Saturday, March 31, 2012

Cuts to SNAP in the Ryan budget

For many years, the Food Stamp Program enjoyed reliable bi-partisan political support.  Even as the U.S. entitlement program for cash assistance was cut and converted to a block grant in the 1990s, food stamps remained largely unharmed.  Leading Republicans such as Senator Bob Dole joined leading Democrats in supporting food assistance for low-income Americans.

Now, at a time when U.S. household food insecurity is near record levels, the nation's largest food assistance program -- under its newer name the Supplemental Nutrition Assistance Program (SNAP) -- is targeted for the most severe cuts ever in its 50-year history.  According to the Center on Budget and Policy Priorities, GOP Congressman Paul Ryan's proposed budget plan for 2013-2022, approved by the House of Representatives Thursday in a partisan vote, would cut SNAP by $133.5 billion, or 17 percent, over ten years.

In reaction to this proposed budget plan, GOP Presidential candidate Mitt Romney said, "It's an excellent piece of work."

Second-place GOP Presidential candidate Rick Santorum, according to the New York Times, said the budget didn't go far enough.

The New York Times editorial yesterday disagreed.  The Times said the proposed cuts "would mean a loss of $90 worth of food a month" for the average household.  If you read the Center on Budget's analysis carefully, clearly the Times should have said "a loss of $90 worth of food stamps in a month" (the distinction arises because SNAP benefits are effectively food support only in part and effectively income support for the remainder).  An economist colleague called me up to criticize this oversight in the Times editorial and to encourage me to post on this topic, saying of the Times: "It's an attempt to be inflammatory.  They wanted a big number."

I see the point, but I might add that the budget cut correctly described, $133.5 billion from SNAP, also is a big number, and perhaps also in this election season an attempt to be inflammatory.

Update (4/1/2012): My colleague encourages me to explain the economic flaw in the Times editorial even more clearly.  The issue has to do with the effect on food spending from an additional dollar of SNAP benefits.  A good rough estimate is that an additional dollar of SNAP benefits generates about 30 cents of additional food spending.  The rest of the additional SNAP benefit substitutes for cash income that the household otherwise would have spent on food, freeing up resources for other household needs such as housing or transportation.  The economic lesson is that a targeted benefit such as SNAP is in part a food subsidy and in part a general income subsidy.  The Times editorial should not have said the Ryan budget would generate a loss of $90 in food, but rather that the budget would generate a loss of $27 in food and $63 in other household needs.

With this correction in mind, my friend writes: "What you did not do in your blog post was tell your reader the nature of the NYT mistake and what the right concept is.... So finally, if I object, it is that you did not take this opportunity to teach a little economics and encourage accuracy.   You came very close to saying, lying is okay as long as the political cause is GOOD.   I do not think you really believe that."

2 comments:

Anonymous said...

"You came very close to saying, lying is okay as long as the political cause is GOOD."

I don't know if it's fair to say the NYT was "lying." In fact, to call it lying seems waaaay over the top.

Regarding the 30% figure, I assume there is a range within the literature or is 30% consistent in the lit? what are some examples of that research?

Parke Wilde said...

Yes, I wouldn't say NYT was lying, nor would I say my original blog post endorsed lying.

For marginal propensities to consume, here (.pdf) is a very old literature review table, but the current research literature is really stumped by this question, so take those estimates with a grain of salt. In a 2009 paper in AJAE, colleagues and I took a different approach to assessing the effect of SNAP on food spending (sorry, may be gated), but the answer is presented graphically and doesn't give a simple marginal propensity to consume. I quote 30% as a ballpark guess.