Level 0 of the digital divide: Algorithmic redlining and online food deserts in Santiago de Chile

While research on the digital divide has traditionally focused on individual access, skills, and benefits, this study proposes a “Level 0” conceptual framework to account for the institutional and supply-side barriers that precede individual agency. Using a reverse-geocoding and web-scraping methodology, we conduct a geospatial market analysis of the three leading food delivery apps (FDAs) in the Santiago Metropolitan Area—PedidosYa, Uber Eats, and Rappi. We investigate how corporate business models and “unwritten rules” produce digital redlining, effectively creating “online food deserts” in specific urban territories. The results of a Binary Spatial Autoregressive (BSAR) model demonstrate that territorial exclusion is a collective geographic phenomenon driven by socioeconomic signals (i.e. residential density and educational levels) and institutional factors (i.e. the role of informal labour), such as residential density and educational levels, rather than the “objective risk” of crime often cited by corporations. By identifying this “Level 0” institutional divide, the study contributes to a comparative theory of digital inequality in the Global South and underscores the need for policy interventions that regulate the supply-side logic of the platform economy. • New framework maps supply-side barriers that limit digital agency before individuals can even act. • Uses novel reverse-geocoding and scraping to map territorial exclusion within Santiago’s food delivery market. • Exclusion is driven by socioeconomic signals like education and density rather than objective local crime rates.