Why Welfare Reform Is Working
On July 1, the “end of welfare as we know it” began in earnest. On that day, the federal legislation that President Clinton had signed nearly a year earlier went into effect, terminating a 62-year-old federal entitlement and creating, for the first time, a limit on how long one can receive federal welfare assistance.
In Washington, however, it seems impossible to leave well enough alone. Clinton himself had promised last year to “fix” troublesome portions of the welfare law, and by the end of July, as Congress passed a balanced-budget plan, it became apparent that the law’s implementation was still susceptible of political manipulation. In the final days of negotiation over the budget, a passive Republican Congress and a politically alert White House began diluting the potent formula conceived and signed a year earlier and in effect for all of three weeks.
These eleventh-hour changes are not insignificant. But they should not obscure the larger achievement. The welfare-reform legislation that went into effect on July 1 is the most far-reaching policy move of the Clinton presidency—and also, to date, the most successful. Not surprisingly, the President used his July 4 national radio address to crow about it. Since he took office in January 1993, he announced, three million fewer people were on the welfare rolls. Even more impressive was the fact that an astonishing 1.2 million had come off the rolls in the first nine months since the welfare-reform legislation passed Congress and before it formally went into effect. Using rhetoric that was once the preserve of conservative polemicists, the President told the nation on July 4 that “we have begun to put an end to the culture of dependency, and to elevate our values of family, work, and responsibility.”
In truth, the legislation itself deserves only part of the credit. Earlier this year, the President’s own Council of Economic Advisers concluded that the drop in the number of people on welfare was due in some measure to the healthy economy and also to the wide variety of initiatives that had emerged over the last few years at the state level. We have, indeed, never witnessed such a fertile period of experimentation, with dozens of state legislatures trying new ways to move people off government assistance and onto a path of self-sufficiency. Most of these former recipients have gone successfully into full- or part-time jobs, while others, recognizing the new demands the local welfare office will soon place on them, have voluntarily dropped out of the system. With the more comprehensive measures of the federal law now taking effect—and notwithstanding the deleterious changes introduced in the balanced-budget negotiations—we have every reason to expect that these trends will continue.
Not everyone is rejoicing, to be sure. The hand-wringing among some conservatives over the last-minute changes smuggled in by the White House in late July is one thing; but it pales in comparison to the deep distress which the legislation, amended or unamended, has brought to liberal policy circles, not to mention the real rifts which the President’s support for welfare reform has caused within his own party. For those who have not wanted to hear that the era of big government is over, the welfare-reform bill has been, indeed, a bitter pill to swallow.
That may explain why, from the beginning, those opposed to the plan repeatedly resorted to a kind of demagoguery that was shameless even by Washington standards. Thus, when the first round of legislation began moving through the Republican-controlled Congress in 1994, Senator Daniel Patrick Moynihan boldly predicted that the result would be “scenes of social trauma such as we haven’t known since the cholera epidemics.” Not to be outdone, Senator Edward M. Kennedy called an early version of the reform bill “legislative child-abuse.”
By the summer of 1996, when it was clear that a bipartisan coalition existed for replacing the federal welfare entitlement with state block grants, time limits, and work requirements, still more alarms were set off. The Urban Institute warned that one million children would fall into poverty, the New York Times condemned the bill as “atrocious,” and Moynihan pronounced it “an obscene act of social regression.” Finally, days after the President signed the legislation, two of his top policy appointees at the Department of Health and Human Services resigned in protest. One of them, Peter Edelman, waited less than six months before publishing an article in the Atlantic calling the welfare-reform plan “the worst thing Bill Clinton has done.” As Edelman saw it, the new legislation offered a grim future for America’s poor:
[T]here will be suffering. . . . There will be more malnutrition and more crime, increased infant mortality and increased drug and alcohol abuse. There will be increased family violence and abuse against children and women, and a consequent spillover of the problem into the already overloaded child-welfare system and battered-women’s shelters.
Today, a year after the legislation was signed, there is scant evidence of a looming crisis, yet the predictions of gloom, albeit somewhat less hysterical in tone, have continued to pour in. In the American Prospect, the eminent social scientist Christopher Jencks has forecast that without a well-funded system of public employment, “states will either have to fudge their time limits or let a lot of destitute families break up.” Writing in response to Jencks, Kimberly Christensen has suggested that he is too optimistic. “Many abused women,” she charges, “will eventually lose their benefits for nonattendance at workfare placements, or reach the [time] limit, with disastrous results for themselves and their children.” To the Washington Post columnist E.J. Dionne, echoing what has become a conventional liberal criticism, “the danger of the reforms is not that they will force people to work but that people will be forced off the rolls without work to go to.” And in the August 4 New Republic, a pair of articles warns that the new law, for all its early success, is unlikely to help the most disadvantaged among us.
Quite obviously, no one knows at this early stage what the ultimate effect of welfare reform will be. But the critics, having lost the legislative battle, now seem determined to convince us that without a massive infusion of new federal funds, the nation’s poor are headed pell-mell toward cruel deprivation and suffering. Fortunately for the country, the facts suggest otherwise.
The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 passed both Houses of Congress with considerable bipartisan support. Like all such sweeping pieces of legislation, it makes changes to numerous federal laws and regulations. But the bulk of the legislation is directed at Aid to Families with Dependent Children (AFDC), the Roosevelt-era assistance program that was the target of most of the growing public dissatisfaction with welfare. The new law effectively repeals AFDC and replaces it with a new program known as Temporary Assistance for Needy Families (TANF). In addition, the law introduces four fundamental changes that distinguish it from every attempt at welfare reform that has come before.
First, it ends the federal entitlement to cash assistance. In the past, eligibility for this assistance was means-tested: anyone meeting the income requirements was automatically qualified. Under the new law, each state determines eligibility. Second, the new law gives a block grant to each of the 50 states, permitting it to design a cash-assistance program as it sees fit. Third, the law establishes a five-year lifetime limit on cash assistance and a two-year limit on receiving assistance without working, thus ensuring that welfare cannot become a way of life. Finally, the law requires each state to craft work requirements as part of its welfare program. By the year 2002, states will need to show that at least 50 percent of those receiving welfare are involved in some form of work or training in exchange for benefits.
These changes all come with a catalogue of exemptions, qualifications, and alternative requirements in special cases—a flexibility that guarantees that the actual programs will vary considerably from state to state. The law will not, for example, “throw a million children into poverty.” States can exempt 20 percent of their caseload from the time limit, and also convert block-grant money into vouchers for children after their families have reached the limit. Even when the federal limits are triggered, states can continue to spend their own money helping poor families (as they do now). And states may exempt parents of infants from all work requirements, while single parents with children under six will be asked to work only part-time.
Although one would never know it from the critics, left untouched by this reform are a host of poverty-assistance programs. Medicaid, a program still in need of reform, continues to provide health coverage to all poor families under the new welfare law. Public-housing programs remain in effect, as do child-nutrition programs and the Earned Income Tax Credit. The food-stamp program will continue to grow, if at a slower rate. Again contrary to what has been charged, children with serious long-term medical conditions and disabilities will not lose their Supplemental Security Income aid; the new law merely narrows the definition of “disability” to exclude some purely behavioral problems.
Most of the bill’s critics have also misunderstood the financing behind it. According to the Washington Post, the new law “hands the problem to the states and fails to equip them with the resources to solve” it. In fact, the bill represents a giant windfall for state welfare spending. The block grants replacing the old, formula-driven AFDC payments have been fixed at 1994 spending levels—but in the meantime, as the President reminded us in his July 4 radio address, the welfare caseload across the nation has been dropping dramatically. (In Maryland, Oregon, Massachusetts, Oklahoma, and Michigan, the AFDC caseload has shrunk by 20 to 30 percent in the past two years alone.) For many states, then, the new block grants, financed on the basis of the more crowded welfare rolls of two years ago, represent a significant hike in funding—hardly the outcome one would anticipate from a Republican Congress routinely described as mean-spirited, heartless, and insensitive to the needs of the poor.
But most confounding of all to critics of the bill, and most heartening to its supporters, is the fact that welfare reform, in its embryonic stages, has wildly surpassed expectations. In April of this year, eleven million people were on welfare, the lowest share of the U.S. population since 1970. Nor have any of the widely predicted nightmare scenarios materialized. Even in cities like Milwaukee, where thousands of welfare recipients have dropped off the rolls in the last two years, local shelters and food banks have reported no new surges in demand for their services.
What accounts for these early signs of success? Following the lead of the Council of Economic Advisers, some have suggested that the drop in caseloads is traceable entirely to the current strength of the economy. But this cannot be right. The economy has indeed been strong; yet previous cycles of prosperity have failed to produce anything close to the reductions we see today.
What is different, clearly, is that the rules governing welfare dependence have started to change. Indeed, they started to change well before the federal law was passed last year. Impatient with Washington’s habitual inaction, both Democratic and Republican governors began introducing time limits, work requirements, and rules designed to promote responsibility in their own state systems. The burgeoning economy has made their work easier, but there is no denying that in states where the rules have changed, the lives and behavior of welfare recipients have also changed, and for the better.
Wisconsin’s much-touted reforms are a case in point. In a detailed study published in Policy Review, Robert Rector has shown how two new programs in that state, Self-Sufficiency First and Pay for Performance, fundamentally altered the relationship between welfare recipients and government. Implemented in April 1996, the programs required recipients to work in the private sector or perform community service, attend remedial-education classes, or participate in a supervised job search in exchange for AFDC payments or food stamps. Those who did not want to work, take classes, or look for a job were no longer eligible for payments. Seven months after the programs began, the AFDC caseload had dropped 33 percent.
Recent experience in Tennessee, though less widely reported, is no less impressive. As it happens, Tennessee is not subject to the provisions of the new federal law, having won prior approval for an equally comprehensive program of its own. Like the federal law, the Tennessee plan, known as Families First, replaces AFDC with a cash-assistance program that requires recipients to work, go to school, or train while working part-time. Tennessee exempts almost a third of its welfare recipients from the time limits (and from some of the work requirements), and in that respect its plan is even more flexible than the federal law. On the other hand, Tennessee imposes a tighter restriction on the number of consecutive months welfare recipients can receive cash benefits. Finally, everyone eligible for benefits, even if exempt from the work requirements and the time limits, must sign a “personal-responsibility contract” outlining the steps to be taken toward self-sufficiency.
In the first six months of the program, 19,000 Tennesseans left the welfare rolls—a 21-percent drop, unprecedented in the state’s history. What makes this reduction more remarkable still is that during these early months no one was being forcibly removed from the rolls by an arbitrary cutoff date. Instead, social-service officials in Tennessee discovered that the mere requirement to show up at a welfare office, sign a statement of personal responsibility, and participate in a work or educational program had a dramatic impact on the lives of people accustomed to receiving a government check without anything being asked of them at all.
Tennessee officials broke down the declining caseload to understand what was taking place. The results are revealing: 5,800 recipients asked that their cases be closed within the first month (“I don’t want to be bothered,” was a common response). Another 5,500 found work and earned enough money to make them ineligible. Almost a third either refused to sign the personal-responsibility contract, or failed to comply with its terms, or refused to attend classes or begin a job search. The rest moved out of the state. As for those still receiving cash assistance, many appear to be enthusiastically pursuing a route to independence. In the first six months of Families First, 18 percent of this group had found full-time employment; 22 percent were in training or were looking for a job; 19 percent were pursuing adult education; 6 percent had gotten some form of employment mixed with training.
Tennessee’s record so far vividly contradicts the most prevalent and longstanding liberal criticism of a decentralized welfare system: that it will spur a “race to the bottom” among the states. Harvard’s David Ellwood, who served as an assistant secretary of Health and Human Services and was a point man for the administration’s welfare-reform plans before quitting in frustration, has made this criticism most explicitly:
History is filled with examples of states choosing to ignore poor families or ignoring racial minorities, regions, or types of families. Moreover, if one state’s rules differ markedly from those of another, there will be an incentive for migration. It is a lot easier to move poor people from welfare to the state border than from welfare to work. Needs and resources also differ widely across states. The states with the smallest tax base are usually the states with the greatest proportion of poor children and families. Fearful of becoming “welfare magnets,” some states may cut benefits and impose more punitive measures than they would otherwise prefer.
On almost every point, the Tennessee example has disproved Ellwood and those who repeat his arguments. Tennessee, a relatively poor Southern state, is also no stranger to racial tensions. With no state income tax, a lean state budget, a recent history of political corruption, and a strong Republican tilt in recent elections, it would hardly seem an ideal candidate for meaningful welfare reform. Yet Tennessee’s program has promoted work and independence without suddenly snatching away the safety net. Moreover, as part of its reform initiative, the state legislature has actually increased spending on welfare by 22 percent since 1994. Nor is Tennessee unique in this respect. The New York Times recently reported that, to the surprise of anti-poverty advocates, state legislatures, flush with federal dollars from the welfare-reform bill, have been spending money on day-care services, emergency loans for car repairs, and free subway passes, all designed to make it easier for welfare recipients to find work.
None of this is to say that the bill initially passed by Congress and signed by the President last year was without significant flaws or risks. Undoubtedly, the most egregious and widely publicized clause was the one denying benefits to poor, elderly immigrants who were already legally residing in the country. At a cost of $12 billion, July’s budget agreement has repaired this defect by applying restrictions to future immigrants only, thereby retaining the legitimate purpose of discouraging future welfare-seekers from coming to the U.S. without penalizing new, legal residents.
As for the risks, no one knows how states will manage if regional economies start to fizzle, depriving former welfare recipients of their jobs. Nor can we be sure state programs will discourage new entrants to the welfare system as successfully as they have been moving people off. There is also reason to question whether work requirements will deter out-of-wedlock births, which is bound to be a major criterion in judging their success.
Still, the early stages of state reform have already told us volumes about the American welfare population itself, and this will be of inestimable value in addressing the dilemmas of the future. Over the decades, as the evidence of a crisis mounted, it had become customary to speak of this population as if it were a single class of people exhibiting a uniform set of behaviors, motivations, and responses to public policy. If the last few months have taught us anything, it is that welfare recipients are a varied lot.
At one end of the spectrum are those who are unlikely ever to find or hold a job, or to lead a life free of government support. They are the disabled, the chronically ill, and those with severe learning problems or long-term histories of drug and alcohol abuse. They have often been on AFDC without interruption for many years, even decades; their families may have been on AFDC for generations. No mix of carrots and sticks is likely to move such people from welfare to independence. In virtually every state they will be exempt from welfare requirements—quite sensibly so—and continue to depend on government support.
Then there is the opposite end of the spectrum: the considerable portion of the welfare population, perhaps as many as 30 percent, who are easily moved off the rolls at the slightest prompting. Many may have already been in the workforce at some point or other (or may have been defrauding the system by working and collecting welfare simultaneously); when compelled to comply with tighter rules, they simply stop showing up for their check. Others move readily into the workforce when they need to. Still others, as we have seen in Tennessee, would rather drop out of the system than be held accountable for their daily activities. Although some of these welfare recipients may find themselves falling back onto government assistance after a short period of time, they are among those best-equipped to move off welfare permanently, and a well-designed program may be all the motivation they need for long-term self-reliance.
Between these two groups are those who are neither highly motivated nor completely incapable of self-sufficiency. Almost all the major efforts in the months ahead will be focused on this group. But consider how much more doable the task has become. To critics, most of the success so far is attributable to “cherry picking.” As the New Republic put it dramatically on the cover of its August 4 issue: “Welfare Reform Has Moved More Than a Million People Off the Rolls. But Those Were the Easy Cases.” This, however, is wide of the mark; the fact that the states have a significantly smaller caseload to work with means they can devote more money to each family struggling to get off welfare. Policy-makers have never enjoyed this luxury before.
How should the money be spent? There is, unfortunately, no proven formula. Many states will no doubt invest millions of dollars in job training and preparedness programs. But the evidence that these can lead to success is, at best, mixed. Thus, the Job Training Partnership Act, one of the largest government-funded training programs, has led to some improvement in women’s earnings but very little in men’s. Similarly unencouraging is a recent study of New Chance, a program operating in twelve U.S. cities and designed to help teenage mothers gain self-sufficiency. Although New Chance spent approximately $9,000 per mother on education, training, child care, parenting skills, and health care, the study (conducted by the same corporation that designed the program) found that these young mothers were no more likely than before to find a job, leave welfare, or delay future pregnancies.
More hopeful are initiatives to replace training and counseling with real work, or at least rigorous preparation for work. In a recent book, It Takes a Nation, Rebecca Blank, an intelligent critic of the legislation passed by Congress, concedes that for welfare recipients with minimal job experience, some time at work often proves far more effective than months of training or classroom education. The work environment introduces into their lives a higher level of accountability and stimulates them to acquire more training if they want to move up the ladder. Most important, job placement and work programs shift the relationship between welfare recipients and their local social-service agency. Putting someone in a job is an entirely different mandate from merely certifying his eligibility for cash benefits, and the change is beneficial to both sides.
The most frequent question raised by skeptics is how welfare recipients will find employment. Following the lead of critics like Christopher Jencks, liberals now propose the creation of thousands of public-sector jobs as the only way to absorb the welfare population into the workforce. States that heed their advice, however, will quickly discover that these programs are expensive and extraordinarily difficult to manage. The history of similar efforts at the federal level—most notoriously, the Comprehensive Employment and Training Act (CETA)—confirms the folly of replacing one welfare system with another that goes by a different name.
Nevertheless, there may be a virtue in putting people in some form of temporary public-sector workfare or community service. As Robert Rector has pointed out, Wisconsin has created thousands of such jobs, not for make-work but to establish the principle of pay for performance. The same model is now to be seen in other states and cities, including New York, where more than half of the public-park workers are welfare recipients earning their benefits. Ironically, though, this quintessential New Deal-style program has also enraged liberals. In July, a coalition of New York churches, synagogues, and nonprofit groups, evidently oblivious to the social destruction the city’s existing welfare system has wrought, announced that its members would not offer workfare positions to welfare recipients, on the grounds that the program is tantamount to slavery.
Such moral posturing aside, it is true that mandatory public-work programs are not always ideal: under them, many older, laid-off skilled workers may be forced to do menial labor in order to gain temporary cash assistance while seeking a new job. But the vast majority of welfare recipients do not have the skills or wherewithal to enter the job market on their own—they are predominantly young women who have never held a full- or part-time job—and for them, programs that demand work in exchange for cash harbor a twofold benefit. First, they habituate a welfare recipient to the norms of the workplace: arriving on time, following instructions, working with others, and so on. Second, by transforming welfare itself, they make it a far less attractive option for a young woman who might once have seen AFDC as a way out of the world of adult responsibility.
It cannot be stressed enough that the current round of welfare reform is different from all that have preceded it. In the past, reform initiatives simply added a labyrinth of incentives to what remained, at heart, a system of entitlements. Work programs, counseling, job searches, child care, and transportation subsidies are surely limited tools if the recipient knows that at the end of the day, there will be no penalty for failing to respond to the rules and incentives. And the most able welfare recipients always knew how to “game” the system.
That is why a legal work requirement and a clear time limit for cash assistance are so crucial. Without the certainty of a fixed cutoff date, workfare programs of the past invariably devolved into another form of open-ended government job training that did little to move the trainee into a real job. The key to the current reform is that it promotes self-sufficiency by removing welfare as a long-term alternative.
And that, regrettably, is also where the changes introduced in this July’s budget agreement are likely to do the most damage. A number of state governors have reacted to these changes by charging that the President has effectively undermined the whole thrust of the legislation. “Even Democratic governors are screaming he’s all but killed it,” wrote the columnist Paul Gigot in the Wall Street Journal.
There is much justice in the governors’ complaints. The administration’s $3-billion Welfare-to-Work program, for example, was stuffed into the July budget agreement as a payoff to big-city mayors who had been left out of the welfare-reform process. Federal funding for yet another unproven job-preparedness initiative like this one runs counter to the main intent of welfare reform, which is (again) to require work, not training, in exchange for a government check. By permitting such alternatives to thrive, the administration has succeeded in creating yet more loopholes for welfare recipients—the very thing that has repeatedly undone past efforts at reform.
But the administration’s attempt to roll back or qualify the progress that has been made has taken on an even more disturbing aspect. Both the President and his Department of Labor have begun to insist that all work performed by welfare recipients, even those in community-service jobs, must be treated as “employment” and therefore subject to the panoply of federal labor regulations. Such an interpretation not only runs contrary to 30 years of sensible precedent, but, by asserting a new and intrusive federal role, it has the very real potential to prevent every new state workfare program from getting off the ground. If the administration has its way, more than two dozen federal requirements would be placed on any workfare position, including the payment of minimum wages (and prevailing wages in construction jobs), payroll taxes for employers, workers’-compensation programs, and so on. It would be hard to conceive a greater obstacle in the path of programs that were intended, after all, to help those most unlikely to find work in the private sector.
As the Republican Congress was notably unenthusiastic about fighting off these rearguard actions against a bill it spent two years struggling to pass, it will now be up to state governments to challenge ongoing efforts to exert federal control over workfare. Still, these legal and technical issues, and others like them, are all that remains of the welfare discussion. Which means that the larger debate that began in the early 1980’s is finally over. That debate has been decisively lost by those who wanted to cling to a system with a guaranteed federal entitlement for the poor. Still to be seen is the extent to which the new state policies will succeed; but whatever the final consequences, it is unlikely we will ever return to the old model.
Indeed, together with the defeat of national health-care legislation in 1994, the passage of welfare reform represents the second major body blow to big-government liberalism during the Clinton presidency. Interestingly, some liberals who have chosen to make their peace with welfare reform have taken to arguing that by ending the country’s old and highly unpopular policy, President Clinton has actually set the stage for a revival of liberalism. Mickey Kaus, for example, one of the most persuasive liberal advocates of the new reforms, is now trying to cast them as a defeat for conservative governance. “Liberals,” he wrote in the New York Times last year, “can now rebuild an active government on a more defensible foundation”—because the states, having gained control of the purse-strings, are outdoing even Washington in spending money to help the poor.
But quite apart from the question of whether levels of spending are the right criterion by which to judge the prospects of liberalism, Kaus’s argument misses the essential point. It is a mistake to believe that the welfare debate was ever about the amount of money the country was spending. If, over the last 30 years, billions of dollars had gone into a poverty-assistance program that had actually helped to foster stable families, safe and clean public housing, higher achievement in education, and a reduction in illegitimate births, no one would ever have complained about a welfare “crisis” in the first place. Money alone was never the problem.
Instead, what distinguishes the current reform is that it has forced both federal and state governments to take seriously the idea that welfare policy can deter, or encourage, behavior. The fact that Tennessee will increase welfare spending this year tells us nothing in itself. But the fact that Tennessee now holds parents accountable for their children’s immunizations and school attendance; that it forces teen parents to stay in school and live at home or with a guardian; and that it provides no additional benefits for single mothers who have additional children while on welfare, means that government is no longer indifferent to the way welfare recipients live and raise their children. All this represents a stark departure from the liberalism that has dominated government policy toward the poor for the last three decades.
Changing the way the poor behave may not make them prosperous, and there will always be critics to insist that until poverty is eradicated, no program can claim success. But by eliminating the certainty that one will be paid whether or not one works or seeks work, we have already taken the most important step on the road toward the end of welfare—and of liberalism—as we have known them.