(Luke Sharrett/Bloomberg)
Opinion writer

There’s a debate raging over reform of the government’s largest nutrition assistance program, triggered by a new federally-funded study showing that possibly up to a fifth of all spending under the $66 billion per year Supplemental Nutrition Assistance Program (SNAP) is being spent sodas, candy and other junk food. A New York Times story called attention to those troubling facts — only to be met with accusations that any suggestion of barring the use of SNAP for junk food is “shaming” the poor or “paternalism.” The House Agriculture Committee held a hearing on the pros and cons of restricting SNAP to food that’s, you know, nutritious on Thursday. Yours truly contributed to the discussion in this recent column.

Aside from condemning the alleged unfairness of limiting poor people’s choices, defenders of the status quo argue any attempt to bar SNAP funds for junk food would produce no health benefit, since most SNAP beneficiaries would just buy the same amount of soda, etc., with their own cash instead of a SNAP debit card.

“Our best prediction is that there will be no consumption change as a result of the SNAP restriction,” Diane Whitmore Schanzenbach of the Brookings Institution told the Agriculture committee. A typical family “can continue to purchase the same basket of goods, but they would have to make certain to pay for the soft drinks out of their own cash instead of their SNAP benefit.”

This is certainly a plausible prediction; money’s fungible. It’s only a prediction, though, and evidence from the operation of a separate but similar government program does not support it.

The Special Supplemental Nutrition Program for Women, Infants and Children (WIC) provides food, nutritional education and medical referrals to about half of all infants born in the U.S., a quarter of 4-and-under kids, and their mothers. Unlike SNAP, it supports the purchase of a narrower selection of nutrient-dense foods; in October 2009, WIC enacted new rules that sharply reduced permissible purchases of sugary juice with program resources, pursuant to dietary recommendations from the Institute of Medicine. The allotment of juice for 1-to-4-year-olds, for example, went down from 288 ounces per month to 128 ounces.

And it worked, according to a Yale School of Public Health study published four years ago in the peer-reviewed journal Pediatrics. The authors examined grocery store records for 2,137 WIC households in two New England states, covering the 9 months prior to the change and the 9 months after. They found that the households bought 43.5 percent less of the newly restricted sweetened drinks with WIC funds in the post-rule-change period, and that this decline was only partially offset by increased purchase of comparable beverages with SNAP or cash. The total volume of juice purchased fell by 23.5 percent according to the study, with no unintended negative consequences such as a shift to soda.

“The public health impact of the shift in beverage purchase patterns could be significant,” the authors noted. The precise reasons for that shift are not clear, though it stands to reason that the new WIC rules sent a strong message about nutrition that affected beneficiaries’ choices about how to spend their non-WIC resources.

A similar virtuous cycle might be set in motion if we tried to purge soda from SNAP — that is, if we tried to apply the lessons of WIC’s experience instead of rationalizing the status quo. Surely data is a better guide to policy than any prediction, even the “best” one.