Saturday, October 3, 2009


Why Military Spending Didn't Get Us Out of the Great Depression

In a thread at ThinkMarkets, the ubiquitous question of WWII spending and the Great Depression came up. I realized that I don't think I've ever succinctly summarized Bob Higgs' research on this point. (I explain all of this in great detail in my book.)

OK so here is what the limited-government, anti-Keynesian is up against:

Well gee whiz, it sure looks like military spending got the US out of the Great Depression, doesn't it? Sure, common sense tells you that directing scarce resources into tanks and bombers, and then shipping them overseas to get blown to smithereens, can't possibly help the domestic economy. But the figures above don't lie, right?

In a series of really first-class papers, Bob Higgs points out the problems with the conventional stats. First of all, GDP includes government expenditures. If we just look at the private component of GDP, we see a different story. (I can't find the data right now, but I'm virtually certain that private sector GDP fell throughout the war years, and jumped way up in 1946.) So already we run into a methodological issue: If the government takes, say, $1 billion out of the hands of private citizens, and spends it on "$1 billion worth" of tanks and bombers, is that really just a change in the composition of total output? Isn't there an objective sense in which $1 billion spent by millions of consumers corresponds to more "total output" than $1 billion spent by military procurement officers in a non-competitive process?

Yet it gets worse. A lot of the wartime expenditures were financed (perhaps indirectly) through the printing press. Just look at how much they jacked up the monetary base during the war; it's shocking to anyone except Ben Bernanke.

Now normally, if the Fed printed up a bunch of money that allowed the government to spend billions (in nominal terms) on military equipment, that wouldn't boost real GDP. Sure, nominal GDP would go up, since you tabulate it by counting up the actual dollar expenditures by everyone, including the government. But because of all the money printing, the prices of milk, eggs, bread, gasoline, and so forth would skyrocket. So because the CPI (or GDP deflator) would be changing so much, the "real GDP" figure would not (in theory) go up just because of Fed-financed war expenditures.

Ah here's the great part: The government made it illegal for the CPI to go through the roof during the war years. And the ridiculous economic statisticians make no effort to adjust for this. In other words, they take the nominal expenditures during the war years at face value, and they take the statutory price schedules at face value.

So yes, if the Fed triples the monetary base in about six years, and for several of those years the government makes it illegal for prices to rise very quickly, then voila! You'll see a big jump in "real Gross Domestic Product." Woo hoo! Stones into bread!

Is Bob Higgs an Austrian?
He's been to a bunch of Mises Institute events. I don't know if he calls himself an Austrian, but he is very sympathetic to the school, if nothing else.
C+I+G+X-I=GDP is and remains the probably best illustration of why Keynesian macro-economics is funny.
I don't know what the numbers are on employment, but it seems like GDP aside, employing everyone and their wife would in fact get us out of the Great Depression.
"I don't know what the numbers are on employment, but it seems like GDP aside, employing everyone and their wife would in fact get us out of the Great Depression."

Sure, it would get everyone employed, but the question is, how much of real value is being produced? Say, if we employ everyone painting their own house, we'll have nice looking houses, but soon everyone will starve to death.

Actually, Bob Higgs also deals with that issue. During the war, unemployment reached historical lows (along the lines of 1%). The problem, of course, is that the draft makes this figure virtually meaningless, since there is no guarantee that a drafted soldier would prefer their "work + wage" to the leisure time they'd have otherwise.

But, from a GDP measurement standpoint, one would guess that GDP would increase... Though, as Gene points out, that doesn't mean much, since the product is of questionable value.
thinkahol again (wanted to just comment before I figured out the password issue)

Are their proposed alternative measures or definitions of the depression however vague?

If indeed employment was closer 100% than its ever been (and I think looking back we can see that everyone certainly didn't starve (I think something like 40% of produce came from victory gardens)), we were in a very popular war against "an existential threat" and when the war ends the U.S. has 45% of total global manufacturing power, I'm not convinced that real value didn't end up getting produced.

In fact, even excluding the weapons created by military spending, the take home wages for employed workers itself would create real value.

I'm not sure you can wave away the difference between the highest unemployment and the highest employment in U.S. history.

The issue of what people prefer, and implying that they would rather go hungry than work doesn't make sense to me. But I think it was probably a moot point considering that WW2 was the most popular U.S. war ever fought.
This comment has been removed by the author.
Thinkahol: Gene was illustrating the problem with using unemployment as a measure of the economy's health. He wasn't saying everyone starved during WWII!

I'm not convinced that real value didn't end up getting produced.

Gene didn't claim that nothing of value would be produced. The question is whether it would result in more things of value being produced, or fewer than would have been produced absent government intervention.

[T]he take home wages for employed workers itself would create real value.

Not much value if there's less stuff to buy with these wages.
Post a Comment

Subscribe to Post Comments [Atom]

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]