Rent Subsidies

Abstract

Several studies have evaluated the experimental housing allowance program authorized by Congress in 1970 and implemented by the Department of Housing and Urban Development. The studies that are relied upon most have nevertheless been considered inadequate because they were essentially case studies. An evaluation design has recently been suggested, the random-comparison-group design, that makes the data from the Administrative Agency Experiment useful for evaluation. The outcomes in the AAE program sites are compared here to projected outcomes based on an analysis of Annual Housing Survey data. The evaluation finds that rent burdens were reduced and housing quality was improved jor many who lived in substandard units. These impacts were achieved without producing inflation. Substantial confidence in the findings is permitted not only by the logic of the method itself, but by corroboration from other research. The design employed is found to be practical and powerful for certain evaluations, and its use in connection with other public programs can be recommended. The random- comparison-group design could substitute for randomized experiments in situations in which such controlled experimentation is not possible or was not carried out.