Were Banks “Too Big to Fail” Before the Fed?

According to Gary Gorton and Ellis W. Tallman, in their  recently released NBER Working Paper, some were.

For several decades prior to the Fed’s establishment, Gorton and Tallman note, “private bank clearinghouses provided lending facilities and assisted member banks when they needed help.”  The pattern of such assistance, they say, reflected a privately-adopted TBTF policy that “was a reasonable response to the vulnerability of short-term debt to runs that could threaten large banks and thereby the entire banking system.”

These clearinghouse bailouts appear, furthermore, to have succeeded in averting more serious crises.  Since, according to Gorton and Tallman’s understanding, “[t]he logic of modern bailouts is the same” as that adopted by 19th-century clearinghouses, they conclude that TBTF remains a reasonable policy today, and not, as many suppose, an invitation to excessive bank risk taking.

But there’s a flaw in Gorton and Tallman’s reasoning, and I’m afraid it’s a lulu.  The flaw consists of their failure to understand what “Too Big to Fail” means.  That meaning has been clear from the time  Congressman Stewart McKinney first popularized the notion during a hearing concerning the Continental Illinois bailout.  “Mr. Chairman,” McKinney said, “Let us not bandy words.  We have a new kind of bank.  It’s called too big to fail.  TBTF, and it’s a wonderful bank.”

In case it isn’t obvious, Congressman McKinney was being sarcastic.  What was “wonderful,” in the sense of “amazing,” was the fact that Continental would remain a going concern despite it’s having been rendered insolvent by bad loans it had made or purchased.  The point is that, if TBTF means anything, it means that certain financial institutions are exceptions to Walter Bagehot’s “classical” rule according to which last-resort loans and other kinds of emergency aid should be confined to solvent financial firms.  It is precisely because TBTF can mean putting insolvent firms on government life-support that critics of the policy see it as a source of moral hazard.

All of this appears, somehow, to have escaped Gorton and Tallman’s attention.  For otherwise they could not have failed to note the crucial difference  between “the logic of modern bailouts” and that underlying the private pre-Fed bailouts that their paper describes.  For while clearinghouses did occasionally rescue “big” banks, they never rescued insolvent ones.  Instead, as is clear from the evidence that Gorton and Tallman themselves muster, clearinghouses went to great lengths to assure themselves of a bank’s solvency before they’d lend it a nickel.  “If a bank possesses good assets and is merely temporarily embarrassed,” a 1901 source cited in the paper explains, “it is good policy of the [clearinghouse] association to prevent [its] failure.”  That’s not too big to fail.  It’s too sound to fail.

And while it’s true that, according to the same source, clearinghouses were especially concerned to prevent “important” members from failing, because such failures try “the weak points of all the banks,” such banks still had to have plenty of  “good assets” to qualify for help.  During the Panic of 1884, for example, the failure of the Metropolitan National Bank might have had disastrous consequences for its many correspondents.  Yet the New York Clearing House agreed to offer it support in the form of clearinghouse loan certificates only after its examiners determined that the Metropolitan “had sufficient assets in good condition” to warrant that support.  In short, “Bigness” (or “interconnectedness”) may have been a necessary condition for clearinghouse support.  But it was never a sufficient condition.

In a TBTF regime, in contrast, “bigness,” or “interconnectedness,” can suffice.  Hence Continental Illinois.  Hence other bailouts of larger, insolvent financial firms since them, including the Fed’s rescues of Citigroup and AIG during the recent crisis.  As Tom Humphrey points out,

the Fed ignored the classical advice never to accommodate unsound borrowers when it helped bail out insolvent Citigroup and AIG.  Judging each firm too big and interconnected to fail, the Fed argued that it “had no choice” but to aid in their rescue since each formed the hub of a vast network of counterparty credit interrelationships vital to the financial markets, such that the failure of either firm would allegedly have brought collapse of the entire financial system.  Fed policymakers overlooked the fact that Bagehot already had treated this argument, and had shown that interconnectedness of debtor-creditor relationships and the associated danger of systemic failure constituted no good reason to bail out insolvent firms.

The difference between the “logic” of the private-market financial firm rescues of the pre-Fed era, in which the rescuers themselves had “skin in the game,” and today’s taxpayer-supported rescues, is a difference of no small importance.  It is why we worry about TBTF these days, while no one worried about it back then.

I hope I won’t be misunderstood as intending to suggest that there’s no merit in Gorton and Tallman’s research, because that isn’t my intention at all.  In fact I consider their inquiry into clearinghouse rescues of the pre-Fed era quite valuable.  The problem isn’t the research itself, but the fact that Gorton and Tallman draw the wrong lesson from it.  The lesson isn’t that TBTF is a dandy doctrine, and that those looking for the causes of financial-system fragility today should look elsewhere.  It’s that private, 19th-century financial actors appeared to have managed last-resort lending more responsibly than their modern, government-appointed counterparts.

[Cross-posted from Alt-M.org]