The Food Desert Myth: Evidence Beyond Proximity

The Measure Being Tested

The Atlas does not reduce food access to one proximity measure. It provides a family of measures that can represent different access constraints.

This article publishes a Santa Clara County comparison using one distance-based measure alongside SNAP participation, a program-use proxy. If a matched rerun reproduces it, the result would motivate tests of distance, vehicle access, transit, eligibility, outreach, and take-up.

"The article reports zero federal food deserts and 4.7x SNAP variation; both figures are under reconciliation."

Published analysis of 6,613 API records, under reconciliation

Methodology

The published article describes 6,613 store records as a Santa Clara County dataset. The current audit indicates that count may represent raw API records across a broader geography, so the store population is under reconciliation.

The article describes calculations for 372 county census tracts. Other project surfaces name 408 tracts, so the tract vintage and inclusion rules are under reconciliation.

• Distance to nearest supermarket (straight-line and road network)
• Number of stores within 0.5, 1, and 3 miles
• A published 2.7 million origin-destination candidate-pair count; the executed routing record is under reconciliation
• SNAP enrollment rates from California Department of Social Services
• Demographic and economic indicators from the American Community Survey

This approach tests whether one distance-based access measure is associated with SNAP participation after accounting for income and demographic factors. SNAP participation is not a direct food-security outcome, and the current public materials do not reproduce the model.

Published Findings Under Reconciliation

Finding 1: Published Zero-Food-Desert Claim

The article reports zero Santa Clara County tracts meeting the USDA food-desert definition. The store and tract populations are under reconciliation, so this claim should not yet be treated as verified.

If confirmed, that comparison would show why proximity alone cannot settle the food-access question. The current audit has not yet confirmed it.

Finding 2: Published SNAP Comparison

The article reports the following SNAP participation range among selected tracts with similar poverty rates and demographic profiles. The population and comparison are under reconciliation. These values describe program use, not direct food insecurity:

• Lowest participation: 12% of eligible households enrolled
• Highest participation: 56% of eligible households enrolled
• Published variation factor: 4.7x, under reconciliation

The article attributes this variation to factors beyond distance. That interpretation remains provisional until the comparison is reproduced from a matched public record.

Finding 3: Published Transit Hypothesis

The article reports a relationship between transit travel times and SNAP enrollment after controlling for income. The current public materials do not reproduce that model, so the relationship should not be treated as verified.

The article uses “mobility deserts” for locations where stores are technically nearby but practically unreachable for residents without cars.

Policy Implications

If the published comparisons survive reconciliation, they would motivate four hypotheses for further testing. They do not rank interventions:

1. Which Atlas measure matches the policy question? A specification should name the income flag, distance threshold, vehicle-access field, and public-transportation assumption before estimating a relationship.
2. Does transit add information beyond distance? A reproduced model is required before treating transit access as a mechanism.
3. What does SNAP participation represent? Variation can reflect eligibility, need, outreach, and take-up. It is not a direct food-insecurity outcome.
4. Does an independent outcome support the claim? A food-security measure not used in the access definition is required before drawing an outcome conclusion.

Limitations

This analysis has several limitations:

• Santa Clara County is wealthier than most U.S. counties; findings may not generalize to rural or high-poverty urban areas.
• We use SNAP enrollment as a proxy for food security; not all food-insecure households are SNAP-eligible.
• Transit calculations assume weekday daytime travel; evening and weekend access may differ.
• We cannot distinguish between enrollment friction and deliberate non-enrollment.

Conclusion

The article's empirical comparisons remain under reconciliation. If reproduced, they would motivate tests of distance, vehicle access, transit, eligibility, outreach, and take-up.

The present evidence does not rank interventions or establish that store placement is less important than another access mechanism.

Frequently Asked Questions

Are there food deserts in Santa Clara County?

The article reports zero Santa Clara County tracts meeting the USDA definition. The store and tract populations are under reconciliation, so the claim should not yet be treated as verified.

How much do SNAP enrollment rates vary across Santa Clara County?

The article reports a 4.7x difference across selected tracts. That article-reported comparison is under reconciliation. SNAP participation measures program use, not food insecurity directly.

What data was used to analyze food deserts in Santa Clara County?

The published article names 6,613 store records, SNAP data, 372 tracts, and a 2.7 million origin-destination pair count. Those populations and the routing record are under reconciliation.

What might explain different SNAP participation rates among tracts with nearby stores?

The article treats transit access, eligibility, need, outreach, and take-up as hypotheses. The current public materials do not reproduce a model that can rank those mechanisms.

What is a mobility desert?

Mobility deserts are locations where grocery stores are within 1 mile but practically unreachable for residents without cars due to poor transit access.