Measuring the Impact of CDBG Spending on Urban Neighborhoods

Measuring the Impact of CDBG Spending on Urban Neighborhoods We empirically investigate Community Development Block Grant (CDBG) Program spending in census tracts in 17 large cities. Regression analysis reveals several strong statistical associations between CDBG spending during 1994-96 and changes between 1993-94 and 1998-99 in three indicators that capture a wide range of neighborhood conditions, though their magnitudes are heavily contingent. First, there is no consistent association between CDBG spending levels and indicators of subsequent neighborhood change unless CDBG expenditures: (1) are sufficiently spatially targeted that they exceed a threshold of the sample mean expenditure level, and (2) are standardized for the number of poor residents in the neighborhood. Second, associations vary according to neighborhood trajectories prior to investment and contemporaneous changes in the local economy. Nevertheless, even in the leasthospitable contexts—highly concentrated neighborhood poverty, pre-existing declines in neighborhood home values, weak city job growth—our estimates are consistent with the hypothesis that CDBG spending at above-threshold amounts produces significant neighborhood improvements according to multiple measures. We discuss implications for spatially targeting CDBG monies and connections of our work with emerging literature on the dynamics of poor neighborhoods.