Abstract
Dollar stores have rapidly expanded their food offerings in recent years. These foods tend to be higher in calories and lower in nutrients, raising public health concerns, especially in rural and low-income areas where food-access challenges are often greatest. However, there is limited empirical evidence evaluating the impact of this expansion on household food purchases on a national scale.
Using data from a yearly, nationally representative panel of approximately 50 000 households, we estimated the share of food purchases from 2008 to 2020 by store type and evaluated the role of dollar stores as food retailers in the United States.
We found that dollar stores were the fastest-growing food retailers by household expenditure share (increasing by 89.7%), with rural growth outpacing growth elsewhere (increasing by 102.9%). Though dollar stores still represent a small share of national household food purchases (2.1% in 2020), they play an increasingly prominent role in food-at-home purchases for certain disadvantaged and rural communities. Understanding the quality of the foods they offer and how this may affect diet-related health outcomes is warranted. (Am J Public Health. 2023;113(3):331–336. https://doi.org/10.2105/10.2105/AJPH.2022.307193)
The absolute number of grocery stores in the United States has been declining since the Great Recession. Supercenters and dollar stores have picked up most of the lost shares of grocery stores, especially in rural areas.1 In particular, dollar stores, traditionally viewed as a destination for discount purchases, offer foods that are mostly packaged, shelf-stable, higher in calories, and lower in nutrients.2 In recent years, dollar stores have rapidly expanded their retail footprint in ways that are highly visible in communities across the country,3 yet there is limited empirical evidence investigating the impact of this expansion on household food purchases on a national scale.
Rural communities especially face a substantial challenge with regard to food access. Rural areas have significantly fewer food retailers than urban areas,4 and rural households with lower incomes are likely to be located farther from the closest food stores.5 Some studies have found that increased access to healthy foods is associated with better health outcomes at the community level, including lower levels of obesity.6,7 However, findings on this relationship are mixed, and the evidence on causal pathways is inconclusive.8,9 Given that rural populations have higher baseline levels of obesity,10 food access and the healthfulness of food purchases in rural areas are of great public health interest. In this study, we analyzed the role of dollar stores as food retailers in rural areas of the United States and the impact on food purchases for at-home consumption. Our results showed that there is substantial growth of dollar stores in the food retail landscape.
The primary data set used in this analysis was the Information Resources Inc (IRI) Consumer Network, a yearly, nationally representative panel of approximately 50 000 households that provide a detailed account of their retail food purchases, including both perishable, random-weight items and consumer packaged goods. We included all currently available years of data (2008–2020) in this analysis. The Consumer Network differentiates purchases made at different store types (e.g., grocery stores, drug stores, and mass merchandisers) and includes dollar stores as a retail channel category. We applied survey sample weights (projection15K) to be geographically and demographically representative of the contiguous United States.
To assess the rurality of households participating in the panel, we matched and merged household zip-code data with rural‒urban commuting area (RUCA) codes developed by the US Department of Agriculture.11 These codes, based on data from the 2010 Decennial Census and the 2006–2010 American Community Survey, classify US census tracts using measures of population density, urbanization, and daily commuting. We categorized communities into 4 groups based on the primary RUCA codes: metropolitan (1‒3), micropolitan (4‒6), small town (7‒9), and rural (10).
Finally, we aggregated household food expenditures by household, store type, and year. We classified store types the same way as IRI: grocery stores, drug stores, mass merchandisers, supercenters, convenience stores, dollar stores, club stores, and other. For the purposes of this analysis, we excluded nonfood items, such as liquor and tobacco products. We also dropped purchases of food items with single-trip costs beyond $500, as these were more likely to be reporting errors or purchases not intended for regular household consumption (e.g., purchasing snacks for an event) or other nontypical shopping events. The use of a $500 cutoff in this study is ultimately an arbitrary one, but we note that it only removed 0.02% (i.e., 1/50 of 1%) of shopping trips for 2008 to 2020.
Our main variable of interest was the share of food expenditures in dollar stores. We first show how food expenditure shares changed from 2008 to 2020, compared with other store types, for all regions and for rural regions; we assessed statistical significance of these changes through an analysis of variance test for each store type treating time as a categorical variable. We then further analyzed dollar store expenditure shares by rurality as well as by income, race and ethnicity, and region.
Figure 1 shows the change in household food expenditure shares by each store type, for all regions as well as just rural regions, from 2008 to 2020; complete statistics on the absolute food expenditure shares by store type and year are provided in Appendix A (available as a supplement to the online version of this article at https://ajph.org). All changes were highly statistically significant (P < .001). In 2008, households spent an average of 62.3% of their food budget in grocery stores. This number declined to 58.3% in 2020. This loss was picked up by club stores (2.4 percentage points), supercenters (1.5 percentage points), dollar stores (1.0 percentage point), and convenience stores (0.2 percentage point). Dollar stores were the fastest-growing retail channel (increasing their share of household food purchases by 89.7%), followed by convenience stores (47.6%) and club stores (30.8%).
FIGURE 1— Change of Share of Household Food Spending in Dollar Stores and Other Formats for (a) All Regions Including Rural and (b) Rural Regions Only: United States, 2008–2020
Note. Projection factors (projection15k) applied to be representative of the US population. “Rural” includes households coded as such by US Department of Agriculture rural‒urban commuting area classifications (i.e., primary code 10).
Source. Authors’ analysis of the Information Resources Inc (IRI) Consumer Network data. The values graphed are available in Appendix A (available as a supplement to the online version of this article at https://ajph.org).
Among households living in rural areas, expenditure shares at grocery stores decreased from 57.4% in 2008 to 50.3% in 2020. This loss was picked up mostly by supercenters (4.6 percentage points), dollar stores (2.5 percentage points), and club stores (1.7 percentage points). Notably, household spending at dollar stores in rural areas increased from 2.5% in 2008 to 5.0% in 2020. In fact, dollar stores were the fastest-growing food-retail channel in rural areas (increasing their share of household food purchases by 102.9%), followed by club stores (49.2%) and supercenters (18.5%). Although supercenters gained a greater share in terms of absolute value, dollar stores increased the most by relative magnitude, doubling their share in household expenditures.
Figure 2 shows the distribution of food expenditure in dollars by urbanicity and a few key demographic indicators. Rural households purchased more foods in dollar stores across almost all demographic groups. The exceptions were rural Hispanic and non-Hispanic Asian households, for whom the sample size might be too small from which to draw meaningful conclusions (Appendix B, available as a supplement to the online version of this article at https://ajph.org). As income decreased, the share of food expenditures in dollar stores increased. Households in the South also purchased more food in dollar stores; within rural areas, households in the South spent the most, and those in the West spent the least. Perhaps the most notable group was rural non-Hispanic Black shoppers; these households spent 11.6% of their food budgets in dollar stores.
FIGURE 2— Share of Household Food Spending in Dollar Stores by Income, Race and Ethnicity, and Region: United States, 2008–2020
Note. NH = non-Hispanic. Projection factors (projection15k) applied to be representative of the US population. We categorized communities into 4 groups based on US Department of Agriculture rural‒urban commuting area classifications: metropolitan (1‒3), micropolitan (4‒6), small town (7‒9), and rural (10).
Source. Authors’ analysis of the Information Resources Inc (IRI) Consumer Network data. The values graphed are available in Appendix B (available as a supplement to the online version of this article at https://ajph.org).
Food purchases in rural areas over the past decade are largely characterized by a shift in expenditures away from grocery stores toward both larger supercenters and smaller dollar stores. Our focus here is on the latter because of the concerns posed by how the foods dollar stores carry differ from those in traditional grocery stores; more research is needed to ascertain whether and how this evolution has also changed the nature of the foods being purchased by consumers.
Dollar stores have experienced the greatest growth in household food purchases over the past decade in terms of relative magnitude, having doubled their market share in rural areas. This rapid growth of dollar stores across the United States since the Great Recession has largely been driven by the rapid expansion of 3 major national chains, which have primarily opened stores in small, rural towns with limited retail options. One chain, Dollar General, planned to open 1100 stores in 2022.12
The concerns surrounding dollar stores and food access center around selection and healthfulness.13 The selection of foods available in dollar stores is typically both less diverse and less healthful than what is found in grocery stores. Historically, dollar stores have only carried shelf-stable beverages and snacks, but now many also carry eggs and dairy, and, more recently, select locations also carry fresh produce.14 Such changes may also partially explain the observed increase in household food purchases at these outlets. Public health advocates have raised concerns that the foods sold in dollar stores are mostly packaged, higher in calories, and lower in nutrients.15,16 Several studies support these claims; they have found that the foods and beverages sold in dollar stores tend to be lower in nutrients and higher in calories.2,4
The recent growth in dollar store food expenditures along with the decline in grocery store food spending, particularly in rural areas, raises concerns that dollar stores may challenge and force out local grocers through competitive pricing, leaving consumers with limited, less-healthy food options. Several localities have already acted on the basis of these concerns. Twenty-five local governments across the country have established policies to curb the expansion of dollar stores. Among them, 9 have specified exemptions if a new dollar store provides certain levels of access to fresh food and produce.17
Alternatively, dollar stores may be filling food voids where local grocers do not have enough business to support maintaining a store,18 providing consumers with food options in low-access areas. Similarly, grocery stores’ consolidation may also leave residents with fewer food options, especially in rural areas.19 In communities where other food storefronts are much farther away, dollar stores may be the only option in terms of food access.
Although both characterizations of the impact of dollar stores on food acquisition may be sensible theoretically, empirical evidence is still lacking. Further research is needed to explore the full impact of dollar stores in areas with low food access and their impact on health and health equity. Our findings suggest that dollar stores are a significant food source for certain disadvantaged populations, especially non-Hispanic Black households in rural areas. We hope our study can instigate more conversation about the role of dollar stores in the food retail landscape and food access across subpopulations.
The lack of studies may partially be explained by the lack of data on food purchases over an extended period. Traditional survey methods, such as the National Health and Nutrition Examination Survey, rely on recall of food purchase and consumption, which may be prone to bias and only captures brief windows of time.20 Innovative methods or the usage of underutilized data sets can create a pathway for future studies in this direction.
This study does have a few limitations. Despite the richness of the IRI Consumer Network data, its sampling strategy is focused on metropolitan areas, leaving the sample size relatively small in rural areas. Similarly, the panel of participants in the IRI data overrepresents non-Hispanic White persons. Interpretation of results pertaining to certain racial minorities (e.g., non-Hispanic Asian persons), especially those in rural areas, should be done with caution. The available data capture a broad variety of shopping trips, and it is not obvious a priori how to distinguish household food shopping trips from those purchases of food by Consumer Network participants that may not be for “typical” household consumption. At the same time, the data also only capture purchases from stores, while nonretail acquisitions such as informal exchanges and community-based meals cannot be assessed, which may be an important part of food consumption for many rural households.21
Our analysis also does not include distance to stores; lower-income households may bypass their closest stores to shop at destinations with lower prices.22 Methods like ground-truthing or robust spatial analysis that go beyond shop proximity are needed to fully understand the role of dollar stores in the American diet.21,23 In addition, while this study only explored food expenditures in dollar stores, future studies should assess the types of foods being purchased and healthfulness of those foods through indicators such as the Healthy Eating Index. Lastly, our study reveals strong regional differences in food purchases in dollar stores, but is not well-positioned to explain those differences.
The increasing market share of dollar stores, especially in rural areas, calls for more attention to dollar stores and their role as food retailers. As dollar stores become a major source of food-at-home purchases for rural communities, understanding the quality of the foods they offer is warranted.
ACKNOWLEDGMENTS
This study was supported in part by US Department of Agriculture (USDA) Cooperative Agreement (58-4000-0-0021) and Tufts University Springboard award (M530868).
The authors would like to thank Bangyao Sun and Divya Saravana for assisting with the analysis, and Hailey Fromkin for editorial assistance.
Note. The findings and conclusions in this publication are those of the authors and should not be construed to represent any official USDA or US government determination or policy. The analysis, findings, and conclusions expressed in this publication also should not be attributed to Information Resources Inc.
CONFLICTS OF INTEREST
No authors have conflicts of interest to report.
HUMAN PARTICIPANT PROTECTION
The Tufts Social, Behavioral, and Educational Research institutional review board reviewed the protocol and deemed this study is not research involving human participants.
