Bryan Caplan  

A Challenge for Anti-Keynesians

Krugman on Default... Time for a Third Party? Let's ...
Many of my favorite people are strident anti-Keynesians.  In their eyes, Keynesianism isn't just false; it's incoherent pseudo-science, a blight on our fair economics profession.  Those who think of Keynesianism as pseudo-science generally hold all Keynesians in one dim view.  They don't distinguish between reasonable and crazy Keynesians because they consider all Keynesians crazy.

If you're one of these anti-Keynesians, I have a challenge for you: Read Truman Bewley's Why Wages Don't Fall During a Recession from cover to cover.  Please.

You know I love you guys.  I think many of you reciprocate my warm feelings.  And I'm willing to bet your esteem for me that, after reading Bewley's book, you will agree that none of the following quintessentially Keynesian claims qualify as pseudo-science.  Furthermore, you will accept that at least two of these claims are, if not true, then at least plausible.

1. Nominal wage rigidity plays a large role in the modern U.S. economy.

2. A substantial fraction of this nominal wage rigidity stems from basic human psychology, not government regulation.

3. Nominal wage rigidity is sufficiently durable to create a long-run inflation-unemployment trade-off at low inflation rates.

If any anti-Keynesian registers for my challenge in the comments, I will happily link to whatever reaction you choose to blog.

COMMENTS (46 to date)

Do any of those ideas actually originate with Keynes? It seems to me that one could agree with all three without thereby being a Keynesian (and all the additional baggage that implies).

8 writes:

"although gold and silver are not by nature money, money is by nature gold and silver...." – Karl Marx

Lawrence D'Anna writes:

Bryan: could you explain, in terms that the Keynsians themselves would agree with, what it is that Keynesians actually believe? I feel like Scott Sumner probably would not object to 1, 2 and 3, but I doubt anyone would call him Keynsian.

When I think of Keynsians, I think of Paul Krugman saying crazy, incomprehensible things about liquidity traps, but that's probably just my own ignorance talking. What do Keynsians beleive? What makes them Keynsian?

Lawrence D'Anna writes:

Would Milton Friedman have objected to 1,2 and 3?

Would Hayek?

Larry writes:

While I agree that these assertions are quite defensible, I think they lead us not to ZLB misery, but rather to NGDPLT.

Glen S. McGhee writes:

"A substantial fraction of this nominal wage rigidity stems from basic human psychology, not government regulation."

To paraphrase Tina Turner, what's "psychology" go to do with it?

Randall Collins' chapter on "The Political Economy of Culture" (1979) covers this topic, with no mention of human psychology. You probably meant to say "sociology."

It's one of the ironies of the "government shutdown": Upper management gets to decide who stays home and who comes to work (upper-upper management decides who gets to make these decisions); middle management re-works the budgets and work schedules to address the changes that the shutdown has caused; and the low-level workers are told to stay home.

Nothing is accomplished, other than the added (and ongoing) internal management of priorities.

The focus is on the organization, and not on responsiveness to constituents -- which is ironic, because the government was shutdown for exactly that reason -- a lack of responsiveness, as a whole, to bloat and to unnecessary expense.

Ironic how the government shutdown perpetuates and legitimates government -- the opposite of what was intended.

Glen S. McGhee writes:

Sorry. I should have added this more recent example.

bill woolsey writes:

I don't think 1 and 2 lead to 3.

We don't have a lot of evidence that a 5% nominal GDP level target will result in lower unemployment on average than a 4% nominal GDP target.

Wage trajectories appear sticky, not just levels.

And further, to the degree zero is important, and that implies more separations, can we really take new hires as being independent?

In my view, it is foolish to try to overcome wage stickiness in order to adjust the real quantity of money to the demand to hold money.

But when we are considering appropriate reallocations of labor between industries, it is not clear that falling real wages in shirking industries, so that workers choose to quit only after they find a new job, is necessarily better.

And as I said above, it isn't clear that the scenario where the firm provides raises for some, keeping their real wages rising, while leaving the workers they don't want with falling real wages, solves the problem from the firm's point of view. Absolute wage cuts might be worse, but if this hurts moral too much, there isn't going to be a significant long run Phillips curve.

That is why wage growth seems so sticky.

liberty writes:

I agree with those above who say that there are many (including myself) who would agree with at least two, maybe all three, of those without considering themselves, and without anybody else considering them, to be Keynesian in the slightest. They are not really controversial claims -- although the third is basically empirical and the policy conclusions one will have will depend on what else you are concerned about re: inflation.

Where I begin to disagree with Keynesians is in their claims about modeling the economy with Aggregate Supply and Aggregate Demand curves, and taking a model built out of such aggregates and using it to "prove" that government must "boost" aggregate demand, and that somehow this will cure economic ills. I think the issues involved are much more complex and there are many more winners and losers than Keynes and his followers ever admit.

Bill writes:

I agree with the commenters. I think I might be hardwired against Keynesianism because it led me down such an intellectual dead end in my undergrad years I.e centralized management of "aggregate demand" for some homogenous "output", NAIRU etc.

I didn't really learn economics and economical thinking until I unlearned Keynesian pseudo science they feed to bored college sophomores.

Dan S writes:

I'm with the chorus on this one. A demand-side model of business cycles is, or should not be, the same thing as "Keynesian." I would say the term Keynesian ought to be restricted only to those models that imply monetary impotence at the zero bound and a large role for fiscal policy in maintaining demand. Although I agree there is an annoying tendency in economics discourse to label any demand-side model "Keynesian," even Milton Friedman's monetarism!

I too get frustrated by those who treat demand-side models as being some kind of voodoo. The basic questions to ask are these:

1. Do you agree that if there is a sudden flight to safe and/or liquid assets, for whatever reason, that is not accommodated by the fiscal or monetary authorities, then this will reduce nominal GDP? If not, why?

2. Do you agree that if wages/prices are sticky, a fall in NGDP (or a fall below trend) will result in unemployment, excess capacity, and below-potential RGDP? If not, why?

Now if you have an issue with wage/price stickiness, that's fine, and let's talk about it, but don't act like demand-side macro is some kind of logically impossible voodoo. Other schools, especially Austrian types, will inevitably jump in and say, "but local knowledge!" "central planning!" "printing worthless paper can't produce real resources!" etc., and then Russ Roberts will chime in and start talking about the wonders of the market, and so on, without ever addressing those two basic premises above.

AS writes:

(3) is debunked by 1970s stagflation. During persistent inflation, wage expectations rise, countering the goal of engineered inflation. Sporadic deflation is only a problem because workers have become indoctrinated to expect consistent mild inflation. Reset long-run average inflation to zero, or even negative, and worker expectations will adjust as well.

Eric Falkenstein writes:

Hayek noted in his book 'Tiger by the Tail' that the 'long run unemployment-inflation at low inflation' leads to high inflation if policy makers think this is something they can manage to achieve full employment. It's like saying there's a "long run opiate-happiness effect at low frequencies of opiate usage." True, but if you rely on that, you'll be hooked soon enough, and then the effect breaks down. Better not to play that game.

Glen S. McGhee writes:

A lot hinges on what you mean by anti-Keynesian.

This account of a massive anti-Keynesian purge during the McCarthy years -- here's how the earliest anti-Keynesians described themselves:

"Under [Dean] Bowen’s new standards, [old guard professor Blodgett's] teaching activities were limited and his textbook replaced by Samuelson’s Economics [!!!]. Blodgett (with the economic historian Donald L. Kemmerer) began a campaign through the press (Blodgett and Kemmerer were close friends of the managing editor of Champaign-Urbana New Gazzette Edward N. Jacquin) accusing the ‘new department’ to be full of «new dealers who expounded welfare-state and deficit spending theories…» Blodgett also accused
Bowen of impingement of his academic freedom (asking him «to use new tools of analysis» (63))."

This goes on to describe the mathematization of economics, something the old guard fought against while Dean Bowen and Everett Hagen put together a first-rank, enviable economics department.

Curt Doolittle writes:

I dont see what is keynesian about the argument.

Anti -keynesians are so because, largely, the externalities in the medium and long term are greater thsn the benefits in the short term because the misallication of capital distorts the function of prices and over-exploits whatener patterns of production and trade are currently being persued by the population. And because people age, they are not fungible resources, and prior opportunities for gaining skill and knowledge are unrecoverably lost, and dependent upon the next generations growth to compensate for.

Therefore keynesian calculations are aggregates divorced from demographics, and mask high opportunity costs with temporary over exploitation. And put the civilization in jeopardy if catastrophic cumulative risk , which we see as increasing bubbles.

Further it transfers the cooperative rules of society from between generations to within generations. Thus creating perverse incentives we have seen boomers demonstrate.

(Stated in ordinary language)

Steve J writes:

@AS "Reset long-run average inflation to zero, or even negative, and worker expectations will adjust as well". Sure they will adjust to the idea that holding money is profitable. How could this possibly be a good thing? Money in a mattress is the equivalent of an idle worker - wasted potential. Deflation creates perverse economic incentives.

RPLong writes:

Unless I have thoroughly misunderstood what I have read, it is possible to be an ardent Misesian Austrian economist while accepting both (1) and (2) and rejecting (3).

And while I don't have enough economic knowledge to speak definitively on this topic, I would venture to say that the principle reason why Misesians would reject (3) is because they don't mean the same thing as you do when they say "inflation."

Andre Mouton writes:

Good comments.

@Steve J: surely walking is wasted potential, in the sense that we could get where we're going faster if we ran. Standing still is the greatest waste of all -- but it's the best thing you can do if lost in the woods. The issue with Keynes isn't sticky wages, but linearity. Keynesians assume that we know where we're going, so it's just a matter of getting there as quickly as possible. It's the "are we there yet" school of economics.

Sam Wilson writes:

I've got a short response here.

The very short version: I already accept wage rigidity, but I do not believe that many of the proposed remedies are consistent with the insights of public choice.

Curt Doolittle writes:

1) Rod Long is correct on the austrian position on 1,2 vs 3, - as usual. Although I think that this direct question with a direct answer masks our problem with Keynesianism.

2) DanS : I don't think 'voodoo' is terribly articulate. We agree that demand side models are accurate. We argue that benefits are short term, numeric, and illusionary. Because I think what we don't agree on is consequences as I've stated above.. And I think in particular, we disagree that it is wise or moral to deprive savers and lenders of natural interest in exchange for reducing unemployment because it destroys the system of cooperation between the generations, and it forces investment and industry to speculate rather than attempt to generate profits. Which exacerbates risk. Hard to argue with this, except for very, very, short corrections. It would be better to directly redistribute taxes in the form of disposable cash to people than to increase the money supply sufficiently to create inflation. (Although that is a more pragmatic positions than most Austrians would agree with.)

3) As I have written consistently, since 2006, the problem is a moral one, as much as an economic one. MONETARY policy is harmful to savers, exacerbates risk, and harms cooperation between the generations, but it is at least, politically, fairly neutral. THe problem with Keynesian FISCAL policy, is that it is not politically neutral. Spending just empowers political action that most of us deem corrupt. Even so far as the rural areas view the cities as corrupt. So Austrian arguments become favored as a rhetorical and moral means of objecting to political favoritism and corruption. And they are good arguments for that.

So, to Bill Wooley, Bryan, and the rest, I think the argument is a straw man. It asks a valid but irrelevant question that has little or nothing to do with the dispute between Conservative economists (austrians) and Progressive economists ( fiscal policy advocates).

The argument is not analytical. It's political.

It was a lot harder for me to make this argument even in 2010 than it is today - when it's just patently obvious from demonstrated political behavior, that the problem is not a failure to understand. It's that progressive economists fail to grasp the fact that voters vote MORALLY, not 'ignorantly' as Bryan tries to advocate.

(Testing? Were you serious? I thought it was a joke.....Seriously.)

Politics is first and foremost a moral not economic enterprise. Otherwise laws around the world would not reflect family moral codes.

Hunter writes:

I just want an explanation of how increasing inflation which is supposed to create jobs by lowering wages is embraced by so many who hold to a demand side theory. Inflation decreases demand by those who have still have jobs by reducing the value of their wages.

R(yan)PLong writes:

@ Curt Doolittle - If your point #1 is in reference to my post at 10:21 AM, I need to clarify that I am a huge fan of philosopher Roderick T. Long, but as far as I am aware, no relation. We seem to share a similarity in names and philosophical interests, but in every other respect I consider myself the lesser of the two R. Longs. :)

Dan S writes:


I'm not sure what you mean by the "short term, numeric, and illusionary" comment. The demand-side model implies that the problem is NOT, for example, a lag in workers switching from building houses to producing something else, but rather the fall in demand is broad-based. There are unemployed workers willing and able to work if only they could be employed, and there are consumers and businesses who would be willing and able to buy their product (be it consumption or capital investment) if only the outlook for employment and nominal growth were better. The idea of demand management is that if you step in with fiscal or monetary policy, consumers and businesses, feeling safer, will consume, and unemployed workers will be employed (of course usually it is the same people wearing different hats). There's nothing illusory about it. Short-term, yes, but believe me, we have never, nor will we ever reach, "the long-term." Life is just a series of short-terms, and if you don't manage them properly, you just get misery for a long time.

As for interest, I don't believe there is such a thing as a "natural interest rate," as the interest rate that prevails is entirely dependent on the monetary context. Is the "natural interest rate" that which would prevail under a policy of 0% inflation targeting? Under a gold standard? Monetary aggregate targeting? Free banking? We should instead be asking, what monetary regime best produces our goal (something like, minimal macro-economic volatility), and then simply let the interest rate do its thing.

MarcusEMann writes:

A few thoughts...

You do realize that Keynes big idea was to look at the components of GDP = C + I + net X + G and to say that when the others aren't having a possitive effect on the economy (G)overnment can play a part. That does not simply imply that govenment spends during a recession, but also that govenment tightens during growth. Ideally, we would concentrate our descretionary spending in rececessions and cut back during times of growth. While this may be politically difficult, that is still the underlying premis.

Daublin writes:

Joining the chorus, when I hear "Keynesian", I think of boosting the economy by spending increases.

The three items you list are not the weak points of that strategy.

AS writes:

@Steve: "Money in a mattress is the equivalent of an idle worker - wasted potential. Deflation creates perverse economic incentives." Money in the mattress is not an unemployed worker. Money is just a piece of paper. Real resources are what matter. Real rates of returns on investments are independent of inflation.

David R. Henderson writes:

I just ordered the book. Thanks. And what did I see on Amazon? The one review, which was quite positive, was by Garett Jones.

MikeP writes:

Money in a mattress is the equivalent of an idle worker - wasted potential.

Indeed, this notion is fraught with error.

As AS notes, money is just paper. Money in a mattress means either (a) that all other money becomes more valuable or (b) that the printer of money can produce more without risk of inflation. Money in a mattress simply means that the demand of the hoarder has less play: but the demands of others have comparatively more.

BZ writes:

It's #2 I have a problem with. Like most planning arguments, it forces me to think "at the margin, people will stare a perfectly ordinary problem in the face, know exactly what they need to do, have the incentive and the means to solve it, but instead will just sit there and wait for the state to solve it for them."

That just doesn't seem real.

Jim Rose writes:

Pissarides, C (2009), The Unemployment Volatility Puzzle: Is Wage Stickiness the Answer?, Econometrica argues the wage stickiness is not the answer since wages in new job matches are highly flexible:

1. wages of job changers are always substantially more procyclical than the wages of job stayers.

2. the wages of job stayers, and even of those who remain in the same job with the same employer are still mildly procyclical.

3. there is more procyclicality in the wages of stayers in Europe than in the United States.

4. The procyclicality of job stayers’ wages is sometimes due to bonuses, and overtime pay but it still reflects a rise in the hourly cost of labor
to the firm in cyclical peaks

from see

how do existing firms survive in competition with new firms who can start workers on lower wages? Industries with many short term jobs and seasonal jobs would suffer less from wage inflexibility

Merijn Knibbe writes:

Another one: 'seven unsustainable processes' from Arch-Keynesian Wayne Godley (1997, not a 'Samuelsonian Keynesian', a real one. He maps the flows of money and shows that some of these flows are unsustainable. Unlike neo classical macro, which relies on the rather vague and ambigious concept of utility, all the concepts he uses are well defined and estimated. And he was right, too...

[url simplified. No need for three lines of code just to get readers to that same pdf file, eh?--Econlib Ed.]

Darryl FKA Ron writes:

[It looks like MarcusEMann may have actually read a little Keynes and comprehended it. Among those that identify with Keynes, both anti and pro, that seems to be a novelty. Kudos.]

Dan S writes:

...Short-term, yes, but believe me, we have never, nor will we ever reach, "the long-term." Life is just a series of short-terms, and if you don't manage them properly, you just get misery for a long time.

As for interest, I don't believe there is such a thing as a "natural interest rate," as the interest rate that prevails is entirely dependent on the monetary context. Is the "natural interest rate" that which would prevail under a policy of 0% inflation targeting? Under a gold standard? Monetary aggregate targeting?..

[It is a small box drawn for us to think within at times. Glad to see an escapee.]

Eugene Patrick Devany writes:

[Comment removed pending confirmation of email address and for crude language. Email the to discuss editing your comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Perplexed writes:

Labor is exempt from anti-trust laws. Lack of government regulation that exists in nearly every other market hardly constitutes "basic human psychology."

Steve J writes:


I get confused when what seem to be simple ideas are twisted to become difficult. My first point was deflation is bad. In general we would expect a higher performing economy with 2% inflation rather than 2% deflation right? My second point was idle money is bad. In general we would expect a higher performing economy when people invest money rather than burn it correct?

MikeP writes:

In general we would expect a higher performing economy with 2% inflation rather than 2% deflation right?

This is not at all obvious. For example, US history saw its greatest annual GDP growth in the latter decades of the 19th century during a time of persistent deflation.

In general we would expect a higher performing economy when people invest money rather than burn it correct?

Loss of money simply makes still existing money more valuable. The economy can produce so many goods and services. Consumers present a demand on those goods and services. Whatever the extant money supply happens to be is used to bid on those goods and services.

Do you think that if everyone were allowed to write another zero on the end of their paper bills the economy would be nine times higher performing? Then why if everyone burned 90% of their money would the economy perform worse?

Note that this does not mean that taxes are harmless! Taxes are not a problem because people have a third less money after the taxman comes. Taxes are a problem because the government actually doesn't burn that money, but rather uses it to bid on less valuable goods and services that people haven't chosen to buy themselves. I.e., the government directly competes with the actually productive economy.

Pierre Fortin writes:

Bryan. See my new evidence on the macro consequences of DNWR in Canada at
Quite compelling.

rick hull writes:

1 & 2 seem perfectly plausible to me. I'm happy to listen to any stories that assume their truth.

Alexander Hamilton writes:

@ anti-Keynesian commenters.

The Chicago/Minnesota rejection of Keynes was all about sticky prices/wages, even though this isn't as central to being a Keynesian as rejecting Say's Law.

Nevertheless, the freshwater hires in my department still deny sticky prices thinking it shows they're rejecting Keynesianism.

Now, faced with even more evidence that sticky prices are undeniable, commenters try to pretend the rejection of Keynes was never about sticky prices, it was about rejecting simple models in favor of more complicated-looking models. Right. In fact the heart of being a Keynesian is a recognition that Say's Law is false -- sticky prices and wages, while true, is really just a side issue.

rick hull writes:

1 & 2 seem perfectly plausible to me. I'm happy to listen to any stories that assume their truth.

Maggie writes:


might I suggest that anyone who is Keynesian or even anti-Keynesian, takes a good long hard look at the period of the late 1960s to mid-1970s, up to about the mid-80s. This is the period that explains why the theories of Keynes failed.

The reason for looking at this particular period is summed up as Stagflation.

No one talks about the Stagflation of the 1970s, or the factors surrounding that Stagflation. This was a world wide problem and it required some rather harsh economic measures to fix our economies.

I do think that Keynes got some things correct for his own time. I think that he was a very good observer of world economies at that point in time. He was correct that the Treaty of Versailles would lead to economic disaster in the future.

Keynes was dealing with the economy in the UK. He was dealing with the specific conditions in the UK during a world-wide depression. His remedy was the dole.

Australia adopted many of the Keynesian ideas and the wheels fell off at the same time as they fell off for other world economies - the 1970s. What preceded the falling off of the wheels happened to be an increase in government spending and a move away from a balanced budget.

There is one quote from Keynes that sticks in my mind, and that is about the necessity to keep the government budget balanced by "raising taxes in times of war" and "reducing taxes and spending at other times". The raising of the taxes in times of war is to finance the extra spending necessary for defense purposes.

A lot of what anti-Keynesians believes in my view is "straw-man arguments". They build up the straw man and then knock down what they have built up and all the while failing to see the Marxist elephant in the room... in this case someone like Krugman who constantly espouses Marxism which is not the same as Keynesian economics.

Philippe writes:

[Comment removed for rudeness. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Robert Nielsen writes:

While I don't qualify as an anti-Keynesian, I will certainly check out that book.

Fair play on your willingness to give credit to the other side.

Slim934 writes:

I am certainly not a professional economist (although I sometimes like to play one after imbibing a few) and consider myself somewhat Anti-Keynesian.

But I think before one can say whether they are or are not Anti-Keynesian, you need to clarify some of the basic tenets of the Keynesianism you are referring to.

My basis of being Anti-Keynesian is basically the writings of Henry Hazlitt when he wrote "The Failure of the New Economics" as well as Hunter Lewis's work on Keynes. If this is the Keynes we are talking about, then yes I am anti-Keynes. If not, then please describe what makes one a Keynesian.

As to "Why Wages Don't Fall During a Recession", I don't really see how this affects the underlying notion of a general decline in wages throughout the economy during a recession. Does the wage fall have to be even over all workers? Why could it not be the case of firms trimming off a bunch of workers and simply loading up their remaining people with more work? Would that not be effectively a drop in wages? Part of the issue is that we need to have a genuine REAL recession free of market mucking before we can really say whether this is true or not from an empirical standpoint.

Furthermore, how far back in history does his analysis go? I mean it seems to me that if one can show a single historical incidence where there WAS a general decline in the wage rate then you've proven the general crux of his argument wrong. It would seem to me pretty easy to disprove if one were to use say....the 1920-21 depression wage data. If there were characteristics drops in the wage rate throughout the economy, then it proves is his point wrong that wages DO fall during a recession.

That being said, it would do nothing to his analysis of the modern economy which he seems to have taken great pains to document. However, the modern economy IS NOT the economy always and forever. Indeed, it could be that the psychology we have now (which influences firing behavior) could be heavily altered by the underlying economic fundamentals of a fiat monetary regime.

Jim Rose writes:

if workers morale is so important a brake, wage cuts should vary systematically with the power of employee effort monitoring technology and the extent of team production and production lines?

are piece rates and bonuses immune to nominal cuts?

Ed Lazear did some good work on what happens after switches to and from piece rates. half the about 44% changes in output comes from self-selection, the rest from discretion over effort.

terrymac writes:

Is wage stickiness as important as Keynesians think it is? Probably not. A free economy would increase productivity fast enough that people wouldn't care so much about the level of wages. Imagine if the rest of the economy were as liberated as the market for electronics and laser surgery. Look at the story about India radically reducing the cost of heart surgery.

Instead of playing games with trying to boost the GDP - about which, another discussion - economists should highlight just how very much the regulations cost us, and how pervasive and irrational those regulations are.

With respect to GDP - whose bright idea was it to include government spending? if it should be included, reverse the sign - it subtracts from far more productive uses of our resources.

In particular, nothing is more stupid, more un-economic, than government education. Children who are homeschooled are running rings around those who are taught by the government, AND they are doing it in a small fraction of the time. This is one of the most obvious examples of how the government is greatly reducing the quality of life through regulation; it is taking excellent choices off the table for most educators.

A recent article in Wired told the story of a Mexican teacher who threw out the rulebook and let his students do most of the driving. At the end of the year, ten of his students scored at the 99.99th percentile level in math. Normally that would require a population of ten thousand students; he surely taught a few dozen, not thousands. Many homeschoolers are having equally remarkable success; their children are starting college at age 12 or 14; those who are taught by government, by contrast, require remedial math and English instruction. Our present system is severely handicapped by government regulations, and we're reduced to deciding how much money to introduce to artificially stimulate spending? Is that the only variable we can tweak?

Comments for this entry have been closed
Return to top