Rising Long-Term Rates Are Usually A Bullish Indicator

by: Scott Sumner

Arnold Kling recently made the following claim:

I would add myself to the list of economists who have some ‘splainin’ to do. I am always willing to be counted among those who doubt the Fed’s power over interest rates, especially long-term real rates. By the way, Scott Sumner used to say that a rise in long-term interest rates could be a bullish indicator. Would he say that now? UPDATE: No.

The correlation between nominal interest rates and NGDP (or RGDP) growth is overwhelming positive. Nothing that happened this week contradicts that. Indeed nothing that happened this week contradicts what I’ve been saying for the past 4 1/2 years:

1. Never reason from a price change.

2. Interest rates are a lousy indicator of the stance of monetary policy, because they reflect both liquidity effects and longer term effects.

I’ve made these points dozens of times.

As far as “bullish indicator,” suppose you were allowed one quick look into a crystal ball, at the level of interest rates in 2016. You are told ahead of time that the yield on Treasuries will be either 2% or 5% in 2016. If you are Paul Krugman, and are rooting for a strong recovery, which is the number you hope to see?

I hope I don’t even need to answer that question. And nothing that has happened this week in any way changes the answer.

So what have we learned this week? To me this is one data point, indicating tight money is slightly more likely to raise long term rates than I had previously assumed. But there is still the enormous stock of previous long term rate changes in response to previous Fed moves. Previous data points. January 2001, September 2007, December 2007, etc., etc. Should we suddenly throw all those observations away?

PS. The fact that Krugman, Cowen, Kling, myself and many others were surprised by the surge in rates over the past few weeks actually speaks well of our understanding of macro. It shows that we’ve been paying attention, that rates don’t usually behave this way. As for those who “got it right,” you have some “splaining to do” about the surging yields in Japan in response to more QE.

The best solution is to remove interest rates from macro, and focus on the three variables that matter; NGDP, nominal wages, and hours worked.

PS. Unlike Kling, I think the central bank has enormous influence over long term rates, but mostly via the income and inflation effects.