More Stores, Less Produce: Philadelphia’s Uneven Food Environment

April 29, 2026

PSCI 1800 - Lily Gessel

When people talk about food deserts, the conversation usually centers on the absence of grocery stores. But in Philadelphia, the data tells a different story, one not of scarcity, but of saturation. In lower-income neighborhoods across the city, residents don’t necessarily have fewer supermarkets than their wealthier neighbors, but rather more competition from unhealthy alternatives.

This analysis draws on two data sources: the Neighborhood Food Retail and Access dataset published by the Philadelphia Department of Health, which catalogs store types across the city’s census block groups, and US Census data measuring median household income by block group.

The city of Philadelphia defines a high poverty area as 20% or more of the block group being below the federal poverty level.

High-produce supply stores, like supermarkets and big box stores, generally offer a larger amount of healthier foods. Low-produce supply stores, like gas stations and dollar stores, have a higher amount of unhealthy food choices, like sugary beverages and sugary and salty snacks.

It is evident from the graph above, that in both high-poverty and low-poverty areas, low-produce stores outnumber high-produce stores. But the degree of that imbalance tells a more revealing story. High-produce options make up only 9.4% of all food retail in high-poverty areas, versus 14.9% in lower-poverty ones. While in both areas the average amount of high-produce options in roughly the same, high poverty areas are saturated in low-produce options.

The average number of high-produce stores per 1,000 residents is nearly identical in both area types - roughly five stores. What differs sharply is the low-produce store count. High-poverty block groups average 36 such stores per 1,000 residents, compared to just 25 in lower-poverty areas.

It is evident from the graph above that as more of the population is below the federal poverty line, or in a high poverty area, the number of low produce stores increases. Low-produce stores show a moderate positive relationship with poverty (correlation: 0.26), meaning they cluster more heavily in lower-income areas.

In contrast, the graph below shows the number of high-produce stores remains stagnant despite a change in poverty level. If grocery access were purely a poverty story, we’d expect that number to fall sharply. It doesn’t.

The above graph features outlier block groups, with low-poverty but exceptionally high amounts of high-produce stores. These block groups correspond to dense, tourist-frequented neighborhoods like Rittenhouse Square and Bella Vista, where heavy foot traffic makes large grocery retailers economically viable. Meanwhile, some low-poverty areas on the city’s outskirts show low store counts simply because their populations are more spread out - meaning fewer stores serve a larger area.

The relationship between income and food access comes into sharper focus in the graph below. As median household income in a block group rises, the share of nearby stores that are high-produce increases steadily. This confirms that income shapes not just how many healthy stores exist, but what proportion of the overall food environment they represent. For a resident doing a weekly shop, that ratio matters enormously, as it determines whether a healthy choice is easy or an effort.

Because this data covers all of Philadelphia’s census block groups, these patterns are representative of the city as a whole, not just isolated pockets. Across neighborhoods it is evident that poverty doesn’t reliably predict fewer supermarkets, but it does predict a food environment crowded with low-produce alternatives.

Improving food access in Philadelphia, then, may be less about attracting new grocery stores to underserved areas and more about grappling with why low-produce retail concentrates so heavily where residents are most economically vulnerable, and what that saturation costs them.