Food insecurity and the irrigation of food deserts

Michelle Obama has reincarnated the food pyramid as a “healthy plate“. The plate is divided equally between fruits, grains, vegetables and protein–a balanced dinner plan to lead the nation towards a leaner and less diabetic future. As part of applying this new plate method to the real world, a patient of mine came to clinic with a map of our city.

“This is what I’d need to do buy everything on a healthy plate,” he said.

How much would he have to spend for this trip to the grocery store? Given public transport and food costs, about $25–twice his post-rent, post-tax daily income. And the trip would require 49 minutes of transportation time, each way.

This patient, like many others, lives in an American “food desert”: a neighborhood that features more liquor shops and gas stations than it does produce stands or grocery stores. In today’s post, we explore the data behind food deserts, some of their politics, and think about how to intervene in the midst of dilemmas about food price spikes and recessions.

Back in the mid-1990s, researchers in the British Low Income Project described “areas of relative exclusion where people experience physical and economic barriers to accessing healthy foods”. As British cities became more segregated and major supermarket chains subsumed regional groceries, the researchers saw food markets close in poor neighborhoods. In the U.S., large mega-grocery stores migrated to white suburban neighborhoods or upscale city centers, with undesirable poorer populations left in semi-industrial areas without access to supermarkets, except by long public transport routes (if those existed). Food stamp recipients who lived farther from supermarkets consumed less fruit and vegetables, and experienced associated increases in obesity, heart disease and diabetes. The implications for race politics and urban geography were obvious. In Northeast Washington D.C., a census tract two miles east of the U.S. Capitol, 61% of residents are both low-income and have little access to healthy food options.

While originally described in cities, the problem of “food deserts” was also recognized in rural zones, where large consolidated food distributors will not venture. Delivery of small amounts of food to local stores with a poor customer base is unprofitable, particularly when delivering to areas far from major highways. In the words of a vice president at the country’s largest food distributor, “When it comes to rural areas and our decisions there…it’s like any business…it’s all driven by profitability and access.”

The subsequent reality for many households is that access to healthy food is limited, and families increasingly rely on foods that are available in gas station convenience stores, bars and casinos…or make a monthly trip to buy pre-packaged foods from Wal-Mart and other “big box” stores that sell many unspoilables. Even though these foods are obviously unhealthy, they make financial sense: energy-dense [junk foods] cost on average $1.76 per 1,000 calories, compared with $18.16 per 1,000 calories for lower-energy nutritious foods like fruits and vegetables. Some folks like Michael Pollan have referred to these energy-dense products as “edible food-like-substances”:

Mapping deserts

The USDA has recently released its national data on the locations of major food deserts. In a recent study, the department published a series of maps identifying food deserts around the country.

The maps reveal a few key points:

(1) In 10% of the 65,000 census tracts in the country, at least one-third of the population lives more than a mile from the nearest supermarket or large grocery store (or more than 10 miles in rural areas).

(2) 82% of the affected people live in urban areas where convenience stores, or even liquor stores, offer the best opportunity for dinner.

(3) 14.7% (17.4 million) of U.S. households were food insecure at some time during 2009. This is essentially unchanged from the prior year.

(4) Households with children had nearly twice the rate of food insecurity (21.3%) as those without children (11.4%). The most food-insecure household type are those headed by a single woman with children (36.6% food insecure).

Re-calculating food security

The USDA data provide a detailed assessment of food insecurity based on extensive surveys. Most of our information about hunger in the U.S. is based on similar repeated survey methods.

But a challenge to this approach is that it takes a long time and serious effort, and can’t always capture the impact of rapid crises like the food price spikes that have occurred during the economic recession (which we discussed in an earlier post). To determine what level of government subsidies or other efforts need to be put in place immediately to buffer new food insecurity,  there’s no time or money to survey every census tract in the country systematically. As a result, several economists have used an alternative method to estimate how many new households become “food insecure” and need extra aid (like food stamps) during crises.

The method calculates how many people are newly food-insecure (and therefore in need of assistance) using older concepts around undernutrition–that is, calculating the number of people who are too poor to consume needed calories. To make the calculation, government and policy agencies construct something called an Engel curve, which is a plot of how many calories people consume at different levels of household income:

They use this curve to look at how much income it would take to purchase the minimum number of required calories per day to avoid becoming malnourished. They then look at the curve of how much income is earned by each sector of the population (a cumulative distribution function of income) to estimate what proportion of the population does not meet the minimum income threshold necessary to purchase the required calories (the “food insecure” portion):

During a food price spike, the Engel curve shifts to the right, such that more income is needed to buy the same number of minimum calories. During a recession, the distribution of income in the population also shifts to the left, so that there are more people below the minimum income level needed to purchase sufficient food (the new curves are subsequently corrected for the price elasticity of food, meaning how much a change in price really results in a change in demand):

While this method gives us a quick measure of how many people might be so poor as to be unable to secure food, it has some considerable drawbacks. First, calorie intake is determined by a lot more than just income. Secondly, in the presence of high-density junk foods that are nutrition-poor but calorie-rich, “food security” might apply more broadly to those who are unable to secure healthy meals rather than those who are simply able to access enough calories. Therefore, as part of the process of calculating government budgets by estimating how many people are newly food-insecure, we may need to drastically modify our methods to account for the impact of food deserts and other factors affecting the actual type of calories and quality of food being consumed. That hasn’t yet caught on in the economics literature, which still prefers the older method of estimation…(anyone interested in describing a new method? drop us a line at EpiAnalysis).

Irrigating deserts

What might be done address the impact of food deserts on food insecurity?

Statistical studies suggests that supermarkets are the most prevalent method of providing reliable healthy food choices to a population, contingent on their produce selection. As a result, many government programs have focused on subsidizing or otherwise incentivizing supermarket corporations to operate in poor neighborhoods. The limited data available suggest that consumption of vegetables among African-Americans in poor neighborhoods increases by 32% when a supermarket appears in their census tract. Among the more widely-cited initiatives is the Pennsylvania Fresh Food Financing Initiative, which provides loans and grants to stores to defray the costs of construction, and has funded 58 new stores.  New York City’s FRESH program (Food Retail Expansion to Support Health) similarly offers an abatement of land or building taxes for a period of 25 years and a sales tax exemption on building materials.

There’s less population-wide data available to support the many case studies that have tried to suggest that farmers markets are a meaningful alternative to supermarkets. It seems that the impact of farmers markets dependents on whether they will accept food stamps (or equivalent vouchers from the poor, see San Francisco’s program for example), how interested vendors are in serving poorer communities, and how interested poor households are in coming to purchase from them. Developing marketing strategies to get poor households interested in fresh product remains a major dilemma. These marketing studies put the onus on the poor to become interested in fresh produce. The economic problem that remains is that farmers markets are sometimes too expensive to alter the fundamental cost equation between junk foods and less-filling produce that requires more food stamps to be used. New York has offered special food stamps that can only be used for produce (beyond regular stamps), which has increased stamp-based sales at farmers’ markets from $40,000 in 2007 to $89,000 in 2008.

Even if markets do not come to the poor, the poor are still trying to access markets. So while farmers markets have been a subject of much acclaim, in part because they are so popular in more elite classes, less attention has been paid to facilitating effective public transport as a critical element of meaningful food desert policy. This point was brought up by the USDA in its testimony to congress in 2008, where public transport affordability was part of an agenda to help people access existing supermarkets, especially since subsidies to build supermarket chain stores were only working in a few major cities and farmers markets remained an even rarer phenomenon. But the topic has been less prominently featured in the 2010 Obama Healthy Food Financing initiatives, which focuses more on subsidies to corporations selling food than to individuals trying to purchase food.

Making supermarkets more available is half the battle; the other half is thinking about how to make healthy foods as cheap as the junk food that is effectively subsidized by corn and soy farming subsidies. In learning from New York, San Francisco and other cities, federal subsidies could not only be used to build supermarkets, but also affect the cost of getting to those markets and the price of food within them.


8 responses to “Food insecurity and the irrigation of food deserts

  1. Is the Engel Curve method relatively new?
    USDA’s food security module has been widely used to evaluate household food security/insecurity. It has shown validity in certain developing countries.
    At a macro level, International Food Policy and Research Institute (IFPRI) classified MENA countries according to food security status, using these indicators: total export/food imports, food production per capita, calories per capita per day and GNI per capita.
    I appreciate your distinguishing food quality from calorie density. The coexistence of food insecurity and overweight is common in today’s world.
    Thank you!

    • Thank you for your comment. The USDA monitors the food security of the nation’s households through an annual, nationally representative household survey. While thankfully this survey-based method has reached some developing countries, the Engel curve approach has been used in many others because of the absence of survey data availability. The IFPRI and related approaches by the Food and Agricultural Organization of the UN have essentially modified the Engel curve method for modern estimates, for example by adding data from surveys; data on age structure and population distribution; and data on production, imports and exports of food commodities. But these modifications are still are subject to a number of uncertainties in the absence of comprehensive surveys and further details about the nutritional makeup of food (some would argue that it is too reductionist to worry about the specific micronutrient contents, but this is subject to ongoing debate and is important in resource-poor settings where food characteristics leave many without core micronutrients).

  2. Pingback: Whew, We Won’t Starve – My Self-sustaining Mind-set and Food Deserts | Fishing for Words

  3. Pingback: Can we design a heart-healthy home? Disease and the built environment | [ EpiAnalysis ]

  4. Pingback: Can We Design a Heart Healthy Home? | The Health Care Blog

  5. Pingback: Can We Design a Heart Healthy Home? | Health Care Jobs in Illinois

  6. Pingback: Can We Design a Heart Healthy Home? | Health Care Jobs In Connecticut

  7. Pingback: Are food deserts relevant to obesity? | [ EpiAnalysis ]

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s