Foreclosures and crime: A city-level analysis in Southern California of a dynamic process

Although a growing body of research has examined and found a positive relationship between neighborhood crime and home foreclosures, some research suggests this relationship may not hold in all cities. This study uses city-level data to assess the relationship between foreclosures and crime by estimating longitudinal models with lags for monthly foreclosure and crime data in 128 cities from 1996 to 2011 in Southern California. We test whether these effects are stronger in cities with a combination of high economic inequality and high economic segregation; and whether they are stronger in cities with high racial/ethnic heterogeneity and high racial segregation. One month, and cumulative three month, six month, and 12-month lags of foreclosures are found to increase city level crime for all crimes except motor vehicle theft. The effect of foreclosures on these crime types is stronger in cities with simultaneously high levels of inequality but low levels of economic segregation. The effect of foreclosures on aggravated assault, robbery, and burglary is stronger in cities with simultaneously high levels of racial heterogeneity and low levels of racial segregation. On the other hand, foreclosures had a stronger effect on larceny and motor vehicle theft when they occurred in a city with simultaneously high levels of racial heterogeneity and high levels of racial segregation. There is evidence that the foreclosure crisis had large scale impacts on cities, leading to higher crime rates in cities hit harder by foreclosures. Nonetheless, the economic and racial characteristics of the city altered this effect.     

The contemporary foreclosure crisis and US crime rates

Foreclosure rates in America reached unprecedented levels during the last half of the 2000s, and many observers have speculated that elevated crime rates were one of the probable negative collateral consequences of this trend. We examine this issue with a comprehensive county-level analysis of the role of foreclosure in shaping contemporary crime patterns, highlighting the possibility of theoretically informed non-linear and conditional relationships. Multivariate regression models that account for the well-documented spatial autocorrelation of crime rates and the possible endogeneity of foreclosure reveal a positive association between rates of foreclosure and property crime that accelerates significantly once foreclosure rates attain historically high levels. Multiplicative models indicate that this pattern holds for burglary across diverse county conditions, but the observed non-linear effect of foreclosure on robbery rates is limited primarily to areas that also exhibit relatively high levels of resource deprivation and limited new housing construction.     

Parks as crime inhibitors or generators: Examining parks and the role of their nearby context

Although neighborhood studies often focus on the presence of some particular entity and its consequences for a variety of local processes, a frequent limitation is the failure to account more broadly for the local context. This paper therefore examines the role of parks for community crime, but contributes to the literature by testing whether the context of land uses and demographics nearby parks moderate the parks and crime relationship. A key feature of our approach is that we also test how these characteristics explain crime in the park, nearby the park, and in other neighborhoods in the city with data from nine cities across the United States (N = 109,808 blocks). We use multilevel Poisson and negative binomial regressions to test our ideas for six types of street crime. Our findings show that nearby land uses and socio-demographic characteristics are a key driver of crime being located within the park or nearby the park. Our results also show a clear distance decay pattern for the impact of various land uses and socio-demographics nearby parks. The results emphasize a need for research to consider the broader socio-spatial context in which crime generators/inhibitors are embedded.     

American crime drops: Investigating the breaks,dips and drops in temporal homicide

ObjectivesWhile a great deal of attention has been given to the 1990s crime drop, less is known about the more recent decline in homicide rates that occurred in several large U.S. cities. This paper aims to explore whether these represent two distinct drops via statistical evidence of structural breaks in longitudinal homicide trends and explore potentially differing explanations for the two declines. Methods: Using homicide data on a large sample of U.S. cities from 1990 to 2011, we test for structural breaks in temporal homicide rates. Combining census data and a time series approach, we also examine the role structural features, demographic shifts, and crime control strategies played in the changes in homicide rates over time. Results: Statistical evidence demonstrates two structural breaks in homicide trends, with one trend reflecting the 1990s crime drop (1994–2002) and another trend capturing a second decline (2007–2011). Time series analysis confirms previous research findings about the contributions of structural conditions (e.g., disadvantage) and crime control strategies (e.g., police force size) to the crime drop of the 1990s, but these factors cannot account for the more recent drop with the exception of police presence. Conclusions: Although both structural conditions and crime control strategies are critical to the longitudinal trends in homicide rates over the entire span from 1990 to 2011, different factors account for these two distinct temporal trends.     

Disentangling the reciprocal relationship between change in crime and racial/ethnic change

Prior research has found that racial/ethnic change and residential instability are positively related to neighborhood crime. However, the process of racial/ethnic change differentially influences crime above and beyond residential instability. While both processes affect crime through the disruption of existing social ties, racial/ethnic change has additional consequences for crime by heightening racial/ethnic tensions and undercutting cross-group interactions. This means racial/ethnic change is a different process than residential instability, and suggests that neighborhoods experiencing high rates of instability and high rates of racial/ethnic change may be particularly susceptible to crime. Therefore, we examine the influence of racial/ethnic compositional change on change in crime across different levels of residential instability. Further, we argue that demographic change and crime may be influencing each other simultaneously: increases in the crime rate and racial/ethnic compositional change impact each other at the same time. To capture this process, we employ a structural equation model (SEM) that accounts for the reciprocal and simultaneous relationship between racial/ethnic change and violent and property crime rates in Los Angeles, California between 1990 and 2000. We also account for the influence of change in spatially proximate communities. Results show robust evidence that increases in racial/ethnic change contributes to greater violent and property crime rates, but the reciprocal influence of crime on racial/ethnic change is contingent upon the degree to which a neighborhood is experiencing residential instability and crime type.     

Growing pains or appreciable gains? Latent classes of neighborhood change,and consequences for crime in Southern California neighborhoods

This study explored the dynamic nature of neighborhoods using a relatively novel approach and data source. By using a nonparametric holistic approach of neighborhood change based on latent class analysis (LCA), we have explored how changes in the socio-demographic characteristics of residents, as well as home improvement and refinance activity by residents, are related to changes in neighborhood crime over a decade. Utilizing annual home mortgage loan data in the city of Los Angeles from the years 2000–2010, we 1) conducted principle components factor analyses using measures of residential in-migration and home investment activities; 2) estimated LCA models to identify classes of neighborhoods that shared common patterns of change over the decade; 3) described these 11 classes; 4) estimated change-score regression models to assess the relationship of these classes with changing crime rates. The analyses detected six broad types of neighborhood change: 1) stability; 2) urban investors; 3) higher-income home buyers; 4) in-mover oscillating; 5) oscillating refinance; 6) mixed-trait. The study describes the characteristics of each of these classes, and how they are related to changes in crime rates over the decade.     

The immigration–crime relationship: Evidence across US metropolitan areas

Despite popular commentary claiming a link between immigration and crime, empirical research exploring this relationship is sparse. Especially missing from the literature on immigration and crime is a consideration of how immigration affects rates of crime at the macro-level. Although individual-level studies of immigrant criminality and victimization tend to demonstrate that immigrants typically engage in less crime than their native-born counterparts, the effect of immigration on aggregate criminal offending is less clear. In this research, we attempt to address this weakness in the literature by examining the effects of aspects of immigration on crime rates in metropolitan areas. We combine 2000 US Census data and 2000 Uniform Crime Report data to explore how the foreign-born population influences criminal offending across a sample of metropolitan areas. After controlling for a host of demographic and economic characteristics, we find that immigration does not increase crime rates, and some aspects of immigration lessen crime in metropolitan areas.     

Immigration and violence: The offsetting effects of immigrant concentration on Latino violence

Despite longstanding interest in the effects of immigration on American society, there are few studies that examine the relationship between immigration and crime. Drawing from social disorganization theory and community resource/social capital perspectives, this study examines the effects of Latino immigration on Latino violence. Data on violence (i.e., homicide, robbery, and Violent Index) and the structural conditions of Latino populations are drawn from the California Arrest Data (CAL), New York State Arrest Data (NYSAD), and U.S. Census data for approximately 400 census places during the 1999–2001 period. Findings suggest that immigrant concentration has no direct effect on Latino homicide or Violent Index rates but may reduce Latino robbery. Immigration also appears to have multiple, offsetting indirect effects on Latino violence that work through social disorganization and community resource measures. These results suggest that (1) immigrant concentration does not contribute to Latino violence and may even reduce some forms of violence, (2) immigration simultaneously stabilizes and destabilizes structural conditions in Latino populations, and (3) it is useful to examine both the direct and indirect effects of immigration on crime.     

Economic deprivation and racial segregation: Comparing Superfund sites in Portland, Oregon and Detroit, Michigan

The research presented here weighs the ability of two major explanations of social inequality—Massey and Denton’s racial segregation explanation and Wilson’s emphasis on economic deprivation (concentrated poverty)—to predict environmental inequality. Two sets of logistic regression analyses are used to predict the location of Superfund sites in Portland, Oregon and Detroit, Michigan providing a conditional understanding of environmental inequality within a larger sociological context. The analysis includes a general examination of the two theories in all census tracts in both cities and a set of analyses focusing upon Black neighborhoods in Detroit. The findings indicate that there is support for explanations of environmental inequality that include both racial segregation and economic deprivation, but that the more powerful of the two is economic deprivation. The results suggest that even though African-American neighborhoods disproportionately house Superfund sites, these facilities are more likely to be located in Black neighborhoods that are economically deprived.     

Urban density and the metabolic reach of metropolitan areas: A panel analysis of per capita transportation emissions at the county-level

We engage a tension in the urban environment literature that positions cities as both drivers of environmental destruction and loci of environmental protection. We argue that the traditional binary view of cities as either harmful or beneficial is too simplistic; we advance a more nuanced understanding of cities to study their internal and external metabolic effects in terms of carbon emissions from on-road transportation at the county-level across the continental United States between 2002 and 2007. First, utilizing satellite imagery from the National Land Cover Database, we create a novel measure of population density by quantifying the number of people per square mile of impervious surface area. Second, we develop a measure of metropolitan adjacency from the rural classifications datasets published by the USDA. In spatial regression models, we find that while higher density reduces emissions, counties that are geographically isolated from metropolitan areas actually have lower per capita emissions, all else equal. We elaborate on the conceptual, methodological, and practical implications of our study in the conclusion.     

Heterogeneity in the rise and decline of city-level homicide rates, 1976-2005: A latent trajectory analysis

This study applies latent trajectory methods to the analysis of temporal changes in homicide rates among large US cities across recent decades. Specifically, annual homicide rates for 157 large US cities are analyzed for the 30 years from 1976 to 2005. We address the fundamental questions: Did all of cities experience similar levels and patterns of rise and decline in homicide rates over these three decades? Or is there hidden or unobserved heterogeneity with respect to these temporal patterns, thus leading to the identification of more homogeneous groupings of the cities? And if latent homogeneous groupings surface, is membership due to specific structural characteristics found within those cities? Evidence is found for the existence of four latent homicide rate trajectories. After identifying and classifying the cities into these four groups, multivariate statistical techniques are used to determine which social and economic characteristics are significant predictors of these distinct homicide trends. Criminal justice measures are also included as controls. It is found that larger cities located in the South with higher levels of resource deprivation/concentrated poverty, higher income inequality, higher percentages of the adult male population that are divorced, higher unemployment rates, higher percentages of youth, higher percentages of the population who are Hispanic and higher numbers of police per capita are more likely to be in a higher than a lower homicide trajectory group. Higher percentages of the population enrolled in colleges and universities and locations in states with higher incarceration rates are characteristics of cities associated with membership in a lower homicide trajectory group.     

Explaining discrepant findings in cross-sectional and longitudinal analyses: An application to U.S. homicide rates

Cross-sectional studies often reach different conclusions regarding the association between key explanatory variables and outcomes than those of longitudinal approaches. This study considers possible explanations for discrepant findings using a decomposition approach with panel data on 404 U.S. counties for the period 1970–1999. The analysis establishes that there are important differences in the effects of independent variables on homicide rates across counties as opposed to within counties over time. Explanations offered for these discrepancies are that variables may have differing temporary (flow) and permanent (stock) influences on outcomes and possible omitted variable bias. The findings highlight the importance of distinguishing among possible stock and flow effects and are significant not only for the study of crime but also for other social phenomena.     

SPATIAL AND TEMPORAL DIMENSIONS OF NEIGHBORHOOD EFFECTS ON HIGH SCHOOL GRADUATION

Research into the effects of neighborhood characteristics on children’s behavior has burgeoned in recent years, but these studies have generally adopted a limited conceptualization of the spatial and temporal dimensions of neighborhood effects. We use longitudinal data from the Panel Study of Income Dynamics and techniques of spatial data analysis to examine how both the socioeconomic characteristics of extralocal neighborhoods—neighborhoods surrounding the immediate neighborhood of residence—and the duration of exposure to disadvantaged neighborhoods throughout the childhood life course influence the likelihood of graduating from high school. Among blacks and whites, socioeconomic advantage in the immediate neighborhood increases the likelihood of completing high school, but among whites higher levels of socioeconomic advantage in extralocal neighborhoods decrease high school graduation rates. Extralocal neighborhood advantage suppresses the influence of advantage in the immediate neighborhood so that controlling for extralocal conditions provides stronger support for the neighborhood effects hypothesis than has previously been observed. Exposure to advantaged neighborhoods over the childhood life course exerts a stronger effect than point-in-time measures on high school graduation, and racial differences in exposure to advantaged neighbors over the childhood life course help to suppress a net black advantage in the likelihood of completing high school.     

Perceptions as the crucial link? The mediating role of neighborhood perceptions in the relationship between the neighborhood context and neighborhood cohesion

This study examines the effects of neighborhood racial in-group size, economic deprivation and the prevalence of crime on neighborhood cohesion among U.S. whites. We explore to what extent residents' perceptions of their neighborhood mediate these macro-micro relationships. We use a recent individual-level data set, the American Social Fabric Study (2012/2013), enriched with contextual-level data from the U.S. Census Bureau (2010) and employ multi-level structural equation models. We show that the racial in-group size is positively related to neighborhood cohesion and that neighborhood cohesion is lower in communities with a high crime rate. Individuals' perceptions of the racial in-group size partly mediate the relationship between the objective racial in-group size and neighborhood cohesion. Residents’ perceptions of unsafety from crime also appear to be a mediating factor, not only for the objective crime rate but also for the objective racial in-group size. This is in line with our idea that racial stereotypes link racial minorities to crime whereby neighborhoods with a large non-white population are perceived to be more unsafe. Residents of the same neighborhood differ in how they perceive the degree of economic decay of the neighborhood and this causes them to evaluate neighborhood cohesion differently, however perceptions of neighborhood economic decay do not explain the link between the objective neighborhood context and neighborhood cohesion.     

Is immigrant neighborhood inequality less pronounced in suburban areas?

We investigate suburbanization and neighborhood inequality among 14 immigrant groups using census tract data from the 2008–2012 American Community Survey. Immigrant neighborhood inequality is defined here as the degree to which immigrants reside in neighborhoods that are poorer than the neighborhoods in which native whites reside. Using city and suburb Gini coefficients which reflect the distributions of groups across neighborhoods with varying poverty rates, we find that the immigrant-white gap is attenuated in the suburbs. This finding applies to most of the nativity groups and remains after accounting for metropolitan context, the segregation of poverty, and group-specific segregation levels, poverty rates, and acculturation characteristics. Despite reduced neighborhood inequality in the suburbs, large group differences persist. A few immigrant groups achieve residential parity or better vis-à-vis suburban whites while others experience high levels of neighborhood inequality and receive marginal residential returns on suburban location.     

Civic communities and urban violence

Civic communities have a spirit of entrepreneurialism, a locally invested population and an institutional structure fostering civic engagement. Prior research, mainly confined to studying rural communities and fairly large geographic areas, has demonstrated that civic communities have lower rates of violence. The current study analyzes the associations between the components of civic communities and homicide rates for New Orleans neighborhoods (census tracts) in the years following Hurricane Katrina. Results from negative binomial regression models adjusting for spatial autocorrelation reveal that community homicide rates are lower where an entrepreneurial business climate is more pronounced and where there is more local investment. Additionally, an interaction between the availability of civic institutions and resource disadvantage reveals that the protective effects of civic institutions are only evident in disadvantaged communities.     

The social organization of criminal justice processing: An event history analysis

Crimes are social events that involve citizens and control agents interacting over time. Prior work neglects the dynamic and interactive qualities of these criminal events. Drawing from the work of Hawley and others, it is suggested that the processing of criminal events is a routine activity socially organized in time and space. Dynamic modeling techniques developed by N. Tuma are applied to longitudinal data collected on over 10,000 criminal events in California cities and used to model rates of transition from arrest to case disposition resulting from police release, prosecutor denial of complaint, or going to court. As the work of Hawley predicts, city size has much to do with the way criminal events are processed. For example, in larger cities it is demonstrated that crime specialists are processed more slowly than nonspecialists, and that each successive police processing of crime specialists results in slower rates of transition relative to nonspecialists; in smaller cities, it is demonstrated that black suspects are processed more quickly than whites, and that each successive police processing of black suspects results in faster rates of transition relative to whites. The former findings are explained in terms of rationalized intelligence gathering, the latter in terms of stereotyping and the harassment of minorities. The systematic form of the observed temporal changes, notwithstanding a large number of legal and extralegal variables taken into account, leads us to believe that we have identified important patterns of police activity. These and other findings convince us that the social organization of criminal justice processing deserves further study.     

Rush to the border? Market liberalization and urban- and rural-origin internal migration in Mexico

In this study we examine the social and economic factors driving internal migration flows in Mexico. We pay particular attention to the effect that economic liberalization has had in encouraging migration to border cities. Our analysis of the origin and destination of migrants is carried out at a finer level of geographical detail than ever before. Microdata files from the 2000 population census allow us to distinguish urban- and rural-origin migrants to the largest 115 cities and metropolitan areas in the country. Our results indicate that economic liberalization, measured by the level of foreign investment and employment in the maquiladora export industry, strongly influences migrants’ choice of destinations. However, economic liberalization fails to fully account for the attraction of the border, as do the higher emigration rates to the United States from border cities. Our analysis also reveals that migrants to the border region and to cities with high levels of foreign investment are younger, less educated and more likely to be men than migrants to other parts of Mexico. Rural migrants are significantly more likely to move to the border and to cities with high levels of foreign investment than urban migrants. The results of our study have important implication for other countries opening their economies to foreign investment and international trade.     

Latent class analysis of response inconsistencies across modes of data collection

Latent class analysis (LCA) has been hailed as a promising technique for studying measurement errors in surveys, because the models produce estimates of the error rates associated with a given question. Still, the issue arises as to how accurate these error estimates are and under what circumstances they can be relied on. Skeptics argue that latent class models can understate the true error rates and at least one paper (Kreuter et al., 2008) demonstrates such underestimation empirically. We applied latent class models to data from two waves of the National Survey of Family Growth (NSFG), focusing on a pair of similar items about abortion that are administered under different modes of data collection. The first item is administered by computer-assisted personal interviewing (CAPI); the second, by audio computer-assisted self-interviewing (ACASI). Evidence shows that abortions are underreported in the NSFG and the conventional wisdom is that ACASI item yields fewer false negatives than the CAPI item. To evaluate these items, we made assumptions about the error rates within various subgroups of the population; these assumptions were needed to achieve an identifiable LCA model. Because there are external data available on the actual prevalence of abortion (by subgroup), we were able to form subgroups for which the identifying restrictions were likely to be (approximately) met and other subgroups for which the assumptions were likely to be violated. We also ran more complex models that took potential heterogeneity within subgroups into account. Most of the models yielded implausibly low error rates, supporting the argument that, under specific conditions, LCA models underestimate the error rates.     

Solamente Mexicanos? Patterns and sources of Hispanic diversity in U.S. metropolitan areas

Rapid Hispanic growth has been a major source of increasing ethnoracial diversity in the United States. However, diversity within the Hispanic population is frequently obscured by the tendency to lump all Latinos together. Our study examines Hispanic diversity at the local level, drawing insights from the Mexican dominance, Caribbean-centric settlement, spatial assimilation, and economic opportunity perspectives. Measures of the magnitude and structure of Hispanic origin-group diversity during the 1990–2010 period are constructed for 363 metropolitan areas based on each area's shares of Mexicans, Puerto Ricans, Cubans, Dominicans, Salvadorans, Guatemalans, Colombians, and ‘others’. We find that diversity magnitude varies markedly across metropolitan Hispanic populations. Although the most diverse metro areas lack a majority origin group, Mexicans often constitute a majority or plurality of local Latinos. Diversity levels and structures have remained relatively stable over time. In both 1990 and 2010, metro areas with more diverse, multigroup Hispanic communities are distinguished by their larger size, smaller proportion of Hispanics, location farther from Mexico and closer to the Caribbean, and greater odds of being a military hub. They also exhibit higher rates of housing construction and lower rates of agricultural and manufacturing employment. We use weighted data to show that Dominican metro dwellers experience the highest Hispanic diversity while the average Mexican lives in an area where four-fifths of all Latinos are Mexican. Overall, our results provide primary support for the Mexican dominance perspective but some support for the other three perspectives as well.