Kenosha and Racine sit next to each other on Interstate 94, share a labor market, and share a supermarket supply chain. The federal government's 2019 food-desert map says Kenosha doubled its desert count while Racine halved it. The federal retailer records, current through December 2025, tell a different story: both counties are losing supermarkets — Racine has lost fifteen supermarket-type stores in the five years since the map was last updated. A short glossary at the bottom of this page explains every program and acronym used here. Toggle between the two counties below to see the evidence for each.
Pritosh Kumar · UW–Parkside SCMLast updated: April 2026Retail data current through December 2025Demographics: U.S. Census 5-year survey, 2019–2023
County view
Click to switch — the map, stats, and store timeline all update.
The Question
Why are Kenosha's food deserts getting worse while the rest of Wisconsin is improving?
Between 2010 and 2019, Wisconsin lost 23 food desert tracts. Most counties improved or held steady. Kenosha doubled.
Wisconsin statewide 2010–2019
−15%
156 food desert tracts in 2010 → 133 in 2019. Milwaukee −7, Racine −4, Brown −3, Dane −2.
Kenosha County 2010–2019
+100%
4 food desert tracts in 2010 → 8 in 2019. The opposite direction from almost everywhere else.
Kenosha's increase from 4 to 8 desert tracts ranks among the most-worsened of Wisconsin's 72 counties — tied with Rock County for the largest absolute increase.
The Paradox
Racine's food access map looks better. Its food supply chain is collapsing.
Two data sources tell opposite stories. One is six years old; the other is current.
Federal food-desert map, 2010 → 2019
8 → 4
The official food-desert count halved. Five tracts were reclassified as recovered. Only one new desert tract appeared.
Federal SNAP retailer records, 2020–2025
−15
Closure records for supermarket-type stores in Racine County since the last food-desert map update. This represents 14 unique locations — Olympic Liquor and Food re-authorized and re-closed inside three months in 2021. Both Racine Piggly Wigglys on the same day. Both Save A Lots inside eighteen months. Both Kmarts in a five-week window.
If you believe the 2019 federal food-desert map, Racine is fine. If you believe the federal retailer records, Racine is already mid-collapse.
Who Lives There
27,808 residents. Half the income. Twice the poverty.
The eight Kenosha food desert tracts share a profile: significantly lower incomes, more than double the poverty rate of non-desert tracts, and more than twice the share of households without a vehicle.
Median household income
$37,797
vs. $70,516 in non-desert Kenosha tracts
Poverty rate
24.9%
vs. 10.5% in non-desert Kenosha tracts
Households without a vehicle
11.9%
vs. 5.3% in non-desert Kenosha tracts
The Prediction
Four of five "recovered" tracts have since lost a supermarket.
The 2019 federal food-desert map reclassified five Racine neighborhoods (census tracts) from food desert to recovered. We pulled the federal retailer records through December 2025 and checked what happened after.
"Recovered" tracts that have since lost a supermarket
4 of 5
Census tracts 2, 3, 5, and 6 each lost at least one supermarket-type store between 2020 and 2025. Only tract 15.04 — a higher-income neighborhood in Mount Pleasant — kept its retail base intact. Compared to the rest of the county (where 7 of the other 39 tracts were affected), this concentration is unlikely to be coincidence (Fisher's exact test, p = 0.011 — less than 1.1 percent chance the pattern is random).
Total Racine tracts hit by a supermarket exit since 2019
11
The exits are not concentrated in the four current deserts. They are spreading through adjacent low- and moderate-income tracts, including every recovered tract except one and two persistent-desert tracts.
The next federal food-desert map update, whenever the U.S. Department of Agriculture releases it, is almost certain to move in the wrong direction. The retail conditions it will measure are not the 2019 conditions. They are the conditions after fifteen supermarket-type closures, with the highest concentration of losses sitting on top of the same neighborhoods the current map classifies as healthy.
The Footprint
The eight food desert neighborhoods of central Kenosha
Click any neighborhood to see its current store inventory, population, and food-desert status. Store counts include every retailer authorized to accept federal food-assistance benefits as of December 2025.
Persistent food desert (2010–2019)
Became food desert
Recovered
Never desert
Supermarket exit since 2020
The Receipts
Four traceable retail-loss pathways
Each event has a date, a store name, and an address. None of this is statistical inference.
Tract 16 (GEOID 55059001600)
Pick 'N Save 6850 closed in November 2017
By 2019 the neighborhood was classified as a food desert. Current stores: two convenience stores. Zero supermarkets.
Tract 21 (GEOID 55059002100)
Festival Foods (Super Store) closed in December 2014
Persistent food desert since 2015. Current stores: two convenience stores. Zero supermarkets.
Tract 9 (GEOID 55059000900)
Lost three supermarkets: Fonte (2005), Quality Grocery (2016), San Luis (2024)
Persistent desert throughout. Current stores: five convenience stores. Zero supermarkets, zero groceries.
Tract 18 (GEOID 55059001800)
Morelli's Deli closed in December 2025 — the neighborhood's last food retailer of any kind
This neighborhood had exactly one food retailer authorized to accept federal food-assistance benefits in its entire history. When Morelli's closed, it became the only Kenosha neighborhood with zero food retailers of any kind.
The Four Current Deserts
What the 2019 map still correctly identifies.
Tract 8 (GEOID 55101000800) Central Racine, west of Interstate 94
The new 2019 desert — still has zero supermarkets in 2025
Three convenience stores. Zero supermarkets. 30 percent Black, 24 percent Hispanic. The household-income profile has improved (poverty rate 7.5 percent), but the retail footprint has not.
Population 4,822 · Median household income $74,545 · 0 supermarkets / 3 convenience
Tract 12.01 (GEOID 55101001201)
Persistent desert since 2010 — still no supermarket
Majority-Hispanic neighborhood (34 percent). Three active retailers: two convenience stores and one combination grocery/other. No supermarket-type store inside the neighborhood.
Population 3,478 · Median household income $56,450 · 0 supermarkets / 2 convenience
Tract 13.02 (GEOID 55101001302)
Persistent desert — one supermarket, six convenience stores, two more closures since 2019
The only desert neighborhood with an active supermarket. The store mix is convenience-dominated and has lost two more supermarket-type stores in 2020–2025. Majority-minority population.
Population 3,458 · Median household income $55,398 · 1 supermarket / 6 convenience · 2 closures since 2019
Tract 9800 (GEOID 55101980000)
A flagged-but-institutional area — should not drive general-population policy
The federal food-desert map flags this area as low-income and low-access, but the U.S. Department of Agriculture's own data file marks it as a "group quarters" location (more than half of residents are in institutional housing such as a prison, dormitory, or care facility). The neighborhood also reports 51 occupied housing units against 727 residents — a ratio of about 14 residents per household, which is characteristic of an institutional facility, not a general-population neighborhood. Zero food retailers. Retained in the four-desert count only for transparency; excluded from policy framing.
727 reported residents · 51 occupied housing units · Group-quarters flag set · 0 active food retailers
What Has Happened Since 2019
The federal food-desert map is six years stale. The story has moved on.
The U.S. Department of Agriculture has not updated its Food Access Research Atlas since 2019. Kenosha has lost nine more supermarket-type stores in the six years since, including three in 2025 alone.
Gordon Food Service · Piggly Wiggly 204 · Alwaha Middle Eastern Grocery
Three supermarket-type stores closed between March and July 2025
The Receipts
Fifteen closure records, fourteen unique stores, five years.
Every row is a date, a store, and an address. None of this is statistical inference. Olympic Liquor and Food appears twice in 2021 because it briefly re-opened then closed again inside a three-month window — the store count is fourteen unique locations.
2020
Kmart 3774 · Sentry Foods 2816 · Kmart 3851
Both Kmarts plus a Burlington Sentry — three Super Stores in thirty-five days
Both Racine Piggly Wigglys closed the same day — August 9, 2021
2022
Supermercado Gran Morelos · Tops Matrangas · Gooseberries Fresh Market · Lee's Deli
Four closures across Racine and Burlington, including an independent Latino-serving supermarket
2023
Supermercado Jimenez · Save A Lot 61326
First of two Save A Lot closures
2024
Max Magic Supermarket · Save A Lot 61327
Second Save A Lot closes Nov 24, 2024 — both Racine Save A Lots gone in under 18 months
The pattern is not random turnover. Chain supermarkets closed in pairs within short windows. Both Kmarts in thirty-five days. Both Piggly Wigglys on a single day. Both Save A Lots across eighteen months.
The Numbers
Two charts that change the framing
Top: supermarket-type exits per year in this county. Bottom: desert vs non-desert demographic comparison.
Supermarket-type exits by year
SNAP Retailer Locator
Desert vs non-desert composition
ACS 5-year
Side by Side
The same mechanism at a different stage.
Racine and Kenosha sit next to each other on I–94, share a labor market, and share a supermarket supply chain. Both counties are roughly 80% convenience-store-dominated on the SNAP retailer side. The FARA trajectory looks opposite on paper. The retail trajectory underneath tells the same story.
Indicator
Kenosha County
Racine County
Total neighborhoods (census tracts)
35
44
Food deserts in 2010
4
8
Food deserts in 2019 (most recent federal map)
8
4
Supermarket-type closures 2020–2025
9
15 records (14 unique stores)
Stores authorized for federal food benefits (Dec 2025)
117
143
Convenience-store share of (supermarket + convenience)
81%
78%
Residents in current desert neighborhoods
27,808
11,758*
* Racine figure excludes the ~1,000 residents of an institutional tract (number 9800; the federal data flags it as group quarters such as a prison or care facility). Including that tract gives 12,719.
Food Environment Index, 0–10 (higher = better)
8.1
8.8
Wisconsin rank (1 = worst of 71 counties)
15
47
Share of residents with limited access to healthy foods
9.5%
3.7%
FoodShare recipients (March 2026)
20,157
28,122
Public-school students who are economically disadvantaged
53.5% (Kenosha Unified)
60.8% (Racine Unified)
Kenosha is where the map already broke. Racine is where it will break in the next FARA vintage.
Wisconsin Sources Confirm It
Three independent state-level datasets tell the same story.
The federal FARA + SNAP retailer pipeline is the spine of this analysis. Three Wisconsin-specific data sources cross-check the picture from different angles — access, recipients, and student-level need — and converge on the same diagnosis.
1. County Health Rankings — Food Environment Index (2025 release)
The University of Wisconsin Population Health Institute publishes the County Health Rankings each year. Their Food Environment Index is a 0–10 score for every U.S. county that combines two pieces: the share of low-income residents who live far from a supermarket, and the share of residents who report not having reliable access to enough food. Higher is better.
Kenosha County
8.1
Ranks 15th worst of 71 Wisconsin counties — bottom 21 percent of the state. 9.5 percent of residents (~15,900 people) classified as limited-access. Food-insecurity rate 10.9 percent.
Racine County
8.8
Ranks 47th of 71 — near the Wisconsin state average of 8.5. 3.7 percent limited-access (~7,300 people), one-third of Kenosha's rate. Food-insecurity rate 10.7 percent — nearly identical to Kenosha.
Read this carefully: Racine has fewer people in limited-access geography (3.7 vs 9.5 percent) but almost the same food-insecurity rate (10.7 vs 10.9 percent). The implication: Kenosha's deficit is on the access side. Racine's residents reach a supermarket but still cannot afford it; Kenosha's residents face both barriers stacked. This is the supply-chain channel the federal data alone cannot resolve.
2. FoodShare enrollment in Wisconsin (March 2026)
FoodShare is Wisconsin's name for the federal food-assistance program known nationally as SNAP (the Supplemental Nutrition Assistance Program). The Wisconsin Department of Health Services publishes monthly county-level FoodShare enrollment counts, which give us actual recipient totals rather than dollar-flow estimates.
Kenosha County · March 2026
20,157
FoodShare recipients across 10,869 households. About 12 percent of the county population. Modest decline from January 2025 (about 21,000).
Racine County · March 2026
28,122
Across 14,494 households. About 14 percent of the county population. Modest decline from January 2025 (about 29,000).
Combined, roughly 48,000 residents across the two counties depend on FoodShare each month. They are the population for whom the shift from supermarkets to convenience stores documented above is most consequential, because food-assistance benefits redeemed at a convenience store buy markedly less fresh produce per dollar than the same benefits redeemed at a supermarket.
3. Public-school economic status (2024–25 school year)
Wisconsin's Department of Public Instruction publishes a school-level dashboard called WISEdash. It reports the share of students classified as economically disadvantaged (household income at or below 185 percent of the federal poverty line) at every public school in the state. This is a leading indicator of household food insecurity and a useful cross-check on the desert-neighborhood findings above.
Kenosha Unified School District
53.5%
9,795 of 18,315 students classified as economically disadvantaged (2024–25 school year). 8 of the district's 31 schools have rates of 70 percent or higher. Highest: Brass Community School at 87.9 percent (406 students).
Racine Unified School District
60.8%
9,589 of 15,776 students classified as economically disadvantaged (2024–25). 8 of the district's 24 schools have rates of 70 percent or higher — one in three. Highest: Knapp Elementary at 85.7 percent (396 students).
Both districts have student bodies in which more than half of children qualify as economically disadvantaged. The Racine district's share is meaningfully higher (60.8 vs 53.5 percent), and the count of high-need schools is similar in absolute terms across districts despite the Racine district being smaller. Hunger inside school buildings tracks food deserts outside them — both districts have elementary schools where roughly nine in ten students live below 185 percent of the poverty line.
Honest Limitations
What this analysis does not claim
This project does not claim a strict cause-and-effect chain from supermarket closures to food deserts. An earlier statistical model on 430 Wisconsin supermarket closures did not survive a methodological audit for known issues with staggered treatment timing and contaminated comparison groups. The corrected version of that model showed no statistically significant effect. We re-framed the project as a descriptive case study, not a causal claim.
What does survive: a placebo test against the other 71 Wisconsin counties, the most recent Census 5-year survey demographic re-check, the verified store-by-store closure timelines for both counties, and the neighborhood-level diagnoses. The forward claim for Racine — that the next federal food-desert map update will push the county back toward or past its 2010 desert count — is a testable prediction, not a causal assertion.
Data-vintage caveat: The 2019 federal food-desert map uses the 2010 boundaries for U.S. Census neighborhoods (tracts). The most recent Census 5-year survey (covering 2019–2023) uses the 2020 boundaries. Five 2010-boundary Racine neighborhoods, including one of the "recovered" ones (number 3), do not have a direct match in the 2023 data and fall back to the 2018–2022 values on this page. A formal 2010-to-2020 boundary crosswalk is a noted follow-up; the four current desert neighborhoods driving the narrative have data in both vintages.
Glossary & Abbreviations
What every acronym on this page means.
This page is written for a general audience. Federal data programs and Wisconsin agencies all use their own short names; the full names are below. Each glossary entry includes the source agency.
Federal programs & data
Food desertA census tract (neighborhood) where a substantial share of residents are both low-income and live more than 1 mile from a supermarket (10 miles in rural areas). Definition set by the U.S. Department of Agriculture. Often abbreviated as "LILA" (Low-Income, Low-Access).
Food Access Research Atlas (FARA)The federal food-desert map. Published by the U.S. Department of Agriculture. Updated in 2010, 2015, and 2019. The 2019 release is still the most recent.
SNAP — Supplemental Nutrition Assistance ProgramThe federal food-assistance program (formerly known as food stamps). Administered by the U.S. Department of Agriculture. Wisconsin's name for the same program is FoodShare.
SNAP Retailer LocatorA federal database listing every store authorized to accept SNAP / FoodShare benefits, with opening and closing dates. Updated continuously. The version used here runs through December 2025.
EBT — Electronic Benefits TransferThe plastic card SNAP / FoodShare recipients use to pay for groceries. The card works only at retailers authorized by the U.S. Department of Agriculture.
U.S. Census ACS — American Community SurveyThe U.S. Census Bureau's annual rolling household survey. Estimates published in 1-year and 5-year averages. This page uses the 5-year averages because the 1-year version is too noisy at the neighborhood level.
FIPS codeFederal Information Processing Standards code. A federal numbering system for U.S. states and counties. Kenosha County is FIPS 55059; Racine County is FIPS 55101 (the leading 55 is Wisconsin).
Census tractA statistical neighborhood drawn by the U.S. Census Bureau, typically containing 1,200–8,000 residents. Both counties are subdivided into dozens of tracts. Tract boundaries are redrawn after each decennial census; this matters because the 2019 federal food-desert map uses 2010 boundaries while newer Census surveys use 2020 boundaries.
Wisconsin agencies & programs
FoodShareWisconsin's name for SNAP. Same program, same federal funding, state-level administration. Run by the Wisconsin Department of Health Services.
Wisconsin DHS — Department of Health ServicesThe state agency that administers Medicaid, FoodShare, and most other public-health programs in Wisconsin.
Wisconsin DPI — Department of Public InstructionThe state agency for K–12 public education. Publishes the WISEdash dashboard.
WISEdashWisconsin Information System for Education Data Dashboard. The state's public reporting portal for school-level enrollment, demographics, and outcomes data.
KUSD — Kenosha Unified School DistrictThe public school district covering most of Kenosha County. About 18,000 students across 31 schools (2024–25).
RUSD — Racine Unified School DistrictThe public school district covering most of Racine County. About 15,800 students across 24 schools (2024–25).
Economically disadvantagedWisconsin's school-level poverty indicator. A student is classified as economically disadvantaged if household income is at or below 185 percent of the federal poverty line. Replaces the older "free and reduced-price lunch" classification, which is no longer reported the same way for districts that use Community Eligibility Provision schoolwide meal programs.
County Health Rankings (CHR)An annual ranking of every U.S. county on dozens of health indicators, published by the University of Wisconsin Population Health Institute and the Robert Wood Johnson Foundation. The Food Environment Index is one of the included measures.
Food Environment Index (FEI)A 0-to-10 score (10 = best) combining two pieces: the share of low-income residents who live far from a supermarket, and the share of all residents who report food insecurity. Calculated by the County Health Rankings team.
Methodological & technical terms
Persistent food desertA neighborhood classified as a food desert in both 2010 and 2019 (and 2015 if data exists). Used in this page to distinguish from "recovered" or "newly worsened" trajectories.
Recovered tractA neighborhood that the federal food-desert map classified as a food desert in 2010 but no longer does in 2019. "Recovery" can be from new retail opening within 1 mile, or from demographic improvement (incomes rising above the low-income threshold), or from a measurement boundary shift — not necessarily a real food-access improvement.
Worsened tractA neighborhood that was not a food desert in 2010 but became one by 2019. These are the "new" food deserts of the past decade.
Group quartersA U.S. Census category for institutional living arrangements: prisons, dormitories, nursing homes, military barracks, etc. Tracts where most residents live in group quarters need different policy treatment than general-population neighborhoods.
Fisher's exact testA standard statistical test for whether a pattern in a small sample (such as "4 of 5 recovered tracts lost a supermarket") is unlikely to be due to chance. The "p-value" reported (p = 0.011) is the probability that the pattern would appear if neighborhoods were affected at random; smaller is stronger evidence.
Margin of error (MOE)The U.S. Census ACS publishes margins of error alongside its estimates because they come from a sample, not a complete count. A change in a neighborhood's median income from one ACS release to the next can be within the margin of error and therefore not statistically meaningful.
Difference-in-differences (DiD)A statistical research design that compares the change in an outcome over time between a treated group and a similar untreated group. An earlier version of this project used a difference-in-differences model that did not survive methodological review; the current page is descriptive, not causal.
People, places & programs cited
Interstate 94 (I-94) corridorThe federal interstate highway running through both Kenosha and Racine counties along the Lake Michigan shore from Milwaukee to Chicago. Both county seats sit east of I-94.
Mount PleasantA Village in Racine County, west of the city of Racine. Site of the Foxconn / Microsoft technology campus and one of the higher-income areas in the county.
BurlingtonA city in far western Racine County (FIPS-wise, the same county but geographically distant from the city of Racine). Two of the post-2019 supermarket-type closures were Burlington stores rather than central-Racine stores.
Pick 'N Save / Piggly Wiggly / Sentry Foods / Save A Lot / Festival Foods / KmartRegional and national supermarket and discount-store chains that have closed locations in Kenosha or Racine since 2014. The page lists each closure individually with date and location.
UW-ParksideUniversity of Wisconsin–Parkside, a public university in Kenosha County. The author is on the faculty there.
Federal poverty lineAn income threshold set annually by the U.S. Department of Health and Human Services. For a family of four in 2025, 100 percent of the federal poverty line was about $32,150; 185 percent (the WISEdash threshold) was about $59,500.
Data Sources & References
Where the numbers come from
Every statistic on this page is traceable to a named, publicly available source.
University of Wisconsin Population Health Institute — County Health Rankings & Roadmaps2025 v3 release · Food Environment Index, Limited Access to Healthy Foods, Food Insecurity · all 72 WI counties · countyhealthrankings.org
Wisconsin Department of Health Services — FoodShare EnrollmentMonthly county-level recipient and assistance-group counts, January 2025 through March 2026 · dhs.wisconsin.gov/foodshare
Wisconsin Department of Public Instruction — WISEdash Public PortalSchool-level economically-disadvantaged share for KUSD and RUSD, 2024–25 and 2025–26 school years · wisedash.dpi.wi.gov
Primary Data — Demographics
US Census ACS 5-Year EstimatesKenosha: ACS 2015–2019 and 2018–2022 · Racine: ACS 2015–2019, 2018–2022, and 2019–2023 (released Dec 2024) · census.gov/acs
Census TIGER/Line Shapefiles (2019)Tract boundaries for spatial joins · census.gov/tiger
Published Evidence — Interventions
The Food Trust — Philadelphia Healthy Corner Store InitiativeYear 3 report: +60% produce sales at participating stores · thefoodtrust.org
RAND PHRESH Project (Pittsburgh Hill District)Natural experiment around a new supermarket opening in a former food desert · rand.org/phresh
Reproducibility
Analysis pipeline — KenoshaPolars/PyArrow ingestion of FARA + SNAP locator + ACS 2015–2019 / 2018–2022; tract-level spatial joins via R sf and tigris. Full code, intermediate Parquet/CSV datasets, and the Kenosha GeoJSON embedded on this page are available on request.
Analysis pipeline — RacineCompanion build adding ACS 2019–2023 5-year (pulled via tidycensus), recovery-indicator joins, post-2019 supermarket-exit table, and the Racine GeoJSON. Available on request.
Visualizations use Leaflet.js with OpenStreetMap/CARTO basemaps and Chart.js. Spatial joins and figure production in Python (polars) and R (sf, tigris, tidycensus).
AI disclosure: Data acquisition, analysis code, figure production, and document drafting for this project were performed with assistance from Anthropic's Claude via Claude Code, under direct supervision by the principal investigator, including an internal two-reviewer methodological audit. All numerical claims have been independently verified against raw USDA Food Access Research Atlas, USDA SNAP Retailer Locator, US Census ACS, and FBI Uniform Crime Reporting sources. A complete audit log and reproducible analysis scripts are available on request.