Strategies for Welfare Reform

By Robert Rector

INTRODUCTION

The United States welfare system has failed. Both the public and decision makers increasingly recognize that the current welfare system has harmed rather than helped the poor. Welfare has un- de rmined the American family and promoted long-term dependency. In the states, however, we see welfare reforms designed to reverse these patterns. These re- forms focus on three themes: requiring responsible behavior from welfare recipients as a condition o f receiving benefits; requiring some able-bodied welfare recipients to work in ex- change for the benefits they receive; and finally, altering the welfare incentive structure by reducing the current rewards for non-work and single parenthood and increasing the relative re- wards for work and marriage.

TWO TYPES OF POVERTY

The welfare system is intended to address the problem of poverty in the U.S. But welfare pol- icy must begin with an understanding of two separate concepts of poverty: "material poverty" and "behavioral poverty." Material poverty means, in the simplest sense, having a family income below the official poverty income threshold, which was $12,675 for a family of four in 199 1. To the average man on the street, to say someone is poor implies t hat he or she is malnourished, in- adequately clothed, and lives in inadequate housing. In reality there is little material poverty in the U.S. in this sense generally understood by the public. 1 Today the fifth of the population with the lowest incomes h a s a level of economic con- sumption higher than that of the median American family in 1955. 2There is little or no poverty-induced malnutrition in the U.S. Persons defined by the U.S. government as "poor"have almost the same average level gf consumption o f protein, vitamins, and other nutrients as per- sons in the upper middle class. Children living in "poverty" today, far from being malnourished, actually grow up to be one inch taller and 10 pounds heavier than the average

Robert Rector is a policy analy st on welfare and family issues at The Heritage Foundation. This is his testimony before the Domestic Task F=e of the Select Committee on Hunger, U.S. House of Representatives, April 9, 1992. ISSN 0272-1155. 01992 by The Heritage Foundation.

I Robert Recto r, "How the Poor Really Live- Lessons for Welfare Reform," Heritage Foundation Backgrounder No. 875, January 31, 1992. 2 Robert Rector, KateWalsh O'Beirne, and Michael J. McLaughlin, "How Poor Are America's Poor?" Heritage Foundation Backgrounder No. 791, September 21, 1990. p. 2. 3 Robert Recur, "Food Fight: How Hungry Are America's Children?" Policy Review, Fall 1991, pp. 38-43. Robert Rector, "Hunger and Malnutrition Among American Children," Heritage Foundation Backgrounder No. 943, August 2,1991.

4 child of the same age in the general population in the late 1950s. The principal nutrition-related problem facing poor persons in the U.S. today is obesity, not "hunger"; the poor are more likely to be obese than are other persons in the U.S. Similarly, a "poor" American has more housing space and is less likely to be overcrowded than is the average citizen in Western Europe.5Nearly all of the American poor live in decent housing that is well-maintained. In fact, nearly 40 percent of the households defined as "poce' by the U.S. government actually own their own homes. "Behavioral poverty," by contrast, refers to a breakdown in the values and conduct which lead to the formation of healthy. families, stable personalities, and self-sufficiency. Behavioral pove r ty is a cluster of social pathologies including: eroded work ethic and dependency, lack of educa- tional aspiration and achievement, inability or unwillingness to control one's children, increased single parenthood and illegitimacy, criminal activity, and drug and alcohol abuse. VVhile there may be little material poverty in the United States, behavioral poverty is abundant and growing.

COMPETING APPROACHES TO WELFARE

There are three distinct approaches to dealing with the inter-related problems of materi al pov- erty and behavioral poverty. 1) The first approach, which could be called "liberal," maintains that decreasing mate- rial poverty leads to decreasing behavioral poverty. Thus raising the incomes of the poor through cash, food aid, and housing assi stance will cause an increase in emotional stability, educational success, and so forth.

2) The second approach, which could be called "redistributionist," posits no clear link between raising incomes and reducing behavioral problems. This theory promotes welfare expansion to raise the incomes of the less affluent for its own sake. While this approach focuses initially on dealing with vital needs such as eliminating mal- nutrition, its aims are open-ended. Thus although welfare spending is already more tha n twice the amount needed to eliminate all poverty in the U.S., demands for more spending are as vociferous as ever. Many advocates -of this position be- lieve strongly that income redistribution is a positive goal in and of itself, and seek to use welfare policy as a means of attaining that goal. The more income re- distributed the better.

3) The third approach might be termed "conservative." It rests on the belief that spending on most welfare programs actually has increased behavioral poverty. In partic ular, welfare has led to an increase in prolonged dependency and has under- mined family structure, thereby contributing to increases in other dysfunctional behaviors. The assumptions behind the first, or liberal, approach to welfare policy are decisively refuted by historical experience. Throughout most of the twentieth century the incomes of Americans of

4 Bernard D. Karpinos, Height and Weight qfMilitary Youths (Medical Statistics Division, Office of the Surgeon General, Department of the Army, 1960), p p. 336-35 1. Information on the current height and weight of youths provided by the National Center for Health Statistics of the U.S. Department of Health and Human Services based on the National Health and Nutrition Examination Survey. 5 Rector, "How the Poor Really Live," pp. 12-13.

all social classes have increased dramatically. As noted, after adjusting for inflation, the per ca- pita economic consumption of the least affluent 20 percent of households today excee ds the per capita income of the median income U.S. family in 1955. In 1950, some 32 percent of Ameri- cans were "poor," having incomes below today's poverty income thresholds adjusted for inflation-in 1990, 13.5 percent of the population was poor. Going b a ck further in time, we find that in the late 1920s the median income of American households was $1,606 (or $11,000 in 1990 dollars); in that year half of the population was probably poor by today's standardO Ac- cording to the axioms of liberal welfare po l icy, as incomes in all social classes rose dramatically throughout the century, we should have seen increases in cognitive ability, increases in emo- tional stability, increases in marital stability, and decreases in crime. 7 Instead we have seen the oppo s ite. The fact is that most people alive today had at least one parent or grandparent who was "poor" by the current government definition adjusted for inflation. But most of these individuals were not poor in the sense understood by the general public beca u se, although their incomes were low, their values, disciplines, and behavior were middle class-as were the values they passed on to their children. Merely raising someone's income does not inculcate middle class values and be- havior; in fact, most welfar e programs do exactly the opposite.8

THE KEY WELFARE PROBLEM

Following the liberal and redistributionist approaches to welfare, the present welfare system is designed almost exclusively to raise the material living standards of less affluent Americans. Th e federal government provides cash, food, housing and medical assistance, and other benefits through 75 separate welfare programs. Total federal, state, and local welfare spending reached $225 billion in 1990, excluding all middle class entitlement progra m s such as Social Security and Medicare. This figure was more than twice the amount needed to raise the income of every American above the current poverty income thresholds. But for the general public the real problem with welfare is not merely the rapidly expanding cost, which now absorbs over 4 percent of the entire national economy-but the sense that wel- fare actually harms rather than helps the poor.

6 In the late 1920s an average size household with an income at the national median would have had an i ncome below today's poverLy threshold measured in constant dollars. 7 In one limited respect, higher income did lead to an increase in the average number of years of school attendance throughout the century. In earlier periods, many young people left scho o l to obtain employment and help support their families. Thus in the past we could expect that an increase in family income would increase average number of years of schooling attained by a child. However, at the present time, leaving school to obtain empl o yment to support one's family is scarcely a major cause of the school dropout problem. 8 The belief that raising incomes reduces behavioral poverty is not grounded in historical reality but is instead based on superficial and misleading statistical analys e s. In many studies "family income" is correlated or regressed against behavioral variables; the analysis then shows that school failure, for example, is more likely among children in families with average incomes of $15,000 than in families with incomes o f $25,000 of similar family type and educational standing. The study then concludes that if we give the lower income families an extra $ 10,000 in income we can expect the average school performance of the lower income children to increase to the same leve l as that of the higher income children. Ibis is spurious reasoning. Income correlates closely with other psychological variables such as parents' family background, self control, motivation, and cognitive and interpersonal skills. Those psychological vari a bles generally will not be quantified and included in the regression analysis, but it is these non-measured psychological variables which cause the higher average school perfornumce or higher behavioral stability among children in higher income families, not the bigger monthly paycheck.

The key dilemma of the welfare state is that the prolific spending intended to alleviate mate- rial poverty has led to a dramatic increase in "behavioral poverty." The War on Poverty may have raised the material standar d of living of some Americans, but at a cost of creating whole communities where traditional two-parent families have vanished, work is rare or non-existent, and multiple generations have grown up dependent on government transfers.

HOW WELFARE UNDERMINES WORK AND MARRIAGE

46 Current welfare may best be conceptualized as a system which offers each single msther a paycheck" worth an average of between $8,500 and $15,000, depending on the state. The mother has a contract with the government: She will continu e to receive her "paycheck" as long as she fulfills two conditions:

1) she must not work; and

2) she must not marry an employed male. 10

The current welfare system has made marriage economically irrational for most low-income parents. Welfare has conver ted the low-income working husband from a necessary breadwinner into a net financial handicap. It has transformed marriage from a legal institution designed to pro- tect and nurture children into an institution which financially penalizes nearly all low-i n come parents who enter into it. Across the nation, the current welfare system has all but destroyed family structure in the inner city. Welfare establishes strong financial disincentives, effectively blocking the formation of in- tact, two-parent families . Example: Suppose a young man in the inner city has fathered a child out of wedlock with his girlfriend. If this young father abandons his responsibilities to the mother and child, government will step in and support the mother and child with welfare. If t he mother has a second child out of wedlock, as is common, average combined benefits will reach around $13,000 per year. If, on the other hand, the young man does what society believes is morally correct (i.e., mar- ries the mother and takes a job to supp o rt the family), government policy takes the opposite course. Welfare benefits would be almost completely eliminated. If the young father makes more than $4.50 per hour, the federal government actually begins taking away his income through taxes. The feder al welfare reform act of 1988 will permit the young father to marry the mother and join the family to receive welfare, but only as long as he does not work. I I Once he takes a

9 This sum equals the value of welfare benefits from different progmms for the average mother on AFDC. 10 Technically the mother may be married to a husband who works part-time at very low wages and still be eligible for some aid under the AFDC-UP program. However, if the husband works a significant number of hours per month even a t a low hourly rate, his earnings will be sufficient to eliminate the family's eligibility to AFDC-UP and most other welfare. I I 71be 1988 federal welfare law required all states to establish an AFDC-UP program by October 1, 1990. Prior to passage of the 1 988 welfare law, 23 states did not have an AFDC-UP program and those states were allowed to limit AFDC-UP cash benefits to six months, but were required to continue to provide Medicaid as long as the family was otherwise eligible for AFDC. 71be 1988 law a l lowed states to require full-time participation by one parent in Job Opportunities and Basic Skills (JOBS) program, while mandating that the states require one parent to spend at least 16 hours per week in work activity. States must enroll at least 40 per cent of their AFDC-UP caseload in work programs by fiscal 1994 and up to 75 percent by fiscal 1997 and 1998. Both the

4

full-time job to support his family, the welfare benefits are quickly eliminated and the father's earnings are subject to taxation. The onset of the War on Poverty directly coincided with the disintegration of the low-income family-and the black family in particular. At the outset of World War 111, the black illegitimate birth rate was slightly less than 19 percent. Between 1955 and 1965 it rose slowly, from 22 per- cent in 1955 to 28 percent in 1965. Beginning in the late 1960s, however, the relatively sl o w growth in black illegitimate births skyrocketed-reaching 49 percent in 1975 and 65 percent in 1989. If current trends continue, the black illegitimate birth rate will reach 75 percent in ten years.12 Generous welfare benefits to single mothers directly c ontributed to the rise in illegitimate births. Recent research by Shelley Lundberg and Robert D. Plotnick of the University of Wash- ington shows that an increase of roughly $200 per month in welfare benefits per family causes the teenage illegitimate bir t h rate in a state to increase by 150 percent. 13 Similarly, high benefits discourage single mothers from remarrying. Research by Dr. Robert Hutchens of Cornell Univer- sity shows that a 10 percent increase in AFDC benefits in a state will cause a decrease in the marriage rate of all single mothers in the state by 8 percent. Welfare programs discourage young men and women from marrying and promote the disintegration of existing two-parent families. Penalizing Work. For the poor, another devastating legacy o f the past 25 years has been the dramatic reduction in work effort. For a growing number of poor Americans, the existence of generous welfare programs makes not working a reasonable alternative to long-term employ- ment. During the late 1960s and early 197 0 s, social scientists at the Office of Economic Opportunity (OEO) conducted a series of controlled experiments to examine the effect of wel- fare benefits on work effort. The longest running and most comprehensive of these experiments was conducted between 1971 and 1978 in Seattle and Denver, and became know as the Seat- tle/Denver Income Maintenance Experiment, or "SIME/DIME." Advocates of expanding welfare had hoped that SHVIE/DIME and similar experiments con- ducted in other cities would prove that gener o us welfare benefits did not adversely affect work effort. Instead, the SIME/DIME experiment found that every $1.00 of extra welfare given to low- income persons reduced labor and earnings by $0.80.14 The results of the SIME/DIME study are directly applica b le to existing welfare programs: Nearly all have strong anti-work effect like those demonstrated in the SIMZ/DlME experiment. The effects of welfare in undermining the work ethic are readily apparent. In the mid- 1950s nearly one-third of poor households were headed by an adult who worked full time throughout the year. Today, with greater welfare benefits available, only 16.4 percent of poor families are headed by a full-time working adult.

requirement for AFDC-Up coverage and the work requirement end Septe mber 30, 1998. 12 U.S. Department of Health and Human Services, National Center for Health Statistics. Note: The black illegitimate birth rate is available only from 1969 on. The pre- 1969 black illegitimate birth rates were calculated using the very simi l ar "Non-White" rate. 13 Shelley Lundberg and Robert D. Plotnick, "Adolescent Premarital Childbearing: Do Opportunity Costs 1&ner?. " June 1990, a revised version of a paper presented at the May 1990 Population Association of America Conference in Toronto, Canada. 14 Gregory B. Christiansen and Walter E. Williams, "Welfare Family Cohesiveness and Out of Wedlock Births," in Joseph Peden and Fred Glahe, The Anzerican Fandly and the State (San Francisco: Pacific Institute for Public Policy Research, 1986). p. 398.

Inter-Generational Dependence. Of the over 4 million families currently receiving assistance through Aid to Families with Dependent Children (AFDC), well over half will remain dependent for over ten years, many for fifteen years or longer. 15 Depe ndency on welfare also appears to spread from one generation to another. Children raised in families that receive welfare assistance are themselves three times more likely to be on welfare than other children when they become adults.16 Ibis inter-generati o nal dependency is a clear indication that the welfare system is fail- ing in its goal to lift the poor from poverty to self-sufficiency. Effects of Family Disintegration. Ile collapse of family structure has crippling effects on the health, emotional stab i lity, educational achievements, and life prospects of low-income children. Children raised in single-parent families, when compared to those in intact families, are one- third more likely to exhibit behavioral problems such as hyperactivity, antisocial be h avior, and anxiety. Children deprived of a two-parent home are two to three times more likely to need psy- chiatric care than those in two-parent families. 17 And they are more likely to commit suicide as teenagers. Absenci of a father increases the proba b ility that a child will use drugs and engage in criminal activity. I Because the father plays a key role in a child's cognitive development, children in single- ar- ent families have lower 1Qs and score less well on other tests of aptitude and achievement . 1? Children in single-parent families are three times as likely to fail and repeat a year in grade school as are children in two-parent families. In all respects, the differences between children raised in single-parent homes and those raised in intact h o mes are profound, and such differences persist even if single-parent homes ar -parent homes of exactly the same in- _,%compared with two come level and educational standing. But the greatest tragedy is that family instability and its attendant problems ar e passed on to fu- ture generations. Children from single-parent homes are far less likely to establish a stable married life when they in turn become adults. VAAte women raised in single-parent families are 164 percent more likely to bear children out of w edlock themselves; they are I I I percent more likely to have children as teenagers. If these women do marry, their marriages are 92 percent more likely to end in divorce than are the marriages of women raised in two-parent families. Sim- ilar trends are found among black women.21

GROPING FOR A SOLUTION

By nature, Americans believe that all problems have solutions. Therefore, American politi- cians and the public have difficulty believing that there are no easy solutions to the

15 David Elwood, Targeting 'Would-be" Long-term Recipients ofAFDC (Washington, D.C.: U.S. Department of Health and Human Services, January 1986), p. 5. 16 M. Anne HUI and June O'Neill, Underclass BehaWors in the United States: Measurement and Analysis of Determinants (New York: Ci t y University of New York, Baruch College, March 1990). 17 Dr. Deborah A. Dawson, "Family Structure and Children's Health and Well-being: Data From the 1988 National Health Interview Survey on Child Health," paper presented at the Annual Meeting of the Pop u lation Association of America, Toronto, May 1990, Table 5. 18 Nicholas Davidson, -The Daddy Dearth," Policy Review, Winter 1990, p. 43. 19 Marybeth Shinn, "Father Absence and Children's Cognitive Development," Psychological Bulletin, Vol. 85, No. 2 (1978) . pp. 295-324. 20 Dawson, op. cit.; Davidson, op. cit. 21 Irwin Garfinkel and Sara S. McLanahan, Single Mothers and Their Children: A New American Dilemma (Washington, D.C.: The Urban Institute Press, 1986), p. 3 1.

anti-marriage, anti-work incentives provided by the current welfare system. But no easy solu- tions exist. In the current public debate there are a number of quick fixes to welfare which fall short of true reform-the most common of these is the current liberal drive to encourage work and re - duce dependency by "making work pay." Under these proposals, the key to welfare reform is to ensure that all single mothers will be financially better off in the job market than on welfare. While a step in the right direction, there are two problems with this idea. First, the average wel- fare mother receives around $11,000 per year in welfare benefits plus Medicaid. Thus the mother must obtain a job with medical coverage paying more than $11,000 per year (or $5.50 per hour) in order to be even slightly b e tter off with a job than on welfare. Second, even if every mother could be guaranteed of obtaining a job with medical coverage paying say $7.00 per hour, the fi- nancial incentives for taking a job would remain slight. Example: if a mother gives up welfar e benefits worth $11,000 per year plus Medicaid and takes a full-time job with medical coverage paying $14,000 per year (or $7.00 per hour), she obtains an annual post-tax income increase of about $2,500 in exchange for working 2,000 hours during the cours e of the year. This is an effec- tive pay rate of $1.25 per hour. The AFDC mother is expected to make a very large increase in labor for very little, if any, financial reward. A similar recommendation is to reduce the disincentives to marriage by raising t h e earnings ca- pacity of low-income fathers. While this would be another step in the right direction, it would not eliminate the anti-marriage effects of conventional welfare. Even if the earnings capacity of all low-income fathers were raised to the poin t where every working father could provide a standard of living for hisfamily higher than the standard of living welfare provides to single mothers- low-income mothers and fathers would still be better off financially if they avoided marriage. The economic logic of welfare is simple and cruel. If a mother and father do not marry, their joint income is the value of welfare benefits for the mother plus the father's earnings. If they do marry their joint income equals the father's earnings alone. Another way o f expressing this di- lemma is that the welfare system imposes an extraordinarily high marginal tax rate (i.e. income loss rate) on the act of marriage. If a man earning $ 10,000 per year marries a mother on welfare their joint income (including the value of the welfare benefits) will fall by some 50 percent. If a man earning $20,000 marries a mother on welfare, the couple's joint income win fall some 30 percent.

PRINCIPLES OF REAL REFORM

There is an emerging consensus on the need to change the welfare in centive structure. How- ever, many current proposals fall short because they are limited to adding small new rewards for constructive behavior on top of the present welfare system while ignoring the huge rewards for idleness and single parenthood already e mbedded in that system. Serious welfare reform must not only provide new incentives for positive behavior, it must also reduce the huge rewards for destructive behavior that exist in the current system. Welfare reformers have four basic tools available to restructure the welfare incentive system. The first two reduce the rewards provided for non-work and single parenthood, while the third and fourth increase the rewards for marriage and work. The four policy tools are:

1) Reduce welfare benefits to non-working single mothers. This is particularly im- portant in states where the combined value of welfare benefits for the average AFDC family greatly exceeds the federal poverty income thresholds.

7

2) Require able-bodied welfare recipients to work or perform community service in exchange for benefits received.

3) Increase the financial rewards available to low-income working families, both married and single, relative to the rewards to non-working parents on welfare by providing tax relief to low-income working families.

4) Increase the financial rewards availabl e to low-income working families, both married and single, relative to the rewards to non-working parents on welfare by providing tax cred its and vouchers for medical insurance to working families who currently lack medical coverage.

Of these four tools for reforming the welfare incentive system, the work requirement is the most important. Under the current welfare system a non-working singl e mother receives an in- come from the government for free; if she becomes employed she must give up all or part of this free income. However, if the welfare recipient is required to work in exchange for benefits, a new cost is attached to welfare depende n ce and the attractiveness of welfare relative to employ- ment is greatly reduced. Indeed, if the work requirement can be coupled with other government policies which ensure the family will be somewhat better off financially when the mother is em- ployed t h an when the family is on welfare, then the anti-work incentives of welfare would be utterly eliminated. However, as long as the welfare recipient has the option of receiving a size- able income from the government without work, then it will be impossible t hrough other means to reduce significantly welfare's anti-work incentives. Surprisingly, a work requirement also eliminates the anti-marriage incentives of the current welfare system. Under the current welfare system, when a single mother marries a fully e m- ployed male she loses most of her welfare benefits. Under a welfare system with a work requirement, a single mother still would lose her benefits upon marrying-but she would now be losing benefits which she had to earn rather than a fi-ee income, so th e loss would be far less sig- nificant. As long as the mother could obtain a private sector job which paid roughly as much as welfare, then marriage would no longer impose a significant financial or personal cost on the mother or her prospective spouse. In d eed, if required to work for welfare benefits, some welfare mothers would prefer to marry and be supported by a husband's income rather than enter the labor force. By converting welfare from free income to income which must be earned, a work re- quirement eliminates most of welfare's anti-marriage incentives and would make marriage economically rational once again for millions of low-income parents. While few states have attempted to establish serious work requirements for AFDC parents, those experimental p rograms which do exist indicate that work requirements can have a signifi- cant impact in reducing welfare dependency. As part of a workfare program operated on an experimental basis in six Ohio counties, AFDC mothers were required to perform community se r vice for twenty hours per week. While only 25 percent of all AFDC mothers were required to participate, the work program reduced overall number of families on AFDC by some 12 percent. In other words, for every hundred mothers that were required to work in exchange for benefits, over forty mothers left welfare entirely.22

22 Bradley R. Schiller and C. Nielsen Brasher, "The Effects of Saturation Workfare on AFDC Caseloads,"

8

While a serious work requirement will change the welfare incentive structure an d reduce de- pendency, it is also vital for other reasons. Society should provide aid to those in need. But aid which is merely a one-way handout is harmful to both society and the recipient. Such aid under- mines the individual's ability to take responsi b ility for his or her own life. If the habit of dependence becomes entrenched, it limits the individual's capability to become a fully function- ing member of mainstream society. Welfare is currently a check in the mail with no obligations. This is wrong. Instead, welfare should be based on reciprocal responsibility: Society will provide assistance, but able-bodied re- cipients will be expected to contribute back to society in exchange for the benefits they receive.

TOWARD COMPREHENSIVE WELFARE REFORM

What is needed is a comprehensive welfare reform strategy. Many elements of comprehensive reform can be implemented at the state level; however, state actions should be complemented by tax relief and an overhaul of the U.S. medical system at the federal level . Although tax policy and medical reform are formally outside the welfare system, reforms in these areas will have a significant impact on the opportunities and behavior of low-income families, and therefore are an important part on any welfare reform stra t egy. Comprehensive welfare reform must strike a balance between two key themes. First, it must seek to increase the rewards for work and marriage among low-income families. Second, it must reduce the incentives currently provided by welfare for non-work a n d single parenthood. Re- forms which fail to follow this balanced approach will be unsuccessful. Comprehensive reform would have seven parts: 1) Reduce Benefits. Welfare benefits for families on AFDC should be reduced. This is particu- larly true in state s with high benefits levels. AFDC recipients are eligible for benefits from nearly one dozen major welfare programs. In all but five states, the combined value of benefits received by the average AFDC family exceeds the federal poverty income threshold. Mo r eover, there is considerable inequality in welfare benefit levels within each state. Because some fami- lies receive aid from many programs, they will have overall benefits much greater than other welfare families of the same size and characteristics with i n the state. Example: AFDC families that also receive housing aid will have overall benefits some $4,000 to $5,000 higher than other AFDC families within the state. In almost every state such families will have combined welfare benefits well above the pov e rty threshold. States should reduce AFDC payments to families that also receive housing aid. 2) Require Work in Return for Benefits. States should require some but not all welfare recip- ients to work in exchange for benefits received. Recipients of Food Stamps and General Assistance who are not elderly and not disabled and who are not directly caring for children should be required to perform community service for at least twenty hours per week. 23 Within

unpublished paper, November 199 1. 23 Requiring so meone to perform community service means that they would perform useful functions in government or in non-profit private sector organizations. Community service is also called "work experience." Many legislators argue that they would like to require welfa r e recipients to work in for-profit private sector jobs. but this expectation is unrealistic because few private sector employers are willing to employ pawns who literally have to be forced to work. However, requiring the recipient to perform community ser v ice with a government organization removes the recipient's option of receiving welfare income without labor. The work requirement makes welfare less attractive relative to employment and will thereby induce many recipients to take. real private sector job s.

9

the AFDC program, mothers who do not have children under age five or who have received AFDC for over five years should be required to perform community service for at least 35 hours per week in exchange for benefits. In all two-parent families recei ving AFDC, one parent would be required to work. For all programs the work requirement should be permanent, lasting as long as the individual or family receives benefits. This policy specifically exempts most mothers with pre-school children from the work require- ment. Because of the high costs of providing day care, work requirements for mothers with pre-school children would almost certainly increase rather than cut welfare costs. Moreover, great caution should be exercised toward any policy which separ a tes young children from their mothers, as this will often have a significant negative effect on the child's development. Thus a well designed work program generally would not include mothers with young children; how- ever, a second rule requiring work fro m mothers who have received AFDC payments for over five years, either continuously or in separate periods, is needed to discourage mothers from inten- tionally having added children in order to avoid their work obligation. If a work requirement of the sort outlined here were established, roughly 50 percent of AFDC mothers would be required to work as a condition of receiving benefits. This would be an enor- mous improvement from the present situation; in the average state only 6 percent of AFDC mothers curr e ntly participate in job search, work, or training programs. 3) Require Responsible Behavior. States should require responsible behavior as a condition of receiving welfare benefits. This would include policies such as insisting that unmarried minor mother s reside with their parents or in some other adult supervised setting, and reducing pay- ments to mothers who fail to provide their children with free immunizations. Most important, mothers who bear additional children while they are already receiving welf a re should not receive an increase in welfare benefits. 4) Enforce Education Requirements. States should rigorously enforce the current federal law requiring all AFDC mothers under age eighteen who have not completed high school or passed a GED to attend s c hool. To avoid the negative affects of separating infants from their mothers, however, mothers with infant children should not be required to participate more than twenty hours per week. 5) Experiment with Wedfare. More mothers leave AFDC through marriage than through em- ployment. States should experiment with "wedfare" programs which provide bonuses to AFDC mothers who marry, leave AFDC and remain off the welfare rolls. However, since the real ef- fects of wedfare programs are uncertain, any such program should be rigorously evaluated through controlled experiments. 6) Provide Tax Credits or Vouchers for Medical Coverage to All Working Families. The current welfare system which provides free medical coverage to single parents and non-working two-parent fa m ilies on AFDC, but does not provide medical assistance to low income working families, discourages both work and marriage. The federal government could reduce the anti- work/anti-marriage effects of welfare by enacting the comprehensive medical reform pro p osed by The Heritage Foundation in A National Health System for America. 24 This plan would, among other reforms, provide federal tax credits and vouchers for the purchase of medical insur- ance to low-income working families not eligible for Medicaid. A proposal similar to the Heritage plan recently was introduced by President Bush.

24 Stuart M. Buder and Edmund F. Haislmaier, eds.,A National Health SystemforAtnerica (Washington, D.C.: The Heritage Foundation, 1989).

7) Provide Tax Relief to All Families with Children. The federal government currently im- poses heavy taxes on low-income working families with children. A family of four making $20,000 a year currently pays $3,780 in federal taxes. 25 This heavy taxation promotes welfare dependence by reducing the rewards of work and marriage relative to welfare. A crucial step in welfare reform is broad family tax relief along the lines proposed in The Heritage Foundation's A Prosperity PlanfQr America: How to Strengthen Family Finances, Revive the Economy, and Balance the Budget. 26 This plan would provide a $ 1,000 tax credit for each school-age child in a family and a $1,500 tax credit for each pre-school child; the tax credits could be used to reduce the family's inco m e tax liability and both the employee and employer share of the Social Secu- rity payroll tax. The effect of this plan -would be to eliminate all federal taxes on working families with children with incomes below 120 percent of the poverty threshold. The revenue loss of these tax credits would be offset by corresponding spending constraint through capping the growth of total federal domestic spending at 5 percent per annum.Thus the plan would not add to the federal deficit.

CONCLUSION

Any attempt to refo rm the current structure of public welfare must begin with a realization that most programs designed to alleviate "material" poverty lead to an increase in "behaviorar'pov- erty. While the poor were supposed to be the beneficiaries of War on Poverty's tra n sfer programs, they instead have become its victims. If policy makers fail to recognize or respond to this relationship, thewelfare state will continue to worsen, rather than improve, the lives of America's poor. The rule in welfare as in other government programs is simple: You get what you pay for. For over forty years the welfare system has been paying for non-work and single parenthood and has obtained dramatic increases in both. But welfare which discourages work and penaliz@s marriage is a system whi c h ultimately harms its intended beneficiaries. Comprehensive welfare reform must combine toughness and refusal to reward negative behav- ior with positive rewards for constructive behavior. Reforms which fail to include both sides of this equation will no t succeed.

25 Figures are for 1991. 26 Scott A. Hodge, ed., A Prosperity Planfor America: How to Strengthen Family Finances, Revive the Economy and Balance the Budget (Washington, D.C.: The Heritage Foundation, 1992).

}}