I should begin by saying that I like the idea of being a financial romantic. But the real reason Paul Krugman is wrong about this hugely important question is that he is confusing (uncharacteristically) issues of liquidity and solvency. He invokes Bagehot in support of a clear-eyed defense of bailing out the financial sector in its times of distress, but Bagehot was observing the need for a lender of last resort.
Lending and bailing are two entirely different animals.
We should expect the Fed to provide essentially unlimited liquidity if there is a run on the banking system—if the system is unable to cover the mismatch between its long-term assets and short-term liabilities. As long as financial institutions have positive net worth at realistic asset prices, this will be nearly costless for the taxpayer and priceless for the economy.
But bailouts are something altogether different. The financial system cries to be bailed out if it faces a solvency crisis, if the value of its assets are plunging and the last shreds of equity are at risk. Examples of this kind of rescue include the injection of vast sums to buy up dubious mortgage-backed securities at face rather than market prices, assuming the liabilities of failing speculators like AIG toward their speculator counterparties, and so on. The US government shouldn’t have done this the first time around, and they damned sure shouldn’t do it a second. I would not be upset to see our legislators sign a blood oath against all manner of bailing.
Yes, a solvency crisis nearly always triggers a liquidity crisis, but you can patch the second without trying to reverse the first.
Another objection arises: how can you let banks fail if they are too big to fail? Very generally, there are two alternatives: you can temporarily take ownership of them—wiping out the equity of their previous owners and sacking the management—while you use public money to resolve their net liabilities, or you simply let them fail while using public money to buy up their assets at liquidation prices to jumpstart a system of public banking. I’d go for the second, but either is feasible and way, way preferable to bailouts.
You can't cleanly separate solvency and liquidity crises. When disaster strikes we are confronted with the possibility of multiple equilibria - 'solvency' can be determined wholly by animal spirits or by liquidity shortfalls, so in times of crisis solvency is a nominal phenomenon rather than a real one.
ReplyDeleteIn those terms I reject the assumption that 'solvency' is a clear-cut case which demands resolution. In crisis banking it is no such thing, and so every case must be evaluated on its individual merits. Small banks may be closed and covered br FDIC, but institutions with large counterparty exposure must be kept afloat, or we risk throwing the economy down into that deep dark hole of the deflationary equilibrium.
Ben,
ReplyDeleteIf you are worried about counterparty domino effects, you can opt for the seize-and-clean option, which does not (intrinsically) entail bailouts. I am less concerned because I am happy to move still-valuable assets quickly over the public ledger, and it's the assets, not the institutions or particular investors, that need to be safeguarded. For more on this second option, look up the "good new bank" idea.
Daniel Davies would point out that the distinction between a solvency crisis and a liquidity crisis is that in a liquidity crisis, the State or the Central Bank has to finance all the assets. Not just the bad ones.
ReplyDeleteWhen the financial sector owns the legal apparatus then you get what we've gotten and will continue to get. The only way to fix the problem is with better fundamental education in the FREE system and with better (more closely coupled) representation of the common people in their government. A four fold increase in the membership of the House would go a long way in improving legislation such that it creates laws more favorable to the populous. Is that not what the 99% actually want?
ReplyDelete