CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press In Economic Theory Folklore Inflation Is Caused by Low Unemployment, Not Fast Growth

In Economic Theory Folklore Inflation Is Caused by Low Unemployment, Not Fast Growth

Thursday, 15 August 2013 07:18

Neil Irwin has a discussion of the growth potential of the U.S. economy that follows the work of two JP Morgan economists. The basic story is quite pessimistic, arguing that we will see rapid declines in labor force participation and much slower productivity growth in the future. I won't comment on these points at length here (the evidence presented is limited in the piece and weak), but will rather focus on the conclusion.

The piece ends by warning readers:

"And if the analysis is right, and we have downshifted to a slower pace of potential growth, we will hit the speed limit of this recovery — the point at which inflation becomes a risk — sooner than forecasters are commonly thinking."

The standard theory about the causes of inflation is based on the unemployment rates. Folks have probably heard of the non-accelerating inflation rate of unemployment (NAIRU). The logic of this theory is that at lower rates of unemployment there is more upward pressure on wages and prices, meaning that the inflation rate increases. At higher rates of unemployment there is less upward pressure on wages and prices, which means the rate of inflation decreases. 

The NAIRU is that wonderful level of unemployment at which the inflation rate will stay constant. For true believers, it is the unemployment rate that the Fed would target since it is the lowest unemployment consistent with a stable rate of inflation.

There are lots of good criticisms of this view, but some permutation of it dominates the vast majority of mainstream thinking about the economy. Note that it does not say anything about growth. The non-accelerating inflation rate of unemployment is a story about unemployment, as the name implies.

Growth can enter into the picture if the unemployment rate is actually at the NAIRU. In this situation, if the economy grows at a rate faster than its potential (it can do this for short periods of time), then the unemployment rate would fall below the NAIRU and we would then see accelerating inflation. But the key variable in this picture is still unemployment, not growth.

This is important for two reasons. First, there is no limit on the economy's ability to acheive rapid growth as long as we have an unemployment rate above the NAIRU. Even if the rate of growth of potential GDP is just 2.0 percent, the economy can still have a spurt of 6 percent, 7 percent, or even 8 percent, as long as there is large amounts of slack in the economy, as is the case today.

The other point is that unemployment is something that people see and experience. Growth is not. The key economic debate is how low the unemployment rate can go -- how many people can have jobs -- not how fast the economy can grow. We will find that out when we get the unemployment rate as low as possible without serious problems with inflation.

Comments (16)Add Comment
Wage Inflation
written by jonny bakho, August 15, 2013 8:57
The inflation everyone worries about is the wage-price spiral. A domestic (because we are really talking about the US) wage price spiral requires that labor be in a strong position to increase wages in response to price increases. How strong is domestic labor in an economy with 7 percent unemployment and a global labor market full of cheap labor? On top of that, a GOP Congress is unlikely to raise the minimum wage. WTF is upward pressure on wages supposed to come? Wage inflation requires an increase in demand for labor which can only come from an increase in demand for goods and services. QE has not greatly increased demand for goods and services, so it has not created wage inflation.

The models that many are using of how inflation is produced in an economy are deeply flawed. Friedman pinned inflation on monetary policy. But Friedman always assumed full employment. That assumption is violated in this recession.
written by Paul Mathis, August 15, 2013 9:34
Right now, core inflation is less than 2%. Why can't inflation accelerate to say 4% without any deleterious effects on the overall economy? Obviously it can and probably should. Therefore, the fundamental premise of NAIRU - non-accelerating inflation - is nuts.

But the larger issue is whether inflation is necessarily determined by the unemployment rate. Obviously, there are many historical examples to the contrary and there has never been any definitive proof that reducing the unemployment rate necessarily causes inflation to accelerate. Conservative economists again have simply assumed such causation.

Bottom line: the U.S. has not had annual inflation over 4% in more than 20 years despite wide variations in the unemployment rates over that time. NAIRU is worthless.
Why Media Fixates on Labor Shortages that Threaten Growth Rather than Unemployment Theatened by Labor Surpluses
written by Last Mover, August 15, 2013 9:51
The key economic debate is how low the unemployment rate can go -- how many people can have jobs -- not how fast the economy can grow.

Most media are drunks under the streetlight looking for the car keys because the light is better there. They frame most economics as too much or too little of something out of context in order to ask the wrong question to get the answer they want to report.

Therefore any well known factor input like labor is immediately pigeonholed in terms of its seemingly positive correlation with growth. Even a dimwit pundit knows more growth requires more labor in one way or another.

From this foundation no more complicated than a child understanding the correlation between more light and a better ability to see, comes the usual nonsense about how growth of the developed world is threatened by running out of labor via population decline combined with increasing ratios of non-working old to working young.

Real economists pile on with explanations that growth is being stymied from the supply side evidenced by fewer joining the labor force, lured away with socialist programs that reduce incentives to work.

So what happens to the direct contradiction of labor surplus of too many workers looking for work who can't find it in this relentless saga of fearmongering that there's not enough workers at the same time?

It's conveniently swept away as easily as a child turns off the light switch to go to sleep. Any dimwit can see even a child controls light from the supply side, because if light came and went from the demand side, that means children can't grow up to be rugged individualists who gather round streetlights to explain it's simply not possible for an economy to come to Keynesian equilibrium below full employment on its own, as a fundamental failure of free market capitalism at the macro level.

Because if it did you see, that puts the lie to the claim that labor itself is the constraint on growth because there's not enough of it, when in fact the constraint is not enough demand to employ that labor which would actually increase growth more.

Explanation? Keynesianism is killing America with dangerous excess debt, specifically debt that creates phony jobs that can never last, can never add value, can never match supply and demand with the surgical precision of working markets that clear regularly.

Wake up America. Find a streetlight meeting near you and get educated on the difference between growth constrained and enabled from the supply side and growth never constrained or enabled from the demand side. Say's law. Supply always creates demand, never vice versa.

Or just read mainstream media instead.
Last Mover - You Are Kidding, Right?
written by Paul Mathis, August 15, 2013 10:31
"Keynesianism is killing America with dangerous excess debt, specifically debt that creates phony jobs that can never last"

Excess debt? We print our own money and can therefore pay any debt at any time.

Phony jobs? You mean like building the Interstate Highway System or Hoover Dam?

Get serious.
He's totally kidding, Paul
written by Bummer, August 15, 2013 12:36
There's a guy named Swift who wants us to eat babies. You should probably go yell at him too.
Last Mover rules
written by pjm, August 15, 2013 1:17
But his humor is not always appreciated.
great idea, in just a few words, pls use and reuse and repeat
written by Benjamin Kupersmit, August 15, 2013 1:32
"The key economic debate is how low the unemployment rate can go -- how many people can have jobs -- not how fast the economy can grow."
Someday this war's gonna end...
written by Citizen AllenM, August 15, 2013 5:37
Ah, but NAIRU needs the spice of the older tech- namely a discussion of the Reserve Army of Labor. Until a significant amount of the long term unemployed are successfully employed, concentrating on the short term unemployment rate to determine any version of NAIRU is simply foolish, because of the tremendous hidden slack in the workforce.

Now, put the blather above in that context, and inflation will be subdued for a long time, baby, a very long time without any fiscal stimulus.
Mr. Inflation is by far Milton Friedman...what did he say?
written by pete, August 15, 2013 6:56
Inflation is a monetary phenomenon. No matter what the unemployment rate is, mo money mo money mo money means some inflation. So yes, holding money constant, there is a relation between unemployment (the inverse of growth) and inflation. But with the monetary base more than tripling, hard to see how this won't lead to an expansion in M2.

Regarding building superhighways, a horrible subsidy for oil and car industries. A very bad idea. Anyway, looking at the 30s, wages were incredibly low, allowing a lot to be done. Workers lived in camps and worked under conditions which would not be allowed today. Sure Skyline Drive in VA is gorgeous, but is that the most efficient use of labor? How about simply repaving streets? Suspend Davis Bacon when the unemployment rate is over 6% and spend some money getting people to work.
written by liberal, August 15, 2013 8:11
pete wrote,
But with the monetary base more than tripling, hard to see how this won't lead to an expansion in M2.

Because money doesn't work that way. The best model of money these days is "monetary realism," which is somewhat akin to MMT but without the claptrap that exaggerates the role of the state.
Re: NAIRU is nuts
written by A Populist, August 15, 2013 8:35

Re: "Why can't inflation accelerate to say 4% without any deleterious effects on the overall economy?"

I am certainly no "inflation-phobe". And, inflation can have some stimulating effects, such as reducing the real value of debts. However, it is not that simple to say that "OK, prices are going up by 4% per year, so investment and consumption will increase, and we will start growing. First off, with high unemployment, wages are not going up in tandem with any rise in prices, so, in terms of actual real consumption, a huge percent of the population (low income workers with stagnant wages) may actually consume *less* as prices rise. And for all those workers, their debt is no more sustainable, if their nominal wages stay the same. Still - I could see how there is some net growth benefit, if enough people's real wages keep up.

I would agree that NAIRU is not operable right now. I believe Dean is pointing out, that since we are so far from NAIRU, that it is silly to worry about run away inflation - which makes perfect sense.

I suppose the fact that housing prices have apparently hit a bottom, and speculators are now bidding (and buying) them up is a good thing, but I don't see that as a sustainable option for reducing U6. Once home prices stabilize, we will still end up with a chronic shortfall of demand (as evidenced by persistently high U6 Unemployment), so long as we have low wages for a majority of the population.

Stimulus will help, but it seems to me, that any stimulus must be large enough to get unemployment low enough to drive wages up, so that when the stimulus is removed, there is more residual demand, and/or more savings in people's pockets to work less.

The other option, is a higher minimum wage, which seems like a more politically plausible option than some super-huge, long lasting stimulus.

And, regarding the likelihood that we will get and sustain inflation without rising wages, I say it is not likely. Low wages limit consumption, and if prices rise too much without rising wages, we could at some point generate some serious political problems - which hopefully will drive the right policy. But with divisiveness and rampant misinformation, who knows?
written by urban legend, August 16, 2013 1:44
We had something like full employment in the early 1950s and the late 1990s into 2000. In neither time was there a big increase in inflation.

With 8% of the private sector unionized, even with some sorely needed increase in wages due to lower unemployment, who here thinks labor could be a cause of a wage-price spiral? It will be many years before the thorough intimidation of the workforce wears off and unions feel comfortable to present "demands" again.

If there were real competition, it would mean little of any wage increase could be passed on in higher prices. Of course, there is probably more oligopoly or outright monopoly today than truly competitive markets, so that is less of a constraint on collective price increases than it might be. That also means, however, a high level of monopsony vis-a-vis buying labor services, which allows more surreptitious collusion, legal or otherwise, in keeping wages low.

There are tons of theories to go around. Experience tells us the NAIRU is a very weak concept at best. The last thing we need in an extensive period of weak demand is chicken little NAIRUism at the first sign of low unemployment with some some healthy and desperately needed wage increases.
I am willing to risk another drubbing
written by Lrellok, August 16, 2013 2:39
SO i have been working on a lot of fringe economic theories. Prof Baker did not like the one i posted before, but i try again.
In order for markets to price efficiently, actors must be free to start and stop buying and selling as prices change. If actors are forced to continue buying or selling the same volume, markets cannot adjust quantity to price, and prices are distorted. In labor markets, however, all able bodied persons are expected to sell 40 hours of work each week, without regard for the changes in wage rates, due the implicit threat of starvation (note the recent comments by republicans on food stamps, IE the ability to get food without having to sell labor).

This would have to mean (in my view) that wages are inherently miss priced, and likely always have been. But if markets are systemically oversupplied in labor, then what does this do to NAIRU? First, there is the possibility that NAIRU is functionally "Post hoc, ergo propter hoc" the logical fallacy of it follows there for it was caused by, and inflation is an outcome of growth, not employment. This is supported by labor force data during the 1970's, when during stagflation labor force surged, so much so that even though the economy was creating a ton of jobs, unemployment rose because labor force increased faster then employment could keep up.

Again though, we get back to the issue of wages. As wages increase, demand for consumer goods should increase. If wages are artificially suppressed, then prices for consumer goods should be artificially low, and distorted relative to each other. If wages where to return to equilibrium, then prices would increase and re-stabilize relative and other prices.
This in my view goes a long way to explaining the conservative business communities gibbering panic over inflation. Because if what inflation is the re-stabilization of prices back to equilibrium, and much (most?) of the current inequity in society is the result of wide scale price distortions, then the business community would have a great deal to loose as wages/prices return to where the should have been.
Lrellock - no drubbing, but another perspective
written by A Populist, August 16, 2013 6:23
NAIRU is a useful concept.

There are some people who believe inflation is an evil, evil thing, no matter what. Not much you can do, to convince those people of implementing policies which will increase inflation - no matter how helpful those policies may be, for large segments of the populace - and even for the system as a whole.

However, there may be people willing to consider policies resulting in a higher, steady rate of inflation (or even a one-time increase in inflation) - so long as they believe that such inflation would be "contained", and that the overall effect on the economy is beneficial.

Well, it is useful to examine, under what conditions inflation can remain steady or dropping, and under what conditions will it simply spiral out of control. For many, NAIRU is a good concept to evaluate whether we are near that point.

Imagine you create a situation where there is a severe labor shortage, and so shortages of goods occur relative to demand (demand being people who want/need stuff *and* have enough cash or credit to buy it), and prices go up. Well, in a good, functioning economy, those higher prices will cause companies to offer higher wages, to meet that demand - in order to profit. But, there are not any more workers, so all this does is raise prices and wages, without any increase in output - a wage/price spiral. Now, obviously, if unemployment is zero, a wage/price spiral is possible. However, in the real world, there is some mismatch of skills, and other factors, so that leads to the concept of NAIRU - an approximate unemployment rate in a given situation, where, if you try to drive Unemployment below this level, a wage/price spiral could result.

Now, some might argue that you could get hyperinflation by other means. I think some people who are concerned about a large money supply may have a point that speculative cash can destabilize prices, as holders of wealth try to secure a profit or "store of value" by buying up commodities they don't need, etc - but I disagree that such would lead to hyperinflation - more likely some short term spikes, distortions, etc - which would decrease over time. And, if speculators were to make commodities markets non-functional, one would hope that government action could put a stop to that.

Maybe there are some other potential causes of never-ending price spirals, but, I think that NAIRU is the main one that we should consider. And, of course, NAIRU is telling us that hyperinflation worries are silly, at this time. (High Unemployment is like an anchor, keeping wages and prices low).
written by Lrellok, August 16, 2013 9:30
"But, there are not any more workers," This statement is nonsensical, unless you are exporting double digit percentages of your GDP. The consumer demand in an area is the sum of the individual demands of the population in that area. But nearly all consumers are potential workers, even if at a very limited capacity. SO in theory the demand for labor can never significantly exceed the potential supply, since the demand for products will only reach the supply of labor and no more.
Further, this is directly refuting not merely the theory but the existence of any and all labor saving technology. If labor was so constrained, then i would assume that labor technologies would be created to redirect low marginal utility labor into other areas. Increasing the overall supply of abstract labor.

The only way i could see that you would end up with your wage price spiral is if you have HUGE export drains on your economy or you are directly constraining markets to prevent labor saving technologies from developing. Actually, you would pretty much need to be doing both. Please clarify what other conditions could produce any shortage of labor assuming wages where free of all bounds?
written by A Populist, August 16, 2013 12:49
Re: " since the demand for products will only reach the supply of labor and no more. "

I don't think that assertion is justified - as a general rule (although right now there is an enormous shortfall of Demand - that has not always been the case).

There are a lot of reasons to believe that, if there is such a thing as NAIRU, that it can indeed be a very low number. I think at one point, U3 was in the neighborhood of 4%, without accelerating inflation. So, if you buy NAIRU as a concept to judge the ability to drive employment lower without accelerating inflation, we have plenty of slack. If you don't buy NAIRU as a limit on accelerating inflation, then fine - lets just try to get unemployment lower.

As far as trying to prove that NAIRU is a real limit - I really don't care at this point, since there are absolutely no policy options on the table which are going to make that discussion relevant anytime soon. So, unless you are talking to some crackpot who believes that, with U6 over 14%, we are too close to NAIRU, and so cannot do anything to lower U6 without risking hyperinflation, a discussion about NAIRU is rather moot.

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.