U.S. flag

An official website of the United States government, Department of Justice.

Crime and Residential Choice: A Neighborhood Level Analysis of the Impact of Crime on Housing Prices

NCJ Number
Journal of Quantitative Criminology Volume: 22 Issue: 4 Dated: December 2006 Pages: 299-317
George E. Tita; Tricia L. Petras; Robert T. Greenbaum
Date Published
December 2006
19 pages
This study determined the prominence of neighborhood crime rates as a factor in neighborhoods' comparative differences in housing prices.
The study found that estimation of the average impact of the total crime index on housing values across Columbus, OH, was negligible; however, analysis of the differential impact of crime by crime type found that the negligible impact of the total crime rate on housing values was misleading. Violent crime did negatively impact home values. The total crime rate was dominated by property crimes, which are likely to be inconsistently underreported. In addition, the study found that total crime, property crime, and violent crime differentially impacted housing prices by neighborhood income level. Homicide rates and their changes negatively influenced housing values across all neighborhoods and across the subsamples of low-income and high-income neighborhoods. Unexpectedly, this impact was largest in the low-income neighborhoods. Possible explanations for this finding are offered. The study used housing, crime, and demographic data at the census-tract level for Columbus, OH. Three data sources were combined to create a panel of 43,577 housing transactions in 189 census tracts for the years 1995-1998. Crime data provided by police consisted of all Part I crimes, including homicide, rape, robbery, aggravated assault, burglary, larceny, and auto theft. Housing data were obtained from the First American Real Estate solutions database. Household demographic and economic characteristics were determined from the 1980 and 1990 Decennial Census summary Tape Files. The crime-housing price literature was extended by disaggregating crime to the census-tract level, using longitudinal data to examine changes in crime in addition to neighborhood crime levels, differentiating between the effects of property crime and violent crime, disaggregating the sample into census-tract income levels, and accounting for the measurement error that is inherent in reported crime statistics. 4 tables and 70 references