Krugman in "I was wrong"

[kpete]

Growth and Interest Rates: I Appear To Be Wrong

In my last post I followed Floyd Norris in criticizing the CBO, which has marked down its estimates of future economic growth without marking down its estimates of future interest rates. I still think that’s a fair criticism. But I also offered a hypothesis: that interest rates fall more than one-for-one with slower growth, so that the crucial difference r-g — interest rate minus growth rate — actually falls, making debt easier, not harder, to handle.

So I’ve taken a quick and dirty look at US history, and it doesn’t seem to bear my hypothesis out. Here’s actual r-g — strictly speaking, interest rates minus the rate of growth of GDP over the previous year — since 1952:




Postwar US history broadly breaks into two eras: a fast-growth generation after World War II, and generally slower growth thereafter. If my hypothesis had been right, r-g should have been lower in the second era than the first. Well, it looks as if the opposite was generally true, even if you ignore the spikes around big recessions.

Now that I think about it, the case of Japan — although complicated by the zero lower bound — also counts in this direction: interest rates have been low, but GDP growth even lower.



http://krugman.blogs.nytimes.com/2014/03/02/growth-and-interest-rates-i-appear-to-be-wrong/

[onpatrol98]

1. But lower growth does appear to make debt harder, not easier, to carry...

I'm not sure what he's apologizing for??? Surely, he never really implied that lower growth would make debt easier to carry. And, if he did, no one would take him seriously...I hope???

[JHB]

2. This part, I think...

You can make the case that US long-term growth prospects have worsened substantially. But it’s hard to make that case without thinking that we will be at least flirting with secular stagnation, which will mean persistently very low interest rates.

One way to think about why this matters this is in terms of the relationship between “r”, the real interest rate, and “g”, the economy’s long-run growth rate. The extent to which public debt is a problem depends a lot of this relationship. If r is close to or even below g, debt is hardly a burden at all; if revenues pay for non-interest outlays, debt as a share of GDP will steadily erode. Only if r>>g should we worry about debt spirals and all that.

So what CBO has in effect done is mark down its estimate of g but not of r. And that’s surely not right. As Floyd says, we should expect lower g to lower r too. In fact, I think there’s good reason to believe that a fall in g will reduce r more than one for one, so that slow projected growth actually reduces the urgency of doing anything about debt. More about that when I have time to get to it.

http://krugman.blogs.nytimes.com/2014/03/01/cbo-mix-and-match/

It's the apology of a reasonable person who pointed out a flaw in someone's reasoning, only to look more closely at it and find out that the flaw isn't as clear as he thought it was.

[onpatrol98]

3. In that case it's admirable...

In that case it's quite admirable and I'm not sure I've seen an economist ever do it. I always got the impression that they were quite comfortable never agreeing.