How to Rescue a Bank

What's the best way to rescue a failing bank? Bear Stearns got bought by JP Morgan with the help of a $29 billion Treasury backstop. Northern Rock was nationalized. Lehman Brothers was allowed to fail outright, enter bankruptcy, and eventually have its operations sold off to Barclays and Nomura. WaMu too was allowed to fail, but only after the OTS had pre-orchestrated a sale of its operations to JP Morgan. Merrill Lynch and HBOS were acquired with the help of regulators happy to waive anti-trust requirements. Fortis is being recapitalized with $16 billion of Benelux government cash. And now there's Wachovia, which is a new model again: it's being acquired by Citigroup, with the FDIC providing a monster $270 billion guarantee that no more than $42 billion of the bank's loans will sour. In return, the FDIC is receiving $12 billion of Citigroup preferred stock and warrants.

Of all these models, the WSJ makes a compelling case this morning that the Lehman one is the worst. Tim Geithner, especially, knew that the consequences of allowing Lehman to fail would be gruesome -- but he did it anyway, and things got much worse than he imagined much faster than he feared.

With hindsight, it would have been a less bad idea to spend $30 billion bailing out at least the senior bondholders of Lehman Brothers: such an action might well have prevented the need for the current $700 billion financial bailout.

I, for one, was a big supporter of the no-bailout-for-Lehman option. But it turns out that Lehman really was too big to fail, or too interconnected to fail, or maybe it was just that the markets were too fragile for any bank to fail.

There are three silver linings to the Lehman failure, though.

Firstly, things are so chaotic right now that a huge bank like WaMu can also fail, leaving its senior bondholders with pennies on the dollar, and the markets barely blink. If Lehman hadn't already failed, it's entirely conceivable that the WaMu collapse could have had much the same repercussions -- but since the repercussions had already happened, it seems that there was a limit to how much worse things could get, at least in the short term. (For a pessimistic view of what the WaMu deal means in the longer term, see John Hempton.)

Second, the consequences of the Lehman failure have forced the US government to implement a massive all-things-for-all-banks rescue plan, rather than continuing to bumble down its erstwhile ad hoc road. Up until now, regulators have been so reactive rather than proactive that there hasn't been a hint of strategy: now, the new Bailout Office Office of Financial Stability actually has the luxury of being able to think things through before it does them.

And finally, the Lehman failure, in forcing Morgan Stanley and Goldman Sachs to accept Federal Reserve regulation, has removed a large potential source of systemic risk going forwards. Goldman has been for many years the largest hedge fund in the world, and for all its vaunted risk-management skills, it could in theory have blown up at any time. Now, that eventuality is less likely, and if it does still happen it will be less harmful.

What's quite clear is that regulators around the world have learned a great deal over the past year. Let's hope they don't again find themselves in the position of having to put those lessons to practical use.