Back in the 1970s, there was much pessimism about whether the Fed could get inflation under control by raising interest rates. Rates kept going higher and higher, and yet inflation also kept increasing. Markets did react to unexpected rate increases as if they were disinflationary, but the longer run pattern (high rates and higher inflation) led people to be very pessimistic about the ability of rate increases to tame inflation.
Today there is much pessimism about whether lower interest rates can boost inflation. Markets do react to unexpected rate cuts as if they are expansionary, but the longer run pattern (low rates and low inflation) leads people to be very pessimistic about the ability of rate cuts to boost inflation.
If I went back in time to 1979 and tried to make the case for rate increases, people would have said; "But the Fed has already raised rates to 12% and inflation keeps soaring. What makes you think more of the same failed policy would succeed?" If I said, "then raise them to 20%" people would have just rolled their eyes, like I wasn't being serious. Well, soon after Paul Volcker did raise them to 20%, and inflation fell from double digits to about 4% after 1982.
There's a lesson here for today. It's now fashionable to complain that negative rates aren't doing any good; after all, the markets continue to show increased pessimism about growth and inflation, "despite" the negative rates. But just as the 12% interest rates were not "tight money" in 1979, the negative rates in Europe and Japan are not easy money. On the other hand, still lower rates of IOR would probably make policy more expansionary.
Having said that, I do agree with Tyler Cowen's recent claim that there are more effective ways of doing monetary stimulus than negative IOR. But I'm not as pessimistic as he is about the effects of negative IOR:
. They are not just a reflection of bad conditions, the tax on intermediation seems positively harmful, and there are ways to run an expansionary monetary policy which don't involve this tax.
I don't see much evidence that negative IOR is not "working" in the article he links to, although I do agree that low rates in general are probably a negative for banks, and indeed for the entire economy.
Given that I've used the "tax on reserves" metaphor for negative IOR, you may be surprised to learn that I don't quite accept the claim that negative IOR is a tax on intermediation. A tax creates a wedge between the equilibrium free market price, and the actual price inclusive of the tax. But we don't see that with negative IOR. Indeed the rate of interest charged to banks on excess reserves is generally lower (I mean less negative) than the rate on risk-free government bonds. So in a sense there is still a subsidy involved.
Recall that in the old days when IOR was zero and T-bill yields were positive, the spread was considered an implicit tax on required reserves. I think that's right. But by that logic we now have a subsidy. I suppose you could argue that the entire structure of negative rates on government securities is artificial, and hence in some sense a "tax." But if you really want to go down that road, you are forced to view positive interest rates on government bonds as a subsidy. After all, a subsidy is just a negative tax.
When I argued that negative IOR was a tax on bank reserves, I meant that it drove a wedge between the rate of interest on cash (zero) and reserves, which are the two forms that base money can take. But if we think about bank balance sheets, they hold reserves and bonds. And if the yields on bonds are lower than on reserves, then I fail to see how negative IOR is the problem. After all, banks could choose to hold only a tiny amount of excess reserves, as they did in America before 2008. Negative IOR often doesn't even apply to required reserves. The fact that Japanese and European banks choose to hold massive quantities of reserves is presumably due to the fact that they pay an above market rate, a rate higher than government bonds. So in that sense there is no tax "wedge" on intermediation. Banks are not being punished for holding reserves; they are being rewarded, relative to market rates.
The real problem at banks is the negative rates on bonds, and those are equilibrium market rates, indeed they are still higher than the Wicksellian equilibrium rate. In theory, banks ought to be able to offset this with even more sharply negative rates on deposits. But for some reason they are not able to do this very well, at least so far. Perhaps this is due to the fear that depositors will switch to cash in safety deposit boxes, or perhaps just due to the public relations hit they would take. In either case, banks tend to avoid negative interest on deposits. So if there is any sort of (informal) zero lower bound on deposit rates, then banks really are hurt by negative interest rates. But not necessarily negative IOR.
So I sympathize with Tyler's concern about negative interest rates in the broader sense. Driving them even further negative won't solve the problem facing banks, unless the policy is able to ultimately lead to much higher nominal rates. So far central banks have been too reactive, they need to get out ahead of the curve. They won't have "done enough" until further monetary stimulus leads to higher bond yields, not lower.
Volcker's high interest rate policy eventually led to much lower rates, once inflation came down. The developed world's central banks need to be aggressive enough to actually raise inflation (and even better NGDP growth) significantly. That's the only durable solution to the low interest rate problem currently facing banks. Anything else will just be a band-aid.
I know my claim seems implausible, given everything going on right now. But just think how implausible I would have seemed in 1979, claiming 12% rates were actually easy money, and that central banks needed to do still more rate increases, much more.
And finally, it's worth noting that Volcker's success was not just caused by sharply higher interest rates, he also changed the Fed's approach to monetary policy, shifting (at least the rhetoric) from interest rates to the money supply, as the indicator of monetary policy. That is, he adopted the monetarist framework (until inflation was under control.) Of course the market monetarist framework would replace the money supply with NGDP growth expectations. That sort of shift would likely be far more effective than driving the IOR ever deeper into negative territory.