-- About CFEPS --
-- Contact Us --








Request a copy of this publication
Historicizing Government Work Programs
A Spectrum from Workfare to Fair Work
Seminar Paper No. 2
Mar 2000
Nancy E. Rose

Note: This paper is still a draft. My presentation will focus on the beginning and concluding sections of the paper -- conceptualizing the spectrum of government work programs and developing an agenda in support of fair work programs -- and will refer to the historical middle sections as part of this broader discussion.

A critical component of welfare policy since the early 1980s has been workfare, programs that require welfare recipients to work in jobs that would normally pay wages in order to remain eligible for welfare payments. Workfare and other "welfare-to-work" programs received a significant boost from the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), the highly touted reform that ended "welfare as we knew it." Although workfare has been ubiquitous for the past two decades, the United States has a history of another type of government work program as well. Fair work, or job creation programs, have been implemented periodically since at least the early 1800s to provide jobs for people normally seen as employed. The best known of these programs are the Works Progress Administration (WPA) in the 1930s and Public Service Employment (PSE) which was part of the Comprehensive Employment and Training Administration (CETA) in the 1970s.

In this paper I investigate the range of government work programs. It includes the following discussions: the differences between workfare and fair work; brief histories of both types of programs; how work programs have been implemented since 1996; and developing an agenda for fair work and related programs, including responses to time-worn criticisms that inevitably emerge when public sector job creation is raised as a possible policy.

The Differences between Workfare and Fair Work

Historically, there have been fundamental differences between workfare, also known as worktests, and fair work programs, generally termed job creation or public employment programs. Workfare, on one hand, and fair work, on the other, can be conceptualized as two ends of a spectrum; increasingly, programs have fallen between these two extremes. With this caveat in mind, the differences between workfare and fair work can be described in the following manner.

Workfare is shameful and stigmatized, mandatory programs for the "undeserving" poor to make them prove that they are not "shirking work," and to therefore end "welfare dependency"; fair work encompasses voluntary programs for the "deserving" poor who become unemployed due to recessions, depressions, automation, and natural disasters, i.e., "through no fault of their own." Workfare ignores the value of caretaking work in the home and requires relief recipients to work for wages or to "work off" their payments; fair work creates jobs as well as education and training for the unemployed, enticing participation by offering relatively decent wages and developing often innovative projects. Workfare helps employers by providing downward pressure on wages, both by directly channeling the poor into low-wage labor markets and by serving as a deterrent so that people "choose" low-wage jobs instead of welfare; fair work programs expose the inability of capitalist economic systems to normally create jobs for everyone who wants one. Workfare involves placements in the public, non-profit, and private sectors; fair work creates jobs in the public and non-profit sectors. And, although through the 1930s workfare targeted both women and men, since that time the main focus of the programs, and of the debates about the programs, has been women, primarily women of color; fair work programs have predominantly been aimed at men, primarily white, male, heads-of-household.

As a result of these differences, workfare has been politically popular, supporting the view that the poor are "lazy" and will work only if forced to do so, while fair work programs have evoked bitter criticisms for interfering with labor markets and with the logic of production-for-profit itself. This has sometimes led to regulations making fair work more like workfare and less like "real jobs," i.e. wage-paying jobs obtained through normal labor market channels. There are three general categories of these policies: determination of eligibility; methods of setting pay rates; and regulations regarding allowable work.

The first set of policies concerns eligibility for the programs. Fewer requirements make the programs more like "real jobs." The most basic criteria is simply a personís status as unemployed or underemployed. Yet in most programs additional requirements have been adopted as well. These have ranged from general criteria, most commonly the unemployment rate in the region or state, to individual criteria, e.g., a personís poverty or welfare status. Importantly, basing eligibility on a "means test," i.e. proving sufficient need to qualify, marks participants as being on relief and reinforces the "ideology of the dole." Additionally, preference has often been accorded based on race, gender, class, marital status, and veteran status. White, male, heads-of-household have historically been seen as most deserving of positions in job creation programs, and special consideration has often been accorded to veterans as well.

A second type of involves determination of payments. When payments are based on market wages for similar work, the programs more closely resemble real jobs. On the other hand, basing payments on the amount of income needed marks the work as relief. Often, total payments are determined by need; hours of work are then figured by dividing this amount by an hourly rate related to the minimum wage or to market wage rates. Further, concerns about keeping payments on work programs below the amount that could be earned through wage-labor have led to regulations limiting the hours of work so that even when pay rates have been based on wages, total payments would be less than the amount that could be earned from market jobs. In other instances, there have been outright limits on the amount people could earn in the programs.

A third set of regulations involves the types of work allowed in the programs. There exists a longstanding tension between the belief that the poor should produce things of value in order to compensate society for providing relief, and sometimes to directly contribute to their upkeep, on one hand, and fears that production of useful goods and services would interfere with the logic of capitalist production-for-profit, on the other. As a result, fair work programs have generally been so constrained by often contradictory rules and regulations that they have easily been seen as failures. It is important to be aware of these contradictions, both because they inform our historical analysis of fair work programs and because they underlie the time-worn criticisms of these programs; indeed, the WPA and CETA continue to be seen as inefficient and unnecessary "make-work" programs, captured during the 1930s in the epithet, "boondoggle."

These contradictions followed from mandates to create as many jobs as possible with allocated funds, but to do so in ways that did not replace "normal government operations" or compete with private enterprise. It led to three sets of problems. Requirements to put as many people to work as possible with the available funds -- i.e., to make work -- meant that the work was often inefficient by industry standards. A notable example concerns the "pick and shovel" projects in the 1930s, as many more men were put to work (women were not considered suitable for construction projects) building and repairing roads than would have been the case if grading and paving machines were routinely used. Directives to refrain from "normal government operations" meant that projects were easily construed as unnecessary -- otherwise, it was reasoned, the government would do them. And restrictions on competition with the private sector mitigated against the production of useful goods.

The reality of fair work programs is often different from the widely-accepted assessments. Contrary to claims of interference in "normal market operations," when useful goods -- primarily food and clothing -- were produced, they were kept out of the market and distributed only to others on relief or used in public institutions such as hospitals. Condemnation of the WPA for unnecessary work such as "leaf-raking" ignores the fact that maintenance projects were prohibited from April 1934 onward. Yet maintenance is necessary work, albeit work that must be done repeatedly. Indeed, it is similar to housework -- underappreciated and only noticed when it is not done. The criticism that has validity concerns inefficiency; however, this was virtually assured by the mandate to create work for as many people as possible. Further, it is telling that while criticisms of make-work and inefficiency have commonly been made toward fair work programs, this has not been the case with workfare -- even though workfare is often clearly inefficient and make-work.

Fears of interference with both the labor market and with the logic of production-for-profit also led to overall restrictions on fair work programs. They have always been temporary, often termed emergency programs and requiring periodic reauthorization. Indeed, attempts to enact a permanent fair work program have never been successful. Additionally, the overall size of the programs has been constrained. Even during the massive programs of the 1930s, no more than one-third of the unemployed were put to work at any one time. Viewed from another perspective, at the height of the 1930sí programs, 4.4 million people were on the work programs, thereby demonstrating that the government has the capacity to create large-scale fair work programs.

A Brief History of Fair Work Programs in the U.S.

Although they have largely been ignored during the past two decades under the deluge of workfare, fair work programs are part of United Statesí history. Since at least the early 1800s, they were established by some cities during the recessions and depressions that periodically ravaged the country. General agreement among relief officials that work was preferable to "charity," especially for white, male, heads-of-household, in conjunction with protest by unemployed workers, led municipalities to increase their expenditures on public works. The work involved normal government activities -- construction (e.g., of culverts, sewers, and schoolhouses) and road repair were most common, and was supplemented by other maintenance and repair work (e.g., whitewashing tenements, laying water mains, improving parks, repairing public buildings, and clearing trees from land about to be flooded by dams, and building culverts, sewers, and schoolhouses). In addition, some "labor-intensive" workrooms, in which garments and rag rugs were sewn, and cloth was spun were also established for women and for men who were unable to labor outdoors. Several characteristics of these early programs place them midway on the spectrum between workfare and fair work. Eligibility was determined by a means-test, with preference given to family heads normally seen as employed. And payments were usually relief, not based on market wages; in fact, the monies were sometimes obtained from private donors. [The following historical discussions draw on Rose 1994 and Rose 1995.]

These types of work programs continued through the early 1930s. As the Great Depression led to measured unemployment rates of 25 percent of the total labor force, and 38 percent of the industrial labor force, protests by the unemployed demanding jobs and relief intensified, and local and state governments in many areas of the country developed work programs. However, the continuing economic collapse led to reduced tax revenues, and these early Depression work programs, along with most other government functions, were hampered by lack of funds.

By the time Franklin D. Roosevelt became President in March 1933, the need for federal relief was apparent. Three major work programs were developed as part of the New Deal. In May 1933, the Federal Emergency Relief Administration (FERA) became the first federal unemployment relief program, providing aid to 1.4 to 2.4 million people each month. It was interrupted from November 1933 through March 1934 by the Civil Works Administration (CWA), quickly set up to put another two million people to work and thereby head off another depression winter of increased need and increased protest. In March 1934 the FERA was resumed as the Emergency Work Relief Program (EWRP) for the unemployed in urban areas and the Rural Rehabilitation Program for the rural jobless. In the fall of 1935 the FERA was replaced by the Works Progress Administration (WPA), which provided work for up to 3.3 million people each month. It decreased during the early 1940s, ending in June 1943, as mobilization for World War II substantially increased production and employment and finally brought the depression to an end. In addition, the Public Works Administration (PWA) was set up to provide construction jobs, all contracted out through market channels, and the Civilian Conservation Corps (CCC) was developed to get young unemployed men out of the cities and put them to work in federal forests. While the PWA and CCC provoked little criticism, the same was not true for the FERA, CWA, and WPA. Indeed, they were targets of a seemingly endless barrage of complaints that sometimes led to significant changes in program policy. Most contentious were policies regarding determination of payments and the types of work done.

Several characteristics marked the FERA and WPA as midway between workfare and fair work. Most importantly, eligibility was based on a "means test," and payments were based, at least in part, on the amount that the family needed. Reflecting class bias, this was determined by local relief departments for "blue-collar" workers, but usually by organizations of their peers for "white-collar" workers. Hours of work were determined by dividing the total amount needed by an hourly rate based on market rates for similar work. [The minimum wage was not enacted until 1938.] Yet there was a brief, but significant, exception. Worries that extremely low agricultural wage rates would do little to increase the "purchasing power of the masses" and help spur recovery, led relief officials, in July, 1933, to set a minimum rate of 30 cents per hour. This provoked a barrage of complaints -- from employers and others claiming that it caused shortages of agricultural and domestic service workers -- until it was ended in November 1934.

The CWA embodied some fundamental differences compared to the FERA and the WPA. Indeed, of all government work programs thus far, it most closely adhered to principles of fair work. Wanting to counteract another Depression winter of increased despair and increased protest, relief administrators developed regulations to make the work similar to "real jobs." This was accomplished by replacing FERA regulations with those of the PWA -- most importantly, the "means test" was suspended and payments were based on local rates for similar work. Workers clamoured for the jobs, and in some areas relief administrators held lotteries to choose the participants. Not surprisingly, the CWA was bitterly attacked by the business community for being too large and too expensive, and paying wages that were too high. These strident criticisms had an effect: the CWA was substantially cut two months after it began and was ended entirely in another two months.

Criticisms were also leveled at some of the projects developed in the work programs. There were three categories of projects: construction, service, and production-for-use. Construction projects were the most common, absorbing approximately ĺ of the program funds. Continuing along the lines of the earlier programs established during recessions and depressions, they provided work for men. The results are indeed impressive. They include construction and repair of: more than one million miles of roads; approximately 200,000 buildings (e.g. schools, libraries, courthouses, firehouses, armories); tens of thousands of recreational facilities (e.g. parks, stadiums, swimming and wading pools); several thousand public utility plants; miles of sewers, drainage ditches, and electric power lines; thousands of flood and erosion control projects (e.g. firebreaks, retaining walls, levees); and several hundred airports. Yet they were routinely attacked as inefficient -- more people were put to work than would have been the case with more capital-intensive production methods commonly used in the private sector.

Service projects provided another important source of work. Commonly known as "white-collar" projects, since they developed work for white-collar workers considered unsuitable for the blue-collar construction projects, they increased in importance throughout the1930s. Projects were developed to reopen schools that had been closed due to lack of tax revenues, and to provide work for the unemployed and services for the public in public health, recreation, libraries, museums, conducting research and surveys, and the arts. In fact, the WPA Federal Arts project for artists, writers, musicians, and actors and actresses is still one of the best known of all the 1930s' projects, with the posters, paintings, murals, guidebooks, and plays remaining as examples of federal support of the arts. These were the main projects criticized as unnecessary.

The third type of projects involved production of consumer goods. Commonly known as production-for-use, or production-for-use and distribution, all goods were kept out of normal market channels and given only to people on relief or used in public facilities. Initially, production-for-use primarily continued the "labor-intensive" sewing rooms that had also been set up during early recessions and depressions, as well as woodyards and canning projects for garden produce. These provoked little criticism. However, the same was not true for the new production-for-use projects. Responding to criticisms of early work projects as "boondoggles," production-for-use was expanded as the CWA was replaced by the EWRP in the spring, 1934. Clearly useful goods were produced: aid was provided aid to "self-help cooperatives" that were set up by the unemployed during the Hoover years of the depression; in the mattress-making project women sewed mattresses and comforters by hand using some of the surplus cotton that had been taken off the market by the Agricultural Adjustment Administration; centers were established to process (slaughter and preserve) animals dying from the severe drought, and to produce and repair goods, primarily shoes and wearing apparel, from the hides, pelts, and wool; and projects were started in which relief recipients were put to work in factories that had been partially or entirely shut down and rented from their owners by state or local relief administrations. The latter was the most contentious. Commonly known as the Ohio Plan since it was most extensive in that state, relief workers produced clothing (e.g. long underwear, coats, knitted goods), furniture (primarily wooden beds and kitchen-type chairs), metal ware (stoves, heaters, and skillets), china, and blankets. In fact, the Ohio Plan most vividly revealed the contradictions of production-for-profit, as the government was combining idle workers with idle factories to produce consumer goods since capitalists were failing to do so.

The new production-for-use projects quickly came under fire. Capitalists complained that the projects were inefficient and, most importantly, that they competed unfairly with private enterprise. Production-for-use was condemned in the Report from a December 1934 conference of 90 business leaders, as they indignantly declared that "(g)overnment competition with private business leads toward socialism." This was simply their worst fear--that production based on capitalist's profits would be replaced by production based on people's needs. As a result, throughout the rest of the 1930s production-for-use was largely confined to sewing rooms.

Opportunities to establish a permanent government work program during this era met with failure. The Social Security Act, passed in 1935 amid great fanfare, provided the first opening. Although the initial version included an Employment Assurance program, establishing the government as "employer of last resort" committed to provide jobs after peopleís Unemployment Insurance benefits expired, the controversy surrounding the FERA and CWA led to cutting Employment Assurance during committee deliberations. Instead, work programs were continued in the temporary WPA, while the permanent Social Security Act included social security, UI, Aid to Dependent Children (ADC, which was changed to Aid to Families with Dependent Children, or AFDC) and other categorical assistance relief programs for the poor, and programs for maternal and child health.

As World War II brought the economy out of the Depression in the early 1940s, a permanent work program with the government as "employer of last resort" was again considered. This recommendation came from the National Resources Planning Board (NRPB), a high-level federal commission established early in the New Deal and charged, in part, with developing overall economic plans for the post-World War II period. The NRPB called for a "new bill of rights" that would ensure the basic necessities of life, including the right to a job provided by the government if the private sector failed to do so, and advocated the "(f)ormal acceptance by the Federal Government of responsibility (emphasis theirs) for insuring jobs at decent pay to all those able to work regardless of whether or not they can pass a means test." This implied a program similar to the CWA, and would have been accomplished through a permanent "Work Administration" to provide "socially useful work (emphasis theirs) other than construction...for the otherwise unemployed." These ideas were embodied in the Full Employment Bill introduced in 1945. However, in place of a large-scale work program and other policies to increase economic security, Congress passed the watered-down Employment Act of 1946, relying heavily on the market mechanism, and promising only to provide maximum opportunities for employment balanced with a commitment to low inflation.

There was a break in fair work programs during the post World War II era through the early 1960s when increasing real wages accompanying the expanding economy gave credence to the view that there would be a chicken in every pot and two cars in every garage -- that is, if you were white -- and anticommunism mitigated against social welfare programs that were the least bit progressive. Yet the idea of a voluntary work program was kept alive through the 1940s and 1950s by organizations such as the National Manpower Council and the Upjohn Institute, which published books and reports discussing job creation, training and education. These ideas came to fruition in the 1960s. Growing concern about long-term unemployment caused by changing requirements for workers' skills led to the 1962 Manpower Development and Training Administration (MDTA), which provided training along with a stipend for unemployed workers. Following historical precedent, it initially targeted white male heads-of-household, although its focus shifted to African-American youths as the War On Poverty got underway in the middle of the decade. The MDTA was soon surpassed by the Neighborhood Youth Corps and other War On Poverty programs, which provided work, education, training, and support services to pull people considered to be on the margins of society into the mainstream.

Fair work programs similar to those of the Great Depression finally returned in the 1970s. The recession of 1969-1970 ended the economic boom of the 1960s and led to the Public Employment Program (PEP) in 1971. Three years later the Comprehensive Employment and Training Act (CETA) replaced the PEP, the MDTA, and the War On Poverty work and training programs with an all-inclusive program. The PEPís job creation program became Public Service Employment (PSE) in CETA.

While PEP and PSE marked the return of job creation, they were far more constrained than their 1930s' predecessors. Most importantly, they were only allowed to develop work in services; the construction and production-for-use projects that were such critical components of the earlier programs were absent. And PEP and PSE were small in comparison to the earlier programs, reaching a high point of 742,000 people in March 1978. Yet this was only one-sixth of the 1930s maximum -- even though the labor force had doubled in the intervening years. While the 1930s' programs created work for one-third of the jobless at their high point, PSE provided jobs for only 12 percent of the unemployed at its maximum. In spite of these overall constraints, PEP and PSE were the targets of many of the same criticisms that had earlier plagued the FERA, CWA, and WPA. Most common were charges that wage rates were too high which attracted people from private sector employment, the work was inefficient and unnecessary "make-work," and the entire programs were too expensive and were riddled with fraud and abuse. Unlike the earlier programs, CETA was also criticized for substituting federal CETA funds for local tax revenues and displacing normal government workers with CETA workers. And, as in the 1930s, contradictory mandates meant that these criticisms were often accurate.

Regulations determining eligibility and payments placed PEP and PSE closer to the fair work end of the spectrum of government work programs than was true of the 1930s-era programs. Although a strict "means test" was not required for participation, there was a mandated focus on providing work for the long-term, low-income unemployed. Eligibility was based on the length of time a person was jobless, family income, and status with respect to UI or AFDC; additionally, preferences for veterans became an increasing concern. These criteria were initially quite loose, but became increasingly strict throughout the 1970s. Payments for program participants proved a continual source of controversy, as public sector unions fought to maintain high payments, in order to prevent replacing public sector workers with CETA workers, and capitalists tried to keep rates down, in order to help maintain a preference for private sector jobs. Payments were supposed to be based on "prevailing rates" for similar work in the local area, with ceilings on the annual amount that participants could be paid. Although local agencies could supplement PSE payments out of their own funds, the ceilings constrained the programís ability to create jobs in high-wage areas of the country.

In spite of the restriction to services, a wide variety of socially useful work was accomplished in the PEP and PSE. Public works projects included: maintenance and repair of buildings, equipment, streets, sidewalks, and sewers; cleaning streets and alleys; collecting garbage; flood and erosion control; and conservation, including weatherization and repair of low-income homes. Parks and recreation projects included the development and maintenance of facilities, as well as operating recreational programs. Educational services, too, involved maintenance of school buildings and grounds, as well as instruction-related projects for teacher's assistants. New programs and services were set up, such as low-cost "meals on wheels" delivered to the elderly in their homes, community arts programs, and conducting surveys of community needs for social services. Additional workers were provided to expand services in existing agencies, including day care centers, shelters for battered women and their children, legal aid, hospitals (e.g. setting up hypertension screening programs and treatment programs for alcoholics), and law enforcement agencies. In fact, many institutions developed during the 1970s, such as food cooperatives and women's health clinics, ran on shoestring budgets and depended on CETA workers to remain in operation.

Confinement to the provision of services was problematic, as it fueled criticisms that the programs were inefficient and unnecessary. This followed in large part from the dominance of ideology extolling the virtues of the market. According to this logic, the private sector is efficient because decisions are based on profits; conversely, the government by its very nature is inefficient because its decisions are based on other criteria, theoretically what the society needs.

Some changes were made in CETA as part of the 1978 reauthorization. It was subjected to increased restrictions regarding targeting and determination of payments. Some of these were beneficial, most importantly, the directive to set up projects for women who had only worked in the home (for no wages), generally called displaced homemakers (DHs), and for women in male-dominated (labeled nontraditional) occupations. Also, in response to charges of inefficiency, a new program was started. The Private Sector Initiative Program (PSIP) established Private Industry Councils (PICs) to develop training programs designed to meet the needs of local employers.

During this era, another attempt to pass legislation mandating that the government serve as "employer of last resort" again met with failure. As originally proposed in 1974, the Humphrey-Hawkins Full Employment and Balanced Growth Act would have included a public service employment program that would go into effect when the unemployment rate fell below 3 percent. However, it met a fate similar to that of the 1945 Full Employment Bill, and was finally passed in an eviscerated form in 1978.

When Reagan became president in 1981 he quickly eliminated the much maligned PSE, and allowed CETA to expire the following year. It was replaced by the far narrower Job Training Partnership Act (JTPA), which continued the PSIP and was designed to provide usually short-term training to meet the needs of local employers. The JTPA lasted until 1998, when it, in turn, was replaced by the Workforce Investment Act (WIA).

A Brief History of Workfare Programs in the U.S.

Historically, workfare has been by far the more common type of government work program. From the colonial era through the early 1930s the poor were typically put to work in workhouses or on "poor farms" in order to prove that they were "deserving" of aid. Sometimes, the work they did was clearly punitive "make-work" -- e.g., moving rocks or digging ditches and filling them up again. Although things of value were sometimes produced, most commonly, food on the "poor farms," the proscription against interfering with the private sector limited the production of useful goods. Additionally, paupers were commonly "auctioned out" to the highest bidder, providing extremely low-cost labor. [The discussion in this section is based on Rose 1995.]

As the Great Depression of the 1930s brought massive unemployment, there was a break in the focus on formal workfare programs. Yet coercive work policies remained, primarily in the practice of closing relief rolls in response to increased demand for low-wage labor. Often carried out during planting and harvesting seasons and in response to complaints of a lack of domestic service workers, it was particularly common in localities with large populations of African-Americans. This practice of routinely closing relief rolls to provide a more than adequate supply of low-wage agricultural labor continued through the mid-1960s, and was finally ended when hearings accompanying the War on Poverty brought it to the attention of the public.

While less common than usual, worktests were not eliminated during the 1930s. They continued in the Transient Program from 1933 through 1935, as participants were required to work in exchange for relief. Additionally, many local relief administrations revived the worktest, as recipients performed maintenance work, primarily cutting grass, collecting garbage, and cleaning sewage disposal plants.

The post World War II era saw the termination of the progressive programs of the New Deal and the expansion of punitive relief policies. As growing numbers of African-American women received Aid to Dependent children (ADC), "suitable home" and "man-in-the-house" policies were applied to restrict their participation -- homes were sometimes judged "unsuitable" if the mother was not married and if a man was suspected of living in her house. Coercive work policies continued through periodically closing relief rolls to force women into (typically extremely) low-wage labor, and were formalized by "employable mother" rules which decreed that ADC recipients could be denied aid at times when additional labor was needed in the fields.

The 1960s was another era of relatively progressive welfare policies. However, it also saw the return of a formal work program that became increasingly punitive through the 1970s. The Work Incentive (WIN) Program, enacted in 1967, required registration for eligible recipients of AFDC [Aid to Families with Dependent Children, which ADC was changed to in 1962]. Reflecting the overall tenor of the 1960s poverty programs, it initially relied heavily on incentives to encourage participation. Most important was the "income disregard," or "work incentive bonus," as the first $30 plus one-third of the remaining earned income was not counted in the computation of welfare payments, and work-related expenses (WREs) were reimbursed. Importantly, WIN included an emphasis on education and training to increase participantsí skills in order to help them obtain and keep jobs.

The 1970s witnessed the development of more punitive work programs. In 1971, the Talmadge amendments ushered in the more restrictive phase II of the WIN program, requiring registration of AFDC recipients with children over age six, and sanctioning noncompliance with the loss of the offending personís portion of the familyís welfare payment. The 1970s also saw the development of experimental work programs. One such program was the Supported Work Demonstration (SWD) project, which was aimed at long-term AFDC recipients who faced severe barriers to employment -- primarily little job experience, training, or education. Providing time-limited (generally twelve to eighteen months) employment in closely supervised and supportive work settings, the goal was to help participants acquire skills, habits, and credentials that would enable them to find and retain jobs. Careful studies of the SWD project found that it was quite successful, both for the individuals involved and for the program as a whole. Compared to those in control groups, participants had higher rates of employment, hours of work, and earnings (although their wages were still low), results that were sustained over several years. Further, the program proved cost-effective, as the benefits outweighed the costs for welfare agencies.

In spite of these beneficial results, SWD remained a small program. Instead, another experimental program, workfare, became the dominant work program for welfare recipients. Commonly called Community Work Experience programs (CWEPs), workfare was initially developed in states as diverse as California, Massachusetts, and Utah. These were traditional workfare programs, as work was required simply in exchange for welfare payments -- not for wages. Hours of work were typically calculated by dividing the familyís AFDC payment (minus any earned income) by the minimum wage. And sanctions were critical, as noncompliance could result in the diversion of payments to a third party, or in suspension or termination of the offenderís portion. One of the most avid proponents of CWEPs was then governor of California, Ronald Reagan, who ignored the consistent findings that the programís costs outweighed its benefits, and extolled the virtues of the purported decline in "welfare dependency," both as governor and in his later capacity as President.

The experimental workfare programs of the 1970s proved to be precursors of the "workfare explosion" of the 1980s. Eschewing analyses that recognized the structural, systemic causes of poverty and bolstered by "poverty-by-choice" arguments -- that millions of women had chosen to go on welfare since it was so generous -- the Reagan administration proposed, and Congress enacted, policies designed to end "welfare dependency" by pushing women into the (usually low-wage) labor market. As part of the steady devolution of control from the federal government, individual states were given wide discretion in designing and implementing policies. Under the 1981 Omnibus Budget Reconciliation Act, states were encouraged to develop WIN demonstration programs, alternatives to WIN. Yet the federal government made some changes as well, as incentives under the WIN program were gutted (e.g., the earned income disregard was reduced and reimbursements for WREs were limited), participation was generally mandated for women whose youngest child was three or older, and sanctions were intensified. The focus on punitive policies was further reflected in the emphasis on job placement, as education and training became increasingly restricted.

From 1981 through 1996, government work programs were characterized by punitive workfare and workfare-like programs. The experimental WIN demonstration programs were institutionalized through the JOBS (Job Opportunities and Basic Skill training) that was part of the 1988 Family Support Act. Although the legislation included options for an array of education and training, as well as support services, funding for these activities was minimal. Instead, the focus remained on immediate job placement -- as well as on collecting child support payments from "deadbeat dads" -- and "job search" became a favored activity of welfare departments. There were some helpful programs, most importantly, programs for women to obtain college degrees. However, although studies consistently showed that a college education was the surest route out of poverty for women, these programs were typically limited and required participants to go through many hoops in order to remain in them.

Government Work Programs since 1996

The PRWORA was an almost logical extension of the policies between 1981 and 1996; yet it went beyond what could have been imagined when the onslaught began. There are two main concerns. First, the PRWORA ended the federally guaranteed entitlement to welfare that had been enacted through the 1935 Social Security Act, and replaced it with a block grant. Thus far, due to the economic expansion, this has not been a problem. However, business cycles are an integral component of capitalist economic systems, and when the next one occurs, it is quite likely that states will use up their block grant funds. What will ensue remains to be seen. The second, and best known, feature of the PRWORA is time limits: families can receive welfare for a maximum of 24 months in any welfare "spell" with a lifetime limit of 60 months. Indeed, the name of the program was changed to reflect this, as AFDC became TANF -- Temporary Assistance for Needy Families. Although up to 20 percent of families on the rolls can be exempted from participation in work activities, as the rolls decline the numbers of people with severe barriers to employment become a larger percentage of those remaining on welfare. Time limits were intended to push welfare recipients into jobs, thought to be the key to ending "welfare dependency"; the other option is marriage.

While the PRWORA is generally a reactionary, repressive piece of legislation, a positive feature was eliminating the prohibition on the use of welfare funds for wage-paying, publicly funded job creation. In conjunction with two additional recent pieces of legislation that concern employment programs and policies, there are consequently new opportunities for job creation. In 1997 the Welfare-to-Work (WtW) program was enacted to set up work activities for hard-to-employ TANF recipients and noncustodial parents of children in TANF households. And in 1998, the Workforce Investment Act (WIA) replaced the Job Training Partnership Act (JTPA) with a new structure of programs for job training and information. Under this act, job training is very broadly construed, so that one of the options includes job creation programs [Johnson and Savner 1999].

Not surprisingly, a variety of programs ranging from workfare to fair work have been implemented. Some are clearly workfare, also called "community service" or Community Work Experience. For example, the Transitional Opportunity Program (TOP) in Hawaii and Wisconsin Works (W2) require work in exchange for welfare payments, with hours of work determined by dividing the grant by the minimum wage. More closely resembling fair work are Community Service Employment (CSE) or Community Service Jobs (CSJ). These include Oregon Jobs Plus, in which employers are subsidized up to the minimum wage and participants are paid labor market wages, and Vermontís Reach Up, in which participantsí TANF payments are diverted and considered as earned income. All of these programs provide only temporary placement -- the objective is always to channel people into unsubsidized employment.

Differences between these current workfare and fair work programs focus on the degree to which the work resembles "real jobs." In those closer to the fair work end of the spectrum, payments are determined by the number of hours worked, and are connected to government programs for wage work, most importantly, eligibility for the Earned Income Tax Credit (EITC) and contributions into the social security system. Therefore, as wage-paying jobs, they stop the time-limit "clock" for assistance through TANF. Further, treating participants as "real workers" improves the "soft skills" of general job readiness and orientation imparted through work programs. On the other end of the spectrum, since workfare programs pay only welfare, not wages, they do not make workers eligible for the EITC or affect future social security payments. However, they usually keep the TANF time-limit "clock" running, thereby constraining future aid.

An Agenda for Fair Work and Related Programs

Finally, we should develop an agenda for fair work and related programs as well as supporting arguments. Since work has been integral to welfare reform, this also involves welfare policies. In doing this, it is helpful to look at both short-range and longer-range goals. This allows us to look at what is possible in the current political climate separately from a more comprehensive vision of what is needed, and thereby maintain a sense of hopefulness about the project.

Short-term goals would focus on turning the lemon of welfare reform into lemonade -- to the degree that this is possible. This includes arguing for job creation at living wages through TANF, as well as through the WIA and the WtW programs. With respect to welfare per se, as Frances Fox Piven [2000] declared in a recent talk, we should push to make welfare administrations interpret ambiguities in TANF in the interests of poor people. Of primary concern is ending the diversion of applicants and reversing the frightening increase in rejection rates of those applying for aid, making sure that those cut off welfare receive transitional benefits, and giving work credit for attending school. Further, punitive policies should be reversed. Most important is to end lifetime limits, stringent work requirements, and the excessive use of sanctions.

There are important issues to note about the focus on wage work as a cure for "welfare dependency." While this may indeed end the dreaded "welfare dependency," it does nothing to reduce poverty. Indeed, it flies in the face of the reality of the low-wage labor market, as these jobs keep people poor, not to mention fail to provide health benefits. Furthermore, low-wage jobs typically entail "flexibility," as hours of work are generally variable to meet the needs of employers. However, this is the antithesis of what mothers need -- predictable hours of work in order to arrange for childcare. As Randy Albelda [2000] recently noted, it is not a case of mothers not being "job-ready" -- rather the jobs are not "mother-ready." Further, if the mandate to work for wages is accepted, then women should be able to earn enough to support themselves and their children. However, this has rarely been feasible. The majority of women are still employed in low-wage clerical work, and in 1998 the female/male wage gap for full-time, year-round workers was 73 percent [Council of Economic Advisors, table B-31].

It is telling that discussions about ending "welfare dependency" have focused on wage work while the other option -- marriage -- has been ignored. This has been the case in spite of the fact that marriage was a focus of the original legislation. In fact, the PRWORA begins with the declaration that: "1) Marriage is the foundation of a successful society." The fact that this age-old method of survival -- economic dependence on a man -- has been largely ignored in the public debates does not diminish its importance in reality.

Longer-range goals promoting fair work and related programs would provide supports for all types of families so that children will be protected and both women and men will have choices about combining wage-work with work in the home. This could be done through an "economic bill of rights" in conjunction with a redefinition of what constitutes work in our society.

The definition of what constitutes work is of critical importance in the current era. There is general agreement that paid labor -- no matter how menial or degrading -- is "real work." However, this is not true of caretaking work in the home -- or of volunteer work in the community. Indeed, poor women typically function as a type of "social glue," providing vital support services for low-income neighborhoods. This unpaid work in the home and in the community needs to be recognized as the critically important, socially necessary labor that it is. And given the consumerist, commodified nature of society, women, and men, should be compensated for this work. A relatively easy way to accomplish this would be to broaden widowís pensions -- one of the survivorís social security benefits -- to cover all caregivers. Another way to do this would be implementation of a Caregiverís Protection Act.

An "economic bill of rights" could draw on the "new bill of rights" initially proposed by the National Resources Planning Board in the early 1940s as well as other proposals from the past -- Employment Assurance from the original version of the Social Security Act, the 1945 Full Employment Bill, and the original Humphrey-Hawkins bill in the mid-1970s. Their visions -- that everyone has the right to socially useful work and freedom from poverty -- should be recaptured and revitalized in the current era. In promoting these ideas, it is instructive to recall the tenor and substance of discussions from the earlier eras during which fair work programs were part of the policy mix, both to counter some of the pessimism that can so easily infuse current debates and to revive some of these proposals and sentiments. Indeed, discussions were so different -- in both style and substance -- from those since the early 1980s that it would behoove us to reclaim them from the proverbial dustbin of history.

From the 1960s we should recall the focus on bringing people on the margins of society back into the mainstream. Indeed, as Clifford M. Johnson and Steve Savner point out in Federal Funding Sources for Public Job Creation Initiatives [1999], one rationale for using TANF funds for job creation programs for teenagers and young adults is to reduce the incidence of out-of-wedlock births. The implication is that having a "real job" increases a personís choices and feelings of hopefulness about the future.

Most relevant for this study, the fair work programs of the 1930s and 1970s can serve as models for future programs. Several lessons emerge. The first set involves basic guidelines. Programs should be universalized, i.e. not confined to specific groups, so that more people have a stake in their viability and to avoid labeling program participants. Programs should be federalized rather than leaving more discretion to individual states, as history shows that this results in more progressive policies. And programs should be funded by the government rather than by individual employers, as the latter increases businessí tax burden and fosters a reluctance to hire new employees.

Most fundamentally, a fair work program would establish the federal government as "employer of last resort," providing jobs when the private sector fails to do so. Participation would be voluntary, and open to people who are unemployed or underemployed. Wages would be based on "prevailing rates" in local labor markets. In order to protect peopleís jobs and prevent the development of a second-class tier of workers, program participants would be allowed to join unions.

A fair work program should be augmented by comprehensive E&T programs -- also universalized -- that would provide a range of education and vocational training. This would also help replace the current work attachment model of welfare reform with a human investment model. The 1970s offers models of creative employment and training programs, , some designed for groups who have been relatively marginal to the labor market, most importantly, women who have only worked in the home. Programs in the 1980s and 1990s to enable welfare recipients to obtain college degrees, and thereby find jobs that can get them out of poverty, could be augmented to encourage this route instead of requiring participants to jump through hoops.

We can see from the massive and innovative programs of the 1930s that large-scale programs providing a range of work are indeed possible. Especially given the cutbacks in both services and publicly funded construction since the early 1980s, a great deal needs to be done, and could easily provide work for millions of people. A partial list includes repair of crumbling infrastructure (e.g. repairing roads and bridges), other construction projects (e.g. low-cost housing and public transportation systems), public sector projects (e.g. teacher's aides in public schools, public health projects, programs for school drop-outs, and conservation), and increased public support for the arts.

Any proposal in support of a large-scale job creation program should anticipate criticisms of fair work programs and develop responses to these time-worn complaints of fraud and abuse, overly high payments, and inefficient and unnecessary "make-work" that will undoubtedly arise.

Charges of fraud and corruption can be dismissed as red herrings. They are routinely made towards government agencies and programs simply because services are provided by government instead of the private sector. Given the ideological dominance of markets since the 1970s, this has become even more problematic. Tellingly, these criticisms have been far more constraining when directed at job creation programs, compared to other government programs such as the military, since these programs expose the inability of capitalist economic systems to create enough jobs for everyone who wants to work for wages.

Charges that payments are too high stem from fears that people will have attractive alternatives to private sector jobs and that women will have alternatives to marriage. The first concern reflects the dominance of the private sector as higher wages cut into a firmís profits -- and have historically constrained wage policies on fair work programs. The second alternative -- an "independence effect" for women -- should be defended as a legitimate choice.

The criticisms that have proven most problematic historically concern inefficiency and unnecessary "make-work." Responses should clarify the contradictions that have caused them. Charges of make-work, commonly made towards fair work programs in the 1930s and 1970s, resulted from directives to avoid substituting work program participants for "normal government workers" or "normal government operations." As a result, the work was easily construed as unnecessary -- otherwise it was assumed that the government would already be doing it. Further, we should point out that much of the work done in the past programs was socially useful. In fact, many of the 1930s' projects, from community facilities to WPA murals, still exist. And PSE projects, from weatherizing low-income homes to working in women's health clinics, enhanced the lives of both the workers and the people receiving the services.

Criticisms of inefficiency and "make-work" were further exacerbated by contradictory regulations. This was particularly clear during the 1930s when requirements to use a maximum amount of labor and a minimum of capital goods -- i.e. to make work -- logically led to these charges. Further, during the 1930s, production of clearly useful consumer goods was hampered by regulations designed to ensure that the programs would not interfere with production-for-profit.

There is an additional argument that can be made in support of fair work programs. They make good economic sense, especially during recessions. In contrast to "trickle-down" policies in which tax breaks were given to the rich, fair work programs are part of "bubble-up" policies that channel money into the hands of working people who will quickly spend it. This results in a "percolator effect": it increases demand for goods and services and induces capitalists to increase production, hire more workers, and thereby expand the economy. Instead of depending on unemployment to hold wages down and maintain profits, we can counterpose wage-led growth, in part through policies that create jobs.

A large-scale fair work program with supporting family and labor market policies would be expensive. Money to fund it can come from rescinding the tax cuts for the wealthy and for corporations passed in the 1980s and 1990s, substantially reducing the military budget, closing corporate tax loopholes, eliminating the ceiling on income subject to social security taxes, and a tax on the sale of short-term assets.

In order to bring these ideas into the national discourse, we need to expand the boundaries of what is now considered feasible. In place of the constricted vision that defines current discussions, we need to recapture the sense of possibility that pervaded the 1930s, 1960s, and 1970s. In place of the supremacy of the market and the politics of greed, we need to argue that real family and moral values means making sure that children are cared for and that parents donít have to undergo humiliation in order to survive. In place of an analysis positing poverty as an individual, psychological problem that destroys a person's "moral fiber," we need a social-structural analysis that sees poverty and unemployment as endemic to capitalist economic systems. History shows that this is possible.


  1. The WPA was originally an acronym that stood for the Works Progress Administration. In July 1939, as part of the growing reaction following the election of many conservatives to Congress the previous November, its name was changed to the Work Projects Administration and it was housed thereafter under the Federal Works Agency.

  2. These studies were carried out by the Manpower Development Research Corporation (MDRC), which was initially established to evaluate the SWD project, and continues to conduct evaluations of government work programs.

  3. 110 stat. 2110, Public Law 104-193, Aug. 22, 1996.


  1. Albelda, Randy, plenary session, Work, Welfare, and Politics, University of Oregon, February 28-29, 2000.

  2. Council of Economic Advisors. Economic Report of the President, 1985. Washington, D.C.: Government Printing Office, 1985.

  3. Johnson, Clifford, and Steve Savner, Federal Funding Sources for Public Sector Job Creation Initiatives (Washington, D.C.: Denter on Budget and Policy Priorities and Center for Law and Social Policy, December 1999.

  4. Piven, Frances Fox, plenary session, Work, Welfare, and Politics, University of Oregon, February 28-29, 2000.

  5. Rose, Nancy E. Workfare or Fair Work: Women, Welfare, and Government Work Programs. New Brunswick, N.J.: Rutgers University Press, 1995.

  6. Rose, Nancy E. Put to Work: Relief Programs in the Great Depression. New York: Monthly Review Press, 1994.

Request a copy of this publication