Transcript
See also: Transcripts of Panel I and Panel II.
MICHAEL ARMACOST: Why don't we begin?
I'm Mike Armacost, and it's my pleasure to welcome you this morning to this event, which we've described as Welfare Reform and Beyond.
I think all of you know that the welfare reform bill passed in 1996 has had profound implications as innovation in social policy. Just briefly, it ended the entitlements to cash benefits to eligible low-income families. It expected work by most parents in these families. It has time limits for cash benefits in the new Temporary Assistance for Needy Families program. It gave more flexibility to states in program management. And it offered more active encouragement through the program to reduce out-of-wedlock births and to reduce teenage pregnancy.
And as you know, many of the components of this legislation will be up for grabs in September 2002. So we can anticipate a very active debate over the next 18 months to determine which aspects of this legislation will be continued and which deserve some fine-tuning.
As policy makers review these issues, undoubtedly, there will be a slew of evaluations of what has happened since 1996. Already one sees that. And at the same time, there are pressures clearly visible already from conservatives on the one hand concerned that not enough is being done by the states to promote marriage or reduce to-of-wedlock births, from liberals concerned not enough is being done to protect the welfare of children, and from some states concerned about the prospect of perhaps losing funds or losing flexibility in the use of those funds.
So we thought the kind of contribution that Brookings can typically make in such cases is to attempt to synthesize the emerging research on this issue in a way that's accessible not merely to policy makers, but to the informed public to encourage thoughtful debate on these issues both in Washington and beyond the Beltway to look at policy options that can address the challenges that will still face the working poor, particularly after they have left welfare, and to provide the media with thoughtful commentary as this debate proceeds.
This represents the kind of formal launch of this initiative, but we have been working very hard, particularly those you will hear immediately after I sit down, assembling a very impressive team to do the work of planning the kind of products that we can deliver—several of which are in your materials this morning—and to work with the foundation community to provide financial support to support the initiative.
I'm very proud of the team that is working on this issue at Brookings, and I think you will be extraordinarily impressed by the diversity of their backgrounds and their experience.
Ron Haskins is a development psychologist; Isabel Sawhill, economist; Kent Weaver, a political scientist; Andrea Kane, who is our outreach coordinator, a public policy specialist. Ron worked for 14 years on the Hill, most recently as I recall, for the majority director of staff for the House Ways and Means Committee. As such, he had a real hand in the drafting of the '96 legislation. Isabel Sawhill was a key member of the team in the Clinton administration that pulled together their package of welfare reform measures. She now serves as president of the National Campaign to Prevent Teenage Pregnancy. Andrea Kane had White House experience on welfare reform and fatherhood issues, and she's also worked with the governors' association and state and local officials on these issues. Kent Weaver has been a long-time member of our government studies staff and has written a book on the politics of welfare reform in '96, which the Wall Street Journal has characterized as a virtual classic in this field.
So I think these diverse perspectives will assure not only an interdisciplinary approach to the problem, but also will assure that we have access to both sides of the aisle as this debate proceeds.
We're extremely grateful for the support which has been provided by a number of foundations, most particularly Annie Casey Foundation, the Foundation for Child Development, Joyce Foundation, MacArthur Foundation and the Packard Foundation.
I wanted to mention that this is also a program that's being webcast live in both video and audio, and it's available through the Brookings home page. We welcome those viewers or listeners at www.brookings.edu. And all our other materials, including these policy briefs, are available on that web site.
So we welcome all of you. It's very satisfying to see such a standing-room-only crowd.
And without further ado, let me invite Belle Sawhill, the co-director of the program to lead off.
ISABEL SAWHILL: Thanks, Mike. And welcome to all of you. It's wonderful to see this much interest in an issue which is probably not going to get debated a whole lot until perhaps a year or so from now. But we think starting this debate early can be very valuable. It will give us a chance to think about what the issues are, to review and synthesize the research that's coming in on them, and hopefully, to produce a more informed debate by the time Congress and others begin to consider what to do about the 1996 legislation.
We have a slide up here that shows you the objectives of the new Brookings project that Mike has just introduced, and I think that what I want to stress here is that there is an incredible amount of research that's been commissioned and conducted by various universities, think tanks, and other research groups. But there is a need to synthesize it and put it out there in a way that can inform this coming debate.
And we see that as our particular niche or role here—not to be producing new research as much as to be synthesizing what already exists.
And I'm very pleased that, later today, we're going to be having a panel in which some of the leaders of some of the other research groups who are working on this are going to be with us to tell us a little bit about some of their own work. And we hope that all of us can be very collaborative.
In addition to synthesizing the research, we very much want to think about new ideas, new issues, new policy options that Congress and the administration may way to consider as we move beyond welfare reform—and that's why we called this project "welfare reform and beyond," because I think it's not just about narrow issues of what to do about the rewrite of the law, but also a broader discussion about the future of antipoverty policy in this country.
We, I think as Mike has suggested, have got a very diverse group of people who are heading up the project. We don't all see these issues exactly the same way so we will internally be having a debate that I think will mirror, to some extent, the debate that the country at large is going to have.
We're going to have two overview presentations right now.
Ron Haskins is going to begin by giving you a summary of what we think the available research tells us about the effects so far of welfare reform. Obviously, that record is going to get modified as more research and more data come in.
I am then going to talk a little bit right after that about some of the issues that we expect the Congress and the rest of the country to be debating.
RON HASKINS: Thank you, Belle. I see a number of people in the audience that I met for the first time in 1995 and 1996. Some of you should be blushing.
We had a good time back then arguing about the finer points of welfare reform. And I would expect that we'll have an equally good time this time around. As Belle says, we'll probably start the debate seriously or at least start the congressional part of the debate next year. But there's a lot of work to be done before then because there is now a huge difference, I believe.
Now, back in '95 and '96, I would argue that we did have more of a research base for some of the policies that we—that Republicans and Democrats put forward—than we do in many policy debates. But that amount of research is very slight compared to what we have now.
We know a great deal about the effects—or some people might even say the impacts in some cases—of this legislation, and it would be a great mistake for Congress or for people who want to influence Congress to ignore this research. So that's a major goal of our project—to try to synthesize the research and then try to see what conclusions flow from the research.
Now, I've always noticed over the years, both when I was in the scholarly world and here in Washington, of course, that people didn't look at exactly the same research and completely agree on the results—totally different conclusions. So we are aware of that and that phenomenon will inform our project.
Let me begin—this is the—I should be able to walk around here. I hope you can still hear me.
One of the things that we want to do is to make sure that we don't just focus attention on TANF on the cash welfare program. As many of you know, the legislation in 1995-96 was extremely broad and so what you can see here is that, in addition to TANF, which I'll concentrate on today but our project will also talk a lot about the reduction in non-marital births. We don't know as much about that as we do about TANF, but we do know some things, and some interesting things. And of course, the goals are not reduced in any way because you don't know much about the impacts of the policies that we provide.
We do know something about the impacts of the Supplemental Security Income Program for children. There were very substantial changes in that program. Child support enforcement—we undoubtedly had the most major changes in that program that we've had in its history. And again, we know something about impacts.
One of the most difficult parts of the debate, as many of you, were the provision on benefits for non-citizens, and we still have very sharp partisan disagreements on what the nation's policy toward non-citizens should be. And I would expect that this will be a very intense, perhaps in some cases even bitter, part of the debate that we have upcoming.
In fact, this debate really has never stopped. We've had lots of debate in the meantime, and Democrats have introduced lots of legislation to undo some of the things that Republicans did in 1996, and I would expect that to continue and reach a crescendo next year and I would not be surprised if there will be very direct and sharp conflicts between Republicans and Democrats on how we should provide welfare to non-citizens.
And finally, child care—we made major changes in our nation's child care policy, and we know a little bit about that as well.
Next slide.
What I would like to do is to talk about the Temporary Assistance for Needy Families program in five regards, and I'd like to begin by reminding you of what the nature of the changes in that program were—for those in the audience who did not participate directly in the debate.
All five of these were major changes in the program. All five of these were the subject of intense discussion, and I think it's probably not too much of an exaggeration to say that there was great disagreement between Republicans and Democrats on all five of them.
First, eliminate the entitlement to cash welfare. The concept that Republicans followed was that people are not entitled to all welfare benefits, particularly cash. They must do something in return. You can't just give people stuff. They have to do something in return. And you can't have a serious program of doing things in return as long as the entitlement is in place. So the entitlement was ended and Sen. Moynihan and others saw this as the single most crucial reform in the legislation.
Second, block-grant funding, which of course could not have been done without changing the entitlement. Rather than give the states more and more money for more and more people added to the rolls, states got a flat sum of money. Therefore, if they could help people leave the rolls, they had more money per family on the rolls because the amount of money stayed fixed. So it provided a great deal of fiscal incentive for the states to help people get off the rolls.
Third, work-participation standards. We had had some participation standards in previous legislation, but these were by far the toughest standards ever put in federal legislation. Work was defined very precisely. States had to meet the standards and, if states or individuals did not meet the standards, we come to sanctions. Again, we had had sanctions in previous law, but never sanctions that had been employed so aggressively by the states. And we have seen since 1996 and even somewhat before that that the states are quite willing to use sanctions for people who do not participate in the work programs, who do not show up for appointments, who do not do the things that the states define as the proper behavior to get ready to enter the labor force.
So there has been quite extensive use of sanctions, and we have quite a research literature on that, and we will be doing—one of our books and briefs will be devoted specifically to the topic of sanctions.
And then finally and most famously perhaps, the five-year time limit, which has not yet hit, although we have had maybe 60,000 or so people who have hit shorter time limits that have been imposed by the states. Under a block grant, the states have dramatic flexibility and many states have chosen to use that flexibility to impose shorter time limits in the states.
So let's see what has happened.
Now, first, we have a whole series—I put this chart up here now for you to read, but just to impress you with the depth of the commitment that Republicans held to what was called then the "illegitimacy agenda" and there were many Republicans. In fact, one of the ironies of the '95-'96 debate is that I always felt that the biggest threat to passing legislation was a breakup of the Republican coalition rather than just the conflicts between Republicans and Democrats because there were very serious and powerful people in the Republican Party that insisted that nonmarital births were more important than work and that, unless we had very strong and aggressive programs to reduce nonmarital births, no matter what we did on work in the long run would not make any difference.
So as a result of that, many, many, many provisions were put in the legislation. I'd like to just mention three of them.
Of course, the bonus—we pay states directly up to $100 million a year—for states that actually reduce illegitimacy. I personally would argue —we're not going to talk about this today but we will in our policy briefs—that that policy has not been effective. States that have won—in Washington, D.C.—you start to scratch your head and think "Did we do this right?" And I think the answer is no, we didn't. And the reason is we don't have very good measures.
So I think this policy will get intense scrutiny during the debate.
The second is abstinence education—again, a very, very controversial policy. Abstinence education is defined very tightly, similar to the definition of work, under the TANF program, so that there was very little wiggle room, and anybody who wanted to use this money had to have a real abstinence education program. All states except California have taken the money. We have an excellent evaluation that's taking place now. We'll begin to get some preliminary results before reauthorization.
As many of you know, President Bush has talked about increasing abstinence education, so my guess is that this also will be an intense part of the debate.
And then, finally, the child support enforcement provisions established—by far the most stringent and tough paternity-establishment requirements that the nation has ever had. And indeed, I will not show data about this, but I can tell you that the establishment of paternity has skyrocketed. And in one year, we actually established more paternities than there were children born out of wedlock in that year, and that was because the states were working the caseload from previous years. So it shows you how the focus on paternity establishment and child support has been very important.
Next slide.
Now, we do have, as I said, important results at four years. And I'm going to talk just about six types of results. First, the impacts on the work—what most people have paid most attention to. And then the little information we have on the rest of these—reduction of nonmarital births, the changes in supplemental security income for children, child support enforcement, welfare for non-citizens, and child care.
Okay. Next slide.
As everybody knows, and I see people in this audience from the media who have made the profound point that the reduction in the welfare rolls is not the only measure of welfare reform. We agree to that. But at the same time, I think it's very important to focus on what has happened here. It is really surprising.
There has never —if you go all the way back to '59—there had never even been three consecutive years in which the welfare rolls had declined. And they have now declined by 50 percent. In the past, the biggest consecutive decline was 7 percent. So this represents a previously unknown phenomenon. There are things going on here that bear careful attention.
It is really remarkable that the rolls have declined this much. It's not the only measure, but it's a very important measure and it tells us a lot that this reform was serious and had major impacts both on states and on individuals.
Next slide
Now, this slide leaves all kinds of complications out. I want to make that real clear. I don't have anything on here about work expenses. I don't have anything on here about Social Security taxes and so forth. But this is very similar to the slides that Republicans brought to the floor of the House—it was really funny seeing senior Republicans hauling these great big charts and putting them up and arguing saying, "Look at these charts."
And this is the basic argument that was made.
Gov. Thompson, now the head of HHS after yesterday, has said many times, "You can't get out of poverty if you're on welfare." This is a fact. We created a system that, as long as you were on welfare, it was virtually impossible to get out of poverty. You had $8,000 in resources plus Medicaid coverage.
But if you even took a low-wage job because of other supports, you could get out of poverty, and you could have much more substantial income. Now, as I say, this leaves out lots of complications.
So let's see what actually happened. Next slide.
The first requirement is that people would have to enter the labor force, which they had never done before despite all of our many, many reforms of welfare—as Gil Steiner used to say, "tireless tinkering" with the AFDC program. But now we see, if you—especially if you look at—these are married mothers and this just continues a trend through much of the last half of the 20th century following World War II where more and more mothers entered the labor force.
Single mothers had then very little movement, and then after about 1995, as the states began to implement their own versions of welfare reform, it increased quite substantially. These are women actually with jobs. And most impressively, these are never-married mothers, and I would argue that these are exactly the mothers that we had many conflicts about on the floor of the House and the Senate that many Democrats said we have to keep entitlements because there are many people on welfare who simply cannot support themselves. They need more education, more training, have lots of barriers and so forth.
Well, here they are—never-married mothers. And it was relatively flat. But after about '95, it just shot up dramatically—a 40 percent increase. So this group—by the way, some of you—I see you looking funny—this group is part of this group. This is all single parents. This is just never-married single parents, the ones most likely to be on welfare, the ones most likely to have long welfare spells
So the first criteria is—did the legislation increase work? And the answer is—without question, it not only increased work but increased work among the most disadvantaged mothers.
Next slide.
Now, income. This is probably the point where we begin to see the types of problems that the states have to deal with and the Congress is going to have to consider and figure out what to do about during the reauthorization debate.
The overall picture from '93 to 1999, even for these mothers in the very bottom fifth—this is the poorest mothers, most mothers on welfare and many mothers leaving welfare. They're below about $11,000 in income in 1999. So this is the very bottom group. Almost everybody in this group is in poverty.
And even in this group, if you look, you can see a very substantial—82 percent —increase in earnings. So even in this bottom group, which is full of mothers who don't work at all, we still on average have had a very substantial impact because there is a great increase in earnings. And of course, right along with that, of course, there's been a big increase in the EITC, the earned income tax credit. If you work more, you make more earned income tax credit, so that fits right together. And earnings go up. The EITC reinforces the earnings and gives you more money.
However, cash welfare has declined quite substantially, and so has food stamps. Now, I would point out, this is what we want. We want welfare to go down. We want earnings to go up. And that's exactly what has happened. But in this bottom quintile, particularly in 1995 and 1996, even though earnings went up and EITC went up, welfare went down more.
This is probably—this is a theory—this is probably because we have mothers who did not take jobs. They left welfare because of the hassle or for whatever reason. These are the sort of things we need to know a lot more about, and they went into the economy, they didn't get a job. So what are they doing? We don't know for sure. We know that they're more likely to be living with other people. They're more likely to be living with people who provide them with in-kind and cash support.
So, the overall picture is positive even in this bottom 20 percent. There is an increase in earnings and there is an increase in total income. But some years, there are problems. And overall, the picture is not nearly as positive as we would want it to be.
And there is no question—mark this—there is no question that there are mothers in this bottom group, and children, who are worse off than they were in 1996 before we passed the legislation. There is no such thing as a policy which produces no losers. And this one has produced some losers down in this bottom quintile.
And these are mothers that we should focus on and we should figure out how to help these mothers.
I can tell you as one step, if they got all the food stamps that they were entitled to—they're eligible for food stamps, but we have a very strange phenomenon that they don't get it. This has always been the case. I see Sheila Zedlewski sitting here, who studied this very carefully, so I should be careful about what I say here. But a lot of mothers are eligible and maybe half don't get the food stamp benefits they're entitled to. And if they got all the benefits they were entitled to, this total income would jump up quite a bit.
So there are lots of strategies to employ here, but there is a problem. We should address it.
Next slide.
Now, this is the same thing except for the second quintile, the second fifth of earnings—of mothers with children. And this is between roughly $11,000 and $23,000. And I would say that this is very nice success story. Earnings are very substantially up—almost 100 percent increase in earnings over the period, almost a 200 percent increase in the earned income tax credit, primarily because more of them get it, and the EITC was expanded very dramatically as a result of legislation that we passed in 1990 and in 1993. And welfare has declined, but not as much as in the previous quintile.
So a lot more work in this second group of low-income mothers, and this is the pattern that we would like to see transferred to the bottom fifth of earners by emphasizing work and emphasizing the kind of supports that we provide for work.
Next slide.
A huge debate in 1995 and 1996 about poverty. I see Paul Offen (ph) sitting here. He wrote many speeches predicting—maybe he didn't write them; maybe he tried to talk Sen. Moynihan out of them. But in any case, Sen. Moynihan and many others said that poverty would increase—a million children more would be put in poverty, and there were op-eds in the Washington Post.
OK, here is the official poverty measure, '95,'96, '97, '98, '99. These are the caseload reductions in those respective years. And by the way, every single one of these caseload reductions are greater than in any previous year of AFDC.
Here is what happened to overall child poverty. As you can see, everything in this chart goes down. So overall child poverty is declining every year, and then black child poverty. And as you can see from this chart, in every year, we have welfare rolls going down, overall child poverty going down, and black child poverty going down.
So those who predicted declines in poverty—it could still happen later. We could have a recession, maybe when the five-year time limit hits, but so far, there have been very dramatic declines in poverty. The declines in black child poverty in '97 and '99 are bigger than in any previous year on record. Right now, black child poverty is at its lowest level ever. And overall child poverty is at its lowest level since 1979.
So not only is there more income, but there's enough income to take these mothers and their children, on average, out of poverty. You still have problems at the bottom that I mentioned. But overall, very positive effects on poverty.
Next slide.
And not only that, but I would say that those official poverty statistics actually underestimate the progress that we made. And here we can see a very important result and I think a hint that the overall strategy we're employing now as a nation is much more successful than our strategies in the past.
This poverty measure includes earned income tax credit, food stamps, other things that are left out of the official poverty measure. So you get a broader assessment of the impact of the work supports that we're supplying to families once they enter the labor force.
And as you can see, 1983-1989, 1993-1999, and what happened to the change in child poverty by this broader measure. And as you can see, in the 1980s, child poverty declined about 15 percent, which is a nice, respectable performance. But look what's happened so far in the 1990s—more than twice as much of a decline in child poverty. This is crucial in my estimation because what it shows is that if you go from a strategy of just giving people benefits as we did in the 1980s and adopt a strategy of doing everything possible to encourage and, where necessary, force people to work, and then subsidize their income even though they have low-wage earnings, we have these very dramatic impacts on poverty.
So the strategy that we have used—next slide—the strategy that we have used is not simply to emphasize work. That of course got the most attention in the debate back in 1996. But now we can see that there is a second part to this strategy that Congress and that the nation is following—and not just Congress but presidents as well—and that is to help low-income families with children by supplementing their income.
This is a study that we asked the Congressional Budget Office to do. It's a little complicated. But let me tell you what the essence of it is.
This shows, if we hadn't changed any of those laws on work support, like the earned income tax credit, child care and so forth—they're listed all here—all these programs were either completely created out of whole new cloth or expanded after 1994—if we hadn't done any of those things, in 1999, we would have given families about $5 billion, a little more, families that were working. We would have supported them to the extent of about $5 billion.
But because of all those expansions, the expansion of the earned income credit, expansion in Medicaid and so forth, in fact, we spent more like $52 billion. And what this tells you is that, on the one hand, we increased work, we encouraged work—some states thrown them off the rolls virtually, much stronger policies than ever in the past to get low-income mothers to work—but once they work, we support their income with a nice work support system that has fallen into place over approximately a 15-year period.
And there is a lot of work, including here at Brookings—and we will have a lot of new information about ways that this work-support system through the tax code and through direct benefits and through the way that states use their welfare money—there are many different avenues here—would expand the support that we give the low-income families so they would have even more income. And I believe that the intensity of this debate is going to go up a few notches as a result of recent research that perhaps Judy will talk about, which shows that children are benefited, apparently in four out of four random-assignment studies, children who had solid work supports, and their families were at a level—Judy wouldn't probably want to use a number, but I'll use a number anyway—families that got to maybe $17,000 or $18,000 a year, there were positive impacts on their children.
The kids are actually better off. They performed better in school in all four of the studies. In some of the studies, they had better health, and their behavior was better. So there could be impacts on the children. And I would ask you to note especially—this is not an educational intervention. This is an intervention where the parents—for the parents to behave like other citizens, to support themselves, and you get these effects on children, which are in every way comparable to direct intervention with the children.
So it's quite a remarkable result. We're at the beginning of this literature. There will be more and more work in the years ahead, and this, too, will be part of the congressional debate.
Next slide.
Now, I just want to say one thing about nonmarital births, and that is, by every measure, they have leveled off. Some Republicans have encouraged me to say at this point that these declines started in 1995 and—show the next slide—these are teen birth rates. They started declining in '91 and are all the way back almost to the level of the 1980s. And these Republicans have asked me to say that both of these are anticipatory responses to the Republican welfare reform bill.
(LAUGHTER)
But I am not going to say that.
We don't have the slightest idea what's going on here, but it's very positive. It's very positive. It might be a result of all those 15 policies I showed you on that slide. Undoubtedly, there are lots of other things going on. A hint is that the decline is much more—is steeper among blacks than any other group. So why? We have a lot to learn here. But this is really one of the most important problems—at least as important as the work problem. We have to figure out as a nation how to get more of our children out of never-married families because it's a disaster for them. It is a disaster.
We have bipa