In this article, the Secretary of the U.S. Department of Health and Human Services (HHS) outlines the key provisions of the 1996 welfare reform act -- written by the U.S. Congress and revised by the Clinton Administration -- underscoring the partnerships among all levels of government and between the public and private sectors upon which the success of welfare reform hinges.

In War and Peace, Tolstoy observed, regarding the limits of policy making, that Napoleon thought he was controlling events but events were controlling Napoleon.

The essence of Tolstoy's insight is that no policy can fully predict or control human behavior. That lesson in humility is especially apparent to policy makers grappling with welfare reform.

In many policy debates, research plays an important, even decisive, role. The U.S. Government banned DDT (a widely-used pesticide) because of overwhelming evidence that it was a carcinogen. In 1964, the U.S. Surgeon General reported that smoking was linked to lung cancer, paving the way for warning labels on cigarettes. In 1996, HHS issued a report from the Surgeon General about the benefits of regular, moderate physical activity, which was based on a review of two decades of research.

And good social-science research has helped shape our policies on immunization, breast cancer, nutrition, drugs, Head Start programs, tobacco use among children, and health-care reform.

But when it comes to welfare, research has cast a shorter shadow, for several reasons. First, social-science research often is not directly linked to actual legislation and, therefore, cannot always answer the unique questions posed by the issues policy makers are considering. Second, policy makers sometimes need current insights into human behavior more quickly than social-science research can provide them. Third, when research about human behavior is fragmentary and conflicting -- as is the case with some issues relating to welfare -- policy makers often have to make leaps of faith.

For example, studies differ about the effects of welfare on premarital birthrates. Some studies suggest that higher benefits are associated with an increase in the number of unmarried Caucasian women having children, but most studies show no correlation at all. When the results of research are uncertain, open to multiple interpretations, or at odds with other research, decisions must be based in large part on the policy maker's judgment, experience, values, attitudes, and training.

Does all this mean that social-science research is an unlit lantern -- offering no useful guidance about how to move people from welfare to work? Absolutely not.

In advising President Clinton and me on our original welfare bill, distinguished academic experts on welfare -- including Mary Jo Bane and David Ellwood, the co-chairs of the President's welfare-reform working group -- relied heavily on current research about trends in the economy and the work force, as well as on research about the behavior of people on welfare. Accordingly, research helped shape our understanding of many important questions, including: Why do people go on welfare in the first place? How long do they stay on? Why do so many return once they've left?

The answers to those questions are important. However, they do not provide the complete answer for policy makers, because welfare reform is fundamentally about changing the culture of our welfare system -- created more than 60 years ago to insure that it accurately reflects 21st-century realities and values. For example, we know that today most women work outside the home. As a nation, we believe that we need to make work pay, that is, that we must change the incentives so that going to work is a more rational choice than staying on welfare. And we believe that children need the emotional and financial support of both their parents.

These realities and values ought to guide our welfare policy because, as President Clinton said recently, "There is a passion in this country ... to move people from welfare to work in a way that enables them to support their children and live in greater dignity." The belief that dependence must be reduced and dignity increased now transcends party lines. In fact, Mr. Clinton vetoed two welfare bills because they defied this consensus.

The President believes that the new welfare law is different, and that it takes strong steps toward reinforcing national values regarding work, family, and independence. In his words, it gives us a chance to "end the terrible, almost physical isolation of huge numbers of poor people and their children." In contrast to the two previous bills, the new law does not replace the national guarantee of food stamps and school lunches with block grants to the states. It drops proposed deep cuts in funds for child-abuse prevention, foster care, and adoption assistance, preserving our national commitment to these vital services. It maintains Medicaid's historic guarantee of health insurance for poor children, the disabled, pregnant women, and the elderly. It is tough on parents who don't pay child support. And, to enable single parents to go to work, the law adds $3.5 billion for child-care services.

Overall, the new law replaces the federal guarantee of income support for families with dependent children with block grants to states, which are required to maintain at least 80 percent of their fiscal 1994 spending on welfare. The law has a five-year lifetime limit on benefits for welfare recipients, although states may exempt for hardship up to 20 percent of their caseloads. At the same time, the law requires welfare recipients, with few exceptions, to work after receiving two years of benefits, to enroll in on-the-job or vocational training, or to do community service. Unmarried mothers under 18 years old are required to live with an adult and to attend school as a condition of receiving welfare.

The goal of our welfare-reform strategy must always be to make work pay. We had begun to fulfill that goal long before President Clinton signed the welfare-reform bill. How? First, by creating the right incentives, so that going to work is a better deal than staying on welfare. We did that by fighting for and enacting budgets that dramatically expanded the earned-income tax credit for 15 million working-poor families and invested in education and training to help people find jobs and keep them. We did it by raising the minimum wage to $5.15 an hour, thereby making it a living wage and giving all full-time minimum-wage workers an $1,800 annual raise. And we did it by expanding Head Start and child care and approving several state demonstration projects designed to expand health-care coverage for the working poor.

Second, over the past three and a half years, we have granted 43 states a total of 78 waivers from previous federal welfare rules so that they can use innovative approaches to reforming welfare. These state demonstrations now are making work and responsibility a way or life for more than 75 percent of all welfare recipients.

Many of the waivers that the governors requested included changes that impose time limits on benefits, create stronger work requirements, and provide expanded child care, thereby demonstrating both a strong desire among states for flexibility and a broad constituency for reform. In fact, many states held extensive public hearings and legislative debates before requesting their waivers. This further increases the likelihood that welfare reform will succeed, since we know from research by the Manpower Demonstration Research Corporation and others that change works best when states have a sense of ownership in their own plans.

When President Clinton signed the welfare bill, he said: "This is not the end of welfare reform, this is the beginning. And we all have to assume responsibility." For the federal government, responsibility includes changing some parts of the bill that have nothing to do with reforming welfare and strengthening protections for children. The two provisions that the President already has identified as needing change are the cutoff of most federal benefits for legal immigrants and the excessive cuts in food stamps, especially for families with high housing costs. These problems are important -- but not insurmountable. And the President has vowed to go back to Congress in 1997 to fix them.

But Washington alone cannot make reform work. The culture and practice of local welfare offices -- indeed, of everyone who works with welfare recipients -- must change dramatically. Instead of focusing primarily on determining eligibility for benefits, officials will need to give welfare recipients the tools that they need to make the transition to work. The greatest challenges to the public and private sectors will be to clear out the roadblocks that keep welfare recipients -- who often lack education and work experience -- from getting jobs and keeping them.

While the first goal of real welfare reform is to make people support themselves, the second is jobs, jobs, jobs. We will need an extraordinary commitment from business, government, universities, unions, non-profit organizations, and religious groups to train and hire people on welfare. To make that easier, the President is proposing a targeted tax credit of 50 percent on the first $10,000 in wages that employers pay each long-term welfare recipient. The credit could be claimed for up to two years. He also has proposed grants to cities to help employers create new opportunities for long-term welfare recipients.

Clearly, policy makers must follow closely the results of the new welfare law and be flexible -- and humble -- enough to make changes if we're not getting the results that we seek: more people working, greater family stability, more parental responsibility, and increased dignity and hope for children.

We will need good research. Researchers have suggested, based on past studies of issues such as job attainment, that it is difficult to change behavior through small, incremental policy changes. But we don't know what effects a fundamental change will have. For example, what are the benefits for young children when they see their parents go to work every day? How successful will the law's work incentives prove? How well are the changes allowed under the state waivers working? And what impact will clear, consistent work-and-study requirements for young welfare recipients have on teenage pregnancy?

Social-science research may not be able to answer these questions fully. But all of us have an abiding duty to try to answer them together. Because, in the end, we must focus on what we do agree on: The best protection for American children is parents who want them, nurture them, and support them. Our efforts -- public and private -- must always reflect that consensus.

When President Clinton signed the welfare-reform bill, he issued a challenge to our entire nation. He said:

"Every person in America ... who has ever said a disparaging word about the welfare system should now say, 'Okay, that's gone. What is my responsibility to make it better?'"

__________

(This article is in the public domain. It originally appeared in The Chronicle of Higher Education, October 4, 1996.