Selling Pens and Proving Algorithms

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.
Read More.
a | A

Market structure.

Here is a story about Aesthetic Integration Ltd., a startup that is pushing the idea of formal verification of trading systems:

Rather than only providing pages of text describing their trading systems, dark-pool operators should give regulators mathematical models that show precisely how they work, Aesthetic Integration Ltd. told the U.S. Securities and Exchange Commission in a letter last week. The code could then be examined with formal verification.

The idea is that a trading system -- an exchange or dark-pool matching engine, a proprietary trading firm's algorithms, etc. -- is a computer program, and you can write a sort of a meta-computer-program modeling how it operates. Then you can run that meta-program through Aesthetic Integration's testing software and make sure that the system does the things you want it to do (match orders by time and price priority, say), and doesn't do the things you don't want it to do (allow latency arbitrage and spoofing, say). Here is Aesthetic Integration's comment letter to the SEC.

I am fond of this idea, which I find pleasingly tidy-minded, but I worry that the things that people most want to test -- Is a trading system fair? Does it allow predatory high-frequency trading? Does it make flash crashes possible? -- are hard to specify precisely. (Also I suspect that they are emergent properties of multiple algorithms interacting, not just the result of a line of code in the matching engine.) Like, if the SEC wrote a mathematical specification for what an exchange needs to do to be fair, and the exchange passed it, plenty of people would complain that the SEC's specification didn't capture all of the ways in which the exchange can be unfair. (That happens now, just without the math.) Still the exercise seems appealing, if nothing else than as a way to force exchanges and dark pools and regulators and critics to think clearly about what precisely it is they want from their trading systems.

Elsewhere in market structure, I very much enjoyed this Bloomberg Markets story about how exchange-traded funds were invented to respond to the SEC's worries about portfolio insurance:

As far as the solution the SEC was envisioning, Most and Bloom seem to have checked that box as well. ETFs have taken pressure off individual stocks. On any given day, the vast majority of ETF trading happens in the secondary market and doesn’t involve creations and redemptions using the underlying securities, which definitely qualifies as a liquidity buffer.

And here is the story of BlackRock's iShares Gold Trust, which has temporarily stopped issuing new shares because it forgot to register with the SEC to issue more shares. Markets are only sort of efficient.

Caveat whatever.

I sometimes hear from readers who work in the financial services industry and who argue that it is held to a higher standard than a lot of other sales industries: Bond salespeople are apparently not allowed to exaggerate when they tell customers what a good deal they're getting, while used-car salespeople can just say whatever. I think there is some truth to that, but in fairness lots of consumer industries seem to be getting stricter about puffery. First there are list prices:

“If you’re selling $15 pens for $7.50, but just about everybody else is also selling the pens for $7.50, then saying the list price is $15 is a lie,” said David C. Vladeck, the former director of the F.T.C.’s Bureau of Consumer Protection. “And if you’re doing this frequently, it’s a serious problem.”

On the Internet, nobody sells things at the list price, so when you claim that you're selling a $260 skillet for $200, and everyone else is selling it for $200, aren't you really selling a $200 skillet for $200? Do you care? I guess the FTC does? The question, to put it in securities-fraud terms, is whether the list price is material to the buyer. The skillet is the skillet; you haven't deceived anyone about it. The buyer values the skillet at more than $200, you value it less, a deal is struck. But you have deceived the buyer about what someone else might have paid for it. Is that fact, irrelevant to the buyer's own situation but relevant to the broader market context, something that the buyer deserved to know? I don't know. We've talked about this before, with blouses, and the parallels to the Jesse Litvak case are clear. 

Then there are law schools. Here is the story of a woman suing Thomas Jefferson School of Law for letting her go there:

As her debt mounted and her job prospects faltered, she filed a lawsuit in 2011, arguing that she would not have enrolled at Thomas Jefferson if she had known the law school’s statistics were misleading.

Thomas Jefferson’s average student indebtedness, then about $137,000 — higher than that at Stanford Law School the same year — was among the highest in the nation. She also pointed to her school’s bar passage rate as consistently lower than 50 percent, which was below the average in California.

But the price and bar-passage stats were public, right? So she knew those things and went anyway? There's a general intuition that the party with more information about a transaction (typically a repeat seller) should have to disclose that information to the party with less information (typically a consumer). Fine. I feel like the next frontier, though, is to expect the seller to tell the consumer whether or not the transaction is a good idea for her. This is called the "fiduciary standard," and is harder.

Meanwhile in financial services proper, here is a pretty sick burn on Deutsche Bank:

While Deutsche Bank rose towards the top of investment banking it became renowned, in a tough industry, as being one of the most poisonous places at which to work, with a culture of back stabbing and blame, as well as that approach common across investment banking of doing what was legally defensible with counterparts, rather than what was always in the best long-term interests of clients.

Materiality.

We have talked a bit before about Valeant's perplexing investor-relations strategy: During times of crisis, it clams up publicly, but talks a lot with analysts (and Bill Ackman). It seems odd to me! But what do I know. It seems odd to other people too, though; here is an article titled "Valeant CEO's Calls With Analysts Raise Hackles Over Disclosure." But those hackles aren't necessarily at the Securities and Exchange Commission:

The mere fact that Valeant’s shares jumped after analysts began speaking about the Pearson chats isn’t proof that the company may have shared material information with them, according to Lynn Turner, a former chief accountant at the SEC who is now a managing director at a legal consulting firm. Nor, Turner said, are the events on their face enough to spur a formal investigation.

If I were rewriting the securities laws I'd probably start from the premise that, if the shares jump after X, then that means that X was material. I mean, maybe not always, but "the shares jump after it happens" seems like an obviously better definition of "material" than what the courts are working with now. Elsewhere: "A Troubling Timeline at Valeant."

White-collar enforcement around the world.

In the U.S., Reuters got a two-page government report about the fact that no individuals were charged in connection with Citigroup's mortgage settlement:

The report, by the Federal Housing Finance Agency's Office of Inspector General, one of the agencies in the Citigroup probe, said following the settlement, prosecutors reviewed the evidence to see if any individuals could be charged and determined "there was not enough compelling evidence."

The report "does not name any individuals that were investigated, nor does it elaborate on why individuals could not be successfully prosecuted." I suppose if there's no evidence that anyone committed a crime, it would be weird to produce a report naming all the people who didn't commit crimes, and all the evidence that doesn't exist against them. But if you really wanted a lot of bankers to go to prison for causing the financial crisis, this report will probably anger you.

In the U.K., on the other hand, there are "new rules to hold bosses responsible for wrongdoing," which might solve those banks' behavioral problems except that the rules are "deterring some bankers from taking on senior management roles and even prompting big-hitters to play down their own importance." The more-bankers-to-prison movement tends to focus on senior bankers, so the trick is to convince everyone that you're not really senior. And then keep doing bad stuff I guess? There are alternative approaches. "We have heard anecdotally that some banks now have a lawyer present at every meeting," says a guy, which is one way to go with it.

In Thailand, "the head of the stock market regulator has called for stronger action on market manipulation after CP All, operator of the 7-Eleven convenience store franchise in Thailand, allowed three board members who were fined for the offence to keep their jobs." Apparently "they did not mean to do anything wrong." I sympathize; no one really understands insider trading law anywhere.

And in Afghanistan, here is the story of Kabul Bank, which was once the largest bank in Afghanistan, but which was nationalized "after its owners were accused of embezzling $825 million using fake loans and spending it on, among other things, 11 villas in Dubai and an airline they used to smuggle cash there." Those owners "were sentenced to 15 years in prison, though they were later spotted at restaurants around Kabul." 

People are worried about unicorns.

How many sexual assault complaints does Uber get? "Five claims of rape and 'fewer than' 170 claims of sexual assault" in almost three years, says Uber, but BuzzFeed reports that "a search query for 'sexual assault' returns 6,160 Uber customer support tickets," and "a search for 'rape' returns 5,827 individual tickets." I do not know what to make of this, though Uber's explanation is weird:

Uber officials suggested that if an email address or rider/driver last name contains the word “rape” like “Jason Rape” or “Don Draper” it will be included when queried. The official noted that 68 riders and 12 drivers have the first or last name “Rape” and claimed that many of them generated multiple tickets during the time period in question. The source also suggested that misspellings of the word “rate” and expressions like “you raped my wallet” accounted for false positives in the search results seen in the obtained screenshots.

First of all, you never want your defense to be "no, see, our customers just use 'rape' as a metaphor for our pricing." But also it only explains one term: Fine, Jason Rape might be a bad Uber driver, but are there also drivers named, like, Richard Sexual Assault? 

In happier news: "Meet The 14 Unicorn Startups That Have Created 25 Billionaires." Uber leads the list with three billionaires, though Airbnb and SZ Dajiang Innovation Technology (a drone unicorn) have also created three billionaires each. And: "Wall Street rebound seen boosting IPO activity."

People are worried about stock buybacks.

Here is the story of a man whose vocal worries about stock buybacks cost him his job, sort of? It's a strange story. Jamie Welch was the chief financial officer of Energy Transfer Equity. He fought with his boss about "Energy Transfer’s decision to burn through more than $1 billion in cash last year by buying back the company’s stock when it traded closer to its peak of $35 a share." But what really burned him up was the cash-and-stock mix in Energy Transfer's deal to acquire the Williams Companies:

In recent months, Mr. Welch has called Williams shareholders, urging them to push the board to reconfigure the deal or vote against it, according to interviews with five shareholders who requested anonymity because they were not authorized to discuss the private conversations.

Mr. Welch argued that the deal’s terms, a mix of stock and a sizable $6 billion cash payout to Williams shareholders, would crush the combined new company under a mountain of debt. In one call, a shareholder said, Mr. Welch referred to the cash payout as “mutually assured destruction.”

A cash-and-stock merger is, of course, economically equivalent to an all-stock merger followed by a stock buyback. And if you were to go to a gas-pipeline company right now and pitch the idea of borrowing $6 billion to buy back stock, you would get a pretty short hearing. (Though Energy Transfer closed on Friday at $8.33, so at least the price is more attractive than it was last year.) So you can see why Welch found the deal a bit daunting. Still a deal is a deal, and Energy Transfer will have a hard time getting out of it, "because the ironclad agreement he helped structure made it nearly impossible to do so."

Strangely, Welch is not the only energy-and-power CFO to depart recently over disagreements about risk: "Electricité de France’s Chief Financial Officer Thomas Piquemal has resigned after a disagreement over the financial impact of the £18 billion ($25.6 billion) project to build new nuclear reactors in the U.K.," arguing that the project "could threaten the financial stability of the company." I like the idea of a league of CFOs who stand athwart financial risk, yelling stop, and who resign rather than compromise their beliefs.

People are worried about bond market liquidity.

The Bank for International Settlements released its quarterly review this weekend, and it would be absurd for a quarter to go by without some bond market liquidity worrying. (Previous quarters have featured worries about "recent dislocations in fixed income derivatives markets," "volatility and evaporating liquidity during the bund tantrum," and "market liquidity and market-making in fixed income instruments.") This quarter's is titled "Hanging up the phone - electronic trading in fixed income markets and its implications."

The increasing complexity of trading algorithms and their possible interactions represent a source of risk that can act as an amplifier in stress episodes. For one, large price movements or price gapping during stressed periods can prove difficult to incorporate in trading algorithms. Liquidity providers' risk monitoring thus often includes measures to interrupt quoting ("panic buttons"). Yet, while suspending liquidity provision may appear rational from an individual market participant's point of view, it raises the risks for the remaining liquidity providers.

Overall, these developments imply that electronic trading may have changed the dynamics - particularly the speed and visibility - of market responses to imbalances in demand and supply. It is, however, important to note that the basic underlying economic mechanism of how illiquidity risks unfold (Borio (2004), Shin (2010)) appears to have remained largely unchanged. Indeed, irrespective of the underlying market microstructure, market conditions remain susceptible to a sudden evaporation of liquidity (Box 2). These are situations in which both human traders as well as PTFs as the "new market-makers" (eg Menkveld (2013)) have always been reluctant to step in as shock absorbers (eg Adrian et al (2013)).

Algorithms are like people, only faster, and people are prone to panic too. Elsewhere:

Large bond funds, including products from BlackRock, Franklin Templeton and Invesco, are holding more than double the cash they held five years ago as fears about liquidity in the fixed income market weigh on asset managers.

But "improving secondary markets have tempted companies to sell more debt and bankers expect that if the markets remain stable then there will be a healthy pipeline of new bond sales." And the 10-year Treasury trades on special in the repo market; in fact, it is "the most special since mid-2008." 

Obviously the rest of the BIS Quarterly Review is of interest too; it includes features on how central banks have implemented negative interest rates (it's not a fan), "the resilience of banks' international operations,"  and the connection between wealth inequality and monetary policy

For most countries in our sample, the net distributional effect of monetary policy depends on its relative impact on the value of housing assets and equities. Changes in house prices and equity returns tend to have opposite effects on inequality if housing assets are concentrated in "poorer" households and equity holdings are concentrated in "richer" ones. In these cases, to the extent that monetary policy has boosted equity prices more than house prices, it has tended to increase wealth inequality.

For illustration, consider the following example. In the United States, the share of equities in the portfolios of q5 households is about 15 percentage points higher than the corresponding share in the portfolios of q2 households while the share of real estate is about 35 percentage points lower. Hence, if monetary policy were to lift equity prices by 10%, it would have to raise house prices by about 4¼% to be distributionally neutral.

Keep running.

Last week I made a little fun of a motivational e-mail sent out by a Goldman Sachs partner named Joe Mauro. "Keep running," it advised, between Bloomberg charts of Goldman's stock price. "What mile are you on?" This weekend Lucy Kellaway made a bit more fun of it, which spurred me to some Googling, which led me to this really quite incredible Financial News article from 2014 about a Goldman Sachs basketball-and-ping-pong (yes) tournament:

Joe Mauro, head of fixed income, currencies and commodities European hedge fund sales and co-head of European macro rates sales at the US bank, somehow pulled a quad muscle kneeling down for his team’s photo, much to the delight of some of the 42-year-old’s teammates.

Brian Levine, co-head of global equities trading and execution services at the bank and a teammate of Mauro’s, said: “That’s the reason we need a strong bench. Guys get injured all the time, including Joe kneeling down for the photo shoot before the first game – he didn’t even get a minute of playing time.”

I hope they yelled "what mile are you on?" at him the whole time.

Me Friday.

I wrote about unicorns.

Things happen.

Investors Fret as ECB Looks Poised to Get More Negative. World's Biggest Bond Traders Give Themselves No Margin of Safety. Credit Suisse Wealth-Management Head Has a Tough Task. "The sale of Barclays’ African unit says more about the bank than the continent." Iron Ore Jumps Most on Record as Market Goes 'Berserk.' Hedge funds’ bearish mining bets turn sour. Fannie and Freddie Shareholders Suffer Stinging Legal Setbacks. The Mystery Madoff Victims Who Left $2.5 Billion on the Table. Puerto Rico: The Multiple Issuer Problem. Europe’s Banks Find a Dumping Ground for Their Losses. GAO to Probe ‘Regulatory Capture’ at New York Fed. Why FinTech is the WORST. Donald Trump "'would last one tournament and then be removed from the team,' said Eric Di Michele, coach of the speech and debate team at Regis High School in New York." Harvard Looks to Yale on Social Life. Bridge cheating. Crossword plagiarism. Hockey hair. Sex spa

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 mlevine51@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net