Did War Based Fiscal Stimulus End the Great Depression?

Paul Krugman believes it is a slam dunk.  It's probably a bad idea to argue with a Nobel prize winner, but my understanding was the science was not quite as settled on that point as Prof. Krugman would have us believe.  Below the fold I'll cite a few pieces of research on the topic.

First piece - Christina Romer, What Ended the Great Depression, (The Journal of Economic History, 1992).  Abstract:

This paper examines the role of aggregate-demand stimulus in ending the Great Depression. Plausible estimates of the effects of fiscal and monetary changes indicate that nearly all the observed recovery of the U.S. economy prior to 1942 was due to monetary expansion. A huge gold inflow in the mid- and late 1930s swelled the money stock and stimulated the economy by lowering real interest rates and encouraging investment spending and purchases of durable goods. That monetary developments were crucial to the recovery implies that self-correction played little role in the growth of real output between 1933 and 1942.

A second piece in the The Journal of Economic History gave a dissenting view - J.R. Vernon, World War II Fiscal Policies and the End of the Great Depression, (The Journal of Economic History, 1994).  Abstract:

The United States economy completed its recovery from the Great Depression in 1942, restoring full-employment output in that year after 12 years of below-fullemployment performance. Fiscal policies were not the most important factor in the 1933 through 1940 phase of the recovery, but they became the most important factor after 1940, when the recovery was less than half-complete. World War II fiscal policies were, then, instrumental in the overall restoration of full-employment performance.

Of course, Paul Krugman is aware of this literature.  Paul R. Krugman, Kathryn M. Dominquez, Kenneth Rogoff - It's Baaack: Japan's Slump and the Return of the Liquidity Trap (Brookings Papers on Economic Activity, 1998):

It seems to be part of the folk wisdom in macroeconomics that thisis in fact how the Great Depression came to an end: the massive onetime fiscal jolt from the war pushed the economy into a more favorable equilibrium. However, Christina Romer contends that most of the output gap created during 1929-33 had been eliminated before there was any significant fiscal stimulus. She argues that the main explanation of that expansion was a sharp decline in real interest rates, which she attributes to monetary policy (although most of the decline in her estimate of the real interest rate is actually due to changes in the inflation rate rather than the nominal interest rate). Indeed, Romer estimates that for most of the recovery period ex ante real rates were sharply negative, ranging between -5 and - 10 percent.

My point is that the end of the Depression, which is the usual, indeed perhaps the sole, motivating example for the view that a one-time fiscal stimulus can produce sustained recovery, does not actually appear to fit the story line too well. Much, though by no means all, of the recovery from that particular liquidity trap seems to have depended on inflation expectations that made real interest rates substantially negative.

Emphasis added.

That final piece is 12 years old, so there may be some research since then that has 'sealed the deal' on the theory that World War II fiscal stimulus 'ended the depression'.  And of course, there is the point that 'by no means all' of the recovery can be attributed to monetary policy.  But I am not aware of any post-1998 studies that have put a great deal of weight on fiscal stimulus as driving the U.S. economy out of the Great Depression.  If there are any macroeconomists in the house that have followed that line of study closer than I have, I would love to hear from them!

Edited to Add: I found links online for all 3 of the papers, so I've added those.  

Comments

[tyronen]

How can you explain the US having unemployment of 14 percent in 1941, and 1 percent in 1942, without the war?

[Mike Moffatt]

"How can you explain the US having unemployment of 14 percent in 1941, and 1 percent in 1942, without the war?"

Well, for one thing I think you mean 1940 for the first number and 1944 for the second.

[Mike Moffatt]

Broadly speaking, though, I would recommend Stephen Gordon's piece in today's Economy Lab:

http://www.theglobeandmail.com/report-on-business/economy/economy-lab/the-economists/economic-policy-is-about-more-than-just-jobs/article1771293/

But as to the specifics of the unemployment rate:

"The number of unemployed workers declined by 7,050,000 between 1940 and 1943, but the number in military service rose by 8,590,000. The reduction in unemployment can be explained by the draft, not by the economic recovery."

http://www.econlib.org/library/Enc/GreatDepression.html

[JP Koning]

Robert Higgs has an alternate view that the economy didn't get out of the depression till 1946.

http://www.econtalk.org/archives/2008/12/higgs_on_the_gr.html

[Kevin Donoghue]

Krugman was asked for an example of a country that spent its way out of a depressed economy. At least that’s what he says he was asked. His blue line is total debt, public plus private. So his is not really a post about fiscal policy. It’s about whether you can spend yourself rich, as the Keynes-bashers so often phrase it.

[Josh]

I think that Romer's paper is the best look at the period that I have seen. She doesn't necessarily suggest that fiscal policy didn't work, but rather wasn't really tried. In fact, she attributes any effects of the war to be monetary-based as the financing of the war in Europe led to gold inflows in the United States.

[Determinant]

How exactly was WWII not a fiscal stimulus when US Government publicly-held debt went from 45% of GDP in 1940 to 110% in 1945?

[tyronen]

So the US government provided jobs to 8.6 million people through military recruitment. These people were paid salaries with borrowed money. Plus there were all the other costs of the war (munitions, equipment, transport etc). Why does that not count as a fiscal stimulus?

Krugman concedes that prior to the war, monetary policy helped more than fiscal, but says that during the war, fiscal policy had the most impact.

More important is what lesson we draw from this time. Obviously in peacetime we won't be able to do a stimulus on the scale of the war. But at the very least we shouldn't be engaging in premature austerity, as the UK, Ireland, and Greece are doing. Nor should a US president who tried to do the right thing be punished because the Senate fatally weakened their original proposal.

US Republicans believe that tax cuts *always* stimulate the economy, spending increases *never* do, and "liquidity trap" is some elitist liberal jargon that means nothing. These notions are what Krugman devotes a large amount of blog posts trying to refute.

[Josh]

Mike,

You might be interested in this new NBER working paper entitled, "U.S. Monetary and Fiscal Policy in the 1930s":

http://papers.nber.org/papers/w16477

Here is the abstract:

The paper provides a survey of fiscal and monetary policies during the 1930s under the Hoover and Roosevelt Administrations and how they influenced the policies during the recent Great Recession. The discussion of the causal impacts of monetary policy focuses on papers written in the last decade and the findings of scholars using dynamic structural general equilibrium modeling. The discussion of fiscal policy shows why economists do not see the New Deal as a Keynesian stimulus, describes the significant shift toward excise taxation during the 1930s, and surveys estimates of the impact of federal spending on local economies. The paper concludes with discussion of the lessons for the present from 1930s monetary and fiscal policy.

[Too Much Fed]

"Did War Based Fiscal Stimulus End the Great Depression?"

Did war based fiscal spending with currency denominated debt destroy enough supply worldwide to end the Great Depression? That means creating more medium of exchange from the demand deposits created from currency denominated debt started working again?

[Too Much Fed]

You might be interested in this one too:

http://krugman.blogs.nytimes.com/2010/10/25/sam-janet-and-fiscal-policy/

[Too Much Fed]

"My point is that the end of the Depression, which is the usual, indeed perhaps the sole, motivating example for the view that a one-time fiscal stimulus can produce sustained recovery, does not actually appear to fit the story line too well."

How about Japan?

[Adam]

So does Krugman really believe that if the US were to send, say, 1 million troops to each of Iraq and Afghanistan (with all the equipment that would require) then the economy would recover and happy days would be here again? Or let's say that in order to avoid unnecessary deaths, the US government simply pretended that it was fighting a massive war and purchased the necessary supplies (then, say, shot them all into space) that such a policy would be economically beneficial?

[Too Much Fed]

"It's probably a bad idea to argue with a Nobel prize winner,"

Well I think it is if he keeps trying to proclaim the solution to too much currency denominated debt is more currency denominated debt, specifically the solution to too much lower and middle class currency denominated debt owed to the rich is more gov't currency denominated debt owed to the rich.

[Too Much Fed]

"However, Christina Romer contends that most of the output gap created during 1929-33 had been eliminated before there was any significant fiscal stimulus. She argues that the main explanation of that expansion was a sharp decline in real interest rates, which she attributes to monetary policy (although most of the decline in her estimate of the real interest rate is actually due to changes in the inflation rate rather than the nominal interest rate)."

Did supply fall?

[Lord]

Not sure what your argument is, that we didn't produce for combatants, that gold inflows were not due to the war, that the monetary authority failed to sterilize the inflows, that draftees are not employed, that government spending shouldn't be counted as output, or what. I would not say we had full employment/output prior to 41 and it seems very unlikely we would have reached that without the war since we had already gone a decade without it, or that we would have had the post war recovery without it, the destruction, the redistribution, or even the relaxed monetary policy. This is not to say the war was necessary for recovery, only necessary for policymakers to do the right thing in the first place.

[Declan]

Well, the question of what policies get us out of a depression is certainly the question of the moment.

The Krugman post you link to, as Kevin says, doesn't really say much about the source of additional spending (printing money or government borrowing), just that we need to spend more. Although certainly, it seems that Krugman places more emphasis on the government borrowing side of things than on the printing money side.

My own 2 cents is that printing money needs to be a big part of the solution - substituting government debt for private debt is a good stalling tactic but won't bring down the overall debt levels which is what needs to happen.

But, like Lord says (if you can't agree with the Nobel prize winner, I guess the Lord is the next best thing), without a war, I'm not sure there will be the political will to undertake printing money OR massive fiscal stimulus. Neither has happened in Japan and they've been at this a long time.

[Sina]

This is the first time I've come across a post on this blog that I felt is so so so deeply flawed -- well not flawed, just unfair.

Just to re-iterate, Lord's comment above:

"Not sure what your argument is, that we didn't produce for combatants, that gold inflows were not due to the war, that the monetary authority failed to sterilize the inflows, that draftees are not employed, that government spending shouldn't be counted as output, or what."

I'll add that the lesson from WW2 is that we need stimulus, regardless of where it comes from.

Real interest rates and inflation had a lot to do with the WW2 stimulus. Fine. So let's just print money then. But then why not use that money towards soemthing productive once we print it, instead of just dropping it from helicoptors?

[Just visiting from Macleans]

Not sure what your argument is

That is the point.

[Alan]

Mike

Any discussion of the effectiveness of f/p (size of multipliers, etc) needs to also consider m/p - the central bank typically moves last. So, I think Krugman's views aren't contradictory with the Great Depression evidence.

In the current crisis, Krugman has diagnosed central bankers as being too conservative to re-inflate the economy using unconventional m/p, but willing to allow reflation through f/p. You can argue with the validity of this (and several people do, notably Scott Sumner), but in this light Krugman's argument is logically consistent.

Also, the evidence from Eichengreen on the Depression further "seals the deal" for m/p vs f/p -- Eichengreen showed that countries that left the gold standard earlier and faster rebounded quicker than those countries that remained on the standard (these countries not only had tight m/p, but also imposed various trade barriers and capital controls, exacerbating their situation)
http://www.voxeu.org/index.php?q=node/3280

Alan

[Arrrrrrrggggggggghhhhhhhhh]

Unemployment wasn't 14% in 1941, but 9% at the start and 6% by the end. "Mainstream" stats like to exclude government workers, but they should be counted.

The depression was over in 1941. Not 1942.

[Colin R Fraser]

Not sure what your argument is

I believe the argument goes something like this:

If we define prosperity as the degree to which we may consume what we want to consume when we want to consume it, WWII did very little to enhance our prosperity. Yes, standard economic measures moved in directions that we would normally equate with higher prosperity, but because during WWII the US ran a command economy, standard economic measures were irrelevant as measures of welfare. For instance, how are we supposed to reliably calculate GDP, the market value of all final goods and services, when the things that we produced, like tanks and ammunitions, have no market value? How are we supposed to calculate inflation when the government is controlling all prices?

Even by generous estimates, real personal consumption expenditures rose by 4% between 1941 and 1944 )and population rose by about that same amount) and by probably more realistic estimates, real personal consumption expenditures decreased. If people are consuming less, how can you argue that they are more prosperous?

Here is where I am getting my stats:
http://www.independent.org/newsroom/article.asp?id=138

[Jim Rootham]

Stats may not lie, but liars can use stats.

There is no conceivable way of telling the story of the 30's and 40's where most people thought they were better off in the 30's (well, being in combat probably wasn't better, but this is the US and that was a small fraction of the population).

[Colin R Fraser]

In what way were they better off?

[Colin R Fraser]

*In what way were they better off in the 40s than they were in the 30s, I mean.

[RSJ]

But these reviews are model dependent and you are citing Romer/Krugman looking back at the great depression through a certain theoretical lense.

What about the accuracy of the forward looking comments from that era?

In 1999, Krugman wrote in "thinking about Liquidity Trap"

"The basic point, once again, is that a credible commitment to expand the future money supply, perhaps via an inflation target, will be expansionary even in a liquidity trap. "

And then also in 1999, Woodford writes, in "How Should Monetary Policy Be Conducted in an Era of Price Stability?"

"What remains true, as stressed by Krugman (1998), is that a central bank in such cir- cumstances [e.g. liquidity trap] would be able to stimulate aggregate demand if it could credibly commit itself to a more expansionary future policy. "

That was prior to the BoJ's highly public and impotent QE efforts. 1998 was also before 11 years of zero interest rates in Japan, accompanied with secularly accelerating deflation. I think ZIRP began right around the time that Krugman said that and they still have ZIRP, with a lower NGDP today than when either of these essays were written.

Here's a VAR study from 2010:

"[A] new quarterly data set for the interwar period that allows development of a VAR model of the U. S. economy over the period 1920‐41. The quarterly data facilitate an estimate of the growth rate and level of potential real GDP, and this in turn allows us to date the end of the Great Depression as the period between 1939:Q1, when the output gap was 21.9 percent, and 1941:Q4, when the output gap was ‐1.3 percent. By using VAR dynamic forecasting and examining the effect of innovations to each variable individually, this paper’s baseline result is that 89.1 percent of the 1939:Q1‐1941:Q4 recovery can be attributed to fiscal policy innovations, 34.1 percent to monetary policy innovations and the remaining ‐23.2 percent to the combined effect of the basic VAR dynamic forecast and innovations in non‐government components of GDP (N). The paper attributes the negative innovations of N in the second half of 1941 to capacity constraints."

http://www.iga.ucdavis.edu/Research/All-UC/conferences/spring-2010/Gordon%20paper.PDF

[Jim Rootham]

They had work, they had incomes, they had places to live, they had food.
That's my understanding of the stories comparing the 30s to the 40s.
I'd say that defines better off.
The aggregate might well not have moved that much, but the distribution almost certainly did.

[Colin R Fraser]

16 million+ of them (in the US) were sent to another country to fight in a war. Some of the ones who were not sent to another country to fight a war were sent to jail for not going to another country to fight a war. For those people, the 40s sounds terrible.

Price controls on goods degraded the quality of goods. Aside from legislated restrictions on consumption, consumers were urged to "Use It All; Wear It Out; Make It Do; or Go Without!". What good is having a job if you can't use the money it pays you to do anything?

[Canadian Drugs]

Although certainly, it seems that places more emphasis on the government borrowing side of things than on the printing money side.These all are great to know about it.

[Jim Rootham]

@Colin Did you arrive here from an alternate history? I didn't live through those periods but my parents did. The history and stories set in those periods are quite clear. Having enough to eat is worth something. England killed off rickets during rationing.

[Determinant]

Might I also add that while we talk about war-based stimulus and the consequent post-war prosperity like they are obvious, they weren't at all at the time. In fact conventional wisdom in 1947 went directly opposite. The thought was that since the depression of the 1820's followed the end of the Napoleonic Wars, the Depression of 1873 followed the US Civil War, and the Great Depression followed WWI, there WAS going to be a lapse back into a depressed economy imminently. The stock valuations in the US in 1947/48 were ridiculously low (in hindsight) due to this belief.

[Colin R Fraser]

What part of what I'm saying isn't accurate? I don't disagree that having enough to eat is worth something, but I find it difficult to accept the argument that there was "wartime prosperity"—not starving to death is hardly prosperity—especially when you consider the things that I've mentioned.

[Jim Rootham]

If you confine the argument to luxury goods available to the upper crust, yes, your description is apropos. Enough to eat and the ability to save sure looks like prosperity to the larger chunk of the population that didn't have that in in the 30's.

The difference after WWII was government policy. The GI Bill especially. Since that was a new thing there was no guarantee it was going to work. Hence, low stock prices.

[Lord]

We could have unilaterally surrendered and not fought the war, but I doubt that would have provided prosperity either. The war was too great and did produce rationing, shortages, death, and destruction, but who is proposing anything near that level of spending or that use of it? Would we have been better off if the war never happened? In the short term, probably so, in the long term, probably not, because we probably would have never done what was necessary to return to trend.

[Colin R Fraser]

The question is whether government spending was the thing that ended the Great Depression. Government policy after the war is kind of irrelevant. The standard argument in favor of this motion, as I understand it, is that government spending was high enough to create noticeable multiplier effects on aggregate demand—as one commenter put it, we spent ourselves rich. Prices went up, real output went up, and unemployment was practically zero, and so, the argument goes, the Great Depression was over.

What I'm saying, and what I think the article is saying, is that it's not as clear cut as that. The biggest reason that it is not clear cut is that all macroeconomic measures were broken during the war. The unemployment rate can't really be trusted, because at its core the unemployment rate is a measure of how the market is able to provide prosperity, and the market was not providing prosperity: government and firms were forced to hire. GDP numbers can't really be trusted, since the government set all the prices of goods, and military spending can't really be said to have a market value.

What is true is that investment spending decreased during the war, and it wouldn't be controversial to say that investment spending is a large source of prosperity. Another measure of prosperity, consumer spending, stayed constant by generous estimates and went down by others.

The question is whether the US and Canada would have pulled out of the Great Depression at the same time or sooner without WWII, and I believe that it's silly to argue that they would not.

[Jim Rootham]

Um, no the unemployment rate is how many people are looking for work, whether they are employed by the government or private industry makes no never mind. Again, GDP is what GDP is, command economy or not.

You are espousing a religious belief that only private actors matter in the economy. The economy is all productive behaviour, (and broad in some definitions of productive at that). You may wish to argue that, in general, private is better than public, and I will disagree. But when you argue that public does not exist, you are being dishonest.

[Lord]

Why would you believe that when it never occurred during the preceding decade? Do you believe government was crowding out the private sector during 15% unemployment? The economy never returned to trend until the war (41 was well into the effects of the war) when it exceeded it and after the war when it maintained it. While I think the economy could have recovered without the war, there was no indication appropriate policies would have been adopted without it, or at the speed it did at least (there was some groping for policies that worked but rarely the will the carry them to the appropriate scale). No, it could easily have been another decade before the depression ended without the war, though the war was to such a large degree caused by the depression, disentangling them is really impossible.

[Colin R Fraser]

GDP is the market price of all final goods produced in a year, no? What is the market price of a tank?

[Determinant]

Surely the profits of war equipment manufacturers matter?

[Mike Brock]

Why do none of these arguments about the causes and breadth of the Great Depression not consider the structural shocks to the economy that were created by Smoot-Hawley which sparked a global trade war, reducing global trade by 60%, etc.

These trade barriers were not removed until 1945.

Yet, economists banter on about the role stimulus and expansionary monetary policy play in fixing market shortcomings -- as if one of the most draconian trade interventions wasn't somehow connected to the length and depth of the economic depression.

[Jim Rootham]

@Colin

GDP is the sum of the transaction price of goods. The transaction price of a tank is the price set by the successful bidder on the tank production contract. Or possibly the price set by authorities. How it is set does not change its contribution to GDP.

[Roland]

One thing about a major war is that you can tax the rich very heavily, without them being able to complain very much. Capital mobility is quite limited during a major war. And "going Galt" seems to lose its appeal under those circumstances.

But the North American situation during the world wars was rather anomalous: a major war fought with mass armies, demanding full labour mobilization, but with no direct threat to home territory, and only very minor demographic cost.

There's no example there for the future. Because even major wars, nowadays, if they ever happened, would never demand that kind of labour mobilization.

[Doc Merlin]

You didn't mention Barro's paper on this. He found the multiplier for this spending was less than 1, which would mean that the spending didn't end the depression.

[Peter T]

I am sorry, but the initial post is simply silly. The war was not about "fiscal stimulus". That was a technical side-show, given general acceptance that the war had to be fought (the UK knew in 1939 that two years of war would bankrupt it, yet still went ahead - there was no other choice for the UK). After May 1941 there was a strong sense in which most Americans (not all) accepted that there was no choice for the US either.

The effects of the war in the US radically restructured the distribution of income, attitudes about classes, race relations, the skills of the much-enlarged working population (most people in modern armed forces have technical jobs), attitudes to government and much else. What came out of the war was a more cohesive society, a new set of managers, a vast new set of production facilities and a skilled workforce with a strong sense of what it was entitled to politically and economically.

The movement of bits of paper was an interesting by-product. It was not the real thing.

[Blikktheterrible]

So, the impression that I get from all of this is that we're seriously limiting the discussion when we discuss the relative contributions of monetary and fiscal policy. In WWII, the economy was affected by forces going way beyond those policy categories. Describing WWII as fiscal policy is a huge oversimplification. I think we should focus on what new categories of economic forces we need to consider and which of those categories are relevant in the contemporary world.