View this issue as a PDF: CrimHighlightsV1N4.pdf
Why is the unemployment rate in the U.S.A. so much lower than the rates in most European countries and Canada? It’s because the “surplus labour” is in prison. “The dramatic increase in the U.S. prison and jail population has made U.S. unemployment rates look lower than they would be otherwise” -- U.S. unemployment rates would jump by about 2% if their (unemployed) prison population were considered part of the work force.
The denominator of most unemployment rate calculations is the size of the work force: those employed and those unemployed (and looking for work). The numerator is those not employed. Another group who might have preferred to be in the work force are those in prison: they are neither employed nor unemployed. For the purposes of unemployment rate calculations, they simply do not exist. Adding them to the numerator and denominator would increase U.S. unemployment figures by roughly two percent.
As long as rates of imprisonment are kept low, this does not create much of a distortion. This paper suggests that (using 1990 data) the difference between “conventional” and “adjusted” (considering prisoners as an unemployed part of the work force) unemployment rates for most European countries is about 0.1 to 0.3%. In other words, if prisoners were to be considered to be “unemployed” in a conventional way, the European unemployment figures would increase by very little. In contrast, using 1990 figures for the U.S. (when there were “only” 1.1 million in jails and prison in that country), adjusting the unemployment rate to take into account the number of unemployed prisoners would increase the U.S. unemployment rate by about 1.5%. By 1994, if prisoners were counted, the male unemployment rate should be adjusted upwards by about 2 percentage points.
The impact of counting prison populations when calculating unemployment figures is dramatic in the U.S. (but not other countries) for a simple reason: the ratio of unemployed to imprisoned residents in the U.S. is much lower compared to most countries. Using 1990 figures, in the U.S. there is one prisoner for every 3.5 unemployed persons. In Europe in 1990, the ratios were quite different: Germany has one prisoner for every 19.5 unemployed people; France has one prisoner for every 20.9 unemployed; Denmark one prisoner for every 35.8 unemployed.
As the authors point out, if prisoners were counted “These modified estimates suggest that unemployment in the economically buoyant period of the mid-1990s was about 8 percent -- higher than any conventional U.S. unemployment rate since the recession of the early 1980s” (page 11).
The effect of prison on unemployment rates is, of course, not equally felt. Blacks and males (and black males in particular) contribute most to these figures. And, in the long term, those who are incarcerated have a lifetime expectation of being less likely to work. In a U.S. national longitudinal study it was found that “even after 15 years, [survey respondents] who were incarcerated as juveniles worked between 5 and 10 percentage points less than did their counterparts who did not experience incarceration” (page 12).
Conclusion. The data are quite clear. A good proportion of the difference in unemployment rates between the U.S. and other countries like Canada comes as a result of the imprisonment policy in the U.S. Prisons not only separate offenders from society, but they can also distort unemployment figures. In the long run, however, one of the (presumably) unintended effects of a high imprisonment policy for youth and adults is that a group of chronically unemployed people are created.
Reference. Beckett, Katherine and Bruce Western. The penal system as labor market institution: jobs and jails, 1980-95. Overcrowded Times, December 1997, Volume 8, No. 6.
When a community has a labour surplus -- a relatively high unemployment rate -- the frequency of imprisonment goes up. This is true even when other factors are controlled for.
The relationship between unemployment and crime is complex -- but the relationship between unemployment and the use of imprisonment turns out to be fairly straightforward. In this study, carried out in the U.S., it was found that counties that had high unemployment rates tended also to have high imprisonment rates.
This study chose the county as the unit of analysis because, as the authors put it, “punishment policies generally are made at federal and state levels, but punishment is dispensed normally by local prosecutors and judges” (p. 360). The authors point out that “increasing anxiety and resentment in the ranks of criminal justice may... contribute to harsher punishment during economic downturns and times of rising unemployment.... [Criminal justice officials] find new merit in the notion that severe penalties are needed to counteract the heightened temptations of illicit activity caused by hard times. The widespread beliefs that unemployment causes crime and severe punishment acts as a deterrent to crime underlie an increasing proportion of their decisions. Day-to-day they do what they can to increase the odds that crime does not become an alternative to economic hardship. The end result of their countless individual decisions is increased severity of punishment” (p. 359). Other studies have shown that “for individual defendants, judicial decisions to incarcerate vary significantly by employment status... Judges apparently view steady employment as an indicator of stability and unemployment as a sign of future trouble” (p. 359).
This study looked at 269 counties in the U.S. where adequate data could be obtained. It found that changes in imprisonment (commitments to prison) between 1980 and 1990 in these counties related to certain predictable factors: changes in the rate of violent crime and changes in the proportions of young males in the county. Other factors (e.g., income, racial composition, property crime rates) were also controlled for statistically. But over and above these factors, “change in unemployment is an independently significant predictor of change in imprisonment” (p. 364).
Conclusion. Communities where unemployment has risen appear to “adjust” their punishment practices so that levels of imprisonment also go up. The paper does not suggest that such a “policy” is explicitly understood by those implementing it. What is clear, however, is that above and beyond standard factors known to relate to crime, unemployment is an important factor in explaining high prison rates.
Reference: Hochstetler, Andrew L. and Neal Shover. Street crime, labor surplus, and criminal punishment, 1980-1990. Social Problems, August 1997, Vol. 44 (3), 358-367.
Watching television news makes people -- particularly middle-aged white women -- more afraid. This is most likely to be due to the “affinity of audience members with victims most often seen on television news.”
Context. For Canada and the U.S., the mid-to-late 1990s have been punctuated with moral panics about crime. Just as things seem to be calming down, an unusual or brutal crime will occur somewhere, and the cycle of crime-blame-punitiveness starts again. In the U.S. in 1993, the national “news” magazines (Time, Newsweek, etc.) all featured crime stories. By late that year television and newspaper news coverage of crime in the U.S. had increased by about 400% in a six month period and the number of Americans ranking crime or violence as the nation’s most important problem increased from 9% (in January 1993) to 49% (in January 1994) according to polls conducted by the Gallup organization.
This study, using data from a 1994 household survey in Tallahassee, Florida, operationalized fear as respondents’ estimates of the likelihood that they would be the victim of a robbery, assault, or burglary.
Findings. Fear in this study was related to the variables that have been found in other studies: fear was generally higher for older people, females, those who think that their own neighbourhoods (or the country, generally) have high crime rates and those who feel less safe in their own homes. For the purpose of this study, however, TV news watching (and radio news) were both significant predictors of fear above and beyond these other variables.
What is probably most important about this study is that the effects of media coverage of crime did not seem to be uniform: the effect appeared to be confined to particular groups of people: women, non-blacks, and those aged 30-54. White middle class women, without recent victimization experience, were particularly fearful if they watched a lot of TV news. It may be, therefore, that these women see the TV victims as being like themselves. They become afraid because the victims portrayed in the news often seem like them. A study of the gender and race of victims portrayed on TV news, carried out just after the present data were collected, suggested that white females were the victims most likely to be portrayed on TV news.
Conclusion. Watching TV news (or perhaps watching American TV news) leads to an increased perception of vulnerability to crime -- particularly for those most likely to be portrayed as victims on TV news. Interestingly enough, the measures of newspaper or magazine reading did not appear to have the same effect. When it comes to explaining fear of crime, look to television.
Reference. Chiricos, Ted, Sarah Eschholz, and Marc Gertz. Crime, news, and fear of crime: Toward an identification of audience effects. Social Problems, August 1997, Volume 44 (3), 342-357.
A detailed study over a ten year period of newspaper coverage of teenage “gangs” showed that coverage of teenage gangs and delinquency increased even though there was no statistical evidence for increasing rates of delinquency. For anyone reading those newspapers carefully, delinquency would mean gangs.
Context. The period from 1987 to 1996 was, for many parts of the U.S., a period when juvenile arrests went up considerably. However, in Hawaii, the increase was modest and, when status offences were excluded, there was, in fact, a decrease in juvenile crime. There was, then, “an absence of an “explosion” or “epidemic” of local juvenile crime” in the state (p. 98).
This study examined newspaper coverage of juvenile delinquency and juvenile gangs during this period. There was evidence of an explosion of coverage of these topics. In the second five years of the period studied (1992-6) there were almost twice as many stories about gangs as there were in the first period (1987-91) and over seven times as many stories focusing on juvenile delinquency. However, juvenile arrests (other than for status offences) were not increasing during this time and survey data (of young people) suggest that gang membership was not increasing.
Not surprisingly, a state-wide survey in 1997 showed that most people (92%) thought that juvenile arrests had increased in the previous few years. Most of these people thought that the increase was large.
Newspaper coverage of delinquency. The majority of newspaper articles were quite unremarkable. However, when the articles did have a point of view, they often used terms like “young killers” and included “advice” from “Los Angeles gang experts” that Hawaii “has all the makings for a youth gang crisis” (p. 106). The gangs were described by the police as “well organized in a paramilitary situation.” The media, then “set the stage for a war that had not yet occurred” (p. 107). At the other end of the spectrum delinquents were described as “blameless victims of circumstance who can be saved by particular, high profile programs.”
The difficulty, as the authors point out, is that “absent from these media accounts are the gritty complexities of life for youth in communities affected by gangs and violence.” Furthermore, the authors suggest that “naive assertions of the blamelessness of youth ultimately fuel public cynicism about constructive approaches to prevention and intervention efforts, while simultaneously permitting “get tough” politicians to characterize all programs as ineffective.” (p. 114).
The conclusion is simple: Crime in newspapers does not necessarily give a reasonable picture of what is happening. Changes over time in what is in newspapers does not necessarily reflect changes in crime. Crime and the coverage of crime are driven by different forces.
Reference: Perrone, Paul A. and Meda Chesney-Lind. Representations of gangs and delinquency: Wild in the streets? Social Justice, Volume 24 (4). Winter 1997.
Crime may have decreased in New York during William Bratton’s reign as chief of police. But it almost certainly did not happen because he endorsed a “zero tolerance” strategy toward minor crime and other irritants. For one thing, murders decreased in other cities (e.g., San Diego where murders decreased 41%) that had completely different approaches to policing. But more importantly, the idea that crime was reduced in New York through a “zero tolerance” approach simply does not fit the facts: there were far too many other things going on in New York to make it plausible that simple changes in police strategies made a difference.
Context. William Bratton was chief of police in New York City for a few years beginning in January 1994. This was good timing: murder rates peaked in or around 1993 in many cities and states in the U.S. and began to decline thereafter. The reductions were as dramatic as the increases in the late 1980s had been. Criminologists have debated, and still are trying to explain, both the 1980s increase and the 1990s decrease in violent crime. William Bratton must find such debates rather senseless. He takes full credit, in this short article, for the decline in crime in New York. As he states the case, “The murder rate has declined by over 50 per cent in New York City because we found a better way of policing” (p. 41). [The reader should understand that the “we” is a “royal we.”]
Most thoughtful analysts disagree with Bratton. One -- Charles Pollard, the Chief Constable of the Thames Valley Police notes that the rhetoric associated with New York’s “new” police style is “concentrated on aggression: on ruthlessness in dealing with low level criminality and disorderliness.... , of confrontational accountability systems.. and on the single-minded pursuit of short term results” (page 44).
These papers. The first of these papers is a simple clear statement by Bratton himself about why he takes personal credit for the crime reduction that occurred in New York. The second, by Chief Constable Pollard is a thoughtful -- and gentle -- critique of Bratton’s paper. He points out, for example, that “zero tolerance” and the “broken windows” theory of crime are not the same. The latter implies that “minor incivilities”, if unchecked and uncontrolled, produce an atmosphere in a community or on a street in which more serious crime will flourish.” Minor problems give a sign that disorder will be ignored. This is quite different from the “zero tolerance” notions so favoured by the right which imply “aggressive, uncompromising law enforcement.”
Pollard points out that “zero tolerance” law enforcement has the unintended effect of taking most of one’s police officers off the street and out of the communities since they will be spending all of their time processing minor criminals through the criminal justice system. Furthermore, it could well undermine the legitimacy of the police since it implies that all problems of order are police problems and should be dealt with harshly rather than sensitively. Finally, Pollard points out that Bratton’s own description of what went on in New York can be challenged on empirical grounds: he noted how easy it is for the data to be manipulated by police highly motivated to do so (page 52-3).
Conclusion. In responding to William Bratton’s self-serving description of policing in New York, Chief Constable Charles Pollard of the Thames Valley (England) Police points out that the “New York miracle” has to be examined carefully and when it is, it is found to be wanting. More important is his observation that “law enforcement on its own has only limited capacity to deal with crime, disorder and fear.” And when the limits of aggressive policing are met, there is nowhere to go: “The police will have lost touch with the community. Confidence will have drained away” (p. 54). Zero tolerance policing provides just one more example of how simple solutions rarely solve complex problems.
References. Bratton, William J. Crime is down in New York City: Blame the police. Pollard, Charles. Zero tolerance: Short term fix, long term liability? Both articles in Dennis, Norman (editor). Zero tolerance: Policing a free society. London, England: Institute of Economic Affairs Health and Welfare Unit, 1997.
Drug programs based on a “harm reduction” approach -- where policy is based on pragmatic rather than solely moral grounds and where the ultimate test is whether the world is a better place -- have been implemented in various countries in response to drug problems. Switzerland has begun a heroin prescription program for 800 seriously addicted persons. It appears to have a crime reduction effect.
Context. Prescribing prohibited drugs as a way of reducing the harmful side effects of criminalizing drugs is a controversial policy. Some of the more liberal European countries have, from time to time, had such programs. Switzerland has recently implemented such a program on the argument that it can be justified if the harmful effects of drug dependency can be reduced even though the dependency itself is not affected.
The program. Switzerland has methadone programs for about 13,000 heroin addicts. However, within this group, there are some who return to heroin anyway. This program targets the “worst” group -- those who have failed in other programs, have “obvious signs” of physical or psychological problems, and tend to have long drug and prison histories. Police suspect that these categories of people are responsible for a disproportionate amount of street crime, where the goal is to get money for drugs.
The goal of the program, then, is to reduce the “need” to acquire money illegally: heroin is administered in small specialized clinics and is injected on the spot.
The preliminary results are very encouraging: criminal activity (violent and non-violent property crime) by this group of people appears to have been reduced considerably at least in the first six months of the program. The participants in the program are, presumably, just as addicted as they were in the past. Their impact on the rest of society is, however, considerably less.
Conclusion. In the area of drugs, as elsewhere, the first policy question one must ask is simple: what is the goal of the policy? In this case, the goal was not, primarily, to save the souls (or, perhaps, even the bodies) of the severely addicted people. It was to reduce the harm to society and, perhaps, as well, to the addicts. Giving them heroin appears to be a successful strategy.
Reference: Killias, Martin and Juan Rabasa. Less crime in the cities through heroin prescription? Preliminary results from the evaluation of the Swiss heroin prescription projects. The Howard Journal, Volume 36 (4), November 1997, 424-429. For “local” coverage of this issue, see also the Globe and Mail (4 April 1998): Appleby, Timothy. “Netherlands swims against the drug tide.” (A1, A12) and a story about the Swiss experience -- “Crime and disease cut by project” (page A12 on the same day).
Estimates of dramatically increased crime (or cases coming to youth court) because of an increased number of adolescents in the population are nonsense. Careful modeling of the demographic changes show that the effect of demographic changes are trivial when compared to year-to-year changes that can occur in the rate at which young people are arrested for crimes.
Context. Obviously if there are more youth in a community, there will be more youth crime. The difficulty, however, is that without careful modeling of age-specific crime rates (and variation over time in these rates), statements which picture an awkward army of adolescents appearing over the horizon are best thought of as hypotheses rather than descriptions. One should question (but neither reject nor accept) headlines such as one that appeared in the Toronto Sun recently (16 March 1998, page 4) -- “Ontario’s attorney general warns: Youth crime wave looms.”
James Q. Wilson has suggested that “just beyond the horizon, there lurks a cloud that the winds will soon bring over us. The population will start getting younger again. By the end of this decade [the 1990s] there will be a million more people between the ages of 14 and 17.... This extra million will be half male. Six percent of them will become high rate, repeat offenders.... Get ready” (quoted on page 6). Wilson was obviously talking about the U.S. as a whole. In that context, the 30,000 “high rate repeat offenders” constitutes about 0.01% of the population which may not be enough even to notice in crime statistics. In California in 1996, young males (age 15-24) made up about 7% of the population. A 15% increase in their numbers will mean that they will constitute about 8% of the population (p.7).
This study looks carefully at one jurisdiction -- California. Studies on this topic should be jurisdiction specific since the two main pieces of data -- the age distribution of the community and the age-specific crime rates -- may vary across jurisdictions. Nevertheless, the lessons learned from one jurisdiction can be helpful for those elsewhere.
The findings are quite clear: as the author states, “Demography is not destiny” (p. 1). The estimate that one gets for crime in the future depends, not surprisingly, more on the crime rates of different cohorts of children than on the (rather small) variation in the proportion of the population that they constitute. Violence rates are highly age specific. Homicide arrest rates in California in 1981, for example were about 25 per 100,000 for the highest group (20-year-olds) and about half that for 40 year olds. In 1994, things had changed dramatically. The peak occurred earlier (age 18) and was much higher (about 48 per 100,000 18 year olds). But the rate for 40 year olds was much the same (or a bit lower) as it had been 13 years earlier. The problem is that different “birth cohorts” do vary in their (eventual) crime rates. There have been times when rates have increased as much as 3% over the previous birth cohort for years at a time. Over a long period of time, this amounts to a substantial increase. Decreases (from one birth cohort to the next) have also occurred.
Conclusion. What lies in store for a community (or in this case, California), “depends on the upbringing of young [people in the community] - today” (p. 2). “If, from this time on, every birth cohort is substantially more violent than the previous one, California [or any jurisdiction] is destined to experience a troubling wave of violence, commencing early in the next century.” There is room for optimism, however, since “One element in the equation that matters -- the future behaviour of today’s children -- is one that could be modified” (page 11).
Reference: Abrahamse, Allan. The coming wave of violence in California. Santa Monica, California: The Rand Corporation, 1997.
Contrary to the views expressed by the far right, public views suggest that rehabilitation should remain as an integral part of correctional policy.
Context. We are repeatedly told (in Canada as elsewhere) that people want a “tougher” criminal justice system and that prisoners should not be coddled. At the same time, however, there are numerous public opinion polls that suggest that rehabilitation should be valued within a correctional setting. Unfortunately, in Canada, we do not have much carefully conducted research on this issue, though we do know that people would rather spend money on alternatives to prison rather than prison construction (for both adults and for youth).
This study was carried out in Ohio, a state that does not have the reputation of being liberal on criminal justice matters. People were asked “general” questions about the relative weight that should be given to rehabilitation in prison (in contrast with “punishment” and “protecting” society). They were also given short “vignettes” -- that varied on a large number of dimensions (gender of accused, criminal record, drug use, employment history, current offence, sentence, and type of rehabilitation program). They were asked a number of questions about the vignette dealing, in effect, with whether they supported the use of rehabilitation with the offender.
The results are simple. People were more likely to list “rehabilitation” than other factors as what they thought should be the “main emphasis” in most prisons (41% listed “rehabilitation” first; 32% listed “protect society” as the “main emphasis” and 20% listed “punishment” as the “main emphasis”, with 7% indicating they were not sure). At the same time, when asked to indicate how important the various purposes were, it should be noted that protection and punishment were each listed as being very important (or important) goals of imprisonment by about 95% of the respondents. Rehabilitation was listed as being important or very important by fewer people -- about 83%. It appears that people are, in effect, saying that one must punish and protect -- these come naturally from being in prison -- but that rehabilitation is also very important and, therefore, needs to be the “main emphasis” of prisons.
The other more specific findings suggested that people valued rehabilitation more for juveniles than for adults and seemed generally supportive of rehabilitative efforts in prison and in the community. The respondents also generally supported the expansion of rehabilitative programs.
Conclusion. Those who suggest that the public is “fed up” with rehabilitation programs for offenders misrepresent the public view. As the authors suggest, although “the public desires punishment and... people want to be protected from predatory criminals, it appears... that the public still is receptive to treating offenders; the appeal of the rehabilitative ideal remains widespread” (page 253).
Reference: Applegate, Brandon K., Francis T. Cullen, and Bonnie S. Fisher. Public support for correctional treatment: The continuing appeal of the rehabilitative ideal. The Prison Journal, Volume 77 (3), September 1997. 237-258.