Money Stuff

Retiring Bankers and Safe Assets


Really, more bank chief executive officers should resign in disgrace. It would take some of the disgrace out of it! I tend to think of a big bank as a roiling random process; the CEO sits on top of it and does the best he can to gently nudge it in a good direction, but ultimately the random process, not the CEO, is in control. If the outcome of the random process is good, the CEO gets paid tens of millions of dollars. If it's bad, then he's the guy who supervised huge losses or widespread fraud or a global financial meltdown or whatever.

The process repeats and is memoryless, and eventually your number is going to come up. (Quick: Name a big bank that hasn't had a big scandal!) And when it does, it seems to me that that right way to think about it is, "well, I've had a good run, time to take one for the team." Like, what are they paying you the tens of millions of dollars for? It's not for swanning into board meetings and saying "our customers now have more accounts than ever." You didn't do that. An army of bankers and salespeople and managers did that; you just got to take the credit. The tens of millions of dollars a year are to compensate you, in expectation, for the couple of hours you're eventually going to spend testifying in front of Congress when something goes wrong. And you'll suck it up, and do the testimony, and then resign, because that is the deal. 

Or that's how it seems to me, but then, I am not a bank CEO. I suppose one worry for the CEOs is that not everyone interprets it this way, and that people who attribute more moral agency to a bank CEO will think that a CEO who resigns after a scandal is personally responsible for the scandal. Which is why they hesitate to resign, and why more of them should. Safety in numbers.

Anyway John Stumpf is leaving Wells Fargo, to be replaced by Tim Sloan, the chief operating officer:

“This was John Stumpf deciding that the best thing for Wells Fargo to move forward was for him to retire -- even though that was a very difficult decision,” Sloan said in an interview. “He wasn’t fired” or even “gently pushed” by the board.

Meanwhile this is kind of wild:

Los Angeles Councilman Paul Koretz is introducing legislation that would require banks working with the second-largest U.S. city to adhere to responsible practices such as barring sales goals, which have been criticized for leading to the bogus account scandal at Wells Fargo & Co.

Imagine banning sales goals in any other industry! But as we've talked about a couple of times recently, banking is not like other industries, and no one is really comfortable with the idea that the people selling us financial products are paid for selling us those products.

Safe assets.

Here are a blog post and related paper by Anna Gelpern and Erik Gerding about the legal construction of safe assets. The paper includes this passage that you should inscribe on your heart if you spend any time thinking about banking and finance:

Unlike the Higgs boson, safe assets are not to be found in nature. They exist in economic theory and in market vernacular. “Safe” in this context does not and cannot mean genuinely risk-free, but rather “safe enough” to ignore the risks for some purposes. We depart from the literature to ask what institutional design elements might make a financial contract safe enough, and what— or who—makes it possible for market participants to ignore the risks, and on what conditions. From this perspective, safe assets look a lot like legal fictions, a familiar device that inserts an assumption known to be false in a chain of reasoning, so as to solve a particular doctrinal problem.

Precisely because there are no risk-free contracts, the law can conjure and maintain safe asset fictions, and place them at the foundation of institutions and markets. It therefore makes no sense to ask whether mortgage-backed securities, bank deposits or Italian government debt are “entirely risk-free.” Their safety, such as it is, ultimately rests on state capacity to regulate, collect taxes, and issue money, and state willingness to deploy these powers in specific ways for the sake of particular constituents and markets. Instead of asking whether a contract is risk-free, it pays to ask what purpose and whose interests might be served by a societal commitment to act “as if,” and to consider the cost of such a commitment. Whether anyone who acts “as if” believes in the inherent safety of safe assets, or chooses to ignore risks believing that she would be made whole, is also a second-order concern. State intervention can justify her actions either way.

So! Here's a story about "A $1 Trillion Paradigm Shift Changes Funding Markets Forever." The paradigm shift is that prime money-market mutual funds were designated as safe (roughly), and now they aren't. A trillion dollars of money moved because we had a societal agreement to act as if money-market funds couldn't lose value, and then regulators decided to stop acting that way. What has changed is not the economic safety or non-safety of those funds -- the volatility or credit quality of their holdings -- but just the legal designation. Before, they were "safe enough" to use as a substitute for cash. Now, they are not. And the repercussions spread out widely and weirdly. To things as important as the calculation of Libor, and as random as mergers-and-acquisitions escrows.

Elsewhere in safe assets, here is an article about Nobel Prize laureate Bengt Holmström's views on money market liquidity:

"I will argue that 'no questions asked' is the hallmark of money market liquidity; that this is the way money markets are supposed to look when they are functioning well," he writes. Attempts to reform credit markets based on insights gleaned from equity markets "can be very misleading" the economist says — and worse, they can stoke the crises that they're trying to prevent.

In particular, the drive for transparency in equity markets is misplaced in markets for safe assets. As I sometimes say, "the point of banking is to conceal risk."

Research troubles.

The job of an investment bank research analyst is to talk to companies, develop opinions about those companies, talk to the bank's traders and salespeople and clients so they know her opinions about the companies, arrange meetings between the companies and the clients so the clients can form their own opinions about the company, and sometimes sit down and publish formal written research describing her opinions about the companies. The written stuff is the most publicly accessible part of the job, but it's not the job. The job is a people job, a sales job, a job of frequent nuanced judgments, not one of occasional oracular Buy or Sell pronouncements.

The Securities and Exchange Commission is skeptical though, and yesterday fined Deutsche Bank $9.5 million because sometimes its analysts talked to traders and customers in ways that disclosed "yet-to-be-published views and analyses, changes in estimates, and short-term trade recommendations during morning calls, trading day squawks, idea dinners, and non-deal road shows." It's stuff like this:

On March 28, 2012, Analyst A hosted a non-deal road show for Company C. Beginning at about 7:30 a.m. and continuing until about 3:15 p.m., the company’s executives met privately with DBSI customers. Analyst A attended all of these meetings. Shortly after the meetings had ended, Analyst A contacted certain DBSI customers, sharing his impressions from the meetings and his view of the company’s prospects for the quarter. That night, Analyst A sent an email to the head of DBSI’s Institutional Client Group for North America, stating that he had a “note hitting in the a.m. [that] will likely rattle the stock.” The following morning, March 29, 2012, Analyst A published a research report on Company C that contained information that was substantially the same as the information that he had conveyed to at least one of the DBSI customers with whom he spoke the day before, as reflected in notes that were taken by that particular DBSI customer

Naughty! Though if he hadn't called any clients, but just kept his impressions to himself until he published his note, then the customers who were at the non-deal roadshow, and heard directly from the company, would still have an edge. Would that be okay? Anyway, this is a trend. We talked a few months ago about a Financial Industry Regulatory Authority action against an Arkansas brokerage called Stephens Inc., which Finra fined for doing similar stuff. I said then that "research analysts who only publish research, and never interact with clients or salespeople, just aren't that useful." Meanwhile in Europe, "unbundling" rules more or less forbid banks from providing free research to win commission business, meaning that banks will have to charge for research and that research will have to earn its keep. There does seem to be a global regulatory agreement to ... kind of get rid of sell-side investment research? Which, I mean, it has not always covered itself in glory, but it hardly seems like the biggest problem in the financial industry? Its big scandal was many years ago, and in the light of subsequent scandals looks positively quaint. 


"Steven A. Cohen is boosting the bonuses he pays to the top traders managing his $11 billion family fortune, but only if they beat the market," and in some sense: Har har har, why would you get a bonus for not beating the market? But also, think about the skepticism that people automatically bring to incentives in finance. (See those Wells Fargo sales goals, above.) When SAC Capital was brought down for insider trading, one thing that was held against it was that employees were encouraged and incentivized to seek out "edge." "Edge," to investigators, was taken as a synonym for "illegal inside information." But it is also a decent synonym for "alpha." A pay structure that rewards employees for making money will also -- necessarily -- reward them for cheating to make money. In the financial industry recently, an assumption has grown up that the incentives are all about the cheating.

Elsewhere in incentives, here are popular and technical versions of an article about incentives by Daniel Barron, George Georgiadis and Jeroen Swinkels. One thing that they find is that "if the agent is risk-neutral and protected by limited liability, then a linear contract implements any effort level at maximum profit": That is, just giving the agent a fixed share of the profits -- a percentage commission, a bonus as a fixed cut of profits, or linear participation in a hedge fund -- is a better incentive structure than any more complicated incentive design.

And elsewhere in hedge funds and insider trading accusations: "SEC Said to Demand That Cooperman Agree to Hedge Fund Suspension."

The CFPB is a little bit unconstitutional.

On Tuesday, a federal appeals court ruled that the Consumer Financial Protection Bureau was the smallest possible amount of unconstitutional: Having an independent agency headed by a single person, not removable by the president at will, violates the appointments clause of the Constitution. But you can fix that by just making the head removable by the president. Which the court did. (Here is the opinion.) So everything is more or less fine. Adam Levitin points out that "the CFPB’s existing rule makings and enforcement actions remain valid and unaffected," and it still has budgetary independence. "The Next President Might Be Able to Fire One of Our Most Important Consumer Watchdogs for Any Reason," worries Helaine Olen, but that seems like a pretty esoteric worry. For most useful purposes, the president outranks the head of the CFPB. If a president decides that consumer financial protection is not worth worrying about, he or she has a lot of ways to gut consumer financial protection, beyond having at-will firing power over the head of the CFPB. Still I suppose there will be an appeal.

Blockchain blockchain blockchain.

Here's a New York Times article about how the blockchain is so cool, so cool, for central banks:

In the systems being discussed, each player in the system would communicate with all the others anytime money moved in the system, allowing everyone to update the ledgers on their computer systems simultaneously. This would provide multiple backups if the central bank’s computers came under attack. It would also hypothetically allow them to complete transactions much more quickly, and would make it easier to spot rogue actors.

On the other hand, "some critics say the public interest in the blockchain has been all talk and no substance — an easy way for stuffy central bankers to appear hip and relevant." To me, it seems fairly straightforward to assume that a central bank's ledger could be ... centralized. Like, the central bank issues the money; it could just keep a list of how much money everyone has, and that could be the official list, and that would be that. You don't need a blockchain. If you keep the list on a computer, you could even complete transactions very quickly. The innovation of blockchain technology in the bitcoin context is that nobody issues the bitcoins: The bitcoins are created by the blockchain itself; the absence of a trusted central issuing party is what required that technology in the first place. The purpose of a blockchain for central banks is a bit less obvious, though, sure, it could create extra backup copies and make hacking harder. That's not nothing.

Elsewhere, Craig Pirrong is skeptical of a proposal for "hyperledger technology designed to disintermediate central counterparties (CCPs) from the clearing process."

Fake robot!

Can't resist a fake robot, sorry. From a Commodity Futures Trading Commission enforcement action:

In yet another aspect of the solicitation fraud scheme, the Complaint alleges that Defendants also offered for sale a “robot” that automatically placed trades via the customer’s trading account according to signals generated by the trading system. According to the Complaint, Defendants claimed that “your robot knows exactly what to do,” and has a “+90% success rate!” In reality, the Complaint alleges, there was no 90% success rate for the robot, and the robot had never been tested using real money. The Complaint further notes that, “during the CFTC’s investigation, Defendants acknowledged that the Robot did not work and that customers lost money attempting to use it.”

People are worried about unicorns.

Well here is "How Not to Strangle Your Unicorn," about now sadly asphyxiated unicorn, and the short version is I guess don't let your unicorn lose money on every sale?

“We kept getting new customers, but we started to see that the profit margins were zero to negative,” said Howard Morgan, an investor at First Round Capital who was on Fab’s board. “The only way the model made money was if people bought multiple items. If you could lose money on each customer, you’re not going to get money out of the whole, and it took us too long to understand that’s what was happening.”

I love the idea that this is an esoteric problem, that Silicon Valley's best and brightest sit around their lavishly appointed offices arguing late into the night: "We keep buying widgets for $10 and selling them for $8, and traffic is great, how are we still losing money?" Anyway yeah I guess the point is that negative gross margins are a choking hazard, for your unicorn.

Elsewhere, the pitches for the Snap Inc. initial public offering have concluded, and "Morgan Stanley and Goldman Sachs Group Inc. will lead the offering and were notified of their role early this week," but I still don't know what dumb stunts the banks pulled to convince Snap that they really "get" the Snapchat brand. If you know ... I guess ... Snapchat me? 

People are worried about bond market liquidity.

Here's a press release about S&P Dow Jones Indices and MarketAxess teaming up "to jointly develop indices that will track the most liquid segments of the U.S. corporate bond market":

"Investor demand for U.S. corporate bonds continues to show strength, and due to this growth, liquidity has emerged as a top-of-mind interest," said Alex Matturri, Chief Executive Officer of S&P Dow Jones Indices.


Me recently.

I interviewed Brad Katsuyama, the co-founder and chief executive officer of IEX, for Bloomberg Markets magazine. (There is also an audio version at the Bloomberg Odd Lots podcast.) One thing that I find fascinating about Katsuyama is that he is so often set up as a symbol of something or other -- the dangers of high-frequency trading, the need for long-term investors, the unfair and rigged nature of Wall Street -- but he doesn't seem to think of himself that way. He built a business, has clients who like it, has a market niche, and doesn't seem interested in being a more generalized Crusader Against the Evils of Wall Street. When asked about, say, reforming listing standards to push companies to focus more on the long term, or expanding into other geographies and asset classes, or reforming retail trading -- all areas where his symbolic clout might give him some traction -- Katsuyama seems interested mostly in focusing on the business that he's good at. "We’re small," he says. "We can’t get distracted. We can only focus on a few things."


Things happen.

Do discounted cash flow models work in a low-interest-rate world? Cliff Asness and Lasse Pedersen on the possibility that the average active manager could outperform the average passive manager. Truly Active Emerging-Markets Stock Funds Have An Edge–Study. Citadel, Millennium Stage Comeback to Reverse Losses in 2016. Bridgewater Headquarters Evacuated After Bomb Threat. Deutsche Bank Said to Implement Hiring Freeze as CEO Cuts Costs. Future of Banking Looks Dark—Why That’s a Problem. The New Plan to Bail Out ‘Too-Big-to-Fail’ Banks. Fed Minutes Suggest Yellen Made the Difference in ‘Close Call.’ "'One in ten directors believes the optimal representation of women on boards should be 20% or less,' the report said, and '97% of those who believe this are male.'" Measuring how bad Twitter is. SEC Breaks Record for Number of Enforcement Cases. Flash Boys Might Keep Trade Code Secret as CFTC Mulls Tweaks. The Leverage Ratchet Effect. Regulatory Leveraging: Problem or Solution? "A computer that is a hundred times more accurate than a human, and a million times faster, will make 10,000 times as many mistakes." Donald Trump "would be the first president since Harry Truman without a pet in the White House." Loose change. Fake ball bearings. A Close Look at the FBI's File on Wu-Tang Clan. Eat, Pray, Tail: I Wore a Tail for 1 Week and Learned Nothing. How to Incorporate Human Remains Into Your Dinner Party. 

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

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