Jobs "saved"? Great, but let's see the math.

Apparently, someone in the administration knows how to keep track of the number of jobs that would have been lost if the this administration's policies hadn't been in effect.

Gee, I think that's just great... but I'd like to see the math, please. Social scientists would call it "quantifying the counterfactual" -- an oxymoron. Regular people just call it "making up data for political posturing" -- a tactic that fools only morons.

Here's an excerpt from the Wall Street Journal's article, The Media Fall for Phony 'Jobs' Claims:

...the inability to measure Mr. Obama's jobs formula is part of its attraction. Never mind that no one -- not the Labor Department, not the Treasury, not the Bureau of Labor Statistics -- actually measures "jobs saved." As the New York Times delicately reports, Mr. Obama's jobs claims are "based on macroeconomic estimates, not an actual counting of jobs." Nice work if you can get away with it.

Let's see the macroeconomic estimates, and the math behind them. It would enable us to add some important additional data to the excellent chart below (one that I wish I'd though of) which I borrowed from the Innocent Bystanders blog. It shows three sets of unemployment numbers: the actual results, and the results the administration projected with and without their recovery plan.

Click to enlarge:

Obviously, the administration's plan isn't delivering the results their experts had projected four months ago. Yet we keep hearing about all the jobs "created and saved" by that same plan. Something isn't adding up. Maybe an unveiling of the math behind the estimates would clear things up(?).

Maybe, but I won't hold my breath.

Comments

[jpniner]

see Art Laffer on the anti-Bernanke, Hyper-inflation bandwagon now?

[Bret]

Did you see Cafe Hayek's post on this?

http://www.cafehayek.com/hayek/2009/06/still-right-for-the-wrong-reasons.html

It seems that the graph isn't quite accurate, though reality may be even worse.

[Optimist123]

I think the graph above is corrected for the timing problem mentioned by Cafe Hayek.

[Bob]

You think this is fuzzy? Wait until the Pres. starts touting all the "green" jobs created.

Don't be too hard on him though. The BLS projects that from 2006 through 2016 the federal government will shed about 90,000 employees, ex post office.

http://www.bls.gov/news.release/ecopro.t03.htm

[woodchuck64]

Given they've sent 40 billion so far (Cafe Hayek link), what's wrong with estimating jobs from that? I mean the money isn't going all overseas, and it isn't going all into some CEO paycheck. Eventually most of that 40 billion will end up in American paychecks.

40 billion / $50k = 800k average paying jobs. Picking $150k seems pretty conservative.

[Bret]

You're right about the graph correction. Sorry, I should have looked more carefully.

[samiran]

Given they've sent 40 billion so far (Cafe Hayek link), what's wrong with estimating jobs from that?

Ethyl:

Because quite frankly, thats a load of BS. Look at the way that Spain has "invested" in green jobs.

For each 4 jobs created, 9 jobs were lost. Money was spent, and the combination of money and new regulations created an environment that actually destroyed jobs.

That's a net negative economic scenario.

Indeed, there are net-positive economic scenarios that *also* result in job losses; this would include spending on capital that allowed for great efficiencies, reducing demand for labor.

Most of this is irrelevant, however; what is important is looking at the actual statistical result of that money spent. Given the current lack of data regarding how the money is being spent, and the anecdotal reports of money lost or wasted, and the anecdotal reports of mounds of red tape, as well as the trade-war consequences of the "buy america" provisions, I would not be surprised to see that for each 5 jobs 'created' by the Obama stimulus, 6 jobs are 'lost'.

Only time will tell, however, and we really don't have any way to make judgments.

The only judgments that we *can* make are that the Obama estimate of unemployment that assumed passage of the stimulus plan were incorrect. We don't know what it would have been without the stimulus plan, and we don't know how the stimulus plan as affected unemployment. As Steve says, being able to examine the math behind those estimates, versus the actual numbers which are playing out, will give us information on how the stimulus is working.

It will also give us some information on the quality of this Administrations economic planners.

The real problem was best identified by Max Baucus, the Montana Democrat:

During a March hearing of the Senate Finance Committee, Chairman Max Baucus challenged Treasury Secretary Timothy Geithner on the formula.

“You created a situation where you cannot be wrong,” said the Montana Democrat. “If the economy loses two million jobs over the next few years, you can say yes, but it would’ve lost 5.5 million jobs. If we create a million jobs, you can say, well, it would have lost 2.5 million jobs. You’ve given yourself complete leverage where you cannot be wrong, because you can take any scenario and make yourself look correct.”

http://allfinancialmatters.com/2009/06/09/how-do-you-measure-jobs-saved/

This sort of Orwellian viewpoints, used to contort traditional economic evaluations of public policy scare the hell out of me.

No government should rely upon a non-falsifiable position as a rational justification for economic policy. That's very, very bad mojo.

[woodchuck64]

"Because quite frankly, thats a load of BS."

I'm not sure what that refers to, but I'll assume it means the claim that the stimulus has saved 150k jobs.

My estimate for perfect efficiency in a perfect world was
40 billion / $50k = 800k jobs saved.

So in some sense the administration is assuming 20% efficiency at this point in time, which seems pretty conservative and hardly much to get worked up over.

"Only time will tell, however, and we really don't have any way to make judgments."

Exactly my view as well. A neutral reaction to the administration's claims makes the most sense to me at this point. They certainly underestimated job loss currently, but that may or may not have any bearing on their claim of jobs saved.

“You created a situation where you cannot be wrong,” said the Montana Democrat. “If the economy loses two million jobs over the next few years, you can say yes, but it would’ve lost 5.5 million jobs. If we create a million jobs, you can say, well, it would have lost 2.5 million jobs. You’ve given yourself complete leverage where you cannot be wrong, because you can take any scenario and make yourself look correct.”

This assumes that we had several equally logical and rational solutions to the economic crisis. In that case, choosing one over the other by justifying it with an unfalsifiable argument would be very wrong. However, that is not the case. Keynesian stimulus was the widely recognized solution on all sides from the beginning. There really was only one logical choice.

The best form of stimulus is debatable, but here too, speed was of the essence, and the Democrats had lots of projects all ready for deployment.

[Bob]

EA and all others who will dis the chart.

The numbers are the numbers!

Maybe that $40 Bill will make a difference. But as of the dates the real world data is worse than the assumptions. Maybe the administration miscalculated the short term losses. If so, what else did they miscalculate? We don't know because we don't know what their calcs are.

[goy]

Two things are more frightening than the actual numbers OR the difference between reality and BHO's fantasy.

First is the trend they imply, given other realities like the rising cost of oil, failure of TARP to do anything measurable so far, federal policies already passed and on the horizon that are antagonistic to small business, etc. You've done trends in the past, Steve - how about one for this?

http://bit.ly/JAtQe

Second is the breathtaking silence in the entrenched, Fifth Column media on this phenomenon. Had this chart been the reality a year ago, you'd be seeing it on every major news network and the front page of the Sunday NYT, in 3D.

[gabriel]

Ethyl:

It sounds like your equation for jobs saved assumes at least two factors.

1) That every dollar goes directly toward saving a job instead of say, toward creating 1 job at the expense of 2 and

2) That the jobs saved are permanent and do not disappear after the stimulus is gone and the work is over.

Given those assumptions, there's no reason anyone is compelled to agree with your equation.

Also, in response to the Montana Democrat, it sounds like you're saying that if only one logical, rational solution exists, then it's ok to justify it using tautologies and incomplete data. Which sounds silly to me. If instead you believe such justifications are always improper, then you are in alignment with the spirit of this blog post.

[woodchuck64]

gabriel:
"It sounds like your equation for jobs saved assumes at least two factors."

My assumption was explicit: perfect efficiency in a perfect world. Since we don't have such perfection, I hoped it was clear that my calculation wasn't a reasonable calculation. The Obama administration's calculation is 20% of that which implies that 80% of the stimulus is not going to job creation, at least not yet. As a rough ballpark figure, it doesn't sound too bad, hardly much to get worked up about.

"it sounds like you're saying that if only one logical, rational solution exists, then it's ok to justify it using tautologies and incomplete data."

No, I'm saying that if only one logical, rational, uncontroversial solution exists (Keynesian stimulus), the government bears no burden of proof to justify implementing that solution. This whole Jobs Saved issue is not meant to be an argument because one isn't needed. It's Marketing: making sure people know that the product they've already bought and paid for is working.

[gabriel]

Ethyl:

"No, I'm saying that if only one logical, rational, uncontroversial solution exists (Keynesian stimulus), the government bears no burden of proof to justify implementing that solution."

I'll grant that in a time of crisis.

"This whole Jobs Saved issue is not meant to be an argument because one isn't needed. It's Marketing: making sure people know that the product they've already bought and paid for is working."

I won't grant that.

There's a difference between justifying an *implementation* and justifying *results*. The purpose of this blog post is to question whether or not the plan is working based on actual data that measures the *results*. No one can be sure if the product they've bought and paid for is working if the logic used to justify *results* is faulty. You can call it marketing all you like, but you either believe that

1) A government should obfuscate the *results* of a program for the purpose of _.

or

2) It should not.

In the spirit of this blog, I'll just assume that you believe 2) and leave it at that.