Class and Crime

Using slightly different kinds of analysis, studies of the geographic distribution of crime in the 1950s and 1960s generally reinforced the findings of Shaw and McKay that official delinquency rates for small urban areas were linked to indicators of poverty and disadvantage (Chilton). Research done in the last two decades of the century continued in both styles. A renewed interest in studies looking at the geographic distribution of crime produced additional evidence in support of a class-crime link. Patterson's 1991 review of twenty-two studies of poverty and crime published from 1976 to 1986 found that some of the studies used data for different sets of cities, for MSAs, and for areas within cites. Although most of the studies showed positive effects of poverty on crime, some did not. In his analysis of fifty-seven areas within Tampa, Florida, Patterson found that levels of absolute poverty were associated with higher rates of violent crime.

During the same period, some researchers using reports of individuals suggested that while social origin might play a minor role in explaining juvenile criminality, the effect of the subject's own social position is important for adult criminality (Thornberry and Farnworth). Others suggested that the correlations between self-reported delinquency and social class are weak and should be weak in part because of the offenses used and in part because traits associated with high and low social class scores are related to different kinds of crime. Responding to the general absence of studies on the impact of social class on adult crime, Dunaway and his colleagues used three different measures of social class to analyze the responses of an adult sample for a single city.

Dunaway and colleagues' "underclass" measure focused on unemployment, receiving public assistance or food stamps, or living in public housing. Another measure used income and education as gradational measures of class. Their third measure of social class focused on a respondent's business ownership and position as an employer or employee. As a measure of crime they used the total number of offenses reported when respondents were asked to check one or more offenses from a list of fifty that they might have committed over the preceding year. This approach gives equal weight to an admission of marijuana possession, illegal gambling, driving while drunk, income tax fraud, threatening to hit a family member, stealing, burglary, robbery, and assault with intent to kill.

Recognizing the problematic nature of this range of offenses, they created a separate violence measure that included some relatively minor offenses but also included serious assaults, rape, and robbery. Using the violence subset as a measure of crime, they reported an inverse relationship between crime and some of their social class measures. When the full set of offenses is used to measure crime, only income is inversely related to crime. While arguing that there was little impact of class on crime if categorical measures of class are used, they note that family income negatively affects crime by both men and women, that the results vary by race in that the class-crime relationship was stronger for white respondents than for black respondents, and that violence is related to social class when income is used to measure class.

In a New Zealand study, Wright and others report that their Socioeconomic Status Score (SES) had both a negative and a positive indirect affect on delinquency. Using data for 1,037 children born in 1972 and 1973 and reassessed eight times since birth, they found no association between parental SES and delinquency at age twenty-one before they looked at several mediating factors. They interpret this as the result of high self-reported delinquency scores for middle-class young people that are high for reasons different from the reasons for high self-reported delinquency scores of lower class young people. They argue that there can be causality without correlation.

While this may explain the results observed in many individual-level studies, another possible explanation of the conflicting results between self-report studies and area studies is the distinctly different locations of the people and situations studied. Studies of geographic location are usually carried out for urban areas, Metropolitan Statistical Areas, urban counties, cities, or census tracts. Confessional studies have frequently been carried out in small towns and areas with very small minority populations. These studies have often been unable to tap both the high and the low ends of the social class distribution. Nowhere is this clearer than in the way the two approaches deal with race. One classic self-report study dropped all black respondents from the analysis (Hirschi). Other self-report studies attempt to hold constant the impact of race. Such procedures are rare in studies of geographic areas. The area studies include minority populations in the crime counts and in the population counts. Whether the areas are census tracts, cities, or Metropolitan Statistical Areas, the populations studied are almost always urban and multiracial.

U.S. public health statistics on homicide as a cause of death indicate that this is a leading cause of death for black males (Anderson, Kochanek, and Murphy). About 40 percent of all homicide victims are black males though black males make up about 6 or 7 percent of the U.S. population (U.S. Bureau of the Census). Although the 40 percent figure has fluctuated some since 1960, the victimization rate for black males has been remarkably consistent for forty years—ranging from 33 to 49 percent. Forty percent was also the figure provided by the Uniform Crime Reports' Supplementary Homicide reports for 1995 (Snyder and Finnegan).

The Federal Bureau of Investigation's Supplementary Homicide Reports (SHR) also suggest that black offenders are responsible for most homicides with black victims. They suggest that 48 to 50 percent of offenders in homicide cases are black males and that most homicides are intraracial (Federal Bureau of Investigation). More importantly, black males have been overrepresented in both the victimization figures and the offender figures for over thirty-five years. During the period 1960 to 1990, the average percentage of homicide victims reported as black males was about 39 percent. In addition, the SHR offender information suggests that, on average, about 44 percent of the people reported as homicide offenders were described as black males. There is little doubt that black males are, and have for some years been, greatly overrepresented as both victims of homicide and as homicide offenders.

The traditional response to any discussion of this situation is the suggestion that these high homicide-offending rates for black males are more a function of social class than biological or cultural differences. However, it is almost as traditional to suggest that we lack sufficient information on social class to claim empirical support for the social class explanation. One way to clarify this murky situation would be through the construction of race- and gender-specific homicide rates for census tracts. Peterson and Krivo analyzed homicide victimization rates for 125 U.S. cities and found that black homicides were linked to racial segregation. Parker and McCall's city-level analysis of interracial and intraracial homicide provides another indication of the probable utility of race-specific data. Using race-specific independent variables for about one hundred U.S. cities, they conclude that economic deprivation affects the intraracial homicide rates for whites and blacks.

In a study that used arrest counts to create race-specific offense rates, Ousey reported a large gap between black and white homicide rates. The black rates were five times as high as the white rates. Although he found that measures of poverty and deprivation had an impact on both black and white homicide rates, he found that the effects of these variables were stronger for whites than for blacks. He suggests that extensive and long-term disadvantage may have produced cultural and normative adaptations that have produced this gap in the rates.

Because social status is the term used in the self-report studies wherein young people are asked about their parents' occupations and their own delinquency, it may be misleading in a discussion of race, class, and crime. Even a term such as "economic conditions" is too vague to describe the ways in which vast differences in income and assets and a pervasive system of racial separatism probably contribute to high homicide rates in some areas of U.S. central cities. For an understanding of this issue, asking why the homicide rates are so high in specific areas of U.S. cities is probably more useful than asking individuals how much crime they have committed and comparing their reports with the social class implied by reports of a parent's occupation.

The patterns of homicide rates by race suggest that the rates are probably linked to exclusion and segregation—economic, racial, and ethnic—but especially to the separation and isolation of large segments of the urban population based on income and assets. This separation is frequently based on race or ethnicity but it is increasingly linked to a combination of racial separatism and poverty. In most studies using census tracts or other relatively small areas, a concentration of the poor in areas with high homicide rates was related to low median incomes, low educational attainment, higher proportions of lowpaying occupations, unemployment, and underemployment in the areas. These indicators in turn are probably closely related to housing conditions, living arrangements, and family composition. In these same areas, additional research will probably show reduced public service facilities (parks, pools, libraries, recreation centers) and reduced expenditures for schools and possibly even for police services. In short, expanded and race-specific studies of the geographic distribution of homicide rates will probably show that areas with high homicide rates are areas with concentrations of poor individuals and poor families, regardless of race or ethnicity.

To the extent that these rates reflect the impact of exclusion, isolation, and impoverishment, a continuing focus on short-term trends will leave the extensive and persistent long-term differences unexamined and unexplained—especially the relatively stable and unusually high rates of homicide victimization and homicide offending reported for black males. To understand this long-term trend we will probably have to look to widespread practices and procedures that persist over time and continue to exclude and isolate a large number of black males from full participation in the economic, political, and social life of American society. It is in this sense that race is closely linked to class as a cause of violent crime in the United States. The class effects are compounded by racial separatism and racial discrimination.

Moreover, as John Hagan has suggested, the relationship between class and crime may be class- and crime-specific. It is also probably race- and gender-specific. He is probably also right in his assertion that not only does class have an impact on crime but some kinds of crime, or at least some responses to crime, have an impact on the social class of some offenders (Sampson and Laub). This is why he is right in his assessment that "the simple omission of class from the study of crime would impoverish criminology."

All of this suggests that the class-crime relationship will continue to generate research, comment, and debate well into the twenty-first century. As more of the research on this issue is focused on specific offenses and specific types of offenses, there may be greater coherence in the results than is now available. The development of standard measures of social class and greater attention to the kinds of questions being asked when using officially aggregated information as distinct from the kinds of questions asked in cohort or confessional studies may reduce some of the confusion surrounding the issue. However, the issue will remain controversial for reasons unrelated to scholarship or social research because of the implications for social policy suggested by any set of clear conclusions in one direction or the other.