There is a culture of poverty which means that many long term welfare claimants lack the capacity to take advantage of employment opportunities even when they arise. The answer, the author claims, lies in a policy of 'help and hassle' which will push them into the labour market. In America, the decline of the family connotes several ways in which parents have become less committed to their children, and to each other. Half of marriages now dissolve in divorce and about a third of children are born to unmarried mothers. Fathers are far less present in the lives of their children than they once were. Also, in many families, parents work more hours than ever before, so children get less attention and supervision than they once did. Some commentators trace the rise in social problems among youth to this.

Is welfare to blame for the decline of the family? Some conservatives, notably Charles Murray, argue that welfare and other social benefits have tempted the poor into dysfunction. They could get support from government if they had babies out of wedlock and refused to work, but not if they got married and supported themselves. If parents neglected their children, they knew government would take over (Murray 1994).

However, the hard evidence for this position is weak. Access to welfare, or the level of benefits, appears to have only a small influence on whether women marry or have children out of wedlock. The decline of the family appears to have deeper roots - in greater cultural permissiveness and perhaps in greater affluence. Today, a single mother often can support herself and her children without a husband, if not through welfare then through employment or help from friends and family.

And government has found no clear way to promote less unwed pregnancy or stronger families. Liberals suggest that to provide better family leave, child care, or other social benefits would somehow strengthen the family - but all these things would make it yet easier for women to raise children on their own, or for parents to neglect their own responsibilities to children. Conservatives say the answer is to take away existing benefits, so that parents will have only themselves to rely on. But in an affluent society with many sources of income, merely reducing government support would change little.

To directly strengthen the family may be beyond the reach of government. Change will hinge mostly on more diffuse changes in the culture to make it less permissive - to promote marriage and stigmatise parenthood outside marriage, as society used to do. Such changes appear already to be occurring in the United States, where the climate on family matters has recently turned more conservative, and divorce and unwed pregnancy are falling.

Public policy can do much more to overcome the other leading problem of poor adults - a failure to work consistently. If there is any way for government to strengthen the family in America, it will probably emerge from current efforts to enforce work among welfare recipients, and also to improve the payment of child support. Below, I lay out the logic behind American work requirements in welfare and then suggest some connections to the family problem.

The work problem and opportunity

In the main, sustained poverty among the working-aged and their children in the United States is due to lack of employment, especially work that is steady and full-time. Around three-quarters of American adults, including single mothers, work at some time during a year, and around half work full-year and full-time. Indeed, the population is more fully employed than ever before in our history. But among the poor as the government defines them, only 42 per cent of adults report any earnings at all in a year, and only 10 per cent of them work full-year and full-time (US Department of Commerce 1998). Among welfare recipients, less than 10 per cent report working at a given moment, although the real level - if unreported work is counted - is perhaps a third. There are working poor - people with large families or limited earnings who remain needy despite working. But the problem of non-working poverty is much larger.

Academics tend to trace non-work and thus poverty back to a set of social barriers that prevent poor adults working - above all, a lack of accessible jobs and child care, but also low wages, racial discrimination, and the disincentives against work set up by welfare itself. But the evidence that such barriers actually prevent work is weak, just like the evidence that welfare creates unwed pregnancy. That is especially true for the long-term poor - those who remain needy for more than two years at a stretch. It does not appear that social unfairness often defeats all employment for the poor, although it does depress the quality of job they can get. Social barriers, that is, cause inequality among workers, but seldom poverty among the nonworkers (Mead 1992).

American policy experience runs parallel. We find that programs that merely give nonworkers better chances to work on a voluntary basis do little to overcome poverty. This includes training for better jobs, work incentives that allow welfare recipients to keep more of their benefits if they work, or even government jobs. Such measures do little to raise work levels. They do not cause large numbers to rush forward and work who were not working before. That is why most experts have abandoned such proposals in America, even on the left.

We have to recognise that today's long-term poverty is seldom the result of steady workers being shut out of jobs or paid starvation wages. Poverty of this structural type was much more prevalent before 1960, when the economy was a lot less generous and systematic bias against racial minorities still existed. Indeed, work levels among the poor ran much higher prior to 1960, before recent equal opportunity reforms and economic expansion, than they do today. Then, the poor were typically needy despite work. Today, they are usually needy for lack of it.

Liberals want to offer new chances to poor people by inventing new social programs. Conservatives want to do it through greater reliance on the private sector. Both sides assume that some change in social structure can liberate the needy to get ahead. Both sides make what I call the competence assumption - that poor adults are able to advance their own self-interest, if not society's. The problem is only to give them chances and incentives that will align their self-seeking with the wider social good. That premise is deeply rooted in the social justice and equal opportunity policies of the past. It was valid as long as most of the poor worked consistently, at least in some job. 

It is no longer valid when many poor adults become dysfunctional - that is, when they cannot achieve even their own self-interest, let alone society's. It is difficult to call failure to work when one is poor anything but dysfunctional. The same goes for failure to get through school, drug addiction, or bearing children that one cannot support. Researchers strain to locate some impediment that would explain these behaviours, for then they would not be irrational. Without stronger evidence for that, additional benefits and incentives to promote work become pointless.

In America as elsewhere in the West, the historic issue between left and right has been the proper size of government. How much should it tax and spend, and how far should it intervene in private society to equalise opportunities and rewards among ordinary people? The left says only bigger government can achieve openness and justice, while the right sees government as the obstacle and wants to cut it back.

But America has learned that merely to change the scale of government does not itself overcome poverty. Large expansions of government aid in the 1960s and 1970s lowered poverty mainly among the elderly and disabled, who are not expected to work. More limited cuts in aid during the 1980s and early 1990s also left working-aged poverty virtually unchanged. That is why America has lately taken more direct steps to move the poor into jobs.

A culture of poverty

To explain serious poverty, we must refer back to what sociologists of the 1960s called the culture of poverty. That is a mindset that I call dutiful but defeated. Studies show that most poor adults have orthodox social values. They want to work, maintain their families, and so on. But they feel unable to do so in practice. They perceive myriad obstacles outside themselves that make them unable to work. These include all the things spoken of by liberal scholars - lack of jobs and child care, lack of training, lack of health care or other support services. But when we inquire how pressing these constraints are, we typically find that they do not actually prevent all employment. Other low-income people somehow overcome them and succeed in working regularly. 

Much more, the problem is a lack of mastery within an individual's personal life - an inability to get organised for work and then to sustain that commitment over time, overcoming problems as they arise. Poor adults typically can find jobs, but they lose them because they fail to show for work regularly, due to health or family problems. Or they lose positions because of conflict with supervisors or co-workers. If the problems were really external, then we would find that benefits to overcome them, such as child care, would have much stronger effects on work levels than they have.

To say this is not to point blame at a personal level. The culture of defeat that surrounds the poor is much a collective force upon them as any lack of opportunity. They are not individually responsible for it. But at the same time, the immediate roots of nonwork usually are not impersonal, so no further social reform, however egalitarian, can overcome it. Even a socialist society would find it had to require the poor to work, because merely treating them fairly does not produce change. This is what even the British Labour Party has discovered, as its New Deal shows.

Inability to function may indeed connect to the opportunity structure, but chiefly well back in time. In America, the root of most of today's serious poverty probably goes back to the historic denials of opportunity for blacks and other minority groups, who today make up a majority of the long-term poor. For blacks, the trouble goes back to the decades of unequal opportunity that followed the defeat of slavery in the Civil War. Jim Crow (the Southern racial caste system) and the exclusion of black workers from the Northern factories took its toll. The majority of blacks kept faith, prepared themselves, and when civil rights came, were ready to advance into the working and middle classes. But many others lost faith in America and themselves. The disillusionment of poor blacks (and, later, Hispanics) caused the family and work levels to crumble in poor areas just when the doors of opportunity opened.

Comparable stories could be told about other countries. The scars of class as well as race figure heavily in the making of dependency throughout the West. But to say this, again, confers no solution. Blighted memories prevent further opportunities in the present from solving the social problem. Today's seriously poor typically do not believe that they will ever have a chance to make it, even if society now wishes them to. Their despair has become immune to social change. Welfare reform must address dysfunction as a behaviour, not because it has no social causes, but because the roots are lost in history where they cannot be reached.

Family and functioning

The family is often the connection between past injustice and present poverty. The opportunity structure appears to have little influence on personal behaviour in the present. Social arrangements affect how people are rewarded if they function, but not whether they function. Rather, the ability to cope stems largely from one's family and upbringing.

Most children acquire a sense of possibility not because society is fair to them but because adults near to them are. By identifying with parents and teachers, they internalise values. By meeting parental expectations, they also derive a sense of mastery that makes them approach life hopefully, without defeating themselves. The wider world has no comparable influence. If parents are effective, children will be well formed even if the surrounding society is unfair. Among blacks, exemplary figures such as Martin Luther King or Colin Powell were the products of an unjust society but also of strong families that upheld demanding standards for them.

Conversely today, a family breakdown has undercut functioning even though society has become much more fair. It is chiefly through dysfunctional parents that the malign influence of past injustice reaches forward to blight our own time. Today's poor adults often neglect or abuse their children because they were mistreated by their parents, as they were by their parents, and so on. Those earlier parents failed, in part, because they were ground down by a hostile or indifferent society. Today society may be more enlightened, providing chances to get educated and get ahead, and to try again if one fails, that were unknown a century ago. But society cannot - without abolishing the family - interrupt the transmission of a heritage of defeat.

Once children leave the family to go to school, there is remarkably little that government can do to change them or to enhance their capacities. By the time they become known to the authorities as problems (for crime or truancy) they may be too impaired ever to take full advantage of the chances actually available to them. Their feeling of hopelessness really arises in the first instance from weak or abusive parenting, even if they project it onto the environment. Thus, no improvement in the wider setting can make them hopeful, because none can undo those early experiences.

The solution lies in rebuilding the family, not society. Government has no easy way to do that, but the best single thing it can do is to restore order in the inner city. It must restrain the behaviour of people who cope poorly, who otherwise would injure themselves and their children, in hopes that the children will function better. It must restore the authority of more constructive authority figures, both parents and others.

This means enforcing the law, but also raising educational standards in the schools and, above all, enforcing work in welfare. It is chiefly by working that parents convey a belief to their children that there is opportunity outside the front door. If welfare parents do not work, no program to help their children is likely to achieve much. To a child, to have functioning parents is worth 25 Head Start programs. Only if parents work and fulfil other civilities, such as obeying the law, can they have the self-respect needed to command the respect of their children.

Work enforcement

American government is enforcing values on all these fronts with rising success, but the most dramatic breakthrough has come in requiring work within welfare. In America, welfare reform mostly means work enforcement. The focus is on the family welfare program now known as Temporary Assistance to Needy Families (TANF). (Programs for the unemployed and disabled have as yet received less reform attention, in part because they are less generous than in other Western countries.)

Traditionally, very few welfare mothers worked, compared to a large majority of mothers not on welfare, including single mothers. Thirty years of research and experience have shown that many more welfare mothers will go to work, or leave welfare, if they are required to enter work programs as a condition of aid. Just participating in mandatory work programs strongly promotes employment, and the effect is greater if the programs emphasise working in available jobs rather than education or training for better jobs. While education and training can help people get better jobs later, it is best in the short run if the nonworkers simply accumulate a work history in jobs they can already get. Since the 1980s, the best programs of this type have shown that employment and earnings on welfare can be raised substantially, despite the social and economic barriers that supposedly prevent work. Mandatory programs outperform voluntary ones chiefly because they cause more recipients to go to work, as against merely raising the earnings of those already working (Mead 1997, ch. 2).

Serious work enforcement in American welfare goes back about ten years. Under the most recent national welfare law, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, the states received more discretion to decide who should be covered by welfare, but they were also required to put even more welfare adults to work than before. Half of the recipients are supposed to be working by the year 2002. The gathering force of work demands, plus a very good economy and many jobs, have driven the TANF rolls down by about half in the last five years - the sharpest fall in the history of program.

Some commentators think low unemployment gets most of the credit, but we should remember that the nation also had a low unemployment rate a decade ago. That was when the welfare rolls began to rise by 30 per cent. What has made the difference today is the growing reality that going on welfare means work. Today, there are many jobs available, as there have been before, but recipients are being driven to take them as never before.

Evidence on the effects of the decline is not all in, but what we know to date is strongly positive. About two-thirds of recipients who have left the rolls are employed, according to state surveys. While some are not, they typically have secured support from friends and family, and active hardship (homelessness, children driven into foster care) is rare. Federal surveys show that work levels among poor adults, including single mothers, are soaring while poverty is falling. The proportion of poor female heads of family rose from 43 to 54 per cent between 1993 and 1997, with most of that coming just in the last year (US Department of Commerce 1993-97). Poverty is falling especially rapidly among blacks, probably because they are a plurality of welfare recipients and have thus been the most impacted by change. 

States differ in how firmly they are enforcing work. Among states with large urban welfare caseloads, Wisconsin has most firmly demanded work as a condition of aid. This state has driven its welfare rolls down by around 90 per cent over the last decade. Yet, according to a recent survey, Wisconsin in 1997 had the highest work level among single mothers and low-income parents and the lowest child poverty rate of 13 states studied (Urban Institute 1999). Nothing but welfare reform could explain this. From what we know now, work enforcement looks like America's greatest success against poverty since the problem first became a national issue almost 40 years ago.

The family connection

Does any of this have anything to do with the family? Very little, one might argue. Requiring single mothers to work for aid infringes the original purpose of family welfare, which was to free them from employment so that they could devote themselves wholly to raising children. Even if one accepts that this goal is impolitic today, requiring work might hurt children. And what does it do to rebuild the family?

But if strengthening the family has not been the main goal of work requirements, that process still has some favourable family effects. Along with other programs built on enforcement, work tests could become the germ of a more constructive family policy, at least for the poor.

Work requirements focus on adults. But we have research that suggests that for a single mother to work is probably good for children too, on average, contrary to the original presumptions of welfare. If a mother works, her children tend to do better in school, controlling for other factors. The reasons probably are that a working mother gives a good example, and her own morale improves, making her a better parent. It is two-parent, not single-parent, families where the child may gain if the mother does not work (Mead 1996). Research on the effects of the recent welfare fall on children is still in process, but it is already clear that for welfare mothers to work is popular, not unpopular, with their children.

The current welfare reform aims much more at raising work levels than at reducing unwed pregnancy among the poor. One reason for that is the weak evidence that welfare is responsible for the problem. Another is that interventions in the family are a lot more controversial among the public than enforcing work. Nevertheless, unwed pregnancy is currently falling in America, and welfare reform is probably one of the reasons. The connection is not that states are cutting off aid to single mothers - they are not - but rather the clear message of personal responsibility that suffuses the welfare reform process. Poor women are being told that if they get pregnant they will have to do more to help themselves, and this causes more of them to avoid pregnancy until marriage. Demands to work impose needed structure on the lives of poor adults (particularly young adults) outside the family, and that brings more order to their personal lives as well.

Some experimental programs have appeared that try to enforce good behaviour in ways beyond employment. One type is programs for teenage mothers that require that they stay in school or attend parenting classes as a condition of aid. It is easier to raise work or school levels than to reduce further pregnancies, but the programs that have the most effect on births are also the most directive, which supports the enforcement model (Besharov and Gardiner 1996). Other programs that combine benefits with requirements - what I call paternalism - have appeared in criminal justice, drug addiction, and homeless policy. Most show promise (Mead 1997).

The work emphasis in welfare supports the family because lack of earnings is itself a major reason why families break up among the poor, or fail to form. Mothers who work are less likely to have to go on welfare than those who do not, and they are more likely to meet and marry men who are working, rather than 'street corner' men who hang around neighbourhoods without working. The potential effects on male work effort are even more important. Failure to provide earnings is the main reason why poor fathers abandon their families, or are kicked out by their spouses. The best single thing social policy could do for the family would be to raise work levels among low-income fathers.

Work requirements in TANF might appear to have little application to fathers, because the program supports mainly single mothers - who are subject to the work requirements, not the departed fathers, who usually do not draw welfare. Actually, TANF does support some two-parent families, and some states have changed the rules to make it easier for such families to qualify. Those fathers, as well as teenagers not in school, are already subject to work tests. More important, the fact that the mother now has to work causes her to put more pressure on the father to work himself, or at least pay child support.

Furthermore, strengthening child support enforcement is a goal of the American reform second only to raising work levels. Efforts to make more absent fathers pay for the families they have helped create go back 25 years. As a condition of federal welfare funding, states have been made to set up special agencies to find the fathers, establish paternity, and then collect support payments. Few low-income fathers pay regularly, in part because they have work problems of their own. That has led to work enforcement efforts for them that parallel the ones in welfare. While few absent fathers can be made to work in return for aid, they can be made to do so on pain of going to jail for non-payment of support. Again, experiments that combine requirements with new support services show promise. The prospect is that better enforcement programs for fathers will be the next frontier for social policy innovation in America.

None of this, admittedly, addresses the family problem other than among the poor. But among the middle class, the problem probably has somewhat different causes. There, parents are probably working too much for the good of their marriages and children, rather than too little, as among the poor. The answer is more likely to come from general shifts in the culture, rather than in changes in social policy.

Enforcement and justice

Poverty would be easy to solve if only resources, and not governance, were required. Both the left and the right naturally search for some way to 'invest' that would overcome the social problem without addressing conduct. For liberals, that means spending more on poor families through welfare or intensive service programs. For conservatives, it might mean special education programs to build up skills, job placement programs, or job creation in the private sector. But the effect of even the best such efforts is limited and long-term. It would be better to invest in high-quality social administration because bureaucracy, unpopular though it is, increasingly must manage the lives of the seriously poor.

What paternalist programs do is mimic the combination of help and hassle for budding citizens that effective families provide. This combination does more to help people function better than either help without requirements or plain severity could do alone. Government must manage lives for only limited purposes - to promote work and the other competences required of citizens. But in so doing, on current evidence, it can also indirectly foster stronger families. As parents are forced to face their responsibilities by the society, they will be more likely to commit themselves more fully to one another. Public expectations must apply the pressure to uphold marriage that the forces of scarcity once did.

Traditionally, social policy sought to promote, not functioning, but social justice. The idea was that society owed some recompense to the needy, whose failure to prosper reflected the pressure of social and economic structures beyond their control. But this presumed that the beneficiaries were already citizens in full standing. And that in turn presumed, tacitly, that they were workers, or lived in a family that had a work history. The very image of welfare as a 'safety net' presumes that individuals are trying seriously to get ahead, to advance and upraise themselves. When they no longer do this, as is true for most of today's seriously poor, then the focus on justice cannot be maintained. The competences assumed by citizenship must first be reaffirmed, above all work. Only then can questions of justice and redistribution get back on the political agenda (Mead 1986).

That process is already occurring in America. Exactly because more welfare recipients are working, society is more disposed to be generous to them. Because they earn more, they also qualify for more help from government, the reverse of what social policy experts expect. State and federal governments are raising welfare payments and other benefits in order to 'make work pay'. The political climate in Washington has shifted to favour, rather than oppose, new social programs. One reason for this is that the booming economy has helped balance the national budget, but another is that today the likely beneficiaries of new benefits look a good deal more 'deserving' than they did a decade ago.

This is the road that the rest of the Western world must also travel. In the short run, welfare reform American-style can look negative, even punitive. But the pay-off is a richer and, above all, a more integrated society.

The moral is that welfare, and the welfare state, cannot only be about rights and claims. Those who claim rights must also have obligations. Those who would be free must first be bound. Parents who would bequeath freedom to their children must first live orderly lives. The chief problem for today's seriously poor is no longer social injustice but the disorders of private life. For these Americans, the way forward is no longer liberation but obligation.

References

• Besharov, Douglas J. & Gardiner, Karen N. (1996), 'Paternalism and welfare reform', The Public Interest, no. 122, Winter, pp. 70-84.
• Mead, Lawrence M. (1986), Beyond Entitlement: The Social Obligations of Citizenship, Free Press, New York.
• Mead, Lawrence M. (1992), The New Politics of Poverty: The Nonworking Poor in America, Basic Books, New York.
• Mead, Lawrence M. (1996), 'Welfare reform and children', in E. Zigler, S. L. Kagan & N. W. Hall (eds) Children, Families, and Government: Preparing for the Twenty-first Century, Cambridge University Press, Cambridge, UK.
• Mead, Lawrence M. (ed.) 1997), The New Paternalism: Supervisory Approaches to Poverty, Brookings, Washington DC.
• Murray, Charles, (1994, 2nd edn), Losing Ground: American Social Policy 1950 - 1980, Basic Books, New York.
• Urban Institute (1999), National Survey of America's Families, Washington DC.
• US Department of Commerce (1993 - 97), Bureau of the Census, unpublished data from the March Current Population Survey.
• US Department of Commerce, Bureau of the Census (1998), Poverty in the United States: 1997, Series P-60, No. 201, Government Printing Office, Washington DC.