EXPLAINING JAPAN'S POSTWAR VIOLENT CRIME TRENDS*

Japan has long been recognized for its low rates of violent crime, rates that usually seem to be declining. The most common explanation for postwar rates links unique cultural characteristics to a system of exceptionally effective informal social controls that, at the macro level, suggest low levels of social disorganization. Other common explanations include low levels of economic stress, a small proportion of young males and a criminal justice system that delivers a high certainty of punishment. In this paper we test these four explanations for Japanese trends using both an annual time‐series national analysis (1951 to 2000) and a pooled cross‐sectional time‐series analysis of the 47 Japanese prefectures from 1955 to 2000 (at 5‐year intervals). The results from the two analyses are largely congruent. They show that measures of economic stress, certainty of punishment and age structure are–compared to common social disorganization measures–more consistent predictors of Japanese postwar violent crime trends. Our results suggest that the remarkable strength of the postwar Japanese economy may play a larger role in explaining Japanese violent crime rates than is usually recognized.     

The Cohort-Size Sample-Size Conundrum: An Empirical Analysis and Assessment Using Homicide Arrest Data from 1960 to 1999

A number of studies use the Age-Period-Cohort Characteristic (APCC) model to address the impact of cohort related factors on the age distribution of homicide offending. Several of these studies treat birth cohorts as spanning several years, an operationalization that most closely matches tenets of cohort theory, yet sharply reduces the number of observations available for analysis. Other studies define birth cohorts as those born within a single year, an operationalization that is theoretically problematic, but provides many more observations for analysis. We address the sample size problem by applying a time-series-cross-section model (panel model) with age-period-specific homicide arrest data from the United States for each year from 1960 to 1999, while operationalizing cohorts as five-year birth cohorts. Our panel model produces results that are very similar to those obtained from traditional multiyear APCC models. Substantively, the results provide a replication of work showing the importance of relative cohort size and cohort variations in family structure for explaining variations in age-period-specific homicide rates. The additional observations provided by our approach allow us to examine these relationships over time, and we find substantively important changes. The year-by-year estimates of the age distribution of homicide offending help us to examine the model during the epidemic of youth homicide.     

Appellate and trial court caseload growth: A pooled time-series-cross-section analysis

This is an econometric study of factors behind filing growth since 1970 in state trial courts and, especially, appellate courts. The model posits two categories of variables: those affecting the supply of disputes and those affecting the costbenefit considerations of potential litigants. The study uses a pooled time-seriescross-section design and a fixed-effects regression procedure. The overall conclusion is that factors determining the supply of disputes overwhelm other factors entered in the model. At the trial level, economic conditions 2 years earlier strongly affect civil filings, and crime rates for the current and prior year have moderate impacts on criminal filings. The output capacity of trial courts, measured by the number of judges, has a strong impact on appeals. Economic conditions and trial court filings influence civil appeals in later years, and prison commitments influence criminal appeals. The great majority of cost-benefit factors, such as simplification of appellate procedure and interest-rate differentials, showed little or no impact on appeals.In view of the rising emphasis on replicability (Dewaldet al., 1986; Campbell, 1986; Cook and Zarkin, 1986), we stress that the data for this research are available from the Inter-Univarsity consortium for Political and Social Research, and we will provide copies of the computer printouts giving the programs (that apply SAS) and the results. In addition, Court Studies periodically updates the data set, and it will supply current data upon request.     

Imprisonment and unemployment: A comment

The current state of research dealing with the relationship between the unemployment rate and the imprisonment rate is assessed, and directions for further research are suggested.     

IMMIGRATION AND THE RECENT VIOLENT CRIME DROP IN THE UNITED STATES: A POOLED,CROSS‐SECTIONAL TIME‐SERIES ANALYSIS OF METROPOLITAN AREAS*

A good deal of research in recent years has revisited the relationship between immigration and violent crime. Various scholars have suggested that, contrary to the claims of the classic Chicago School, large immigrant populations might be associated with lower rather than higher rates of criminal violence. A limitation of the research in this area is that it has been based largely on cross‐sectional analyses for a restricted range of geographic areas. Using time‐series techniques and annual data for metropolitan areas over the 1994–2004 period, we assess the impact of changes in immigration on changes in violent crime rates. The findings of multivariate analyses indicate that violent crime rates tended to decrease as metropolitan areas experienced gains in their concentration of immigrants. This inverse relationship is especially robust for the offense of robbery. Overall, our results support the hypothesis that the broad reductions in violent crime during recent years are partially attributable to increases in immigration.     

Unemployment and Crime Rate Fluctuations: A Comment on Greenberg

Several ways in which the specification of the Cantor and Land (1985) conceptual model of transient relationships between aggregate unemployment and crime rate fluctuations differs from that of Greenberg (2001) are noted. It follows that we do not accept Greenberg's Eq. (1) as a valid theoretical representation of the processes of interest. We briefly review the substantive context from which our investigation began in the mid-1980s. We also review the time series properties of our model and of the aggregate unemployment and crime rates used in its estimation. We note how the time series behavior of various crime rates determines which parts of the Cantor and Land model are and are not likely to be estimated as statistically significant for those series. We conclude with some comments on the limitations of aggregate time series research designs for testing the behavioral hypotheses used to generate expected relationships between aggregate unemployment and crime rates and suggest some alternative research designs.     

Theory, Operationalization, Identification, and the Interpretation of Different Differences in Time Series Models

The principals in this exchange, David Cantor, David Greenberg, and Kenneth Land, have each contributed outstanding methodological and substantive scholarship to criminology and to the social sciences more generally. In exchanges such as this we are engaged in a collective process of refining and advancing our knowledge in the areas of both criminology and statistical methods. The issues raised by Greenberg provide an opportunity to reflect on time series modeling and to revisit some issues concerning operationalization and model specification. I do not expect any of us to agree fully on these issues, but we may be able to shed some light on these issues, and collectively this may help the larger community of criminologists and social scientists.     

Prison commitments,crime, and unemployment: A theoretical and empirical specification for the United States, 1933–1985

This paper examines sources for the changing commitment rates to U.S. state prisons (PCR) from 1933 to 1985 using a variety of time-series techniques. Theoretically, it resolves ambiguous interpretations of how crime, unemployment, and imprisonment are related. Hypotheses that crime and punishment are in equilibrium are rejected. Our final specification supports theories integrating institutionally endogenous and socially exogenous causes of prison use and includes feedback effects between crime and punishment. We reached several general conclusions. (1) Changes in PCR are due partly to changes in the levels of unemployment, age composition of the population, and military active-duty rates. (2) Effects of the criminal justice system, captured as autoregressive institutional drift, account for approximately half of the year-to-year fluctuations in the PCR. The contemporaneous prison discharge rate also influences the rate of prison commitments. (3) Neither the specified nor the unspecified institutional effects mediate the effects of other exogenous variables. (4) Under most simultaneous-equation specifications, the crime rate is moderately influenced by the contemporaneous unemployment rate and strongly influenced by prior levels of prison commitments. The preferred simultaneous causal model estimates a modest positive coefficient for the unemployment-crime causal path and a substantial positive coefficient for the unemployment-prison commitment causal path.     

Alternative Strategies for Identifying the Link Between Unemployment and Crime

National-level time series data are a crude tool for distinguishing between two alternative behavioral explanations for a link between unemployment and crime. Consequently, inferences drawn from aggregate time series estimates are likely to be misleading. A more fruitful approach to learning about the link between unemployment and crime would be to utilize a menagerie of different methodological approaches such as cross-section and panel data analysis of less geographically aggregated areas, natural experiments, international data, individual-level data, and ethnography.     

COUNTING CRIME BOOMS AMONG NATIONS: EVIDENCE FOR HOMICIDE VICTIMIZATION RATES, 1956 TO 1998*

Constructionists argue that crime booms are rare, modernizationalists predict that booms have been limited to industrializing nations, and globalizationalists claim that booms are universal among nations since World War II. We define crime booms as rates that increase rapidly and exhibit a positive sustained change in direction and use econometric methods to test for booms with homicide victimization rates for 34 nations, 1956 to 1998. Twelve nations satisfied our criteria for booms—too many to support constructionists, but too few to support globalizationalists. In support of modernizationalists, 70% of industrializing nations qualified as having booms, but fewer than 21% of industrialized nations did. Future research should explain industrializing nations that do not experience booms and industrialized nations that do.     

Delay reduction in the federal courts: Rule 50(b) and the Federal Speedy Trial Act of 1974

During the 1970s, 94 federal district courts implemented two major policy initiatives, Rule 50(b) of theFederal Rules of Criminal Procedure and the Federal Speedy Trial Act, that were designed in Washington to combat delay in the processing of federal criminal cases. Both of these initiatives established a national priority of delay reduction in criminal cases, encouraged local district court planning for delay reduction; established reporting procedures for monitoring local compliance, and provided for the determination of quantitative goals for the time to disposition of criminal cases. Neither initiative mandated specific activities for delay reduction; this determination was left to the discretion of local federal district courts. This research examines the effectiveness of Rule 50(b) and the Speedy Trial Act by constructing a 150-month time series of three measures of case processing time. A multiple-intervention time-series model found that both of these initiatives contributed to the dramatic reduction in the time to disposition in federal criminal cases. These effects persisted after controls for changes in case characteristics and judicial resources were introduced.Points of view expressed in this research are those of the author and do not necessarily represent the official position of the U.S. Department of Justice.     

TESTS OF NONLINEAR DYNAMICS IN U.S. HOMICIDE TIME SERIES,AND THEIR IMPLICATIONS*

Attempts to explain temporal patterns in U.S. homicide data usually assume that a linear process accounts for the variation. A nonlinear process is an obvious alternative, however, and reasonable arguments suggest that nonlinearity may in part underlie homicide series dynamics. This paper applies tests for nonlinearity to national time series of homicide rates and counts. The results provide relatively little evidence of nonlinear structure; instead, a linear (random walk) process appears to generate most of the change in the series. Although this supports the unstated assumptions of current theories, it also raises questions about why homicides should follow a linear time path in the first place.     

The “war on cops,” retaliatory violence,and the murder of George Floyd*

The police murder of George Floyd sparked nationwide protests in the summer of 2020 and revived claims that public outcry over such high-profile police killings perpetuated a violent “war on cops.” Using data collected by the Gun Violence Archive (GVA) on firearm assaults of U.S. police officers, we use Bayesian structural time series (BSTS) modeling to empirically assess if and how patterns of firearm assault on police officers in the United States were influenced by the police murder of George Floyd. Our analysis finds that the murder of George Floyd was associated with a 3-week spike in firearm assaults on police, after which the trend in firearms assaults dropped to levels only slightly above that which were predicted by pre-Floyd data. We discuss potential explanations for these findings and consider their relevance to the contemporary discussion of a “war on cops,” violence, and officer safety.     

Time Series Analysis of Crime Rates

A methodological critique of Cantor and Land's (1985) approach to the time series analysis of the crime–unemployment relationship is developed. Error correction models for U.S. homicide and robbery rates for the years 1946–1997 are presented to illustrate procedures for analyzing nonstationary time series data. The critique is followed by a discussion of methodological problems in work by Devine et al. (1988), Smith et al. (1992), and Britt (1994, 1997) that builds on Cantor and Land's approach.     

Trends in victimization in schools and elsewhere, 1974–1981

Trends in the rate of victimizations of juveniles in three settings-schools, homes, and streets/parks-are examined monthly during the period 1974–1981. The relationship between in-school victimization rates and those occurring outside of school are analyzed with multivariate ARMA models informed by previous research on school victimization (Gottfredson and Gottfredson, 1985) and an importation perspective on the source of crime and victimization in institutions such as schools. Results indicate that the overall in-school victimization rate remained relatively stable during this period but that victimization rates of juveniles in other settings had significant effects on in-school victimizations. This suggests that underlying causes of victimization in general are important determinants of victimization in schools. These results are limited, however, as we examine these sources of victimization only indirectly via relationships among the different victimization rates in dynamic models and by the aggregate nature of the monthly data from the National Crime Survey.     

Testing Theory and the Analysis of Time Series Data

Although the relationship between unemployment and crime has been a longstanding interest in criminology, there is little agreement about appropriate models for analyzing this relationship. David Greenberg's (this issue) discussion highlights two issues that raise questions about recent research on the unemployment–crime relationship. First, he extends the work of Hale and Sabbagh (1991) and argues that cointegration methods should be used instead of first-differenced regression models to analyze unemployment and crime time series data. Second, he argues that previous attempts to test his strain theory linking unemployment to the age distribution of crime rely on flawed hypotheses, inappropriate data, and faulty measurement. In this paper, I address both of Greenberg's claims. I begin with a discussion of the relative utility of cointegration analysis and of first-differenced regression models for the analysis of the unemployment–crime relationship, focusing on the link between theory and statistical model. I then discuss the possibility of ever testing and falsifying Greenberg's strain theory.     

A War on Cops? The Effects of Ferguson on the Number of U.S. Police Officers Murdered in the Line of Duty

Police agencies in the U.S. are currently facing a major legitimacy crisis resulting from a spate of high-profile use of force incidents, many involving minority citizens. Recent headlines emphasize that there is now a “war on cops” and that police officers are facing increasing levels of hostility and violence fueled by a growing anti-police sentiment. In the aftermath of events in Ferguson, Missouri in August 2014, some commentators claim that the number of police officers feloniously assaulted and killed in the line of duty has increased sharply. Using time series analysis of data from the Officer Down Memorial Page, we test whether events in Ferguson were associated with an increase in the number of police officers murdered in the line of duty. Our results provide no evidence for a “Ferguson Effect” on the number of U.S. police officers murdered in the line of duty as of March 2016.     

Measuring the Convergence/Divergence of “Serious Crime” Arrest Rates for Males and Females: 1960–1995

It is now almost a quarter of a century since Adler (1975) and Simon (1975)stimulated a debate about the convergence of crime rates for men andwomen. The ensuing debate generated literally dozens of papers. Given theexistence of a series that now extends from 1960 to 1995, this papersuggests an appropriate way to examine the convergence hypothesis usingtime series techniques. These techniques take into consideration the effectsof the following factors: (a) random shocks orinnovations, (b) the potentially lasting effects of suchinnovations, and (c) the autocorrelation that time series oftenexhibit. Using time series techniques on annual data, we examine trends inthe arrest rates for males and females for six Part I crimes (homicide,robbery, aggravated assault, burglary, larceny, and motor vehicle theft)for the years 1960 through 1995. We test for convergence, divergence, notrend, and a special condition of equilibrium between series calledcointegration.