FT premium subscribers can click here to receive Free Lunch every day by email

This week I attended a lecture by Laurence Ball, the US macroeconomist, who presented his new book on the Federal Reserve’s handling of the Lehman Brothers failure 10 years ago in September. His argument has circulated in less developed form for a while — as a working paper and as a summary column — and is newly relevant with the anniversary of the financial crisis. Its scathing criticism of the Fed is a contrast, to put it mildly, to the remembrance conference with the three principal policymakers hosted by Brookings last month. That conversation between Ben Bernanke, Tim Geithner and Hank Paulson is well worth watching (or reading as a transcript). On Lehman, they agreed that they knew the bank’s failure would be bad, but that the Fed could not have lent to it without an external business willing to buy the bank and thereby inject equity.

Ball makes a twofold counterargument. First, he says the archival record shows that questions of the legality of Fed lending to Lehman (whether the collateral was good enough or the bank was solvent) did not play a big role in the deliberations about what to do with the bank at the time. Second, he argues that in economic fact, there was ample collateral on Lehman’s books that was eligible for Fed lending under the Primary Dealer Credit Facility, which the Fed did extend to other big investment banks as the fallout from the Lehman bankruptcy began to unfold.

Why, then, did Lehman’s private creditors — in particular its “repo” lenders whose loans were secured by assets given them by the bank as collateral — stop rolling over their loans in that second week of September 2008? There were legitimate doubts about Lehman’s solvency. Ball himself cites estimates of its equity between minus $4bn and $13bn, ie either negative or vanishingly small for a balance sheet of more than $600bn. But why did secured creditors care about solvency, given that they had collateral to secure their claims? Ball, too, seems puzzled by this, but the answer seems to be that for practical and reputational reasons, they would rather not have to wait for a bankruptcy process to get their money back.

That, of course, constituted a reason for the Fed to be cautious about lending, too. But behind all this lies the presupposition that not rescuing Lehman — in the sense of keeping it going in its then incarnation by helping it make all its creditors whole — was a big policy mistake. That is why it matters whether the responsible officials’ explanation for their actions is credible. Paradoxically, however, they may be protesting too much. That’s because Ball’s analysis of why it is not credible gives reasons to think a rescue was not desirable even if it was both legal and practically do-able.

The detail here matters. Lehman was, of course, a network of different legal entities. At the top was Lehman Brothers Holdings Inc, which was the main entity filing for bankruptcy on the morning of September 15. But as Ball highlights, different subsidiaries were treated differently. In particular the Wall Street broker-dealer subsidiary, Lehman Brothers Inc, was not immediately forced into bankruptcy. LBI was given rescue loans until part of it was sold to Barclays and the rest liquidated not through the normal bankruptcy process but a special resolution regime for stockbroking companies. (This regime prevents fire sales and aims to engineer a transfer of customer portfolios and the brokerage activity to another broker.)

That seems to be the right approach: if you can identify the system-critical parts of a teetering financial company, the best policy is not to rescue the whole thing but do just what is necessary to keep those parts operating. That can be done without making every last creditor whole. One such system-critical part was the broker-dealer subsidiary, which was a key hub in the Wall Street repo financing market. Another was the London-based broker-dealer, Lehman Brothers International Europe, which was not given Fed financing — though one can sensibly argue that this UK-registered company was the responsibility of UK regulators.

Another systemic effect of the Lehman bankruptcy was to cause a money-market fund, which many savers treated as safe bank deposits, to “break the buck”, or return less than 100 cents on each dollar of nominal fund value. This made the whole market freeze up, and cut off a main channel of short-term financing for corporations (MMFs invest in companies’ short-term debt securities). The US government quickly had to put in place a guarantee scheme for MMF investments. But this is no argument for rescuing Lehman either. Keeping it afloat would just postpone losses on its unsecured debt in an eventual insolvency; even if it could get back on its feet, the money market was exposed to other failing businesses, such were the losses to be realised in the system. It is, in general, a bad rule to bail out debtors to protect their creditors — it is always cheaper to help the particular creditors that matter.

While the treatment of Lehman can be faulted on many grounds, not rescuing it in toto is not one of them. The system-critical parts could have been better segregated; the Fed could even have required more assets to be transferred to them from other parts of businesses. The need for a money market guarantee could have been anticipated. The consequences of Lehman’s wind-down amount to a case for managing bank liquidations better, not for avoiding them altogether.

Other readables

  • When government policy manages to spur innovation, it is not by trying to correct existing markets but by creating and shaping new ones.
  • In case you missed it: the New York Times investigation of how Donald Trump’s wealth largely came from his father, in ways that minimised the tax due.

Get alerts on Global financial crisis when a new story is published

Copyright The Financial Times Limited 2019. All rights reserved.
Reuse this content (opens in new window)

Follow the topics in this article