The 1996 welfare reform legislation—the Personal Responsibility and Work Opportunity Reform Act (PRWORA)—dramatically eliminated a central component of the New Deal social policy framework. By replacing the Aid to Families with Dependent Children “entitlement” program with state block grants to provide Temporary Assistance to Needy Families (TANF), it ended the federal government’s long-standing commitment to helping needy children. This new law expires in October 2002. The emerging debate over its reauthorization presents the nation with a fundamental policy choice: should we tinker with the new welfare policy at the margins, or should we introduce a fundamentally different approach to the nation’s broader social policy framework.

We think that there is a compelling case in favor of a different approach. All the major proponents of the 1996 changes (from the Clinton Administration to key Republican leaders) agreed that the success of “welfare reform” would not be determined by the number of people left on the welfare rolls, but by improvement in the economic conditions of poor Americans. By this standard, PRWORA has failed. Moreover, marginal reforms are not going to remedy those failings. Welfare policy is really only one part of a much larger issue, and the basic problem we face is not how to reform the welfare system, but how to promote the long-term well-being of all American households, including the welfare poor and the millions of vulnerable working families only an income level or two above them. To address this broader issue, we need some new ideas about economic security—in particular, policies that emphasize asset development. Whereas welfare policy has traditionally focused on income transfers to people who are not earning income in the market, we need instead to build an alternative policy framework that ensures that all Americans have the assets they need to succeed economically.

The Politics of Welfare Reform

The 1996 policy change was the answer to a perceived “welfare problem,” situated at the intersection of work and family, responsibility and reward, race and the proper role of government, federal and state. By the mid-1990s, national opinion polls indicated public support for fundamental change. As a general matter, Americans recognized that individuals’ dependence on welfare had roots in both larger socioeconomic issues and individual circumstances. They accepted that the government has a duty—although perhaps not for an unlimited time—”to care for those who cannot care for themselves.” The public may have thought that the government spent too much money on welfare, but probably because it was not spending the money well. People were less concerned about the size of the rolls than about recipients making an effort to help themselves. They wanted to move recipients—even mothers of young children—to work, but were more concerned about giving recipients the skills they needed to be “self-sufficient,” than about saving tax money. Indeed, they expressed a willingness to pay more taxes for supports such as child care and transportation. Public opinion also endorsed time limits and work requirements, but paired them with support for the provision of public sector or other jobs and exemption from the limits where no job was assured.

Public opinion, then, supported reform, but did not determine its precise contours. Instead, the 1996 reforms reflected a specific political landscape. In 1996, welfare rolls had only recently ebbed from their historic peak, so concerns about welfare were running high. Moreover, in 1992, Presidential candidate Bill Clinton found political advantage by promising to “end welfare as we know it.” After the 1994 “Contract With America” congressional elections, Clinton confronted a Republican congress that would make him pay dearly if he did not keep his promise. To the dismay of erstwhile supporters, he sacrificed job guarantees, training, and other supports that had been his condition for agreeing to impose work obligations and time limits. At the same time, many states had already received regulatory waivers before Clinton took office, which permitted them to run their systems with increased freedom from federal direction.

Apart from this political conjuncture, the debate in 1996 was also driven by years of organized, sustained, and effective attacks on the rationale for welfare and the outcomes ascribed to it. High out-of-wedlock birth rates, it was argued, demonstrated irresponsibility in sexual matters and social mores. Low rates of child support among single-parent families indicated a failure of parental responsibility. Long-term receipt of welfare benefits became associated with a desire to sit at home rather than to make a productive contribution at work. Some critics even argued—largely contrary to empirical evidence—that welfare was self-defeating, and even spurred family break-ups. These arguments were given special cogency by race-based assumptions about who were the chief beneficiaries of welfare. In short, critics argued that dramatic change was needed, and the answers were best left to the states.

PRWORA was the result, but not an inevitable one. It represented a political triumph of those in power at the time. A different, more thoughtful outcome was possible, and entirely compatible with public attitudes. This is particularly regretable because by most indications PRWORA has failed to realize the most basic goals of its architects.

Did Welfare Reform Work?

Has the 1996 welfare reform been a “success”? The answer, using four different criteria—each focused on outcomes attributed to reform—seems clearly to be “no.”

1. Caseload numbers have dropped, but those numbers were falling long before PRWORA became law (in August 1996). Between January 1996 and June 2000, the caseload dropped steadily from 4.6 to 2.2 million families, roughly the level of 1970. But the decline in the rolls began earlier: the numbers fell from an historic peak of 5.0 million in January 1994 to 4.6 million in January 1996. Moreover, the percentage of eligible families who actually enrolled and received benefits also fell continuously from 84.3 in 1995 to 55.8 in 1998.

2. With respect to employment, many people who left the welfare rolls (“leavers”) worked some after their departure, but often at sporadic jobs and typically at inadequate rates. On one national estimate, 50 percent of 1996 and 1997 leavers worked in the first month after departure, a figure in the low range of estimates for leavers during the prior, pre-reform decade. At the state level, the numbers of leavers who found work in the first three months after departure ranged from 47 to 67 percent; for the first year, between 62 and 75 percent. Moreover, the employment for leavers was not very stable. Only about 40 percent of the group worked at some time during each of the four quarters of the year after exit. Fewer worked during every month, and still fewer every week. Only about 12 percent are estimated to have worked 35 hours or more on a weekly basis. Moreover, estimates of the percentage of welfare leavers who return to the rolls within the first year after departure range from 20 to 40 percent.

3. With respect to earnings and overall income, gains are quite limited. At the national level, median wages for 1997 leavers were estimated at $7.08 per hour (in 1999 dollars); the median monthly earnings for their families was estimated at just $997. The wages, earnings, hours of work, and work schedules of the 1999 leavers showed little improvement. State studies show median quarterly earnings, even for those who worked, ranging from only about $500 to $1,000 per month. Relatively few had employer-sponsored health insurance; many had no insurance at all. One study of 1997 leavers suggests that only 29 percent had improved their average monthly personal income (during their first post-welfare year) more than $50 by leaving welfare. Even when account was taken of the income of all members of the leavers’ households, fewer than 50 percent showed such improvement. Net of taxes and work expenses, the number are probably even lower. These figures are consistent, incidentally, with reports of significant food insecurity and other forms of material hardship among these post-welfare families.

4. Finally, the most important criterion for evaluating the success of PRWORA is the long-term economic outlook for former welfare households. By this standard, welfare reformers have little positive to claim. A Wisconsin study of 1995 leavers over three years, for example, showed that although 88 percent worked at some point during that period, fewer did in successive years; and for many, employment was not consistent. Median annual earnings (of those who worked) increased, but their median aggregate earnings over the three years were only about $25,000 (or about $8,400 per year). Moreover, over one third received temporary assistance (TANF) benefits during the three years and nearly two-third received Food Stamps. After three years, about 59 percent of such families had after-tax income, combined with cash assistance, below the poverty level; for 89 percent, it was below 150 percent of the poverty level.

Even during the nation’s strong cycle of economic growth in the late 1990s, welfare recipients who moved from welfare to work were likely to do little or no better than low-income mothers who had already joined the work force: their long-term prospects were little better; time limits aside, many would cycle through welfare as they had done before; and the vast majority continued to find it very hard (nearly impossible, without continuing supports) to earn enough money to move above the official poverty line. In short, welfare reform did not succeed in improving the long-term economic prospects for poor households.

The Larger Issue: Economic Insecurity

Persistent problems after the 1996 reforms suggest a need for a different and more comprehensive solution. The welfare issue cannot be treated on its own because it is part of a much larger problem concerning the well-being of the working poor—and of those who work yet struggle to stay in that nebulous area between poverty and barely “middle class.”

Shared circumstances now place a broader segment of the population at risk. For example, having a job, even a full-time job, hardly guarantees an escape from poverty. More of the poor (on the official poverty measure) had a full-time worker in the family in 2000 than in 1993 (44.5 percent compared with 36.0 percent); more had two or more workers in the family in 2000 than in 1993 (9.4 percent compared with 9.0 percent). The economic prospects for literally millions of households with breadwinners in the labor market—male and female, single and married—are dim. In 1997, over one fourth of all workers were low-wage workers, earning less than $7.50 an hour. About one third of low-wage workers (nearly one tenth of all workers) were members of low-income families (with incomes below $24,600, about 150 percent of the poverty level for a family of four in 1999). Although nearly 40 percent of low-wage and low-income workers were employed full-time, year-round, many others could not secure such work: nearly half because they could not find it and many others because they had only seasonable employment, suffered involuntary layoffs, had health problems, or lacked childcare.

Nor does marriage guarantee an escape from poverty. In 1999, 4.3 million of America’s 10.9 million poor children lived in married-couple families. Shifting from the official poverty measure to a more realistic, “self-sufficiency” income standard (roughly 200 percent of the poverty level), we find that 14.0 million of the 25.3 million children in families that were not self-sufficient lived in married-couple families.

These outcomes reflect larger changes experienced by many individuals who enjoy less security now and reduced opportunity over the long term. Real wages for many male wage earners have stagnated or even decreased. While women have, on average, made gains relative to men, inequality for both women and men has dramatically increased. The recent boom restored some, but not all, of the losses in median real wages suffered since the early 1970s. Moreover, evidence shows lower prospects for wage growth over the working lives of many men as compared to those who first entered the workforce in prior decades; and the underlying causes point to similar results for many women. Some—largely college graduates—have held their own. Those who invested in some college-level training suffered less than those who did not, but returns on investing in college education have fallen. If such loss of real lifetime earnings continues unabated, the cumulative effect will be to raise the level of workers’ indebtedness or lower their lifetime savings for retirement and other purposes.

Declining wage growth has also been linked to increased job instability. This instability reflects a shift of available jobs toward those in the service industries, particularly low-wage, high-turnover ones, and also decreasing job stability within industry. Workers find fewer opportunities for wage growth, promotion, and training during the tenure of the jobs they hold, and changes in job tenure yield decreased returns, with an increasing number ending up trapped in low-wage careers. Moreover, the link between increased productivity and wage growth has weakened, especially for men. While there are more high-skill, high-wage contingent workers, most temporary work is still low-skill, with lower wages and benefits than those accorded full-time workers. Many of these jobs reflect an increasing decentralization and outsourcing of work and production.

These problems of job instability and vulnerability to catastrophic risks are not unique to the welfare poor or even the working poor. Even households that were once considered safe now often face economic calamity as a result of forces beyond their control. Bankruptcy, for example, often is the result of unemployment or underemployment, i.e., the loss of high-paying, good-benefit jobs and the necessity of taking low-paying, poor-benefit ones. For others, credit card debt serves as an easy, though ultimately very costly, means to maintain their standard of living, enabling them to gamble that such debt will keep them from falling over the financial ledge. Large bills and reduced incomes as a result of illness and injury—the loss of breadwinners’ income and the cost of treatment and care not covered by insurance (which may be lost along with a job)—often result in financial failure. The interplay of family and financial stress has been linked to marital instability and is compounded by the almost inevitable financial troubles resulting from divorce and separation, particularly devastating for women and children. For many, ownership of a home—middle class families’ principal asset—is not sufficient because they are “house poor.” They have little equity (tapped out by second mortgages or home equity loans) or suffer when market values fall. The cost of financing college loans for their children and the burden of student loans on young people that may extend into their working years (and may be onerous if they do not get good jobs) also threaten the economic stability of the middle class.

In summary, economic insecurity is not simply the burden of the so-called “welfare poor,” or even of struggling blue collar households. It has spread well up the economic ladder. Suffering from common structural forces largely beyond their control, American families need a solution much broader than welfare reform.

New Directions: Asset Development

For reasons we turn to next, we believe that an asset-based framework provides the right foundation for future domestic policy. It also provides an important angle of approach to the debate over welfare reauthorization.


Assets refer to the capacities and resources that enable individuals to identify and choose projects for their lives: to define what a good life is, and to pursue it.This definition fits widely-shared beliefs about the keys to achieving economic well-being: earnings to sustain growth during a working lifetime; knowledge and skills to enhance those earnings; pensions for support in retirement; insurance to protect against risks; financial resources to complement and enhance all three; and a network of personal and professional connections that supports and empowers.

Assets can be classified as individual and collective. Individual assets include income, human capital, and financial assets. The first of these, income assets, are rooted in employment-based income assets—jobs that are the source of cash income and benefits. These are essential because, for most households, an adequate employment-based income is still the principal means of attaining economic well-being. Minimum wage laws, tax subsidies for employer provided benefits, and the protection of workers’ rights to organize and bargain collectively—all directly enhance such income. Transfer income assets assure cash and non-cash supports when jobs are lost (unemployment and workers’ compensation) or when employment income is inadequate or lacking (Medicaid, Food Stamps, TANF benefits, and the EITC). Human capitalprimarily encompasses the knowledge and skills that are critical for gainful employment (and perhaps also the physical capacities that comprehensive health care helps to ensure). Building human capital can be the key to a virtuous cycle of increasing opportunity, establishing the base for continued acquisition of greater knowledge and skills and further movement up the economic ladder. Finally, financial assets, such as savings and checking accounts, stocks and bonds, and equity in a home, are critical to economic security and opportunity: to support life transitions, to provide a buffer in face of crises, to enable new ventures, and to sustain retirement income.

Several types of collective assets are also of great importance: they supply a critical community-based physical and service infrastructure; provide networks of support; serve as community resources in which workers, suppliers, customers, and neighbors have a stake; and can be a source of sustainable benefit within and across generations. But we concentrate here on individual assets, which are the primary indicators of economic well-being.

Insurance-Opportunity State

To appreciate the distinctiveness of the asset framework, consider the differences between it and the post-New Deal framework of social policy: what is sometimes called the insurance-opportunity state. Such a state has two main components. One is insurance, protection in the face of commonly shared risks of harm. Some risks are the inevitable concomitants of the human condition, such as age (youth or old age), illness, and injury. Others are the likely outcomes associated with the society’s particular economic, social, and political structure. For example, a principal risk in a dynamic market economy is involuntary unemployment. The corresponding goal of social welfare has been to guard against threats to income: to protect against absolute inadequacy and against sudden or massive losses. The focus is on stability and preservation.

The other goal is opportunity, which focuses on change and growth. It assumes a basic level of stability and well-being. But it also entails identifying and making choices about the potential for greater well-being, planning to achieve that goal, and building and then employing the capacities and resources necessary to realize it. These endeavors entail a relevant measure of initiative and self-reliance. Moreover, with every attempt to seize opportunity there is a risk (of failure). But it is a risk different in kind from that associated with social insurance: it is self-initiated, in some measure controlled, and, ideally, informed.

The contemporary American social welfare state originated during the New Deal and linked insurance with opportunity in a virtuous circle. For the vast majority, individual well-being was tied to opportunity through employment. Individuals had an obligation to seize job opportunities and strive to succeed when they were seized. Government’s role was to assure that such opportunities were available and that the rewards accorded by the workplace were meaningful. This arrangement enabled many individuals to enjoy some measure of economic security even when adversely impacted by limited opportunity, including old age and even disability. For many, the circle was complete: the government drew upon and pooled a portion of the productive wealth created when individuals successfully seized opportunity at the workplace and used it to maintain the system of insurance that sustained them when misfortune occurred. Many others, though, remained outside of this circle, including minorities, women, and those who worked hard but were simply left out of the structure of opportunity.

New Challenges

Changed realities now challenge the capacity of the insurance-opportunity state to assure economic security and opportunity. Employment and its rewards remain crucial to opportunity, but they must be thought of more in terms of a career than a job. A shrinking and shifting time horizon for jobs, even measured down to segments of days, demands that people identify, choose, plan and prepare themselves for the sequence of jobs they may hold over their working lifetimes. However much any particular job may offer basic economic security and minimum economic well-being, people’s long-range interest is economic mobility over a working lifetime. That requires a commitment to lifelong learning that can enable movement along ladders of opportunity through career-building. It highlights the importance of networks that have often provided access to job opportunities, but also the need for new intermediaries to identify and shape those opportunities as well. If dynamic economic change challenges individuals to find opportunities, it also challenges the larger society to ensure that they are available.

Changed roles in the family and their relationship to work have also altered the calculus of economic security and opportunity. Whatever its contemporary configuration, the family remains a locus of material, psychological, and emotional supports that afford its members the security that sustains them and enables them to seize opportunities that lay outside the family. It remains the principal place where kids develop essential intellectual skills, as well as the capacity to enter into relationships, and the cultural confidence and financial wherewithal to enter into a complex society with some hope of success. Both men and women now enjoy an expanded universe of opportunity; the worlds of work and family are no longer exclusive spheres for each. But they require the capacities and resources necessary for them to fulfill the responsibilities that accompany participation in both worlds, for their own sake and that of their children.

Increased life spans have extended the timeline for security and opportunity. Many people can now anticipate expanded choices about an active life at work, as an alternative or complement to other activities. But taking advantage of these opportunities now requires much greater financial and other resources. Moreover, longer life spans have increased the responsibilities of family members of working age to care for and support the elderly and infirm, making it more difficult to seize opportunities and enjoy economic security.

The calculus of risk has changed. More “flexible” employment relationships mean greater labor market risk for individuals and strains on the traditional public system of insurance to protect against it. They mean a weaker system of private, employer-based protection against the risks of ill health, disability, and aging, rendering less relevant the public system of tax subsidies that helped sustain it and straining the system of government protection that complemented it. People must be enabled to plan and provide for themselves against those risks and government and enterprise must each make a contribution to providing protection.

Asset-based policies can meet these challenges through both traditional and new approaches.One approach would be to provide income assets that make work truly pay, by integrating the EITC with the tax and income transfer system; raising and indexing the minimum wage to inflation; updating the unemployment compensation system in light of workplace changes such as the shift to part-time and temporary employment; and linking income supports to a realistic measure of well-being such as a “self-sufficiency” standard. Other approaches could build financial assets through such vehicles as seeded children’s opportunity accounts at birth; matched savings accounts perhaps modeled after Individual Development Accounts; and universal retirement accounts to complement the traditional social security system. Still other avenues could invest in human capital through vehicles like individual lifelong learning accounts with government or employer contributions, to enable worker mobility within and across enterprises.

An asset development framework is critical, then, to the task of reconstructing the insurance-opportunity state in light of new realities. It takes account of the capacities and resources with which all must be equipped to enjoy meaningful opportunity, despite all the new uncertainties and risks. It identifies the conditions that enable self-reliance (properly understood) and initiative, and the resources people need in order to plan and make choices. It provides a way to relate the productive contribution to society that might be expected of individuals with their capacities and resources to make it. It offers a basis upon which to justify protections in the face of incapacity and loss and the need to start anew after losses due to commonly shared risks. It serves to highlight how universal access to means that can enable people to provide for themselves can be afforded in a manner consistent with mutual dependence and social responsibility.

An asset policy framework also appeals to fundamental values that have sustained this nation’s greatest social policy achievements. Such values include opportunity (a meaningful chance for all to achieve well being); choice (the autonomy to select among opportunities); personal responsibility (initiative in developing the capacity to contribute to one’s own well-being and using it not only on one’s own behalf but also that of one’s family and the larger community); fairness (no unfair advantage in the pursuit of opportunity); and social responsibility (sufficient reward for those whose lives embody these values and basic support for those not capable of seizing opportunity or who have failed in the attempt).

What Should Be Done?

If we are right about the limits of welfare reform, and the promise of an asset-based policy framework, what are the conclusions for the reauthorization debate?

Retreating to the pre-1996 welfare system will not do: too many families on and off the rolls are struggling. Certainly, the pre-1996 system of cash assistance and income in kind was critical to families’ survival. Clearly, hunger and privation have no place in any system, old or new. But the old system never had the political support to stave off losses in the real value of cash assistance, let alone to effect long-term and dramatic change for the better. Recipients were stigmatized as being different from folks “who worked hard and played by the rules.” Stigma lingers today and the race card can always be played. But people who have now moved “from welfare to work” are more likely to be seen as similar to many others who are losing in the game both so earnestly play. Little support is evident among many constituencies for a return to the old system.

But embracing the post-1996 status quo will not do either. Some families have greater security, but many have landed in the low-wage labor market, where people work hard, live in poverty, and know hunger and all the other manifestations of economic deprivation. Still others—more challenged in capacities and resources—have hardly made it into the world of work. For them, time limits loom larger with each passing day. Indeed, some enthusiastic supporters of welfare reform admit those difficulties. A swelling economy aside, many who have “made it” owe the outcome to the EITC, Food Stamps, and expansions in health care coverage. Significant changes to the new status quo are required.

Marginal changes to remedy the most harmful aspects of the new policies would help for the short run. For liberals, such changes would include softening sanctions, easing time limits, reducing dead-end work requirements, and increasing overall support to the states for the benefit of program participants, notwithstanding lower case loads. For conservatives, the right changes would include more aggressive policies relating to non-marital births and marriage.

But such reforms should serve only as a starting point for building a new framework of social policy based on the idea of asset development. The 1996 reforms, suitably modified, would preserve the kernel of truth implicit in “work first,” that one of the keys to well-being for the vast majority of Americans is consistent, long-term employment in good-paying jobs with good benefits. The assumption was that if recipients joined the game and played as hard as the “rest of us,” they would be well on the way to being like the rest of us. But if so, their efforts should afford them a standard of well-being not unlike that which the rest of us expect and enjoy.

For thoughtful observers, then, the issue is not merely one of entry into the workforce but rather, a rewarding and sustained connection with it. The initial challenge is economic stability (or security) but only as the basis for economic mobility as well. And mobility requires assets: updated income asset policies; expanded, effective human capital policies; new financial asset policies; as well as others linked to the economics of community and place.

The coming welfare reauthorization debate offers the occasion to open up discussion about the promise that asset development policies have for fostering both security and opportunity.1 But how are we to get from here to there?

First, we must refashion the goals—and corollary measures of success—for reform. Traditional but more realistic income-based terms are important. But other meaningful asset-based measures of well-being with broader appeal are required, e.g., ones that identify economic opportunity gained and mobility achieved.

Second, we can identify the key roles that assets play in both protecting individuals from risk and affording them opportunity. Certainly, the focus should be on those whose lives are lived within or at the margins of economic deprivation. But the focus must be broad enough to encompass the working poor and the working (and struggling) middle class. The experiences and aspirations they share arise from the common role assets play in their lives and those commonalities can be the basis for a shared political agenda. We need to examine policies that support asset-building for the more affluent, which in turn justify asset-based policies that work foreverybody.

Third, we must focus on the common needs of American households, pointing out the avenues by which governmental policy can promote greater opportunity for all. With the fundamental values of fairness and opportunity as the backdrop, this would include the need for meaningful ladders of progress to enable those on welfare not only to get out of poverty but to advance into the economic mainstream. It also means exploring the ways in which policies that promote asset development for middle class youngsters can be applied to children and youth born into poverty.

Finally, we can offer concrete proposals that are relevant to welfare reform but have a broader reach, even if the short-term political calculus offers little prospect for immediate success. Writ small or large, they can embody an asset-based vision and point to its promise.

The reauthorization debate is the domestic policy opportunity of the near future, but it will be a lost opportunity if it devolves into an argument over whether this or that element of the 1996 changes succeeded. No honest analyst should feel good about discussing the minutiae of an economic security policy that clearly has not been a credible success. Reauthorization will also be a lost opportunity if it focuses only on the poor to the exclusion of other low-income working families, or even the conditions of tenuously “middle class” families. The upcoming debate offers a tremendous occasion to focus the nation and its leaders on the needs that all households have for a meaningful chance to achieve economic well-being, and it can start a discussion that one day results in a new domestic framework with asset-building policy as its common core. An asset policy framework appeals to fundamental values: opportunity, choice, personal responsibility, fairness, and social responsibility.


1 See Larry W. Beeferman and Sandra H. Venner, Promising State Asset Development Policies: Promoting Economic Well-Being Among Low-Income Households, Asset Development Institute, Center on Hunger and Poverty, Heller School for Social Policy and Management, Brandeis University, April 2001. Available online at