Should Government Control What Low-Income People Eat?

In Free to Choose, Milton Friedman argued “There is no place for government to prohibit consumers from buying products the effect of which will be to harm themselves.” Is there still broad agreement on this principle—and should it extend to all Americans? Or should those receiving government assistance be subject to particular restrictions? In the case of the food stamp program, should Uncle Sam be dictating what people can eat and drink?  And if so, would it prevent people from consuming unhealthy foods, regardless?

That’s a question at the heart of recent calls to try to alter the Supplemental Nutrition Assistance Program (SNAP, formerly food stamps) to attempt to encourage healthier eating among low-income Americans. According to a recent front-page article in the New York Times based on a Department of Agriculture study, the number one commodity purchased by recipients is soft drinks—recipients spend 5 percent of their food dollars on soft drinks, and 9.3 percent on the broader category of “sweetened beverages” that includes energy drinks and other categories. Sweetened drinks are thought to be a contributor to our country’s obesity epidemic, which is why public health lawyer Michele Simon argued that SNAP “is promoting all the wrong kinds of foods,” and Marion Nestle, a prominent nutritionist at New York University, is quoted as saying the program is a “multibillion-dollar taxpayer subsidy of the soda industry.” So the argument goes, since sugar-sweetened beverages are harmful, we should ban their purchase using SNAP dollars with the hope that the ban will reduce soda consumption and improve public health.

Such a ban is a popular proposal from public health advocates who want to interfere with the market-based approach currently taken by SNAP, which provides additional resources for food purchases but generally leaves the recipients free to choose what and where to buy, subject to prices, their budget constraints, and their tastes and preferences. The argument put forth in the Times article, and the proposed policy reforms banning the purchase of certain commodities with SNAP dollars, are misleading and potentially even dangerous to this fundamental U.S. safety net program.

Let’s not forget that SNAP has been a clear policy success. It lifted nearly 5 million people out of poverty in 2014 (the most recent data available), is efficiently targeted to families who need benefits the most, reduces the likelihood that families have trouble affording food, serves as an automatic fiscal stabilizer in times of economic downturns, and has extremely low rates of both error and fraud. SNAP also has long-term benefits. My own recent research study found that those who had access to SNAP benefits during childhood grew up to be healthier, and women in particular were more likely to become economically self-sufficient due to childhood access to SNAP benefits.

The recent Times article does a disservice to the public understanding of SNAP in several ways.

First, it is misleading to discuss SNAP recipients’ consumption of soda without noting that the average American not on SNAP also spends substantial amounts on soda. The same USDA report cited in the New York Times article finds that non-SNAP recipients spend 4 percent of their grocery dollars on soda—about the same as SNAP recipients.  Clearly, soda consumption is a national problem, not just one impacting the poor.

Second, the debate needs to address the important policy question of how SNAP changes the types of groceries that participants buy. This is simple economics: by increasing a family’s resources to purchase groceries, SNAP should increase both the quantity and the quality of foods purchased, and it has. When SNAP increases a low-income family’s grocery purchasing power, they are able to buy more nutritious foods they otherwise could not afford. While this is a surprisingly hard question to study empirically, a recent study found that SNAP benefits increase participants’ consumption of nutritious foods such as vegetables and healthy proteins, and reduces consumption of fast food.

Third, it is not at all evident that banning soda purchases with SNAP benefits would actually be successful in decreasing soda purchases. SNAP benefits are modest—approximately $4.50 per person per day—and as a result nearly all families supplement their SNAP purchases with groceries purchased from their cash income. The USDA study indicates that households spend only about $14 per month on soda, or $24 if all sugar-sweetened beverages are included. These numbers are well below the amount of out-of-pocket spending on groceries done by SNAP families. What is the likely impact of banning soda purchase with SNAP dollars, then? Consumers would change the payment method used for buying soda, but likely will not change the amount of soda they buy. It would be a lot of regulatory red tape—defining what counts as a sugar-sweetened beverage, and programming grocery scanners nationwide to exclude those UPC codes—for no actual behavioral change.

Reducing consumption of sugary drinks is a laudable goal, given their likely impact of the nation’s health and waistline. But the recent USDA study shows there is no need to single out SNAP recipients for their consumption of sugar sweetened beverages: Americans at all income levels should drink less soda. If we really want to get Americans of all incomes to drink less soda, policymakers would do much better with a broad-based educational campaign, or a tax on sweetened beverages.