A Hidden Truth: Understanding U.S. Food Insecurity from 2009-2017

Andrew Leung, Evan MacArthur-Waltz, and Emily Byers

December 9th, 2022

Introduction

Due to recent downward economic trends and geopolitical conflicts, food insecurity has taken a back seat in the nation’s conscience, with just 2% of Americans naming it as an important issue, according to the latest Gallup polls. However, food insecurity is still as prevalent as ever with more than 34 million people, including 9 million children, in the United States being food insecure (Feeding America).

In this blog, we want to explore three key questions:

  1. What are the trends of food insecurity in the U.S. over the past decade? At the state level? At the county level?
  2. Which states/counties have the highest increase and decrease in food insecurity over the past decade? Which have the highest and lowest prevalence of food insecurity?
  3. What about these states/counties have led food insecurity to persist or improve?
Source: [Everyday Health](https://www.everydayhealth.com/coronavirus/pandemic-exacerbating-mental-health-problems-in-food-insecure-americans/)

Source: Everyday Health

Defining Food Insecurity

The United States Department of Agriculture defines food insecurity as “a household-level economic and social condition of limited or uncertain access to adequate food” (USDA). In other words, as Feeding America describes on their website, food insecurity is “the lack of access, at times, to enough food for an active, healthy life” (Feeding America). At once a public health and social justice issue, food insecurity is associated with many social and health outcomes and is largely driven by unemployment, poverty, and income shocks. Interventions such as food assistance programs and “broader societal-level improvements in economic stability” have been proven methods of food insecurity intervention (Feeding America).

About the Data

We requested Map the Meal Gap data from the non-profit organization Feeding America, a nationwide network of food banks across the United States. Since 2013, Feeding America has released an annual report on the state of food insecurity in the United States. Upon our request, the Feeding America research team provided us with all of their data since the founding of the Map the Meal Gap project. We focused on the percentage of state and county populations who are food insecure to gain a better understanding of how food insecurity prevalence has shifted in the United States from 2009-2017. While we were also supplied with data from 2018-2020, we decided not to incorporate this data into our project because the Feeding America research team shifted their methodology and metrics after their 2019 report (on 2017 data)––meaning we would not be able to make a direct comparison between the pre-2017 and post-2017 data.

Apart from the food insecurity data, we also relied on U.S., state, and county shapefiles from the tigris package to make our geographic visualizations.

Most of our data comes from the Feeding America Map the Meal Gap research project

Most of our data comes from the Feeding America Map the Meal Gap research project

Food Insecurity Over Time Across the U.S.

We start our exploration of trends of food insecurity in the U.S. over the past decade through looking at prevalence of food insecurity at the nationwide and state levels.

U.S. Map

Across the United States, food insecurity has generally decreased since 2009. The map below displays this trend at the state level. States lighter in color have lower rates of food insecurity, while darker colors display the opposite. The effects of the 2008 economic crisis were felt across the country in the three years following, as shown by the increase in food insecurity across the country from 2009-2011(USDA). After 2011, however, median food insecurity steadily declined.

Change in Food Insecurity by State Between 2009 and 2017

This bar graph displaying the absolute change in percentage of state populations who are food insecure highlights Louisiana and California as exhibiting the greatest increase and greatest decrease in percentage of state populations food insecure between 2009 and 2017. While Louisiana’s food insecurity prevalence grew by 1.9%, California’s food insecurity prevalence decreased by 5.7%. Interestingly, only 4 states experienced an increase in food insecurity and only one state remained at the same level of food insecurity. The other 45 states and D.C. all experienced a decrease in food insecurity between 2009 and 2017.

Highlighting States of Interest

Food Insecurity Prevalence by State Over Time

The following line graph displays the percentage of state populations that are food insecure between 2009 to 2017. As seen in the previous visualizations, it is evident that food insecurity prevalence across the United States is trending downwards. However, the fluctuations in food insecurity prevalence among different states over the 9-year period varies widely. That being said, through highlighting certain states, this line graph makes clear that there are four states worth investigating further. While there is a lot of movement among different states’ food insecurity populations, Mississippi and North Dakota stand out as consistently having respectively the highest and lowest percentages of their populations being food insecure between 2009 and 2017. Secondly, upon highlighting Louisiana and California, it also becomes clear that these are the states with the greatest increase and greatest decrease respectively in food insecurity prevalence between 2009 and 2017.

Four States Stick Out

This box and whisker plot confirms many of the findings shown in the previous visualizations. From 2009 - 2011, the median rate of food insecurity in the US increased slightly from 14.8% to 15.7%. From that point on, however, the median rate of food insecurity steadily decreased from 15.7% to 12.2% in 2017. Interestingly, despite national trends, there are two states that have remained outliers on both ends of the food insecurity spectrum: Mississippi and North Dakota. In this 8-year period, North Dakota remained the state with the lowest insecurity, ranging between 7.3% to 8% of the population. On the other hand, Mississippi remained the state with the highest rate of food insecurity, ranging between 19.2% and 22.7% of the population. Mississippi was a statistical outlier from 2012 to 2016.

Understanding at the Local Level

Having established four states as worth further investigation, we developed county-level maps to see how food insecurity prevalence has shifted between 2009-2017 at the local level.

Each state is divided by county. Each point is size-adjusted to represent population centers.

Lowest and Highest

North Dakota
North Dakota has a below average poverty rate and low unemployment rate that results in more people being able to afford food (Map the Meal Gap Report 2019, 20). North Dakotans are also able to redeem their SNAP benefits from more locations than most states, including farmer’s markets. Lastly, North Dakota’s small population relative to other states might mean that the issue of food insecurity is more manageable.

North Dakota has a below average poverty rate and low unemployment rate that results in more people being able to afford food (Map the Meal Gap Report 2019, 20). North Dakotans are also able to redeem their SNAP benefits from more locations than most states, including farmer’s markets. Lastly, North Dakota’s small population relative to other states might mean that the issue of food insecurity is more manageable.

Mississippi
Mississippi has remained the hungriest state in the United States since 2009, with 19.2% of its population (574,272 people) being food insecure as of 2017. One driving factor for the enduring nature of food insecurity in the state is the high rates of persistent poverty, defined as places that have had poverty rates of 20% or more for over three decades. Over half of Mississippi’s counties (42) are currently in persistent poverty. ([CRS](https://sgp.fas.org/crs/misc/R45100.pdf)). Exacerbating food insecurity in the state is the uniquely strict requirements for the Supplemental Nutrition Assistance Program (SNAP), the program meant to alleviate food insecurity for impoverished families. Families must have an annual income that is at or below 130% of the federal poverty line ([MDHS](https://www.mdhs.ms.gov/economic-assistance/snap/)), a low bar that leaves many of the families above this threshold unassisted. Furthermore, Mississippi requires an in-person interview when applying for SNAP, a burden on those with inflexible jobs and/or without a car.

Mississippi has remained the hungriest state in the United States since 2009, with 19.2% of its population (574,272 people) being food insecure as of 2017. One driving factor for the enduring nature of food insecurity in the state is the high rates of persistent poverty, defined as places that have had poverty rates of 20% or more for over three decades. Over half of Mississippi’s counties (42) are currently in persistent poverty. (CRS). Exacerbating food insecurity in the state is the uniquely strict requirements for the Supplemental Nutrition Assistance Program (SNAP), the program meant to alleviate food insecurity for impoverished families. Families must have an annual income that is at or below 130% of the federal poverty line (MDHS), a low bar that leaves many of the families above this threshold unassisted. Furthermore, Mississippi requires an in-person interview when applying for SNAP, a burden on those with inflexible jobs and/or without a car.

Most Changed

California
California has a mix of very rich and very poor areas, with some rural and very developed locations. This contrast means that it is difficult to point out specific reasons for a decrease in food insecurity between 2009 and 2017. That being said, California is a generally wealthy and progressive state supportive of funding efforts to reduce food insecurity, as evidenced by the greatest decrease in food insecurity in the covered time period among all U.S. states. Another possible explanation for the decrease in food insecurity in California is the influx of wealth and wealthy people reducing the relative number of food insecure people without changing the actual number. This could be a possible confounding factor worth further investigation.

California has a mix of very rich and very poor areas, with some rural and very developed locations. This contrast means that it is difficult to point out specific reasons for a decrease in food insecurity between 2009 and 2017. That being said, California is a generally wealthy and progressive state supportive of funding efforts to reduce food insecurity, as evidenced by the greatest decrease in food insecurity in the covered time period among all U.S. states. Another possible explanation for the decrease in food insecurity in California is the influx of wealth and wealthy people reducing the relative number of food insecure people without changing the actual number. This could be a possible confounding factor worth further investigation.

Louisiana
Like Mississippi, Louisiana has similarly strict state programs that limit access to resources like SNAP (just about double the amount of spending on SNAP). There is also a similar lack of job opportunities in rural areas. Many of the same social factors are shared between Mississippi and Louisiana because of geographic proximity and relatively similar populations (~3m and ~4m, respectively). The rate of food insecurity is actually increasing faster than population is growing, suggesting that in fact there are changes in the economy, or social policies in Louisiana that are causing this growth, rather than it being due to population change. Despite this, it has not caught up to Mississippi, possibly because it is still wealthier.

Like Mississippi, Louisiana has similarly strict state programs that limit access to resources like SNAP (just about double the amount of spending on SNAP). There is also a similar lack of job opportunities in rural areas. Many of the same social factors are shared between Mississippi and Louisiana because of geographic proximity and relatively similar populations (~3m and ~4m, respectively). The rate of food insecurity is actually increasing faster than population is growing, suggesting that in fact there are changes in the economy, or social policies in Louisiana that are causing this growth, rather than it being due to population change. Despite this, it has not caught up to Mississippi, possibly because it is still wealthier.

Conclusion

Limitations of Our Work

Our investigation into our four highlighted states is preliminary. This was especially true for North Dakota, given the lack of prior research on food insecurity in the state. In general, it seems that relatively little effort has been put into considering if North Dakota’s success has been due to well implemented programs, demographic factors or other reasons. Overall, because we wanted to provide only a high level perspective on food insecurity in the U.S., it was impossible to make direct conclusions about the benefits or costs of specific plans or modes of action. Therefore, our explanations are limited to apparent connections between equity and food security.

Another limitation of our work is the timeline of available data. Because of a shift in Feeding America’s methodology of data collection, we decided that it would not be acceptable to compare more recent data (2018-2020) with the data we used (2009-2017). We chose to use the larger of the two subsets of the data (2009-2017) since it seemed better for the type of generalized comparison, and overview that we were interested in. Because of this, we did not attempt to consider the effects of the COVID-19 pandemic on rates of food insecurity, which seems to have worsened food insecurity due to economic struggles and supply chain issues. If we were to continue this project, we would want to examine the impact of the pandemic on food insecurity.

Final Thoughts

Our visualizations suggest that, while food insecurity has generally improved, it is still an issue that strongly affects some regions of the country. Additionally, our research into specific states suggests that many of the underlying causes for food insecurity relate to broader issues of social inequity, which may be part of why it has been a difficult issue to address over the past decade.

By looking more deeply into how food insecurity prevalence in four states has changed over time, we uncovered some potential explanations for why food insecurity has increased, decreased, or remained stagnant between 2009 and 2017. For North Dakota, food insecurity has remained the lowest among all U.S. states likely because of the low unemployment rate, small population, and widespread SNAP access. For Mississippi, food insecurity has remained the highest among all states likely because of high poverty rates and strict requirements for food assistance programs. Louisiana experienced the highest increase in food insecurity between 2009 and 2017, which we hypothesize is because of a lack of attention on the issue in the state government and a consistent lack of job opportunities over the nine year period. And lastly, California experienced a major decrease in food insecurity prevalence between 2009 and 2017 probably because of the influx of wealthy residents moving to the state and increased state funding for addressing food insecurity.

Overall, through this project we gained a better understanding of the broad trends in food insecurity prevalence at the national, state, and local levels from 2009-2017. While we noted an overall decrease in food insecurity across the United States, it is clear that hunger is still a major issue that deserves more attention. Achieving a high level understanding of food insecurity trends is a step towards addressing what is a highly complex and challenging problem.

Methodology

We used plotly to make a U.S. map that is both interactive and animated. plot_geo() initializes a plotly-geo object, essentially an interactive base map in the shape of a chosen location. From there, by adding what plotly calls “traces” on top of the object (add_trace()), we added food insecurity data to correspond with the location data. By adding the additional argument frame = … in the plot_geo() function, we created an animated map with a time slider automatically.

We used gganimate alongside gifski (as a renderer) to create GIFs out of maps created in ggplot2 that display food insecurity data in each state over time. This works by using the function transition_states() within a ggplot2 object, and the function animate(renderer = gifski_renderer) after the object is created.

We used ggiraph to easily make plots interactive. We used this package to overlay an interactive scatter plot on top of an interactive line graph for the line graph showing food insecurity prevalence by state over time. We also used this package to create the interactive horizontal bar graph showing the change in food insecurity prevalence between 2009 and 2017.

References

Packages

  1. Beger A (2021). states: Create Panels of Independent States. R package version 0.3.1, https://CRAN.R-project.org/package=states.
  2. C. Sievert. Interactive Web-Based Data Visualization with R, plotly, and shiny. Chapman and Hall/CRC Florida, 2020.
  3. Firke S (2021). janitor: Simple Tools for Examining and Cleaning Dirty Data. R package version 2.1.0, https://CRAN.R-project.org/package=janitor.
  4. Gohel D, Skintzos P (2022). ggiraph: Make ‘ggplot2’ Graphics Interactive. R package version 0.8.4, https://CRAN.R-project.org/package=ggiraph.
  5. H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2016.
  6. Hafen R (2020). geofacet: ‘ggplot2’ Faceting Utilities for Geographical Data. R package version 0.2.0, https://CRAN.R-project.org/package=geofacet.
  7. Henry L, Wickham H (2020). purrr: Functional Programming Tools. R package version 0.3.4, https://CRAN.R-project.org/package=purrr.
  8. Hester J, Bryan J (2022). glue: Interpreted String Literals. R package version 1.6.2, https://CRAN.R-project.org/package=glue.
  9. Hester J, Kuhn M, Pedersen TL, Miller E, Bache SM, Müller K, Ooms J, Robinson D, Seidel DP, Spinu V, Takahashi K, Vaughan D, Wilke C, Woo K, Yutani H (2019). “Welcome to the tidyverse.” Journal of Open Source Software, 4(43), 1686. doi:10.21105/joss.01686 https://doi.org/10.21105/joss.01686.
  10. Meissner P, Ren K (2020). robotstxt: A ‘robots.txt’ Parser and ‘Webbot’/‘Spider’/‘Crawler’ Permissions Checker. R package version 0.7.13, https://CRAN.R-project.org/package=robotstxt.
  11. Ooms J (2022). gifski: Highest Quality GIF Encoder. R package version 1.6.6-1, https://CRAN.R-project.org/package=gifski.
  12. Pedersen T (2022). transformr: Polygon and Path Transformations. R package version 0.1.4, https://CRAN.R-project.org/package=transformr.
  13. Pedersen T, Robinson D (2022). gganimate: A Grammar of Animated Graphics. R package version 1.0.8, https://CRAN.R-project.org/package=gganimate. R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
  14. Sjoberg D (2020). hablar: Non-Astonishing Results in R. R package version 0.3.0, https://CRAN.R-project.org/package=hablar.
  15. Walker K (2022). tigris: Load Census TIGER/Line Shapefiles. R package version 1.6.1, https://CRAN.R-project.org/package=tigris.
  16. Wickham H (2022). rvest: Easily Harvest (Scrape) Web Pages. R package version 1.0.3, https://CRAN.R-project.org/package=rvest.
  17. Wickham H, François R, Henry L, Müller K (2022). dplyr: A Grammar of Data Manipulation. R package version 1.0.9, https://CRAN.R-project.org/package=dplyr.
  18. Wickham H, Averick M, Bryan J, Chang W, McGowan LD, François R, Grolemund G, Hayes A, Henry L,
  19. Yihui Xie (2022). knitr: A General-Purpose Package for Dynamic Report Generation in R. R package version 1.40.
  20. Yutani H (2022). gghighlight: Highlight Lines and Points in ‘ggplot2’. R package version 0.4.0, https://CRAN.R-project.org/package=gghighlight.

Data

  1. Gundersen, C., Waxman, E., Engelhard, E., Brown, J. Map the Meal Gap 2011: Food Insecurity Estimates at the County Level. Feeding America, 2011.
  2. Gundersen, C., E. Waxman, E. Engelhard, T. Del Vecchio, & A. Satoh. Map the Meal Gap 2012: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2010. Feeding America, 2012.
  3. Gundersen, C., E. Waxman, E. Engelhard, A. Satoh, & N. Chawla. Map the Meal Gap 2013: Food Insecurity Estimates at the County Level. Feeding America, 2013.
  4. Gundersen, C., E. Engelhard, A. Satoh, & E. Waxman. Map the Meal Gap 2014: Food Insecurity and Child Food Insecurity Estimates at the County Level. Feeding America, 2014.
  5. Gundersen, C., E. Engelhard, A. Satoh, & E. Waxman. Map the Meal Gap 2015: Food Insecurity and Child Food Insecurity Estimates at the County Level. Feeding America, 2015.
  6. Gundersen, C., A. Dewey, A. Crumbaugh, M. Kato & E. Engelhard. Map the Meal Gap 2016: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2014. Feeding America, 2016.
  7. Gundersen, C., A. Dewey, A. Crumbaugh, M. Kato & E. Engelhard. Map the Meal Gap 2017: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2015. Feeding America, 2017.
  8. Gundersen, C., A. Dewey, A. Crumbaugh, M. Kato & E. Engelhard. Map the Meal Gap 2018: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2016. Feeding America, 2018.
  9. Gundersen, C., A. Dewey, M. Kato, A. Crumbaugh & M. Strayer. Map the Meal Gap 2019: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2017. Feeding America, 2019.
  10. Gundersen, C., A. Dewey, E. Engelhard, M. Strayer & L. Lapinski. Map the Meal Gap 2020: A Report on County and Congressional District Food Insecurity and County Food Cost in the United States in 2018. Feeding America, 2020.
  11. Gundersen, C., Strayer, M., Dewey, A., Hake, M., & Engelhard, E. (2021). Map the Meal Gap 2021: An Analysis of County and Congressional District Food Insecurity and County Food Cost in the United States in 2019. Feeding America.
  12. Gundersen, C., Strayer, M., Dewey, A., Hake, M., & Engelhard, E. (2022). Map the Meal Gap 2022: An Analysis of County and Congressional District Food Insecurity and County Food Cost in the United States in 2020. Feeding America.

Informational sources

  1. Congressional Research Service (2022), “The 10-20-30 Provision: Defining Persistent Poverty Counties.”
  2. “Overall (all ages) Hunger & Poverty in the United States | Map the Meal Gap” (n.d.). Available at https://map.feedingamerica.org.
  3. Smith, Z. (n.d.). “Pandemic Is Exacerbating Mental Health Problems in Food-Insecure Americans | Everyday Health,” Available at https://www.everydayhealth.com/coronavirus/pandemic-exacerbating-mental-health-problems-in-food-insecure-americans/.
  4. Stewart, M. (n.d.). “SNAP,” Mississippi Department of Human Services.
  5. “USDA ERS - Definitions of Food Security” (n.d.). Available at https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-u-s/definitions-of-food-security/.