Can we please stop talking about the zero lower bound?
Krugman today insists that we do, in fact, face a problem of inadequate demand. And he’s right! But he glosses this as an “excess supply of savings even at a zero interest rate,” which isn’t right at all.
Let’s be clear: There is not “an” interest rate, certainly not a zero one. There are various interest rates, and the ones that are relevant to saving and investment remain high. The BAA corporate bond rate (the red line in the figure below) is currently at 5.7 percent — pretty much exactly where it was in the first half of 2005. And given that inflation is substantially lower than it was five years ago, that particular real interest rate is not only not zero, it’s gone up.
The real question is, can reducing the federal funds rate reduce the economically important interest rates? Now, obviously the answer is No if the fed funds rate (the blue line in the graph) is as low as it can go; in this sense the ZLB is real. But the answer can also be No when the fed funds rate is well above zero, if there’s no reliable link between the overnight Treasury rate and the rates businesses borrow at; and that seems to have been the case since sometime in the ’90s. As the figure shows, the Fed’s recent rate reductions didn’t reduce bond rates at all, even before the Fed Funds rate hit zero; and all the hikes earlier in the decade didn’t raise bond rates either. You’d see a similar picture if you looked at any other economically relevant interest rate. In general, as my friend Hasan Comert shows in his just-defended dissertation, the Fed lost control of the important interest rates some time ago. So the best thing you can say for the zero lower bound, is that arriving there has dramatized a truth that should have been evident for some time already.
Just to spell out the argument a bit more, the point is that a central bank able to control interest rates and/or the "money supply" requires a particular set of institutional and regulatory arrangements. In the case of the US it was (a) statutory reserve requirements; (b) the lack of a secondary market for many commercial bank assets, making the holding of reserves necessary for liquidity purposes; (c) the lack of high-return short-term investments for commercial banks, so less pressure to mobilize reserves; (d) the absence of competing institutions not subject to reserve requirements. The latter three factors made the statutory requirement effective, by reducing incentives to evade it; they were all connected with the partitioning of the financial system into differentiated institutions serving different functions and markets.
Without those conditions, central banks can't do monetary policy as conventionally understood. But that doesn't mean they're unimportant! It just means we're back in the world of the Bank of England under the 1844 Bank Act, when the lender-of-last resort function was the sole or at least main purpose of central banks.