Network News

X My Profile
View More Activity

Research desk: What's a dollar of stimulus worth?

By Dylan Matthews

BHeffernan1 asks:

How many jobs does a federal gov dollar buy? Provide optimistic (most efficient -- extending unemployment? building a smart power grid?), median and pessimistic (tax cuts for wealthy?).

For reasons Ezra has laid out in the past, it's best to focus not on job creation narrowly, but on the general economic impact of a policy. Policies that create jobs have other benefits (and costs) as well. If you're evaluating a proposal to pay workers to build a railroad, you want to know what the value of that railroad is, not just how many workers were hired. If you're giving laid-off workers unemployment insurance, the point isn't just job creation, but also helping them pay rent. So if you want raw job estimates for different policies, see Moody's analysis of the cost per job of job tax credit proposals, or CAP's look (PDF) at the job creation potential of various clean-energy investments. But keep in mind they leave out important effects of various proposals.

We can get a better picture of the overall effectiveness of a stimulus method by looking at its contribution to GDP. The most recent numbers come from April's Senate testimony (PDF) from Mark Zandi of Moody's. Zandi calculated the change in GDP caused by a dollar spent on various stimulus policies. Here are the results, grouped from least effective to most effective. You might want to click on the graph to see the larger, and more readable, version.


The pattern is striking: Direct government spending -- through unemployment benefits, food stamps, work sharing or infrastructure spending -- top the list, giving you more than a dollar's worth of stimulus for a dollar's worth of spending, while cuts to taxes affecting businesses and upper-income individuals -- such as the corporate, dividend, capital gains and alternative minimum taxes -- give you less.

The reason there is clear: A tax cut that ends up with upper-income folks gets saved rather than spent, and a dollar saved doesn't stimulate the economy. That's why some tax breaks, such as the job tax credit and payroll tax holiday, are fairly effective, though still less so than direct spending. The problem, of course, is that the politics of tax breaks are easier than the politics of spending, even though the tax breaks are actually more expensive. But if the government wants the maximum stimulus at the minimum deficit cost, direct spending is the way to go.

By Ezra Klein  |  June 17, 2010; 4:39 PM ET
Categories:  Stimulus  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Gallup poll: Americans favor more stimulus, partial health-care repeal
Next: Knives come out


Moody's top two items (food stamps and workshare) include the term "temporary", which is not a term in the federal legislative vernacular. The third item (extended federal unemployment entitlement) implicitly includes the fictional term "temporary".

Infrastructure spending then becomes the top contender, with general [non-targeted] aid to states and specific tax cuts following. There's a lot of agreement on these items, yet those at the legislative helm seem wholly uninterested, instead favoring entitlements without foreseeable sunset.

Posted by: rmgregory | June 17, 2010 4:57 PM | Report abuse

Notice something else about that chart:

Why don't we RAISE the corporate tax rate and LOWER payroll taxes. Even if we did that revenue neutral, there would be a GDP boost.

Posted by: stevie314 | June 17, 2010 5:04 PM | Report abuse

well considering payroll taxes are supposed to fund specific programs AND the fact that increasing corporate taxes can lead to greater excess burdens (due to inelasticity differences b/w workers and corps) than comparable payroll changes, long term fiscally, such a tradeoff seems pretty complex...

Posted by: stantheman21 | June 17, 2010 5:10 PM | Report abuse

I think stevie is on to something, but rather than raise the corporate tax, we ought to bring forward the expiration of the Bush tax cuts for people earning over $250K. And/or pass a financial transactions tax. And then spend the proceeds on a combination of aid to states for home care, child care, and other social services, along with federal jobs programs targeted at the unemployed young people and over 55s. Even if you don't phase in the tax increases and spending increases differently (ie both start right away), you still get a significant net positive multiplier (around 1.3 by the Zandi chart) even though the deficit does not increase. Deficit Neutral Fiscal Stimulus!

Also, bear in mind that Zandi's estimates are, I believe, from a time period when the US economy was not facing a liquidity trap and deflation. Economist Gauti Eggertson has argued that under those (ie our current) circumstances, spending has a higher multiplier (he estimates 2x), and tax cuts have a negative multiplier (in his model they increase deflationary expectations by making it easier for businesses to cut prices). If Eggertsson is right, then Deficit Neutral Fiscal Stimulus would have an even more expansionary impulse than one would think from the Zandi estimates (closer to 2.5 than 1.3).

Posted by: rwclayton7 | June 17, 2010 5:23 PM | Report abuse

I'm with rwclayton7. Accelerate the end of the Bush tax cuts. The rich always claim that tax cuts create jobs, but they don't. Until the top personal income tax rates came down below 40% it was more advantageous for a corporation to retain earnings and plow them back into the company. Invest and expand, in other words. Now they funnel more and more of the corporation's earnings to top management as compensation, usually regardless of how the company or the stock is doing. And these folks don't spend like lower income people do. Many stash or spend it abroad.

But most of the Congress people respond to 1) their own and their friends' economic interests and 2) the interests of their donors. So all we hear from both sides of the aisle is tax cuts for the rich. Right now Susan Collins and Olympia Snowe are happily cutting aid to states, just as they did with the original stimulus, in order to preserve tax breaks for some favored constituents and/or donors. Unless we figure out a way to lessen the influence of money in our politics, we are not going to get the economy moving or solve any of our longer-term serious problems, because the small numnber of people who profit from the status quo are just too powerful.

Posted by: Mimikatz | June 17, 2010 6:17 PM | Report abuse

We all know the best use of tax dollars is on more F-35s!

Posted by: AZProgressive | June 17, 2010 6:31 PM | Report abuse

As Klein said two days ago: "This is a nice object lesson in the inadequacy of GDP as a measurement of societal well-being."

It's akin to churning a stock account.

Also, without being able to see Zandi's methodology (I was unable to find a reference to it, though there are innumerable rapt references to Zandi's claim) one is left wondering as to any basis in reality.

Posted by: msoja | June 17, 2010 7:13 PM | Report abuse

Hi Dylan,

Thank you very much for the response. But I worry putting it in terms of GDP makes it less meaningful (though undoubtedly more accurate)? I believe people care more about the unemployment rate than about GDP, even if they're two sides of the same coin.

For the past few days, Ezra has made the point several times that there should be more stimulus to get the economy back on track and speed job growth. He also argues that it is money well spent (a healthy lunch) as far as Fed spending goes. We in the peanut gallery / comment section fall into our central casting rolls of "spend more to speed growth" vs. "we're putting our children in debt".

My question was meant to try and quantify deficit spending in terms of job growth. Most regular folks have a very good sense of what the country feels like at 5%, 7%, 9% unemployment and a poor sense of what the country feels like at 2%, 4% 6% GDP growth. After all, you can have a pretty high rate of GDP growth (acceleration) and still have a very modest impact on the unemployment rate.

The best way to make the case for case for more stimulus, is in terms of jobs. But the hard part is the assumptions that allow a translation from GDP growth to jobs. And, what do you get for the money. It's just undeniably more appealing to pay for a smart power grid than for extended unemployment benefits. Even if it's less effective as stimulus, the product at the end is something used by all.

Anyway, thanks again for your response and work on the blog.

Brendan Heffernan

Posted by: BHeffernan1 | June 17, 2010 7:58 PM | Report abuse

Just went back and read your link to Ezra's post on stimulus vs. job bill. And I agree. It was a stimulus bill not a jobs bill and GDP is the clearer way to look at it.

But, I still think further stimulus will be sold on the basis of job creation primarily and infrastructure secondarily.

Now, how to make that graph. Hmmm.

Posted by: BHeffernan1 | June 17, 2010 8:14 PM | Report abuse

Not sure from whence rwclayton7 gets the idea that Zandi's proposals do not keep in mind the liquidity trap, so I cannot say if the argument is accurate. I would note, however, that these are new numbers that differ significantly, if not relatively, from those he offered 21 January, 2009, q.v., (.pdf)

Posted by: | June 17, 2010 8:16 PM | Report abuse

This makes a lot of sense, until something occurred to me that I'm getting hung up on--

"The reason there is clear: A tax cut that ends up with upper-income folks gets saved rather than spent, and a dollar saved doesn't stimulate the economy."

Doesn't a dollar saved (unless it's shoved under a mattress) go to a bank that can lend it out, stimulating GDP? Maybe I'm way off, but can someone explain this to me?

Posted by: aarhead | June 17, 2010 8:34 PM | Report abuse

--"Doesn't a dollar saved (unless it's shoved under a mattress) go to a bank that can lend it out, stimulating GDP?"--

Yeah. I've pointed out the same thing here, before. What Klein calls savings, and dismisses, are inevitably invested somewhere, ie., in businesses, creating jobs, or building homes. Even the money in people's ordinary checking accounts. Rich people are rather more proactive in making sure their extra money is working for them, and, as a byproduct, generating real value for others.

Posted by: msoja | June 17, 2010 9:33 PM | Report abuse

--"If you're evaluating a proposal to pay workers to build a railroad, you want to know what the value of that railroad is, not just how many workers were hired."--

You know, that's just more nonsense. If you're just paying workers to build a railroad you don't get to count all the other expensive stuff that goes into making a railroad. To count the track and locomotives that someone else paid for as part of your stimulus is just plain dishonest. With that thinking, you could give the railroad scheme a dollar and count the billions that the thing winds up being worth. It's just stupid, and yet it's what this place is awash with.

Posted by: msoja | June 17, 2010 10:42 PM | Report abuse

"But if the government wants the maximum stimulus at the minimum deficit cost, direct spending is the way to go."

It's also the way to go if you want to further consolidate power within a small group of elite politicians who control the national purse strings.

I have to say, I'm with msoja on method. When you come up with numbers that so exactly dovetail with liberal expectations and desires, it looks like a neckbearded, basement-dwelling comic book geek's chart on why every supermodel actually wants to sleep with him, and will, given half-a-chance.

The laborious path to reach such conclusions needs to deconstructed down to the nano-particle, or the conclusions won't be credible. Because, chances are, once so deconstructed, the conclusions won't be credible. And even supporters might have to acknowledge that there is some fudging going on.

@aarhead: "Doesn't a dollar saved (unless it's shoved under a mattress) go to a bank that can lend it out, stimulating GDP? Maybe I'm way off, but can someone explain this to me?"

The rich (a) tend to spend a lot, irrespective of the constant accusation that they are hoarding the pie so that nobody else can have any. And, when they do save, they like their money to work (or to fantasize that it is working, and (b) they invest. In start-ups, in VC firms, in stocks and bonds--in things that end up buying durable goods or employing people or inventing new products that benefit everybody. The zero-sum-game thinking on the left about rich people (and I include all the liberal rich people, of which there are many, who you'd think would know better) amazes me.

But, I'm sure the problem (as I'm often told) is just that I've been brainwashed by Rush Limbaugh or Sarah Palin or Zombie Ronald Reagan.

Because, you know, you can't dispute hard, imminently obvious facts like a temporary increase in food stamps doing the most of any kind of spending or tax cuts to directly stimulate GDP.

And, once all the complaints about rich people sticking their tax savings in mattresses is combined with a comment about slashing the taxes for the middle class, because they'd go and spend it (I know I would), I'll take that lamentation seriously. Until then, it's just posturing by people who believe, against all the lessons of history, that central planning and command economies are superior to messy, unequal capitalism.

Posted by: Kevin_Willis | June 18, 2010 9:40 AM | Report abuse

I certainly agree with msoja and kevin_willis. A dollar saved or invested has a very real multiplyer effect. The other thing is I don't trust congrss to write a prudent stimulas bill. How much is a Nancy Pelosi's husband's business contract worth? How much is it worth to have turtles cross a highway, or to give money to Acorn as Barney Franks tried to do in the first TARP bill. How much is it worth for Montana to spent stimulas money building a tennis court? That what you get when congress passes a stimulas bill.

Posted by: hfarmer2 | June 18, 2010 11:40 AM | Report abuse

I think if the last 10 years have taught us anything, it's that is entirely possible to invest one's money in things that don't "end up buying durable goods or employing people or inventing new products that benefit everybody". And to make an a**load of money doing so.

Posted by: julie18 | June 18, 2010 1:15 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company