Bryan Caplan  

Should Make-Work Count?

Meritocracy Without Borders... Krugman Killer Quote...
According to official numbers, unemployment stayed extremely high through the New Deal.  But some of Roosevelt's defenders say the numbers are misleading.  Bob Murphy, author of The Politically Incorrect Guide to the Great Depression and the New Deal, explains their argument, then effectively ridicules it:
...The controversy centers around the status of workers who had "make-work" jobs (through the WPA, etc.) under the New Deal.  The "official" government numbers do not count those people as truly being employed, whereas if you did count them, then obviously the unemployment rate would drop a lot more due to the New Deal. For example, the unemployment rate would get back into the high single digits by 1937 (before shooting up again) if you remove WPA workers from the ranks of the "unemployed."

So which unemployment series is correct? This is a somewhat philosophical question; Bob Higgs says that there is no objectively "right" answer, it rather depends what you are using the series for. If the issue is to assess the plight of Americans and gauge how bad the human suffering was during the 1930s, then it would be misleading to say that 19 percent of the workforce had no job during 1938. After all, these people weren't starving because they were getting government checks.

On the other hand, if you are trying to assess the ability of the New Deal to restore soundness to the American economy, then surely it is relevant that the private sector was incapable of finding useful employment for 19 percent of the workforce in 1938, five full years into the New Deal...

Let's use an analogy to make the point: Imagine that in the last days of his presidency, George W. Bush declared a new policy. His spokespeople explained, "It is rather misleading to say that the unemployment rate in October 2008 was 6.6 percent, because that implies millions of Americans are destitute. But in fact they are all receiving generous assistance from the government in various forms. If we say that their 'job' is filling out the paperwork for unemployment claims, then the true unemployment rate is more like 0.4 percent. Those are the people who truly have no source of income, and need to be helped."

Would any left-liberal sign on to that rationale?
By the way, if you find employment-data-related humor hilarious, Greg Mankiw's got some equally effective ridicule of Obama's job creation/salvation promises.

Comments and Sharing

COMMENTS (7 to date)
Greg Ransom writes:

My grandfather worked for the WPA. The WPA was his employer. And my grandfather would tell you that "working" for the WPA was a make-work program, not a real job.

That's what my grandfather told me, and I can tell you my grandfather was a much more honest man than anyone leftist I've ever met in academia.

Taras Smereka writes:

I am not sure what you intend to connote by the word "ridicule". Reductio ad absurdum is a very commonly used and accepted method of argumentation.

El Presidente writes:

Does it count? I suppose it depends on how you define the problem and whether it solves the problem as you define it. It would be nice for everyone to receive pay equal to value added and to earn a living by doing so. But that's not how life really works. Some people get more, some less. So, that will probably change how we specify the problem. The basic problem is that people go without necessities when they lack income and that the "generous" benefits are constrained by law to a certain duration within a person's lifetime. If they can't secure what they need through private employment then our economy is failing them. Why wouldn't we count make-work? Perhaps because it screws with or dogma?

I share your preference for private sector employment, but I'm not willing to restrain government so that wages can be bargained into absurdity. You know that will cause the most pain to those who already hurt.

El Presidente writes:

Ya know the old proverb about giving a man a fish versus teaching him to fish? Well, that works as long as he has something to eat while he learns. Otherwise, it's a load of self-righteous crap. My apologies to the author.

By the way, Mankiw was criticizing the measurement prospects, not disclaiming the effects. Probably because he knows it would be just as ridiculous for him to say he could prove it didn't work as for others to say they could prove it did. What did you think the President should say? How about, "We're gonna spend some money and do some stuff and it should work out better than not spending money or doing stuff, but there's really no way we can be sure either way. So, we're damned if we do and you're dead if we don't. God bless America." Maybe that would be funnier.

Jesse writes:

Do economists have real jobs?

spencer writes:

Since the 1950s less than half of the US unemployed are eligible for unemployment benefits.

The ratio you should look up is called the "recipiency" rate

Boonton writes:

I think the best solution would be the 'does it look like a duck' test. Building the Hoover Dam was a job. Building an aircraft carrier was likewise a job whether your check came directly from the gov't or from a private company with a gov't contract. We could argue that it wasn't a great job or the type of job a healthy economy should produce but that's neither here nor there. There are other metrics that measure those things like per capita income, gov't spending as a % of GDP, civilian versus gov't labor forces etc. Collecting a check is not a job so shouldn't count as such even if you have to spend a half hour a week filling out a form.

On the other hand, if you are trying to assess the ability of the New Deal to restore soundness to the American economy, then surely it is relevant that the private sector was incapable of finding useful employment for 19 percent of the workforce in 1938, five full years into the New Deal...

It would be nice if we are going to have an honest discussion about the New Deal to actually have one.

1. If by 'New Deal' you mean Keynesian demand side policies, you have to accept the fact that FDR was not consistent, especially before WWII. A lot of the New Deal were somewhat incoherent supply side policies to force prices up by limiting competition based on the idea of the time that 'ruinious competition' had driven prices too far down.

2. Using 1938 as a point to measure unemployment is disengenous becaue it does not consider the recession that hit in 37-38 driving up unemployment.

3. provides a useful chart showing the G portion of GDP during the Great Depression. In 1936 Gov't purchases went from 152.5 to 147.0 ($B 1992 dollars) indicating a retreat from classic Keynesian type policies. It would be tricky to assert measuring the New Deal at the 1938 is fair because the the New Deal caused the recession.

4. Post 1938, gov't spending quickly ramped up to $174.2 by 1940 and 288.0 in 1941. Unemployment fell finally back into the single digit zone. Post 1942 we get the massive WWII spending zooming up to $692.

5. Interestingly, while unemployment did spike again in 1938, GDP continued to grow and as far as GDP goes 1938 was not horrible compared to 1932 or 33.

So if by New Deal we mean the actual historical New Deal with its starts and stops, public spending mixed with non-Keynesian supply side planning like the NRA and so on then yes we can say the New Deal's record is spotty. But economists are not historians. The argument over the New Deal isn't whether or not the NRA was a great idea, the argument was does the evidence support or undermine the classical Keynesian argument that massive stimulus spending is the receipe for getting out of a depression.

Taken as a whole, the evidence does seem to support it. Yes you can argue that other evidence from other time periods needs to be taken into account. You can also argue that monetary policy might have been quicker, more effective and less expensive etc. But that's not what seems to be happening here. Here it seems like the critics are acting less like economists and more like politicians or Fox pundits. Hence the cherry picked data (1932 unemployment versus 1938). Hence ignoring the post 1938 history which followed a much more clear cut Keynesian policy of fiscal stimulus (even if this was a partially unintentional by-product of the brewing world war)

Comments for this entry have been closed
Return to top