Kathy Beckett, Indiana University
The low rate of US. unemployment contrasts strikingly with very high levels of joblessness in Europe. Official statistics show that U.S. unemployment is now around 4% while European unemployment is over 10% and has been persistently high throughout the 1980s and 1990s. Policymakers and analysts often attribute low unemployment in the United States to a highly unregulated labor market in which there is very little state intervention. In Europe, on the other hand, governments negotiate with unions and employers in setting wages, and welfare states provide generous benefits for unemployment, training, health care and education. Without these institutional protections, so the theory goes, wages are allowed to fall to a level at which everyone can find work and indeed everyone must look for work because government assistance is so meager. In Europe, wages are rigid and unemployed workers don't have incentives to seek work because of the generous welfare benefits at their disposal. In short, the American model of market deregulation has apparently generated low unemployment and the European case stands as the cautionary counter-example.
The estimates of unemployment upon which these arguments rest do not include the prison and jail population, which in the U.S. is now estimated at more than 2 million. Our research suggests that when this population (comprised mostly of young, able-bodied men) is included in estimates of unemployment and joblessness, European rates of unemployment are actually lower than those in the United States for eighteen of the twenty years between 1975-1995. These adjusted figures also show that unemployment rates among African-American men have not improved much since the recession of the early eighties. Thus it appears that apparently low rates of unemployment in the United States are in part due to the dramatic expansion of its penal system.
Industrial Relations and the Welfare State
It is often observed that the United States lags far behind Western Europe in industrial relations policy and welfare state development. The comparative weakness of U.S. industrial relations is illustrated by statistics on collective bargaining coverage. While the wages of most European workers are set by some kind of collective agreement, fewer than twenty percent of American workers have union representation. Social policy is also less developed in the United States. Approximately one-quarter of the gross domestic product (GDP) of the large European countries is devoted to public spending on health, education, and welfare. In the United States, social spending accounts for only 15% of GDP. In areas that are related specifically to employment, only about one-third of U.S. workers are covered by unemployment benefits in contrast to nearly all European workers. The United States also spends less than a quarter of the European average on active labor market policies that train and mobilize workers into jobs. In short, traditional indicators of government intervention show the U.S. government only weakly regulates its labor market in comparison to the extensive state institutions of Europe.
Penal Institutions and the Labor Market
Although welfare and industrial relations statistics do illustrate the weakness of social protection mechanisms in the United States, this does not justify the claim that market principles alone drive the superior U.S. employment record. Labor markets are embedded in and affected by a wide array of social arrangements that extend beyond the regulative mechanisms of the welfare state. In the United States, for example, market deregulation and welfare state retrenchment have been accompanied by rapid expansion of the criminal justice system. At the peak of the recession of the early 1990's, criminal justice spending exceeded $91 billion, dwarfing the $41 billion spent on all unemployment benefits and employment-related services. By 1992, the public cost of correctional facilities totaled more than $31 billion and by 1996, 1.63 million people were being detained in American prisons and jails (Gilliard and Beck, 1997). This was a significant and costly state intervention comparable in size to the large social programs of European welfare states. Indeed, due to low levels of unemployment insurance coverage, more American men were incarcerated than received unemployment benefits in 1995.
U.S. incarceration rates are even more striking when compared to those of other industrialized democracies. In 1993, the U.S. incarceration rate was 5 to 10 times greater than other DECD (Organization for Economic Cooperation and Development) countries. These high rates correspond to large absolute numbers. In the United States, prison and jail inmates are counted in the millions. In European countries, prison populations number in the tens of thousands.
The Short-term Effect on Incarceration on Unemployment
By moving more than one million able-bodied men of working age into prisons and jails, U.S. criminal justice policy has had profound effects on estimates of employment trends. Because they are institutionalized, prisoners are not counted by population surveys as members of the civilian labor force, or even among those "not in the labor force." In the short-term, then, incarceration lowers conventional unemployment measures by removing significant numbers of able-bodied men from estimates of joblessness. These labor market statistics therefore give a falsely optimistic picture of labor utilization when incarceration rates are high. To remedy this limitation, we analyze unemployment trends that take account of the size of the incarcerated population. This may even be a conservative strategy for assessing the short-term effect of incarceration on employment because it ignores the job creating effect of the penal system.
To assess the short-term effect of incarceration we distinguish two unemployment statistics. First, the usual unemployment rate is just the number of unemployed as a percentage of the total labor force--the unemployed plus those employed. Second, we can account for the incarcerated population by simply adding them to the total jobless count and to the figure for the total labor force. This incarceration-adjusted unemployment rate can be interpreted as either a more general measure of joblessness than the conventional unemployment rate, or a measure of the causal effect of incarceration. In this causal interpretation, adjusted unemployment--which includes prison and jail inmates among the jobless--describes what the unemployment rate would be if the incarceration rate were zero. The importance of incarceration as a source of hidden unemployment varies by sex and across countries. More than 90% of prison and jail inmates are male in the United States, so we focus on trends in the labor market conditions of men. From a comparative perspective, the short-term effect of incarceration is tiny in Europe because incarceration rates are so low. In most European countries, unemployed males outnumber male prison inmates by between 10 and 20 to 1. In the United States in 1995 this ratio has fallen to just under 2.2 to 1. Differences between the United States and Europe are also reflected in the relative size of the conventional unemployment rate and the adjusted figure that includes the incarcerated. In most European countries, counting prisoners in estimates of unemployment only changes the unemployment rate by a few tenths of a percentage point. By contrast, prison and jail inmates in the United States added 1.5 points to the usual unemployment rate in 1990 and over 2 points by 1994.
Conventional estimates of U.S. and average European unemployment suggest that unemployment in the United States peaked in 1983 at about 10%, rose again in the early nineties, but recovered fairly quickly after each of these recessions. Although European unemployment rates were low compared to the United States until 1984, estimates of unemployment suggest that recovery from the recessions of the mid-eighties and early nineties in Europe was relatively weak.
However, employment performance in the United States looks less impressive once we take the size of the prison and jail population into account. An adjusted estimate that adds prisoners to the male unemployment count shows that labor market inactivity in the United States never fell below about 7% in the 1980's. By 1994, the prison and jail population had become so large that it added about 2 percentage points to the male unemployment rate. These modified estimates suggest that unemployment in the economically buoyant period of the mid-1990's was about 8%, higher than any conventional U.S. unemployment rate since the recession of the early 1980's.
A more detailed examination of the U.S. data allows us to specify the impact of incarceration on the labor market experiences of black and white men. In 1983, when the prison population is added to the unemployment count, the resulting unemployment rate for all men is just 1 percentage point higher. However, estimates of unemployment among black men in 1983 increase by 4 points to 23%. The effect of incarceration on white male unemployment is smaller, raising the unemployment rate by only about half a percentage point. As the prison population grew through the 1980's, the labor market effects of incarceration become much larger. For all men, average unemployment in the 1990's is lifted to nearly 8%. When we include the incarcerated population in estimates of black unemployment we find that nearly 1 in 5 African-American men were without a job throughout the 1990's. Incarceration has a similar effect on estimates of black joblessness, a category that includes those no longer looking for work.
In sum, the growth of U.S. incarceration through the 1980's and 1990's conceals a high rate of persistent unemployment and joblessness. Adjusted unemployment figures that include the incarcerated population suggest the United States labor market has performed worse, not better, than European labor markets for much of the past two decades. Incarceration has particularly strong effects on estimates of black unemployment: when prisoners are added to jobless statistics, rates of joblessness among black men have remained around 40%. (It should be noted that only approximately 5% of prisoners in the U.S. worked in 1996 producing goods or services for external consumption.)
The Long-Term Effect of Incarceration on Unemployment
Thus it appears that by removing large numbers of men from the labor force count, incarceration artificially lowers the usual figures for labor inactivity. At the same time, it is likely that the expansion of prisons and jails will increase unemployment in the long run. Research suggests that job applicants with no criminal record are far better off than demographically similar persons who were convicted and incarcerated. While convicts who acquire educational and vocational skills in prison are able to improve their chances of employment (Irwin and Austin 1994), resources for education and vocational training in prisons and jails have declined in both absolute and relative terms, and the recent decision to deny prisoners Pell grants to pursue higher education suggests that this trend is likely to continue in the future.
Our analysis of data from the National Longitudinal Study of Youth (NLSY) supports the argument that incarceration increases the likelihood of future joblessness. The results of this analysis indicate that youth incarceration reduces annual employment by about 5 percentage points, or about 3 weeks per year, controlling for education, work experience and local labor market conditions. The effect is larger for blacks, whose employment is reduced by about 8 percentage points (more than 4 weeks in the year) by juvenile incarceration. In fact, the effects of youth incarceration on adult employment are larger than failure to graduate from high school or living in a high unemployment area. Even after fifteen years, respondents who were incarcerated as juveniles worked between 5 and 10 percentage points less than their counterparts who did not experience incarceration. The effects of adult incarceration on employment status are even greater, reducing employment by about one fifth or about 10 weeks per year. A wide variety of models thus strongly supported the conclusion that incarceration has large and extremely long-lasting effects on the job prospects of ex-convicts.
Comparative labor market research and recent policy debates attribute low levels of unemployment in the United States to an ostensibly de-regulated labor market. In contrast, our research suggests that the U.S. has made a significant intervention in the labor market by expanding the penal system in the 1980's and 1990's. As a result of the policies associated with the wars on crime and drugs, prisons and jails grew to detain around 1.6 million people by 1995. Consisting mostly of young, unskilled, able-bodied men of working age, these large prison and jail populations conceal a high level of joblessness that, if included in labor market statistics, would contribute about 2 percentage points to the male unemployment rate by the mid 1990's. These effects are especially strong for African-Americans: labor inactivity is understated by about two-thirds, or 7 percentage points, by the conventional measure of black male unemployment. Despite claims of "Eurosclerosis" and the successful deregulation of the U.S. labor market, our revised estimates show that unemployment in the United States exceeded average European rates for all years between 1975 and 1993.
While incarceration has the immediate effect of lowering conventional estimates of joblessness and unemployment, it significantly increases the chances of unemployment among ex-convicts. With over two million men currently in prison or jail, current levels of incarceration annually generate the equivalent of a full year of unemployment for more than 400,000 American men. In the aggregate, then, it appears that the high U. S. incarceration rate will greatly reduce the productivity and employment of the male workforce.
How can these findings be reconciled? If incarceration lowers conventional measures of joblessness in the short-term but increases unemployment in the long-term, why does the U.S. labor market still perform well according to conventional indicators? The steady expansion of the prison and jail population combined with high rates of recidivism and re-incarceration helps to explain this paradox. About two-thirds of young state prisoners are re-arrested within three years, removing many of those at risk of unemployment from the labor force. With high rates of recidivism and intensified surveillance of ex-convicts, the short-term negative effect of incarceration on unemployment dominates the long-term positive effect. Under these conditions, the appearance of strong employment performance has been assisted by an ever increasing correctional population.
It has been argued that some European welfare states may also conceal unemployment. However, the dangers of U.S. prison expansion are significantly greater than those posed by European welfare policies. In contrast to welfare institutions, the penal system has unambiguously negative effects on the job prospects of its clients. While many job training programs and employment related services expand human capital and strengthen social networks, incarceration devastates the market power and productive capacity of potential workers. Moreover, penal expansion exacerbates rather than alleviates racial and class inequalities. In sum, the massive expansion of the penal system is a uniquely American mode of state intervention that improves conventional indicators of labor market performance in the short-term--but will exact a high social cost in the long run. While some policy analysts celebrate the free market principles of the U.S. model, these same principles should be assessed in light of the significant and coercive reallocation of labor through the expansion of American prisons and jails.
Bureau of Justice Statistics (BJS).1995. Census of Federal and State Correctional Facilities. Washington D.C.: Department of Justice.
Gilliard, Darrell K. and Allen J. Beck. 1997. "Prison and Jail Inmates at Midyear 1996." Bureau of Justice Statistics Bulletin.
Irwin, John and James Austin. 1994. Its About Time: America's Imprisonment Binge. Belmont, Ca.: Wadsworth.
Mauer, Marc. 1994. Americans Behind Bars: The International Use of Incarceration. Washington, DC: Sentencing Project.
This is an abbreviated version of a paper published by Western and Beckett called, "How Unregulated is the U.S. Labor Market? The Penal system as a Labor Market Institution." American Journal of Sociology 104:1030-60 (1999).
[Editor's Note: This article outlines the impact of mass imprisonment on the US. labor market. However, the paper actually understates the impact of mass imprisonment because it only focuses on half the equation, the prisoners who are locked up. Also impacting the job market are the people who are hired to guard and contain the prisoners. Recent Bureau of Justice statistics indicate that at least 600,000 people are directly employed by prison systems. If it were not for mass imprisonment this is a large work force that would either have to be employed elsewhere or would find itself unemployed. If people indirectly employed by prison systems are included, the impact is even higher. This includes the construction workers who build prisons, medical contractors, prison bureaucrats, etc. Staffing levels in prison systems vary, but Washington state has roughly 15,000 prisoners and its Department of Corrections employs about 7,400 people. This results in hiring one prison employee for every two prisoners received into custody. If the authors' analysis had encompassed people employed by prisons and jails, the labor market impact of mass imprisonment would
be even higher. Not a part of the study, but also relevant, is the fact that the US. maintains a large military of some 1.4 million people. Most European countries have small armies and small prison populations. If the U.S. had a military proportionate to that of other countries its unemployment figures would also be higher. The subject of another article, but there is a direct correlation between global capitalism, mass imprisonment and a large military force.]
As a digital subscriber to Prison Legal News, you can access full text and downloads for this and other premium content.
Already a subscriber? Login