Ezra Klein says “yes.” How so?
[A]t this point, [deficit-neutral emergency spending is] worth trying. It’s best to do jobs legislation using deficit dollars. That way you’re not taking money out of one part of the economy to put it into another. But as Dylan Matthews wrote yesterday, money spent in different places does provide different levels of stimulus. It’s plausible that you could move cash from, say, tax cuts for the wealthy, which tend to get saved, and use it instead for a payroll tax holiday, or infrastructure projects, or a tier of unemployment benefits for people in states with unemployment rates above 9 percent and who’ve been out of a job for more than 99 weeks.
This is pretty close to what I wrote in a comment to my own post on Greenspan’s apparent reversal on the Bush tax cuts (emphasis added):
[I]t’s all about: (i) timing, and (ii) distribution. Considering both, I think we can address both the short-term and long-term economic concerns by extending only the middle-class portions of the Bush tax cuts only temporarily. And yes, unemployment benefits still have more stimulative bang per buck than even “middle class” tax cuts in terms of immediately boosting demand for goods and services, because unemployed people have even less income (are more cash-constrained and hence will immediately spend all that you give them) than “middle class” working people.
As for any offsets one might use to “pay” for stimulus (vs. deficit finance it), the distribution of the burden of those offsets affects how detrimental to the stimulus effort the added costs to households or businesses would be. For the purely short-term goal of immediately boosting aggregate demand, if I pay for extended unemployment benefits by immediately raising taxes on the rich, or cutting government spending that is purely wasteful (as if it is thrown away and benefits no one), such offsets would not “harm” the stimulative effect as much as if I pay for those benefits by cutting some other spending that truly benefits/assists low-income households (and hence would undo the helpful effect of additional unemployment benefits).
Most economists assume an effective stimulus requires that on net the spending be deficit financed, because we presume that a “Robin Hood” deficit-neutral policy (where we would take from the rich to give to the poor) would not be politically viable. Still, that is an example of a deficit-neutral policy that would effectively boost aggregate demand. If it continued too long, however, that effective stimulus would turn into a policy very harmful to longer-term economic growth via the effects on the aggregate supply (labor supply, saving–the inputs that add to our economy’s productive capacity) of a full-employment economy.
And later in his same blog post, Ezra mentions “stimulus” as another “bad word.” I agree, not just for the knee-jerk, visceral reactions it might incite from people who don’t like typical “stimulus” policies, but because it’s somewhat inaccurate and pretty insufficient in describing short-term countercyclical fiscal policy even among economists who support such policies. Way back when the American Recovery and Reinvestment Act was first passed in February of last year, I explained that what the recovery act sought to accomplish seemed to me to be a variety of very different (and very ambitious) economic goals. “Stimulus”–what I interpret as policies designed to provide an immediate boost to (aggregate) GDP–was just one of the goals. “Assistance”–providing help to households suffering from severe drops in their economic well-being–was another, and that was regardless of whether such assistance at the household level boosted (aggregate) GDP effectively or not. Fortunately, with spending on extended unemployment benefits, it does both, so the fact that we are deficit financing perhaps the final bit of this recession’s deficit-financed “stimulus” spending, seems a reasonable–and decent–thing to do. (I tried to suggest so in my short live radio interview with Marketplace Morning Report’s Steve Chiotakis yesterday morning.)