Isaac Martin gives us an account of the American welfare state, at a time when the current economic crisis seems to pave the way for a new New Deal. According to him, however, it is unlikely that the Democrats push back on the welfare reforms of the 1990s: despite their conservative origin, they have made some social and political progress possible.

It is an exciting time for those of us who would advocate a more solidaristic and generous social policy in the United States. But it is also a perilous time for social scientists who make our living pretending to predict the future. American social policy appears to be on the threshold of a major transformation whose contours are not yet visible. There have been a few such transformative moments in the history of American social policy, and they have produced new policies and political alignments that were not, and perhaps could not have been, anticipated beforehand. The two best known and most recognized examples are the New Deal (which gave us old age pensions, unemployment insurance, and mothers’ pensions in 1935); and the Great Society (which gave us public health insurance for the poor and the elderly in 1965). Historians of American social policy have likened these transformative events to the Big Bang, and the import of this analogy is not only that they were big, but that they were epistemological watersheds, thresholds beyond which prediction and scientific laws break down.

The United States now appears to enjoy many of the same conditions that made these prior big bangs of social policy possible—Democratic majorities in both houses of Congress; an economic crisis that has thoroughly discredited the pious orthodoxy of the previous administration; a Democratic presidential administration; and, if the early statements of that administration can be trusted, the political will to pursue an ambitious social policy agenda.

But I would like to suggest that we should not expect boundless innovation. The most profound social policy transformation of our generation may already have been accomplished, and much of the agenda of the new administration will be constrained within the bounds established by the welfare reform of the mid-1990s.

“Welfare reform” is usually used euphemistically to refer to a single piece of 1996 legislation with the unpronounceable name of PRWORA (the Personal Responsibility and Work Opportunities Reconciliation Act). This law transformed social provision for lone mothers by reducing benefits, imposing new time limits on receipt of cash assistance, conditioning cash assistance on recipients’ labor market participation, and restructuring the financing of the program in ways that were designed to encourage the further restriction of benefits and eligibility over time. Welfare reform in this narrow sense was not much of a big bang. It seems to have been an incremental change—the law adjusted a few quantitative policy parameters in a way that tightened the screws on poor single mothers, but otherwise continued the traditional approach to social policy.

But this narrow usage of “Welfare Reform” as a synonym for PRWORA is misleading. Welfare reform was not just PRWORA—because social welfare policy is not just government transfer spending. It is a commonplace among scholars of comparative social policy that such spending is only one of the institutional devices by which capitalist societies reconcile a market economy with the need for social protection. The clusters of related institutions that Gosta Esping Andersen famously called “worlds of welfare capitalism” also include patterns of regulation, wage bargaining, and even family and charitable institutions. The American world of welfare capitalism, to a degree that is distinctive among wealthy countries, relies on instruments other than direct government spending on cash assistance. Particularly important are regulatory instruments—characteristically enforced by private litigation rather than government action—and selective tax privileges that blunt the competitive edge of the market. It is also characteristic of American welfare capitalism that much of this social protection takes place at sub-national levels of government (and is therefore subject to substantial regional variation within the United States).

When we attend to these other domains of the American welfare regime, we see that PRWORA was only one part of a qualitative transformation that took place across the board. In the 1990s, policy changes in all of these domains reoriented social protection away from stay-at-home parents and toward labor market participants and their dependents, or “working families” in the jargon of the day. This new emphasis on social provision for the working poor was not confined to any particular part of the political spectrum. The new policy approach should not be understood as a victory for the left or the right, but as a pivot in the line of struggle—involving more generous social provision for some traditional constituencies of the left, and less generous provision for others. To equate welfare reform with PRWORA is to misrecognize the extent and character of this transformation.

In the sphere of regulation, the clearest examples concern minimum wage legislation. The same Congress that passed PRWORA legislated the first real increase in the federal minimum wage for decades. Several state legislatures also increased minimum wages in the mid-1990s. And more than 100 local governments, including municipalities and various administrative divisions of the states, enacted minimum wage laws of their own that exceeded the state and federal minima. All but a handful of these local minimum wage laws were so-called living wage laws that mandated higher minimum wages for employees of certain firms that had a financial relationship with the local government. Although no living wage law covers more than a small fraction of the workforce, the cumulative effect of hundreds of such laws in the largest urban labor markets in the United States is substantial.

In the sphere of tax policy, the clearest example is the earned income tax credit (EITC). This tax privilege amounts to a negative income tax for people who participate in the labor market but whose earnings are less than a given eligibility threshold. The EITC was created in 1974, but Congress legislated the first major expansion of the program in 1993. By the end of the decade it was providing more income support than traditional government spending on cash assistance for the poor.

What is striking with hindsight is how much all of these policies had in common. In political debate, these different policy instruments are sometimes described explicitly as substitutes for one another, as when employers oppose living wage laws because they were inferior to the EITC; and sometimes as complements for one another, as when living wage laws incorporate provisions for informing workers about their EITC eligibility. But regardless of whether they are complements or substitutes they are close kin. Across many different policy domains and levels of the political system, all of these policy changes of the mid-1990s favored increased social provision for low-income working families.

This transformation has at least two important consequences for the future of social policy in the United States.

First, it has caused the fates of the poor to diverge in ways that may make coalition-building difficult. American economists and sociologists have investigated the effects of PRWORA on earnings and income exhaustively, with greater methodological rigor than they have approached almost any other question in the applied social sciences. Most of this research finds little effect for good or ill—because it looks for average effects, and averages here are misleading. Welfare reform has caused real increases in well-being for some lone mothers. The package of work incentives in PRWORA (especially when combined with work supports such as child care and transportation, along with minimum wage regulations and the EITC) has measurably increased the average income of mothers in some samples. In at least one suggestive experimental study, the push into the labor market has also decreased the self-reported exposure of some lone mothers to domestic violence; if this finding reflects a real decrease in violence then it is surely an increase in well-being. But welfare reform has also exacerbated the social exclusion of other lone mothers. For those who are not able to find or keep steady work in the labor market—whether because of local labor market conditions, because of employer discrimination (on the basis of race, say, or previous criminal record), because of illness, because of domestic violence, or indeed for any reason at all—the withdrawal of state support is effectively a sentence of banishment to the underground economy on pain of starvation.

Like much else in American politics and policy the divergence in the fates of former recipients of state support is patterned by region. Analysts consistently find that PRWORA was implemented in a more punitive and less generous fashion—with correspondingly less beneficial outcomes—in the southern states. This is also the region of the country that is least likely to provide other tax or regulatory supports for working families. The pattern is, of course, consistent with the region’s historical reliance on labor-repressive agriculture supported by an ideology of racial supremacy.

The second important consequence is that welfare reform has co-opted the American left. To speak of co-optation may sound unduly harsh; I may soften the sting by pointing out that I include myself in the left, and that I do not regret this co-optation because I do not see a way around it. I do not wish to accuse anyone on the American left of selling out, and even less do I wish to make any claims about false consciousness. My point is simply that welfare reform has created important new tactical opportunities for the organized left, and we should not expect any part of the left to undo welfare reform if it will require surrendering these opportunities.

These new opportunities resulted from the administrative provisions of the new laws. Any form of social provision requires the state to determine which individuals are eligible to receive social support. The American welfare reforms of the 1990s—including not only PRWORA but also tax credits and minimum wage regulation—devolved this responsibility from the public sector to private actors, and this devolution created new policy handles that trade unions and other organizations of the left have seized. Provisions of PRWORA led many states to devolve the work of determining eligibility from state caseworkers to local intermediaries, and some of these intermediaries are progressive organizations that are otherwise involved in social justice advocacy; they will advocate for increased funding for public assistance, but not for reducing the amount that is targeted to labor market participants. Minimum wage legislation is implemented by employers, under threat of litigation, and lawsuits to enforce wage standards are now a staple tactic of American labor unions. Eligibility for the EITC, like the American income tax, is assessed by the individual taxpayer (subject to control by the authorities). Confronted by the bewildering income tax rules, many low-income Americans rely on tax advisors to assess their eligibility for tax credits, and some unions and civic organizations of the left have begun to dispense EITC-related tax advice as a way to identify and recruit potential members. All of these new policy handles will be hard to let go.

In the electoral arena, the Democratic Party may also have developed a vested interest in welfare reform. By targeting social assistance to working poor people, welfare reform has disarmed one of the most potent rhetorical weapons in the Republican arsenal—the threat that welfare will undermine the American work ethic. This was a pervasive and effective theme in Republican campaign rhetoric in the 1980s and early 1990s. It was largely absent from the 2008 elections, and when it was deployed it seems to have been ineffective. Few Democratic politicians today are eager to re-arm the Republican party by advocating social provision for the non-working poor.

Welfare reform, like the New Deal and the War on Poverty, may thus come to be seen as a watershed in American social policy development. To be sure, welfare reform was less spectacular than these other big bangs. On balance it was less solidaristic. And unlike these earlier big bangs it was not the project of any unified administration or group of political actors. It is better conceptualized as a shift in the terrain of American social politics. But like these prior big bangs, welfare reform was a transformative moment when a package of new policies created new and lasting political alignments. The question is now whether those alignments will survive the deepening economic crisis—or whether instead the growing ranks of the non-working poor will ultimately push the United States, and the Obama administration, toward a more solidaristic social policy.

To quote this article :

Isaac William Martin, « Which direction for the welfare state?. Legacies of welfare reform in the United States », Books and Ideas , 6 March 2009. ISSN : 2105-3030. URL :

Nota Bene:

Si vous souhaitez critiquer ou développer cet article, vous êtes invité à proposer un texte au comité de rédaction. Nous vous répondrons dans les meilleurs délais :

by Isaac William Martin , 6 March 2009

Related articles

The Fall of the House of Obama?

Can Obama still defend an agenda for prosperity?

by Pauline Peretz

Rethinking the New Deal

About : Ira Katznelson, Fear Itself : The New Deal and the Origins of Our Time, (...)

by Jean-Christian Vinel

Deplorable, yourself

About: Carol Anderson, White Rage: The Unspoken Truth of Our Racial Divide, (...)

by Isaac William Martin

A redrawing of the Red/Blue map in November 2010?

Five questions to Andrew Gelman about the coming midterm elections
by Pauline Peretz

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet