Bethany McLean and "Capital Punishment" [updated]

Slate carried an article by Bethany McLean on why stricter capital requirements would not have prevented our most recent financial crisis.  I made this point in two posts that pre-date her article: bank capital's moment of truth, and debunking the myth of bank capital.

What caught my eye in this article was her solution in an ideal world for preventing financial crisis.  She championed asset-level disclosure as required under the FDR Framework.

Pretty much everyone—everyone, that is, except bankers and a handful of their supporters—agrees that there's one surefire way to make the global financial system safer: Require banks to increase capital reserves. 

But there's a risk in seeing stricter capital requirements as a catchall solution. Think about the past (the financial meltdown in 2008) and the present (the fear that a default by Greece will ignite another financial crisis ). If banks had held more capital, would we have avoided either mess? I'm afraid the answer is no. 

... A curious paradox of capitalism is its aversion to capital. When you think about it, capital is money (classically in the form of common stock) that isn't owed to anyone, so it can protect those who are owed money, like bondholders, by absorbing losses first. It's expensive, because the providers of capital want to be paid for putting their necks on the line. 

Bankers, whom we think of as the ultimate capitalists, can't stand the stuff. Indeed, the new capital standards being contemplated in Basel have bankers up in arms. 

At a House Financial Services Committee meeting last week, industry representatives argued, among other things, that large capital requirements will raise the cost of the loans they make. Prominent economists, including Simon Johnson, former chief economist of the International Monetary Fund, andAnat R. Admati of Stanford, say this is bunk. 

Regardless of who's right about that, what seems to get lost in the public debate is what capital isn't. It isn't a literal pile of cash or stock certificates, but rather an accounting construct based on the difference between what the bank says its assets are worth, and what its liabilities are. 

If the bankers are wrong about how much their assets (i.e., their loans), are worth, then they're wrong about how much capital they have.

Her description of what capital is needs to be re-read.  She is very explicit in telling you that capital is an accounting construct.  As a result, it is a number that is subject to being gamed by the banks and their regulators - think suspension of mark-to-market accounting at the start of the credit crisis.

The reason this blog has been down on capital is because it is an accounting construct.  It may or may not reflect reality.

This blog has favored solvency (a bank is solvent if the market value of its assets is greater than the book value of its liabilities) instead because it does reflect current reality.

... Historically, most, if not all, bank failures have resulted from a run on the bank, meaning that short-term lenders or depositors yank their money because they doubt the value of the bank's assets. Once that fear sets in, it doesn't matter whether the bank is well-capitalized or not: It runs out of cash anyway.  

The long form of this point was made in the posts on bank runs found in the guide to FDR Framework.

Bear Stearns and Lehman Brothers could plausibly argue, at the time of their demise, that they were well capitalized, but they still suffered a run because investors no longer trusted the value of the assets. The very situation in which a bank most needs a lot of capital is when investors start to suspect that the bank might be lying about the value of its assets. At that point, no amount of capital is enough.

... More broadly, think about how the crisis was finally averted. The common perception is that the Troubled Asset Relief Fund, or TARP, which put billions of dollars of capital into the biggest banks, did the trick. But that's not quite right. TARP showed that the government wasn't going to let the banks fail, and it was that demonstration which averted the crisis of confidence. What really made a difference was the other things the government did, most notably the FDIC's blanket guarantee of all the banking sector's debts.

This blog has repeatedly made the point that it was the US government putting its credit on the line, not programs like stress tests, that stopped the crisis of confidence.

... Not only would higher capital requirements have failed to prevent these crises; conceivably they might have made them worse. 

The risk weighting that the regulators apply to assets encourages banks to hold more of the assets that are supposed to be low-risk. That's why banks all owned a lot of mortgage-backed securities—they were purportedly low-risk, and banks didn't have to hold much capital against them. Sovereign debt like Greece's was also purportedly low-risk; that why banks owned a lot of it. 

Subsequent events showed that the risk weightings left something to be desired. 

... If banks had more capital, wouldn't they take less risk, because the bankers would have more to lose? I'm not sure the answer is yes, because the capital is still other people's money.

Previously, using Merrill Lynch as an example, it was shown that more capital could in fact lead to more risk taking.  By simply substituting low risk mortgage-backed securities for low risk sovereign debt, Merrill would have improved its return on equity without having changed the risk weightings under the capital requirements.  As a result, Merrill could have lost far more money.

A better argument in favor of stricter capital requirements is that if there were more capital and less debt in the system, then the likelihood would diminish that short-term lenders and depositors would get scared enough to create a run. 

But in the gigantic, murky, interconnected world of global finance, I'm not sure there's any amount of capital—any amount that still allows the system to function, that is—that would provide enough comfort under all circumstances. 

In an ideal world, we would do a lot more than require banks to hold more capital. We'd mandate transparency in the valuation of their assets, so that outsiders had a prayer of knowing what the stuff on banks balance sheets was actually worth.

With transparency in the valuation of their assets, we would also get each of their individual positions.  This would allow outsiders the ability to value these assets for themselves and determine what they are actually worth.

We'd force banks not to rely on short-term funding, so they couldn't have a liquidity crisis. And we'd make sure that regulators understood all the unique risks in each institution that they were regulating.

As discussed at length in the post on the new model for bank supervision, it is the disclosure of the asset and liability-level data that allows the regulators to understand all the unique risks in each institution.  This disclosure allows the regulators to harness the analytical capabilities of the market to understanding each institution.

But none of that is going to happen. So maybe requiring more capital is the best we can do. But we should be aware that it's an imperfect solution, and that getting the details wrong risks creating catastrophic problems down the road.

Do not be so pessimistic Bethany!

When politicians (see UK Prime Minister David Cameron), central bankers (see Bank of England's Mervyn King and the Financial Policy Committee), regulators (see the Bank for International Settlements, otherwise known as the central bankers' bank) and the mainstream media (see the Economist magazine) call for asset-level disclosure, it is a clear signal that they too know that disclosure is needed.

To paraphrase Winston Churchill, the overseers of the financial system always get it right with disclosure, they just have to try everything else first.

Update
The U.S. General Accounting Office issued a report that weighed in on the short-comings of bank capital that result from it being an accounting construct.

More than 300 insured depository institutions have failed since the current financial crisis began in 2007, at an estimated cost of almost $60 billion to the deposit insurance fund (DIF), which covers losses to insured depositors. Since 1991, Congress has required federal banking regulators to take prompt corrective action (PCA) to identify and promptly address capital deficiencies at institutions to minimize losses to the DIF. 

The Dodd-Frank Wall Street Reform and Consumer Protection Act requires GAO to study federal regulators' use of PCA. This report examines (1) the outcomes of regulators' use of PCA on the DIF; (2) the extent to which regulatory actions, PCA thresholds, and other financial indicators help regulators address likely bank trouble or failure; and (3) options available to make PCA a more effective tool. 

... Although the PCA framework has provided a mechanism to address financial deterioration in banks, GAO's analysis suggests it did not prevent widespread losses to the DIF--a key goal of PCA. 

Since 2008, the financial condition of banks has declined rapidly and use of PCA has grown tenfold. However, every bank that underwent PCA because of capital deficiencies and failed in this period produced a loss to the DIF. Moreover, these losses were comparable as a percentage of assets to the losses of failed banks that did not undergo PCA. 

... Since the 1990s, GAO and others have noted that the effectiveness of PCA, as currently constructed, is limited because of its reliance on capital, which can lag behind other indicators of bank health. That is, problems with the bank's assets, earnings, or management typically manifest before these problems affect bank capital. 

... GAO tested other financial indicators, including measures of asset quality and liquidity, and found that they were important predictors of future bank failure. These indicators also better identified those institutions that failed and did not undergo the PCA process during the recent crisis.

An unbiased observer like the GAO realizes that looking at a bank's assets and their quality is of more predictive value in measuring the solvency of a bank than looking at a bank's capital position.  This is not surprising, because deterioration in the assets shows up before accounting losses are recognized.

The FDR Framework and its requirement to disclose the asset-level data incorporates this practical reality.