2008 to 2015. (That's not quite right -- they plummeted, and then recovered smartly -- but I guess it's right on average.)
  2. Instead throwing more money into that declining business, the company gave the money back to shareholders to invest in something better.
  3. And so, despite its declining business, Jack in the Box gave shareholders a return on their investment that grew 6 percent per year.
  4. It could have done even better had it invested its money in a robust growing business.

But what would that be?

People are worried about bond market liquidity.

If I am worried about bond market liquidity when I am 80 years old, then ... well honestly I guess I won't be that surprised? Anyway the Securities and Exchange Commission asked for comments on a new rule about same-day reporting of Treasury trades to Trace, and it got this:

Jane Carson, who identified herself as an “80 year old individual investor,” was the first to weigh in, in a letter received Aug. 5 that expresses exasperation with the opacity of the world’s biggest debt market.

Here's her letter, which is one page long and shows deep familiarity with Treasury market structure, the history of Trace and the relevant academic literature. It's a refreshing change from the individual-investor comment letters that the SEC gets about equity market structure.

Elsewhere in Treasury liquidity, here's Treasury's blog on "Examining Changes in the Treasury Repo Market after the Financial Crisis." Based on previous Fed and Treasury blog posts on bond market liquidity, you can probably guess the conclusion. ("The fact that repo volumes and pricing have remained within a relatively stable range in recent years demonstrates that," like the Treasury market, the repo market "continues to function well.")

Things happen.

Visium Trader-Turned Whistle-Blower Tied to Hedge Fund’s Demise. Banks’ CoCo Bond Payouts Gain More Protection in EU Proposal. PwC sued for $5.5bn over mortgage underwriter TBW’s collapse. Robert Hockett and Saule Omarova: "On this view, the sovereign public acts effectively as franchisor, licensing private financial institutions to earn rents as franchisees in dispensing a vital public resource: the public’s monetized full faith and credit." Izabella Kaminska on Libor. Resolution on CCP Resolution? Tronc tronc tronc tronc. Trump and Bayrock. AT&T beats Citi challenge over saying ‘thanks’ to customers. America’s Golf Courses Are Burning. Eric Clapton broke an Icelandic fishing record. F-22 Raptor Grounded by Swarm of Almost 20,000 Bees.

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:
    Brooke Sample at

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