Welfare reform is a smashing success; welfare reform poses a host of new challenges and promises. Both conclusions could readily be drawn from the research, analysis, and opinion industries that have been spawned in the wake of the 1996 federal welfare reform law that ended the federal welfare entitlement and substituted the “workfare” model.
Evidence detailing the success of the workfare approach (which builds on a number of state-initiated models, most notably Wisconsin’s workfare initiative) is not hard to find. Reams of interim studies and press accounts have reported reduced welfare caseloads around the United States and have provided strong evidence that many welfare recipients actually get living-wage jobs. But the problems and challenges arising from welfare reform are harder to pin down in part because the very swiftness of welfare devolution has taken many people by surprise.
The immediate question for philanthropy and other supporters of outcomes research is “Are we asking what we need to ask in order to find out?” While the voluminous research on welfare dependency is too sweeping to allow easy generalizations, it may be useful to define a few broad “fields” of this research in order to understand the priorities of policymakers and researchers.
Measures of Success
The first and largest of these research fields involves various projects designed to quantify, in some meaningful way, whether welfare reform is a “success” or “failure,” based on some reasonably objective criteria. While definitions of “success” can vary with the policy bias of the person asking the questions, a significant number of these projects ask straightforward, non-ideological questions. How many people are leaving welfare rolls? How many of those are finding jobs? Of those who find employment, how many persist in the workforce?
Such simple questions, though, don’t give researchers a very challenging assignment, other than ensuring that their databases and methodologies are sound. As Jack Tweedie writes in State Legislatures magazine, “Nationwide, caseloads dropped by 36 percent from January 1994 to March 1998. . .. Reducing caseloads was one goal of welfare reform [but] legislators’ concerns do not end when a family leaves the rolls . . . . To judge how reform is working and how they can make it work better, legislators need to know what is happening to these families.”
To help find out “what’s happening,” Tweedy identifies a few issues that states have already investigated, including earnings of former welfare recipients, other sources of family support, measures of hardship following loss of cash assistance, and percentage of former recipients who end up back on the welfare rolls. Tracking parameters like these requires more sophisticated research methods than the “core” success questions, but nothing unmanageable for the social scientist. Of course, once one opens the analytical door of tracking post-welfare beneficiaries, the universe of potential welfare research projects explodes.
The reason is simple. Once you reach beyond basic questions of size of welfare rolls, job placement, and job retention, absolutely anything that might go wrong (or right) with a former welfare recipient is fair game for the researcher. For example, the Economic and Social Research Institute, in association with the Urban Institute, surveyed employers to determine their basic expectations of workers, and their specific attitudes toward welfare recipients. The Labor Department funds different approaches to job placement and provision of “employment support services” such as training and child care, to determine what works.
The National Governors’ Association Center for Best Practices reports that most states are tracking not just job placements, but whether former welfare recipients “are able to adequately support their children.” The Center suggests other questions that researchers should pursue in the years ahead, such as: weekly hours worked, type of employment benefits received, potential for workforce advancement, causes of job loss, tracking changes in family and living arrangements, ability to make rent payments, and risk of abuse or neglect of children living with former recipients. Although these questions are valid, taken together they imply a massive, if not permanent, research effort that will delve deeply into the lives of post-welfare recipients for years to come.
Could it be that we want to know too much about the former welfare population? Each increment of research is subject to the law of diminishing returns, because the farther out in time researchers try to track welfare recipients, and the more detailed information they try to gather, the harder they find it is to collect that information. On top of all this, many “down the road” post-welfare questions require the analyst to make very subjective value judgments. For example, how can one demonstrate that a former welfare recipient is “unable” to make rent payments? After all, “inability” is a function of both income and other claims on that income.
None of these problems are unique to social science, of course. But the point for potential funders is that while we can be reasonably confident regarding caseloads and job placement numbers, the bulk of welfare research tackles more complex, long-term problems where the real return for the research dollar is uncertain and both research design and measurement of outcomes more subjective.
Nuts and Bolts
The difficulty with “follow-up” studies illustrates another research deficit, namely the lack of in-depth study in key management and administrative issues. Several key players in this research, including Jason Turner (a human resources administrator for New York City) and the Urban Institute’s Michael Wiseman, emphasize the need for more and better studies of staffing for welfare programs, risks of politicization, development of professional performance standards, and program integration (as Wiseman points out, welfare recipients also eligible for other programs, such as food stamps, now face radically different rules and procedures for each program).
In fact, the NGA’s Center for Best Practices has been addressing these questions. Indeed, the welfare reform community is virtually unanimous in its eagerness to find success in welfare reform. Despite the intense opposition of liberal critics when the 1996 reform bill passed, the statistics on caseloads and job placements alone since then have given all involved a significant incentive to make the best of the new system. This means not just looking at the nuts and bolts of what’s effective in getting people off the rolls and into stable employment, but building up ancillary support programs to transform the world of welfare into the world of “managed work.”
“Managed work” is a situation where the employee, if only because of the welfare stigma, is assumed to require extensive support systems in order to succeed in the workplace. These support systems include not just corollary government aid programs (food stamps, child care aid, training, education, courses on how to behave socially in the workplace, lessons in financial management), but a complete information network (how to get transportation, how to apply for other forms of aid, how to upgrade work skills) derived from both government and philanthropic sources. If the welfare research agenda is a reliable guide, we are in the process of replacing cash-grant welfare with a much more complex “guaranteed lifetime employment” system of aid.
What Is Undone
The research implications of all this for both public policy and private philanthropy are murky at best. One might expect that, given all the talk about budget savings from drawing down welfare caseloads, research would focus on whether savings from smaller caseloads are merely being shifted into the training/placement/caseload management budgets. While there is evidence that caseload reduction provides real net savings to the public, that needs to be documented by looking at all the alternative programs that are being offered to former welfare recipients, including those offered by private employers and charities (cost-shifting is cost-shifting even if it moves costs off the government’s books).
We will no doubt discover much that is useful to know from the current crop of research in welfare reform. We will also pay a high dollar price for that research, whether public or private, and can only hope that the benefits of what we learn will justify the cost.
George A. Pieler is a contributing editor to Philanthropy.