Using Linked Administrative Data to Profile a City’s Rental Stock and Landlords and Guide a Lead-Safe Housing Initiative

Claudia Coulton
Michael Henderson
Francisca García-Cobián Richter
Jeesoo Jeon
April Urban
Michael Schramm
Robert L. Fischer
Case Western Reserve University

By the time they enter kindergarten, an estimated 25 percent of Cleveland, Ohio, children have at least one test showing an elevated blood lead level, and to address this high rate of lead exposure at its source, the city committed to a lead-safe housing strategy. Most families with young children rent homes in the private rental market, making these properties and their owners key elements in moving forward on a lead-safe agenda. This article describes how parcel data, property tax rolls, deed and foreclosure records, housing code violations, rental registry information, building permits, evictions, and Housing Choice Voucher program records were used to evaluate lead risk in the rental housing stock and develop a typology of landlords. Deterministic and probabilistic methods were used to link the property data sources, resulting in the identification of 103,386 rental units, 54,786 rental properties, and 36,659 landlords for the analysis. More than one-third of the rental properties were found to be at high risk of failing to meet lead safety standards. A latent class analysis uncovered three classes of landlords, characterized as having different capabilities to comply with the lead safety ordinance. Small, under-resourced landlords who would likely require the highest level of support from the lead safety coalition owned approximately 25 percent of the rental properties. This study guided the lead-safe Cleveland strategy and is being updated to evaluate progress toward reducing lead hazards in rental housing.