Draft: Not for circulation or distribution Do Housing Choice Voucher Recipients Import Crime? by:

Recent public attention has focused on whether and the extent to which Housing Choice Voucher Recipients (HCVRs) influence crime. Journalistic accounts of HCVR mobility imply that they do cause crime thus shedding negative light on the largest federal housing assistance program. This paper adds to the growing empirical literature in this area by examining in places among the 100 largest metropolitan areas in the country whether changes in HCVRs presence over time are followed by changes in crime rates. To do so, data is used on voucher holders from the U.S. Department of Housing and Urban Development’s (HUD) Picture of Subsidized Housing, on crime from the Uniform Crime Report (UCR) Offenses Known and Cleared by Arrest compiled by the FBI, and on important demographic data from the Census. The results of the analysis do not support the hypothesis that HCVR cause crime. Where positive relationships between HCVRs and crime are found, more careful analysis using leads and lags of important variables demonstrate reverse causation at play; that is, that HCVRs tend to move to areas where crime is increasing. Draft: Not for circulation or distribution Introduction The question of whether Housing Choice Voucher Recipients (HCVRs) cause crime has gained increased attention in recent years. This is partly because relatively recent changes in the HCV program (HCVP) that relaxed where HCVs can be used led to expansion in the types of neighborhoods in which recipients search and rent housing. The HCVP is the largest federal rental assistance program, serving a little over 2 million low-income households. Prior to the mid1990s, it was extremely difficult for HCVRs (or at the time Section 8 recipients) to use the certificate/voucher anywhere other than in the jurisdiction of the issuing housing authority, resulting in the concentration of recipients living in mostly racially segregated central city neighborhoods that were also economically disadvantaged (Wallace et al. 1981). However, during the 1990s, the Department of Housing and Urban Development (HUD) implemented a number of policy changes that affected the locational choices of HCV recipients, thereby influencing recipients moves across the broader metropolitan area including to the suburbs (Covington, et. al., 2011). As HCVR move to new neighborhoods, a number of claims through mostly journalistic accounts have been leveled that they import crime in doing so. The most important of these is arguably Rosin (2008) whose Atlantic Magazine article “American Murder Mystery” implicates the HCV program (as well as all other HUD subsidized housing programs) as a source of crime increases in Memphis, Tennessee as a result of low-income households using vouchers to move 1 The HCVP is administered nationwide by approximately 2,500 local public housing agencies (PHAs). Voucher recipients may spend up to 40 percent of adjusted gross income at the initial lease, and the remainder is paid by the voucher usually up to fair market rent levels in the metropolitan region. 2 These policy changes included the implementation of the Mobility Counseling Program, through which counselors encouraged movement from high-poverty to low-poverty neighborhoods, the introduction of portability that led to the streamlining of the process through which HCV could use their voucher in a jurisdiction other than the one that issued the voucher, and the shift from certificates to vouchers that arguably provided assisted households considerably more choice about where to live. The former placed a ceiling on the amount of rent a household can pay (which could act to limit locational choices), while the latter vouchers do not, as long as the household is willing to pay the difference, and pays no more than 40 percent of its income toward rent (Lubbell 2001).