RACE AND PRISON VIOLENCE*

Data from 58 male institutions in the federal correctional system were used to test for racial differences in both violent and alcohol/drug misconduct, controlling for a large number of individual, prison environment, and community background variables. Because “structurally” the in-prison station of black and white inmates is essentially identical, the data provide a unique methodological opportunity to test deprivation versus importation models of prison adjustment as well as more encompassing structural versus cultural theories of violence. The major findings are that, net of controls, black inmates have significantly higher rates of violent behavior but lower rates of alcohol/drug misconduct than white inmates. These patterns parallel those of racial differences in the larger society. We interpret these findings as supporting the importation theory of prison adjustment and the subculture of violence thesis regarding high rates of black violence in the larger society.     

THE IMPACT OF OUT-OF-STATE PRISON POPULATION ON STATE HOMICIDE RATES: DISPLACEMENT AND FREE-RIDER EFFECTS*

Past studies of the impact of prison population on homicide rates have produced widely divergent results. Those using state-level data find small impacts, but those using national data find very large ones. We use displacement/free-rider theory to explore the difference between these results. Displacement, in the current context, refers to a criminal's movement away from states with higher imprisonment rates. Free riding occurs when a state benefits from criminals being incarcerated in other states. If the displacement effect holds, a state's prison population has a stronger impact on crime within the state than would be accomplished by deterrence and incapacitation alone. If the free-rider effect holds, higher prison populations outside the state reduce homicide in the state because criminals are incapacitated elsewhere. Using vital statistics data for 1929 to 1992, we conduct separate homicide regressions for each state using both in-state and out-of-state prison population as independent variables. We find that the out-of-state variable has a much larger (negative) association with homicide, indicating substantial free riding. We also find evidence of a small displacement impact.     

SELECTIVE INCAPACITATION,THREE STRIKES,AND THE PROBLEM OF AGING PRISON POPULATIONS: USING SIMULATION MODELING TO SEE THE FUTURE*

Research Summary: The growth of prison populations over the last three decades is a great source of concern for policy makers and observers. One mechanism by which this growth occurs is via sentencing reforms that extend length of stay for certain categories of offenders. This has the effect of aging prison populations, which is problematic for many reasons. Apart from the increased financial burdens entailed in caring for older prisoners, it is also important to consider the intent of reforms in evaluating them. Of late, sentencing reform has become increasingly focused on the selective incapacitation of dangerous offenders. Policies that have the effect of aging the prison population are problematic from this perspective due to the diminishing returns realized with respect to incapacitation as offenders age. Dynamic systems simulation analysis is employed to investigate the likely consequences of recent sentencing reforms that increase length of stay for some offenders. These analyses indicate that the effects of recent reforms may not be as dramatic as some observers have predicted, but they suggest that the consideration of alternatives to incarceration for elderly offenders is warranted from the standpoint of cost considerations as well as that of selective incapacitation. Policy Implications: The results indicate that California's Three Strikes law will not accelerate the rate of growth of the elderly prison population. However, even without increasing the proportional representation of elderly prisoners, the number of elderly prisoners is expected to grow substantially over the next three decades. These prisoners will strain criminal justice system resources while presenting little public safety threat. State criminal justice policy makers and their constituents should closely examine laws that impose very long stays without discretionary release, as these statutes may contribute to the production of elderly prisoners. This problem is particularly pronounced in Three Strikes and other habitual offender laws that use retrospective methods to identify habitual offenders. Additionally, the effects of reforms lengthening stay for some offenders must be considered in light of cumulative effects of sentencing reform resulting in changes to the demographic structure of the prison population overall. Dynamic systems simulation modeling is presented as a valuable policy‐making tool, as it allows the policy analyst to examine the potential impacts of laws in the absence of data suitable for conventional statistical analyses.     

NEIGHBORHOOD HOUSING INVESTMENTS AND VIOLENT CRIME IN SEATTLE, 1981–2007*

Despite significant advances in the study of neighborhoods and crime, criminologists have paid surprisingly less attention to the extralocal forces that shape violence. To address this issue, we draw on an emerging body of work that stresses the role of home mortgage lending—a resource secured via interaction with external actors—in reducing neighborhood violence and extend it by addressing concerns that the lending–violence relationship is spurious and confounded by simultaneity. We explore the longitudinal relationship between residential mortgage lending and violence in Seattle with a pooled time series of 118 census tracts over 27 years, and we instrument our endogenous predictors (home mortgage lending and violent crime) with changes in their levels from prior periods. Employing Arellano–Bond difference models, we assess both the effect of mortgage lending on violent crime as well as the effect of violent crime levels on mortgage activity. We find that infusions of home mortgage lending yield reductions in subsequent violent crime; yet the impact of violent crime on subsequent lending is not significant. Results underscore the importance of incorporating external forces such as home mortgage lending into explanations of neighborhood violence.     

“CRACK,” STREET GANGS,AND VIOLENCE*

This investigation considers the connections among street gangs, “crack” cocaine, and violence associated with crack distribution during the initial years of crack proliferation. Data were extracted from the narcotics investigation files and homicide fires of five Los Angeles Police Department and Sheriff's Department areas where both crack and gangs were prominent. The aims were to compare for 1983–1985, when crack first emerged as a significant problem, hypotheses about (1) gang involvement in crack distribution and (2) concomitants of gang involvement, particularly violence. The analyses confirm a dramatic growth in crack sales, an accompanying increase in gang members involved, but a declining rate of involvement, and inconsistent evidence on the impact of gang involvement on sales events. We infer that crack distribution, while including many individual gang members, was not primarily a street gang phenomenon.     

RACE AND CRIME IN POSTWAR AMERICA: DETERMINANTS OF AFRICAN-AMERICAN AND WHITE RATES, 1957–1988*

The idea that crime and deviance are explained mostly by access to opportunities—especially those provided by employment, income, education, and family stability—is one of the most powerful assumptions about crime in postwar America. However, despite its importance, the actual relationship between opportunity measures and crime during this period remains little understood. while cross-sectional studies of these issues have become common, few longitudinal studies exist and those that do include a limited number of variables. Moreover, despite important differences in the history and experiences of African-Americans and whites during this period, researchers have assumed similar dynamics by race. In this paper, we use annual time-series data from 1957–1988 to examine the effects of economic well-being, educational attainment, and family stability on rates of robbery, burglary, and homicide for blacks and whites. Our results show that these measures have different—usually opposite—effects on black and white crime rates during the period. In general, measures of opportunity have expected effects on white but not black rates. We consider the implications for policy and research.     

RACIAL THREAT,PARTISAN POLITICS,AND RACIAL DISPARITIES IN PRISON ADMISSIONS: A PANEL ANALYSIS*

Why are racial disparities in imprisonment so pronounced? Studies of alternative outcomes in the criminal justice system find positive relationships between minority presence and punitive outcomes. Therefore, it is puzzling that the studies of racial incarceration ratios find negative relationships between this presence and such discrepancies. We use a pooled time‐series design to resolve this dilemma. Successful Republican attempts to link crime with public concerns about a dangerous racial underclass also suggest that where these racial appeals are successful, African Americans should face higher incarceration rates than whites. In contrast to prior research, our results are consistent with findings about other criminal justice outcomes. They show that an inverted, U‐shaped, nonlinear relationship is present between African‐American presence and racial disparities in imprisonments. Additional results indicate that the presence of African Americans in deep southern states and greater support for Republican presidential candidates together with increases in the most menacing crime (which often is blamed on African Americans) also help to explain these discrepant racial prison admission rates.     

CHANGES IN POLICE NOTIFICATION FOR RAPE, 1973–2000*

We use data from the National Crime Survey (NCS) and the National Crime Victimization Survey (NCVS) to explore changes in the likelihood of police notification in rape incidents. The findings indicate that during the 1970s and 1980s there was a significant increase in police notification by third parties and by victims raped by non‐strangers. During the 1990s the increase in rates of police notification in rape incidents accelerated and broadened in scope. In addition, differences in police notification between stranger and non‐stranger incidents diminished during the 1970s and 1980s and, by the early 1990s there was no significant difference.     

A COMPARISON OF CHANGES IN POLICE AND GENERAL HOMICIDES: 1930–1998*

This paper presents a new data series for homicides of law enforcement officers. Available for more than two centuries, it is much longer than series previously examined. Police killings had two extreme peaks, one in the 1920s and another in the 1970s. We use the post‐1930 part of the series in a time‐series regression to explore structural conditions that affect police killings in the short term. Economic conditions, prison populations, and World War II have considerably larger impacts on police killings than on homicide generally. Police killings are less affected by demographic changes and by the crack epidemic.