Robots, Rogues and Regulation

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

Robots vs. humans.

We've talked before about the branding potential of the term "index fund." These days, index funds are viewed as good, sensible, efficient, low-cost investment products, in contrast with actively managed funds, which are viewed as expensive and dumb. If you run an actively managed fund, one option is to push back on that perception by touting the benefits of active management. But there is another option. You can just call your actively managed fund an "index fund." It takes some work -- you have to have some process to turn "a list of stocks that I picked" into "an index" -- but it is surprisingly achievable. I mean, who's to say that one list of stocks is an index and another list of stocks isn't? Standard & Poor's? Who put them in charge of indexes? Really any list of stocks, viewed in the right light, is an index, so any fund can be an index fund. You don't even necessarily have to lower your fees. 

What index funds are to the fund industry, robo-advisers are to the financial advising industry, so it is not surprising -- thought it still made me giggle -- to read that "ETrade Financial Corp. is launching a robo-adviser, with a twist: human beings." 

Previous robos have embraced passive investing, using index funds that include exchange-traded funds (ETFs) rather than pricier actively managed funds. They use a consistent set of formulas to determine how to spread money across different asset classes. At ETrade, a nine-member investment team will tweak portfolios in response to market conditions.

Sure, whatever. "Some of it is your gut," explains ETrade's senior vice president of investment product management. And a pure robot doesn't give you much in the way of a gut. For gut you need people. Anyway the fees seem to be competitive with real robo-advisers.

It's an odd financial moment we live in. It seems obvious that the future of a lot of commoditized investing products will be algorithmic, whether the algorithm is as simple as an S&P 500 index or as (slightly more) complicated as an automatic-rebalancing robo-adviser. A lot of humans provide services that are competitive with those algorithms -- active fund management, portfolio-rebalancing financial advice -- at higher prices and with less reliability. Those humans see the robots coming for them. Some of them are putting on robot costumes and hoping to hide out. I suppose it could work, for a while. But it's hard to see it as a long-term solution.

Meanwhile here is Hugh Son in Bloomberg Markets on the algorithms coming for your job:

What all of this means is that the number of front-office trading and dealmaking jobs has been in decline since 2007, just before the financial crisis, when it peaked at 64,521. It was 14 percent lower last year at the world’s largest investment banks, according to Coalition. Even if revenues recover because of higher interest rates, improving economies, and a rebound in debt trading, new platforms will simply scale up to the higher volumes without needing many more flesh-and-blood operators. Wall Street has reached peak human.

The advice is basically, like, learn to code.

Rogues.

One of the biggest cultural differences between the U.S. and Europe is that, in Europe, if you are fired from a bank for doing grossly illegal things, it is considered cool to sue for back pay. In the U.S., committing massive financial crimes is considered a good reason to lose your banking job. In France ... meh, it is more complicated. Jérôme Kerviel, who rogue-traded his way to a 4.9 billion euro loss at Société Générale and a three-year prison sentence (of which he served a few months after walking from Rome to France for some reason), naturally thought to sue Soc Gen, and naturally won:

On Tuesday, the tribunal ruled that his illegal actions presented “no real and serious cause” for his dismissal, his lawyers said. The panel ordered Société Générale to pay Mr. Kerviel roughly €450,000.

Hmm right "no real and serious cause." Soc Gen said it would appeal and called the decision "incomprehensible," which is not entirely wrong, but let's try to comprehend it anyway. This seems relevant:

The majority of Mr. Kerviel’s award — €300,000 — represents a performance bonus for 2007. The sum was linked to a €1.4 billion profit that the bank booked in the fourth quarter of that year from his rogue dealings.

The theory seems to be that Soc Gen knew what Kerviel was up to, booked profits from his rogue trading, and therefore should have to share those profits with him in the form of a bonus. (And, by the way, if you can get a trader who can make 1.4 billion euros of profits and demand only a 300,000 euro bonus, that is a great deal, though I suppose also a red flag.) This has been "Mr. Kerviel’s argument from the start: namely, that his managers — many of whom quietly left the bank after the scandal — turned a blind eye to his activities and even tacitly encouraged them, as long as his deals were profitable." There is a certain rough justice to it. If the bank benefited from his rogue trading, he should share in the benefits. And if the bank later lost all those benefits and then some, well, that was no longer his problem, he was gone. 

But that is ... not how we do financial regulation these days? Even in Europe? Regulators are now focused on bonus deferrals, clawbacks, "malus" -- rules that are aimed at precisely this problem. If you do a big trade and make a lot of money and get a big bonus, and then it turns out the big trade was actually very bad and loses several times as much money as it made, they're supposed to take your bonus back. The goal is to discourage people from doing big risky roguish trades to get a big bonus, because they know the bonus will be at risk as long as the trades are. Even if the bank knew about them! Kerviel has become "a sort of folk hero in France," and is viewed as a scapegoat for broader systemic problems. That might be true. But the general rule here -- that traders should get to keep the profits of their trades and not be penalized for their losses -- is not a great one.

Elsewhere in scapegoats, here's a profile of Rebecca Mairone (now Rebecca Steele), the former Countrywide executive who was sued by the Justice Department for running the "Hustle" mortgage program that sort of looked like fraud but turned out not to be fraud

Mergers and acquisitions.

One of the main battlegrounds in modern finance is over when and how you're allowed to deceive your counterparties. The key question in the Hustle case was, if a bank promises to deliver good mortgages and then ends up delivering bad ones, is that fraud? The key question in the Jesse Litvak case is, if a bond trader lies to customers about the price he paid for bonds, is that fraud? These questions are harder than they sound because trading financial products has some analogies to poker, and bluffing and deception are accepted practices in many areas. "It is no answer to the question to say this was regarded as honest by other bankers," said the prosecutor in a U.K. Libor trial yesterday, but I think a lot of bankers and traders would disagree with him.

The same issues come up in mergers-and-acquisitions banking. The job of a sell-side M&A investment banker is to try to get the highest possible price for his client. One way to do that is to create a sense of competitive pressure, to get potential buyers to raise their bids. If there is only one bidder, well, you don't want to just come out and say "hey you're the only bidder so whatever price you want to pay is probably fine." On the other hand, it is considered a bit of a breach of decorum to just make stuff up. You can't be like "hey there are six other strategic bidders and they're all higher than you so you have to raise your bid a lot to get to the next round." But you should probably hint at something like that.

Anyway, Guy Hands of Terra Firma Capital Partners bought EMI for 4 billion pounds in 2007 and has been suing Citigroup more or less ever since, claiming that "he was misled by David Wormsley, Citigroup’s head of investment banking in Britain and a longtime banker to Mr. Hands, about plans by Cerberus to offer a competing bid in an auction for EMI." (Terra Firma ended up the only bidder, and would presumably have paid less if it knew that.) His latest lawsuit, in London, adds new claims that other Citi bankers deceived Hands not only about the auction dynamics but also about the quality of EMI's business:

During a May 2007 phone call, Klein "said he thought so highly of EMI, but for the fact Citi was conflicted as sell side, Citi would want to invest alongside Terra Firma and would be interesting in co-investing once deal was done," Terra Firma’s lawyer Anthony Grabiner said in his opening remarks during the first day of the trial in London. "Citigroup thought the quality of the business was in fact poor."

The evidence for that is that other Citi bankers sent dumb e-mails after the deal signed:

“I am amazed you got them to pay up for that old pup,” one said in May 2007. “At long last you sold the pig,” another said. Terra Firma argues the exchanges suggest the bank knew EMI was in worse condition than Hands believed.

This was in 2007, before all the settlements driven by dumb e-mails; I hope everyone is wiser now. But, even in 2007, those e-mails were so unnecessary. "Congratulations on getting the deal done": fine, good, polite. "Congratulations on selling that terrible company to a buyer who was no doubt deceived": no, bad, wrong. At least say it in person!

Elsewhere in M&A, Steven Davidoff Solomon is not really sweating the Dell appraisal ruling:

The Dell case is unlikely to be a game changer. The market price is likely to continue to be the price used because it is so difficult to compute “fair value” otherwise.

The opinion may affect management buyouts, but that may not be such a bad thing — forcing would-be managers to work hard to justify buying their own companies would be better for shareholders. In any case, management buyouts are rare these days; only two were proposed in 2015, according to Factset MergerMetrics.

And here is a story about a lawsuit arguing, not that the 2014 merger between Herman Miller and Design Within Reach was misguided or bad, but that it never happened, because Design Within Reach did not follow the proper corporate formalities to complete a reverse share split that was a precondition for the merger.

Financial regulation.

I hate writing about politicians' financial-regulatory proposals because they are all politics and no financial regulation. Here is Republican Representative Jeb Hensarling's press release, and speech, and executive summary, outlining his "Financial CHOICE Act," which would replace the complicated regulatory soup of Dodd-Frank with a different complicated regulatory soup. I count 32 bullet points, plus eight sub-bullets and seven numbered paragraphs, in the executive summary. One of them is "Repeal the so-called Chevron deference doctrine," which has very little to do with financial regulation but which does underpin all of U.S. administrative law. It seems like a big change, for a bullet point! Two more are "Incorporate more than two dozen Committee or House-passed capital formation bills" and "Incorporate more than two dozen regulatory relief bills for community financial institutions." It gives the impression that he went around asking people for stuff to include in the soup, and they were like "I have two dozen ideas," and he was like "fine but you only get one bullet point." There's also an audit-the-Fed bullet point. It is a kitchen sink of grievances.

To the extent there is a signature financial-regulatory proposal, it's that banks with equity equal to at least 10 percent of their non-risk-weighted assets can get out of basically all other regulation:

Under our plan, banks that maintain a simple leverage ratio of at least 10 percent and, at the time of the election, have a composite CAMELS rating of 1 or 2 may elect to be functionally exempt from the post-Dodd-Frank supervisory regime, the Basel III capital and liquidity standards, and a number of other regulatory burdens that pre-date Dodd-Frank.

Any bank that chooses to maintain a 10 percent leverage ratio in order to qualify for regulatory relief under our plan will be significantly better capitalized than Dodd-Frank or any U.S. or global regulator currently requires them to be. Under the Basel accords, banks must maintain at least a three percent leverage ratio, and U.S. prudential regulators require six percent for those considered globally systematically important.

Collectively, we estimate that those U.S. banks currently identified by regulators as G-SIBs will need to raise several hundred billion dollars in new equity to qualify for regulatory relief in our plan, assuming their asset size remains constant. 

The idea that all financial regulation can be replaced by a single simple leverage ratio, if the leverage ratio is high enough, is weirdly bipartisan. I normally think of it as a leftish idea -- it's most frequently associated with Anat Admati -- though there the emphasis is on the high enough leverage ratio; in the Republican version, the emphasis is on getting rid of all the other regulation. The appeal of it strikes me as essentially magical:

  1. Bank regulation is complicated.
  2. It would be nice if it could be made simpler without sacrificing any effectiveness.
  3. Therefore it can be made simpler without sacrificing any effectiveness.

There may be some form of that magical syllogism that is true, but I have never seen much evidence for the leverage-ratio version of it. Here is an exceptionally smart post from Mike Konczal refuting it by explaining how each aspect of financial regulation covers a different part of a bank's balance sheet:

There are three relevant Dodd-Frank capital requirements here. The first is leverage requirements, which set equity, or the balance between overall debt and overall assets a firm can have. The second is risk-weight requirements, which set the amount of debt a firm can have relative to a reasonable estimate of how risky those assets are. The third is regulating the composition of debt, which sets how much short-term versus long-term debt a firm should hold. As you can see, each of these requirements touches a different part of the balance sheet.

You could quibble -- arguably risk-weighted capital requirements really regulate equity -- but the intuitions seem right, and it's a good quick guide to what sort of disaster Hensarling's plan could create. The Optimized Hensarling Bank would have $10 billion of equity and $90 billion of run-prone overnight wholesale debt, and would use the money to buy $100 billion of illiquid high-yielding junk-rated loans.

Though by the way the banks don't even want this: The fact that they'd need to raise "several hundred billion dollars in new equity" to meet the 10 percent standard "is likely to dissuade many of the top financial institutions from offering broad support for the measure." But if it did become law, someone would find a way to optimize into it.

People are worried about unicorns.

Not a ton of unicorn news today, but Anna Wiener has a little article "On Reading Issues of Wired From 1993 to 1995," which is at least as good:

Stories of tech-enabled social change and New Economy capitalism weren’t in competition; they coexisted and played off one another. In 2016, some of my colleagues and I have E.F.F. stickers on our company-supplied MacBooks—“I do not consent to the search of this device,” we broadcast to our co-workers—but dissent is no longer an integral part of the industry’s ethos.

This reminds me that I seem never to have mentioned Wiener's "Uncanny Valley" in Money Stuff; I am now very late to it but if for some reason you have not read it, you should. There is not much better writing about unicorns out there.

People are worried about bond market liquidity.

Here is Bloomberg Gadfly's Lisa Abramowicz on Pine River Capital Management, which is offering lower fees in exchange for longer lock-ups because of you-know-what:

There's been "a one-way move toward markets becoming less liquid," the firm's CEO Brian Taylor wrote in a letter on Monday. As a result, "We view it as entirely reasonable for us to reduce fees for those investors who are willing to commit longer-lock capital to our investment ideas."

Elsewhere in bond market liquidity, happy first day of European Central Bank corporate bond buying! Here are some charts.

Things happen.

Paul Singer’s New Target: Mossack Fonseca. Sovereign debt: Curing defaults. Europe Prepares to Pick Who Can Run Greece’s Banks. Hedge Funds’ Fast Money Not Welcome as Iceland Bolsters Defenses. Saudi Arabia considers income tax for foreign residents. Australian Regulator Accuses NAB of Manipulating Rate. LendingClub Stumbles Anew, Delaying Meeting With Investors. Valeant is bad at conference calls. Activist Investors Have a New Favorite Target: REITs. Police Probe Finds Nothing to Suggest Aubrey McClendon Committed Suicide. Behind Peter Thiel's Plan To Destroy Gawker. These Are the Best CEOs in America, According to The People Who Work For Them. "You cannot loosen a tie if you wear no tie; you cannot roll up your sleeves if you have no sleeves." Eagle vs. goose. Frosé. A Restaurant That Only Serves Its Food In Jars Is Opening Near NYU. There's a Martin Shkreli musical in the works.

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