More NAIRU bashing

4 Mar, 2017 at 18:23 | Posted in Economics | 8 Comments

As yours truly argued in a post the other day, NAIRU does not hold water simply because it hasn’t existed for the last 50 years. But still  today ‘New Keynesian’ (a monstrous misnomer) macroeconomists use it — and its cousin the Phillips curve — as a fundamental building block in their models. Why? Because without it ‘New Keynesians’ have to give up their (again and again empirically falsified) neoclassical view of the long-run neutrality of money and the simplistic idea of inflation as an excess-demand phenomenon.

The NAIRU approach is not only of theoretical interest. Far from it.

The real damage done is that policymakers that take decisions based on NAIRU models systematically implement austerity measures and kill off economic expansion. The unnecessary and costly unemployment that this self-inflicted and flawed illusion eventuates, is something its New Classical and ‘New Keynesian’ advocates should always be kept accountable for!

According to the  [NAIRU theory], unemployment differs from its natural rate only if expected inflation differs from actual inflation. If expectations are rational, we should see as many quarters when inflation is above expected inflation as quarters when it is below expected inflation. That suggests the following test of the [NAIRU theory]..

74-7495-LTNQ100ZBecause a decade contains 40 quarters, the probability that average expected inflation over a decade will be different from naverage actual inflation should be small. If the [NAIRU theory] and rational expectations are both true simultaneously, a plot of decade averages of inflation against unemployment should reveal a vertical line at the natural rate of unemployment … This prediction fails dramatically.

There is no tendency for the points to lie around a vertical line and, if anything, the long-run Phillips is upward sloping, and closer to being horizontal than vertical. Since it is unlikely that expectations are systematically biased over decades, I conclude that the  [NAIRU theory] is false.

Defenders of the [NAIRU theory] might choose to respond to these empirical findings by arguing that the natural rate of unemployment is time varying. But I am unaware of any theory which provides us, in advance, with an explanation of how the natural rate of unemployment varies over time. In the absence of such a theory the [NAIRU theory] has no predictive content. A theory like this, which cannot be falsified by any set of observations, is closer to religion than science.

Roger Farmer

For more NAIRU critique see Merijn Knibbe’s post on the rwer blog today.

8 Comments

  1. Jerry,

    You say “NAIRU assumes that there is a specific rate of unemployment that can be determined ahead of time and that therefore it can be used in policy targeting to avoid accelerating inflation.”

    Well one of the first papers to suggest the NAIRU idea was Franco Modigliani and Lucas Papademos’s Brookings paper “Targets for Monetary Policy in the Coming Year.” (1975). The authors certainly do not suggest NAIRU can be estimated with any huge degree of accuracy “ahead of time”. For example they suggest it is “somewhat over 5.5%”. They also admit that NAIRU varies over time depending on sundry factors like terms of trade (p.142).

    And another of their phrases reads: “The shading of an area on either side of NIRU indicates both uncertainty about the exact location of NIRU and the implausibility that any single unemployment rate separates accelerating and slowing inflation”. Having said that, I think even those authors are too optimistic on how accurately NI(A)RU can be estimate.

    Quite distinct from that however, is the question as to whether it is legitimate to write equations or have discussions which assume NAIRU has a very specific value. I say it is, because in principle, though obviously not in practice, it can be accurately determined. Likewise it is legitimate to assume the Moon is composed of a very specific number of protons and to an accuracy of say one in a trillion even if human beings can only estimate the number to say one in a billion. I.e. ask God the exact value for NAIRU or the number of protons composing the Moon, and he’d give you the answer.

    I have often found it useful to “assume for the sake of argument that NAIRU = X%”, e.g. in the works linked to below.

    Re JG, it certainly should not be designed in such a way as to cause excess inflation. I am pleased to say the ideas I was promoting on JG 20 years ago have actually been put into effect in the UK. That’s in contrast to the current crop of JG enthusiasts who I regard as incompetent and whose ideas have no chance of being put into effect: certainly not under a Republican regime in the US.

    In Sweden (Lars’s homeland) they were very keen on JG schemes 20-50 years ago. They’ve subsequetly cut expenditure on them as they often turn out to be a waste of money. My ideas on JG (way beyond the comprehension of the current crop of JG advocates) are here:

    http://mpra.ub.uni-muenchen.de/19094/
    and:
    http://mpra.ub.uni-muenchen.de/14206/

    Comment by Ralph Musgrave— 6 Mar, 2017 #

  2. Jerry,

    “First, it assumes the existence of NAIRU for that time period.” Well if NAIRU has no “existence” at all, i.e. if there is quite literally no relationship at all between inflation and unemployment, I repeat the question: why don’t NAIRU bashers (you included) advocate a massive increase in demand with a view to reducing unemployment to zero or thereabouts? Reason is that you know perfectly well there is some sort of relationship between inflation and unemployment: i.e NAIRU does have “some existence”. That is why you are reluctant to answer that question.

    “It assumes that NAIRU could be accurately measured”. I have never said, nor have I ever meant to suggest that “accuracy”. It strikes me as blindingly obvious that NAIRU cannot be accurately measured. Life and the universe is full of instances where it is obvious beyond reasonable doubt there is some sort of relationship between two variables, while the relationship is not accurately known.

    “It assumes policymakers can effectively target that NAIRU rate.” I assume by that you mean the authorities can control the unemployment rate to the nearest 1% or 0.1% or whatever. Again it is blindingly obvious that unemployment cannot be controlled that accurately.

    “And it assumes that there are no other causes of inflation during that time period except NAIRU.” Seems I need to keep repeating myself, but here goes. Advocates of NAIRU are perfectly well aware that there are factors contributing to inflation other than the amount of unemployment: e.g. there are expectations and shortages of capital equipment. I’ll repeat that.

    Advocates of NAIRU are perfectly well aware that there are factors contributing to inflation other than the amount of unemployment: e.g. there are expectations and shortages of capital equipment.

    In fact if you want a specific example of that (one you ought to be aware of but quite obviously are not) consider the fact that about 5 years ago the Bank of England reckoned the well above target inflation at that time WAS NOT caused by excessively low unemployment. The BoE therefor did not raise interest rates. I thought at the time they were right, and it turned out they were.

    You don’t seem to have got the point that the world is a complicated and messy place, hence the fact that a relationship cannot be accurately measured is not a good reason for saying the relationship does not exist. The effect of a human being inhaling flu viruses is very unpredictable: the effects range between death and no observable effects at all. Luckily for the human race, medics have more common sense than economists: medics don’t conclude that because a relationship is highly uncertain that therefor the relationship does not exist.

    Comment by Ralph Musgrave— 5 Mar, 2017 #

    • Ralph, I was answering your question as to what you missed in your argument in your first comment. That one particular argument you made suffers from assuming inflation is only caused by unemployment falling below NAIRU. Along with all the other things I mentioned in my last comment.

      Saying that there can be “some sort of relationship between inflation and unemployment” is a lot different from saying NAIRU is a useful concept. NAIRU assumes that there is a specific rate of unemployment that can be determined ahead of time and that therefore it can be used in policy targeting to avoid accelerating inflation. Saying there can be a relationship between full employment and excess demand and inflation seems reasonable to me- it just isn’t NAIRU and you should make up a new name for it. I will be happy to call it Musgrave’s Theorem or whatever else you decide.

      Another thing- I do support the Job Guarantee which would lower involuntary unemployment to about the zero range. I believe a JG might very well cause a large increase in AD and lead to an initial burst of inflation. So I do advocate that, which you probably already know anyways.

      As to the inflation rate in the UK 5 years ago. I don’t live in the UK and don’t follow all their stats but is this when the VAT was raised and prices rose also? If so, I would congratulate the Bank of England for realizing they were connected somehow. Although I would expect almost anyone to realize that as well.

      Comment by Jerry Brown— 6 Mar, 2017 #

  3. Roger Farmer’s PDF is absolutely fascinating. I wish I could leave it at that. But I got to the part where Farmer describes his Chart 1:

    “There is no tendency for the points to lie around a vertical line and, if anything, the long-run Phillips curve revealed by this chart is upward sloping, and closer to being horizontal than vertical.”

    It is true that on Farmer’s chart the data points do appear to lie closer to horizontal than vertical. But the claim is meaningless. It can be proved incorrect simply by making the chart taller and less wide.

    The average values of inflation and unemployment that Farmer uses are not like coins where, when you stack them up, the pile gets taller. Average values don’t have thickness. They don’t have width. Farmer’s “closer to horizontal” claim is naive.

    As noted, I find Farmer’s paper very interesting. I want there to be things in it that I can use. If I raise questions it is because I need to test certain points for validity so I can be sure my arguments will not be made weaker by quoting him. So, I have another concern. Setting up his test of the NRH, Farmer says:

    “a plot of decade averages of inflation against unemployment should reveal a vertical line at the natural rate of unemployment.”

    But his chart shows “no tendency for the points to lie around a vertical line”.

    I think Milton Friedman would say there is a vertical line thru each of Farmer’s dots, and that each vertical represents a different natural rate. Apparently, Roger Farmer thinks so, too. He tries to preempt the “many verticals” argument, saying:

    “I am unaware of any theory which provides us, in advance, with an explanation of how the natural rate of unemployment varies over time.”

    Sadly, one’s inability to explain how the natural rate may vary is not evidence that it doesn’t vary. Nor is it evidence that the rate doesn’t exist. Farmer’s ignorance does not turn his chart into strong and valid evidence.

    I think the ‘natural rate’ is probably hokum. But if my concerns are valid, Chart 1 is definitely hokum.

    Comment by The Arthurian— 5 Mar, 2017 #

  4. Jerry, Thankyou for pointing out that the UK has not had an inflation target of EXACTLY 2% for the last 70 years. Strictly speaking you’re correct, but frankly that’s a bit of a trite point: I mean if inflation had shot up to 10% prior to the arrival of NAIRU are you seriously suggesting the authorities would have done nothing about it? I.e. prior to NAIRU there was in effect an inflation target, but it was a ROUGHER one than the more precise inflation targets that governments aim for nowadays.

    Moreover prior to the existence of NAIRU there was the Phillips curve, which comes to roughly the same thing. And before anyone makes the blindingly obvious point that the “A” (acceleration) element in NAIRU is not present in the Phillips curve, I am well aware of that.

    And prior to the existence of the Phillips curve, anyone with half a brain realized there is a relationship between inflation and unemployment: David Hume writing about 200 years ago was obviously aware of the fact. And the ancient Romans when they had their credit crunch around 33AD would doubtless have noticed that unemployment was relatively high and inflation relatively low in the deflationary period before 33AD.

    Next, you say “you are assuming that inflation is always and only caused by the unemployment rate falling below NAIRU estimates.” I did not say that, nor did I say anything that suggested that. Anyone with a grain of common sense (and that includes me) is aware that there are other factors involved, e.g. expectations, shortages of capital equipment, etc.

    Next, in your final paragraph you say “if as Lars maintains, NAIRU is a wrong idea” it will result in governments pitching demand too low. Well clearly that’s true if by “wrong” you mean there is no relationship WHATEVER between inflation and unemployment. But if that’s the case, why (as suggested for example by Simon Wren-Lewis) do NAIRU bashers not advocate a MASSIVE increase in demand with a view to giving us a HUGE reduction in unemployment and associated social problems? Reason is that NAIRU bashers know perfectly well THERE IS a relationship between inflation and unemployment.

    On the other hand if Lars is simply saying that the NAIRU relationship exists but is not as precisely known as we would like, I have no quarrel with that, and nor do most NAIRU advocates. In that case the best we can do is what governments actually do, which is to keep testing the relationship: i.e. keep trying to raise demand till inflation gets uppity, then rein in demand a bit.

    Comment by Ralph Musgrave— 5 Mar, 2017 #

    • Ralph, perhaps I misunderstood your argument in your first comment-

      ““The real damage done is that policymakers that take decisions based on NAIRU models systematically implement austerity measures and kill off economic expansion.”
      Well if that were the case, then we’d see inflation persistently below the 2% target. But that’s not what we see.!!! At least not in the UK where I live. In the UK over the last 70 years or so, inflation has averaged well over 2%.
      Now if that does not neatly demolish Lars’s point, what have I missed?”

      It seems to me that you are trying to discredit Lars Syll’s observation that policy based on NAIRU will likely harm the economy by cutting expansions short out of fear for inflation.

      You say that this is obviously false because over the last 70 years inflation in the UK has averaged over 2%. Since inflation has averaged over 2%, policy makers have not cut expansions short and Lars is wrong. This is not a good argument.

      First, it assumes the existence of NAIRU for that time period. It assumes that NAIRU could be accurately measured. It assumes policymakers can effectively target that NAIRU rate. It assumes policy makers have actually targeted that rate. And it assumes that there are no other causes of inflation during that time period except NAIRU. What you have missed is that all of these assumptions are wrong.

      Comment by Jerry Brown— 5 Mar, 2017 #

  5. “The real damage done is that policymakers that take decisions based on NAIRU models systematically implement austerity measures and kill off economic expansion.”

    Well if that were the case, then we’d see inflation persistently below the 2% target. But that’s not what we see.!!! At least not in the UK where I live. In the UK over the last 70 years or so, inflation has averaged well over 2%.

    Now if that does not neatly demolish Lars’s point, what have I missed?

    Of course it could be argued that the 2% target is too low. But that’s AN ENTIRELY SEPARATE POINT. IT HAS NOTHING TO DO WITH THE PROS AND CONS OF THE BASIC NAIRU IDEA.

    In any case, the consensus is that 2% is about right, though many Germans argue for less than 2% and the official ECB target is not 2% but 2% as an absolute maximum.

    Comment by Ralph Musgrave— 4 Mar, 2017 #

    • Ralph Musgrave, I wasn’t aware that the UK has had a 2% inflation target for the last 70 years. Or that the NAIRU theory was that old either.

      But more importantly, you are assuming that inflation is always and only caused by the unemployment rate falling below NAIRU estimates. I don’t think even most NAIRU supporters imagine that there are no other possible causes of inflation, or at least I hope they don’t.

      And if as Lars maintains, NAIRU is a wrong idea in the first place, policy attempting to make unemployment fit the NAIRU level is going to be misguided and probably harmful and may have absolutely no effect on the actual inflation rate in any event.

      Comment by Jerry Brown— 5 Mar, 2017 #


Sorry, the comment form is closed at this time.

Blog at WordPress.com.
Entries and Comments feeds.