Food Stamps as Fiscal Stimulus?

Megan McArdle, in an uncharacteristically badly-reasoned (and really rather mean-spirited)  post, ridicules an increase in food-stamp benefits as a fiscal stimulus measure. I responded in her comments, but I’m so inordinately proud of my response that I can’t resist posting it here:

Geez, Megan, I’m sorry but that ranks as one of the lamer posts that
you’ve blessed us with. Unworthy of your fine mind. It (combined with
your follow-up posts) is digressive, obfuscatory, wildly anecdotal, and
barely touches the central issue of fiscal stimulus.

Point by point:

1) The poor don’t need more food.

The point of a temporary food-stamp bump is not to provide more
food; it’s to stimulate demand. If it also helps out the subset that
does need more food, or lets all food-stamp recipients buy somewhat
better food, in aggregate (perhaps reducing obesity), that’s just a

But (acceding to your digression) c’mon: a three- or six-month bump
in food-stamp benefits is not going to have any even vaguely
significant effect on obesity among the poor, in either direction.

2) Food stamps only imperfectly translate into increased cash income, meaning that the poor will spend . . . more money on food.

And spending more money on food will increase demand. Which is the goal. (Not increasing "cash income.") Another digression.

3) If the increase in food stamps takes the form of expanded
eligibility, rather than larger grants, the administrative issues and
public outreach will delay your stimulus until well after it is no
longer needed.

Agreed. Should just push a button (figurative speech) and add money
to the existing debit cards. This is one of the best arguments *for*
this method of fiscal stimulus. *Very* little friction or delay in the
system. Straw man.

4) (it’s hard to say how much) of the food stamp spending will
simply draw down perishable stocks rather than generating new economic

Well duh. Producer and consumer elasticity means it’s not a steel
rod. But the limited research we have shows that $1 in increased
food-stamp benefits yields $1.73 in increased demand over the next
year–more effective than any other fiscal stimulus measure.

>>Eventually this will probably generate more economic activity, but probably well after your stimulus is needed.

This statement is *less true* for food stamps than for any other
fiscal stimulus measure. And see recent Krugman on timing of fed cuts
vis-a-vis unemployment and NBER "recessions."

5) The economy doesn’t need a food sector more distorted by daft government programs than it already is.

Agreed, wholeheartedly. But a short, sharp bump would have limited
distortionary effects, because it would not affect expectations. And
when it comes to government programs causing distortions in the food
market, the food-stamp program is *not* the first thing you think of.
It’s agricultural subsidies, whose pernicious effects make food stamps
looks like the tooth fairy. Red herring.

>If you want to give money to the poor, give it to them.

Helicopter drops targeted by income are extremely difficult to
implement. When there’s a system in place to do that targeting,
effectively and efficiently, why not use it?

>Even if they spend it all on drugs, it will hardly be much worse
than spending it all on increasing their already astronomical obesity

Agreed. Either way, it increases aggregate demand.

Demand-induced employment can take pressure off the Fed, which is
already starting to feel the looming and inexorable power of the
numeral zero (or, more accurately, has some reasonable probability of
feeling that looming constraint sometime this year). A diversified
portfolio of stimulus measures can make the best tool in the toolbelt
(monetary policy) more effective and flexible, arguably
counterbalancing or outweighing the undeniable inflationary effects of
fiscal stimulus.

Thanks for listening,


Oh, and I overcame my laziness and searched my hard drive to find the source for the $1.73 figure. A July 2004 analysis (PDF) by Moody’s of Bush economic policies–notably the 2001 stimulus measures.

It’s important to note that these were the only fiscal stimulus measures that were instituted promptly (ever? certainly in many decades), so they’re sort of the only body of evidence we have to determine whether fiscal stimulus early in a downturn actually works.

It does seem to increase demand significantly over the ensuing year. Whether that’s worth the tradeoffs (stimulating inflation, increasing debt, etc.) is another question.

A related post, responding to Greg Mankiw’s objections, here.