Monday, October 25, 2010

Thursday, October 7, 2010

Why Banning Soda Purchases for Food Stamp Users is Misguided

I woke up (early) this morning to the following story: A Push to Ban Soda Purchases with Food Stamps. I just about fell out of my chair, not because of the story itself, but because I had just had a practice exam session with a colleague and we had discussed something very similar.

Essentially, Mayor Bloomberg has requested that the USDA allow NYC to ban the purchase of sodas and other sugary beverages (referred to from here on as SSBs, or sugar-sweetened beverages) using Food Stamps. First, I applaud the stance that Mayor Bloomberg has taken on health and the obesity epidemic. And I agree with him that we need to get the American public to drink less SSBs (but not just those who are using Food Stamps, a program which is now actually called SNAP, which is going to be relevant to my argument). I just don't agree that restricting Food Stamp purchases this way is either the right way to do it (in terms of stigmatization) or the correct way to do it (in terms of behavior change). And here's why...

First, how does Food Stamps, formally known now as the Supplemental Nutrition Assistance Program work? Food Stamps is a nutrition program that is a means-test targeted program, meaning your eligibility to receive benefits is based on your means (your income plus other assets). It's actually got another layer of targeting built in, in that you have to be interested enough in receiving the benefits (need them enough) that taking the time to apply and submit proof of income and assets is worth it. Once you qualify to receive benefits (the main test is having a gross income at 130% of the poverty line or less), then your benefit is calculated.

Here's where the "Supplemental" aspect of SNAP/Food Stamps comes in. Food Stamps are designed to be supplemental to your food budget. The underlying assumption is that, assuming you have income, you are putting some of that income towards food purchases. The benefit formula is as follows:

Benefit= Maximum Benefit- (0.3 times your income level)

Therefore, as your income goes down, your benefits will go up, and at a level of zero income, you will receive the maximum benefit. The current maximum benefit for a family of four is $668 per month. At levels of income above zero, it is assumed that 1/3 of your income will be spent on food.

This is a crucial point, as there are really two main buckets of people who receive Food Stamps benefits in terms of how the program affects their spending: Constrained and Unconstrained. Unconstrained individuals, because they are spending some of their income on food in addition to the benefits of SNAP, are not changing their spending habits- meaning that they are not spending more than they would like on food, and that Food Stamps is not increasing their purchasing of food to a level greater than they would prefer. In this case, Food Stamps acts as an income transfer: the cash that they would have already preferred to spend on food can now be freed up in their budgets to be spent on other needs. It is like giving these recipients a raise- their budget is increased, and there are no restrictions on their spending (aside from the fact that some things are banned under food stamps: non-food like liquor, and some prepared foods). The key point here is that they are spending more on food than their Food Stamp benefit alone.

Constrained individuals on the other hand are forced to buy food that they wouldn't otherwise buy. This indicates that they have needs greater than food (medical care, housing) but the government has essentially chosen for them that they will spend this benefit on food. It's not an income transfer in this case. Even at the same level of income as someone else, this group has different preferences (higher priorities than food spending) but are forced to spend Food Stamp dollars on food. This group would rather have cash that could be spent on their higher priority items.

The point I am trying to make here is that, for unconstrained individuals, banning the purchases of SSBs using Food Stamp dollars will not have the intended effect. If a given recipient prefers to purchase SSBs, it's simply a matter of shifting their food budget so that their own dollars are spent on SSBs and Food Stamp dollars are spent on something else. This group can still buy Sugar-Sweetened Beverages at the same rate previously, given how cheap SSBs are relative to other foods, particularly nutritious foods, in the diet. Only those very constrained individuals would have their purchases changed by this policy.

This is my first problem with this policy: it won't meaningfully change behavior among the targeted population.

My second issue is that it stigmatizes Food Stamp recipients. It restricts their choices, and in so doing, feels punitive. Food Stamp recipients are not the only Americans who are drinking far too many SSBs. I feel this policy is a lot of noise and attention for one that won't work in the way intended, and that time, money and effort could be better spent on policies that have a more effective design.

If we REALLY want to change the food choices that Americans generally, and Food Stamp recipients specifically, make, then I propose that we make fruits and vegetables cheaper. It's non-punitive. It's got potential to change behavior. One of the complaints that is often heard from Food Stamp recipients is that they would like to buy healthier choices, but that fruits and vegetables (and other non-processed foods) are so much more expensive. I could devote a hundred pages to how this happened (Thank you, Earl Butz), but the bottom line is that people's purchasing decisions are based on the relative costs of different foods, and if we make fruits and vegetables competitive with the reconstituted soy/corn blend junk that it competes with (re: pretty much all processed foods, including SSBs), we just might have the START to a solution. (By no means do I think this will be enough to fix the obesity epidemic in this country.)

Food Stamps, like all policy interventions, is not a perfect solution, but it is a rather elegant one. Like all policies, it's a compromise. It's designed to not distort the markets (by increasing food demand or imposing different consumption choices on the recipients), but at the same time allow the government to put a priority on food spending for low income families. From a free-market perspective, giving cash benefits would be the ideal market solution, but for obvious reasons, this is not politically very popular. Food Stamps tried to bridge this gap, and while it has its flaws, it does the best it can.