How do you define a food desert? Probably not the way I do.

(I’m having an ongoing conversation with my friend and fellow blogger Leah from Brazen Kitchen about food deserts. We discovered this week that we’ve both been thinking of this topic since it was in the media earlier in the summer. Keep your eye on her blog for a different angle on the issue. It’s so much more fun to be a geek when you’re not alone!)

You may have heard the phrase “food desert” repeatedly this summer, when First Lady Michelle Obama stood up with several major retail chains and announced a deal to plop down some big ole big box stores in food deserts. Which at face value is great – low-income communities, in both urban and rural areas, often have trouble attracting and maintaining grocery stores and other retailers because of a mix of low returns and high risks. Although the program does include a few independent retailers, the highest volume of stores in the initiative are national or regional chains. Commentators were perplexed – did we pretty much guarantee WalMart shareholders huge gains at taxpayer expense? Why not an emphasis on independent, locally owned businesses, or expanding farmers markets and cooking classes focusing on the foods you can buy there?

It looks like it’s grounded in the USDA’s definition of a food desert: “a low-income census tract where a substantial number or share of residents has low access to a supermarket or large grocery store.” Low-income* communities are defined as a census tract having “either: 1) a poverty rate of 20 percent or higher, OR 2) a median family income at or below 80 percent of the area’s median family income;” and a low-access community qualifies if “at least 500 people and/or at least 33 percent of the census tract’s population … reside more than one mile from a supermarket or large grocery store (for rural census tracts, the distance is more than 10 miles).” So access to a supermarket or large grocery store is the only way to get good quality food, and you have to be within a mile (or ten). And it’s the entire census tract that is or isn’t a food desert, regardless of how densely populated it is (population density determines the census tract size). And the definition of grocery store or supermarket is one that has at least $2 million in sales, which rules out smaller, usually locally owned stores.

This isn’t the only definition of a food desert – the CDC says that a food desert is an area that lacks “access to affordable fruits, vegetables, whole grains, low-fat milk, and other foods that make up the full range of a healthy diet.” They also say (the best sentence I’ve found so far on a government website) that “[i]dentifying food deserts is not an exact science….” The 2008 Farm Bill defined a food desert as an “area in the United States with limited access to affordable and nutritious food, particularly such an area composed of predominantly lower income neighborhoods and communities.” Wikipedia, my source for all things not otherwise easily defined, has a two-paragraph definition section on its food desert page.  And has a slew of definitions and uses on its page in many countries. So while the definitions are similar, they’re not exactly interchangeable. The recent USDA definition certainly seems to be the most specific.

I know how I shop – and especially in the summer when local fruits and vegetables are flying fast and furious, I do what I can to avoid major supermarkets (especially the Whole Foods parking lot – this parody is spot on). I don’t feel as if I live in a food desert (which I don’t, according to the USDA) but even so, I can avoid most major grocery stores and still get my basics sorted out. So I started to think about tweaking the definition of a food desert using my neighborhood, because it’s easier to do this in theory than it is to mess with data on a national scale in practice. But I’m just trying to prove a point here.

My neighborhood is an urban one of moderate density. In the past ten years, its access to stores that sell food has vastly increased. In addition to two long-time grocery stores within a mile or two (one which has been greatly expanded), my end of town now has three more big box stores plopped nearby. I’m within spitting distance (relatively speaking) of two Giant Eagles** (one a Market District), a Whole Foods, a Trader Joe’s, and a brand-spanking-new urban Target – with expanded produce section! This region of Pittsburgh (Shadyside/East Liberty/Larimer) is fast returning to its historical roots as a regional shopping district after spending many years languishing in misguided urban planner hell. Oh, plus there’s another Giant Eagle across the river. And my favorite grocery store, the East End Food Co-op, is a little further away.

So, not likely to be a food desert anywhere close to this place. Except that there are three census tracts that are still pink on the food locator’s website nearby.

Exhibit A (general lay of the land):

And closer to the core area I’m referencing:Here’s the grocery stores (that I know of – this is not meant to be exhaustive) added in:

Let’s ignore for a moment that one of the census tracts noted as a food desert actually *has* a grocery store in it, which brings up the question of grocery store data sources (and having good data in general). I’m curious about the one-mile rule, and if a significant chunk of these tracts are actually within a mile of a grocery store. That little tract – 1203 – seems a little suspect:

So at least 500 people and/or 33% of the people in this tract, have to be more than a mile away from a grocery store to be considered low access. This data set says that absolutely everyone fits into that category. This tract also has 57.8% of its population that is both low income and has low access – 1291 people.***

But it looks *so* close to the grocery stores plotted above, doesn’t it? Let’s pull out every planner’s friend, the scale bar, and draw a one-mile radius around Target, Trader Joe’s, the co-op, and the Waterworks GE, since those are the closest grocery stores to that tract (instead of calling the entire tract a desert because of its center point).

A little different when you use a one-mile radius, I’d say. At least half of census tract 1203 (the more populated half, judging by the street grid on the map) is covered by Target or Trader Joe’s. And if you can get to Trader Joe’s, you can get to the Giant Eagle across the street.

What about seasonality? If you put farmers’ markets or farm stands into the mix, how does that change the landscape? Here’s the seasonal markets, a year-round farmers’ market, a farm stand, and several community gardens:

And their coverage in a one-mile radius: ****

At least one of the larger markets takes food stamps (SNAP? WIC? I can’t keep up with what they’re called anymore), which makes purchasing even easier. Not all of the community gardens sell to the public, but some do. So, best case scenario, several months out of the year, more people have access to (fresh, local) food. And two out of the three food desert tracts are now covered by the existing options. In theory, anyway. Note that I didn’t take a look at the transit routes, for example, because I’ve bored you enough already.

So what’s my point to this exercise? Well, I have several:

  • Recognize the limitations of a national definition in a local area. I probably understand the area around these census tracts better than someone crunching numbers in Washington does. And this is a far from comprehensive assessment of the issue at hand. If I’m worried about food access in these areas, I know there’s a lot more to take into consideration than just what I see on the Food Desert Locator.
  • Make sure you have the most up to date data. Self-explanatory, that one.
  • Determine if a definition is the best one for the job. Do we need another huge big box? Or would something else work better instead – maybe a smaller grocer to stabilize a neighborhood business district and augment the existing large supermarkets?
  • Use this as a tool, not as gospel. I direct this one to the USDA and Mrs. Obama – please, please, PLEASE do not develop a new funding stream for food deserts based solely on this tool. It’s a great tool, but it’s just a tool. Build the flexibility in at the local level so people on the ground who understand the context can do good things in their neighborhoods, not just watch as another big box retailer tromps on in.
You may see me hit on this topic again – the issue of food access is so much more complex than the ‘easy’ solution of throwing a WalMart Express in a poor neighborhood makes it out to be.

* Don’t get me started on the fact that the federal government doesn’t define a low-income area consistently. There is a difference between median household and median family income, for example.

**Giant Eagle is the dominant regional grocery store chain.

*** Just so you know, the population data in this tool is from 2000. For this census tract, the 2010 population is down to 1,650. Here’s another question about data – it’s out of date (probably because Census 2010 income-level data has not been released). Will this be updated? Was it possible to use the American Community Survey data to get more recent (and therefore relevant) data on this issue, at least in urban areas?

**** assuming that both a farm stand and a big market would draw from a mile away, that is.