Levine on Wall Street: Bad Culture and Fake Factors

Banks want to follow regulations, they just want them to be the right regulations. Also: Buffett and IBM, factor mining, dollar stores, money funds.

Are banks evil?

Yesterday the Fed had a big conference to harangue banks to be less evil, with New York Fed President William Dudley sort of leading the charge by telling bankers that if they do not "do your part in pushing forcefully for change across the industry," i.e. less evil, then "the inevitable conclusion will be reached that your firms are too big and complex to manage effectively." So that's fun. But for my money the more interesting speech was from Fed Governor Daniel Tarullo, who got into nuance about how regulation actually works:

In some firms the attitude we perceive is one of a mere compliance exercise. The firm proceeds to address the deficiencies identified by the Fed in a discrete, almost check-the-box fashion. To oversimplify a bit, I would say that our sense is that management at these firms wants the hurdle to capital distribution removed, but once the specific problems have been remedied, they want to move on. If this is the attitude we perceive, I suspect the working level employees of such firms do the same. The supervisory reaction in such cases is quite likely to be an inclination toward greater scrutiny.

Other firms, by contrast, seem to have internalized the aims of the risk-management processes and systems that we expect of them. In these firms the dialogue can be quite different, with supervisors observing that, even as a specific problem is addressed, such as deficiencies in estimating losses for a particular loan portfolio in a tail event, the firm has gone back to think about how the identified shortcomings fit into their overall risk decision-making and management processes. These firms will, on their own, then consider whether changes in other areas are needed. Again, I suspect that the line employees in these firms are also hearing a different message and presumably, to at least some degree, will behave in accordance with that message when they encounter risk-management issues not covered specifically by Fed communications.

It's fun to guess which banks are which. But this is also important:

Before leaving this topic, I want to observe that regulators can unwittingly reinforce what I have termed a mere compliance mentality. I would first note that the detail of many regulations means that attention to narrow issues of compliance is sometimes wholly understandable and, indeed, essential. Banks, like other regulated entities, need to be able to determine how a regulation actually applies to them. Beyond that kind of unavoidable focus on narrow compliance, however, management and line employees are more likely to adopt a mere compliance mentality where regulations appear to them to have been poorly drafted or implemented.

If the rules are dumb, Tarullo is saying, they won't foster a compliance culture. This sort of thing would not sound particularly sympathetic coming from a bank: "The Volcker Rule is not a sensible way to reduce the risks that led to the 2008 financial crisis, so we committed a bunch of fraud in Mexico." But it's surely true, right? If bank supervision is not a collaborative process where everyone at least partially agrees on the goals of reducing risk and the means of getting there, then it will be a game where banks try to pull one over on regulators. Consider the Carmen Segarra tapes for a minute. On the one hand, what sort of compliance culture did Goldman Sachs have there? On the other hand, what sort of compliance culture is fostered by nitpicky arguments over whether a conflicts policy was "bad" or "nonexistent"?

Should stocks go up?

Yesterday IBM's stock price lost $12.95 a share, and Berkshire Hathaway owns 70.2 million shares, so Berkshire was down about $909 million, and $909 million is awful close to $1 billion, and the internet is what it is, so there was a lot of "Warren Buffett lost a billion dollars, ha!" type stuff yesterday. But Dan McCrum makes the excellent point that Buffett is on record saying that he wants IBM stock to stay low: He's a long-term holder and the company is buying back stock, so the more it buys back cheaply, the more efficiently Buffett will consolidate his position. In fact Buffett has just about broken even on his stock so far, has received three years of dividends, and has seen his stake rise from 5.5 percent to 7 percent just on buybacks. If Buffett is right about the long-term outlook, then he's in great shape. I suppose if the market is right about the long-term outlook then he's in less great shape, but still. Andrew Ross Sorkin is skeptical.

Elsewhere in tech buybacks, this is sort of a fair question:

How can Apple “return” capital to shareholders if those shareholders never supplied Apple with capital in the first place? As I pointed out in my earlier post, the only funds that Apple ever raised on the public stock market was $97 million (about $274 million in today’s dollars) at its IPO in 1980.

I've said it before but I'll say it again: If your model of public equity markets is that they're where companies go to raise money to fund their businesses, your model is wrong.

Moneyball, but for money.

Did you know that you can use statistics to pick stocks? Hmm, you did? Well, did you know that you can sort of squint at those statistics and pretend that they're baseball statistics? Why would you want to do that, you ask? I don't know. Baseball! Sports metaphor! Just buy some stocks. That seems to be the thesis of a Goldman Sachs equity research note, and good lord. Elsewhere, Gawker's giving stock tips.

Here, on the other hand (via Tyler Cowen), is a new NBER paper (ungated PDF here) about research into the cross-section of expected stock market returns, finding that "most claimed research findings are likely false." The intuitive idea is that hundreds of academics (and hedge funds) write hundreds of papers trying to find factors that explain stock market returns and that can be used to outperform the broad market:

We observe a dramatic increase in factor discoveries during the last decade. In the early period from 1980 to 1991, only about one factor is discovered per year. This number has grown to around five in the 1991-2003 period, during which a number of papers, such as Fama and French (1992), Carhart (1997) and Pastor and Stambaugh (2003), spurred interest in studying cross-sectional return patterns. In the last nine years, the annual factor discovery rate has increased sharply to around 18. In total, 162 factors were discovered in the past 9 years, roughly doubling the 90 factors discovered in all previous years.

But since they keep mining the same data for the same sorts of factors, their statistical thresholds for significance are probably too low. If you use a t-ratio of 3, as the authors advocate, rather than the more popular 2, you find that most of the factors that have been identified don't actually work, for some value of "actually." Fun exercises include: Does the failure of factor mining contribute to recent underperformance by hedge funds? Or are the measures of hedge fund underperformance themselves wrong?

What's up with the Dollars?

Ronald Barusch's view is similar to mine, which is that the Family Dollar/Dollar Tree/Dollar General merger battle is a game of timing, and Elliott Advisors' proxy fight is too far back in the queue to matter. By the time Elliott gets to vote on its director nominees at the 2015 annual meeting, Family Dollar will have been bought by Dollar Tree or Dollar General or some dollar to be named later, and the proxy fight will be moot. So the proxy fight isn't serious; it's just a way to register Elliott's displeasure with the board in a public way, and push it to the General rather than the Tree.

How Fidelity kept money market funds safe.

Basically everyone thinks that money market funds should have a floating net asset value, so it's a little weird that the Securities and Exchange Commission declined to make that the rule. (I mean, it sort of did, but with key exceptions.) Here is the story of how that happened, which involved a ton of lobbying by Fidelity, the leading provider of money market funds. Fidelity is a great candidate to be the moral center of the financial world -- that name, it's in Boston, it's got so much retail money, the business is pretty straightforward -- which of course gives it a lot of cover to lobby for stuff like this.

Things happen.

Basis arbitrage on Greek credit default swaps seems fun. The SEC likes using administrative law judges (cf.). Floating-rate debt is great when interest rates go down. Sears keeps borrowing more money from Eddie Lampert. Fannie and Freddie shareholders keep suing the government. AbbVie/Shire is totally dead. You couldn't pay anyone in pounds for a while yesterday, seems bad. Soon you'll be able to invest $250,000 with the Carlyle Group. Short-selling is fine, and criticizing journalists for talking to short sellers is dumb. Foreign exchange rigging fines could be $41 billion, including $6.5 billion for Deutsche Bank, says Citigroup, and I guess I'll take the under? I guess? "We find that discounts for illiquidity can be surprisingly large, approaching 30 to 50 percent in some cases." "72 percent of the for-profit programs at 7,000 schools produced graduates who on average earned less than high school dropouts." Wine baths. Humanity’s Last Great Hope: Venture Capitalists. Sad Guys on Trading Floors is back. The Internet: A Glossary

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Matt Levine at mlevine51@bloomberg.net

    To contact the editor on this story:
    Zara Kessler at zkessler@bloomberg.net

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