Compliance Therapy and Software Rules

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


One simple plausible story about the U.S. banking industry is that in 2006 it was fun and aggressive and wild and dangerous, and in 2026 it will be boring and sleepy and safe and utility-like, but in 2016 it's a roiling sea of psychosexual conflict as it transitions from one state to the other. What would Freud make of this amazing Wall Street Journal article about bankers, regulators and compliance? Here's the kicker, an anecdote about a training program for bank compliance officers:

Instructors made light of the tension between banks and regulators, performing a skit parodying the Wizard of Oz during one class. A male instructor dressed as Dorothy wondered why one of her regulators hadn’t told her something.

“Oh you gullible, innocent girl!” replied an instructor dressed as the Wizard. “Don’t you know there’s no such thing as a good regulator?”

The punch line here is not that "there's no such thing as a good regulator." The punch line is that the banks' coping strategies for dealing with their new regulatory status include role-playing and cross-dressing. This is how the banks are processing Dodd-Frank: By dressing up like Dorothy and complaining. These bankers are not swaggering risk-takers, but neither are they quiet boring utility operators. They have been stripped of their identity and are struggling, through deep emotional pain, to find a new one.

So they draw pictures of how regulation makes them feel:

At a Barclays PLC town hall after Dodd-Frank rules began to go in place, bank compliance executives shared images of how each group thinks of the other, said someone familiar with the meeting. To represent bankers, compliance executives showed an image of the Wild West: cowboys on horses with guns.

On the other side, to show how bankers view compliance officials, the executives revealed a picture of nuns carrying guns, an indicator that the group was seen as ultraconservative but still dangerous.

Again I don't want to focus on the substance of the pictures, or their oddly gendered nature. The point is that bankers -- well, compliance officers, anyway -- are spending their time on these transparently therapeutic activities. No one in "Liar's Poker" went around sharing images of how mortgage bonds made them feel. The new world of banking is in touch with its emotions, but those emotions are raw and confused.

(Freud would probably have something to say about the repression of excretory functions too: "At one midsize bank in the Western U.S., a compliance officer was reprimanded after complaining about an individual regulator to a colleague in the bathroom, only to find out the regulator had been in a stall.")


After nearly two decades of breakneck expansion into ever more countries and ever more businesses, global banks are in retreat. For most of them, it is no longer a viable strategy to try to be all things to all customers around the world.

That article talks in blandly businessy terms about shrinking footprints (the top 10 global banks were in an average of 65 countries in 2008, versus 55 in 2015) and revenue diversification, but to truly understand the impact of the retreat from universal banking you might have to give the bankers a set of crayons and ask them to draw what they're feeling.

Market structure.

Here's a new working paper from the Bank for International Settlements called "Who supplies liquidity, how and when?"

Who provides liquidity in modern, electronic limit order book, markets? While agency trading can be constrained by conflicts of interest and information asymmetry between customers and traders, prop traders are likely to be less constrained and thus better positioned to carry inventory risk. Moreover, while slow traders' limit orders may be exposed to severe adverse selection, fast trading technology can improve traders' ability to monitor the market and avoid being picked off. To shed light on these points, we rely on unique data from Euronext and the AMF, the French financial markets regulator, enabling us to observe the connectivity of traders to the market, and whether they are proprietary traders. We find that proprietary traders, be they fast or slow, provide liquidity with contrarian marketable orders, thus helping the market absorb shocks, even during a crisis, and they earn profits while doing so. Moreover, fast traders provide liquidity by leaving limit orders in the book. Yet, only prop traders can do so without making losses.

I think there are two broad ways to think about high-frequency trading. There's the market-microstructure view, in which high-frequency trading is the electronic successor to classic market-making, and in which most of the actions of high-frequency traders -- frequent order cancellations, payment for order flow -- can be explained by standard models of how a market-maker should set its prices and deal with adverse selection. And there's the conspiracy-theory view, in which high-frequency trading is a complicated theft that no one can quite explain because, you know, it is so fast. I suppose it is not surprising that academic and official economists tend to come down on the market-microstructure side. Anyway the BIS did: Fast proprietary traders -- that is, high-frequency traders -- provide liquidity by using "contrarian strategies, buying against downward price pressure, and selling against upward price pressure," just like market makers are supposed to.

Elsewhere in BIS working papers, here is Andrew W. Lo on "Moore's Law vs. Murphy's Law in the financial system: who's winning?" It gets a little weird:

It is fruitful to think of the law as the software of the American operating system – yet if a team of software engineers were to analyse the corpus of federal law, they would see thousands of pages of poorly documented code, with a multitude of complex, spaghetti-like dependencies between individual modules.

When you think about the regulation of, say, electronic trading, some of it is made up of laws and regulations about fraud and position limits and spoofing and so forth. And some of it is made up of literal computer code with which electronic traders' systems interact: The New York Stock Exchange is basically a computer program that sets the rules by which electronic traders can trade stocks. You could imagine that the near future of financial regulation might involve putting relatively more of the regulation -- the "operating system" for financial activities -- into actual computer code instead of background law. (You can see an example of this in the U.S. equity markets' move away from breaking "clearly erroneous" trades after the fact, and toward preventing them in advance through "limit up/limit down" rules that are coded directly into the software.) That would probably be a good move: The software industry is relatively young, compared to the legislation-and-regulation industry, but it seems to have a much clearer idea of what constitutes good design than the legislators/regulators do.


Speaking of putting rules into software, here is a funny story from Nathaniel Popper about how "a group of computer scientists released a paper on Friday describing a number of security vulnerabilities in" the DAO, which is sort of a distributed venture capital partnership for blockchain and smart-contract enthusiasts:

The authors of the paper argue that the money that has been put into the project, known as the Decentralized Autonomous Organization, could be frozen or stolen by attackers as a result of flaws in the way that the venture, known as the D.A.O., was set up. The money is all in a digital currency called Ether, which is a newer alternative to Bitcoin and exists entirely online.

When we last talked about the DAO, I contrasted its corporate-governance innovations with other Ether projects whose innovation was building more efficient pyramid and Ponzi schemes. But we are still in the early days of blockchain adoption, and it is not completely impossible that all blockchain innovations -- or, at least, all innovations in truly distributed blockchains, as opposed to the proprietary blockchains being developed by groups of banks -- will just turn out to be new ways for techno-libertarians to steal from each other. In any case the DAO has raised over $100 million and I am rooting, not necessarily for it to be stolen by hackers, but at least for them to do something funny with it. It would be a bit sad if they funded, like, an on-demand laundry startup

Elsewhere: "Chinese investors are pumping up bitcoin again, sending prices up nearly 16% in the past four days." While enthusiasm about bitcoin as the future of finance has cooled, and has moved on to second-generation blockchain projects like Ether and the DAO and the bank-run blockchains, the currency is still hanging in there. "On Monday, prices moved up as high as $525.49 per bitcoin."

Merger feuding.

It's funny to think about the takeover fight between Energy Transfer Equity LP and Williams Cos. as a problem of smart contracting. In September, they signed a merger agreement in which Energy Transfer would buy Williams for about $33 billion in cash and (mostly) stock. And then energy prices crashed and the deal stopped looking like such a great idea for Energy Transfer. But it had a signed deal and no obvious way out. So things sort of ... descended into murk? "Behind the scenes, executives at Energy Transfer are suffering from a giant case of buyer’s remorse and frantically searching for a way to pull out of the deal," wrote the New York Times in February. Energy Transfer found a funny little hook in April, claiming that it couldn't get a standard tax opinion from its lawyers. And last week it actually had the chutzpah to sue Williams -- which had already sued Energy Transfer -- claiming that "Williams has breached the merger agreement entered into with ETE on September 28, 2015, by, among other things," "refusing to cooperate with ETE's efforts to finance the merger," "failing to use reasonable best efforts to complete the merger," and -- and this is my favorite -- "consistently making public statements implying that the Williams Board supports enforcing the merger agreement as opposed to completing the merger." It is odd to think that saying you support enforcing the merger agreement might be a breach of the merger agreement? And it does seem like Williams wants to enforce the merger agreement by, you know, completing the merger? Like ... that is what enforcing the merger agreement means? I don't know. I'm sure some lawyer somewhere thought the difference was important.

Anyway Williams's lawsuit is seeking a ruling that it can terminate the merger agreement and get a $1.5 billion breakup fee, which, again, chutzpah.

I suppose that in the distant future of finance, merger agreements might be smart contracts. The idea of smart contracts is that they are self-executing code: If some external objective conditions are satisfied, then the money is wired automatically, and the Williams shares poof automatically into Energy Transfer shares. A deal is a deal, and a benefit of putting it into code is that it makes it harder for one party's cold feet to throw the whole thing into murk. But that's a benefit of putting it into a written contract, too! There are some vague terms in the merger agreement -- "reasonable best efforts," for instance, which lets Energy Transfer maintain with a straight face that Williams is the one preventing the deal from closing -- but for the most part it's intended to set out clearly when Energy Transfer can walk away and when it has to close. But it didn't quite work, not because of the vague terms, but because of what I'm sure seemed to everyone at the time like an objective external condition: Energy Transfer doesn't have to close the deal if it can't get a tax opinion from its lawyers. And now, it says, it can't. 

If you were coding up that contract, would you code it up as just "if (no TaxOpinion) then (no Closing)"? Would you write a TaxOpinion function that takes all the underlying objective facts that Energy Transfer's lawyers would consider and spits out a tax decision, or would you just take whatever Energy Transfer's lawyers give you as itself an objective data point in the world? Would you defer to a distributed blockchain court for a ruling on whether the tax opinion is being withheld in good faith? The weird hard murky questions wouldn't go away: They'd just be pushed up, and you'd have to decide them in advance, without knowing why the parties might end up with cold feet, before signing the deal. If you had to hash this stuff out before signing the deal, though, I'm not sure the deal would ever get signed.

Ukraine fighting.

It is probably less funny to think about Ukraine's fight with Russia in terms of smart contracting, but they do have a contract -- a loan agreement under which Ukraine owes Russia $3 billion -- and they are now in arbitration in a London court. Russia thinks Ukraine should pay because the contract says it should pay. Ukraine thinks it shouldn't have to pay because Russia invaded Ukraine, took the Crimean Peninsula, imposed economic sanctions and generally has not behaved the way you'd hope for from a good-faith contract counterparty. Here is Mark Weidemaier, who thinks that "none of these arguments are slam dunks," but that Ukraine has a chance, and that some of its arguments are "clearly meritorious if -- and it's a big if -- the court makes the appropriate findings of fact." But those findings of fact are about war, not contract law, which makes them awkward for a contract dispute:

The argument under prevention doctrine is straightforward: If one party to a contract makes it difficult or impossible for the other to perform its duties, the latter's failure to perform is excused. Russia's annexation of Crimea, support for armed rebels in the east, and general policy of destabilizing Ukraine's economy has made it impossible for Ukraine to repay, at least while complying with its IMF support program. If one accepts that version of events, the argument is a slam dunk. The problem is that the court will have to weigh in on precisely the kinds of politically-charged questions that (in my view) judges would prefer to avoid. 

People are worried about unicorns.

After the revelations about Peter Thiel bankrolling a campaign to shut down Gawker, the media's worries about unicorns seem to have been replaced with meta-worries about reporting on unicorns. Here's Nellie Bowles:

After six years as a reporter in Silicon Valley, I’ve found that a tech mogul will generally call anything unflattering I write “clickbait” and anything flattering “finally some real journalism”.

A macabre parlor game among reporters here now is to guess which billionaire will, as Thiel did, wait 10 years with a grudge before seizing an opportunity to bankrupt you and exact maximum revenge. It’s a paranoia that seems more fitting for reporters covering characters like Vladimir Putin than the latest startup.

Everything's great in the Enchanted Forest, where all the unicorns are kind, all the unicorn herders are geniuses, and nothing ever goes wrong.

People are worried about bond market liquidity.

Nah, bond market liquidity worrying takes some time off for Memorial Day weekend. So does the bond market: Friday had the lowest investment-grade bond volumes of the year so far.

Things happen.

The Untold Story Behind Saudi Arabia’s 41-Year U.S. Debt Secret. Gunvor boss used $1bn payout to sever ties with Russian oligarch. Greece and Creditors Spar Over Legislation Changes. Top C.E.O. Pay Fell — Yes, Fell — in 2015. Where More Women Are on Boards, Executive Pay Is Higher. MetLife Suit Raises Questions of Extent of Corporate Liability. Miner Ends Quest for Gold to Unearth Strongest Material in World. Bridgewater secures $22m in state aid. LendingClub Founder Turned to Mack for Emergency Loan Help. "We’ve never had a president who not only thinks the government will be a toy for him to play with and push people around — wow!! how wild is that!! — but who tells us, in advance, over and over again, that that is his game." Man deletes Twitter account. Lanzhou New Area has a life-size Parthenon and Sphinx but no people. ‘We’re struggling to get by on £200,000 a year.’ The NSA guide to the internet. Is Everything Wrestling? Siberian surfing. Ice cream wars.

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