Why I (Still) Think Shadow Banking is Key to Financial Reform

Two weeks before a contested primary is probably not the time for subtlety and details, but I want to contest the arguments in these pieces. Though Dayen tries to catch me in a contradiction, I’ve long thought that the project of combating shadow banking was to extend banking regulations to financial activities rather than silo them. So there’s no inconsistency there. Though I’m supportive, I also think that “breaking up the banks” is being overplayed as a financial crisis issue, doing more work as a problem and a solution than its proponents say it does. It’s also a useful check how my mind has and hasn’t changed since 2010.

Shadow Banking

Dayen points out that I co-wrote a 2010 report titled “Creating a 21st Century Glass-Steagall,” in which I said that “the real problem of this crisis is in the overlap between investment banking and commercial banking.” [2] Dayen’s claim is that, since I don’t prioritize Glass-Steagall now, I must be contradicting myself.

In that report, we didn’t call for isolating investment banking from commercial banking, which is what people generally mean by restoring Glass-Steagall. Indeed, the whole point of the report is to focus on the activities of finance, the way mismatches between commercial bank assets and investment bank funding can creating the conditions for panics, as seen in this figure from it:

The report argues that we must extend the regulatory environment of regular banks to shadow banks. This would tackle the regulatory arbitrage that helped shadow banks grow. I still fight for this.

Specifically, the recommendations were: “shadow banks should be subject to the same limits on risk-taking as banks,” “shadow banks should be subject to the same capital standards as banks,” and “[s]tress test liquidity positions.” These were the central ideas for me then, and they are central ideas for me now. (Given how the liquidity stress tests have become one of the central attacks against Dodd-Frank, with whisper campaigns about how the LCR is causing “bond market liquidity,” I think we were on the right track.)

Rather than “trying on contradictory criticisms,” I thought then that splitting investment banking from commercial banking wasn’t related to tackling shadow banking, and I still think so now. It’s remarkable to me how much what I think now is reflected in the 2010 piece, though little has changed that would change my mind.

I have no problem changing my mind, of course, but the biggest argument Dayen presents here is that “shadow banks [benefit from] government subsidies [in deposits]. We shouldn’t cross-subsidize risk-taking with public money.” But that’s not the arbitrage. The arbitrage is being able to do these activities without following reasonable banking rules. From the figure above, the funding stream in question in the crisis wasn’t deposits, and it won’t be in the future, either. What was in question were money market mutual funds, repo markets, commercial paper, and other short-term lending that was subject to panics.

This kind of criticism matters for those who want to take a “both, and” approach here. To the extent that a focus on Glass-Steagall causes us to think we’ve solved the problem if deposits are simply not in the mix, it’s a distraction from the core issue.

Size, Then

I have a column in the next issue of The Nation about how and why Bernie Sanders can simply absorb these issues into his platform, but for now I want to establish a quantitative point: “breaking up the banks” is probably not as much of a solution as it’s being portrayed.

Barney Frank said that Lehman was “very small” when it failed. The Huffington Post writes that this statement is “preposterous.”

Okay, so let’s break up Lehman Brothers, Bear Stearns, and Goldman Sachs. How do we do this? The main push both in 2010 and since then has been the Brown-Kaufman amendment (also called the SAFE Banking Amendment). This would have forced financial firms to have nondeposit debt no larger than 3 percent of GDP. That is about $520 billion now, or $440 billion in 2008.

When Lehman failed, it had debts around $610 billion. If it had been broken up, it would be about 30 percent smaller. Bear Stearns would have been below that cap, so wouldn’t have been impacted. Was the problem with Lehman that it was 30 percent too big? Goldman Sachs currently has about $770 billion in liabilities. When we hear “break up Goldman!” are we thinking that it should slim down 30 percent, to around half a trillion dollars, and that would end the “Too Big” part of Too Big to Fail?

You could argue Lehman’s counterparties were too big. But since most of the big players like JP Morgan and Citigroup have significant deposit funding, they also would be broken up less than you’d think to get their nondeposit liabilities down to $500 billion. Even if the counterparties were also $500 billion in size, that doesn’t seem small enough to end the risks.

For every big firm, there was a small one that caused just as much, or more, of a panic. One of the primary panic-inducing counterparties of Lehman was the Reserve Fund, a money market mutual fund with $65 billion in assets whose $785 million commercial paper exposure to Lehman caused it to break the buck.

Size, Since Then

Brown-Kaufman failed with 33 votes in early 2010, not coming close to the 50, much less 60, votes it would have needed. I was a big supporter of it then, but in the past years I’ve focused more on capital requirements. Why? Some developments that have happened since 2010:

1. Regulators have pushed capital requirements further than I had expected, and capital requirements have gained cachet among reformers as a way to limit the size and concentration of the banks going forward. This approach leaves breaking up the banks to the market, not to Congress. Even in 2009, Congress wasn’t able to pass a bill saying GE had to stop being a financial company or that JP Morgan should have to shrink 6 percent. But both of those things were accomplished by forcing those companies to obey reasonable banking rules and fund themselves with sufficient capital, and letting market forces do their work.

More can be done through this channel, and the lack of discussion of capital requirements (except by Martin O’Malley) is a disappointing part of the Democratic primary.

2. A crucial element of breaking up the banks has been to shift the funding sources. It’s why the SAFE Banking Act has looser restrictions on deposits than on nondeposit liabilities. However, we need more granular differentiation of how short-term debts are, which we can get through capital requirements. Some of the funding source shuffling that breaking up the banks was meant to accomplish has happened, and more can happen through capital requirements. As mentioned above, the liquidity requirements, which focus on this, are very much a target of the finance industry.

3. I have no idea how a single-point-of-entry resolution would work, and more importantly how clean of a process it would be. But the avenue it uses, interjecting at the holding company level, does make me think the institutional and business line makeup of firms is less important than in 2008. Crucially, regulators at the FDIC are showing signs of success in using living wills to try and simplify business lines to make this process credible. I think it is essential to push further there.

That’s how my mind has and has not changed since 2010. I will say it’s great that financial reform is getting a much deeper and more public debate than I would have expected. Here’s to keeping it up through the election and beyond 2017.

[1] I was hoping that I’m a minion like those yellow creatures, both because I am adorable and because I communicate in an inscrutable language nobody can ever understand. But probably not.

[2] Was that an oppo dump? I was hoping my oppo file would have something scandalous, like being caught in bed with either a dead girl or a live boy, rather than a pdf I co-wrote six years ago which doesn’t even argue what the gotcha point says it does.