Wed, Apr

America’s Great Experiment With Jobs For The Underclass

Taxi Driver was one of a number of movies in the 1970s that presented a menacing underclass, unemployed and threatening social stability. FILMPUBLICITYARCHIVE/UNITED ARCHIVES VIA GETTY IMAGES


SUPPORTED WORK - This month marks the fiftieth anniversary of the National Supported Work Demonstration project, America’s experiment to end the underclass through jobs. Supported Work is little remembered today. But it was one of the largest employment demonstrations undertaken by the federal government up to that time, involving 10,000 workers. Its outcomes continue to be felt throughout the anti-poverty and job training fields.

Supported Work laid bare the difficulties in integrating the underclass into the social and economic mainstreams. The Supported Work participants were drawn from four groups identified in the 1970s as the “hard to employ”: women who had been on welfare for at least 30 of the previous 36 months, ex-offenders, former drug addicts and unemployed out-of-school youth. The participants were provided with paid work experience, an array of training and social services, and placement assistance into regular jobs.

Despite these supports, two thirds of the participants either did not complete the training/work experience phase or stay employed for more than a short time. The high failure-to-complete rates were linked more to behaviors than lack of skills. Skill deficits, especially limited reading and math skills, played a role. But more often the participants who did not complete were unable to meet workplace requirements or protocols or were not sufficiently motivated by the entry level jobs.

The Underclass Threat and the “Supported Work” Experiment 

In the years leading up to 1974, the welfare rolls and crime rates were rapidly rising throughout the country, especially in the largest cities. In September 1966, 439,394 persons were receiving Aid to Families with Dependent Children (AFDC), the main welfare benefit program, in New York City. By September 1972 that number had jumped to 913,087 persons. Crime rates, particularly violent crime rates, increased dramatically from 1965 to 1974. Popular culture at the time, in movies such as Taxi Driver and Fort Apache, The Bronx depicted a dangerous urban landscape, with a menacing underclass, unemployed and threatening social stability.

A blue ribbon committee of the time, chaired by Eli Ginzberg of Columbia University and Robert Solow of MIT sought to convey the gravity and urgency of addressing this underclass, as they wrote:

“The nation faces few problems as formidable as the presence of a group of people, largely concentrated in its principal cities, who live at the margin of society. Whether because of distortions in the economy, lack of training or motivation, or the attitudes of employers, these people have been excluded from the regular labor market and find, at most, sporadic employment. Though relatively few in number, they have become a considerable burden to themselves and the public—as long-term recipients of welfare and as the source of much violent crime and drug addiction.”

In early 1974, federal officials, foundations and university-based researchers, began planning on a large scale employment initiative that would test an “opportunity model” of integrating the underclass into the mainstream economy. The project, termed Supported Work, grew out of a pilot program by the Vera Institute, built around transitional jobs. The transitional jobs provided income for participants and modeled workplace behaviors needed for regular jobs. In general, participants were assigned to small work crews (10 workers or fewer), with a supervisor who served as foreman and counselor. The worksites included residential construction and rehabilitating homes in disrepair, furniture manufacturing, park maintenance and street maintenance. Participants would be employed in these transitional jobs for up to 12 months and then move into jobs in the mainstream economy.

Forty non-profits throughout the country applied to be sites for the program, and by late 1974, fifteen sites were chosen.The first Supported Work participants enrolled in March and July 1975. In addition to the transitional jobs, participants were provided with weekly peer support workshops, as well assistance with transportation, emergency food or housing. Upon completion of the transitional jobs, participants were assisted in landing a job outside of the program.

Supported Work was distinguished not only by its scope and ambition but its careful research effort. A national management organization, the Manpower Demonstration Research Corporation (MDRC) was formed to oversee operations and the extensive research effort that accompanied the program. Supported Work was the first major job training effort that utilized the technique of random assignment. Workers were recruited from among the four targeted groups, and randomly assigned either to program participation or a control group that did not receive program services. The employment and earnings of the two groups were tracked for three years after enrollment. Total costs of Supported Work reached $82.4 million, of which $11 million represented research costs.

Supported Work operated from 1975 through 1979. During this time, around 10,000 workers were enrolled, split between the participant group and control group. Less than 30% of the enrollees had graduated from high school. Outside of the welfare group, the arrest rates for the other three groups (out of school youth, ex-addicts, and ex-offenders) ranged from 54% to 100%.

The Majority of Supported Work Participants Do Not Complete

Only around a third of the 10,000 enrollees completed the program and went on to unsubsidized employment or to additional education/training. Further, the employment and income gains of the participants were not far above those of the control groups.

Among the four “hard to employ” populations that were targeted, the welfare recipients and ex-addicts achieved the better outcomes in employment rates compared to the control groups—though the differences were not dramatic. The participant ex-offenders and out of school youth showed no significant gains in employment or crime reduction over the control groups.

A young New Yorker writer, Ken Auletta, spent time at one of the project sites, the Wildcat Service Corporation in mid-town Manhattan. He attended the life skills and counseling classes with the 26 participants and shadowed them at work sites, and chronicled the participant work journeys in his 1982 book, The Underclass. The results at Wildcat were similar to those nationwide: six of the 26 trainees drop out in the first few weeks and others drop out over time, so that only about a third complete and go on to additional training or jobs—as private security guards, clerk typists, offices aides, cashiers.

Auletta recognized that observers would see the results in different ways. He would write in The Underclass:

“Some readers may think that the MDRC’s five-year national supported work experiment was a failure because two thirds of the 10,000 enrollees did not complete the program and go on to unsubsidized employment jobs. Other readers will think the program a success because this group is so difficult to reach yet one third managed to enter the world of work.” 

The Supported Work advisory committee, focusing on the one-third of participants who entered the world of work recommended that Supported Work be extended and even expanded in size. It was not. By late 1979, the program ended.

Supported Work Ends But Leaves Several Legacies for Employment Programs Today

Though job training practitioners today have moved away from the term “hard to employ”, they still focus on the targeted populations of Supported Work (as well as the homeless and other workers with greatest difficulties in finding and retaining jobs). And programs today have been and continue to be shaped by several legacies of Supported Work. Chief among these legacies are the following four:

Transforming the moribund welfare offices: In showing the limits of an opportunity model of integrating the “hard to employ”, Supported Work set the stage for the welfare reforms by states in the 1980s that led to the landmark federal 1996 welfare reform under President Clinton. Welfare reform is often thought of in terms of its addition of work requirements to the welfare system. Less recognized is the many work supports for welfare recipients that were added by welfare reform, rooted in part in Supported Work. Additionally, welfare reform transformed the moribund welfare offices of the 1970, which served mainly to disperse benefits, into active, employment-focused centers, recognizing the strengths that welfare recipients brought to the job market and society. NYU Professor Lawrence Mead, one of the architects of welfare reform, would go on to expand the approach of mixing work requirements with work supports/opportunities to low income men in his 2011 book, Expanding Work Programs for Poor Men.

Growing the use of transitional jobs and “employment social enterprises”:Supported Work significantly expanded the use of transitional jobs, from the pilot projects of the Vera Institute. In the years since Supported Work, the use of transitional jobs has grown and been sharpened by groups such as the Center for Employment Opportunities (transitional jobs for ex-offenders) and by a series of pilot projects with community groups and local governments under the federally funded Department of Labor Enhanced Transitional Jobs Demonstration.

Additionally, a sub-industry of Employment Social Enterprises (ESEs) has emerged that provides transitional jobs in businesses, connected to non-profits, that are revenue-generating and use this revenue to offset the transitional jobs costs. REDF, the nation’s leading investor and expert in these ESEs, has invested in more than 280 ESEs since 1997, employing more than 87,000 participants from the “hard-to-employ” groups and generating $1.6 billion in revenue. Today more than 300 ESEs are operating in California alone. 

Revealing how workers in the “hard to employ” groups, who were not program participants, were quietly finding jobs on their own: While Supported Work suffered from high drop-out rates, it also showed how much other workers in the “hard to employ” populations, the control group members, were finding jobs on their own. The small difference in employment and earnings between the Supported Work participants and the control group participants was due in part to the ability of the control group participants to find employment without Supported Work. In this regard, Supported Work, like welfare reform, would demonstrate the savvy and strengths among the “hard to employ”: a theme that leading job training groups today—America Works, Goodwill, Hopeworks—have picked up.

Establishing MDRC and serious research protocols for improving the employment of the “hard to employ”: MDRC, the research arm established in Supported Work, would not end in 1979. Instead, it would go on to become what it is today: America’s leading applied research center on workforce and job training efforts. Its rigorous use of control groups set it apart from nearly all other think tanks and policy groups. It may be the most influential legacy of Supported Work.


I was new in the job training world in 1980, when the Supported Work report was issued, and can recall some of the discussions the report generated regarding the possibilities of behavioral change, the values and limits of counseling, the interaction between work behaviors and the jobs available to the hard to employ. These discussions are similar to ones we have today. I still keep in my office the four volumes of the final Supported Work report and reference it from time to time. Its concerns regarding the “hard to employ” remain our concerns fifty years later.

(Michael Bernick served as Director of the state labor department, the Employment Development Department, following eight years as a Board member of the BART transit system. He currently is an employment counsel with the international law firm of Duane Morris LLP, Milken Institute Fellow and Fellow with Burning Glass Institute. His newest book is The Autism Full Employment Act.”)


Get The News In Your Email Inbox Mondays & Thursdays