Soc DOI 10.1007/s12115-016-9993-8
SOCIAL SCIENCE AND PUBLIC POLICY
Food Deserts: What is the Problem? What is the Solution? James D. Wright 1 & Amy M. Donley 1 & Marie C. Gualtieri 1 & Sara M. Strickhouser 1
# Springer Science+Business Media New York 2016
Abstract The theory of food deserts is that poor people eat poor diets in part because fresh, healthy food is not accessible in areas where they tend to live. We review evidence from a number of disciplines on various elements of this theory and find it wanting. Access to a car is, for most, a more important consideration than access to a full service supermarket. Moreover, a number of cases are reviewed where full service supermarkets were opened in food deserts, usually with little effect on shopping or eating habits. Keywords Fooddeserts . Diet and nutrition . Food insecurity . Access to healthy foods . Access to transportation . Nutritional knowledge . Food costs . Food cultures
What is a Food Desert? There seems to be agreement that the term Bfood desert^ was first used in Scotland in the early 1990s (Cummins and Macintyre 1999) to refer to areas (neighborhoods, census tracts, communities, etc.) that lacked access to healthy, nutritious and affordable food. Once conceptualized, it became obvious that food deserts were not distributed randomly across the landscape. Rather, they tended to be concentrated
* James D. Wright
[email protected]
1
Department of Sociology, University of Central Florida, 4000 Central Florida Blvd, Orlando, FL 32816-1360, USA
in low-income or minority neighborhoods. Could access issues and food deserts more generally therefore be the cause of disparate dietary patterns, levels of food insecurity and health outcomes across races, economic classes, and geographic regions? Although seemingly a simple question, no one has yet come up with a completely satisfactory answer. The very definition of a food desert has been debated; indeed, some have wondered if such things even exist. Do people who live in these Bdeserts^ in fact eat a different diet or pay higher prices than people who live in areas where fresh, wholesome food is abundant? And if so, are the food deserts responsible for this or are other causal factors present? Would putting full service supermarkets in the food deserts make a discernible difference in diets, health outcomes, or the overall rate of food insecurity? These are the questions taken up in this article. According to the United States Department of Agriculture (USDA), Bfood deserts^ are defined in theory as Burban neighborhoods and rural towns without ready access to fresh, healthy, and affordable food. Instead of supermarkets and grocery stores, these communities may have no food access or are served only by fast food restaurants and convenience stores that offer few healthy, affordable food options^ (BFood Deserts^ 2013). Using Census tracts as the unit of aggregation, there are two further operational thresholds in the USDA definition: (1) BThey qualify as ‘low-income communities,’ based on having: a) a poverty rate of 20 % or greater, OR b) a median family income at or below 80 % of the area median family income; AND (2) They qualify as ‘low-access communities,’ based on the determination that at least 500 persons and/or at least 33 % of the census tract's population live more than one mile from a supermarket or large grocery store (10 miles, in the case of non-metropolitan census tracts.^
Soc
This definition, although now used more or less universally in Bfood desert^ research, has been controversial since it was first formulated for the following reasons: &
&
&
&
&
Food deserts are defined in terms of Census tracts. Tracts, in turn, correspond only loosely to what most people would recognize as Bneighborhoods.^ Moreover, the definition seems to imply that most people shop for groceries within the Census tract where they live, but in reality, fewer than one in five do (The Economist 2011; Drewnowski et al. 2010; Aggarwal et al. 2014). The designations for when a tract qualifies as a low-income tract are arbitrary and hard to justify. Tracts average about 4500 in population. Deciding a tract is Blow income^ if 20 % or more of its residents are below the Federal poverty line would be hard to justify on any conceptually meaningful basis. The same is true of the Federal poverty lines themselves or the B80 % of area median income^ standard. The distance standards seem equally meaningless. Why is the distance standard one mile in urban areas but ten miles in rural areas? Is one mile assumed to be a reasonable walking distance? For seniors, children, the obese, the disabled, or pretty much anyone lugging a full bag of groceries, a one mile walk would be a challenge. Why is distance the sole determinant? What about the conditions of area sidewalks, how many major thoroughfares one has to cross, the presence or absence of crosswalks, and on through a long list of things that constrain the walkability of urban neighborhoods? What about local crime rates that may well Bconstrain^ walkability at least as much as distance or physical limitations? As we discuss later, the distance standards are effectively trivial for households with cars – if you have a car, it doesn’t really matter how far away the nearest fullservice grocery is. The ease with which a distance can be traversed is the critical issue, not the distance itself. The nominal definition of a food desert specifies that they are to be conceived as areas Bwithout ready access to fresh, healthy, and affordable food,^ but there is nothing in the operational definition that addresses fresh, healthy, or affordable. The apparent assumption is that there are never any healthy options available in fast food restaurants and convenience stores, whereas Bfresh, healthy and affordable^ foods abound in full service supermarkets. But even McDonalds’ menu features numerous grilled chicken and fish sandwiches, a couple of salads, fresh apple slices, and bottled water; many convenience stores and even some gas stations stock items such as eggs, milk, some fresh fruit, cereal, trail mix, cheese, peanut butter, whole grain crackers, yogurt, and even hummus. And as for the dizzying array of junk food, alcohol, soda pop and other nutritional catastrophes that adorn every full-service super-market in the US, nothing further need be said.
The concern with food deserts in low-income communities is that residents of these areas experience a special type of food insecurity – not that food is unavailable, but that the food that can be easily had is not nutritionally adequate and is overpriced to boot. The food that is available lacks variety and proper nutrition. The unavailability of nutritious food is in turn linked to an abundance of health issues among residents of these areas. (Whether these health issues are a direct result of living in food deserts or a result of the characteristics of people who happen to live in such areas [poor, racial minorities, etc.] is a key question.) The USDA definition of a food desert is not the only one, just the most commonly encountered. Other definitions include, in addition to geographic proximity, problems accessing nutritious foods due to physical and economic barriers. Others focus on the quality of food available to residents and the affordability of those foods based on the socioeconomic condition in the community. One study operationalized the term Bfood desert^ as Burban areas with ten or fewer grocery stores and no stores with more than twenty employees.^ These alternative definitions are sometimes employed because rather than being concerned with just the factor of nutrition, researchers are sometimes more concerned with the type, size, and number of stores available to residents and not necessarily the quality of available food.
Theories of Food Desert Formation Researchers and commentators have proposed different theories to explain how food deserts emerged in the United States. One theory is that chain supermarket stores have put smaller neighborhood Bmom and pop^ grocery stores out of business. These local Bmom and pop^ stores are forced to close by the enticements of the large supermarkets: large parking lots, longer hours, lower prices, more variety, and some even bag your groceries and bring them out to your car, which small stores rarely offer. The downside is that these suburban megasupermarkets are only available to those with cars or those who can to them via public transportation. Without transportation, many are left to do their grocery shopping at convenience stores where nutritious food is less available. Other researchers argue that food deserts arose in inner-city areas because the median income dropped when affluent residents migrated to suburban areas in the 1960s through the 1980s. Studies show that this migration to the suburbs caused half the supermarkets in the three largest cities of the United States to close. The same migration caused a significant decrease in the median income of the inner city areas which equates to less purchasing power and fewer stores. Shaffter (2002) explored this process and found that middle- and upper-income (and white) suburban communities (in
Soc
California) had twice as many supermarkets as low-income (and black) communities had. Other studies analyzing access to grocery stores in communities by racial composition and poverty have produced similar. Morland et al. (2002) collected data on the dietary intake of residents participating in the Atherosclerosis Risk in Communities study to examine the local food environment’s impact on nutrition. These communities were located in Maryland, North Carolina, Mississippi, and Minnesota. Findings suggested that predominantly white census tracts had five times the number of supermarkets that black neighborhoods had. Findings were similar in Zenk et al.’s (2005) study examining the accessibility of food stores by income level and racial composition of neighborhoods in Detroit. They found that residents of high-income and white neighborhoods had greater access to supermarkets and large chain food stores. Low-income and nonwhite neighborhoods had mostly small grocery and convenience stores. Compared to suburban areas, land prices in urban areas are high and often only small plots are available. This makes it more expensive and sometimes completely impossible for retailers to build large supermarkets in urban areas. In order to profit, supermarkets must move large quantities of goods. This proves to be difficult in urban communities where population density is high but poverty is also high. Space is also an issue in rural areas. In this case, plenty of relatively cheap land is available but population density is low so moving large volume of product through rural stores is problematic. The optimal combination of land, affluence and people is found in the suburbs, and that is where the supermarkets tend to be. Despite some variation in the details, in the end all theories depict food deserts as the creation of voracious, profitmaximizing capitalism. The large supermarket chains and the many ancillary businesses that sustain them (distribution, marketing, production, etc.) see themselves less in the business of feeding the world and more in the business of generating profit. The money is in the suburbs (mainly) and so that is where the stores also want to be – not just food stores but all stores. It is hard to make money on poor people, whether the commodity being offered is food, medical care, housing, day care or transportation. As political commentator Matt Bruenig put it, BWhy do food deserts exist? Poor people do not have very much money. That is, supermarkets do not open up in poor areas because poor people do not have very much money to spend on their wares.^ Indeed, one food activist objects to the current conceptualization of Bfood deserts^ as places that lack convenient access to full-service supermarkets because that idea Bproposes large corporate grocery stores as the solution and often doesn’t encapsulate the reasons why these black-and-white solutions are so difficult^ in low income communities. The real problem, many argue, Bis poverty, and solutions should be communitybased rather than corporate.^ It does seem fair to ask why
more full-service supermarkets are the answer to a problem largely caused by the business decisions of the full-service supermarket chains. But if that is not the right answer, then what is?
Do Food Deserts Even Exist? At some level, this is a trivial question that asks only whether there are any Census tracts that satisfy the USDA definition. And there are – in fact, there are 6529 of them, about 9 % of all 74,134 tracts. Three fourths of these food deserts are urban, the remainder rural; and within these 6529 Census tracts live an estimated 13.6 million food insecure people. The USDA has published detailed data on the demographic characteristics of the food desert tracts compared to other tracts. These data contain no real surprises. BRelative to all other census tracts, food desert tracts tend to have smaller populations, higher rates of abandoned or vacant homes, and residents who have lower levels of education, lower incomes, and higher unemployment. Census tracts with higher poverty rates are more likely to be food deserts than otherwise similar low-income census tracts in rural and in very dense (highly populated) urban areas. For less dense urban areas, census tracts with higher concentrations of minority populations are more likely to be food deserts, while tracts with substantial decreases in minority populations between 1990 and 2000 were less likely to be identified as food deserts in 2000.^ A subsequent multivariate analysis showed that Bminority population, poverty rates, and region of the country are consistently significant predictors of food desert status^ (2012: p. 23). As with virtually all other negativities in the modern world, the poor and African Americans get the short end of the stick. A more meaningful question is whether the USDA definition demarcates a real and serious problem, and here the answer is less obvious. If a Census tract lacks a large full-service supermarket, does that imply per se that there are no food outlets other than fast food restaurants and convenience stores offering crappy non-nutritious food? Clearly not: One obvious alternative are small full-service grocery stores (Short et al. 2007). BFood systems researchers and activists have paid scant attention to studying the potential role of small fullservice retail food stores… These stores meet many of the criteria for community food security (CFS) by providing a wide variety of relatively low-cost foods…^ (2007, p. 352). Also overlooked are the tens of thousands of larger and smaller food retailers, farmers’ markets, corner groceries, roadside food vendors, bodegas, and such, many of which specialize in fresh produce, locally grown meat, fish and dairy, or in short, fresh, healthy food. Contrary to a widespread impression, these not-large food retailers and specialty shops collectively account for about half the total US retail food market.
Soc
BYes, but…^ say the critics. Surely these small and medium sized markets are more expensive and carry low-quality or no fresh produce, little better than the convenience stores. Are they not high-cost nutritional wastelands? The data say otherwise. First, Bthe prices at all three small markets in the Mission and the midsize grocery in Bayview are substantially lower than nearby chain stores.^ Across all stores studied and all food items examined, prices in general were either lower or about the same as prices at nearby national chains. As for quality, the study examined fresh produce available in the markets and compared the results to those at nearby national chains. In one case, produce quality was clearly lower (more wilted veggies, more over-ripe and bruised fruits, etc.), but in all the other cases the produce quality was rated by observers as excellent or at least Bgood,^ comparable to the quality ratings in the chains. It was also clear that the smaller stores can and do offer foods necessary for nutritionally adequate diets and that the small neighborhood food markets easily outdid the big chains in offering more culturally appropriate foods. The smaller markets studied by Short and associates do have problems, of course. They operate on razor-thin profit margins across-the-board and often survive mainly by employing family members at reduced wages and locating in sites where rent or land is extremely cheap. So these smaller stores are no panacea. Also, as we see in more detail later, having nearby stores that offer high quality food at reasonable prices does not mean that people will buy and eat it. This turns out to be the biggest problem with the entire Bfood desert^ concept. What we do learn from the Bay Area study just reviewed is that in at least some cases, the food situation in certain neighborhoods is probably not as dire as the term Bfood desert^ suggests. Blithely accepting the Bfood desert^ designation without further investigation would lead one to conclude that some neighborhoods are much worse off than they actually are. Another study used concept mapping to ascertain local perceptions of food price, availability, etc. in neighborhoods identified as food deserts and compared those perceptions to the equivalent perceptions in Bfood oases^ (where highquality food is abundant). This led to a measure of food buying practices, which turned out to be largely the same in both food deserts and food oases. People whose food buying practices led them to the junk food aisles tended to buy junk food even if they were around the corner from a large supermarket; and people whose food buying practices led them to seek out fresh produce and dairy also did so even if it entailed a long ride on the bus. Availability was largely unrelated to food choices. One of the best studies of the relationship between Bneighborhood food environment^ and dietary intake is Ruopeng and Sturm (2012). These analysts looked at selfreported diets of some 13,500 California children and adolescents. Dietary measures were based on Bdaily servings of fruits, vegetables, juice, milk, soda, high-sugar foods, and fast
food, which were regressed on measures of food environments. Food environments were measured by counts and density of businesses, distinguishing fast-food restaurants, convenience stores, small food stores, grocery stores, and large supermarkets within a specifıc distance.^ Making a long story short, Bno robust relationship between food environment and consumption is found. A few signifıcant results are sensitive to small modeling changes and more likely to reflect chance than true relationships.^ And later, BThe present study found no evidence to support the hypotheses that improved access to supermarkets, or less exposure to fast-food restaurants or convenience stores within walking distance, improves diet quality or reduces BMI among Californian youth.^ Many studies reaching similar conclusions can be found throughout the literature. But so too can one find studies that reach opposite conclusions, studies whose results seem to indicate that residents of food deserts do eat poorer diets than those in other neighborhoods. One case in point is Davis and Carpenter (2009), who also studied California children but used data from the California Healthy Kids survey and found a positive relationship between nearness of fast-food restaurants to a child’s school and both soda consumption and obesity. Another study (Hendrickson et al. 2006) studied residents in four lowincome communities, two urban and two rural, in Minnesota and reported that Bresults of the food inventory show that foods within the communities were costly, of fair or poor quality, and limited in number and type available… Through focus groups and surveys, participants expressed concern that healthy food choices were not affordable within their communities and believed that people in their community suffered from food insecurity. The absence of quality, affordable food for low-income residents in these four Minnesota communities prevents or diminishes their ability to choose foods that help maintain a healthy lifestyle.^ Likewise, a well-known study in Chicago (Gallagher 2006), based on the City’s 18,888 Census blocks or block groups, calculated the distance from Bevery City of Chicago block to the nearest grocery store and fast food restaurant^ and on that basis identified Chicago’s food deserts. As always, the blocks so designated were disproportionally non-white and low income. The findings Bpoint to one conclusion: communities that have no or distant grocery stores, or have an imbalance of healthy food options, will likely have increased premature death and chronic health conditions, holding other influences constant. … it is clear that food deserts, especially those with an abundance of fast food options, pose serious health and wellness challenges to the residents who live within them and to the City of Chicago as a whole.^ Researchers frequently encounter research literatures where one study concludes X and another concludes not-X, and so to the question, BWhich is it?^ the only honest answer is, BIt depends!^ – not a very satisfactory answer but the only
Soc
honest answer available. Some general points of criticism pertinent to most or all of the studies we have (or could have) reviewed are as follows: &
&
&
Food deserts, however defined, are always found to be heavily populated by persons of low and moderate incomes and by ethnic and racial minorities, and they are probably distinctive in many other ways as well (per cent of abandoned housing, local crime rate, low education, high unemployment, etc.) Absent statistical controls for these demographic and socio-economic factors, not much of value can be concluded, since we generally cannot decide which outcomes are in fact the result of living in a particular place (the neighborhood food environment) and which are the result of social, demographic, and economic variables that happen to be correlated with living in particular places. The observed relationships, that is, may be spurious, not Breal.^ Single site case studies are particularly deficient in these regards. Even if it can be shown that a particular relationship is real and not spurious, correlation is not cause. If people who live in food deserts eat too much junk food and not enough fresh produce, it does not necessarily follow that their lousy diets are the result of living in a food desert. For all we know, their diets might be equally deficient even if they lived across the street from a full-service supermarket. The studies we have reviewed rarely specify why one set of research sites was chosen rather than another set, and that raises the issue of observer bias, i.e., the possibility that sites are chosen, whether consciously or not, to sustain an a priori conclusion to which the investigator is already committed. With a little digging, anyone can locate a neighborhood or two in any US city that completely fulfills the food desert stereotype – poor, black neighborhoods whose streets are lined with gas stations, convenience stores, fried chicken restaurants, fast food franchises, drug dealers, hookers, and disaffected teens. And with near-certainty, one will also find high rates of obesity, diabetes, premature mortality, and severe food insecurity in these same places. But with equal ease, one can also find other low income neighborhoods, equally poor, equally black, equally distraught, that are also, as one commentator put it, Bhome to a roadside stand serving organic fruit and vegetables, a health-food shop packed with nutritious grains and a superstore that attracts flocks of shoppers from well outside the desert…^ One can, in short, find anecdotal evidence – cases of uncertain generalizability – to support nearly anything. What we need is a national study of a large probability sample drawn from the list of recognized food deserts, one that uses observational data to assess the local availability, accessibility, quality and price of food and deploys focus group and
survey data to obtain direct individual- and family-level data on food shopping, preferences, preparation, consumption and food insecurity. Only then will we know how (if at all) Bplace matters.^ So far, no such study exists.
Race and Class Disparities in Access to Healthy Food Food insecurity is highest in low income families and among racial and ethnic minorities, the same populations that are sharply over-represented among the residents of USDAdefined food deserts. Thus, there is no disputing that there are racial, ethnic and social class disparities in access to adequate supplies of healthy food. The issue is not whether this is true but why. Is it exclusively a matter of retail locations (place)? Is this mainly a matter of economics? In what ways does Bculture^ fit into the mix? As for the Bwhether,^ Hilmers et al. (2012) systematically reviewed 24 recent, quantitative studies in the Englishspeaking high-income countries that examined Bneighborhood disparities in access to fast-food outlets and convenience stores.^ Eighteen of these studies specifically investigated income disparities in exposure to fast-food outlets; fourteen reported a statistically significant relationship (lower income = greater exposure). Nine studies, all of them US studies, examined racial and ethnic disparities in exposure to fast-food outlets. All of them found that Bthat unhealthy foods were more heavily promoted in African American communities.^ The same general patterns were evident in access to convenience stores. BNeighborhoods where economically disadvantaged and minority populations reside were more likely to have abundant sources of foods that promote unhealthy eating.^ Still other studies show the consequences for obesity and various chronic diet-related diseases. But we must always be aware that food retail location is, as the Institute of Medicine has put it, Bonly one component of the total food environment that affects how people eat and, more fundamentally, their health. Another caveat is that the supply of healthy food will not suddenly induce people to buy and eat such food over less-healthy options, especially when relative prices of the healthier foods are high.^
Food Deserts, Food Choices and Dietary Well-Being Whether exactly true or not, assume that people who live in food deserts tend to eat worse diets and to suffer from more chronic diseases than those who live in non-deserts. The assumption throughout the food deserts literature is that if the people living in food deserts were somehow given access to better food, they would buy it and be healthier as a result. If this assumption is wrong, then in some sense the entire
Soc
hullabaloo about food deserts is Bmuch ado about nothing^ or in yet other words from the Bard’s pen, Ba tale told by an idiot, full of sound and fury, signifying nothing.^ Although there are some exceptions, most studies that we have reviewed suggest that the assumption is indeed wrong. Specifically, the diets of food desert denizens do not tend to improve very much, or at all, when better food and different food outlets are made available. This literature at least intimates (Bproves^ would be far too strong a word) that the dietary issues and food insecurities that beset poor, black communities are generally not the result of neighborhood foodretailer locations. The most recent study to reach this conclusion is Elbel et al. (2015), undertaken in Brooklyn by a research team from New York University. Under the auspices of the New York City Food Retail Expansion to Support Health (FRESH) and with funding from the federal Healthy Food Financing Initiative, a community in the South Bronx (Morrisania), a largely black and Hispanic community with high poverty rates and one of the worst health profiles of any community in the city, was chosen as the site for a new, large (17,000 square feet), fullservice supermarket. A demographically similar nearby community, Highbridge, was selected as the comparison community. Detailed data on dietary intakes were collected from parents and caregivers of children ages 3–10 in the several months leading up to the new store’s opening in August, 2011, then again in the 2 months following the opening, and finally approximately 1 year later. The data collection methodology was the street-intercept survey, a technique known to be effective in surveying low income communities. The design, thus, is Ba natural experiment.^ If the absence of a full-service supermarket in a neighborhood causes people to eat sub-optimal diets, then opening one should cause dietary behaviors to improve. So the prediction would be significant changes in diet, consumption and purchasing behavior subsequent to the store’s opening, if not immediately, then surely within a year. Also, since advocacy groups from the First Lady down to local food policy councils have been urging people to eat better diets for years, a comparison community assures that any changes observed in the Btreatment^ condition are not common trends occurring everywhere. As non-randomized or so-called quasi-experimental designs go, this one is pretty good. The findings were pretty unequivocal: BWhile there were small, inconsistent changes over the time periods, there were no appreciable differences in availability of healthful or unhealthful foods at home, or in children’s dietary intake as a result of the supermarket… The introduction of a governmentsubsidized supermarket into an underserved neighborhood in the Bronx did not result in significant changes in household food availability or children’s dietary intake.^ One commentator on the study remarked, BIncreasing access to fresh food does not guarantee that people have the money, let alone the time and knowledge, to take advantage of it.^
The Elbel et al. paper is by no means the only recent research supporting the same conclusion. A similar study in Philadelphia (Cummins et al. 2014) found that Bthe intervention moderately improved residents’ perceptions of food accessibility. However, it did not lead to changes in reported fruit and vegetable intake or body mass index. The effectiveness of interventions to improve physical access to food and reduce obesity by encouraging supermarkets to locate in underserved areas therefore remains unclear… simply improving a community’s retail food infrastructure may not produce desired changes in food purchasing and consumption patterns.^ Another recent study analyzed data on almost 100,000 adults who completed the 2007 and 2009 California Health Interview Surveys. Findings showed pretty clearly that having fast food joints nearby Bincreased the frequency of consuming fried potatoes, sugar-sweetened soft drinks, and fast food, decreased [the] frequency of consuming vegetables, and [increased the] probability of being overweight or obese^ (2013: pp. 3–4). Otherwise, BWe found no strong evidence that food outlets near homes are associated with dietary intake or BMI^ (p. 1). In short, whether there was or was not a fullservice supermarket, a convenience store, a corner grocery, and so on within a walkable distance of respondents’ residences made no difference in what they ate, how they shopped, or their weight. These are not uncommon or unusual; indeed, they were anticipated in a review essay published in Obesity Reviews in 2007 (White 2007) on the topic of BFood Access and Obesity.^ Although White focused more on UK studies, plenty of US studies were also reviewed. One section of the paper dealt with food deserts, at the time rather a new concept. BThe idea of food deserts had immediate appeal to the media and policymakers, and rapidly became enshrined in government policy; it was mentioned in the [UK] National Health Strategy and the Government’s independent enquiry into health inequalities. However, it is important to recognize that it has little scientific basis.^ Serious studies of food availability are a morass: Some show higher quality and lower prices in the large supermarkets but Bmore recent studies have failed to replicate these findings, showing instead that ‘healthy’ foods tend to be as, if not more, available in poorer areas and are lower in price.^ With respect to distance vs. mode of transit, BThis research consistently demonstrates that car ownership and use of a car to buy food is socioeconomically patterned and that this is a key determinant in choice of main food stores.^ In conclusion, a causal link between access to full-service supermarkets and dietary health is yet to be established. As one commentator put it in a polemic published in The Economist (2011), BMerely improving access to healthy food does not change consumer behavior. Open a full-service supermarket in a food desert and shoppers tend to buy the same artery-clogging junk food as before—they just pay less for it.
Soc
The unpalatable truth seems to be that some Americans simply do not care to eat a balanced diet, while others, increasingly, cannot afford to.^ Numerous studies also confirm that not owning a car is a much more serious barrier than the distance to the nearest supermarket. All studies confirm that very few people (on the order of 10–15 %) grocery shop in the same Census tract (or neighborhood) where they live, so the very definition of a Bfood desert^ is called into question. Cultural background, tradition, education, custom and habit are far more telling predictors of what people choose to eat than how far they are from the nearest full-service supermarket or how many fast food outlets and convenience stores they pass along the way. As Margot Sanger-Katz once remarked in the New York Times, BGiving the poor easy access to healthy food doesn’t mean they’ll buy it.^
Why are the Poor Relatively Under-Nourished? The food desert literature often seems to assume that food deserts matter because consumers often make decisions on their food intake based on the food available to them in their neighborhood. But as the convenience stores begin to offer healthier options and the supermarkets continue to offer numerous non-healthy options, the persistence of unhealthy food choices in the food deserts comes more and more to reflect factors other than retail food locations. Three hypotheses suggest themselves: (1) Many people are uneducated about nutrition and make poor food choices even when they don’t have to. (2) People know what the healthy options are but can’t afford to buy healthy food. And (3) food choices are more a function of custom, habit, tradition and culture than they are the result of nutrition information or economic constraints (or market locations). In all likelihood, all of these play some role. Class, Race and Nutritional Knowledge Unless you possess rudimentary nutritional information, it is not immediately obvious that an apple is a better snack choice than, say, a bag of potato chips or a Snickers bar. The educational deficits created in schools serving poor and minority populations are wellknown. It is thus reasonable to assume that nutritional knowledge of low income and minority populations may be lacking, at least in comparison to affluent whites. Several studies can be found to support this conclusion. One such (Cluss et al. 2013) studied nutritional knowledge among parents of Medicaid-insured obese children and concluded that BParents’ understanding of food’s nutritional value is variable. Black race, less education, and very low income are associated with poorer nutrition knowledge.^ Parents of these children tended to believe that foods were healthier than they in fact were. Another study (Acheampong, Haldeman and Erausquin, 2011) surveyed a sample of urban low-
income, low-education Latino immigrants with preschoolage children and reported a significant relationship between low nutrition knowledge and less healthy child feeding practices. Other studies, however, lead to different conclusions. Both Lucan et al. (2015) and Lynch et al. (2012) studied concepts of healthy foods and healthy diets among low-income African Americans. Lynch and associates found that low income African-American women had generally accurate understandings of which foods were healthful and which not but lacked awareness of why some foods were healthier than others. Lucan and associates undertook largely unstructured interviews with low income blacks in a community in west Philadelphia, whose BParticipants generally expressed views consistent with nutritional guidelines and dietary recommendations that were in place at the time of our study. Participants discussed dietary concepts in ways, and using terminology, that most nutrition authorities would not challenge.^ Acheampong and Haldeman (2013) studied nutrition attitudes, knowledge and beliefs among low income black and Hispanic women and found that BThe majority of African Americans had good knowledge in nutrition while Hispanics had fair knowledge.^ They also found that Bthe most common barrier to consuming a healthy diet reported by both groups was the cost of healthy foods.^ A final study deserving mention is Minkoff-Zern (2014) analysis of California farmworkers. She notes that BFood assistance providers in the USA often treat farmworkers’ inability to afford healthy food as a lack of knowledge about healthy eating, reinforcing racialized assumptions that people of color don’t know ‘good’ food.^ The assumption, in short, is that racial and ethnic minorities and the poor make Bbad^ food choices because they don’t know any better. Minkoff-Zern’s data and analysis show instead that Bfood security and healthy eating, rather than being a matter of consumers making healthy ‘choices,’ is a matter of class-based and racial differences in the food system.^ Rather than assuming that proper education about nutrition would solve the food problems of the poor, her insistence is that income inequality is the heart of the matter. As Matt Bruenig has said, BGenerally speaking, when the problem is economic inequality, the solution is less economic inequality. As the food desert research has shown, nothing else will do.^
The Economics of Eating Healthy Most people who shop at farmers markets or healthy food stores such as Fresh Market or Whole Foods (both of which stock plenty of non-nutritious food, by the way – just check the bakery aisle) will agree that healthy eating is indeed more expensive than eating diets of low or middling nutritional efficacy. At one level of analysis, this is obvious: High quality
Soc
protein is more expensive than canned pasta products; fresh fruits and veggies are more expensive than frozen produce; etc. But what does contemporary food science say about the difference? By general consensus, the best recent study of the topic is a meta-analysis of 27 prior studies chosen because each reported average prices for foods that were characterized according to healthfulness. The analysis showed that on average healthier diets cost about $1.48 more per person per day. Inevitably, this became Ba buck and a half^ in the popular commentary on the study and we use that convention here. Now, a dollar fifty may not seem like a lot of money to readers of this journal, but within the food budget of the poor, it is not a trivial amount. In 2013, the poverty line for a family of four was $23,050. $1.50 per day times 365 days times four people = $2,190 per year. So going from a less healthy to more healthy diet would add 9.5 % to the annual cost of living for a four person family right at the poverty line, and even higher percentage increases for families below the poverty line. A family of four whose annual income was half the poverty line (annual income of $11,525; this is about half the total poverty population) would see their annual cost of living increase by 19 % by switching from a less healthy to more healthy diet. Or consider that cost from the viewpoint of SNAP. An adult male is considered SNAP eligible if his weekly food budget (generally calculated as one-third of income) is $35 per week or less. $1.50 a day x 7 days in a week = $10.50 – an increase in weekly cost of living of about a third. A related study in the UK (Jones et al. 2014) used longitudinal data from the UK’s National Diet and Nutrition Study to examine food prices over time. The trend indicator was the cost per 1000 kcal. In 2012, the average price per 1000 kcal was three times higher for the healthiest diet than for the least healthful diet. Further analysis confirmed that from 2002 to 2012, all food prices had risen but the price of healthful food had risen faster. So not only is healthy food more expensive, the price difference between more and less healthy food has increased. Researchers at the University of Washington (Monsivais and Drewnowski 2007) compared prices for 372 foods and changes in the prices over a 2 year period (2004 and 2006). Foods were categorized from Blower-energy-density^ to Bhigher-energy-density.^ Low-energy-density foods include whole grains, lean meats, low-fat dairy products, and vegetables and fruits; high-energy-density foods are mainly sweets and fats. The results showed, first, that low-energy-dense (healthier) foods were far more expensive per calorie provided than high-energy-dense foods (by approximately an order of magnitude); and secondly, that the cost of healthier foods had increased far more rapidly than less healthy foods. BBeing able to replace fats, sweets, and snacks with less energydense options is becoming an ever greater economic
challenge^ (2007: 2074.) As one commentator on the study put it, BFor people on limited incomes, this is so much more than an annoyance. You weigh the decision, ‘Do I fill my child’s empty stomach so they don’t feel the pang of hunger with what I can afford? Or do I try to feed them nutritious food?’^ (David Bobanick, quoted in Mapes 2007). The economics of healthy eating are thus pretty straightforward. Healthier food is more expensive than less healthy food by a significant amount. Most poor people seek out foods with the most energy density at the lowest cost, and those are foods high in sugar and fat. Thus, the poor gravitate towards highsugar, high-fat, cheap but unhealthful diets. Cheap food is fattening; good food costs more money that many lowincome people can afford. The preceding depicts the food choices of the poor as a tradeoff between cost and nutrition, but the matter is considerably more complicated than that. Indeed, many public health interventions are based on the assumption Bthat people prioritize health when making food choice decisions.^ As a result of this premise, public health interventions tend to focus on increasing people’s nutritional knowledge and making healthy choices easier by increasing availability or decreasing costs of healthy foods. But the emphasis on cost and nutrition ignores the many other factors that drive low income people’s food choices. To illustrate, Antin and Hunt (2012) conducted 2-h interviews with each of 20 low-income African-American women ages 18–25 to identify salient factors involved in food choices other than price and healthfulness. Examples of key factors that recurred throughout the results: &
&
&
&
Familiarity: Young low income African American women tended to prefer foods with which they were familiar whether they were particularly healthful or not. One had an obsession with whole milk that resulted from being given whole milk every day by her mother. BShe was aware that she needed to cut down on calorie intake, yet the social meaning associated with consuming certain types of high calorie foods like whole milk trumped health considerations (p. 338). Convenience: Convenience was often more important than health considerations. Women juggling child care, work, searching for work, familial obligations, school and the like saw fast food and prepared foods as essential time-savers, even though, as one participant put it, B[I know] that burrito just went straight to my thighs.^ Nutritional Quality: BMost women in the study recognized a strong connection between health and the nutritional qualities of foods. However, in spite of this connection, they did not always prioritize nutritional quality when selecting foods^ (2012: 338). Enjoyment: Many of the women in the study, and in the world at large, eat the things they enjoy eating or that
Soc
&
&
Btaste good,^ regardless of their healthfulness. All of us know and cherish Bcomfort food^ and tend to buy and consume food items that provide comfort (chocolate, bread, alcohol, whatever) even when we know full well that healthier choices are available. Satiation: BIn spite of its importance, satiation as a distinct factor in food choice remains underdeveloped in the literature yet appears to play an important role in food choice for many of the women in the study^ (p. 338). One 25 year old woman in the study remarked, BMy kids really enjoy raviolis. And I buy cans and cans and cans of it. It’s so cheap. You know what I mean? It’s so cheap, and I know it’s not all that healthy, but it fills them up. And that’s the main thing. I want my kids to be full.^ Cost: The amount of money on hand when these women grocery-shopped was a highly relevant consideration. BI’m going to go towards whatever is on sale even if that’s the worst thing for me. I have to eat somehow and still be able to pay my phone bill, cable, all this other stuff. I feel like for the government to be putting bulletins and having health alerts and stuff, it’s just making people more depressed knowing they can’t afford that^ (p. 338).
None of this is meant to imply that the poor are predestined to eat unhealthy diets. Aggressively cautious food purchasing and preparation could lead to better diets even if incomes did not increase. But so doing increases the time and effort required to shop for and prepare food, and for many poor people, these other factors of satiation, convenience, and the like assume greater importance.
Class, Race and Food Cultures Factors in food choices other than health and price (assumed to be universal motivators!) typically result either from previous personal experiences or from differences in culture, custom, habit, and food traditions. It is overwhelmingly obvious that culture is a huge factor in determining what we think is appropriate to eat and what is not. In some cultures, diet consists almost entirely of meat and fish – no produce allowed! Some Native American tribes ate diets comprised mainly of seeds, roots and nuts. In Europe and North America, meat comes mostly from beef, pork, lamb and chicken, but in other cultures, snakes, monkeys, anteaters, mice and rats are all acceptable protein sources. South American Indians eat monkeys, iguanas, grubs, bees and head lice; Aboriginals of Australia eat lots of insects. In the same vein, culture even determines what parts of an animal may or may not be eaten. Observant Jews and Muslims do not eat pork, but Christians do. Jews will only consume the front quarters of beef, not the hindquarters (where all the T-bones and other steaks are found). Long sections of the Christian Bible and equally long
sections of the Muslim Qur’an describe what believers can and cannot eat. In the United States, beef, pork and lamb are acceptable meats, but dog, cat and horses are not; in France, horse meat is often on the menu. Koreans relish dogs; Vietnamese and Chinese eat cats; Papua New Guineans eat bats. In the US and elsewhere, it is also OK to eat, say, the leg or back muscles of a cow but considered at best yukky to consume muscles such as the heart, tongue or pancreas, although all of these are perfectly safe and tasty. In short, a great deal of what humans eat, or refuse to eat, is determined by culture and food traditions. Food consumption also carries cultural identity and when people are told that they need to change what and how they eat, some piece of their cultural identity gets chipped away and discarded. Many people, it seems, resent being told that their cultural inheritance is somehow flawed or dysfunctional. No better example of cultural influences on food preference can be found than the diet of contemporary African-Americans, whose present-day culture reflects the legacy of slavery. As slaves, African-Americans learned to eat whatever was available to them. This included whatever slave owners did not want to eat (the offal from slaughtered pigs, for example), food that slaves were allowed to raise on their own (okra, chickens, other vegetables from home gardens), and such foods as could be scavenged from the land (dandelion greens and the like). Lard could be rendered from unwanted hog fat which provided a high-energy-density food, and soon enough, frying in hog lard became the near-universal method of cooking food. In many cases, these available food products were rendered more palatable by adding lots of salt, sugar and fat. From these and related experiences in the slave era, the so-called Bsoul food diet^ emerged and is a predominant fixture in the African American diet even today, a century and a half after slavery ended. The present-day impact of slave culture on African American diets is well-illustrated by the African-American BSunday dinner,^ a firmly established cultural tradition. A typical African American Sunday dinner might include fried fish, yams, ham hocks, fried chicken and mashed potatoes with plenty of gravy. Studies have shown that AfricanAmericans sometimes feel that eating healthier food means giving up part of their culture and resent people who would suggest such a thing. And while there are indeed many ways to make soul food Bhealthier,^ many African Americans believe that so doing would alter the flavor or be discourteous. Similar points could be made about, e.g., Hispanic diets, Asian American diets, the diets of poor whites – indeed, the diets of any identifiable American subculture. And the point, of course, is that many elements of Btraditional^ cooking and food culture are nutritionally sub-optimal but persist at least in part because food carries cultural identity and people cling to these identities even when healthier alternatives exist. Many people from working class backgrounds struggle with the advice to eat more veggies and less meat, even knowing that
Soc
excess meat consumption is bad for many reasons, simply because during their childhoods, the only reason not to eat meat at every meal was that the family could not afford to. Culture and economics are far greater barriers to healthy eating than the distance to the nearest full service supermarket, a conclusion now sustained by a sufficiently large body of evidence to make one wonder whether Bneighborhood food retail location^ matters very much at all.
What is to be Done? One serious issue with the Bfood desert^ concept is that it does not lead to any plausible solution. The theory of the food desert is that people living in neighborhoods without access to a full-service supermarket will be more food-insecure and eat a worse diet than others. So the obvious solution would be to encourage more full-service supermarkets to open in food deserts. But the main reason they are not already there is because it is difficult to make profit providing food to lowincome populations. Efforts to overcome the economics have not been encouraging. And even when supermarkets (or other, presumably healthier, food outlets) do open in previously under-served areas, local food choices and food buying habits generally do not change. Many other alternatives to locating full service supermarkets in poor areas have been suggested but none seem all that promising. Some suggest community gardens, communitysupported agriculture, farmers’ markets, and urban agriculture more generally but studies suggest that these interventions would probably not be very effective in turning food deserts into food oases. All Bfood sovereignty^ innovations are probably worth encouraging for other reasons but not because they will (or even can) do much to reduce food insecurity among the poor. Overcoming cultural resistance to dietary change is presumably a matter of better education, and surely, the dietary knowledge of the average American could be improved. But there is not a lot of evidence suggesting that poor diets result from inadequate knowledge and a small mountain of evidence that economics lies at the heart of the matter. For most, a lousy diet results from inadequate income far more than from deficient knowledge or distance to a full service grocery store. As Matt Bruenig has it, Bthe food justice advocates have the entire thing backwards. The same thing that causes food deserts to exist is what causes them to persist — in effect — even when supermarkets come into the area. It is not proximity to grocery stores that matters; it is income. Providing the poor with higher incomes will eliminate food deserts and increase their genuine access — not just geographical access — to healthy foods.^ Unfortunately, Americans do not trust poor people with their money, so strategies to decrease income inequality by
spreading the wealth around more widely are usually dismissed as utopian, radical or hopelessly naïve – and usually all three. One alternative that might be more workable and, given the state of the research, more effective, is to give poor people cars rather than cash. Of all the factors that seemingly limit access to food, the most important is no money but second in importance is no car. There are many programs that provide free or low-cost cars to qualifying low income families in Alabama, California, Colorado, Florida, Georgia, Indiana, Indiana, Minnesota, Massachusetts, and many other states. These programs go by names such as Online Car Donation, Working Cars for Working Families, Vehicles for Change, Free Charity Cars, Car Ministry, Wheels to Work, With Causes Charitable Network, 1–800-Charity Cars, Good News Garage, Wheel Get There, Wheels for Hope, etc. All of them have eligibility restrictions so it is not as though cars are given away to anyone who asks. But qualifying individuals and families whose path to self-sufficiency is thwarted by unreliable or non-existent personal transportation can turn to these kinds of programs to find relief. Such a strategy might strike some (or even many) as unworkable on a large scale and prone to scamming, but is it any less realistic than the thought of persuading large national grocery chains to open new stores in low-income minority urban neighborhoods? Probably not.
Further Reading Acheampong, I., & Haldeman, L. 2013. Are Nutrition Knowledge, Attitudes, and Beliefs Associated with Obesity among LowIncome Hispanic and African American Women Caretakers? Journal of Obesity Article ID 123901, 8 pages, doi:10.1155/2013/ 123901. Aggarwal, A., Cook, A. J., Jiao, J., Seguin, R. A., Moudon, A. V., Hurvitz, P. M., & Drewnowski, A. 2014. Access to supermarkets and fruit and vegetable consumption. American Journal of Public Health, 104, 917–923. An, R., & Sturm, R. 2012. School and residential neighborhood food environment and diet among California youth. American Journal of Preventive Medicine, 42(2), 129–135. Antin, T. M. J., & Hunt, G. 2012. Food choice as a multidimensional experience. A qualitative study with young African American women. Appetite, 58(3), 856–863. Bruenig, M. 2012. Confusion Around Food Deserts. Retrieved August 12, 2015. (http://mattbruenig.com/2012/04/14/confusion-aroundfood-deserts/). Cluss, P. A., Ewing, L., King, W. C., Reis, E. C., Dodd, J. L., & Penner, B. 2013. Nutrition knowledge of low-income parents of obese children. Translational Behavioral Medicine, 3(2), 218–225. doi:10.1007/ s13142-013-0203-6. Cummins, S., & Macintyre, S. 1999. The location of food stores in urban areas: a case study in glasgow. British Food Journal, 101(7), 545–553. Cummins, S., Flint, E., & Matthew, S. A. 2014. New neighborhood grocery store increased awareness of food access but did not alter dietary habits or obesity. Health Affairs, 33(2), 283–291.
Soc Davis, B., & Carpenter, C. 2009. Proximity of fast food restaurants to schools and adolescent obesity. American Journal of Public Health, 99(3), 505–510. Drewnowski, A., Aggarwal, A., Moudon, A.V. 2010. The Supermarket Gap: How to Ensure Equitable Access to Affordable, Healthy Foods. A Research Brief. University of Washington Center of Public Health Nutrition. Retrieved December 19, 2013. (http:// depts.washington.edu/uwcphn/pubs/reports.shtml). Elbel, B., Alyssa, M., Dixon, L. B., Kiszko, K., Cantor, J., & Courtney Abramsand Tod Mijanovicha. 2015. Assessment of a governmentsubsidized supermarket in a high-need area on household food availability and children’s dietary intakes. Public Health Nutrition, 18(15), 2881–2890. Gallager, M. 2006. Examining the Impact of Food Deserts on Public Health in Chicago. Full Report. Retrieved November 28, 2013. (http://marigallagher.com/site_media/dynamic/project_files/1_ ChicagoFoodDesertReport-Full_.pdf). Hendrickson, D., Smith, C., & Eikenberry, N. 2006. Fruit and vegetable access in four low-income food deserts communities in Minnesota. Agriculture and Human Values, 23, 371–383. Hilmers, A., Hilmers, D. C., & Dave, J. 2012. Neighborhood disparities in access to healthy foods and their effects on environmental justice. American Journal of Public Health, 102(9), 1644–1654. Jones, N. R. V., Conklin, A. I., Suhrcke, M., & Monsivais, P. 2014. The growing price gap between more and less healthy foods: analysis of a novel longitudinal UK dataset. PLoS ONE, 9(10), e109343. doi: 10.1371/journal.pone.0109343. Lucan, S. C., Maroko, A. R., Sanon, O., Frias, R., & Schechter, C. B. 2015. Urban farmers’ markets: accessibility, offerings, and produce variety, quality, and price compared to nearby stores. Appetite, 90(1), 23–30. Lynch, E. B., Holmes, S., Keim, K., & Koneman, S. A. 2012. Concepts of healthful food among low-income African American women. Journal of Nutrition Education and Behavior, 44(2), 154–159. Mapes, L. 2007. Healthier Foods Getting More Costly, Study Says. The Seattle Times Retrieved November 23, 2013. (http://www. seattletimes.com/seattle-news/healthier-foods-getting-more-costlystudy-says/). Minkoff-Zern, L.-A. 2014. Knowing BGood Food^: immigrant knowledge and the racial politics of farmworker food insecurity. Antipode, 46(5), 1190–1204. doi:10.1111/j.1467-8330.2012.01016.x. Morland, K., Wing, S., & Roux, A. D. 2002. BThe contextual effect of the local food environment on residents’ diets: the atherosclerosis risk in communities study. American Journal of Public Health, 92(11), 1761–1767.
Pablo, M., & Drewnowski, A. 2007. The rising cost of low-energydensity foods. The Journal of the American Dietetic Association, 107(12), 2071–76. Shaffer, A. 2002. The Persistence of L.A.’s Grocery Gap: The Need for a New Food Policy and Approach to Market Development. UEP Faculty & UEPI Staff Scholarship. Retrieved on November 30, 2013 (http://scholar.oxy.edu/uep_faculty/16). Short, A., Guthman, J., & Raskin, S. 2007. Food deserts, oases, or mirages? Journal of Planning Education and Research, 26, 352–364. The Economist. 2011. Food Deserts: If You Build It, They May Not Come. July 7. Retrieved November 27, 2013. (http://www. economist.com/node/18929190). United States Department of Agriculture. 2013. Food Deserts. Retrieved November 29, 2013 (http://apps.ams.usda.gov/fooddeserts/ foodDeserts.aspx). White, M. 2007. Food access and obesity. Obesity Reviews, 8, 99–107. Zenk, S. N., Schulz, A. J., Israel, B. A., James, S. A., Bao, S., & Wilson, M. L. 2005. Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in Metropolitan Detroit. American Journal of Public Health, 95(4), 660–667.
James D. Wright is a Professor at the University of Central Florida, Director of the UCF Institute for Social and Behavioral Sciences, and a former Associate Editor of Society (1988–1997). Amy M. Donley is an Assistant Professor in the Department of Sociology at the University of Central Florida and the Associate Director of the UCF Institute for Social and Behavioral Sciences. Her research focuses on inequalities, specifically food insecurity, poverty, and homelessness. Marie C. Gualtieri is a Doctoral Student in the Department of Sociology and Anthropology at North Carolina State University. Her research focuses on the experiences of food insecurity among seniors in the United States and the impact of food insecurity on health and well-being. Sara M. Strickhouser is a Doctoral Candidate in the Department of Sociology at the University of Central Florida. Her research interests include food insecurity and applied and evaluation research. She is a 2008 graduate of Stetson University and is currently completing her dissertation on the social and health outcomes related to food insecurity in the US.
