Recently, I’ve seen some discussion to the effect that negative interest rates are in some sense “natural”, not weird. The basic idea is that it was difficult to store wealth throughout most of human history. Thus the return on wealth was generally negative. (JP Koning and Joe Weisenthal provide nice examples.)
I think that’s a perfectly fine argument, as long as people understand that the interest rate being referred to here is the real interest rate. I’m a bit worried, however, that people might confuse that argument with a completely different question—is there something weird or undesirable about the negative nominal interest rates that we see in Europe and Japan?
Back in April 1977, 3-month T-bills yielded about 4.5%, while the inflation rate over the previous 12-months had been running at close to 7%. Indeed when I was young, negative real interest rates were quite common. If you add in the fact that nominal interest is taxable, then the real after-tax rate of return during the 1970s was usually negative, at least for those who held taxable bonds.
So there’s nothing at all unusual about negative real rates. We also saw them in 2003-05. They are quite common. What is unusual is negative nominal interest rates. So are negative nominal rates natural?
It turns out that nominal variables do not have “natural rates”. We get to choose the level of nominal variables when we choose our monetary policy. We can have negative nominal interest rates or positive nominal interest rates; it’s entirely up to us. If Congress doesn’t want negative nominal interest rates, it’s perfectly OK for them to pass a law instructing the Fed not to allow negative interest rates . . . UNDER ONE CONDITION.
Long time readers know where I’m going with this. If Congress wants to instruct the Fed to disallow negative nominal interest rates, they need to know that in doing so they are also instructing the Fed to set an inflation (or NGDP growth) target high enough so that the Wicksellian equilibrium nominal rate never goes negative. Otherwise the Fed can’t achieve its dual mandate. And that’s fine, if Congress wants to go down that road.
In other words, because of the dual mandate, Congress needs to know that legislation banning a Fed policy of negative interest on reserves will be interpreted (I hope and presume) as an instruction for the Fed to use expansionary monetary policies to maintain positive interest rates. The Fed will presumably raise their inflation target, and also do enough QE to hit the higher target. Whatever it takes. After all, Congress would have told them to do that.
You might say that Congress doesn’t understand this. But in any debate over this sort of radical legislation, Fed officials would almost certainly make it very clear to Congress that (in their mind) the implication of this legislation is a slightly higher inflation target, unless Congress provides some other method for achieving the goal, such as authorization for the Fed to buy unlimited amounts of any asset, anywhere in the world, if that were necessary to hit the 2% inflation target.
If you don’t believe me, consider the following. In 1977, Congress gave the Fed a mandate for price stability, high employment and “moderate long-term interest rates”. So Congress has already passed the sort of law I am discussing, but intended as more of a cap on rates than a floor. You might wonder how the Fed interprets their moderate long-term interest rate mandate. Fed officials say that they believe the only way to fulfill the instruction to maintain moderate long-term interest rates is by keeping inflation low. Just as I claim they’d interpret a no negative interest rate law, except in the opposite direction. Most of the time, the Fed does not even directly target long-term rates. Rather, they target inflation, and assume that doing so will keep long-term rates “moderate”, as Congress instructed them to do.
With a law banning negative nominal interest rates, Congress would be instructing the Fed to keep interest rates within a sort of corridor. But this would be nothing like the corridor associated with short-term interest rate targeting. This would be a long-term interest rate corridor. Control would be done via the Fisher effect, not the liquidity effect. (Steve Williamson as Fed chair?)
And the only way to keep rates within that corridor would be to select an inflation target high enough to keep nominal interest rates above zero, but low enough so that nominal long-term rates never rose above “moderate”. To do that, the Fed would probably have to switch to level targeting (of prices and NGDP) and perhaps slightly raise the inflation target.
PS. As always, I buried the lede. It seems to me that this post is a powerful argument for the Fed asking Congress whether negative rates are OK. Whichever way Congress answers, the Fed immediately has much more power, more “ammo”. They either get to use negative IOR, or get to do unlimited QE, or they get to set a policy target path high enough to avoid the zero bound and still hit the dual mandate.
Because Congress told them to.