'Employable mothers' and 'suitable work': a re-evaluation of welfare and wage-earning for women in the twentieth-century United States.

U.S. welfare policy has yet to adequately address a mother's two work roles - care-giving and wage-earning. These two responsibilities have produced conflicting policy responses, sometimes within the same historical period. Citizens and legislators have raised concerns about mothers who worked too much outside their homes; or conversely, mothers who did not work enough to support their families. These contradictions are reflected in the language and practice of welfare policy. During the twentieth century, programs such as mothers' pensions gained public support by promising to subsidize some mothers to raise their children. That language changed after World War II to reflect an expectation that women must enter the workforce to earn, as in the workfare initiatives of the late twentieth century. In contrast to the shift in language over time, the practice of encouraging wage-earning has demonstrated remarkable continuity.

Similarly, the scholarship on gender and welfare has focused on one, not both, of women's responsibilities.(1) Little scholarly attention has been paid to the relationship between wage-earning and welfare for the years preceding World War II. Instead, the standing interpretation for both mothers' pensions and Aid to Dependent Children explains these programs as policies that supported women to stay at home. Whether viewed as a maternalist policy designed to protect motherhood, as presented in Theda Skocpol's recent work; or as a restrictive policy that regulated women's behavior as described by Linda Gordon, the leading scholarship discusses these programs as a relationship between government and families.(2)

In contrast, a secondary theme in welfare literature, the labor regulation thesis, focuses on wage-earning to the exclusion of gender or family issues. According to Frances Fox Piven and Richard Cloward, welfare policy encourages earning regardless of sex, family status, or race and cares little about the protection of family life or the maintenance of traditional gender relations. In fact, both scholars dismiss gender as a central element in welfare's operation. Piven and Cloward's focus on the intersections between markets and welfare policy could have filled this gap in the literature, but their insistence on gender's irrelevance and the primacy of the market over all other factors has kept the interpetations divided between class conflict and gender conflict.(3)

This article argues that policies for impoverished, mother-only families can best be explained by considering the public sector's relationship to both markets and families. This survey of the evolution of such policies illustrates that the expectation to earn has long been a component of public aid for women, even during the early twentieth-century when the greatest support for a maternal endowment existed. The connection between wage-earning and welfare, I argue, redirects us to understand the role of politics in the development of the programs, in the debates over public spending, and its impact on potential beneficiaries of public aid. By following the transitions in the policies from state-level work rules to federal policy, and eventually to the Supreme Court's legitimation of work requirements, this article offers new perspectives on the relationships among gender, race, and welfare over the twentieth century.

The article is organized around three eras of significant transition in public provision. It begins with the origins of public aid to impoverished mothers through the mothers' pension programs of the early twentieth century. The policy that rapidly passed state legislatures, largely due to its appeal to support women to rear their children at home, operated as little more than a wage subsidy. Faced with contests over public spending, the program applied exclusionary criteria, among them a woman's ability to earn. The second period examines the Social Security Act and the development of two provisions for mother-only families: Aid to Dependent Children (1935) and Survivors' Benefits (1939). The Social Security Act retained the language of maternal support but federal policy administrators never enforced this goal. Rather, state and federal governing bodies jockeyed for a workable position between federal regulation and states' rights. The compromise that resulted allowed states to implement work rules. These rules expanded the definition of parental support beyond the male breadwinner to include maternal provision as both wage-earner and care-provider. The "able worker" and "employable mother" criteria worked hand in glove with moral and racial criteria to limit access to benefits. The third section of the article discusses the development of federal work rules in the context of post-war transitions in the economy and civil rights. By the 1960s, welfare policy accommodated state work rules and progressively strengthened earning requirements within federal law - not as a mediator of labor, but as a mediator of political pressure and public spending.

Mothers' Pensions, 1911-1935

Throughout the nineteenth century, two groups made up the largest number of poor relief recipients. Mother-only families were consistently the single largest group using both public and private agencies. The exceptions to this pattern existed during periods of massive unemployment when the second group, unemployed men, turned to poor relief for aid for their families. The long-term need of female-headed families and the intermittent use of relief by male-headed families has been recognized by scholars as the economic consequences of historic social relations. Men, it was assumed, would leave relief as work opportunities returned. Women, however, were more likely to need aid for longer periods due to childrearing obligations and the low wages received for women's work.(4)

The first major transition in provision for impoverished women came with the mothers' pensions laws. Advocates recognized that charity was insufficient, demeaning, sporadic, and inadequate to meet the needs of families in an industrial nation. Social scientists noted the correlation between wage-earning, poverty, and mother-only families. Juvenile court judges and settlement workers wanted to change child welfare policy that took children from their families and placed them in institutions. A gender-conscious mobilization of traditionalist women's groups and progressive settlement house workers called for maternal subsidies to support women to raise their children at home.(5)

The arguments for pensions varied. A traditionalist appeal could be found among members of the National Congress of Mothers. One member, Mrs. G. H. Robertson, urged her coworkers to support the mothers' pension laws to strengthen the bonds of family life. "These friendless mothers are our especial responsibility." Having grown up "to believe in marriage and home-building," Robertson asked her supporters to not "let them find it a tragedy." She called for a national commission and appropriations to aid mother-only families, by which she meant deserted wives, never-married women, and widows with young children.(6) Others on the progressive edge of this movement called for social insurance for mother-only families. They wanted the government to provide insurance to poor mothers that was similar to private insurance benefits. Citing legal precedent for a wife's claim of support when the husband had property, Sophonisba Breckinridge asked if the same premise of a woman's entitlement would not apply to the woman whose husband died and left no property?

Whether it be termed a pension for the mother, ... or pay for her services as agent of the court, or a grant in aid of family life - it should be available, sufficient in amount, regular in payment, dignifying in its assurance of the community's concern for the well-being of her group.(7)

A diverse group of supporters with a wide range of motivations built support for the new policy and succeeded in passing legislation in a majority of states.

Illinois was the first to pass a state-wide enabling law in 1911, but other states followed quickly. Two years later, eighteen additional states passed mothers' pension legislation. By 1921, forty states and the territories of Alaska and Hawaii had some form of provision for the support of dependent children in their own homes.(8)

Grace Abbott, Director of the U.S. Children's Bureau during the 1920s, explained the rapid spread of mothers' pension legislation in terms of the debate over the value and appropriate location of women's work. Abbott believed the public accepted the new programs because they recognized that "the contribution of the unskilled or semi-skilled mothers in their own homes exceeded their earnings outside of the home." She added that "it was in the public interest to conserve their [women's] child-caring functions." Abbott's assessment emphasized the well-documented evidence on women's poor working conditions, and raised the issue that different causes of poverty could demand different solutions. But she presented an overly optimistic view of the state's interest in providing funds for these programs.(9)

Abbott's statement outlined three problems in creating social policy for mother-only families. The first issue referred to the disparity women faced in the workforce. Occupational segregation and job crowding limited women's work options. Years of research by Abbott's colleagues in the Department of Labor and the nation's settlement houses had documented that the majority of wage-earning women worked in sectors characterized by long hours and low wages. These features made self-sufficiency, let alone support of a family, difficult if not impossible. Abbott's sister, Edith Abbott, had written the pioneering book on the economic status of women workers in the U.S. - a book that highlighted the structural barriers faced by women wage-earners. In addition, Grace Abbott understood very well the opposition women faced from trade unionists in improving these labor conditions for working women. The majority of women could not work their way out of poverty as men theoretically were able to do, and useful social policy would need to recognize these facts.(10)

Second, Abbott recognized that the prevalent cultural view of women's primary duty as a mother had played a significant role in the rapid spread of mothers' pension laws. The National Congress of Mothers and the General Federation of Women's Clubs - two organizations with the largest membership of any women's groups in the early twentieth century - promoted the maternal role over any other. Abbott acknowledged the views of these traditionalist women's groups with her reference to the "public's interest to conserve [women's] child-caring functions." She also knew that social theorists blamed working women for family fragmentation and juvenile delinquency. Any social policy for mother-only families would have to incorporate these ideas into its plan.

The part that Abbott portrayed too rosily - the third problem in creating new policy for mother-only families - was the public's willingness to pay for a subsidy to female-headed families. As Director of the Children's Bureau, she was intimately aware of meager family stipends, requirements for women to earn, and the differential treatment of mothers based on their race, marital status, and in some cases, family size. At the level of implementation, the public discourse which endorsed aid to impoverished mothers met a serious challenge from an equally strong demand for families in poverty, whether male-headed or female-headed, to earn their way to self-sufficiency. A conundrum evolved at the level of implementation. Aid for mother-only families would be endorsed, but a municipality could not possibly support all mothers in need of aid. Consequently, in county after county, administrators of the program developed a variety of screening mechanisms, including one that considered the mother's "ability to earn."(11)

A specific example from Chicago demonstrates the use of work requirements. When first passed in 1911, the mothers' pension law authorized that public funds could be provided to parents for the support of their underage children. That provision included single as well as two-parent families, it listed no limits on the size of grants, nor did it describe other eligibility requirements. The Chicago program became the focus of national attention and debate due to its size and the involvement of its notable reform community. By the next meeting of the state legislature, political pressure had forced a revision of the law and created one of the most restrictive measures at the time (1913). New eligibility rules restricted grants to widows and wives of disabled men; children under the legal work age, which was fourteen years at the time; and citizens.(12)

Wage-earning became intertwined with public aid in two ways - through the law and through denied eligibility. The law stated that a mother "may be absent [from home] for work a definite number of days each week to be specified in the court's order," and women received an examination to determine the degree to which they could earn wages.(13) The history of Mary Legaikas, referred to in the case records of the Juvenile Court as Mrs. C, illustrates a common situation for pensioned mothers.

In June, 1913, Mrs. C, a Polish woman, applied for a pension for her two children aged 8 and 5 years because she found it impossible to earn enough to support them. Her husband had died of heart disease in 1909, leaving some insurance; but the money had been used for paying funeral bills, debts, and living expenses. The family had been compelled to ask help from the county agent and the united charities a number of times during the four years following the death of the father. A stepson had gone to work at the age of 14, but Mrs. C found him so unmanageable that in 1911 she sent him to his uncle in Tennessee. Mrs. C had been earning only $10 a month by sweeping in a school.(14)

The Juvenile Court's caseworkers estimated this family's budget at thirty-four dollars a month, and granted the two children a pension of ten dollars each. The pension, along with Legaikas' earnings of ten dollars a month, left the family only four dollars short of their need. When Legaikas developed health problems from cleaning the school, caseworkers helped her find a job cleaning in a bank where she earned two more dollars a month. Mary Legaikas' story was recorded as a "success" among pensioned families, but her wage-earning was not exceptional. Approximately one-half of the women who received pensions in Chicago during the first fifteen years of its operation engaged in wage-earning.

The second way in which wage-earning became a component of the mothers' pension program was the practice of denying women aid on the basis of their "potential to earn." Although all able-bodied women were urged to find some type of employment, African American women were believed to be more able to find work. A combination of job segmentation which retained certain classes of jobs for blacks and the perception that "black women had always worked" led to this racialist perception. For example, take the case of an anonymous African American woman who had received a pension for her five children until the 1913 revision made her ineligible. At that point, she turned over the home responsibilities to her fifteen-year old daughter and went out to work. Two years later, when the law became more inclusive, the woman reapplied for a pension, but was denied on the basis that she was "a good wage earner." The court gave precedence to her role as an earner over that of a caretaker for her dependent children, and refused even a subsidy for the children, such as that received by the Legaikas family.(15)

The expectation that women who received mothers' pensions would also work for wages was not unique to Chicago. A Children's Bureau study published in 1923 found that local administration frequently encouraged wage-earning by mothers to meet the gap between family needs and pension stipends. The report noted:

In each place studied there were some mothers who did not have as much free time to devote to their children as was considered desirable. Some mothers were working because there was no other way to get an adequate income for the family, although the physical strain of work in addition to the care of the house and the children was probably more than they could long endure ... In many instances, however, it was believed that some money-earning occupation on the part of the mother was a wholesome influence in the family life.(16)

Slightly over half - 52 percent of the 942 women included in the study - earned to support the family. Twice as many worked outside the home for wages as those who did piecework at home. Furthermore, nearly half (48 percent) of those who worked outside the home worked three or more full-time days per week.(17)

The mothers' pension program in practice diverged quite significantly from the initial proposals of its proponents, and its public benefits were distributed differentially on several grounds. Fiscal and political pressures combined with moral and racialist ideologies to shape the policy in its implementation. Rather than a right or entitlement, mothers' pensions became a needs-based program. From this perspective, it is apparent that using economic criteria like one's ability to earn allowed local administrators broad discretion in providing or witholding benefits. This perspective on the linkage between public aid and work requirements places the failure of women's employment policies in stark relief. The competition between self-sufficiency and care-taking contributed to the weakness of the program at this early stage. To what degree did these ambiguities accompany the program in the transition from locally administered, state-level programs to a federally-supported policy in the Social Security Act?

Aid to Dependent Children, 1935-1962

Historians have looked upon the great safety net of social provision set in place by the Social Security Act as the foundation of the modern welfare state, albeit a limited one. The exclusion of domestic service and agricultural work from social insurance benefits resulted in a two-tier system that protected some workers and not others. Furthermore, many New Deal programs inscribed with family wage ideals limited women's access to social security by emphasizing their primary responsibility within the home, as Linda Gordon has argued. Yet, as this section will illustrate, the practice of welfare continued to blur women's dual roles. Most important, the meaning of parental support - as a male contribution to the family - changed. In the decades following passage of Aid to Dependent Children, state and local governments increasingly passed rules that linked welfare with wage-earning.

The Depression all but dismantled the state mothers' pension programs. The policy had existed in 1600 counties in 45 states by 1931, but during the next two years, counties began to drop the program. Declining revenues limited the number of families able to receive the aid, and social services went through a period of considerable fluctuation during the first years of the economic crisis. The Children's Bureau reported that 90 counties had stopped their programs between 1931 and 1933. Some counties retained the program, but added rules that excluded families on the basis of race or family size. Many cities simply kept eligible, but unfunded people on a waiting list. In Chicago, 1,434 families received mothers' pensions in 1935, but the waiting list for the program contained an additional 7,942 families. The public relief department carried 3,870 additional families who could have qualified for mothers' pensions, if additional funds had been available.(18)

During the first few years of the Depression, social service agencies made repeated attempts to convince the federal government of the severity of the economic crisis and the necessity for federal intervention for unemployed families. By 1935, federal funds helped states establish relief and employment programs. That same year, after much debate and compromise, Congress passed a set of provisions designed to provide a safety net for American workers and their families called the Social Security Act. The act had two components: one for social insurance and one for public assistance. Promoters of social insurance met with some success particularly in the provisions for unemployment insurance and old age insurance, but the program had many shortcomings. The social insurance provisions covered limited areas of work and by contemporary estimates excluded about one-half of the working population. In addition, the categories of coverage excluded a majority of women and minority workers. These omissions led to a compromise position to maintain the relief programs; thus the public assistance section of the Social Security Act included categorical aid to the disabled, aged, and dependent children.

Title IV included provisions for Aid to Dependent Children (ADC) and specifically authorized funds for the support of families "who have been deprived of parental support or care by reason of the death, continued absence from the home, or physical or mental incapacity of a parent."(19) As such, supporters maintained their ideal "to enable the mother to stay at home and devote herself to housekeeping and the care of her children."(20) In addition, ADC expanded the benefits of mothers' pensions in two important ways. First, a state's acceptance of the program and its federal funds obligated it to implement the program in every county in the state, share costs with counties, and coordinate the program from one central agency. Mothers' pensions never had this breadth of coverage. Second, ADC expanded eligibility to include deserted, separated and unmarried mothers. Only a few states had previously passed such inclusive eligibility.

On the other hand, ADC passed Congress with a series of compromises that reduced the scope of the Act and left considerable control to local administration. Southern Congressmen had effectively eliminated a provision in the original legislation that would have standardized benefits to all recipients by requiring states to provide "a reasonable subsistence compatible with decency and health." That stipulation in the policy would have authorized an equitable provision of public aid regardless of race, a goal which they perceived to be incompatible with southern social relations. Congressional compromises also allowed localities the flexibility to limit eligibility for ADC relative to their resources. Some states continued to use the "ability to earn" criterion as a mechanism for limiting benefits, and in some cases, aiding local labor markets.(21)

The creation of categorical aid to dependent children complicated policy for female-headed families. Congress divided public provision into two categories during its debates over the Social Security Act. Social insurance covered those determined to be "employable" while an extended welfare system covered those people determined to be "unemployables." This policy decision continued to muddy the definition of women's work and fostered a situation in which mother-only families remained outside provisions in employment programs or social insurance, yet had access to only the minimal resources through ADC. The 1939 amendment to the Social Security Act resolved this situation for one group of mothers. Survivors Insurance for Widows of Workingmen secured benefits for the wives and children of male workers in covered industries. This provision of insurance became the "counterpart of the protection of the wage-earner and his aged wife or widow."(22)

The Social Security Act's contributory system, therefore, institutionalized a division between female-headed families. Wives of industrial workers received a social security insurance payment sufficient to maintain the family. There would be no economic means test, no periodic inspections of the home, no rules governing the behavior of the widow or her children. Justification for the grant came entirely from the male earner's contributions, and it was seen as an extension of the social insurance package for those men working in the specific covered industries. The second group of mother-only families received benefits from the categorical relief program of ADC. If one's husband had worked in agriculture, service, or trades that remained outside the umbrella of social insurance, one received no survivor's benefits. If a woman's husband divorced, deserted or separated from her, or if she never married, she must first apply for benefits and then be subject to regulation and scrutiny.(23)

To what degree did the earlier practices, which expected mothers on aid to supplement family income with their wages, continue during these years? To answer that question we need to turn first to the relief programs that characterized New Deal measures well beyond the passage of the Social Security Act in 1935. Several states took years to put these plans in place. As categorical relief was transferred from the federal emergency programs to state and local governments, some areas failed to raise funds for the aged, blind, and dependent children. In general, work projects under the Federal Emergency Relief Administration (FERA) or the Works Progress Administration (WPA) characterized mother-only families as "unemployable" and designated work relief projects predominantly for male-heads of families. Exceptions existed, however.(24)

The line between government work relief, welfare, and low-wage work could be particularly narrow. In Minneapolis, employers' associations frequently pressured the public welfare department to limit eligibility and reduce stipends. In 1936, hundreds of women who received direct relief had their grants cut. An investigation into the welfare board's action revealed that the department had "made an organized effort to force single girls who are on relief to accept jobs as domestics in homes at starvation wages." The opposition charged the state with forcing women to accept work at sub-standard rates.(25) In Virginia, Washington, D.C., South Dakota and New Jersey, local relief agencies closed welfare cases when local agricultural employment or domestic service was available. FERA supported workers' choices to refuse private sector employment when wages were too low, conditions poor, or the prospect of being paid doubtful. At the same time, the agency maintained that available seasonal employment "such as harvesting of tobacco and cotton" should be taken by the able-bodied in lieu of relief.(26) Urban workers also blended direct relief with wage-earning. A study of city workers in 1934 found over one-half of those who used both relief and earnings worked nearly full-time. Service occupations, particularly servants, hotel and restaurant workers, cleaners, and laundresses exhibited higher incidences of this pattern. The combination of private wages and public relief raised the concern that federal funds served as wage subsidies. These incidents anticipated similar charges by civil rights lawyers thirty years later.(27)

The Woodrum Act of 1939 changed the eligibility for the WPA and explicitly excluded anyone who could meet ADC eligibility. This effectively eliminated work programs as an avenue of support for female-headed families at a time when ADC administration was overwhelmed by demand. Women who received ADC probably continued to supplement it when and wherever possible, despite legislative language that discouraged earning.

The intermingling of race and gender ideas with labor segmentation was also evident. A study of the administration of ADC in southern states explained the underrepresentation of African American women on welfare as a result of local culture and attitudes. A field supervisor explained the discrepancy between need and coverage as "a unanimous feeling on the part of the [welfare] staff and board that there are more work opportunities for Negro women.... The attitude that 'they have always gotten along' and that 'all they'll do is have more children' is definite." Referring again to the local southern welfare board, the study reported that local administrators "see no reason why the employable Negro mother should not continue her usually sketchy seasonal labor or indefinite domestic service rather than receive a public assistance grant."(28) Over the next twenty years, other states would expand their expectations on mothers to earn.

Between 1935 and the early 1960s, state governments contested, challenged, and changed the meaning of "parental support" in Aid to Dependent Children. The original provision intended to aid children whose father had died, deserted, or suffered from a long-term disability. Consequently, the language of the act defined eligible children as those who had "been deprived of parental support or care." To the authors of the Social Security Act, this phrase meant that fathers would provide the income and mothers would provide the care. However, parental support came to mean a mother would assume both roles as the pattern of mixing relief and wage-earning crystallized into policy in some locations during the 1940s and 1950s. Louisiana adopted the first "employable mother" rule in 1943, only eight years after passage of the SSA. The state maintained that all capable women with children over the age of seven years who received public aid could be denied assistance when field work was available. Georgia implemented a similar rule in 1952 but lowered the children's age to three years.(29) The federal Bureau of Public Assistance did not support compulsory earning by mothers, but the agency proved unable to control the practice or maintain desired standards of care. Having been given the leeway to set limits on aid, states used criteria such as family size or structure, residency, and "employable mother" rules to reduce public costs and address local political demands. Local control over eligibility also gave governing bodies the flexibility to address taxpayers' complaints and to retain "customary" practices of race relations. The Bureau of Public Assistance responded with additional regulations, but was unable or politically unwilling to challenge the many and varied state departures from federal rules. It has been estimated that "two-thirds of eligible dependent children" remained uncovered by ADC in 1940.(30)

The dual roles imposed upon poor women can also be seen in the competing directions of federal welfare legislation during the 1950s. Federal initiatives attempted to strike an uneasy balance between "strengthening family life" through reinforcement of women's traditional caretaking roles, while encouraging parents "to attain the maximum of self-support." To move toward the first goal, Congress passed a caretaker grant that made a specific provision for "the mother or other relative" in its 1950 Social Security Act Amendments. The grant brought ADC in line with other categorical aid provisions, such as Old Age Assistance or Aid to the Blind, and some promoters hoped that this additional stipend would decrease the need for women to earn.(31)

But efforts "to attain the maximum of self-support" may be found in two restrictive measures passed with the same Amendments. The Notification of Law Enforcement Officers had a direct effect because it required welfare officials to notify the police when a deserted child received aid. The Knowland Amendment legitimated a reduction of public aid during labor force shortages. Although it was directed at unemployment insurance, not ADC specifically, the measure paralled those mentioned above in Minneapolis during the Depression, or in the various counties that closed welfare cases when field work became available. It foreshadowed the ability of states to circumvent federal initiatives.(32)

By 1962 when federal legislation first adopted employment policies, it followed the practice of a majority of states. Thirty-three states had work requirements in their ADC regulations, and many included a provision that could deny aid if employment was refused.

Federal Employment Policy in Aid to Families with Dependent Children, 1962-1972

Most commentators refer to the Family Support Act of 1988 as the turning point in work requirements for women on Aid to Families with Dependent Children (AFDC). That statute placed responsibility for the financial support of children upon both parents. It implemented firm measures to collect child support from fathers, and it provided funds for employment and training. The Family Support Act has been characterized as a profound change in welfare policy that created a new dilemma about the role of mothers. Yet, as this section demonstrates, the federal government made a commitment to mandate earning from women on AFDC much earlier and the courts legitimated requirements to earn. These precedents by the legislative and judicial branches validated the authority of localities to control public costs.(33)

Why did federal policy change during the 1960s? The literature offers three explanations. Social scientists most frequently cite post-war demographic patterns, and argue that ever increasing numbers of female-headed families relied on public support which resulted in a welfare crisis. This occurred at the same time that greater numbers of women entered the workforce at all economic levels. A shift in public values about working mothers corresponded with the rise of mother-only families on AFDC and led to the decline of protective, maternalist sentiments. Yet, even though demographics reshaped post-war welfare, the poor had always been asked to contribute to their support. The shift in social values toward wage-earning for mothers applied more to middle-class women than to poor women. The second explanation places the motor of change in labor regulation. Evident throughout U.S. history, labor controls became a justified practice in the post-war years as local administrators expanded or contracted welfare eligibility in accordance with local labor demand and public budgets. Labor segregation accounted for the disproportionate exclusion of certain groups and fostered racial discrimination. As this article discusses, lawyers challenged labor regulation in the states during the 1960s and 1970s. Yet as a comprehensive explanation of change, labor regulation does not explain the absence or outright failure of work policies in some areas. The third explanation argues that welfare attempted to regulate women's sexual behavior particularly through the use of "suitable homes" and "fit mother" criteria, regulations that cut women off grants if their personal behavior deviated from official rules. Welfare policies clearly tried to regulate women's sexual behavior, however the bulk of evidence demonstrates that the policies sought to reduce the size of public aid rather than control household labor or childbearing.(34)

None of these explanations alone explains the shift in federal policy at the time, and they only marginally address the political context of the change. Furthermore, the sketchy historical accounts of "employable mother" rules mistakingly suggest that these state laws were overturned. In fact, the opposite occurred.(35) This section will demonstrate that by the 1960s support for female-headed families had become politically and economically unfeasible. The needs of states to control their expenses featured prominently in both political arguments and court cases. Federal welfare policy placed wage-earning expectations upon both parents and as such accommodated itself to and expanded upon state "employable mother" rules. While exemptions from work rules still existed when family conditions or child care demands necessitated, these care-based ethics also could be suspended by local decisions. The increase in work requirements for all women receiving AFDC in the decades following World War II had its roots in the political as well as economic struggles of welfare policy.

During the 1960s, the central assumption guiding federal welfare policies was continued economic growth. John Kenneth Galbraith, in his influential 1958 book, The Affluent Society, predicted that the expanding post-war economy provided the opportunity for increased governmental spending in the public sector - spending which could reduce the degree of poverty in America. Michael Harrington described specific sectors of poverty in the U.S. in The Other America published in 1962 and also called for massive increases in federal funding. Influential members of the Kennedy and Johnson administrations lobbied for targeted interventions, and yet, the plans employed relied on fiscal measures. The Council of Economic Advisors designed conservative measures like tax decreases to spur economic growth, and job training - not job creation - to address unemployment.(36)

Social science research, available to policy makers at the time, contradicted the thesis that economic growth alone could reduce the incidence of poverty. Structural barriers to employment emerged during the post-war era that made it more difficult to work one's way out of poverty. The mechanization of agriculture, beginning in the 1940s and continuing for two decades, reduced the need for manual labor in the cotton fields of the South and displaced thousands of African American and white field workers. Thousands of people migrated to the cities of the northeast, upper midwest, and Pacific Coast to fill the wartime demand for labor, but lost those jobs in the post-war demobilization. The aged and children, groups outside the labor force, made up significant sectors of the poor. Policy makers had firm evidence by 1963 that "technological change, shortages of decent paying jobs, and racial discrimination" rested at the core of rising poverty rates, as James Patterson has noted. Yet, the policymakers adopted fiscally conservative anti-poverty strategies and created programs that relied on individual initiative.(37)

William Julius Wilson described the increased rates of poverty for urban racial minorities relative to the effects of migration, economic restructuring, and social isolation. The migration of southern blacks to northern cities during the 1940s and 1950s expanded the population of those cities with disproportionate numbers of unskilled workers, many of whom were young. At first, higher paying industrial jobs existed and later civil rights legislation opened those jobs to blacks, but in about a generation the national economy shifted from one based on manufacturing and industry to one that focused on "higher-order service provision." The inner-city poor typically attained less education, which placed them at a disadvantage for these new jobs. Manufacturing jobs left the industrial northeast and midwest and relocated in other regions of the country or outside the U.S. entirely. The expansion of lower-order service jobs between 1975 and 1985, which could have supplied new jobs to low-skill workers, occurred outside the inner city. Wilson's structural analysis moved the discussion of rising poverty rates away from behavior to focus on larger economic transitions and their relation to poverty rates. His thesis raised questions about the inherent inadequacies of welfare reforms that relied on wage-earning for self-support during periods of declining job availability.(38)

Wilson's study concentrated on male unemployment in inner cities. The increased rates of poverty for female-headed families, which had risen significantly in the post-war decades, would decline, Wilson argued, as the economic prospects of African American men improved and expanded the "marriageable pool." This part of his thesis has been challenged on two specific points. It perpetuated an analysis of women as economically reliant on men, and it consequently obscured the two challenges historically faced by women: their responsibility for child care and their weak position in the wage-economy.

Women were certainly affected by the declining financial fortunes of men, but additional economic trends after the war had an impact. The expanding service sector created a demand for workers in this low-skill, low-wage sector, and women continued to move into the labor market in greater numbers. But the characteristics of these jobs - wage ceilings, absence of benefits (particularly health care and child care), and little correlation between productivity and wage increases - made it difficult, if not impossible, for women supporting families to get ahead.

Federal welfare reform proposals began to discuss openly work incentives for women receiving AFDC in 1962. Over the next thirty years, continuing to date with the Clinton administration's welfare reform proposals, the central premise of these reforms has been to increase women's wage-earning contribution. The 1962 Social Security Amendments allowed states to add unemployed fathers to AFDC benefits, and in so doing provided the first language for work requirements in federal AFDC rules. The Kennedy administration wanted to emphasize services that would enhance both parents' efforts to become self-supporting. It provided funds for the reimbursement of child care expenses to encourage women to work, but the amount allocated remained relatively small. Disincentives remained in the program, in the form of a 100 percent wage deduction from welfare benefits.

By the mid-1960s, the growing number of families receiving AFDC, the rise in the number of never-married mothers receiving assistance, and the increased costs to localities created what academic and popular commentators referred to as a "welfare crisis." States and cities cut welfare budgets to reduce caseloads, but despite these attempts budgets continued to expand. The War on Poverty programs included employment programs, but they solicited primarily unemployed men and teenagers. Neither the employment programs nor the work requirements for mothers proved effective in moving people into self-supporting jobs. By 1967, public support for social programs waned in the context of an expensive war in Vietnam, continued social protest, and unrest in the nation's cities. Politicians found that their constituents were a receptive audience for calls to reduce welfare costs. Their arguments focused on personal behaviors - illegitimacy and a lack of the work ethic - as the source of the enlarged social programs. Wilbur Mills, Representative from the state of Arkansas and chair of the House Ways and Means Committee, played a major role in the campaign to restrict benefits to AFDC and require wage-earning.

The 1967 Social Security Act Amendments attempted to correct the ineffectiveness of the earlier provisions. Those amendments, referred to as the Work Incentive Program (WIN), created a plan for comprehensive services, job training, and employment. Congress authorized funds for job training and child care. WIN's incentive allowed a woman to keep part of her earnings in addition to her AFDC benefit, thus eliminating the 100 percent tax on earnings. Women could keep the first thirty dollars and one-third of any additional amount earned. The measure included stiff penalties, however, and benefits could be cancelled for anyone who refused to participate or refused to accept employment. WIN also allowed exemptions for mothers with children sixteen years or younger in deference to caretaking responsibilities, yet by the end of 1971, Congress passed the Talmadge Amendments which expanded the work component. They required all recipients to register for jobs and lowered the age exclusion to six years of age. States received a mandate to submit fifteen percent of their AFDC families for jobs or training.(39)

The discussion surrounding work requirements in 1967 and 1971 illustrated clearly that lipservice to motherhood aside, the priority in the law was to reduce AFDC rolls and the corresponding public costs of those programs. The House Ways and Means committee report on the SSA in 1967 expressed its concern "over the rapidly increasing costs to the taxpayers" and the expanding numbers receiving public aid. The report linked the program's expansion to greater numbers of female-headed families and an increase in "out of wedlock" children receiving aid. The report also made it clear that the committee members were aware of an "increasingly critical public attention" to AFDC. Yet, WIN had limited success placing AFDC recipients in jobs. Too few people received referrals to work programs, less than two percent obtained jobs, and the majority of those jobs were at low wages. Furthermore, WIN prioritized recipients for training and placement. Fathers received preferential treatment and were thought to be the most able to re-enter the market. Single mothers and married women followed. While it was politically and fiscally imperative to increase work requirements, the federal government left the details to the states.(40)

Federal policy did not seek to macro-manage labor supply, but at the same time it did not enforce rules that would prohibit states from passing work rules. The following discussion of two seminal court cases reveals that in some locations, state and local public welfare administrators managed the size of their programs using "employable mother" criteria to exclude otherwise eligible families. As mentioned earlier, thirty-three states had work requirements in their welfare codes before 1962. A close examination of these two cases: one in Georgia that provided the first constitutional test on employment as an eligibility requirement, and one in New York that eventually led to the Supreme Court decision, illustrates the emerging legal position on the relationship between women's wage-earning and receipt of AFDC. In both cases, the courts reaffirmed a state's right to make wage-earning a condition of aid.(41)

During the late 1960s and early 1970s, civil rights and anti-poverty lawyers presented cases before state courts and the U.S. Supreme Court in an effort to eliminate racial discrimination in welfare policy and to expand the rights of the poor. The Supreme Court decided in favor of expanded legal protection for welfare recipients and overturned several state laws between 1968 and 1973. These cases ended the "substitute father" laws that had required financial support from non-related men, and residency requirements. They also entitled recipients to due process through hearings regarding their case. But state and Supreme Court decisions that involved issues of employment and the control of public costs fared less favorably. The majority of the Supreme Court believed that welfare represented an experiment in "cooperative federalism" and as such required a delicate balance between state and federal powers. Success in these relations depended upon "considerable latitude" granted to the states. Nowhere is that more apparent than in the "employable mother" rules. The first constitutional test of the "employable mother" rule occurred in Georgia in 1968 as part of a larger civil rights legal campaign.(42)

Georgia legislators had historically resisted public provision for female-headed families. The state had the dubious distinction of being one of two states that had not enacted any form of mothers' pensions as late as 1931. By 1938, the state temporarily lost federal matching funds for ADC because it had established a racial quota system. Over the next two decades, African Americans consistently received less access to ADC than whites relative to their population and need. In 1951, the Georgia state legislature passed the first law in the country that denied aid to "more than one illegitimate child of a mother." The following year, the law was repealled on constitutional grounds, but the state legislature implemented a plan that would achieve similar goals. The plan comprised several measures that would establish paternity for all children, enforce "substitute father" liability, enact suitable home rules, and require wage-earning. The "employable mother" rule is the only limitation among these that has not been discussed in the literature.(43)

Georgia's "employable mother" rule required all able-bodied mothers to work when "suitable work" was available, and when they were not needed in the home. This policy had many characteristics of earlier practices in various regions of the country; yet, the Georgia case was distinctive for its subtext of gender and racial expectations regarding employability. Georgia defined "suitable work" as employment in line with a woman's "training and experience." It allowed for individual aptitude, but it also classified types of work appropriate to women workers, for example, domestic service. The rule further defined work as that which kept "with the prevailing wage scale and working conditions in the community." The County Boards of Welfare could use their discretion about "the habits and customs prevalent in the community" to determine if a woman was able to engage in part-time or full-time work. The language of the rule and the flexibility it granted to local boards to make decisions based on existing community "habits and customs" created an opportunity for discriminatory practices in the context of Georgia's previous history of aid to dependent families and the period's racial politics.(44)

In the fall of 1966 and spring of 1967, several African American welfare recipients working with C. B. King, an Albany, Georgia attorney known for his litigation of civil rights cases, and Jack Greenberg, Charles Jones, and Stephen Ralston of the NAACP Legal Defense Fund, challenged Georgia's regulation. The lawyers for the plaintiffs argued that county welfare boards cut black families off AFDC when agricultural work or domestic service was available. The boards assumed the work to be full-time employment, and eliminated any supplemental aid from public funds to those families. The rule violated their constitutional rights, compelled women to leave their young children, maintained "a stable underpaid agricultural labor force," and fostered racial discrimination.(45)

The plaintiffs' case focused on three parts of the "employable mother" rule. First, the rule made work mandatory for any mother who received aid. It stated that she "must accept available 'suitable' employment" if it provided necessary "safeguards" for her children. Second, the rule allowed county welfare boards located in areas with seasonal employment to "designate such periods as periods of full-time employment," and to deny aid to AFDC applicants and close current cases during those months. The fact of actual employment was secondary to the premise that work was available. Third, the rule distinguished between full-time and part-time employment and denied supplementary aid to full-time workers regardless of their need. The defendants in the case, officials of the county and state welfare departments, argued that Title IV of the SSA called for parental responsibility for the support and care of the child. Mothers, they argued, had the same standard of responsibility as fathers. This position collapsed any division of labor between parents and placed the burden of both on the single mother.

The Fifth Circuit Court of Appeals found Georgia's Public Welfare "employable mother" rule valid and within constitutional protections. Women must accept employment when it was available. The case did result in a change of policy on the second and third points, however. Before the case went to court, the state department of welfare revised two parts of the rule. Actual employment replaced presumed seasonal employment and the distinction between full-time and part-time workers was modified. Shortly before the Court of Appeals finding, the plaintiffs added charges against the federal and regional representatives of the U.S. Department of Health, Education, and Welfare who had approved the plan, and filed the case in U.S. District Court. The District Court agreed with the Court of Appeals and supported the work requirement in Georgia's plan. The judges found "there is no federally protected right of [a] mother to refuse employment while receiving assistance and remaining at home with her children." But the District Court did agree that Georgia had violated the Fourteenth Amendment's equal protection clause by denying aid to full-time workers and thus "discriminating on [the] basis of source of income."(46)

The plaintiffs' lawyers succeeded in eliminating discriminatory practices in the receipt of AFDC. The decision allowed women to combine public aid with full-time wages, a benefit previously denied. This had particular importance considering the low wages of the work. But the plaintiffs had hoped to challenge broader premises of welfare law - the "right not to work outside the home" and the segregation of employment patterns.

The issue of work requirements in AFDC reached the Supreme Court in 1973. The case involved public assistance recipients who had challenged New York State's requirement that people must accept jobs as a condition of eligibility. This case followed the implementation of WIN, and the plaintiffs sought to have the federal provisions supercede the state work rules. WIN offered greater support services and broader exemptions from employment. The District Court agreed, but the state appealed the decision.

The U.S. Supreme Court upheld New York State's work program, but the majority and minority reports offer an illustrative comparison of the issues surrounding employable mothers. At the center of the majority decision was the state's right to control work policies as a way to manage costs. Justice Powell, writing for the majority, rejected the claim that WIN superseded state plans. If Congress wanted that to be the case, it could have stated so directly, he argued. Furthermore, a system that gave the federal government sole jurisdiction over work programs presented several problems. The federal government would have "the exclusive manner of applying the carrot and stick" in employment policies. To give the federal government such powers, Powell continued, "could impair the capacity of the state government to deal effectively with the critical problem of mounting welfare costs and the increasing financial dependency of many of its citizens." The opinion affirmed the state's authority to set guidelines requiring self-support, limit expenses to those "most in need," and adopt mechanisms "to cope with the fiscal hardships enveloping many state and local governments." The decision set forward the Supreme Court's belief in the necessity of flexibility in allowing states to manage the fiscal politics of their welfare programs.(47)

Justice Thurgood Marshall wrote the dissent with Justice William Brennan joining. The dissent articulated two major objections: the Justices had misinterpreted Congress' intent in establishing a federal work program, and they granted more extensive powers to states than the Court's earlier decisions had authorized. On the first point, Marshall argued that Congress had specifically designed a program of training, education and employment in the 1967 WIN program. It represented a "carefully coordinated system" built upon a foundation of support services, designed to shape skills which could lead to real jobs and self-support. Congress recognized that the success of such a plan demanded supervision and evaluation, and made such provisions, Marshall continued. It did not envision a system of state make-work programs, such as New York State's, nor that employment would become a condition of welfare. On the second point, Marshall chastised the Justices for their interpretation of state's rights. States had been granted flexibility in the SSA in two areas: the allocation of resources and the establishment of standards of need. The majority opinion erred when it allowed states to propose "additional conditions of eligibility" such as work requirements, Marshall argued.(48)

Marshall's dissent also cautioned his peers and the public about the problems inherent in work programs. He acknowledged that the original intention of ADC had failed in practice, and mothers had often left home to earn additional income. But he questioned the purpose, efficacy, and potential abuses of work programs. What kind of job could be required - work for which one was trained and from which one could derive self-support, or any job designated by the welfare authority? The latter had proved to be the case in state experiences and as ex-director of the NAACP Legal Defense Fund, Marshall was well aware of the way local boards had operated. Yet, in the context of growing impatience with increased public expenditures, rising unemployment, and inflation, the American public expressed overt hostility toward welfare, and the courts reinforced the authority of states to mandate earning by mothers.(49)

Two points may be drawn from this overview of twentieth century policies for impoverished mother-only families. First, no analysis of public aid to mother-only families will ever fully evaluate the development of policy if it only focuses on the relation between states and families. It must add markets to that analysis. Women had to balance wage-earning with welfare receipt even though policy did not provide many resources to do so. The findings presented here challenge the standard interpretations that public aid removed women from the workforce. In fact, the studies from the early twentieth century show a rate of employment among women receiving mothers' pensions at nearly 50 percent, much higher than scholars have previously thought. The evidence also indicates that the practices of the 1910s and 1920s continued through the following decades, although more case studies are needed for the 1930s and 1940s. The New Deal social policies that privileged male workers and identified mothers as "unemployables" because of their home responsibilities contributed to a false dichotimization of poor women's experience. So too did the attitudes of organized labor and social insurance proponents that failed to incorporate mother-only family issues into plans for universal social security. The flexibility granted to local welfare boards allowed officials to deny full or partial grants to women on the basis of their "ability to earn." Local administrators determined eligibility, not through standardized budgets or uniform rules, but in consideration of fiscal management and, in some areas, labor needs. Furthermore, an examination of wage-earning and welfare explains links between race and welfare in specific local economies. Some women, like southern African Americans, had historically been marked as "more employable," and the "employable mother" rules required wage-earning in racially segregated areas of the labor force. The continued use of work rules allowed state policies to operate in tandem with these racially constructed labor practices.

During the 1960s, the issue of wage-earning became more visible at the national level. Civil rights lawyers challenged laws at the same time that city and state officials charged welfare recipients with fraud and castigated them for lacking initiative. Within this context, a federal policy emerged that insisted upon wage-earning requirements for mothers. Its history raises implicit concerns about the success of employment solutions in welfare reform. While welfare has changed in many ways over the century, the barriers women face supporting themselves - locating child care and working in low-wage sectors - have remained fairly constant. The provision of supplemental services may have the greatest long-range impact on self-sufficiency in families, but the additional costs of these programs continue to raise political opposition.

The second point to be drawn from this overview is the significant presence of fiscal politics in shaping programs at the local and federal levels. Politicians, policy administrators, and constituents repeatedly attempted to limit the size and cost of their programs with varying success. By 1981, fiscal politics and budget reduction became so central to federal welfare reform that the Omnibus Budget Reconciliation Act (OBRA) eliminated the "work incentives" built into the policy since the early 1960s. OBRA eliminated deductions for work-related expenses including daycare and gave no incentive for earned wages. The requirements to earn stayed in place, however. At the state level, experimental programs in work training and education have seen periodic cutbacks if not complete reduction of their budgets. Ironically, at a time in American politics when "family values" dominated political discourse, welfare policy expressed those values by eclipsing women's care-giving role in the family. If welfare policy continues to emphasize wage-earning, politicians and planners will need to convince the public of the long-term benefits that additional expenditures in child care, training, and education will bring. They will need to address, finally, the specific challenges presented by the poverty of mother-only families.

Department of History Las Vegas, NV 89154-5020

ENDNOTES

This research received financial assistance from the Nevada Humanities Committee and the UNLV Graduate College. I would like to thank Terry McDonald, James Patterson, Colin Loader and the anonymous readers of this journal for their helpful comments on earlier versions of this article. David Anderson provided helpful research at an early stage of this project.

1. The research on gender and the welfare state has developed into a complex and theoretically varied body of literature. For the most fully developed argument on maternalist social policy, see Theda Skocpol, Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (Cambridge, MA, 1992); and Seth Koven and Sonya Michel, eds. Mothers of a New World: Maternalist Politics and the Origins of Welfare States (New York, 1993). The most helpful overview of Aid to Dependent Children may be found in Linda Gordon, Pitied But Not Entitled: Single Mothers and the History of Welfare (New York, 1994). Recent work on gender and welfare states in Europe offers an interesting comparison to the U.S. on this point. For example, see Alisa Klaus, Every Child a Lion: The Origins of Maternal and Infant Health Policy in the United States and France, 1890-1920 (Ithaca, 1993); Susan Pedersen, Family, Dependence, and the Origins of the Welfare State: Britain and France, 1914-1945 (New York, 1993); and Miriam Cohen and Michael Hanagan, "The Politics of Gender and the Making of the Welfare State, 1900-1940: A Comparative Perspective," Journal of Social History 24:3 (Spring 1991): 469-484.

2. Linda Gordon and Theda Skocpol discuss their different understandings of gender and welfare in Contention 2:3 (Spring 1993): 157, 185.

3. Frances Fox Piven and Richard Cloward, Regulating the Poor: The Functions of Public Welfare (New York, 1971), and "Welfare Doesn't Shore Up Traditional Family Roles: A Reply to Linda Gordon," Social Research 55:4 (Winter 1988): 631-647. Michael Katz, The Undeserving Poor: From the War on Poverty to the War on Welfare (New York, 1989). For analyses of work and welfare since the 1960s, see Mimi Abramovitz, Regulating the Lives of Women: Social Welfare Policy from Colonial Times to the Present (Boston, 1988), 338-342. Roberta Spalter-Roth and Heidi Hartmann have written several pieces on the inadequacies of contemporary work policies as anti-poverty plans. For example, see Roberta Spalter-Roth, Heidi Hartmann, and Linda Andrews, Combining Work and Welfare: An Alternative Anti-Poverty Strategy, A Report to the Ford Foundation from the Institute for Women's Policy Research (Washington, D.C., 1992).

4. Excellent overviews of the history of welfare may be found in James Patterson, America's Struggle Against Poverty, 1900-1935 (Cambridge, MA, 1986); and Michael Katz, In the Shadow of the Poorhouse: A Social History of Welfare in America (New York, 1986). For particular studies on women in poverty, see Linda Gordon, Heroes of Their Own Lives: The Politics and History of Family Violence, Boston 1880-1960 (New York, 1988), and Beverly Stadum, Poor Women and Their Families: Hard Working Charity Cases, 1900-1930 (Albany, 1992).

5. For the origins of mothers' pensions, see Mark H. Leff, "Consensus for Reform: The Mothers' Pension Movement in the Progressive Era," Social Service Review 47 (1973): 397-417; Theda Skocpol, Protecting Soldiers and Mothers.

6. Quote is from Mrs. G. H. Robertson, "The State's Duty to Fatherless Children," Child Welfare Magazine (January 1912): 156-160. A good source on the contemporary debate surrounding mothers' pensions may be found in Proceedings, National Conference of Charities and Corrections (Fort Wayne, IN, 1912).

7. Sophonisba P. Breckinridge, "Neglected Widowhood in the Juvenile Court," American Journal of Sociology (July 1910): 67.

8. U.S. Department of Labor, Children's Bureau, Mothers' Aid, 1931, Publication No. 220 (Washington, D.C., 1933), 2.

9. Grace Abbott, The Child and the State, Select Documents, vol. II (Chicago, 1938), 229, cited in Winifred Bell, Aid to Dependent Children (New York, 1965), 6.

10. Edith Abbott, Women in Industry: A Study in American Economic History (New York, 1910).

11. The problem of insufficient funding and work requirements was discussed in several Children's Bureau reports. For example, see U.S. Children's Bureau, Minimum Standards for Child Welfare Adopted by the Washington and Regional Conference on Child Welfare, 1919, Publication No. 62 (Washington, D.C., 1920), 10; U.S. Children's Bureau, Proceedings of the Conference on Mothers' Pensions, Providence, Rhode Island, June 28, 1922, Publication No. 109 (Washington, D.C., 1922), 11, 24; U.S. Children's Bureau, Standards of Public Aid to Children in their Own Homes, by Florence Nesbitt, Publication No. 118 (Washington, D.C., 1923), 17-20. For a discussion of settlements' racially-based views on working mothers, see Ruth Crocker, Social Work and Social Order: The Settlement Movement in Two Industrial Cities, 1889-1930 (Urbana, 1992).

12. Laws of Illinois, Forty-Seventh General Assembly, 1911, 126; Laws of Illinois, Forty-Eighth General Assembly, 1913, 129.

13. Laws of Illinois, Senate Bill No. 300, 48th General Assembly, 1913, 129.

14. U.S. Department of Labor, Children's Bureau, The Administration of the Aid-to-Mothers' Law in Illinois, by Edith Abbott and Sophonisba Breckinridge, Publication No. 82 (Washington, D.C., 1921), 35-36.

15. U.S. Department of Labor, Children's Bureau, The Administration of the Aid-to Mothers' Law in Illinois, 109-110. The majority of denied pensions were refused on the assessment of adequate means of support. Forty-nine percent were deemed economically sufficient, 37 percent were legally ineligible, and 12 percent were either uncooperative or unfit. Economic sufficiency also provided the single largest reason for removing women from the program once they had been accepted. Cook County, IL, Board of County Commissioners, Charity Service Reports, County Agent Report (1911-1917) and Juvenile Court, Mothers' Pension Department (1918-1927).

16. The study took data from large, medium, and small cities and one rural county. U.S. Department of Labor, Children's Bureau, Standards of Public Aid to Children in Their Own Homes, 17.

17. U.S. Department of Labor, Children's Bureau, Standards of Public Aid to Children in Their Own Homes, 20.

18. Grace Abbott, "Recent Trends in Mothers' Aid," Social Service Review 8:2 (June 1934): 208; Grace Abbott, From Relief to Social Security: The Development of the New Public Welfare Services and their Administration (Chicago, 1941), 183.

19. Grace Abbott, The Child and the State, vol. 2 (Chicago, 1938), 309-312.

20. U.S. Committee on Economic Security, Social Security in America: The Factual Background of the Social Security Act as Summarized from Staff Reports to the Committee on Economic Security (Washington, D.C., 1937), 223.

21. For an overview of ADC, see Winifred Bell, Aid to Dependent Children; and Linda Gordon, Pitied But Not Entitled. Grace Abbott describes the circumstances surrounding the construction of the provision in Grace Abbott, From Relief to Social Security, 279.

22. "Final Report of the Advisory Council on Social Security," 30, Grace and Edith Abbott Papers, Special Collections, Joseph Regenstein Library, University of Chicago, Chicago.

23. Linda Gordon, "Social Insurance and Public Assistance: The Influence of Gender in Welfare Thought in the United States, 1890-1935," The American Historical Review 97 (February 1992): 19-54.

24. Some women did receive work relief, but their participation was restricted by quotas, restrictions on the types of jobs, such as sewing, and a wage scale that remained lower than men's wages. Elizabeth Faue, Community of Suffering and Struggle: Women, Men, and the Labor Movement in Minneapolis, 1915-1945 (Chapel Hill, 1991), 132-133; Nancy E. Rose, "Gender, Race, and the Welfare State: Government Work Programs from the 1930s to the Present," Feminist Studies 19:2 (Summer 1993): 319-342.

Also, see Claudia Goldin, "The Changing Economic Role of Women," Journal of Interdisciplinary History (Spring 1983): 707-733; Ruth Milkman, "Women's Work and the Economic Crises, Some Lessons from the Great Depression," in A Heritage of Her Own, Nancy Cott and Elizabeth Pleck, eds. (New York, 1979), 507-537. Grace Abbott citing a Children's Bureau study of the 1921-1922 depression and its effect on families, in Grace Abbott, From Relief to Social Security, 125.

25. Faue, Community of Suffering and Struggle, 119-120, 154-155.

26. Federal Emergency Relief Administration, Monthly Report, July 1 to July 30, 1935 (Washington, D.C., 1935), 49.

27. Federal Emergency Relief Administration, Monthly Report, June 1 to June 30, 1935 (Washington, D.C., 1935), 9-11.

28. Mary S. Larabee, "Unmarried Parenthood Under the Social Security Act," Proceedings of the National Conference of Social Work (Washington, D.C., 1939), 447-449.

29. Bell, Aid to Dependent Children, chp. 6. For the Georgia case, Anderson v. Burson, 300 F. Supp. 401 (1968).

30. Scholars disagree on the explanations for the Bureau's ineffectiveness. Piven and Cloward argued that the principle of local responsibility which retained significant control at the local level was deliberate and functional to local labor demands. Piven and Cloward, Regulating the Poor, 128-130. Patterson found the federal bureau unable to act alone in the absence of "potent lobbies" or organizations of the poor. Patterson, America's Struggle Against Poverty, 68, 87. Yvonne Zylan's political analysis of ADC in the post-war years suggests that the "backlash" on ADC programs, particularly in the states, might have been tempered had Organized women mounted some political opposition. Unlike organized women's support for mothers' pensions in the early twentieth century, Zylan has found no evidence for support of the ADC program at mid-century. Yvonne Zylan, "Constructing the Patriarchal Welfare State: Aid to Dependent Children and the Politics of Gender, 1945-1960." Unpublished paper in the author's possession.

31. New York Times, 18 August 1950, p. 14, c. 2. Abramovitz believes states pressured the federal government to add the caretaker grant to help relieve some of the costs, Regulating the Lives of Women, 316, 322; Patterson, America's Struggle Against Poverty, 88.

32. The Knowland Amendment effectively allowed states to cut workers' benefits during a labor shortage, even a strike. President Truman and organized labor spoke out against the measure as one that would seriously challenge a worker's right to reject employment at sub-standard wages. New York Times, 12 August 1950, p. 1, c. 4; 29 August 1950, p. 20, c.6.

33. Aid to Dependent Children was renamed Aid to Families with Dependent Children in the 1962 revisions. Family Support Act, Statutes at Large, 100th Cong., 2d Sess., 102 Stat. 2343 (1988).

34. Social scientists have explained the shift toward increased federal work expectations as part of a transition that replaced social workers' influence over policy with that of economists. Irwin Garfinkle and Sara McLanahan, Single Mothers and their Children: A New American Dilemma (Washington, D.C., 1986); Katz, The Undeserving Poor. For labor regulation theory, see Piven and Cloward, Regulating the Poor. For sex regulation theory, see Mimi Abramovitz, Regulating the Lives of Women.

35. The 'employable mother' rule has been included among the state welfare laws overturned by the courts. Instead, as this article discusses, the courts reaffirmed a state's right to require work. Piven and Cloward, Regulating the Poor, 308; and Patterson, America's Struggle Against Poverty, 179.

36. John Kenneth Galbraith, The Affluent Society (Boston, 1958); Michael Harrington, The Other America (New York, 1962); Katz, The Underserving Poor, 91-94.

37. For example, see Nicholas Lemann, The Promised Land: The Great Black Migration and How it Changed America (New York, 1991), 345. The impact specifically on African American women is discussed in Jacqueline Jones, Labor of Love, Labor of Sorrow: Black Women, Work, and the Family from Slavery to the Present (New York, 1985), 235-268, 301-310. Also, see James Patterson, America's Struggle Against Poverty, 136.

38. William Julius Wilson, The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy (Chicago, 1987), chp. 2, especially 33-41; and chp. 4, especially 160-161. 39. Wilson, The Truly Disadvantaged, 39; Patterson, America's Struggle Against Poverty, 175, 179.

40. Report of the House Committee on Ways and Means, Social Security Amendments of 1967, H.R. Rep. No. 544, 90th Cong., 1st Sess. 95-96. On the limits of WIN, see Patterson, America's Struggle Against Poverty, 176; Abramovitz, Regulating the Lives of Women, 332.

41. Anderson v. Burson, 300 F. Supp. 401 (1968); New York State Department of Social Services v. Dublino, 93 S.Ct. 2507 (1973). For a legal review that applies a patriarchal analysis, see Sylvia A. Law, "Women, Work, Welfare, and the Preservation of Patriarchy," University of Pennsylvania Law Review 131:6 (May 1983): 1249-1339.

42. Jack Greenberg, Crusaders in the Courts: How a Dedicated Band of Lawyers Fought for the Civil Rights Revolution (New York, 1994). The court cases are respectively, King v. Smith, 88 Sup. Ct. 2128 (1968); Shapiro v. Thompson, 89 Sup. Ct. 1322 (1969); and Goldberg v. Kelly, 397 Sup. Ct. 254 (1970).

43. U.S. Department of Labor, Children's Bureau, Mothers' Aid, 1931. South Carolina was the other state. Act No. 445, Georgia Laws 1951, pp. 692-693, cited in Bell, Aid to Dependent Children, 71, 35; Georgia, Department of Public Welfare, Public Assistance Manual, Pt. III, Sec V, p. 15 (1952).

44. Regulations, Georgia Manual of Public Welfare Administration, Part III, Section V-C(3) (b) (2), cited in Brief for Plaintiffs (October 12, 1966), p. 11, Anderson v. Burson RG 21, U.S. District Court, No. District of Georgia, Atlanta Division, Box 880.

45. According to Stephen Ralston, King contacted the NAACP Legal Defense Fund. Personal correspondence, Stephen Ralston, Senior Staff Attorney, NAACP Legal Defense and Educational Fund, Inc., December 24, 1993. Complaint, June 5, 1967, Anderson v. Burson RG 21, U.S. District Court, No. District of Georgia, Atlanta Division, Box 880.

46. New York Times, 8 October 1967, p. 38 c. 3; Anderson v. Burson, 300 F. Supp. 401 (1968).

47. New York State Department of Social Services v. Dublino, 93 S.Ct. 2507 (1973), 25122513.

48. New York State Department of Social Services v. Dublino, 93 S.Ct. 2507 (1973), 2522.

49. New York State Department of Social Services v. Dublino, 93 S.Ct. 2507 (1973), 2520.