7

Incorporating Data on Crime and Violence into the Assessment of Fair Housing

MICHAEL C. LENS

Why Fair Housing and Exposure to Violence Matter

Fair housing remains vital to U.S. housing policy. Fifty years after the Fair Housing Act, disparate outcomes by race are persistent in the United States, and income and wealth inequality continue to grow. For example, while in 1983 the median white family had eight times the wealth as the median Black family, in 2017, that ratio was 10 to 1, and disparities in homeownership rates and home values are a central contributing factor to these dramatic disparities in wealth.¹

Further, small but steady declines in segregation by race (particularly Black-white segregation) have been largely offset by increases in segregation by income. Since 1970, high-income households have increasingly segregated themselves into richer and richer enclaves, while the poor had just one decade—the 1990s—in which they became more integrated with other income groups.² In all other decades since 1970, concentrated poverty has been on the rise.

To bolster federal efforts to further fair housing and hold local governments accountable for fair housing outcomes, the U.S. Department of Housing and Urban Development (HUD) published its landmark Affirmatively Furthering Fair Housing (AFFH) Rule in 2015. A key part of the AFFH process, and the topic of this chapter, is the Assessment of Fair Housing (AFH), which was expected to be conducted every five years by jurisdictions receiving HUD funds. Reflecting what we now know about how segregation and neighborhoods affect life chances, particularly for low-income and minority households, the AFH process got a lot right. Specifically, the AFH guided jurisdictions to collect a wide range of data that go well beyond simplistic measures of segregation by race and income. Importantly, the AFH focused explicitly on disparities in access to neighborhood opportunities, such as quality schools, employment, and transit, in addition to concentrations of racial and ethnic groups and households in poverty.

However, while weighing such a wide range of attributes is virtually impossible to do perfectly, I argue that two problems exist with the guidance provided under the 2015 rule. In turn, these key areas need improvement if there is to be a more robust revival of the rule’s aspirations. First, the AFH process largely treats all spatial concentrations equally, yet the literature on residential decision making and neighborhood effects suggests that these opportunities should not be weighed equally. In particular, neighborhood violence and school quality rise to the top in two strands of research that are vitally important to consider in evaluating the AFH process—the literature on residential location decisions (or preferences) and the literature on neighborhood effects. Second, and the explicit focus of this chapter, is the complete absence of data on crime and violence in the AFH process. Again, the research on residential location decisions and on neighborhood effects suggests that exposure to neighborhood violence frequently outweighs other neighborhood characteristics in terms of importance.

This chapter reviews the research on neighborhood preferences and neighborhood effects to identify the key attributes that low-income and minority households weigh most heavily when making residential location decisions and the neighborhood attributes that have the strongest effect on life outcomes. Although there is much uncertainty in each of these domains—our understanding of household preferences and specific neighborhood effects is imperfect and incomplete—each strand of literature points to neighborhood violence as being the neighborhood attribute that households care the most to avoid and the one that appears to have the greatest effect on life outcomes. Given this confluence of empirical research and household instinct, it is paramount that we find ways to incorporate data on crime and violence into the fair housing framework.

The Assessment of Fair Housing

In 2015, HUD published its new AFFH Rule. From a data and neighborhood opportunity standpoint, the key is that the Analysis of Impediments (AI), filed by all jurisdictions receiving HUD funding, was replaced with an AFH, to be conducted every five years.³ HUD also produced the AFFH data tool to guide local governments and public housing authorities (PHAs) in conducting their fair housing analyses (for more detail, see the Introduction in this volume). The data tool and the AFH encourage jurisdictions to collect and report data in several domains. First, given that fair housing laws prohibit discrimination by race or ethnicity, the AFFH tool and the AFH are heavily focused on measuring the spatial distribution of a region’s population by race and ethnicity—specifically, isolation and dissimilarity indices. Second, jurisdictions are to identify racially and/or ethnically concentrated areas of poverty (R/ECAPs), census tracts where at least 50 percent of the population identify as ethnic or racial minorities and the poverty rate is either 40 percent or more or greater than three times the poverty rate of the metropolitan area.⁴ Third, localities are to assess the spatial distribution of population sorted by other demographic attributes, including national origin and limited English proficiency, disability status, sex/gender, families with children, and households living in publicly supported housing. There are also data on housing problems and disproportionate housing need, such as the distribution of high-rent burdens. Finally, the AFH focuses on the concentration of neighborhood attributes through the creation of several indices: the School Proficiency Index, the Jobs Proximity Index, the Labor Market Engagement Index, the Low Transportation Cost Index, the Transit Trips Index, and the Environmental Health Index. Notably missing are data on crime, about which HUD states:

Residential Preferences and Fair Housing

The initial impetus for the Fair Housing Act was to hold local governments accountable for reducing segregation by race and income, which was often explicitly encouraged at the local level. As segregation by race and income has persisted over time and continues to result in non-white and poor households disproportionately occupying low-amenity neighborhoods, the focus has slowly shifted from explicit segregation policies to disparate neighborhood outcomes. An important first step in assessing disparities in neighborhood outcomes is to identify the neighborhood attributes that households prioritize when making residential location decisions.

Several studies examine the motivations and preferences of low-income households with respect to residential location, and they provide important insights into what matters for these residents when choosing to locate in a neighborhood. We have two general sources of data on residential preferences. The most straightforward is data on revealed preferences, those that we observe based on the attributes of people’s residential locations. These preferences are clearly constrained by household resources and housing market discrimination, at a minimum.⁶ Moreover, there are also limits to conclusions we can make about neighborhood preferences from studies that speak directly to households and capture their expressed preferences, chiefly because of tradeoffs between dwelling and neighborhood quality that they are forced to make and the limited menu of options that these families often consider to be realistically attainable.⁷

With these caveats in mind, it is useful to summarize what we know about the neighborhood preferences of low-income families. Notably, crime and violence are often cited as a concern for those who wish to move out of distressed neighborhoods. In surveys of Moving to Opportunity for Fair Housing (MTO) participants, crime was consistently offered as a primary motivation for wanting to enroll in those programs and move out of their original neighborhoods, which were typically quite unsafe.⁸ For participants in the other major housing voucher demonstration program—Gautreaux, in Chicago—we can surmise that the astonishingly high crime rates were among the reasons why participants were motivated to enter that program as well. In fact, drastic improvements in neighborhood safety can be highlighted as one of the clearest success stories in the MTO and Gautreaux programs. In the Gautreaux program, for example, participants starting out in the Robert Taylor Homes came from an environment that in 1980 comprised only 1 percent of Chicago’s population yet experienced 10 percent of the city’s murders, aggravated assaults, and rapes.⁹ Before moving to the suburbs, nearly half of Gautreaux participants “told of dangerous and frightening incidents that occurred regularly on the streets of their inner-city neighborhoods.”¹⁰ Criminal victimization rates were twice as high among Chicago public housing tenants overall as in the city as a whole. Micere Keels, Greg J. Duncan, Stefanie Deluca, Ruby Mendenhall, and James Rosenbaum estimate that violent crime rates in Gautreaux participants’ original neighborhoods were three times as high as those in Chicago.¹¹

Unfortunately, Gautreaux participants who moved to other parts of the city continued to face higher crime rates than in the city overall, and the same outcomes were found for suburban movers. However, those crime rates were still significantly lower than what they left behind. Suburban movers had a violent crime rate about 5 times as high as the crime rate in the Chicago suburbs at that time, and those who moved to the city faced violent crime rates about 1.5 times as high as the Chicago crime rate at that time. More promisingly, many years after their initial move, the Gautreaux households tracked by Keels et al. lived in neighborhoods with very comparable violent and property crime rates to Cook County (which includes Chicago and surrounding suburbs) as a whole. Mark Votruba and Jeffrey Kling estimate that in a sample of 2,850 Gautreaux participants moving to better-educated, safer neighborhoods, program participation saved up to seventeen lives, with thirteen of those averted deaths due to homicide.¹²

In the MTO program, an astonishing 77 percent of household heads cited “moving away from crime and drugs” as their primary or secondary reason for wanting to move.¹³ Similarly, 45 percent reported that a household member had been a crime victim in the previous six months, and 50 percent described their streets as very unsafe. Despite living in housing units and neighborhoods with multiple negative aspects (such as substandard housing conditions and neighborhoods with very high poverty rates and limited retail options), MTO participants overwhelmingly cited crime as the top motivation to move.

School quality is another neighborhood aspect that likely has clear ramifications for quality of life, and once again the Gautreaux and MTO programs provide insights on participants’ preferences for better schools. Positive educational outcomes for Gautreaux participant children have been consistently cited as one of the main success stories of the program and an incentive to expand the housing voucher program.¹⁴ James Rosenbaum, Marilyn Kulieke, and Leonard Rubinowitz examine the link between educational outcomes for youth Gautreaux participants and the quality of schools in suburban neighborhoods where participants moved. They find that educational standards were higher in suburban districts where the participants moved and that student-movers also received additional educational assistance.¹⁵ However, they experienced significant obstacles to adjusting to these new school environments, including increased racial discrimination in their new schools, which were predominantly white. Ultimately, the students were able to rise to the higher standards in their new schools—their grades were equivalent to those received in their previous schools. Further, children’s attitudes toward school were found to be higher in their new schools.

As with crime, MTO baseline surveys cite the search for school quality as a common motivation for entering the program. Lisa Sanbonmatsu et al. report that just under half (49.4 percent) of participants cited better schools as a primary or secondary reason for moving. Given that the study was limited to families with children, this is almost certainly an overestimate of the overall public housing population who seeks to move to neighborhoods with better schools, but it is telling that school quality is the second-highest priority (behind crime) among this population.¹⁶

Despite the strong evidence for preferences for neighborhoods with low crime and quality schools, even substantial interventions such as MTO have had limited success in placing families in neighborhoods with these attributes. Jens Ludwig et al. report clear gains for the experimental group in terms of reduced exposure to neighborhood crime. At follow-up four to seven years after moving, a gap of 30 percentage points was observed between the experimental and control groups in their response to whether they felt safe in their neighborhood at night. A 9 percent gap was found in the proportion of families who reported a member of their household being victimized in the previous six months. However, by the second follow-up ten to fifteen years after baseline, no statistically significant differences were found in recent criminal victimization history between the experimental and control groups. On the other hand, adults in the experimental and Section 8 groups were more likely to report feeling safe in their neighborhoods, were less likely to report that the police did not respond, and were less likely to report drug activity in their neighborhoods than the control group. While it is troubling that criminal victimization rates were equal across these groups in the long term, improved perceptions of neighborhood safety were persistent positive outcomes in the MTO study.¹⁷

It is telling that households so often point to crime as an impetus to participate in mobility programs designed to change neighborhood locations, but the fact remains that when we look at the revealed preferences of low-income households, they commonly occupy higher-crime neighborhoods than the average. Much of this circumstance has to do with the increased costs of living in lower-crime environments, but qualitative research suggests that these households face additional constraints.

In this vein, Peter Rosenblatt and Stefanie DeLuca provide a much-needed mixed-methods investigation into this phenomenon in an effort to explain the divergence between stated and revealed preferences of low-income renters. They use data on Baltimore MTO participants and extensive features on Baltimore neighborhoods to paint a fuller picture of the trade-offs, constraints, and priorities of these households that help determine their residential locations. In interviews, a desire to move to safer environments is consistently mentioned as a motivation for entering the program and a chief benefit of participant moves. Parents talk frequently of increased feelings of safety when their children left their homes to play. However, problems with the housing unit also surface as a reason why participants frequently made subsequent moves back to less-safe and/or higher-poverty neighborhoods. Often participating in the private-rental market for the first time, MTO movers sometimes find it difficult to identify landlords who would accept their voucher, and then when they do finally find one, sometimes they discover that those landlords are neglectful. Through this research, Rosenblatt and Deluca identify explanations in addition to limited financial resources that may explain why we observe households occupying high-crime areas, despite their stated preferences. Along with neglectful landlords and substandard housing units, they uncover a process of evaluating neighborhood crime conditions in potential destination neighborhoods that may subject low-income households to unsafe spaces. Specifically, they describe a process that they label “telescoping,” in which the homeseekers look at the immediate block of their unit—which may be relatively safe—while discounting or ignoring high-crime conditions on neighboring streets. In quantitative research undertaken at the census-tract level, we would observe such households as being highly exposed to neighborhood crime, although the households—which unfortunately may have experience navigating even more dangerous conditions—are less likely to view their blocks as immediate threats.¹⁸

What Is It about Neighborhoods That Matters?

Research on residential preferences tells us that low-income households often value lower-crime neighborhoods and better schools first and foremost and that they make trade-offs between these and several other neighborhood attributes, including access to employment and green space and proximity to environmental toxins. What does the neighborhood effects literature tell us about which neighborhood attributes matter the most in affecting key household outcomes, such as employment, education, and health? A better understanding of the key mechanisms in neighborhood effects will allow us to better prioritize and assess neighborhood attributes in the fair housing context.

In The Truly Disadvantaged, published in 1987, William Julius Wilson focuses on the segregation of Black households into jobless ghettoes as a result of racial discrimination in housing, manufacturing decline, white flight, and the flight of the Black middle class. Wilson identifies this concentration of joblessness as the main factor for deteriorating social conditions among urban Black families, such as high school dropouts, criminal involvement, and a rise in poor single-parent families.¹⁹ The neighborhood effects literature has largely consisted of efforts to test and extend Wilson’s hypotheses using a variety of data sources and methods. Specifically, this work often seeks to evaluate the mechanisms through which concentrated poverty and race affect a host of other household outcomes. Although segregation scholars often debate whether segregation by income or by race is more harmful, it is likely that these two forces of spatial stratification interact in important ways, particularly with respect to inherited neighborhood disadvantage and exposure to neighborhood violence.

Ingrid Gould Ellen and Margery Turner, writing in 1997, take stock of the neighborhood effects literature in the first ten years following publication of The Truly Disadvantaged, a point when Gautreaux had given way to MTO as the policy lever designed to test the effect of neighborhood on household outcomes. Ellen and Turner conclude that empirical research generally confirms that neighborhood environment has an influence on important outcomes for children and adults, but they find that efforts to identify which characteristics matter most and to quantify their importance are inconclusive at that point. Further, they note that neighborhood effects are much less important than family characteristics, although there is typically a very high correlation between neighborhood and family characteristics.²⁰

The results from MTO speak to how neighborhood poverty affects a variety of household outcomes, but it is difficult to connect MTO outcomes to specific neighborhood attributes other than poverty. In all, the impact of moving MTO households out of high-poverty, dangerous neighborhoods was less profound than many expected, particularly for adults. Adults in the experimental group were no more likely to be employed at the first or second follow-up than those in the control and comparison groups, and being in the experimental group had no positive effects on children’s schooling or employment outcomes. Children were also no less likely to engage in risky or criminal behaviors. The experimental group did experience statistically significant declines in adult obesity relative to the comparison groups, as was the case with mental health problems for female adolescent participants. Raj Chetty, Nathaniel Hendren, and Lawrence Katz look at long-term outcomes by using tax returns to estimate earnings and identify current neighborhood locations and whether younger MTO recipients subsequently attended college between the ages of eighteen and twenty.

They continue to find no effects on adult outcomes, even when testing for a dosage effect (the length of time spent in low-poverty neighborhoods). However, they find substantial, positive effects on long-term outcomes for children who moved when they were young—specifically, an increased likelihood of college attendance and higher earnings.²¹ We can conclude from the extensive research on MTO that neighborhood poverty rates matter, particularly for children who are exposed to these neighborhoods the longest. But many adult and youth outcomes do not respond to changes in neighborhood poverty. Further, several neighborhood attributes are highly correlated with poverty rates in these areas, including racial segregation, crime and violence, and low-quality schooling, meaning it is unclear whether escaping concentrated poverty is truly the driver of these outcomes.

A few studies have used MTO data to examine the effect of neighborhood attributes other than poverty rates to determine their effects. Ludwig and Kling isolate the effects of neighborhood crime rates to determine whether the presence of crime leads youth to engage in more crime themselves.²² They find little evidence for such a contagion effect and instead find that racial segregation has a statistically significant impact on the propensity to commit crime. Evelyn Blumenberg and Gregory Pierce test whether access to public transit plays a role in employment outcomes for MTO participants and find that improved access to public transit is associated with the increased likelihood of keeping a job but not of finding and securing one.²³ Michael C. Lens and C. J. Gabbe test the spatial mismatch hypothesis by using MTO data and find that MTO did not improve spatial proximity to jobs for program participants, but that it would not likely have mattered anyway; they find no connection between job proximity and increased likelihood of employment or increased earnings at follow-up.²⁴

The findings by Lens and Gabbe reaffirm the often-inconclusive findings regarding the role of employment accessibility in determining employment outcomes. This phenomenon is persistently studied and yet has a decidedly mixed record. Michael Stoll finds that Blacks and Latinxs live in areas of Los Angeles with poor job growth, which results in their spending more time and effort to find work.²⁵ Also in Los Angeles, Paul Ong and Evelyn Blumenberg find that the job-poor neighborhoods lived in by welfare recipients make it less likely that they will find work.²⁶ By contrast, Robert Cervero, Onésimo Sandoval, and John Landis find no relation between regional job accessibility and employment outcomes for welfare recipients in Alameda County, California—a finding echoed by Thomas Sanchez, Qing Shen, and Zhong-Ren Peng, who look at Temporary Assistance for Needy Families recipients in six U.S. cities.²⁷ Given this information, it is unclear whether employment accessibility should be a central feature in the fair housing discussion.

Some of the more groundbreaking work on the role of segregation by income and race examines the confluence of these two concentrations rather than isolating them, as has been past practice.

Patrick Sharkey focuses explicitly on the confluence of concentrated poverty and racial segregation to examine the role of what he terms “inherited neighborhood disadvantage.” Using the Panel Study of Income Dynamics (PSID), Sharkey finds that many of the wealth and income disparities observed between whites and African Americans can be explained by the incredibly stark differences in neighborhoods that these different racial groups occupy.²⁸ Among the PSID cohort born between 1955 and 1970, only 4 percent of white households lived in relatively high-poverty neighborhoods, where the poverty rate was 20 percent or higher. For African Americans born at the same time, that number was fifteen times higher, or 62 percent. These differences barely changed in thirty years; among the 1985 to 2000 cohort, those numbers were 6 percent and 68 percent, respectively.

In other words, higher-poverty neighborhoods that are commonplace for African Americans are almost unheard of for white Americans. Importantly, these disparities hold when controlling for income differences between whites and Blacks.

Sharkey links living in a high-poverty neighborhood to two key outcomes—inherited neighborhood disadvantage and economic mobility. He finds that neighborhoods are largely inherited across generations: the correlation between the income level of parent and child neighborhoods is quite high (about 0.67). However, he also finds that when white families live in high-poverty neighborhoods, it tends to be for a single generation, and whites tend to live in affluent neighborhoods for multiple generations. The opposite pathways are typical for African-American families—multigenerational exposure to neighborhood poverty is common, and multigenerational exposure to affluent neighborhoods is rare.

The exposure to neighborhood disadvantage, Sharkey argues, contributes to the remarkably persistent gaps in income and wealth between Black and white families. Sharkey finds that the neighborhood poverty rate of a child explains a great deal of the income he or she earns as an adult and also explains much of the economic mobility gap. The latter fact reflects the reality that Black children are more likely than white ones to experience downward mobility (moving from a high-income category to a lower one) and that Black children are less likely to experience upward mobility (moving from a low family-income category to a higher one).

Sharkey’s conclusions have been reaffirmed by Raj Chetty, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez, who examine the geography of intergenerational mobility, looking at the regional scale rather than the neighborhood. They find that movement up and down the economic ladder across successive generations varies dramatically by metropolitan area. They describe the United States as “a collection of societies”—in some metro areas, economic mobility across generations is common, whereas elsewhere, movement out of poverty is a rare event.²⁹ Importantly, they find that the spatial concentration of particular demographic characteristics, such as college attendance and teenage birth rates, is strongly linked to rates of economic mobility.

Further, the persistence of racial and economic segregation is continuing to lead to substantial inequities in terms of public services that people consume and exposure to crime and violence. Higher-quality services and other amenities are concentrated in particular locations within metropolitan areas, and these concentrations map onto patterns of economic and racial segregation. Jorge De la Roca, Ingrid Ellen, and Katherine O’Regan use Census data, a unique tract-level dataset on crime in ninety-one U.S. cities, and geocoded school-zone data at the Census-block level to estimate the effects of racial segregation on the exposure of different racial groups to low-socioeconomic-status neighbors, crime, and low-quality schools.³⁰ They find substantial racial disparities in exposure to disadvantaged neighborhoods. Specifically, whites and Asians are much less likely to live in low-status neighborhoods or neighborhoods with high crime or low-quality schools than Blacks and Hispanics. Further, these disparities are not fully explained by differences in income; they find “that the average poor white person lived in a neighborhood with a lower violent crime rate than the average non-poor black person.”³¹ They also find that metropolitan-area segregation levels (i.e., dissimilarity and isolation indices between various racial groups and whites) are strong predictors of these racial gaps in exposure to all three domains of neighborhood disadvantage—neighborhood socioeconomic characteristics, quality of the zoned school, and violent crime.

Robert Sampson’s Project on Human Development in Chicago Neighborhoods (PHDCN), which, like MTO, began in the early 1990s, has collected perhaps the most extensive set of neighborhood characteristics in the pursuit of identifying their effects on household outcomes. Sampson’s book Great American City: Chicago and the Enduring Neighborhood Effect summarizes this work.³²

PHDCN researchers combine field observations with data on housing, crime and violence, residential mobility between neighborhoods, contacts between public officials and leaders in different neighborhoods (to measure communication between neighborhoods), administrative records, and a letter-drop survey to measure social altruism and civic cooperation—key components of what Sampson and his colleagues term “collective efficacy.”³³ This method is best suited to a project like the PHDCN that is able to collect data on a broad scope of variables, but more data are available at small levels of geography than ever before. Furthermore, some countries, such as Sweden, have better individual-level data available to tie neighborhood opportunity measures to individual outcomes.³⁴

Using these extensive data on neighborhood domains and over multiple time periods, Sampson finds substantial overlap between various measures of disadvantage and also finds that neighborhood disadvantage is very persistent over time. Neighborhoods with high rates of violence also have low health indicators and poor collective efficacy. Further, disadvantaged neighborhoods tend to remain disadvantaged for decades—and neighborhood poverty is particularly persistent in neighborhoods with high proportions of African Americans—stressing again the interaction between race and poverty concentration. Sampson’s data allow him to tie together the role of community social capital in protecting neighborhoods from becoming violent and disadvantaged.

Finally, research from Sharkey suggests that neighborhood violence is particularly influential on children’s outcomes. Using data from the PHDCN, Sharkey finds strong evidence that local homicides affect children’s performance on verbal and reading assessments taken shortly after the homicides occurred.³⁵ He exploits the exogenous variation in the timing of the homicides to strengthen the causal linkages between violence and assessments. In another paper, Patrick T. Sharkey, Nicole Tirado-Stayer, Andrew V. Papachristos, and C. Cybele Raver find more evidence that geographically proximate homicides have a negative impact on several youth outcomes, including pre-academic cognitive skills, such as impulse control, and vocabulary and math assessment scores.³⁶ Further, they find that parents’ mental health conditions are negatively affected by local homicides. In Sharkey’s recent book Uneasy Peace, he succinctly sums up the effects of violence: “Local violence does not make children less intelligent. Rather, it occupies their minds.”³⁷

Sharkey provides further evidence that the slow declines in the Black-white education achievement gap may be attributable to declining violence in urban America by linking data from the National Assessment of Educational Progress (NAEP) to local crime rates. Once more, in Uneasy Peace, Sharkey reports work with colleague Gerard Torrats-Espinosa that examines the link between crime and upward mobility, a key outcome that fair housing policy should try to influence. They find that in areas where violence declined faster, upward mobility became much more likely.³⁸

Incorporating Crime and Violence Data into the Assessment of Fair Housing

Strong evidence shows that neighborhood matters, and persistent disparities in neighborhood quality are key justifications for why fair housing laws continue to be necessary. Segregation research, which has consistently summarized myriad justifications for fair housing law, has typically focused on the attributes of those who live around you rather than on the spatial sorting of structural characteristics that shape opportunities for individuals and families. Although the precise mechanisms through which neighborhoods affect people’s lives are often unclear, I argue that we should emphasize the structural characteristics of neighborhoods that reflect spatial inequalities in the location of amenities and disadvantages, particularly neighborhood violence.

When MTO and Gautreaux participants discussed their motivations for leaving behind public housing for new housing locations, they did not talk much about there being too many Black or poor neighbors around them. Time and again, participants talked about neighborhood safety as a primary reason for neighborhood dissatisfaction. They also discussed school and housing-unit quality and a lack of jobs or retail options. In other words, people in high-poverty, racially segregated neighborhoods do not tend to talk about the people who live around them.

They talk about the structural, rather than demographic, characteristics of their neighborhoods. Research by Sharkey and others suggests that these motivations are particularly well justified—living in violent neighborhoods not only is dangerous and exposes you to risk of bodily harm or worse but affects your ability to function and excel in life.

The AFH process generally reflects what we know about neighborhood effects. Although considerable attention is paid to neighborhood demographic analyses, quite a few domains would also be considered structural: school proficiency, job indices, transportation costs, and environmental health indicators are all included. What is clearly necessary and missing is an emphasis on collecting and assessing the spatial distribution of crime and violence.

Importantly, we now often have the data to do this. Neighborhood crime data have been collected before on a large scale, suggesting that collecting data in a large number of cities is feasible. Two decades ago (1999–2001), Ruth Peterson and Lauren Krivo (2010) conducted the National Neighborhood Crime Study (NNCS), a nationally representative sample of crime data for 9,593 Census tracts in ninety-one U.S. cities.³⁹ The resulting public dataset includes an average of the major crime categories developed by the Federal Bureau of Investigation’s Uniform Crime Report System over the entire three years for each Census tract.

More recently, tract-level crime data have been used with increasing frequency, covering a variety of years and cities. Lens, Ellen, and O’Regan, writing in 2011, collect neighborhood-level crime data for ten U.S. cities for a purpose closely related to assessing fair housing—measuring the neighborhood crime rates faced by housing subsidy recipients.⁴⁰ Brent Mast and Ronald Wilson investigate the relationship between vouchers and crime in Charlotte, North Carolina, by using data on property, violent, residential burglary, and street crimes from 2000 to 2009.⁴¹ Elizabeth Griffiths and George Tita use tract-level data on homicides in Los Angeles to explore whether public housing is a “hotbed” for crime.⁴² John MacDonald, John Hipp, and Charlotte Gill also use tract-level crime data in Los Angeles to investigate the effects of immigrant concentration on crime. Los Angeles, like many U.S. cities, has crime data available online for anyone to access.⁴³

John Hipp and Daniel Yates use tract-level data to study how returning parolees affect crime in Sacramento.⁴⁴ This breadth of research suggests that technology and a greater appreciation for data sharing among public agencies—including police—are helping foster an era in which crime data are increasingly available at small levels of geography, including Census tracts.

Some may suggest that neighborhood crime is so highly correlated with many of the other factors that are captured in the AFH process that it is unnecessary to exert extra effort to collect these data. However, many have shown, including Sampson, Raudenbush, and Earls, that demographic features of a neighborhood do not determine violent crime rates.⁴⁵ Hipp finds that the association between poverty and murder rates becomes insignificant when neighborhood inequality is accounted for, suggesting that the relationship between poverty and violent crime is much more complicated than people think.⁴⁶ It is likely that concentrated disadvantage coupled with racial segregation explain much of the variation in neighborhood crime, but these urban features do not map perfectly onto one another.⁴⁷ Specifically, Robert Sampson, Jeffrey Morenoff, and Stephen Raudenbush find that although the proportion of a neighborhood’s population who identifies as Black is highly correlated with violent crime in neighborhoods, particular neighborhood characteristics, such as immigrant concentration, percentage of professional/managerial occupations, concentrated disadvantage, and residential stability, wash out the link between race and crime.⁴⁸

Fair housing protections target low-income and minority households. These populations are deeply concerned with crime and violence in their communities. The research on neighborhood effects consistently points to neighborhood violence as a key mechanism through which neighborhood affects life chances. Given this significance of exposure to violence for individual outcomes and the fact that the neighborhood characteristics identified in the 2015 version of AFH process are inadequate proxies for neighborhood violence, we need to find ways to make crime data accessible to more jurisdictions undertaking these assessments. As noted, crime data on several cities and years have already been collected and utilized for research purposes, suggesting that collecting neighborhood crime data on a wide scale is feasible. Further, given that municipalities conduct the AFH, they are particularly well positioned to obtain data from local police agencies.

An important limitation may always be the lack of neighborhood crime data in most suburban areas. Thousands of suburban jurisdictions and police departments exist across the country, and it is simply not feasible to collect crime data from all those areas. To conduct assessments of entire metropolitan areas will inevitably be difficult. However, in some cases, police agencies cross city boundaries. For example, there are eighty-eight cities in Los Angeles County, most of which are suburban. Many of these jurisdictions are policed by one agency—the L.A. County Sheriff’s Department. Scholars doing work on the AFH process can help assess the feasibility of collecting neighborhood crime data across metropolitan areas and begin the task of overseeing data-collection efforts.

Further, the lack of availability of crime data is reflective of a lack of transparency and accountability in U.S. municipal police departments, a problem laid bare by countless instances of police brutality, which have received heightened attention in recent years. Housing and segregation scholars should join the chorus of voices seeking increased police transparency, specifically through the dissemination of accessible data. Housing segregation means that not only violence is concentrated in communities of color but also oppressive policing practices ranging from everyday harassment to murder.

Looking forward to a future, more progressive, HUD effort to revise the AFFH Rule and the accompanying AFH, it will be important to pay more attention to crime data. Extensive research has made clear that one of the neighborhood characteristics that has the most significant effect on the educational performance and socioeconomic mobility of individuals is their childhood exposure to neighborhood violence. The ability of the AFFH Rule to reduce socioeconomic disparities and enhance the well-being of young people could be improved, therefore, by adding data regarding disparities in exposure to crime to the existing measures that HUD is already providing, where such data are publicly available. Where such data on crime are not already publicly available, HUD could encourage municipalities to make them public or, at a minimum, to analyze the spatial patterning of the crime data that municipalities should be able to obtain from their own police departments. The lack of attention to crime and violence in the current AFFH Rule is arguably its most significant shortcoming. Remedying that absence by including analyses of disparities in exposure to violence should be a priority for HUD, for block grant recipients, and for civil rights advocates.

ENDNOTES

1. Signe-Mary McKernan, Caroline Ratcliffe, and C. Eugene Steuerle, “Nine Charts about Wealth Inequality in America,” October 5, 2017, https://apps.urban.org/features/wealth-inequality-charts/.

2. Kendra Bischoff and Sean F. Reardon, “Residential Segregation by Income, 1970–2009,” US2010 (Russell Sage Foundation, October 16, 2013), available at https://s4.ad.brown.edu/Projects/Diversity/Data/Report/report10162013.pdf; Paul A. Jargowsky, Stunning Progress, Hidden Problems: The Dramatic Decline of Concentrated Poverty in the 1990s (Washington, DC: Brookings Institution, May 1, 2003), available at https://www.brookings.edu/research/stunning-progress-hidden-problems-the-dramatic-decline-of-concentrated-poverty-in-the-1990s/.

3. Robert Mark Silverman, Li Yin, and Kelly L. Patterson, “Siting Affordable Housing in Opportunity Neighborhoods: An Assessment of HUD’s Affirmatively Furthering Fair Housing Mapping Tool,” Journal of Community Practice 25, no. 2 (April 2017): 143–158.

4. U.S. Department of Housing and Urban Development (HUD), “AFFH-T Data Documentation-HUD Exchange,” September 2017, available at https://www.hudexchange.info/resource/4848/affh-data-documentation/.

5. Ibid.

6. S. J. South, K. Crowder, and E. Chavez, “Exiting and Entering High-Poverty Neighborhoods: Latinos, Blacks and Anglos Compared,” Social Forces 84, no. 2 (December 2005): 873–900.

7. Peter Rosenblatt and Stefanie DeLuca, “‘We Don’t Live Outside, We Live in Here’: Neighborhood and Residential Mobility Decisions among Low-Income Families,” City and Community 11, no. 3 (September 2012): 254–284.

8. Maria Hanratty, Sarah McLanahan, and Becky Pettit, “The Impact of the Los Angeles Moving to Opportunity Program on Residential Mobility, Neighborhood Characteristics, and Early Child and Parent Outcomes,” Working Paper (Bendheim-Thoman Center for Research on Child Wellbeing, Princeton University, 1998); Lisa Sanbonmatsu et al., Moving to Opportunity for Fair Housing Demonstration Program—Final Impacts Evaluation (Washington, DC: HUD, 2011), available at http://www.huduser.org/portal/publications/pubasst/MTOFHD.html.

9. Leonard S. Rubinowitz and James E. Rosenbaum, Crossing the Class and Color Lines: From Public Housing to White Suburbia (Chicago: University of Chicago Press, 2000).

10. Ibid., 83–84.

11. Micere Keels, Greg J. Duncan, Stefanie Deluca, Ruby Mendenhall, James Rosenbaum. “Fifteen Years Later: Can Residential Mobility Programs Provide a Long-Term Escape from Neighborhood Segregation, Crime, and Poverty?” Demography 42, no. 1 (2005): 51–73.

12. Mark Edward Votruba and Jeffrey R. Kling, “Effects of Neighborhood Characteristics on the Mortality of Black Male Youth: Evidence from Gautreaux, Chicago,” Social Science and Medicine 68, no. 5 (March 2009): 814–823.

13. Sanbonmatsu et al., Moving to Opportunity for Fair Housing Demonstration Program.

14. Ibid.; James E. Rosenbaum, Marilyn J. Kulieke, and Leonard S. Rubinowitz, “White Suburban Schools’ Responses to Low-Income Black Children: Sources of Successes and Problems,” Urban Review 20, no. 1 (March 1988): 28–41.

15. Rosenbaum, Kulieke, and Rubinowitz, “White Suburban Schools’ Responses to Low-Income Black Children.”

16. Sanbonmatsu et al., Moving to Opportunity for Fair Housing Demonstration Program.

17. Jens Ludwig et al., “What Can We Learn about Neighborhood Effects from the Moving to Opportunity Experiment?” American Journal of Sociology 114, no. 1 (July 2008): 144–188; Sanbonmatsu et al., Moving to Opportunity for Fair Housing Demonstration Program.

18. Rosenblatt and DeLuca, “‘We Don’t Live Outside, We Live in Here.’”

19. William Julius Wilson, The Truly Disadvantaged (Chicago: University of Chicago Press, 1987).

20. Ingrid Gould Ellen and Margery Austin Turner, “Does Neighborhood Matter? Assessing Recent Evidence,” Housing Policy Debate 8, no. 4 (January 1997): 833–866.

21. Raj Chetty, Nathaniel Hendren, and Lawrence Katz, The Effects of Exposure to Better Neighborhoods on Children: New Evidence from the Moving to Opportunity Experiment (Cambridge, MA: National Bureau of Economic Research, May 2015).

22. Jens Ludwig and Jeffrey R. Kling, “Is Crime Contagious?” Journal of Law and Economics 50, no. 3 (August 2007): 491–518.

23. Evelyn Blumenberg and Gregory Pierce, “A Driving Factor in Mobility? Transportation’s Role in Connecting Subsidized Housing and Employment Outcomes in the Moving to Opportunity (MTO) Program,” Journal of the American Planning Association 80, no. 1 (January 2014): 52–66.

24. Michael C. Lens and C. J. Gabbe, “Employment Proximity and Outcomes for Moving to Opportunity Families,” Journal of Urban Affairs 39, no. 4 (May 2017): 547–562.

25. Michael A. Stoll, “Spatial Job Search, Spatial Mismatch, and the Employment and Wages of Racial and Ethnic Groups in Los Angeles,” Journal of Urban Economics 46, no. 1 (July 1999): 129–155.

26. Paul Ong and Evelyn Blumenberg, “Job Access, Commute and Travel Burden among Welfare Recipients,” Urban Studies 35, no. 1 (January 1998): 77–93.

27. Robert Cervero, Onésimo Sandoval, and John Landis, “Transportation as a Stimulus of Welfare-to-Work: Private versus Public Mobility,” Journal of Planning Education and Research 22, no. 1 (September 2002): 50–63; Thomas W. Sanchez, Qing Shen, and Zhong-Ren Peng, “Transit Mobility, Jobs Access and Low-Income Labour Participation in US Metropolitan Areas,” Urban Studies 41, no. 7 (June 2004): 1313–1331.

28. Patrick Sharkey, Stuck in Place: Urban Neighborhoods and the End of Progress toward Racial Equality, 1st ed. (Chicago: University of Chicago Press, 2013).

29. Raj Chetty, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez, Where Is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States (Cambridge, MA: National Bureau of Economic Research, January 2014).

30. Jorge De la Roca, Ingrid Gould Ellen, and Katherine M. O’Regan, “Race and Neighborhoods in the 21st Century: What Does Segregation Mean Today?” Regional Science and Urban Economics 47 (July 2014): 138–151; Ruth D. Peterson and Lauren J. Krivo, “National Neighborhood Crime Study (NNCS), 2000: Version 1,” May 5, 2010, available at http://www.icpsr.umich.edu/icpsrweb/RCMD/studies/27501.

31. De la Roca, Ellen, and O’Regan, “Race and Neighborhoods in the 21st Century,” 143.

32. Robert J. Sampson, Great American City: Chicago and the Enduring Neighborhood Effect (Chicago: University of Chicago Press, 2012).

33. Robert J. Sampson, Stephen W. Raudenbush, and Felton Earls, “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy,” Science 277, no. 5328 (August 15, 1997): 918–924.

34. George Galster, Roger Andersson, Sako Musterd, and Timo M. Kauppinen, “Does Neighborhood Income Mix Affect Earnings of Adults? New Evidence from Sweden,” Journal of Urban Economics 63, no. 3 (May 2008): 858–870.

35. Patrick Sharkey, “The Acute Effect of Local Homicides on Children’s Cognitive Performance,” Proceedings of the National Academy of Sciences 107, no. 26 (June 29, 2010): 11733–11738.

36. Patrick T. Sharkey, Nicole Tirado-Stayer, Andrew V. Papachristos, and C. Cybele Raver, “The Effect of Local Violence on Children’s Attention and Impulse Control,” American Journal of Public Health 102, no. 12 (December 2012): 2287–2293.

37. Patrick Sharkey, Uneasy Peace: The Great Crime Decline, the Renewal of City Life, and the Next War on Violence, 1st ed. (New York: Norton, 2018), 87.

38. Ibid.

39. Peterson and Krivo, “National Neighborhood Crime Study (NNCS), 2000.”

40. Michael C. Lens, Ingrid Gould Ellen, and Katherine O’Regan, “Do Vouchers Help Low-Income Households Live in Safer Neighborhoods? Evidence on the Housing Choice Voucher Program,” Cityscape 13, no. 3 (2011): 135–159.

41. Brent D. Mast and Ronald E. Wilson, “Housing Choice Vouchers and Crime in Charlotte, NC,” Housing Policy Debate 23, no. 3 (July 2013): 559–596.

42. Elizabeth Griffiths and George Tita, “Homicide in and around Public Housing: Is Public Housing a Hotbed, a Magnet, or a Generator of Violence for the Surrounding Community?” Social Problems 56, no. 3 (August 2009): 474–493.

43. Office of Mayor Eric Garcetti, “Los Angeles Open Data: Information, Insights, and Analysis from the City of Los Angeles,” A Safe City: Crime Data from 2010 to Present, available at https://data.lacity.org/A-Safe-City/Crime-Data-from-2010-to-Present/63jg-8b9z; John M. MacDonald, John R. Hipp, and Charlotte Gill, “The Effects of Immigrant Concentration on Changes in Neighborhood Crime Rates,” Journal of Quantitative Criminology 29, no. 2 (June 2013): 191–215.

44. John R. Hipp and Daniel K. Yates, “Do Returning Parolees Affect Neighborhood Crime? A Case Study of Sacramento,” Criminology 47, no. 3 (August 2009): 619–656.

45. Sampson, Raudenbush, and Earls, “Neighborhoods and Violent Crime.”

46. John R. Hipp, “Income Inequality, Race, and Place: Does the Distribution of Race and Class Within Neighborhoods Affect Crime Rates?” Criminology 45, no. 3 (August 2007): 665–697.

47. Lauren J. Krivo, Ruth D. Peterson, and Danielle C. Kuhl, “Segregation, Racial Structure, and Neighborhood Violent Crime,” American Journal of Sociology 114, no. 6 (May 2009): 1765–1802.

48. Robert J. Sampson, Jeffrey D. Morenoff, and Stephen Raudenbush, “Social Anatomy of Racial and Ethnic Disparities in Violence,” American Journal of Public Health 95, no. 2 (February 2005): 224–232.