For 99 percent of human history, life was precarious. For everyone. There were no rich and poor; only poor and not-so-poor. Everyone lived off the land, either by hunting and gathering or by farming. Those unable to care for themselves were nurtured by their blood-clans, or abandoned. There was little or no surplus. In times of drought or other natural disasters many died.

Welfare as a social, rather than only a familial, responsibility is a relatively recent innovation, an outgrowth of powerful social, institutional and technological developments. The rise of nation states and powerful central governments made possible widespread poor relief based on mandatory taxation. The rise of larger cities created populations no longer blood-related. The emergence of the market economy displaced hundreds of thousands of rural dwellers that could no longer feed themselves without having a job. An increase in crime and social instability resulted. Finally, new production-oriented technologies multiplied output, generating a surplus that both widened the gap between rich and poor and created the possibility of distributing part of the surplus as public relief.

The U.S. welfare system is a direct descendant of the Elizabethan Poor Laws, enacted in the late 16^th century and widespread by the 18^th century. Strategic, not moral considerations were behind the Poor Laws. Their objective was to maintain public order and reduce the number of vagrants and "masterless men" wandering the countryside.

The introduction of the Poor Laws sparked a spirited debate that continues no less vigorously 400 years later. Some prominent opponents of public relief argued that it aggravated rather than ameliorated poverty. Thomas Malthus gathered vast amounts of data to prove that populations inexorably grow faster than the food supply.[1] Helping the poor increases the population and thus increases hunger and poverty and social unrest. Malthusians "were disposed to consider all giving as pernicious, to distance themselves from the poor and to allow market forces to effect a wholesome discipline".[2]

Poor relief opponents believed the new theory of evolution strengthened their case. Charles Darwin himself insisted, "It was wrong to ameliorate the condition of the poor, since to do so would hinder the evolutionary struggle for existence." British scientist Herbert Spencer, who coined the term "survival of the fittest," argued that the human world, like the natural world, benefited from the fierce and often deadly evolutionary struggle. If not forced to compete for scarce resources the human species would be weakened. Spencer’s work spawned a new movement called Social Darwinism that repudiated all state assistance to the poor, needy, physically feeble, or infirm.

Some opposed public giving because it undermined private giving. Edmund Burke, for example, argued that forcing people to pay taxes to support the needy deadens the spirit of voluntary giving and thus undermines social virtue.[3]

Still others argued that poor relief, even if well intentioned, worsens the very problems it seeks to reduce: idleness and vice. "A person may be endowed with a heart ‘overflowing with the milk of human kindness’ and be the occasion of much more extensive mischief than the most hardened villain," argued the British Charity Commissioners in 1823, "Such a man by an indiscriminate alms-giving may be the promoter of idleness and beggary, the patron of deception and vice, and so far as he holds out a premium for what is bad, an actual diminisher of the sum of good."[4]

Despite these arguments and their formidable proponents, by the mid 19^th century, many governments has established parish or county-based programs to assist the poor and disabled. These programs had two key features:

1. Relief efforts were designed to encourage recipients to leave the relief rolls quickly. "While relief should not be denied the poor, life should be made so miserable for them that they would rather work than accept public aid," explained the 1834 British Royal Poor Law Commission Report. The conditions of welfare should always be less attractive than the conditions of the lowest paid laborer.[5]

In the United States, local governments established residential poorhouses. "Living conditions were supposed to be harsh enough to convince paupers that only through hard work would they escape the atmosphere of penury," observes one historian. [6]

2. Programs distinguished between the "deserving" and "undeserving" poor. In England, the "visiting" movement, spearheaded by churches and later integrated into government welfare programs, went into the houses and neighborhoods of welfare applicants to evaluate their character. In the United States, " Individuals applied to the overseers of the poor for admission and only gained entry to the poorhouse of they were deemed to be ‘worthy poor’, people not responsible for their poverty and incapable of improving their own life. Children sometimes lived at the poorhouse for short periods, but the overseers usually apprenticed them to a farmer or tradesman."

Welfare in Europe

In the 1880s, Germany’s Otto von Bismarck created the first welfare state by introducing a comprehensive social insurance system consisting of accident, sickness and old age benefits.[7] Austria adopted Germany’s system in 1888 and Hungary in 1891. These became the model for welfare programs in many European countries in the 20^th century.

In Europe many welfare programs are entitlements available to everyone regardless of income levels. In the United States most welfare programs are means-tested. For example, virtually all Europeans have some form of universal health care, universal long-term care for the elderly, and universal access to free pre-school. France, Iceland, Norway and Finland offer child care assistance to everyone with children. Many European countries offer birth and adoption grants, and allowances for families with disabled children that are not means-tested.[8]

Welfare in the United States

In the United States welfare programs are means-tested. These include Temporary Assistance to Needy Families (formerly Aid to Families with Dependent Children), Medicaid, Supplemental Security Income, Food Stamps, the Earned Income Tax Credit, and Section 8 housing.

Indeed, in the U.S. the very word "welfare" has come to have a pejorative connotation as a government handout, an unearned and therefore suspect form of assistance. We do not call entitlement programs like Social Security, unemployment insurance and Medicare "welfare".[9]

Even less likely are we to use the term "welfare" to describe many public subsidy programs that could be described as "reverse means-tested". This occurs where the wealthier receive the greater public assistance. For example, Americans can deduct the interest paid on their mortgage from their taxable income. Those who pay no taxes gain no benefit. Those in the top tax bracket receive the largest benefit.

The first widespread effort U.S. programs for the poor and infirm took place after the Civil War.[10] Between 1911 and 1920, 41 states enacted Mothers’ Aid laws to prevent children from being separated from their mothers and to reduce child labor.

The New Deal

With the coming of the Great Depression in the 1930s many states discontinued their Mothers’ Aid laws due to a lack of resources. The federal government stepped in. The New Deal initiatives took many forms. The least favored was direct cash assistance. Indeed, in his 1935 message to Congress, Franklin Delano Roosevelt declared, "The Federal Government must and shall quit this business of relief…Continued dependence upon relief induces a spiritual and moral disintegration, fundamentally destructive to the national fiber."[11]

The largest New Deal initiatives by far involved direct job creation. In 1933, unemployment hovered around 25 percent. Virtually all of the unemployed were able-bodied men. The Works Progress Administration (WPA), Federal Emergency Relief Administration (FERA), Civil Conservation Corps (CCC) and other federal programs directly employed more than 5 million people.[12] To bolster workers, Congress also enacted minimum wage and maximum hour legislation.

The second largest category of New Deal relief initiatives involved creating a series of safety nets based on the insurance concept. Everyone paid premiums and those who met certain criteria were paid from these funds. The Social Security Act of 1935 created two such social insurance programs – old-age insurance and unemployment compensation.

The third New Deal relief initiative, and by far the smallest, consisted of direct cash payment to children in families without a "breadwinner". Aid to Dependent Children (later changed to Aid to Families with Dependent Children, or AFDC) was an entitlement program, available to anyone who met the eligibility criteria. States set their own eligibility criteria and determined benefits. The federal government provided matching funds. Single parents were considered "deserving poor" in the 1930s because most were widows. Relatively few married mothers worked.

In 1939, widows and surviving children became eligible for a husband/father’s social security benefits. That removed most widows from AFDC, transforming it into a program that primarily served single parents and their children and African American widows who did not qualify for Social Security benefits because farm and domestic workers were excluded from the Social Security system until 1954.[13]

State Eligibility Rules

The AFDC legislation gave states the authority to take into account the "moral character" of applicants when deciding eligibility. Many states inserted "suitable home" provisions that denied AFDC to children on the basis of alleged immoral behavior by their mothers.

Critics charged that these provisions punished children for their mother’s behavior, were a disguise for racial discrimination and inevitability forced destitute mothers into further immorality to support their children. In 1945 the federal government recommended the abolition of suitable home provisions; 15 states refused to drop them.

In the 1950s at least 18 state legislatures attempted to deny AFDC to illegitimate children. Only one effort succeeded. Louisiana specified that any home in which an illegitimate child had been born subsequent to the receipt of public assistance would be considered unsuitable. Approximately 23,000 children were dropped from the state’s welfare rolls that summer; almost all of them were African American children whose parents were in common-law marriages.[14]

In response to the Louisiana law, the Secretary of Health, Education and Welfare issued a ruling that, as of July 1, 1961, states "may not impose an eligibility condition that would deny assistance with respect to a needy child on the basis that the home conditions in which the child lives are unsuitable... Assistance will therefore be continued during the time efforts are being made to improve the home conditions or to make arrangements for the child elsewhere."[15]

In 1962 Congress extended AFDC to children in foster homes and child-care institutions and to two-parent families in which the parents were both unemployed.

The War on Poverty

The 1960s and Lyndon Johnson’s War on Poverty laid the foundation for a major expansion of public assistance. In 1965 Medicare and Medicaid were established. That same year food stamps were expanded from a trial program to a permanent program.[16] In 1972, the third element of public assistance, Supplemental Security Income (SSI) was established. It provided cash assistance to the needy aged and blind and disabled.

Both Medicaid and AFDC require state contributions. Food stamps and SSI do not. In Medicaid, states pay health care providers directly and are reimbursed by the federal government for 50 percent to 80 percent of costs, depending on the state’s per capita income.[17] States are required to provide a certain range of services for certain categories of people.[18] Beyond that states set their own criteria for eligibility, services and payment rates.[19]

For AFDC, states defined eligibility and set payment levels. There was no limit on the duration of the aid. AFDC was an entitlement program in the sense that a person who met the eligibility criteria could not be denied assistance.[20] In 1970 the federal government covered slightly more than 50 percent of AFDC costs. Over the years AFDC payment levels have varied by more than 500 percent from state to state. In 1995, the last year of AFDC, Mississippi paid $120 per month to a family of three with no earnings while Connecticut paid $636.

Between the late 1960s and the mid-1970s the number of AFDC recipients increased from about 6 million to 11 million; the number of food stamp recipients increased from about 1 million to 19 million.[21] Welfare spending became a matter of national debate.[22]

The elimination of racial restrictions on welfare, and moral restrictions that were often used to deny benefits on the basis of race, increased AFDC participation by poor black families, many of which were headed by never-married single mothers.[23] After Daniel Patrick Moynihan’s 1965 report on black families, welfare policy debates increasingly focused on dependency, illegitimate births and divorce.[24] Policy makers increasingly viewed welfare programs as encouraging dependency and idleness.

Poverty, Civil Rights and the Negative Income Tax

Around the same time, many in the Civil Rights Movement were making poverty a civil rights issue. The National Welfare Rights Organization, through direct action and legal strategies, succeeded in increasing welfare benefits and improving the atmosphere in welfare offices across the country.[25] In 1968 Martin Luther King called for a Poor People’s March in Washington patterned after the 1963 civil rights march.[26]

To deal with perceived flaws, in 1969, President Richard Nixon proposed to replace AFDC with the Family Assistance Plan (FAP), a national minimum income.[27] The initiative gained support across the ideological spectrum, from conservative economists like Milton Friedman to liberal economists like James Tobin. Many viewed it as a way to eliminate the welfare bureaucracy, reduce administrative costs and eliminate the invasive and demeaning nature of welfare applications. The program was coupled with a work requirement. However, as Nixon insisted, "…a welfare mother with pre-school children should not face benefit reductions if she decides to stay home. It is not our intent that mothers of pre-school children must accept work."[28].

Congress rejected FAP.[29] It subsequently established Supplemental Security Income (SSI) as a negative income tax (NIT)[30] for a narrower group of recipients. SSI focuses on the poor who are unable to work.

In 1977, President Carter proposed a universal negative income tax with one level of income guaranteed for those not expected to work and a lower level guaranteed for those expected to work. The proposal included guaranteed public service employment as a last resort for those who were expected to work, making it the first welfare reform proposal to guarantee jobs for recipients. As with President Nixon’s proposal, Carter’s would not have required women with children age six or younger to work.[31] The initiative was unsuccessful.

Although Congress was unwilling to establish a universal negative income tax, in 1975 it did establish such a mechanism for workers. The Earned Income Tax Credit (EITC) offered poor wage earners as much as $400 per year in a refundable tax credit. The EITC was raised twice in the 1980s and again in the 1990s to become by far the country’s major anti-poverty program.

Welfare-to-Work

Under Ronald Reagan the welfare rolls declined. Many states tightened welfare eligibility requirements.[32]

In 1981, Congress changed the 1967 rule that allowed recipients to earn $30 a month before losing any benefits and then reducing them by $2 for every $3 earned. After 1981, this was changed to one dollar removed for every one dollar earned. As a result by 1983 nearly 14 percent of all beneficiaries were removed from AFDC rolls. Removing most work recipients from the caseloads increased the percentage of welfare recipients who were not working.

In the 1980s there was increased emphasis on welfare-to-work programs. It was no longer unusual for women, even single mothers, to work outside of the home. By the early 1960s about a quarter of married mothers and over 50 percent of all single mothers worked outside the home. By the mid 1980s employment rates of married mothers exceeded those of single mothers.

The EITC allowed welfare reform policies that emphasized putting welfare recipients into any job, regardless of the low wages, because it raised incomes higher than the level of AFDC in most states.

The 1988 Family Support Act (FSA) introduced work requirements for welfare recipients, along with subsidized childcare and expanded job training. It also allowed people who left the welfare rolls to retain Medicaid.[33] The FSA took effect just as an economic recession began. Many states were unable or unwilling to provide the matching funds necessary to draw down their share of federal money.[34]

The recession swelled the welfare rolls from about 11 million to 14 million. Public dissatisfaction with welfare rose again. Interestingly, by the time the new debate about welfare broke out, AFDC had become a minor part of the anti-poverty effort. From 1970 spending on the poor had increased from about $11 billion to $70 billion, in real dollars. Half the increase was a result of the expanded EITC.[35] The other half was a result of soaring Medicaid costs. [36]

In the 1980s states had begun experimenting with their welfare rules and policies under waivers from federal regulations, which had been introduced through amendments to the Social Security Act in 1982. States began implementing time limit and "work first" policies that required recipients to search for employment in conjunction with or in the place of education and training. California imposed strict requirements for training and work. Florida instituted time limits. Wisconsin made eligibility conditional on numerous expectations for adults and children receiving AFDC. By 1995, 40 states had submitted waiver requests for various policies and some 75 percent of the welfare population was living in states that had gained waivers.

1996 Welfare Reform

In 1996 Congress passed, and the President signed, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA). This legislation dramatically changed the structure and the philosophical orientation of poor relief [click here for breakdown of the vote in Congress].[37]

It replaced Aid to Families with Dependent Children (AFDC) with the Temporary Assistance to Needy Families (TANF) block grant. Under AFDC, families who met eligibility requirements were guaranteed cash assistance, and federal matching funds automatically increased if states spent more on welfare. Under TANF there is no entitlement to cash assistance. States can allocate TANF money as they choose among any programs that aim to accomplish the goals of the block grant: providing assistance to needy families, ending the dependency of poor parents on government aid, preventing and reducing out-of-wedlock pregnancies, and encouraging two-parent families.[38] And the federal payment, in the form of a block grant, is capped. If caseloads increase or if benefit levels rise, the federal payment does not increase. If, on the other hand, caseloads decrease or benefit levels fall, the federal payment does not decrease.

The 1996 law made other changes. Those on welfare are not automatically eligible for and enrolled in Medicaid. They must apply for Medicaid separately. The definition of disability for children under SSI was narrowed. Welfare recipients incarcerated for more than 30 days are denied eligibility. A previous law had made it much more difficult for those with an alcohol or drug abuse problem to gain welfare.

PRWORA restricts legal immigrants’ access to public welfare benefits for the first five years they reside in the country. Food stamps were limited to three months out of every three years for people under 50 who are not raising children. States were allowed to contract more services out to private, charitable and religious organizations or to provide vouchers for services that can be redeemed with a contract organization.[39]

Families are restricted to 60 months of cash assistance in a lifetime (although states may choose to use state funds to extend this time limit). Among each state’s TANF recipients, 50 percent of single parent families and 90 percent of two parent families must meet work requirements.[40]

The block grant format gives states greater flexibility to design and administer their own TANF programs. As a result benefits, eligibility rules and program requirements vary more than ever across states.[41] At the same time it limits the ability of the Department of Health and Human Services (HHS) to regulate states’ programs.

The TANF block grant to states was set at $16.5 billion annually for five years beginning in 1997 and remains at that level today. There is no built-in inflation adjustment but Congress can increase the amount of the block grant when the program is reauthorized.[42]

Each state’s share of the block grant is based on the amount they received from the AFDC program in the mid-1990s. The states are required to continue "maintenance spending" of at least 75 percent of the amount they spent on AFDC and related programs (education and training, job placement, childcare and other services) in 1994.

State spending on cash assistance decreased from 58 percent of TANF expenditures in 1998 to 33 percent in 2002, while the proportion spent on childcare increased from 9 percent to 19 percent and workforce development spending from 7 to 10 percent. States were also able to put aside some money for future spending, which they used to support TANF programs when states began experiencing budget shortfalls in 2001.[43]

From March 1994 to July 2001 welfare rolls declined by 59 percent from their historical high of 5.1 million families to about 2 million families.[44] A heated argument in academic and policy circles has occurred as to whether the unprecedented economic expansion of the late 1990s was the case of this decline or whether it was a result of welfare reform, and whether the decline in caseload reflects an improvement in the situation of poor families or not.[45]

In 2003 the Bush administration proposed converting Medicaid to a block grant program. As with welfare reform, states would receive a fixed annual amount for Medicaid rather than matching funds. A task force assembled by the National Governor’s Association disbanded without coming to an agreement, in effect rejecting the proposal. The administration’s 2005 budget proposal included the option for individual states to convert their Medicaid program to a block grant. New Hampshire announced in February that it will be the first to test a block grant program; the governors of California, Connecticut and Florida have indicated that they are considering the move. In New Hampshire and Connecticut, however, the legislatures have prohibited the governor from acting on the matter without their approval.[46]

Conservatives maintain that we are spending more and more money to address a poverty problem neither as extensive nor as severe as it is represented to be. "The living conditions of persons defined as poor by the government bear little resemblance to the notions of ‘poverty’ held by the general public," write Robert Rector and Kirk Johnson of the Heritage Foundation.[47] A typical American defined as "poor" has a car, air conditioning, a refrigerator, a stove, a clothes washer and dryer, a television and a stereo. Almost half own their homes. Hunger is rare and declining, and the principal nutrition-related health problem among the poor is over consumption, not under consumption. "Roughly a third of poor households do face material hardships… However, even these households would be judged to have high living standards in comparison to most other people in the world."[48]

Conservatives argue that by granting a right to welfare, the state restricts the rights and liberty of those who are taxed to fund that entitlement. David Kelly, author of A Life of One’s Own: Individual Rights and the Welfare State, argues that the rights guaranteed in the U.S. Constitution "reflect the Enlightenment ideas that individuals are ends in themselves and that relationships among people should be voluntary." Expanding these to include the right to food, shelter, education, medical care and retirement benefits "express the idea that clients of the welfare state own the people who produce the wealth on which welfare clients depend."[49]

If the U.S. is to have social welfare programs, conservatives prefer an "older, masculine, paternalistic version of the welfare state" directed toward "those who want to help themselves and need a helping hand."[50] "Fathers want their children to grow up to be self-reliant, self-supporting, and able to cope with a recalcitrant world," explains conservative commentator Irving Kristol. "Mothers want their children to be as completely protected as possible from such a world and to be gratefully attached to them as long as they live".[51]

Conservatives maintain that social welfare policies undermine economic growth, and thereby fail to reduce poverty in the long-term even if they provide some short-term relief.[52]

Conservatives point to the drop in the poverty rate, the rise in single mothers working and the unprecedented drop in welfare caseloads from 1996 to 2000 as proof that the legislation has benefited poor people. The poverty rate among black children has dropped from 41.5 percent in 1995 to 30 percent in 2001 – its lowest point in U.S. history – after 25 years of remaining relatively unchanged.[53] The poverty rate of single-mother families declined from 41.9 percent in 1996 to 33.6 percent in 2001.[54]

For single mothers who left welfare in 1996, average incomes increased from a range of 10 to 40 percent above the poverty level to 50 to 70 percent above poverty after two years off welfare.[55] After four years, the poverty rate among women who left welfare in 1996 declined by about half.[56]

Conservatives say the decline in the number of families was not just the result of economic conditions; welfare reforms played an important role. From 1950 through the mid-1990s there were eight periods of economic expansion but no sustained declines in the number of families on welfare.[57] During the expansion that began in 1982 the welfare caseload grew by 13 percent even as the economy added 17 million jobs over 8 years.[58]

Conservatives maintain that "tough love" welfare policies work. One study comparing declines in caseloads from January 1997 to June 1998 found that states with strong work requirements had average reductions of 50 percent, while those with lenient requirements had average reductions of only 14 percent.[59] The relative vigor of states’ economies had no statistically significant impact on caseload decline

Wisconsin implemented a program that denies benefits to able-bodied adults who refuse to work, cuts benefits for people who don’t show up for their jobs, and provides community service jobs as a last resort to ensure that all welfare recipients work for their benefits. From the early 1990s, when the program was implemented, to 1998 Wisconsin experienced a 90 percent decline in its welfare rolls. [60]

Those states that continue to give benefits to recipients who refuse to work and have weaker sanctions experience reduction rates about 25 percentage points lower than in states with strong sanctions.[61]

Conservatives say entitlement welfare programs undermine work ethics and reduce employment. Given the choice between work and leisure, at the same level of income, people will choose leisure. The Seattle-Denver Income Maintenance Experiment, a controlled experiment on the effects of welfare conducted in the early 1980s, proved this. Every additional welfare dollar led to 80 cents less in earnings from employment.[62]

About half of the 2 million mothers on welfare are idle – increased participation rates and required work hours are needed to encourage productive activity that leads to self-sufficiency.[63]

Conservatives maintain that low work levels by parents are one major cause of child poverty. In half of all poor families with children the adult(s) work less than 1,000 hours per year (full-time work of 40 hours per week, 50 weeks per year is 2,000 hours). Another quarter work less than 2,000 hours per year.[64] The Heritage Foundation reports that in 2001 black families with children in extreme poverty had, on average, only 405 hours of work per year, and 60 percent did not work at all.[65]

Conservatives believe welfare is bad for children. Anne Hill and June O’Neill found that children of mothers with extensive welfare participation have lower cognitive abilities than children of mothers who have never been on welfare.[66] This was true even after adjusting for family income, race and parental IQ.[67]

Conservatives say the second major cause of child poverty, in addition to having parents who don’t work much, is the absence of a father in the home. They believe that welfare spending should put a high priority on encouraging marriage and two-parent households. "Nearly two-thirds of poor children reside in single parent homes; each year an additional 1.3 million children are born out of wedlock," writes Robert Rector of the Heritage Institute. "If poor mothers married the fathers of their children, almost three-quarters would immediately be lifted out of poverty."

Data from the Princeton Fragile Families and Child Well-Being Survey shows that of couples who are unmarried at the time of a child’s birth, the poverty rate is 55 percent among women who remain unmarried. But the poverty rate drops to 17 percent among women who marry the father of their child.[68] In 2001 the poverty rate for black single mothers was 35 percent, but only 8 percent for black married couples.[69]

Conservatives say that welfare encourages out-of-wedlock births, a precursor to long-term dependency. In 1970 less than one in one hundred children lived in households headed by a never-married single mother – today that number is one in ten for all children, and one in three for black children.[70] Three-quarters of government aid to children goes to single parent families.[71] Children from these families are more likely than children from intact families to become involved in crime, to fail in school and to end up on welfare as adults.[72]

"Until welfare is no longer a viable option," says Lisa Oliphant of the Cato Institute, "young women will continue making harmful choices and the country will continue to spend endless amounts of energy and taxpayer money trying to clean up the resulting dependency problem."[73]

Conservatives believe that means-tested welfare programs discriminate against marriage. Since benefits are based on household income the benefits a women and her children receive are reduced or eliminated if she marries an employed man. Thus, conservatives support marriage promotion efforts as part of welfare. Since studies show that day care is bad for children,[74] marriage also allows one parent to care for the children while the other parent works.[75]

Conservatives maintain there is no need for additional federal spending on childcare. Federal funding through the Child Care Development Fund increased nearly 120 percent from $2.2 billion in 1997 to $4.8 billion in 2002. About $940 million in CCDF funds were neither spent nor obligated at the end of 2002. In addition, declining welfare caseloads from 1997 to 2002 have resulted in $59 billion in surplus block grant funds for states. In 2002 alone the amount was $13.4 billion, of which $2.7 billion was neither spent nor obligated at the end of 2002.[76]

Liberals maintain that welfare programs are necessary to help people who, through inadequate opportunity or bad luck, need assistance to lift themselves out of poverty, and that the majority of Americans receive welfare at some point in their lives. Data from the Panel Study of Income Dynamics reveals that two-thirds of Americans use at least one means-tested welfare program at some time between the ages of 20 and 65.[77]

Liberals point out that most people who receive welfare do so for a short period of time. Job loss, illness or divorce can trigger a relatively brief spell of poverty. Half of all families who entered AFDC in 1993 (prior to enactment of time limits) received assistance for less than 20 months; more than one-quarter received assistance for less than four months.[78] For families who entered in 1996, 80 percent used welfare (TANF, food stamps, SSI) for less than 20 months; 40 percent for less than four months.[79]

Liberals maintain that most people want to work and accept welfare only because their circumstances force them to do so. Studies suggest that almost all Americans would choose to work even if an adequate level of income were provided without work requirements.[80] Work force participation of the poor is determined by factors other than the level of welfare payments: inadequate education or training, poor health, welfare programs that penalize recipients for work, etc.[81]

Liberals argue it makes no sense to increase the work requirements for TANF recipients. The problem is not unwillingness on the part of most TANF recipients to find work or inaction by state welfare administrators. The problem is higher unemployment in a weak economy, and the fact that the best-qualified welfare recipients have already left the rolls. Those who continue to receive welfare experience higher levels of poor health, depression, domestic violence and weaker social network support than those who have left or who have never received welfare.[82]

Liberals maintain that the much-used term "welfare Queen" reflects a misunderstanding of the level of welfare. Welfare has never been enough to facilitate "leisure" among recipients and the situation has steadily worsened. State-paid AFDC benefits eroded from $792 for a family of four in 1970 to $435 in 1994 (in 1994 dollars).[83]

Liberals maintain that the 1996 act made caseload reduction, not poverty reduction, the objective. States are rewarded for significantly reducing caseloads whether or not the reduction is linked to an improvement in the lives of former recipients. They note state programs begun under waivers prior to 1996 increased labor force participation among welfare recipients without significantly reducing caseloads. Welfare reform after1996, on the other hand, greatly reduced welfare caseloads but did not affect labor force participation.

Liberals argue that the strong economic expansion of the late 1990s accounted for increased work participation by welfare recipients while the pre-1996 welfare waivers and aggressive investments by states in worker training and education explains the increased work participation in that era. "Increases in work appear to be explained entirely by the strong economy after 1996," explains Rebecca Blank, economist and Dean of the Ford School of Public Policy. "This is consistent with the idea that the 1996 legislation focused much more on getting people off of welfare – through sanctions, time limits, and diversion activities – while the waivers focused more on running strong welfare-to-work programs."[84]

Liberals note that TANF caseloads began to decline in 1994, two years before the 1996 welfare reform act was passed and three years before it began to be implemented in the state, further indicating that the reductions were a result of the economic expansion.[85]

Liberals maintain that reducing welfare rolls is not the same thing as reducing the need for assistance. Over 12.8 million children lived in poverty in 2003 – 1.6 million more than in 2000.[86] Food stamp caseloads increased by 38 percent during this period and the number of children receiving health insurance through Medicaid or SCHIP increased by 4.8 million. But welfare caseloads, nationwide, fell by 4 percent. "The number of jobless women with children not receiving welfare rose by 188,000 in one year," reported the Children’s Defense Fund in 2003, "leaving a record three-quarters of all single mothers without public assistance and causing a sudden surge in extreme child poverty."[87]

Forty percent of the homeless population is now made up of families with children, up from 36 percent in 2000.[88] A recent report from the Princeton/Columbia Fragile Families and Child Well Being Study found that about 60 percent of all groups of mothers who were eligible for TANF, whether or not they received it, reported experiencing material hardship in the previous year, compared with 44 percent of mothers who were not eligible for TANF.[89] Fourteen percent of mothers who left involuntarily and 8 percent who left voluntarily had been evicted in the past 12 months.[90]

Liberals note that a number of states have made it more difficult for the needy to apply for TANF. Others have reduced caseloads by throwing people off them for violating one of a growing number of requirements, like attending meetings. They note that in Georgia, for example, before someone can even begin the application process, she is required to get a form signed by six employers saying that she applied in good faith for a minimum wage job and was turned down. Once enrolled, if she is penalized twice for failure to meet some administrative/work requirements, she is barred for life from seeking state aid. No wonder caseloads in Georgia have dropped dramatically.[91]

In the early 1990s about 75 percent of those who walked into the welfare office to apply wound up on welfare rolls. Today it is closer to 25 percent.[92]

TANF applicants who are "diverted" from participating tend to be those in the most need. They have lower education levels and are more likely to be disabled and have other health problems.[93]

Caseloads are also falling because of increasingly strict state eligibility rules. In 14 states, the entire family loses welfare after the first instance of non-compliance. Twenty-one states eliminate the adult share of grant.[94]

One study concluded that "a large portion" of the decline in the caseload in Texas from 2002 to 2003 was driven by a policy change that terminated assistance to the entire family, rather than just to the adult recipient, when an adult did not meet certain program expectations.[95]

A growing number of poor families and children do not receive TANF assistance. Department of Health and Human Services data show that the percentage of families poor enough to qualify for TANF cash assistance that actually received assistance dropped from 52 percent in 2000 to 48 percent in 2001.[96] Only 33 percent of children living in poverty were receiving TANF in 2002.[97] In the mid-1990s for both families and children coverage was over 80 percent.[98]

Liberals argue that a reduction in caseloads does not necessarily translate into an improvement in the lives of those formerly on the rolls. One 2003 study found the proportion of families leaving welfare who are not employed has risen from 50 percent in 1999 to 58 percent in 2002.[99] Another study found that 50-75 percent of welfare leavers remain poor 2-3 years after leaving the welfare rolls.[100]

Nor does caseload reduction mean that people are finding good jobs. The Urban Institute found that median monthly wages for families that had left welfare were under the federal poverty level for a family of three.[101] A study using national data to track families who leave welfare found a 55 percent poverty rate in the first year after leaving; 42 percent remain poor five years after leaving welfare.[102]

Researchers at the University of Chicago tracked Illinois welfare recipients as they left cash assistance programs. They found that it took on average a year to have earned income close to the poverty level for a family of three ($13,880 in 1999) even for those who had a job at the time they left cash assistance and stayed employed for a year.[103] Very little change was seen in the incomes of those who had jobs when they exited, even though the economy improved over the five-year period considered (1996-1999). Of those who did not have a job when they left cash assistance, 65 percent were still unemployed 4 years later.

Liberals note that programs that focus on "making work pay" rather than simply moving people off the welfare rolls can be more successful. Manpower Demonstration Research Corporation did a study of a 1994 to 1998 pilot program in Minnesota that provided generous rewards for work by increasing the amount of work earnings that are disregarded in determining welfare benefits, mandatory employment services, childcare, and various other services.[104] Sanctions for non-compliance with work requirements were mild, yet participation in the program led to "substantial increases in parental employment and earnings…notable not only for their size, but also because they lasted into the third year of follow-up."[105]

Participants in the Minnesota program experienced reduced poverty, "a dramatic decline in domestic abuse, a modest increase in marriage rates, and, for children, better performance in school and fewer behavioral problems."[106]

Liberals argue that while marriage is often beneficial, welfare recipients often have a very high rate of domestic violence.[107] They argue that exhortations to marry will not be as effective in creating more two-parent families as would increasing the incomes of poor families, through living-wage jobs or welfare assistance. A 2003 study by the Princeton/Columbia Fragile Families and Child Well Being Study concludes: "Generous welfare is positively related to cohabitation, a union status that has been little examined in previous research. Consistent with recent evidence from randomized experiments, we find scant evidence that generous welfare benefits affect marriage."[108]

Liberals point out that when Aid to Dependent Children was begun in the 1930s its purpose was to support single mothers so they could stay home to raise their children. It was not until the 1980s that policy makers began to use AFDC, and now TANF, as a mechanism to move mothers from the home into the workforce on the grounds that it was good for children. But forcing single mothers to leave their children at an early age may not benefit either the mother or the child. Research conducted in the early 1990s indicated that families headed by single mothers do not experience the improvement in well being that two-parent families do after moving from welfare to work.[109]

Liberals argue that if single mothers are to work, more money is needed for subsidized childcare – currently only one out of seven children who are eligible for childcare assistance under federal rules receives assistance.[110]

Liberals argue that because of structural differences between TANF and AFDC, states now bear the financial risk of increases in poverty and unemployment associated with economic downturns. The 1996 welfare reforms replaced a program in which federal dollars automatically increased when states spent more with one in which states receive a fixed amount of federal funds. According to Rebecca Blank, "The need for public assistance is countercyclical, increasing when the economy deteriorates. However, states are not well-suited to finance countercyclical programs since almost all operate under balanced budget requirements."[111]

In both 2001 and 2002 states spent about $2 billion more in TANF funds than they received in federal block grants, drawing down reserves accumulated in previous years. At this rate, states will have will have spent down their TANF reserves by the end of 2004 and will need to make program cuts to balance their budgets.[112] Almost half of the states have already made cuts in their childcare program, imposed waiting lists or are no longer accepting new applicants since January 2001. In addition, states are requiring low-income families to pay more for childcare.[113]

Liberals point out that AFDC/TANF represents less than 1 percent of federal expenditures and that percentage has been declining steadily for more than 20 years. They argue that federal "handouts" to wealthier families dwarf cash payments to the poor. For example in 2002 the mortgage interest tax deduction cost the federal government $66 billion.[114] More than half of the mortgage interest subsidies goes to the wealthiest 11 percent of taxpayers – that is more than the entire Department of Housing and Urban Development budget for low-income housing (which is about $31 billion).

Liberals maintain that spending on social welfare programs can and does decrease poverty but the structure of the U.S. welfare program undermines its effectiveness.[115] One study compared poverty rates in 15 countries before and after income transfers to the poor from 1961 to 1991. It found the reduction in poverty resulting from U.S. welfare programs was much less than every other country except Australia, the U.K. and Ireland. In part this is because the U.S. had the second lowest spending (after Australia). But the U.S. and Canada were almost identical in pre-transfer poverty and government welfare expenditures as a share of GDP, yet the post-transfer poverty rate in the U.S. was nearly double that of Canada. Canada’s cash assistance program was more widely available, its benefit levels are higher relative to the average income, and the child tax credit is available to working and non-working families alike.