The Money View

What a liquidity crisis looks like

Bloomberg's reporters continue their diligent work looking back on the Fed's lending in the subprime crisis. Matt YglesiasYves SmithPaul Krugman, and others have picked up the story. Meanwhile, the world looks ahead to the next development in the eurozone crisis. To read these crises correctly, liquidity should be front and center. It is missing in Bloomberg's work, and it is missing in European policymaking.

A bank, or any firm, or indeed any economic entity, must constantly evade the survival constraint (a term from Minsky). When it can't pay its bills, its creditors can resort to the legal system and shut it down. This is the moment of illiquidity—the inability to come up with the cash to meet obligations.

The payment system has facilities for ensuring that the failure of payment by one party does not lead to the failure of payments by other parties. The private money market allows firms to postpone this day of reckoning, at a price. The central bank, at the center of the payment system (in the US as in Europe), ensures the availability of bank reserves so that one day's postponement can be obtained at its target price. In an everyday, non-systemic liquidity crisis, a single payment fails, and the system flexes just enough so that this failure does not spread. Little crises like this happen all the time, and usually the system contains them.

A liquidity crisis becomes systemic when many payments fail, enough that large parts of the system are squeezed up against the survival constraint. Mortgage defaults come in higher than expected, so payments to holders of mortgage-backed securities come in lower than expected, so guarantor banks have to come up with cash quickly.

In both cases, solvency is a concern at some level. But what finally gets the lights shut off is that the electric bill doesn't get paid—illiquidity, that is, whether or not the household, or bank, or country is solvent. Our understanding of the central bank's role as lender of last resort is that, when the risk is that a large part of the financial system will go dark, so to speak, it is better to bite the bullet, provide liquidity, and keep the lights on. This is from Bagehot, of course, but also Minsky, Kindleberger, and Hawtrey.

What is missing from Bloomberg's extensive report is an understanding of the liquidity dimension of the crisis of 2008. Bloomberg argues, for example, that banks earned $13B of profit on these loans, a figure based on net interest margin, the difference between the interest earned on a bank's assets and the interest paid on its liabilities, a claim that has been uncritically accepted by many in the blogosphere.

The figure assumes that the funds were just routine deposits, that were turned around and lent out to routine borrowers. But this view makes the entire crisis look like one of solvency, when it was evidently one of liquidity—Bear and Lehman were gone, AIG was on life support, and no one would extend credit beyond overnight for fear that the money would not come back. What if the Fed's lending had not been there? Deleveraging would have been the likely result—asset sales into a falling market to raise cash. Below-market interest rates may not have been too high a price to pay to induce banks not to contract their balance sheets at the height of a crash.

As I have argued before, the Fed made far graver failures than can be seen in Bloomberg's documents. The Fed failed to adapt to a changing financial system, it failed to protect the payment system from excessive risk-taking, and it failed to react quickly enough as the crisis erupted. The liquidity shortage was of a different kind than the Fed was prepared to meet.

Today in Europe, policymakers dwell on austerity measures and the idea of a fiscal union, assuming that if these solvency concerns are resolved, the liquidity problems will solve themselves. But illiquidity is what will take down major European financial institutions first, solvent or not. Before too much longer, let us fully absorb the liquidity lessons of the last crisis.