Roger, The productivity norm cannot solve all problems. If unions drive real wages above their equilibrium value, this may increase structural unemployment. But this would be a problem under any stable money rule.

Follow this…

The only piece I’ve co-authored with Garrison in the JMBC (Nov. 1997) was a critique of the Sumner (JMCB Feb. 1995) scheme to have the Fed create a GDP futures market and then tie its policy to the net positions taken by speculators in that market. Summer (Nov. 1997) replied.

You said:

“So, with the tech bubble and the 1987 stock bubble there was no deflation because the Fed lowered rates and kept on propping the money supply, as you would advocate. The difference between us is that you think this is a good thing; I look a few years down the line and see the disasters that this unleashes, and conclude that it’s a bad thing.”

Where were the disasters from propping up the economy after the 1987 crash? I thought the 1990s one one of the most prosperous decades in world history. Or was that just an illusion too?

The longest inflation I know of was 1955-2008 in the US. But countries like Australia may have had longer ones. In economics 50 years is virtually forever, all long run adjustments occur in well under 50 years.

You said: “But this doesn’t explain anything.”

-Yes it does. You offered, as an argument against Selgin’s productivity norm, Keynesian “sticky wages” and said that there is no other explanation of why deflation causes unemployment. I offered you a coherent explanation that does not need to resort to the fiction of “sticky prices”. Inflationary monetary expansion causes misallocations of capital and labor and an unsustainable boom, which is followed by a deflationary bust that causes unemployment.

You said: “Deflation is by definition always going to be preceded by rising prices or stable prices.”

–No, productivity rises cause deflation, and they don’t have to be proceeded by inflation. Deflationary busts, however, are preceded by inflationary expansion.

You said: “In fact, deflation is not caused by the bursting of bubbles.”

–Again, deflation could be caused by productivity increases (as Selgin explains) or through the bursting of bubbles, unless monetary authorities counteract it, as you suggest. So, with the tech bubble and the 1987 stock bubble there was no deflation because the Fed lowered rates and kept on propping the money supply, as you would advocate. The difference between us is that you think this is a good thing; I look a few years down the line and see the disasters that this unleashes, and conclude that it’s a bad thing.

“Do you think Hayek would have favored the deflationary monetary policies last year, which reduced NGDP? I don’t.”

This is your biggest problem. You continue to shift the discussion to the aftermath of the bubble and refuse to ever consider any evidence about the causes of the bubble. You’ve completely dodged all my questions and points about how this comes about.

George,

You said: “There’s nothing unsustainable about 3 percent NGDP growth, and no reason to suppose that it will lead to hyperinflation. After all, the Fed and many other central banks have maintained average NGDP growth rates several percentage points above 3 percent for many years now, without raising inflation rates much above 2 percent.”

And Scott also said: “You say mild inflation inevitably leads to Zimbabwe. Not true. Check out the 20th century in most developed economies.”

That’s a good point, but in retort, I’d argue that inflation has been very high considering how much productivity has been increasing. And I’d also point out that these low inflation periods were regularly checked with deflationary depressions that slowed down inflation. Further, these have not been very long periods of time, and they are becoming worse with time. They won’t be good examples if massive inflation happens in the next few years.

I have a serious question here on which I’d love to hear both your takes: Historically, what could we see as the longest continuous period of sustainable low inflation that did not lead to high or hyper-inflation?

A caveat: crashes, bubbles and depressions are deflationary corrections of the moderate inflationary booms, and so they check this mild inflation. In that sense, any significant deflationary depressions cannot count as part of a period I’m looking for. Whenever central banks lets deflation take place then this is a stop of continuous inflation. I am looking for the longest period in history which would receive the Scott Sumner seal of approval—with no deflation or contraction of the money supply.

David,

I agree with you entirely. I do not believe the Nobel prize confers any intellectual authority, and I only invoke it when arguing with people who do, to get them to consider the work of someone who’s earned it but is curiously ignored in the mainstream.

And I agree with you on the problems with invoking sacred texts, and hate doing so myself. In my defense, I did it this time because our argument was specifically about what Hayek said, since Scott specifically argued that Hayek had never said something he had said. This invoking of sacred texts is particularly problematic in the case of Hayek since he was a voluminous writer who changed his mind on a few things over the decades. For me, it’s not his authority; it’s the argument itself that matters.

“–I apologize; I overestimated how much you understood Hayek and so thought this was obvious. The deflation is only caused by the bursting of the bubble.”

But this doesn’t explain anything. Deflation is by definition always going to be preceded by rising prices or stable prices. So if you define either rising prices or stable prices as an inflationary bubble, then your theory become tautological. In fact, deflation is not caused by the bursting of bubbles. When the tech bubble burst we did not get deflation, nor did we get it when the 1987 stock bubble burst. Deflation is caused by tight money, not bubbles.

You might be right about the Swedish academy, but Hayek’s current reputation is mostly based on his micro work, which is far more respected than his macro work. Whether this is fair is another issue.

You say mild inflation inevitably leads to Zimbabwe. Not true. Check out the 20th century in most developed
economies.

I do not favor targeting the money supply, so I have no idea what your comment about predicting money refers to. If you are talking about predicting NGDP growth, people certainly do make such predictions all the time, at least implicitly. Most people I know expect roughly 4% pay raises each year, if they are doing an average job. Where do you think they get these predictions from? If we have stable 5% NGDP growth that people will simply assume that what is the “normal” is 4% pay raises each year (assuming population rises at 1%.) If we have 3% NGDP growth, then 2% raises will be regarded as normal. People can and do make those sorts of calculations all the time. There is no reason this can’t work forever, without ending up in Zimbabwe.

Do you think Hayek would have favored the deflationary monetary policies last year, which reduced NGDP? I don’t.

David Beckworth, I see what you are saying. But it is odd that Bernanke fought vigorously against the risk of deflation, and indeed deflation that might have been benign, and is not fighting vigorously against recent deflation caused by falling AD.

One complication is inflationary expectations. Even if the optimal rate of inflation is minus 1% in the steady state. It is not minus one in an economy where inflation expectations are 3%. You need to get there gradually.

Greg, Thanks for pointing that out.

George, Thanks. And unlike me you have read your Hayek.

David Glasner and George, It isn’t quite clear what Hayek is saying there. If he’s saying that stable prices would avoid something like the Great Depression, or even a Great Recession, then I agree. At the same time I agree with George that a stable price level policy can create some malinvestment, and perhaps to some extent this explains the housing boom (although I think part of that was either stupidity or bad regulation). But let’s say George is right. What problem did this cause? Did it cause wasteful investment? Yes, Did it cause subprime bubble to burst and damage banking? Yes. Did it cause some macroeconomic difficulty between mid-2006 and mid-2008? Yes. Did it cause unemployment to rise from 5.5% to 10% in the past 17 months? No. That was the secondary deflation that would have been prevented by a stable price level policy (or more accurately a stable 2% if that’s what the policy rule was prior to the bubble bursting.)

So I am sympathetic with both your arguments. I think George is right that stable prices are inferior to the productivity norm, and that the housing fiasco may partly reflect the problems with stable prices. But I think Hayek’s statement is defensible if viewed as saying the worst sorts of problems could be avoided with a stable price level. Recall that unlike us three Hayek lived through the Depression, so his dividing line between small and big macro problems may be different from ours.

Greg. That might be it. Also recall that inflation had become so bad in the post war period (especially the 70s) that perhaps Hayek thought that stable prices were the best we could hope for. But here is another thought. Friedman’s natural rate hypothesis (which David pointed out was anticipated by Hayek) may have slightly changed Hayek’s views on the real effects of different trend rates of inflation. So Hayek might have still preferred a productivity norm, but I could easily imagine him being OK with Woolsey 3% rule, under the influence of “natural rate” theory that says the trend rates of inflation doesn’t matter for real variables in the long run. Obviously I have no evidence for this, but I don’t think it is implausible. I’ve noticed that most great macroeconomists subtly shift their views as macro conditions change.

George, Yes, and didn’t he and Garrison also have a recent piece on Hayek in the JMCB?