Asking Scott Sumner for Clarification
As I repeatedly say, I think Scott Sumner is a very sharp guy who understands technical economics models. (For some of the technical papers I’m working on here at Texas Tech, Sumner may be one of the few people I personally know who can give me feedback on the drafts.) But his latest EconLog post really underscored something that has been concerning me over the years, and so I left this comment (which I hope he sees and answers, if he has time):
Scott,
I would like to request that when you get a chance, you devote an EconLog post to avoiding potential confusion among its readers. Here’s my concern:
==> From a 30,000-ft view, I think a lot of free market economists in 2005 would have endorsed the following: “The Keynesian economists thought the proper policy in a recession was to print money and generate (price) inflation, in order to bring down unemployment. But Friedman and others pointed out that this was quite shortsighted. Once unions and other workers adjusted to the new policy, the so-called Phillips Curve would simply shift. Then we’d have the worst of both worlds, of high unemployment and inflation. That’s why the central bank shouldn’t try to solve a recession by the printing press, but should instead focus on monetary policy rules that provide a stable framework that, in the long run, maximizes real GDP growth and minimizes volatility.”
==> So I know you know all of this stuff, and can get much more nuanced, but I’m concerned the average EconLog reader will take away the lesson from your typical posts here that: “For the last 8 years we have had tight money and that’s why the economy has been bad. The way to fix this recession is to inflate. The way to fix the next recession is to inflate. If unemployment is too high, you need more inflation to fix it.”
Again, I realize that’s not literally what you are saying, but at times it sounds like that’s the take-away message. So if you could at some talk about this, I think that would be helpful.
“The Keynesian economists thought the proper policy in a recession was to print money and generate (price) inflation, in order to bring down unemployment. But Friedman and others pointed out that this was quite shortsighted. Once unions and other workers adjusted to the new policy, the so-called Phillips Curve would simply shift”
I think this is a bit mis-leading. What Friedman was objecting to was the idea that level of employment even outside of a recession could be increased by generating inflation. Friedman (I think) also subscribed to the view held by Sumner that correct monetary policy could avoid recessions- but this has little to do with generating inflation and much more to do with creating the right monetary conditions for the economy to stay in equilibrium.
So Friedman rejected the idea that the Central Bank should not let policy mistake bygones be bygones?
I’m not sure what his views on that would have been. As he favored a fixed money supply growth rule and didn’t seem to worry much about variations in velocity – he may not have had a strong opinion on it.
That’s what I thought, and that’s what makes Market Monetarists seem so much more like advocates of Keynesian discretionary policy.
I don’t know if I’m average, below average or way below average but, here’s the impression I get from reading Sumner. Economic policy can reduce the hardships from recessions by stabilizing the path of the economy – smoothing out the booms and busts. The most effective tool is monetary policy. Stabilizing the path of the economy means stabilizing something: employment, price level, money supply, nominal GDP or something. Keynesians said we should stabilize the path of the price level and that has the added benefit that you can reduce unemployment if you pick a high enough rate of increase for the path of the price level.
Friedman said no, the relationship between unemployment and inflation is more complicated than a simple Philips curve . You need to consider the natural rate of unemployment which can shift over time so, inflation is a bad indicator. It’s much better to stabilize the path of the money supply. Sumner says stabilizing the path of the money supply would be great but the usual measures of money supply growth: price inflation or interest rates are bad indicators. You should really look at NGDP growth, actually the market expectations of NGDP growth.
So, at the end of the day is what Sumner really want’s to do is just print money until you get inflation and the economy improves? (or technically, print money until the market expects inflation and the economy improves?) I dunno. Maybe so. BUT I’ll say three things.
1) The idea that some recessions are caused by a crash in the money supply is very compelling for me. My parents, aunts and uncles would talk about their young adult years during the Great Depression. When I would ask what was the Great Depression the answer was ” The stock market crashed and after that everyone was out of work.” My high school history text said the same thing. It never made any sense. What was the cause and effect? I wish someone at the time would have pointed me to the “Monetary History of the United States”
2) Will NGDP futures targeting work better than past inflation targeting etc. Maybe. and engineer will tell you that the feedback system will work well or go unstable depending on the amount of lag or lead there is in the feedback loop. NGDP futures is an attempt to get some forward looking lead into the feedback loop. Could be a big difference from a backward looking inflation measure.
3) Saying that Sumner wants to print money to get inflation and that will improve the economy is a mischaracterization similar to saying Hayek wants the recession to just run it’s course – take your lumps – things will eventually get better. Both are wrong. Both men want feedback mechanisms to stabilize the economy. For Sumner the feedback mechanism uses market expectations (NGDP futures) and adjustment of the money supply. For Hayek the feedback is market expectations (entrepreneurial profits) and redirection of investment dollars. The problem with Hayek is that it is so much harder to redirect investment dollars if the money supply has already collapsed.
Sumner will always be a perma-inflationist as long as NGDP is growing less than 4.5071648493535638283737% annualized.
“The way to fix this recession is to inflate. The way to fix the next recession is to inflate. If unemployment is too high, you need more inflation to fix it.”
Dr. Murphy, I assume you mean price inflation in this context.
The way to fix a recession is to go back to the NGDP growth path, if and only if, the recession was caused by a drop in NGDP or so called Agg. Demand (i.e. Great Depression and Great Recession).
For example, if Real GDP would always increase faster than Nominal GDP, by some productivity shock that keeps it that way for ever, we would always have deflation and rgdp growth under NGDP targeting.
Increased Inflation rates would only occur during supply side negative shocks.
Keynesians are wrong, among many reasons, because they always assume Agg. Supply is fixed, which is really the inverse of the Short Run Phillips Curve. So every recession must always be accompanied by deflation, negative output rate, and an increased unemployment rate. So recessions in their world view are always be about lack of spending. That is crazy.
So, MMs propose to keep NGDP on target, so if the rate inflation increases, it means rgdp is not growing like before. If the inflation rate falls, rgdp is growing more than before. The inflation rate and rgdp would have a perfect correlation of -1.
Fix NGDP rate = rgdp rate +inflation rate, with rgdp rate and inflation rate having a correlation coefficient = -1.
So, a better descriptor of MM is: “If unemployment is too high [and NGDP has fallen below trend], you need [to increase NGDP through monetary policy, and not government spending like Keynesians want] to fix it.”
And: “If unemployment is too high [because of a negative supply shock, just keep NGDP on target, so the market (real factors) will adjust] to fix it.”
Your comments RE: correlation are not true as regards Level Targeting.
The target level is designed by percent change. So yes, the correlation between %rgdp and %inf.rate is -1 under level targeting.
Level targeting just means that if one year the NDGP rate is below or above. Next yeat rates had to be sufficent to catch up. Then that relationship comes back to normal, r =-1.
No it is not, you’re conflating level targeting with growth rate targeting, they’re not the same thing.
“The way to fix a recession is to go back to the NGDP growth path, if and only if, the recession was caused by a drop in NGDP or so called Agg. Demand …”
It’s not a recession when people voluntarily reduce their demand – voluntary reduced demand is not a mistake and not a market failure.
It’s a recession when you’re losing money on investments in projects the profitability of which was not assessed on the basis of consumer demand, but rather on “liquidity” (stolen purchasing power) injected into the higher-order stages of production (e.g. capital sector) ex nihilo.
The problem can not ever, logically, be lack of consumer demand because you’re not supposed to be producing what consumers don’t want.
If consumers all of a sudden decide they don’t want your, or any number of, goods, the producers are the ones that are out of touch with consumer demand.
The point of trade is NOT to create employment but to satisfy consumer demand; Consumers’ wealth does not belong to the producer or the worker.
A belief in the concept of aggregate demand is functionally equivalent to a belief that consumers owe producers and laborers (and investors) a portion of their wealth so that trade helps everyone.
But if I’m the one whose wealth is being stolen/redistributed, then I’m not really a part of the “everyone”, and so there’s no incentive for me, personally, to agree to a system that does this. It’s robbery by one portion of the population of another portion. That’s not sharing and that’s not a collective in the sense its proponents would need it to be.
“t’s not a recession when people voluntarily reduce their demand – voluntary reduced demand is not a mistake and not a market failure.”
Sure, Agg Dem reduction is not the definition of a recession. Agg. Demand drop can cause production to fall and unemployment to rise, which is the actual definition of a recession, not the drop in nominal spending.
“The problem can not ever, logically, be lack of consumer demand because you’re not supposed to be producing what consumers don’t want.:
Sure, NGDP Targeting is de facto stabilizing nominal and rrwal investmwnt, not consumwr soending. The component of spending that meaningfully falls in resessions is investment. That why Keyes in as are dead wrong because they think consumer spending can ipso facto substitute spending on investment.
We just don’t say ‘investment targeting’ because if consumers change their preferences, NGDP targeting respects that preference proportion. On more reason I want small, like % to 2 %, targeting NGDP because 5% can more easily disrupt that pattern.
Free banking would do that.
Dr. Murphy,
it’s all about the expectations about the future path of policy and whether there are money supply shocks.