Quantifying the second stimulus

December 8, 2010
Michael Linden and Michael Ettlinger have a good overview of the cost in dollars of the tax-cut compromise, and its benefit when it comes to employment numbers:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google,mail" data-share-count="false">

Michael Linden and Michael Ettlinger have a good overview of the cost in dollars of the tax-cut compromise, and its benefit when it comes to employment numbers:


David Leonhardt then boils their numbers down even further:

Of its estimated $900 billion-plus cost over two years, roughly $120 billion covers the high-end tax cuts and the estate tax cut, $450 billion covers Mr. Obama’s wish list and $360 billion covers the tax cut extensions both parties favored.

Both in terms of dollars and in terms of jobs, then, this deal is heavily weighted towards progressive options. Yes, the tax cuts for the rich are expensive, at $60 billion a year. But the payroll cut as proposed will cost $120 billion a year, and it does look as though the Obama administration has found its second stimulus.

As Catherine Rampell reports, cutting taxes is almost never the first-best form of stimulus. But it’s the only form of stimulus which is politically feasible—and what’s more, there are diminishing marginal returns to extra spending-related stimulus, in terms of jobs created, since the first stimulus more than covered the low-hanging fruit.

Incidentally, Leonhardt disagrees with my take on payroll-tax cuts. I reckon it makes perfect sense to give them to employees rather than employers, but he says no:

The ideal package would have been larger than the current one, and it would have been better tailored. The $120 billion cut in the payroll tax, for example, will apply to the portion paid by workers, not companies. The Congressional Budget Office and other analysts have said that cutting the workers’ portion provides less bang for the buck because individuals are likely to save some portion of the money. Cutting the employers’ portion subsidizes hiring.

But politics prevented the best kind of payroll tax cut. Republicans did not want one larger than the $120 billion, one-year cut in the package. Administration officials wanted the political benefit of having that whole sum apply to individual workers. The resulting compromise will help the economy, but not as much as it could have.

John Carney made a similar point yesterday:

Temporary tax relief tends not to increase consumer spending by very much. What’s more, tax relief that comes in the form of a temporary payroll tax cut is even less likely to stimulate spending.

Carney has a certain amount of academic literature on his side, especially a recent paper based on a poll of consumers, asking them how much of any tax cut they would save rather than spend.

Here’s the more detailed testimony of CBO director Douglas Elmendorf. He calculates that reducing employers’ payroll taxes would increase GDP by between 40 cents and $1.20 for every dollar spent, while reducing employees’ payroll taxes would boost GDP by somewhere between 30 cents and 90 cents. On the jobs front, every million dollars spent on the employer side would create between 5 and 13 jobs in the first year; if the cuts were applied to employees, the range is between 3 and 9 jobs.

Elmendorf writes that reducing employees’ payroll taxes “would have effects similar to those of reducing other taxes for those workers”; he doesn’t go into any detail about the behavioral economics of quietly reducing withholding versus sending out splashy rebate checks.

So, let me apologize to Greg Mankiw for calling him disingenuous yesterday: there’s clearly much more of a consensus here than I thought.

I’m not yet persuaded that employer-side tax cuts are better: I still think that for three main reasons, employee-side cuts make sense right now. The experience of the last cuts shows how invisible they are and therefore how likely they are to be spent; the size of corporate cash piles shows how unwilling companies are to reinvest extra temporary cashflow; and in general employers are richer than employees, so giving the tax cut to them seems regressive. But it’s clearly credible and intellectually honest to believe otherwise.

That said, I love Elmendorf’s idea of cutting employer-side payroll taxes only for those employers who increase their payrolls. That policy would surely have a very large bang-to-buck ratio.


Comments are closed.