Keynes vs. Reality-2

by Russ Roberts on July 14, 2011

in Stimulus

In this earlier post, I noted this 1943 Paul Samuelson prediction:

When this war comes to an end, more than one out of every two workers will depend directly or indirectly upon military orders. We shall have some 10 million service men to throw on the labor market. We shall have to face a difficult reconversion period during which current goods cannot be produced and layoffs may be great. Nor will the technical necessity for reconversion necessarily generate much investment outlay in the critical period under discussion whatever its later potentialities. The final conclusion to be drawn from our experience at the end of the last war is inescapable–were the war to end suddenly within the next 6 months, were we again planning to wind up our war effort in the greatest haste, to demobilize our armed forces, to liquidate price controls, to shift from astronomical deficits to even the large deficits of the thirties–then there would be ushered in the greatest period of unemployment and industrial dislocation which any economy has ever faced.

From Paul Samuelson, “Full Employment after the War,” in S.E. Harris, ed., Postwar Economic Problems, 1943.

Samuelson was wrong. When the war ended, there was massive demobilization of the armed forces and a large reduction in government spending. Yet the economy thrived, unemployment remained very low and there was a huge expansion of private sector employment.

One defense of Samuelson is that this quote is mainly about the dynamics of the labor market–it doesn’t say much about aggregate demand. It doesn’t focus particularly on the drop in government spending. But if you go back to the original article (which you can find by taking a line or two and using Google books–I hope to post more on this later) you will find other passages that make it clear that he saw this as a problem caused by a sudden drop in aggregate demand.

There were a number of interesting comments on the post. Some noted the drop in GDP after the war but forgot that wartime price controls and large amounts of government output made measuring of GDP during the war quite difficult. There was a mild recession in 1945–during the war. The most dramatic change was the one I highlighted–a large expansion of private sector employment with very little unemployment. It was an incredible transition achieved with surprising speed.

I’d like to highlight one comment by Daniel Kuehn:

Samuelson was wrong in forecasting. Keynes wasn’t wrong in theorizing.

Why did Samuelson end up being wrong? Have you ever thought that through?

Because post-war demand was substantial and he didn’t expect that.

High levels of private sector demand and a full employment economy… where the hell do you get “Keynes vs. Reality” from that? That has Keynes written all over it.

All you’re demonstrating is that Samuelson didn’t have a crystal ball. That isn’t exactly a news flash.

This defense of Keynes unintentionally highlights a flaw in Keynesianism at least as practiced by Daniel but I don’t think he’s alone or I wouldn’t single him out. (And I would add that I usually appreciate Daniel’s comments, particularly their tone in the face of criticism from the rest of you.) One, would an economist presume that a change in government spending would have no effects on other types of spending? If government spending shrinks and there is suddenly more resources available to the private sector, why would you not foresee a change in private sector spending? Secondly, the fact that full employment corresponds to high levels of spending (regardless of their source) is very close to a tautology. The challenge is to understand causal linkages and the direction of causation.

In the last few sentences of the previous paragraphs, I had originally referred to private demand. I changed it to private spending. I don’t know what private aggregate demand means, or the phrase “pent-up” demand. The usual way that Keynesians explain the post-war expansion despite the huge cut in government spending is to say, well of course the economy boomed, there was a lot of pent-up demand. What does that mean? There is always pent-up demand in the sense there is a stuff I wish I could have but can’t. But the standard story is that people couldn’t buy washing machines or cars during the war–they were rationed or simply unavailable or unaffordable. So when the war ended, and rationing and price controls ended, people were eager to buy these things. But the reason these consumer goods were rationed or unavailable is because all the steel went into the tanks and planes during the war. So when the war ended, there was steel available to the private sector. That’s why cutting government activity can stimulate the private sector. Fewer resources are being commandeered by the public sector. As the Hayek character says in the Fight of the Century when answering the Keynesian claim that WWII ended the Great Depression:

Wow. One data point and you’re jumping for joy
the Last time I checked, wars only destroy
There was no multiplier, consumption just shrank
As we used scarce resources for every new tank

Pretty perverse to call that prosperity
Rationed meat, Rationed butter… a life of austerity
When that war spending ended your friends cried disaster
yet the economy thrived and grew faster

This earlier post where I discuss Krugman’s claims about rationing and the austerity of the early 1940′s may also be of interest.

Be Sociable, Share!



71 comments    Share Share    Print    Email


sargasm July 14, 2011 at 8:33 pm

Demand is magic. Demand somethin through legislation, and it magically appears.
Supply is meaningless. People (government to be specific) must demand something before it is supplied.
Entrepreneurship is a myth. To think that average people could anticipate demand and provide… preposterous! Only government can do that.
Demand comes first, in the form of government mandates, and supply comes second.
That is how we got the wheel, the lever, the steam engine (probably the most important invention of the last thousand years), and everything else we use.

GP Hanner July 15, 2011 at 11:01 am

and unicorns. Don’t forget the unicorns.

Ryan Langrill July 14, 2011 at 8:37 pm

I can kind of see where Samuelson is coming from. Before WWII there was not significant private demand, why would he have expected anything to change in the meantime? Leijenhuvfud, who I like quite a bit, argues something like: ‘within a certain corridor, market self-corrections work great, but once you are outside it the price system stops correcting itself to full employment. The result of WWII was not to increase aggregate demand, but to put capital into the hands of the private sector and push the market back into the self correcting corrodor.’ It’s in a footnote, so he doesn’t elaborate or test the hypotheses against, say, Higgs, but it’s an interesting interpretation of Keynes (like he and Clower were, generally).

Ryan Langrill July 14, 2011 at 8:39 pm

I am not saying I buy either Samuelson or Leijenhuvfud, I was just throwing that out there as food for thought.

vidyohs July 14, 2011 at 9:07 pm

“Before WWII there was not significant private demand,”

I don’t understand.

Are you saying that prior to WWII people took things like fabric, ready made clothing, shoes, houses, autos, tractors, food stuffs, perfumes, shaving creams, razors, telephones etc. etc. for granted and were ambiguous about their acquisition? They could take or leave those things? People routinely went naked or semi-naked because there was no significant private demand for clothes?

So, how did the end of WWII change that and stimulate sudden desire for those consumer goods. Were people suddenly embarrassed to go semi-naked, live under bushes, and eat bark off trees because the war ended?

Ryan Langrill July 14, 2011 at 9:24 pm

I am attempting to understand, from his own point of view, what Samuelson was thinking. Before the war, Samuelson might have argued, there was an excess supply of goods and services because there was an excess demand for money. In other words, insufficient aggregate demand for full employment – you are sitting perpetually inside the ppf (I don’t know why that would cause unemployment, but whatever). In the Keynesian model, the war would have caused a decrease in private consumption – but not at a one-to-one rate, since some resources are involuntarily unemployed. Unless something changes in the business environment, why would Samuelson have thought 1946 would be better than 1937?

I think professor Roberts is right, I was just throwing Leijenhuvfud out there for the sake of discussion, so I don’t really get the point of your post.

vidyohs July 15, 2011 at 10:44 am

Okay, when you wrote the words I quoted above, they weren’t your thoughts but a quote of someone else? If that is correct, then my comments weren’t directed at you but at the person who originated that thought.

If it was a quote, can suggest that in the future you put quotation marks around what you’re quoting, when you’re quoting, so people will know. Don’t assume all of us will know, show us with the marks.

Subhi Andrews July 14, 2011 at 8:39 pm

BTW, there is a paper(actually, there are more than 1 paper) by Christina Romer that shows how flawed Kuznets measurements of output fluctuations in the pre-WWI period are. This is a paper that George Segin & White cites in there paper on the failure of the Fed. All these Keynesian monetary & Fiscal interventions have only made the recessions worse – in depth & length. Besides, long term inflation rates were near or below zero prior to these Keyensians “improvements”. Keynesians have only managed to make things worse, not better. Christina Romer has the proof. I am sure that she doesn’t read her own proof that way.

Ken July 14, 2011 at 9:14 pm


“And I would add that I usually appreciate Daniel’s comments, particularly their tone in the face of criticism from the rest of you.”

I’d rather someone be an asshole and do the right thing, rather than be a really nice tyrant. Civility matters, but not as much as not destroying other people’s lives because you think they’re too stupid to live their lives properly.


Methinks1776 July 14, 2011 at 10:12 pm


Michael July 15, 2011 at 8:38 am

You can’t be civil and do the right thing?

Tim July 15, 2011 at 9:42 am


SweetLiberty July 15, 2011 at 10:44 am


Ken July 15, 2011 at 1:14 pm

You can, but when talking about the two, the former just isn’t that important.


Jack of Spades July 15, 2011 at 12:26 pm


Am I reading you correctly? Are you referring to Daniel as a would-be tyrant?

I disagree with Daniel on a variety of issues. He is comfortable with a larger role for government than I am. But…tyrant? Oh, come on! Caligula was a tyrant. Hitler and Stalin were tyrants. Pol Pot was a tyrant. If you can call someone with Daniel’s political views a “tyrant,” you are stretching the meaning of the word beyond belief.

I can’t help but be reminded of how Rothbard would refer to Milton Friedman as a “statist.” If someone with with Milton Friedman’s libertarian convictions is a “statist,” then the word has lost all meaning.

Anotherphil July 15, 2011 at 12:58 pm

If you can call someone with Daniel’s political views a “tyrant,” you are stretching the meaning of the word beyond belief.

Barack Obama, his executive orders, czars and compatriots are the first iteration of statism, its a “soft tyranny”, but it bears an unmistakable pedigree.

Anotherphil July 15, 2011 at 1:00 pm

Also, Obama’s not the first President with imperial tendencies..

Jack of Spades July 15, 2011 at 1:14 pm


First, political views are not a “grab all.” Because Danial has been generally supportive of various economic policies of the current President, it does not follow that he supports *every* policy. Secondly, I’ve visited Daniel’s blog off and on, and he does not hesitate to be critical of many of the President’s policies.

Ken July 15, 2011 at 1:17 pm

It stretches the imagination that someone who flaks for larger government, disdains personal choice when it can be made by bureaucrats, and who cheers for Obama, a man who believes that businesses and individual choice are evil and that politicians should be given more power isn’t a would-be tyrant.

This is exactly my point. Removing people’s liberty is tyrannical, even when the person doing it has a smile on his face and uses polite language.


jehu July 14, 2011 at 9:28 pm

Your argument amounts to a defense of the FDR State managed economy. You allow whatever stats you cite completely mask what took place during the war — namely, the total reorganization of economic activity under a fascist regime. Even as a communist, I am absolutely astonished by your flippant use of facts.

Have you ever read Stromberg, Rothbard or Carson? Or, do you spend you entire time in a theoretical bubble?

Don Boudreaux July 14, 2011 at 10:12 pm

What ARE you talking about?

Methinks1776 July 14, 2011 at 10:15 pm

If he knew he would not be a communist :)

Don Boudreaux July 14, 2011 at 10:24 pm

It’s as if he’s referring to some post written by lord-knows-who, but not the post above written by Russ – or any post that I’ve ever read written by Russ.

Greg Webb July 15, 2011 at 12:20 am

It is the typical incoherent ramblings of the looney left…

jehu July 14, 2011 at 10:50 pm

Well, I know enough to know if you replace the gold standard by ex nihilo dollars (as was done in 1933) “demand” no longer has any meaning, nor does “spending” — or prices for that matter. Now, your entire argument, which assumes prices, and, therefore, that demand and spending are real, is itself meaningless.

I know this despite being a communist. The question is, why you don’t know it.

vikingvista July 15, 2011 at 12:00 am

“The question is, why you don’t know it.”

Because it is so obviously false. In this place where you learned to become a communist, did you also learn that 2+2=5?

Urstoff July 15, 2011 at 1:24 am

Look at all the trouble having an obviously wrong theory of value is getting you in to.

jehu July 15, 2011 at 7:36 am

Really? Are you willing to actually deny empirical evidence? Okay, fine. When did the contraction phase of the Great Depression end? Was it not 1933-34? So, what happened that year? Did not FDR debase the dollar from gold? Coincidence? Pure chance? Of no theoretical significance?

If I were making facts up, your resistance to my conclusion might seem reasonable to an disinterested observer. But, these facts are available for everyone to examine. As are the arguments of Romer and Friedman.

vikingvista July 15, 2011 at 11:50 am


” “demand” no longer has any meaning, nor does “spending” — or prices for that matter”

Is undoubtedly one of the all time dumbest things ever uttered.

jehu July 14, 2011 at 10:42 pm

You are speaking of the post-WWII economy as if it were in some fashion a “normal” free market from which rules could be drawn relative to the scope of changes in government spending. However, that sort of economy disappeared during the Great Depression and the war, and was replaced with a highly managed fascist economy.

Now, how do you draw any lessons from an economy which is being fundamentally managed by the State? What of the responses to a sudden change in government spending can be said to be typical of a “free market”? Did the economy resemble, in any reasonable fashion, the economy as it existed before 1929, and especially 1933?

Remember, Samuelson’s “prediction” is not prediction, but an argument for the very sorts of interventionist state activity that has been typical of the entire post-war period.

Greg Webb July 14, 2011 at 10:59 pm

So Franklin Delano Roosevelt was a fascist. Does George know this?! I believe that George thinks FDR was a progressive/liberal. Who would have guessed that FDR was a fascist!

Greg Webb July 14, 2011 at 10:52 pm

Jehu is another statist who claims to be advocating for “the people”. How lucky can “the people” get?

jehu July 15, 2011 at 7:37 am

If you are going to claim I am a statist, you should at least be required to offer evidence to that effect.

brotio July 16, 2011 at 12:26 am

If you are going to claim I am a statist, you should at least be required to offer evidence to that effect.

Even as a communist, I am absolutely astonished by your flippant use of facts. – Jehu, July 14, 2011 at 9:28 pm

How’s that for evidence?

Greg Webb July 16, 2011 at 1:29 am

I’d say that is conclusive evidence. Thanks, brotio.

vikingvista July 16, 2011 at 3:50 am

Anarchocommunists believe that their notions of property abolition and coercive democracy do not entail a state. This is as amusing as their belief that scarcity can be abolished.

Daniel Kuehn July 15, 2011 at 6:27 am

The other thing I’d just add is that economics is a science of a complex system. We don’t judge results by forecasting. We forecast to the best of our ability for the same reason we forecast weather – if we can get close, it’s useful information.

But we don’t test meteorology by the accuracy of forecasts
We don’t test evolutionary biology by forecasting what’s going to evolve next
We don’t test abnormal psychology by forecasting when a psychopath has a psychotic break
We don’t test seismology by forecasting the exact dates and magnitudes of earthquakes

However, we do test physics by it’s ability to forsee certain planetary and celestial motions
We do test chemistry by it’s ability to predict what will happen when we mix two chemicals

What’s the difference between the top set and the bottom set? Roughly speaking, the top set are sciences of complex systems and we assess theories by their ability to explain past observations. The bottom set are sciences of more mechanical systems, so predictions are more feasible and we use them more often.

Where does economics lie?

I’m of the view that the economy is an emergent order that is formed by the decentralized interaction of billions of agents following roughly consistent patterns of behavior and interacting to form a complex whole – a complex order – that is characterized by emergent behaviors and laws. That’s just my view, and that view leads me to the conclusion that using forecasts to test theories is a bad way of testing theories in economics.

Josh S July 15, 2011 at 7:53 am

Meteorology makes testable, empirical claims. Weather models are constantly being revised in light of predictive failure. Keynesianism makes testable, empirical claims, too. They just happen to almost all be false except when they’re borderline tautologies.

The problem, as this “pent-up demand” example illustrates is that no matter how the theory fails, a Keynesian can always come up with an ex post facto reason for why the failure actually validates rather than contradicts the theory. This isn’t all that unlike Keynes himself, whose General Theory is little more than a series of hunches with the gaps filled in with hand-waving.

But that’s not science, and it’s certainly not empirical. That’s a prioristic reasoning, and all Keynesians really do is add more and more hypotheses and hunches rather than revisiting their old ones, much like Ptolemaists had to keep adding more epicycles every time the planets didn’t show up where they were supposed to.

Daniel Kuehn July 15, 2011 at 10:07 am

You can improve forecasting by looking at predictive failure and rerunning models with new data. Macroeconomic forecasters use essentially the same statistical tools to do this as meteorologists. But meteorologists don’t use forecasts to evaluate their theories. No science that is associated with forecasting evaluates their theories on the basis of forecast accuracy that I’m aware of.

re: ” Keynesianism makes testable, empirical claims, too. They just happen to almost all be false except when they’re borderline tautologies”

Could you provide a little more detail on what you mean by this. I was under the impression there was quite broad empirical literature that supports all sorts of Keynesian claims. What, precisely, do you have in mind here?

re: “The problem, as this “pent-up demand” example illustrates is that no matter how the theory fails, a Keynesian can always come up with an ex post facto reason for why the failure actually validates rather than contradicts the theory.”

This isn’t true at all – Russ asked if I wanted to respond to this post and I gave him a lot of thoughts – six reasons why strong post-war growth could have been expected ex ante, and some thoughts on why Samuelson and others didn’t remark on these reasons. If Russ doesn’t post my thoughts, I will on my blog and I’ll let you know.

SweetLiberty July 15, 2011 at 11:00 am

“But meteorologists don’t use forecasts to evaluate their theories.”

Of course they do! Without measuring the accuracy of forecasting versus models and theory, you would have nothing but a bunch of disparate weathermen making predictions based on rain dances, the toss of chicken bones, and myriad other superstitious mechanisms which could never be falsified if not evaluated against their forecasts.

“No science that is associated with forecasting evaluates their theories on the basis of forecast accuracy that I’m aware of.”

It is only when scientists empirically gauge the accuracy of various predictions against theory that that field can truly be considered a science. Otherwise, it is nothing more than opinion which cannot be falsified and, if that is the case, then why bother?

Daniel Kuehn July 15, 2011 at 11:10 am

Sweetliberty -
There’s a big difference between improving forecasting and testing theories.

re: “It is only when scientists empirically gauge the accuracy of various predictions against theory that that field can truly be considered a science”

Absolutely. I strongly agree.

But how do you test theory against prediction? In an extremely complex system forecasting the future is usually a poor way of doing it. We don’t prove evolutionary biology by forecasting what the next species will be. We do it by making predictions based on theory about what we’d expect to see in the data, and then we evaluate it with the data. But that’s different entirely from forecasting.

Josh S July 15, 2011 at 11:07 am

You can’t compare Keynesianism with meteorology. In meteorology, nearly *all* the physics that determine weather patterns are well-established; what is lacking is sufficient information to make accurate predictions (and we’ll always lack this information), hence the use of stochastic modeling. A failure of a meteorological model doesn’t cause a re-examination of, say, the conservation of energy, because it is itself a well-tested principle of physics.

Keynesianism doesn’t have that luxury. The assumptions of the system–animal spirits, cost-push inflation, homogenization of capital, the idea that monetary aggregates are sufficiently explanatory and causal in themselves, etc–are intrinsically untestable. They’re naked a priori hypotheses. And if the system fails (like when the 1970s featured both high unemployment *and* high inflation, or when repeated failures of QE bust the predictions of chartalists, or when none of the stimulus measures in Japan worked), you just come up with a new hypothesis, a new exception to the rule. But what you never, ever, ever, ever do is revisit the founding hypotheses and question them.

That’s why Keynesian economics is best compared to Ptolemaic astronomy.

re: Could you provide a little more detail on what you mean by this. I was under the impression there was quite broad empirical literature that supports all sorts of Keynesian claims. What, precisely, do you have in mind here?

The most important and most widely discredited empirical claims specific to Keynesianism are monetary. Driving interest rates to zero doesn’t lead to prosperity or nonscarce capital. High inflation and high unemployment can coexist quite comfortably. I know there are a couple chartalist hangers-on out there espousing MMT quackery, but when it comes to explaining inflation, Keynesians are about as relevant as Marxists these days.
And I recall that when Germany cut its budget a couple years ago, Keynesian economists predicted ruin. Didn’t happen.

Metryll July 16, 2011 at 8:19 pm

Re Keynes with Ptolemaic astronomy :

If one wants to take a scientific analogy then he would better use heuristic. Pratically what is an heuristic ? A rationalized bet. If an heuristic is correctly set up, it can save a lot of calculus. It’s precisely how work an A* for example.

Real genius of Keynes was to recognize that his era mathematics (and our btw) were not up to describe human behavior. Instead he used what he had at hand : psychology.

Before jumping at me with left/pink/socialist think of it :

A: What branch of maths deal with human emotions ?
Q : None.

Austrian seem to state that economy complexity is such that no one can understand it (please correct me if I’m wrong about Austrians here).

Neo classical think that economy must be mathematically defined.

Both are wrong. Inabilty to use mathematical or use of mathematical formalism with an reduction of reality (Homo Economicus) prove nothing but that current mathematical progress is not up with complexity of problem to be solved.

This said while economy is a complex activity, it’s certainly not the sole, the other being humanity great passion : war.

Never asked yourself why military use Art of War and not Science of War ?

Eduardo July 15, 2011 at 11:05 am

“much like Ptolemaists had to keep adding more epicycles every time the planets didn’t show up where they were supposed to.”
So what is the old economic theory economists should revisit ? Yours? Which is it…?

Regarding the Russ Robert post, as Daniel Kuhen said, there are a lot of explanations that are consistent with Keynes’ views.

Josh S July 15, 2011 at 1:05 pm

Revisit the basic monetary assumptions, rework through what money is using marginal utility and subjective theory of value. Understand that the whole of economic theory rests on money and interest–since Keynes got money and interest so wrong, the entire theory needs to be discarded..

Methinks1776 July 15, 2011 at 10:54 am

So where does your planning and social engineering come into this complex system? You have declared that you are a fan of both and your politics support this view.

My view is pretty much the same as your stated view (except I’m not sure exactly what you mean by “roughly consistent behaviour”) and my view leads me to the conclusion that I’m barely capable of running my own business, let alone engineering entire societies. I’m curious why your similar view leads you to an entirely different conclusion about your abilities.

Daniel Kuehn July 15, 2011 at 11:12 am

You have a long history of misrepresenting what I think my abilities are, so it’s not wonder you’re confused.

Josh S July 15, 2011 at 1:06 pm

Here’s the thing, Daniel…you’re a pretty smart guy, overall. But you don’t seem to think you have the ability to manage or engineer the economy. What makes you think that elected politicians are better than you? For my money, you are about ten times smarter than almost anyone on Capitol Hill, with few exceptions. If you don’t trust yourself to run things, then why do you expect me to trust Bawney Fwank to run things?

Methinks1776 July 15, 2011 at 1:58 pm

I can’t get to the specific comment anymore because it’s locked up in Disqus, but you did indeed in the last few months flat out say that you support some social engineering – in those words.. I found your frankness shocking, frankly.

I remind you that this is not the first time you’ve denied saying something you did actually say. It’s difficult to know which face you’re presenting at any given time.

Daniel Kuehn July 15, 2011 at 2:10 pm

Depending on what you call “social engineering” I definitely support some social engineering.

When did I deny that?

Your ability to make absurd claims and get defense remains breathtaking, Methinks. All I said is that you misrepresent what I think my abilities are. You often act as if I think I can (or tothers can) engineer the whole economy.

Daniel Kuehn July 15, 2011 at 2:10 pm

*get defensive

Methinks1776 July 15, 2011 at 4:26 pm

I can’t control how you interpret how I “act” with regard to your or Gosplan’s ability to engineer an economy. I’ve never accused you of wanting to either. Talk about misrepresenting views.

Yes, I am talking specifically about this comment:

“I definitely support some social engineering.”

How does that jive with what you wrote above? As importantly, how do you define “social engineering”? In your interpretation may be the key. I don’t know.

You give a lot of lip service to markets and liberty but seem to favour a lot – if not complete – state intervention to create a society more to your liking. That which you don’t openly favour, you justify or poo-poo. That is my impression of how you “act”.

SweetLiberty July 15, 2011 at 11:37 am

“We don’t prove evolutionary biology by forecasting what the next species will be.”

Actually, they do. Evolutionary biologists often use the short lived fruit fly to test their forecasts of manipulating longevity across generations. Selective breeding and genetic engineering are used to make new plants and animals. These experiments are not random “let’s throw these things together and see what happens”, but purposefully guided, goal oriented endeavors based on current understanding and forecasting.

Again, without forecasting and evaluation, all you have is an irreconcilable difference of opinion.

But perhaps that’s all economics truly is.

Daniel Kuehn July 15, 2011 at 11:48 am

Right but controlled experimentation and forecasting are two entirely different things!

If they’re carefully controlling it, it’s NOT a forecast, it’s an experiment!

Experimental empiricism is great.
Forecasting is great.
Non-experimental empiricism is great.

But let’s not mush the three together or assert that one is required for “science”. Science is indeed empirical – it’s about testing theories against data. But the nature of the empiricism is heavily dependent on the nature of the subject of study.

SweetLiberty July 15, 2011 at 12:45 pm

Perhaps are disagreement is one of semantics. Someone who proffers a hypothesis for a future outcome based upon an event or events whether controlled or otherwise is making a prediction or calculation in advance. This is, by the definition, a forecast. It matters not whether the hypothesis is based upon a controlled experiment or observation of events out of the forecaster’s control – both scenarios make a prediction, a forecast. A scientific forecast is falsifiable, which is to say that IF the forecast turns out to be false, THEN the hypothesis is somehow in error.

The problem with social sciences is that forecasts (or predictions if you prefer) are made quite often, but due to the complexity of variables, the data can be interpreted to support most any pre-position (i.e., the stimulus WOULD have fixed the economy if only it would have been larger according to Krugman, et al.)

Economics has the curious distinction of proving every hypothesis correct no matter who makes it.

The non-Keynesian economists on this blog represent a minority of economists, yet no Keynesian has produced any evidence which Russ or Don will accept as proof. The same is clearly true going the other way, as witnessed in the main body of this post.

Samuelson, a Keynesian, made a prediction. It turned out to be false. Yet this does not sway Keynesians. Okay, if this particular forecast which was falsified does not sway Keynesians, what would? About two years ago Obama promised that with his Keynesian stimulus, unemployment would not exceed 9%. Oops. But this is not enough to falsify Keynesian economics. The Fed has kept interests rates at rock bottom for several years in order to “stabilize” the economy. Hasn’t worked yet. Banks and too-big-to-fail industries have been bailed out, homeowner susbsidies and cash-for-clunkers have all been tried, but still the economy is in the toilet. Yet this is not enough to sway a Keynesian.

If no experiment, no artificial manipulation, no data, no observation based upon forecast can falsify a Keynesian’s point of view, then the hypothesis is not scientific but ideological.

What on earth would constitute a falsifiable forecast between the likes of Don and Russ vs. Krugman and Geithner? If economists with opposing hypotheses cannot agree on what outcome would disprove one theory over another, then all you have is blind faith. This is not science – not even social science.

SweetLiberty July 15, 2011 at 12:46 pm

EDIT: Perhaps *our* disagreement…

Jim July 15, 2011 at 2:37 pm

Well written sweet.

Economists currently have it both ways: they have much power but also claim extreme fallibility. Exhibit A: The Fed.

vikingvista July 15, 2011 at 5:18 pm

No observation can be neat enough to falsify any economic theory. Fortunately reason provides us with enough knowledge to make economics even possible. Unfortunately, keynesiacs consider that essential knowledge to be unnecessary for their macroeconomics.

So, yes, nothing imaginable can falsify keynesianism to a keynesiac.

Henri Hein July 15, 2011 at 1:29 pm

“We don’t judge results by forecasting.”

Since the issues at hand are used to justify massive interventions in the economy, the ability to forecast takes on significance quite more severe than your other examples.

Earlier in the week I watched a whole series of news clips on, almost every single one lamenting why congress doesn’t “get it” and are even considering austerity now. If Keynesians were so wrong in 1946, why should I believe the alarmism over austerity in 2011?

Jim July 15, 2011 at 2:07 pm

The other thing I’d just add is that economics is a science of a complex system. We don’t judge results by forecasting. We forecast to the best of our ability for the same reason we forecast weather – if we can get close, it’s useful information.

Picture Greenspan and later Bernanke as late as 2008, pontificating on the health of the economy in the face of fiscal disaster even as interest rates and Fed monetary policy abetted the size of the crisis as opposed to containing it.

Picture the hundreds of millions of millions the Fed spends on modeling you imply should not be used as a bar for their effectiveness.

Keynes aside, the problem with economics may not be that they can not forecast (they clearly can not using current economic assumptions and models). The problem is that so many economists have so much power despite their failure.

By your own argument, economists should have far less to say about monetary policy, and for the Fed, instead of markets, to control interest rates is absolutely insane.

A market driven system would in the aftermath of the Great Recession be overhauled, and heads would roll. The Fed trundles on. The failure of the Fed, perhaps the largest employer of economists in the world, in the financial crisis demands a long and fundamental appraisal of the whole discipline. So does my quotation of your first paragraph. What are the chances of that happening, given bureaucratic inertia?

I submit your quotation is yet another illustrative example of smaller less powerful government, lest men caught up in the hubris of their ideas spread the pain of their mistakes across an ever larger plane.

Daniel Kuehn July 15, 2011 at 10:44 am

Links never seem to make it through the comment filter here, but if you google the Social Democracy for the 21st Century blog, it has a good response up to this too.

Samuelson might have feared post-war stagnation, but apparently Keynes didn’t.

carlsoane July 15, 2011 at 1:28 pm

A couple of extracts caught my eye.

“There was the massive surge in consumption after 1945 using accumulated savings. ”
“In other words, Americans now had the security of welfare programs that allowed them to free up more of their income in spending.”

We still have the security of welfare programs, in fact they are much larger than they were in 1945. But we’ve stopped spending. Is that because we no longer have the accumulated savings we had in 1945?

It sounds a bit like ABCT to me.

Slappy McFee July 15, 2011 at 11:42 am

I would like to second Russ’ apprieciation of the contribution that Daniel Kuehn makes to cafehayek. His responses are much more reasonable and thought out that those that shall remain nameless.

I may not agree or understand Mr Kuehn but at least he remains respectful in the lions den.

SweetLiberty July 15, 2011 at 11:43 am


Henri Hein July 15, 2011 at 1:19 pm

Hmm, I meant to reply up here. Much as I disagree with Daniel, I think his intellectual curiousity is real.

Daniel Kuehn July 15, 2011 at 11:49 am

I find that lions are dangerous to piss off.

Henri Hein July 15, 2011 at 1:16 pm

Me too.

Jim July 15, 2011 at 2:34 pm

Surely the economy is a complex system. FWIW, since PSST seems to me to be mostly a rewording of complexity theory, I do not understand why Mr. Kling invents a new language, rather than incorporating his work with complexity research. Surely Hayek’s refocus on complexity from economics was anything but a coincidence.

Regardless, complexity and PSST imply that relationships and inter-causality and substitution of actors are as important as their functions in any economic models of a complex system. For it is those relationships that define utilization (or full employment) as much as original function.

In that light, ‘aggregate’ supply or demand are meaningless concepts. So is the idea of equilibrium. That is, the system is continually evolving to adapt, sometimes faster than others. But it is always evolving and surviving.

If a complex system is not producing to its potential, however that is defined (many would say full employment), the most important question is not how to affect total output (aggregate demand). The most important question is how to destroy any impediments that keep it from evolving faster.

Some may respond that a complex system might find an ‘equilibrium’ that falls short of full employment. My response is that there is no such thing as a complex system in equilibrium, and second, in a global marketplace with rising standards of living, there must be impediments to full utilization or they would not exist very long.

Conclusion: Post WWII prosperity was at least partially achieved by the removal of regulation impediments, not just reduced government spending. That is surely the case today.

juan carlos vera July 15, 2011 at 3:45 pm

I have repeated several times, here in Cafe, that the Keynesian problem is not an economic problem. Mankind has trouble to meet their needs for goods, all of which are scarce, forcing its members to maintain a constant battle against a shortage, so that prosperity expresses small advances in that battle. A theory of this problem is an economic theory. But Keynes did not theorizes about this problem. Keynes, and his fanatical followers, says that the obstacle to overcome is not scarcity but abundance because the goods are here but men can not afford them. Then for Keynes the problem is solved if men have the means to acquire these goods. Clearly, Keynes’s theory “is not” an economic theory. Or, as Robert say, the Keynesian statement is tautological.. This fact should limit the debate between Keynesian and Austrian, because its theories are theories of two different problems and therefore they are not consistently comparable…

Metryll July 16, 2011 at 10:56 pm

Nice critic of Marxism, totally irrelevant for Keynes.

It’s a common and quite widely false idea that Keynes theory is a demand one. Demand only is Marx, Keynes is supply/demand. Basically Keynes take clasical frame but refute Say law. Demand is more covered not by preference but as result of neglect by classical.

About Keynes nature as economic theory, it’s that and only that .Contrary to Hayek and Marx, one will find very few philosphy or social analysys. The sole noteworthy point being that Keyenesian economy work only in demcracies.

Ironist July 18, 2011 at 3:00 pm

Russ, please forgive a non-economist for asking some naive questions here. What was the state of private savings immediately after WWII? If many people were essentially forced to buy war bonds and forgo consumption, might the level of savings after the war have been significantly higher than before? Also, wasn’t household formation very high after the war (speaking as a 1946 neonate myself)? If savings and household formation were high, this should result in higher demand. Again, if all this is so, I’d expect rising prices for many consumer goods and services, followed quickly by rising production/supply and employment. Was Samuelson thinking that the economy would not function again unless Washington ordered it to?

Previous post:

Next post: