Welfare Reform and State Flexibility

Transcript

MS. ISABEL SAWHILL: Good morning.

Let me welcome all of you on behalf of the Welfare Reform and Beyond Project at Brookings. I'm Belle Sawhill and I co-direct the project along with Kent Weaver who is sitting over here in the front row.

We are in the final stages of the project since one of the major objectives of the project was to synthesize the research that was available on how the 1996 Welfare Reform law was working and to help structure and provide background on the debate surrounding the reauthorization of the law this year.

We will be doing more meetings and producing more policy briefs over the remainder of this year. This is actually the 11th in our series of public forums on these issues. We have produced 23 policy briefs so far on various issues. And most of them are contained in a volume which is out on the table in the hall called Welfare Reform & Beyond.

We have three more policy briefs that are in the pipeline as I speak. One is on Workfare and what we've learned about its successes and failures. The second is on the fiscal challenges facing the states. And the third is on tax policy and low income families and will be done under the auspices not only of Welfare Reform & Beyond, but under the auspices of the new Tax Policy Center that is joint between The Urban Institute and The Brookings.

As most of you know, this is a very important time in the reauthorization process. The House has passed a bill. As we speak this morning the Senate Finance Committee is hoping to mark up its bill.

That said, the purpose of today's forum is not so much to revisit the nitty-gritty debates about what should or should not be in this bill, although I'm sure some of those issues will come up, but rather to step back and consider some of the broader questions that underpin or provide background on the ongoing debate in this country about state flexibility and about the role of the federal government vis-à-vis the states.

We are very pleased to have as a moderator for this event and discussion Bob Reischauer. Bob is President of The Urban Institute. He is formerly a Senior Fellow here at Brookings. He is formerly Director of the Congressional Budget Office. And if I remember correctly, a long time ago he wrote his doctoral dissertation on the issue of state flexibility.

We have a distinguished group of other panelists. Senator Tom Carper will be joining us shortly in addition to others sitting in the front of the room here.

With that, let me turn it over to Bob so we can get into the discussion.

MR. ROBERT D. REISCHAUER: Thank you very much Belle, and let me add my welcome to all of you.

In federal systems of government a perennial issue of debate is how much flexibility or control should be given to subnational units of government over national programs. This issue arises because almost always the national or the central authority has more resources than the subnational units of government and therefore is usually the wellspring of new policies, while the subnational units of government often have the administrative capacity to operate whatever programs are thought up at the national level. To an unusual degree even among federal systems, the United States has relied on programs that share responsibility to one degree or another between the national and the subnational units of government meaning usually the states.

In some of these instances it really is a situation in which the federal government is using the state or the local government as simple agents to administer its programs and very little in the way of flexibility is afforded to those units of government. Here the Food Stamp program would be a classic example.

In other cases the federal government has shared its fiscal responsibility with the states and has given the states a modest amount of flexibility, and Medicaid would be an example of that.

At the other extreme we have, or we had programs like the general revenue sharing program in which there were very few strings attached and money was put on the stump and the states could do pretty much whatever they wanted.

Welfare, meaning AFDC and the TANF program, has always been one of those in-between type programs and therefore has been an area in which we have battled year in and year out over the two roots here. The one size fits all policy or the let a thousand, it's not a thousand, 51 flowers bloom.

This isn't really surprising considering that welfare programs operate in a policy area where opinions are very strongly felt and the fundamental values are at stake and financial responsibility is shared and the objective conditions with which the program must operate vary tremendously across the states, namely the nature of the caseload, the fiscal capacity and administrative capacity of the states, the labor market, and so on.

We're here today to discuss and debate this issue how much flexibility is appropriate in welfare programs at the state level, and to carry on the discussion or to start it off, you and your questions will be a big part of this gathering, we have a very distinguished panel each of whom will speak for five minutes.

MS. SAWHILL: Can I just interject, Bob, since I forgot to say that this is being Webcast live. If there are people in the home audience who want to submit questions they can, and the address is "question@brookings.edu".

MR. REISCHAUER: Let me repeat that just because Belle obviously I don't think was on the camera.

If you're watching this over a Webcast you can also submit a question at "question@brookings.edu".

We, as I said, have a distinguished panel with different perspectives on this issue to start off the discussion. First we will hear from Tom Gais who is the Director of the Federalism Research Group at the Rockefeller Institute at SUNY Albany. A political scientist who has been in charge of examining field research in 22 states and is a prolific author in this area.

He will be followed by Larry Mead who is a Professor of Political Science at NYU and has been a visiting professor at numerous other institutions and the author of a number of well-known books in this area.

Bob Greenstein will be our third speaker. Bob is the Director of the Center on Budget and Policy Priorities and before that, which was many years ago, was the Director of the Food Stamp Program. Bob has been one of the most influential policy advisers in the area of welfare for at least the last decade.

Finally, in the empty seat which will soon be filled, we will have Senator Tom Carper from Delaware, a two-term governor and a leader in the welfare reform effort and a five-term member of the House of Representatives. Tom also was in charge of the National Governors Association when he was governor, and has been one of the leaders in this area.

He also, I can't resist a digression. He also has one of the most interesting bios for a member of Congress that I've ever seen. Usually the purpose of the bio is to show how your genetic material came out of the center of your state, drawn from a well. His says that he was born in West Virginia, grew up in Virginia, educated in Ohio, and now represents Delaware. So it's really an interesting thing.

Without further ado, let me turn it over to Tom.

MR. THOMAS GAIS: Thank you very much, Bob.

TANF was the biggest block grant actually ever enacted and it gave states enormous flexibility. If anything, that flexibility has increased dramatically since the enactment in 1996. The two most important factors I think were the 1999 non-assistance rule, and because this is a five minute talk I'm not even going to begin to try to explain that. Also the big declines in caseloads among the states which among other things sharply reduced the amount of spending on cash assistance and freed up a lot of money for a lot of other kinds of spending among the states.

I think a large part of this flexibility will in fact continue to exist no matter what happens under reauthorization. Although attempts to clamp down on the flexibility may create quite a few other problems. Just to give you an example of what I'm thinking about, in fiscal year 2001, 22 states actually could have shifted all of their cash assistance spending to what we call separate state programs under state maintenance of effort funding. It's complicated, but essentially it means that they could have shifted their entire program away from TANF funding and avoided all restrictions, work requirements, time limits, imposed by the federal government.

The flexibility I think has been quite important. It has moved social policy in the United States in an important direction well beyond the sort of limited debates that existed in 1996 around more of a punitive model of welfare reform, and well beyond the AFDC waivers that existed as developed in the early and mid 1990s.

There's been a major shift, for example, away from cash assistance to a wide variety of services. To take Florida as one example, in 1994-'95 fiscal year, 80 percent of Florida's spending on AFDC and related programs was cash assistance. In fiscal year 2001-2002, 20 percent of their spending is on cash assistance under TANF and what we call state maintenance of effort funding.

Child care has been one of the biggest elements of this shift and in an analysis that we've been doing on changes in fiscal patterns among the states, the typical state has more than doubled its childcare spending between 1995 and 1999, just four years. An historic shift in priorities and patterns of spending on human services.

There's been also a major expansion of the clientele served by these programs. Away from the poorest of the poor which were primarily served by AFDC to a much wider range of working families, some who are on cash assistance and many of whom are off cash assistance. Again, especially in the case of child care, but also child support enforcement, post-employment programs and a wide variety of others.

There has been an incredible experimentation and a lot of innovation in structures and processes, what we call second order devolution, front loaded processes that are much more different than they existed before under AFDC, how people come into welfare offices. There's a wide variety of new collaborative, integrated systems that some states have been trying to impose, although there's been fairly limited success so far in actually making them work.

So the limits on this flexibility in large part have been mostly administrative capacity, the funding formulas for certain states as well as state politics, not federal rules. Still I do think there is quite a bit of congruence between national purposes and state behavior, not because of oversight and accountability mechanisms which have been largely ineffective, but largely because of the availability of money and the fact that states have already supported these goals pretty strongly on their own.

I do, however, think that there are real accountability problems, but those problems are not going to be solved, I think, by stronger federal direct oversight, but I think it's problems that have to be dealt with more by creating better systems of governance at the state and local levels. The state/local programs are becoming much more complex, much harder to understand, and especially now that the media are much less interested in sort of a new welfare, it's really critical to provide more information about the nature of these programs, whether they're being implemented, where the money is going, who is being served and who is not being served under these programs. I think the federal government could play a much stronger role actually in helping states, state politics improve the disclosure of these basic characteristics of the programs and who's being reached and who's not.

MR. REISCHAUER: Larry?

MR. LAWRENCE M. MEAD: It's a pleasure to be here. I want to compliment Brookings for hosting much of the discussion surrounding welfare reform. It's been a major contribution I think to the debate in Washington and outside. Much that I'm going to say is based on a new book that I've written about welfare reform in Wisconsin in which I also compare Wisconsin to other states, and I'm very heavily dependent on case studies done at The Urban Institute and also by Tom Gais' state capacity study at the Rockefeller Institute. So I start off by thanking my other speakers literally for providing the evidence for much that I say here.

Let's go on to the next slide.

The way we usually talk about state flexibility is in terms of federal goals. We usually say that Washington has some kind of priorities bur the states demand flexibility. And state flexibility is not itself a federal goal. That's how we think of it. So we compromise. We have a set of federal priorities that we think are very important and then we allow the states discretion on other matters. And in TANF we have a combination of strong federal goals, particularly in an area of work requirements, and extensive discretion in other areas.

So this perspective leads to an assumption that the states choose on matters of policy that are not top federal priorities.

We think of TANF, for example, in terms of the choices states have made. Tom Gais and Kent Weaver, for example, have a Brookings brief, part of the current series about the state paths, approaches that states have taken toward reform. Some states emphasize limiting dependency, others promote work or reducing poverty as their chief goal.

As I look at the case studies I also find it. I find that the big urban states are the most ambivalent about reform. These are states that typically pay high benefits, but they're not sure that they want to enforce work as firmly as Washington does so they tend to be cautious. Their caseloads have fallen, but they typically have limited sanctions and a somewhat more limited approach to enforcing work than elsewhere.

The Southern states, on the other hand, are doing what they've done historically and that is restricting dependency not only by low benefits but by strong sanctions and work requirements as well. That's why we tend to see a strong caseload fall in those areas.

Then there's a group that I'm most interested in, what I call the good government states, those that have the tradition of running government efficiently and effectively. They typically are engaging in what I would call enforcement. They typically have high benefits but they also try to use the benefits plus stringent administrative procedures to enforce work and also child support. We see this in Wisconsin but also in Michigan, Minnesota, Oregon, Vermont, a number of other states of this sort. This personally I find to be the most effective of the reform policies and it's also the one that the public appears to support. So in this sense, in this very limited sense, the good government states perform best.

But in all this analysis we assume that the states are able to choose and that they are all equal in their decisions.

Let's go to the next overhead.

In fact the state capacity to choose is in fact variable. Not every state is able to take clear-cut decisions on welfare reform. As I read the case studies I find that some states quite simply cannot choose or they cannot implement their choices.

As I mentioned already, some of the urban states appear to be seriously divided about welfare reform. New York and California, which are the largest states, are both so deeply divided about this that they had to delegate many of the key issues to their counties. So the major decisions are being taken at the very local level in the two largest states.

There are some other states like Massachusetts would be an example and also New York City within New York, where divisions about reform, and particularly about work requirements and sanctions and time limits are so severe that the community groups who oppose reform are quite simply irreconcilable. They are at a point of lying down in front of the bulldozer to stop this any way they can. So there are deep divisions.

Other states have serious administrative problems. Either there are divisions within the welfare agency where now the agency is supposed to take on a work mission which traditionally it has not taken as a high priority. Some states have tried to merge those two functions. Few have been able to do it as fully as Wisconsin.

And then there are serious problems relating the welfare agency to the various employment agencies that are supposed to carry out the work mission. The most straightforward way to implement these requirements appears to be to let the welfare department do it, and that pretty much was the setup under the Family Support Act. But many states have tried to turn the welfare to work mission over to the voluntary employment agencies. The agencies are now grouped under the Workforce Investment Act, the Employment Service and Federal Training Program. The problem there is that those agencies are not accustomed to enforcing work. They're accustomed to working with voluntary job seekers. They don't have referral procedures. And in the majority of states that try to do this, as far as I can see, there have been serious difficulties.

In some cases, notably in Texas and Florida, there's a virtual paralysis of reform. In fact in the Southern states is where we find the most serious bureaucratic problems, typically because they had the most limited administrative structure ahead of time.

It's the good government states that appear to deal with these problems most thoroughly. They have a less divisive political process and they also have stronger administration. This basically is the Wisconsin advantage as well as other states like Oregon.

Let's go to the next overhead.

So I see a federal interest in not simply promoting federal priorities, but also in doing things that might enhance the state capacity to carry out their role in reform. We should be concerned not simply with whether the states carry out federal priorities but whether they can carry out and execute their own priorities.

Now in the past Washington has tried to do this through a number of suasions. Through trying to raise administrative standards in the '60s. After that, measures to promote the development of management information systems. And more recently, the use of performance standards in Family Support Act and in a more limited way so far in TANF. The performance standards for work of course in TANF were undercut by the caseload fall credit. It seems to me that that has to be taken out in order for there to be serious participation standards.

The level of those standards is probably less important than that they be serious and that the states really have to meet them and that there be publicity. It's a combination of publicity plus standards that it seems to me has shifted the states, including the Southern states, towards a more effective approach to welfare reform. They have been forced really to take on much of the approach of the good government states because they've had to deal with welfare as a problem to be solved. Not simply as a moral preachment, not simply as an interest that has to be satisfied which is often the case in the urban states. It had to look at it as a policy problem. And I say keep the heat on. Let's have there be serious performance requirements. Let's have there be evaluation. Let's have there be publicity, and it seems to me that's the best approach so far to moving the states towards a greater capacity to carry out their own intentions.

MR. ROBERT GREENSTEIN: I think there is a broad agreement that more flexibility is needed in a number of areas, and more than sometimes is recognized, progress can be made in this area.

If you take, for example, the case of the Food Stamp program it now has about 20 state options that as recently as two or three years ago didn't exist. But if there is a need for more state flexibility in various areas the discussion of this issue, at least in recent months, is often too simplistic.

I think we need to step back and ask what the purpose of more flexibility is. Usually it is to serve families more effectively. Then think about the fact that under some of the current discussion there is a bias that federal rules always impede that goal, and that if only the feds would get out of the way states would always include coordination between the programs.

I think that's too simplistic. A few examples.

Following enactment of the welfare law in '96, in a number of states large numbers, probably hundreds of thousands of families that work their way off welfare were terminated from Medicaid and Food Stamps when they left welfare despite remaining eligible for it. It took federal standards requiring that states not terminate families from Medicaid and Food Stamps unless they determined they were now ineligible for Medicaid and Food Stamps, followed by federal enforcement of those rules over state complaints that this was intruding on state flexibility to begin to address the problem.

Similarly, following enactment in '97 of the CHIP program, the Child Health Insurance Program, problems cropped up in a number of states whereby a family would apply for health insurance for its kids for either Medicaid or for the CHIP program, and in many states different state agencies, the state's choice, ran the two programs. And in a number of states if the family's kid was eligible for one program but not the program for which they wanted to apply, they were simply rejected from that program and not enrolled in the other.

Along came a federal screen and enroll requirement to require that if the child applied for one program the state had to screen and see if the child was eligible for the other insurance program and enroll the kid.

Again, there were complaints from a number of quarters that this was an inappropriate intrusion into state flexibility. The feds held firm and the states eventually responded in most cases by doing the right thing and setting up a single joint application for the two programs.

Another example is that in a number of states that have CHIP programs as well as Medicaid, the states have different income limits for when a child loses eligibility for Medicaid and becomes eligible for CHIP. Different income limits for different age kids in the same family. A kid under six and a kid over six have to apply for the separate programs.

We have pleaded with states for five years to have a single income limit, set it wherever they want, but a single income limit for all kids in the same family. Fifteen to 20 states have still declined to do that. I say this not to criticize states or to say the federal government is the fount of all wisdom, but simply to say that the issue is more complex and nuanced than is sometimes presented.

In thinking about this complex area of state flexibility and federal standards, it seems to me that it is useful to distinguish between services and benefits. In services such as what kind of job training is needed, what's the most effective way to move people from welfare to work? Well a search of local issues including but not limited to the local job market come into play, and broad flexibility is in order.

In the area of benefits, federal standards or at least federal minimum standards take on more importance. We have these in the supplemental security income program, we have a national Food Stamp benefits structure which incidentally was established under President Nixon after in the early years of the Food Stamp program some states set Food Stamp income limits as low as 50 percent of the poverty line.

Federal standards are important in other areas as well. For example, some federal benefit programs have rules that prohibit supplantation, that is to say use of federal benefit dollars intended for low income families to replace state dollars in ways that then allow states to use the freed-up state dollars to fill potholes, give tax abatements, or do anything of the sort.

With regard to federal TANF dollars there is no such rule to prevent supplantation and the GAO has now documented this supplantation is occurring.

Another example would be in the WIC program, the Women Infants and Children nutrition program. There is an extraordinary feature in this program whereby competitive bidding is used to procure the infant formula provided in the program. Competitive bidding is now saving $1.6 billion a year and funding the service of more than two million women, infants and children every month through this program.

The competitive bidding is in place in all states only because of the federal requirement which was resisted by states again as an intrusion into state flexibility but despite the fact that the benefits are 100 percent federally funded.

Prior to the imposition of this competitive bidding requirement, in some states the big infant formula companies had deals with state health commissioners whereby they gave discretionary grants to the health commissioners, the commissioners could use as they wanted in return for an agreement that competitive bidding not be used.

In the area of low income housing there are federal standards that give broad flexibility to local public housing authorities but require that minimum percentages of federal voucher and public housing funds be targeted on households that are roughly below the poverty line. Particularly in the area of the housing voucher program, in the absence of such standards in some particularly suburban areas it is likely that a larger proportion of the federal voucher dollars would go to more middle income and wider households than is the case under the standards.

So what do we do? It seems to me that the issue was best framed in a Brookings session a few weeks ago by Sheri Steisel of the National Conference of State Legislatures who said first you try and build more flexibility where appropriate into the program; then you look for places where you can add more state options to the program; and only third do you go to waiver authority. And I would add when you go to waiver authority you try to figure out which basic federal standards to maintain.

The Food Stamp program went from having virtually no meaningful waiver authority prior to the '96 welfare law to having a fairly broad waiver authority, but it has basic standards that do things like precluding supplantation and preventing the elimination of eligibility for whole categories of families that are complying with all work and other requirements and otherwise meet all federal standards.

State options can be promising. In the recently enacted Farm Bill, for example, there's a provision that will allow states now for the first time to use a single definition of income and a single definition of assets across Food Stamps, TANF and Medicaid combined. We should provide states more options to use single definitions across a broader array of programs beyond those three.

But a key here I think is to recognize that there's no silver bullet such as a super waiver. There's no substitute for the hard work of looking carefully, program by program, and across groups of programs to figure out how to expand flexibility without undermining the basic purposes of the program. There is a fair amount that can be done and substantial progress that can be made through such a process.

Part of the process entails figuring out areas where more waiver authority is needed but which basic standards one needs to protect in doing that.

Taking what I would regard as the lazy person's approach, which is what I think the superwaiver is. Not doing the work program by program and simply laying out there blanket waiver authority across everything with hardly any of the safeguards I've just mentioned. I think it's unfortunately likely to do more harm than good.

Finally, as we proceed in this debate I think it's important for the debate to be based on as accurate information as possible. I've been dismayed in recent months to see rhetorically cases being made, in this case it's for the super waiver, based on one piece of misinformation after another. In recent months Governor Engler and Secretary Thompson, I won't go into the details here, have each held up aspects of the Food Stamp program as things that were absurd and showed why you needed a super waiver. For example, Secretary Thompson said we ought to be able to put Food Stamp and cash on the same EBT card. It can already be done and many states are doing it. President Bush in his one example for the super waiver in a speech in Ohio about two months ago or a month ago focused on an obscure Food Stamp rule and said this rule doesn't make sense. Here's why we need a super waiver. That rule is waivable under current waiver authority and at least one state is already waiving it.

So we need to have accurate information which again is [erudite] and understanding what you can and can't do in individual programs. We need to do careful work in identifying what the problems are and what the solutions are. I think if we can get past the buzzwords ranging from the feds are always dumb, bureaucratic interlopers who impede the proper delivery of service to the opposite view, that no flexibility should be allowed to the states, and we do this more careful program by program examination, I actually think substantial progress can be made.

MR. REISCHAUER: Thank you Bob.

While we wait for Senator Carper let me turn to the panel and ask a few questions.

First to Tom. You in a sense described the golden age of flexibility and I am thinking that over the last five years because of the economic circumstances which by and large led to the significant decline in caseloads, states were awash in money and therefore we have seen what happens to flexibility in the best of all conditions. States now are feeling some fiscal strains. Some states are seeing their caseloads increase. You're out there with your toes in the dirt in the fields, what do you expect to happen with this, now that states have the flexibility which gives them the opportunity to do harm as well as good?

MR. GAIS: Yes, actually we are getting some field research reports in right now. States are feeling a financial pinch. They were spending a lot more money in the last couple of fiscal years with pressure to spend down the surplus both in Washington as well as in their home states. They have, in fact, done that yet their budgets are being crunched quite badly right now. The programs, the TANF programs are being cut. Some of the states are, I know Michigan for example is cutting back on the fatherhood program, clothing allowances for kids, a variety of other relatively small programs individually, but taken together it's quite a few programs total.

The money is the critical issue for flexibility. And what states have not been able to do so far is be able to understand the changes if they were to plan well for the volatility and the amount of money available to them over the years, and the sustained programs over a large period of time. What's happening is we have programs that get rolled out for awhile and then get cut back or eliminated.

So again, one of the restrictions on the exercise of state flexibility really is administrative capacity as well as the money. States have a long way to go before they really develop that ability to do so.

And you're right, the period of time in the last several years with the large amount of surpluses that they had has permitted this great innovation.

So I guess in the future if there are stricter controls, stronger work requirements, so on and so forth, as a result of reauthorization and if the economy doesn't pick up, then the state's flexibility will be decreased pretty dramatically. But if the economy recovers pretty well then there will be quite a bit of flexibility still even under some of the more restrictive, reauthorization proposals.

MR. REISCHAUER: Larry, you sang the praises of flexibility, then said we should have performance standards and be tough on performance standards. We have a long history in this country of establishing performance standards, having sanctions placed on the books, and yet never enforcing them. When you have a great variety of activities going on in the different states, it's hard to really match the activities with the standards in a one to one match. How realistic do you think it is to rely on sanctions?

MR. MEAD: Well, the argument for performance standards is that they fall between a posture of detailed regulation on the one hand and outright state discretion on the other. The advantage of employment standards is that they pose welfare as a problem that states have to solve. They have to achieve a certain level on the standards, but they are less free to decide the best way to do that.

So I think a posture performance standard places more pressure on the states to develop their own policymaking capacity than either the posture of regulatory details or outright flexibility.

Now Bob, you're right to say that we often haven't enforced the standards in the sense that we haven't always levied fiscal sanctions against states that fail to fulfill them, but the adverse publicity that surrounds that and the fact that the states are put on record as being more or less successful in achieving a certain performance standard still exerts tremendous pressure on them to clean up.

In the case of fraud abuse standards, I doubt that we ever actually imposed a penalty on a state and yet the states cleaned up. The same is true for other performance measures we have.

I would say the performance measures under the Family Support Act, under TANF did exert some pressure on the states to get their act together and improve, even if we didn't actually take any money away, or not much money away from the states.

So when standards combine with publicity it really does the job.

MR. REISCHAUER: Senator, as soon as you get wired up I'm going to give you the floor. The other three panelists have had an opportunity to provide some opening comments, and I will give the same to you —

SENATOR THOMAS CARPER: I really don't want the floor. I just want to go right to the question, okay? I have heard myself talk plenty of times about this stuff and I really want to hear what other people have on their mind.

MR. REISCHAUER: Bob Greenstein, you spoke a lot about waivers and there is really a basic choice between providing flexibility in the basic law, in other words giving states the freedom to operate on a broad band of policies, and having states apply to the federal government for a waiver in which there is some stricter individual review, there is a feeling that this will be monitored over time. Are you suggesting that we should rely more on waivers while tightening up the fundamental law, the amount of flexibility in the law?

MR. GREENSTEIN: Just the reverse. What I'm suggesting is the laws themselves should be made more flexible. Again, taking as an example some of the recent changes in the Food Stamp program. If you look at the changes over recent years we moved some of what were rigid federal rules such as what the vehicle limit is and so forth, from automatic, uniform federal rules every state had to comply with to minimum standards and then given states the flexibility to alter those rules and conform them to what they do in TANF or Medicaid itself.

I think what I'm saying is that there are certain areas that are of pretty fundamental importance, and for two reasons, both because Congress should have some role and because waivers where states have to come and apply are not the ideal approach. Where you can identify things, that you can build flexibility into the basic law, either through its underlying structure or through state options, and you can do that without undermining the purposes of the program, that's preferable to waivers.

I also think an issue on waivers Bob you just referred to while ago, monitoring. That depends on how the waiver authority is set up. In something like the super waiver proposal, rather than the notion of testing some innovative approaches, first in a limited number of areas in a careful manner with an independent evaluation, and then if the results show certain things, spreading them or giving states the authority to spread them nationwide, it effectively takes almost everything in all of the covered programs and says an unlimited number of states can change anything on a state-wide basis without even a requirement for an independent evaluation.

So I also think that the term waiver is a broad term, and it can accomplish anything from a carefully evaluated demonstration project from which we seek to learn things to guide national policy, to sweeping changes without much evaluation. There's a role for both but each one is more applicable to certain kinds of changes.

SENATOR CARPER: As an old governor — Fifty states, 50 laboratories of democracy, we ought to use them and find out what works and what doesn't work. At the National Governors Association we established a Center for Best Practices to try to identify what's working and to provide a mechanism so the rest of us could learn to steal the good ideas and claim them as our own.

With respect to waivers, President Clinton in his Administration gave a huge number of waivers to a bunch of states including to Delaware. In fact far more than two-thirds of the states had the opportunity to experiment boldly.

In Delaware we used our own money to provide for, to let people get some extra educational time, for — Sometimes we think only the federal government has money in this game. The states have their own money. Some of them have quite a bit of money and they ought to be able to use — They can use their own money to experiment boldly as well, outside of the federal laws.

This guy right here, I think he used to do something with the Food Stamp program, didn't you? In your youth?

The Food Stamp [inaudible] part of the Farm Bill we just passed. It had some provisions I didn't like. One of the provisions I liked was to try to simplify the Food Stamp program and provide a little more flexibility there. There are a couple of different ways we can go at this.

One, the states that have existing waivers, we're going to extend your waivers. The ones that are in can extend their existing waivers.

Two, states use their own local money, own state money and the flexibility that that entails.

We can provide more flexibility in the Food Stamp program, more flexibility in TANF for our states, or we can go down the path which the Administration has proposed with the super waivers, which the governors I think like generally. I think the Republicans put it in their bill.

In the end I think there's probably going to be enough consensus that we want flexibility that we could figure out what to keep and what to throw out.

There are concerns about the super waiver authority that might lead one to want to tear it back a bit. Again, the idea is to see what works, to test it, to demonstrate it, to model it, and then to ride with it.

MR. REISCHAUER: Is there a point where we should worry as a nation about inequities that might arise between the states when benefits, opportunities, services begin to vary tremendously between one state or another, although they're all being financed largely by federal dollars? I think that was one of the worries with the AFDC program and with more flexibility —

SENATOR CARPER: I think we should be mindful of that. We also need to keep in mind there's another side to the coin. States like Delaware and Connecticut and other states who have very high incomes, we get relatively modest federal dollars in all kinds of programs. But we have more ample resources on the state side and we spend them.

MR. GREENSTEIN: Larry just mentioned, continuing the conversation, reiterate a finding made earlier that I also think that there are some differences between the delivery of services and the delivery of benefits.

In the case of delivery of services, education. How do you move people from welfare to work? What's the right job training program? It doesn't really make sense to have, I think, extensive federal standards. Job markets vary, capability of education and training providers vary, all sorts of things vary by states.

In the case of benefits, I think there is a strong case for federal minimum standards, and to use the Food Stamp program again as an example, and I agree, the Senator largely designed most of those dozen new options that are in the new law.

VOICE: We should have let you design the rest of the Farm Bill. [Laughter]

MR. GREENSTEIN: In the case of the Food Stamp benefit structure, here's the only national benefit structure we have that is a floor under nearly all the [inaudible] and in which children have the same benefit structure essentially as the elderly. In every other federal and previous state benefit program that were more categorical, the elderly did much better than the children.

Now I wouldn't for example want to see a waiver authority that would enable, because this was particularly popular in a given state, benefits to be reduced for children and shifted to the elderly or reduced for children and used for supplantation to replace state dollars.

So I think, and I don't think we're saying different things. The question is sort of figuring out what are the standards in the federal rules that are important and need to be changed, and then where are the areas where one can provide the flexibility, and where one can do it in a way that a state doesn't even need to ask for a waiver. You build a basic option into the law. I presume governors would always prefer to just have the option of having to get the waiver in the first place.

MR. GAIS: I do think there is an important issue here having to do with inequality and differences among the states, and it is somewhat troubling. One of the issues had to do with the funding formula. States for some very good reasons have very different levels of funding per child under the TANF block grant. Even though there are very understandable reasons, because they're related to historical patterns of spending, they do have consequences in terms of the strategic choices and the policies that have been adopted by the states. Kent Weaver and I have been working on that and the analyses have found that states which have relatively large block grants for a poor child are the ones that are getting higher, that are offering higher earning disregards that are more likely to offer state earned income tax credits and also to increase their discretionary spending on child care under the TANF block grant.

On the other hand the states, generally the Southern states but also some of the Western states that have the smallest block grants per poor child, are the ones that are more likely to [inaudible] in what we call the stick, have shorter than federally required time limits, stronger sanctions, and a variety of other mechanisms to try to get people off the caseloads.

I don't think that's going to be addressed except to the extent that there are some efforts to increase the supplementary spending package, the supplementary spending under the reauthorization bill which I think is a very good thing and begins to deal with the question, although it doesn't resolve it completely. But it does underline some of the points I think that Bob Greenstein was making though about the importance of maintaining a kind of pluralism within a system. Where some programs do give states a lot of flexibility and some programs really maintain a safety net and pretty strong federal controls.

Just one aspect of that, to take an example, is what you can find in the South. The South has had the biggest decline in TANF welfare caseloads. However, they have had smaller than average declines in Food Stamp caseloads. In fact the average Food Stamp benefits have increased in the South in the last few years.

So the pluralism in the country has actually allowed some of the inequities in one part of the system to be dealt with and compensated slightly by having an entitlement in another part of the system. For that reason I think it's very important to maintain this kind of pluralism which allows both flexibility as well as uniformity in the U.S. social policy structure.

MR. REISCHAUER: Senator, as a former governor and as a leader of welfare reform efforts in the Senate, there is a bill that's come out of the House. You're now dealing with it in the Senate. Do you hope to see more or less state flexibility in what the Senate produces than what the House has produced?

SENATOR CARPER: I think the House has included in it the super waiver that we talked about earlier and that gives states a whole lot of flexibility. That flexibility is so great that it frankly causes some discomfort in some quarters, including with some folks here on the podium.

I mentioned earlier, we can provide states flexibility by extending their existing waivers. We can provide states flexibility by making more flexible the TANF programs as we reauthorize it, the kind of changes in the Food Stamp program that Bob alluded to. States can use their own dollars for flexibility. Or we can do some variation of a super waiver. I don't think in the end we're going to pass what is in the House Bill with respect to super waiver, but we've got to be smart enough to figure out what there makes sense and what doesn't.

As an old governor I found the four things that really are the most critical in helping people move from welfare rolls to payrolls and stay there: one, a job; two, a way to get to the job; three, who's going to help me with my childcare; and four, health care. If you expect somebody to move from welfare where they have Medicaid to go to a low-paying job that has no health care, forget about that being a successful experiment for long.

We've got to make sure, especially, this is the major difference between where Evan Bayh and I and others are, and the White House and the House. We put a significant amount of money in child care, a significant amount of money in child care. And whatever comes out of the Senate Finance Committee maybe later this year I think will provide significantly more money for child care than was thought by the Administration or included in the House bill.

MR. REISCHAUER: But we're being more prescriptive with respect to how you get from welfare to work, how much you have to —

SENATOR CARPER: Not in the —

MR. REISCHAUER: I know. In the House proposal. I wanted you to talk a bit about the Bayh/Carper proposal.

SENATOR CARPER: In terms of the key differences, one, adequate resources for child care, a significant difference.

We increased the money in something called the social services block grant which used to be funded at about $2.8 billion a year, it's been ratcheted down over the years to about $1.7 billion. We use that money in my state real flexibly. We use a lot of it for child care, but it can be used for some of the flexibility we want to give to states. Evan and I ratcheted it back up to $2.8 billion.

Among the other differences, I like to talk about Delaware. Please excuse me, it's what I know best. But in Delaware we comply with the existing federal welfare law. There's a requirement that says you've got to have half of your welfare caseload working or doing some kind of work-like activity. Those activities are actually defined in the '96 law.

We also say to states like Delaware, you can offset that work requirement by counting it as a credit against that 50 percent work requirement, you can credit the people that have dropped off your welfare rolls since 1996.

As it turns out, over 55 percent of the people who were on the welfare rolls in Delaware in '96 are gone. They either timed off, they found a job, they've gone on to other employment, they've moved, they got married, quit working, but they're off — So our caseloads are down by about 55 percent. It more than meets the 50 percent work participation rate requirement under the federal law.

As it turns out, a bunch of other states meet their work participation requirements, the 50 percent requirement under federal law because they've reduced their caseloads. In some cases they've reduced their caseloads by helping people get jobs and they're still working those jobs. In a lot of cases they just timed off. Those people are going to be timed off, they dropped of the rolls, and states don't have an effective work participation rate requirement.

Evan and Tom say here's what we ought to do differently. We ought to [inaudible] the work participation rate requirement from 50 percent of your caseloads to 70 percent over the next five years, and states, rather than being able to count the reduction in the caseloads as a full credit to offset that work participation rate, they say let's exchange that with an employment credit so the states get credit for helping people actually move to a job. They get a bigger credit if they move people to a job that pays at least 50 percent of the average wage in a state. You get credit if you help somebody go to a job, even if they're still getting child care or transportation money out of your TANF dollars you get credit for that as well. If you make sure that a non-custodial parent is providing for assistance, supporting their custodial child who used to be on welfare, you get a credit for that.

The idea is to use the workplace credit to incentivize states, to move people off of welfare into real jobs, particularly into better-paying jobs, to incent folks who are non-custodial parents who have an obligation to kids on welfare, to incent them to provide their child support. And to even let people count as a credit for sending some of their caseloads to school, post-secondary training.

Those are the kind of things we ought to be doing, and that's what we do in our proposal.

MR. GREENSTEIN: I wanted to get back to your question because I think it's a fascinating one about flexibility versus prescription, in legislation before the congress, particularly the House Bill. As Senator Carper mentioned, the House bill has the super waiver proposal, but in almost every other respect the House bill is a remarkably prescriptive bill. It greatly reduces the flexibility that states currently have to structure their welfare to work program.

If you take those state programs that based on independent evaluations are the most effective. Take the Minnesota program. There's probably no program for which there's stronger independent evaluation of positive impact on employment, earnings, child well being, even potentially marriage and family [inaudible] than the Minnesota Family Investment Program. The Minnesota Family Investment Program, would have to be ripped up under the House bill. It doesn't meet the much more rigid requirements for how to do welfare to work programs under the bill which led your former colleague, Governor Ventura, to write an OpEd piece recently saying we're doing it, it works. Why should we have to change it?

If you look at the markup today in the Finance Committee, what's interesting is that some of the people who will offer proposals that include the super waiver are going to vote against almost every other amendment that would provide greater state flexibility. There will be an amendment offered by Senator Graham of Florida to give states the option, just an option, to serve legal immigrant pregnant women and children in their first five years in the country in Medicaid and SCHIP. I think a similar provision is in the Carper/Bayh Bill.

SENATOR CARPER: We don't mandate it, but as you say, if you want to use some of your TANF money for health care for illegal immigrants, you can do that. You don't have to. We'll say the same thing with respect to cash assistance.

MR. GREENSTEIN: The provision is generally opposed by the same people who favor the super waiver who tend to be the same people who favor the more prescriptive House provisions on the welfare to work program.

So we kind of have a conundrum here.

It reaches its zenith when the argument gets used, don't worry about all those prescriptive welfare to work requirements in the House bill because you could waive them away with the super waiver.

First, it's not clear you could. The federal government will have to approve such waivers, and if they feel as strongly about the prescriptive work requirements in the House bill as they say, there's no assurance they would grant those waivers.

Which gets back, I think, to the point of why building flexibility into the basic programs is a superior mode to waivers in the first place. Senator Carper mentioned that as governor he got a lot of waivers from the Clinton Administration. An area where a particularly large number of waivers were granted, and I don't know about Delaware in particular, was in the area of Medicaid and health care.

When you look at the waivers the Clinton Administration gave, they tended to operate on the philosophy that there were certain minimum federal standards that needed to be protected, and states could then experiment in ways that enhance things.

Today there are starting to be waivers developed in Medicaid that would actually pare back Medicaid coverage and services and use savings to fill holes in state budgets. There's a question of whether those would be allowed, but that's a very different kind of waiver and there are reports, which I must say we've been told this by state officials, I've not confirmed this in writing and on paper. We have been told in the last month by a few state officials that it has been indicated to them by the Administration that certain waiver requests would be looked upon more favorably if the state backed administration legislative proposals on Capitol Hill in the welfare debate in return, which is in my view an abuse of power in terms of the waiver authority.

So all these are reasons — Not that one shouldn't do waivers, but where you build the basic flexibility to the degree you need it into the programs in the first place. You don't make them more prescriptive as in the House bill and then do a completely wide open super waiver and say that's the answer.

MR. REISCHAUER: I'm going to turn to Larry and then go to the audience for questions, so be thinking of good questions to ask these gentlemen.

MR. MEAD: This discussion mostly has taken the traditional form of federal priorities versus state flexibility which you've seen [inaudible]. But as I was arguing earlier, it is also a federal priority to in some way empower the states to use well the power that they have to make their own welfare policies. The problem with the caseload fall credit in the '96 Act was that it freed the states from the need to do that. They could meet the federal, the new TANF work participation requirements simply by having the caseload fall, and it might fall for any reason unrelated to policy. Of course it fell for a number of reasons unrelated to policy, particularly good economic conditions.

Now it seems to me absolutely critical, and apparently this is going to happen under all the bills, to withdraw the caseload fall credit and have the states really subject to some kind of participation requirement under the new Act.

IF they do that then they have to actually make policy to decide how to do that. They have to decide what's going to be expected of people; and secondly they have to implement it. Those are the two key decisions. And it seems to me they ought to be under pressure to do that and it should not be some new cop-out, some new caseload fall credit.

Now the governor is suggesting another kind of credit where you get credit for employment or for other good outcomes. That strikes me as moving in the right direction. When there's some concrete thing that you have to try to achieve, and then you have to make policy for that. You have to carry out policy for that.

I think if we put the pressure on the states to perform in this manner we not only advance federal objectives but we also build up the state capacity to achieve their own objectives, which it seems to me is just as important.

MR. REISCHAUER: The floor is open for questions.

QUESTION: Hi, I'm Mark Milsoff from the state of Maryland. My question is for Mr. Mead and maybe a follow-up to Senator Carper.

I heard you talk about performance standards and how that could be a way the federal government can get some accountability out of states who are implementing their flexibility. But everything you've referenced I think has to do with process measures. Process accountability measures. And I think the employment credit, although I'm supportive of that concept, is a way of mitigating an existing performance measure that is process based.

Why is there not a movement towards outcome-based performance measures where states are rewarded for moving people off welfare and into jobs regardless of putting people into work-like activities while they're on welfare?

MR. MEAD: I think it's a good idea to have performance measures of that kind also. I'm not saying those aren't important. In fact there were going to be such measures, you may remember, on the Family Support Act, and again under TANF. We haven't seen them yet for various reasons.

I do think the work participation standard is independently important because it is a requirement I think of the politics of welfare reform that more of the adults who are getting aid should be employed, should be involved in what is called work activities. You can dispute how that can be defined. You may call it a process measure. I think it's also an outcome measure. Work in and of itself is an important outcome. People are participating in something that makes demands on them, involves some kind of give-back to the community. That strikes me as valuable and important. I would also agree that performance measures having to do with the private sector job placements, wage rates, retention rates, these things are all of value.

SENATOR CARPER: The only thing I would say, if you look at the employment credit that's in the bill that Senator Bayh and I have authored with others, we replace the reduction in caseload that lets people sort of wipe out, states wipe out their work participation rates, with an employment credit that actually says state, you get credit against your work participation rates by actually moving people off of welfare into a job. And they can't just be there for a day or a week, they've got to be there for awhile.

Secondly, states get more credit if you move people into a job that's paying at least 50 percent of the average wage in the state.

So the idea there is to incent, not just to see people time off the welfare rolls, be sanctioned off. The idea is get credits for the work participation rates by getting people into jobs, better jobs, [inaudible].

QUESTION: Hi. Mary Guiden with StateLine.Org.

My question is for Mr. Greenstein. You mentioned that there's no silver bullet solution to this and that state programs really need to be looked at on a case by case basis. You also made the argument that there's really a need for federal intervention. So I just wondered when you have to craft one program that fits all, what's your solution for how Congress and the Administration should address the issue?

MR. GREENSTEIN: Let me clarify. I didn't mean to say if I did, case by case basis. What I was talking about was you need to, when you're designing a broad remedy such as a broad cross program waiver authority you have to be careful to look program by program to determine what are, if any, and in some cases there might not be any, in other cases there surely will be, what are the federal standards in that program that you need to be careful not to waste? Or if for example you want to say well we want to be sure that this flexibility we're granting doesn't simply become a license for large scale supplantation.

If you simply wrote a blanket provision in, there shall not be supplantation, you probably couldn't enforce it. You have to do the work of saying where are the vulnerabilities? Where are the program intersections that if we don't pay attention to we're going to open the loophole for supplantation. You figure out what they are and you kind of close them to begin with.

So for example if you look at the Food Stamp program which prior to 1996 had no significant waiver authority, when the '96 Welfare Law substantially broadened that waiver authority among the things it wrote in you don't actually ever find words that say no supplantation. But they figured out where that could occur under the waiver authority and they close it off to those possibilities.

So I'm just saying that you need to do the work.

We've talked about the Food Stamp provisions of the Farm Bill. Two or three years ago there was a fair amount of complaints by a number of states about various rigidities in the Food Stamp program and a lot of it was at the level of it's too rigid, it's not flexible enough.

Let me give credit to a couple of outstanding people who are generally in the room from our staff at the Center, went state by state and talked to people, state officials, and tried to understand what really are the specific problems here and then figure out how we could craft a whole array of state options to cut through a number of those problems. That's the kind of thing that can be done. It can't be done simply by legislating a one paragraph provision in the law. I think it's a combination of figuring out what you want to do perhaps in welfare law that doesn't throw the baby out with the bathwater but it also is a process we need to start that probably can't be done in 30 or 60 days to sort of look carefully across program by program, across programs, and identify those areas.

I think where one can identify them there is now a very receptive bipartisan atmosphere on Capitol Hill to providing that additional flexibility. I think there was nothing more bipartisan in another life than for instance Farm Bill than the added flexibility in the Food Stamp program. Everybody, right, left, center, both parties, was for that, states were for it, groups that work with clients were for it. There really is I think a lot of progress that can be made there.

QUESTION: Hi. My name is Jennifer Laverall. I work in Housing Policy and I'm partly here because TANF has been proposed as a model by some for the reform of housing policy, particularly the devolutionary aspects of TANF.

Dr. Gais, you said something that struck me as odd from a dynamic standpoint, and I may have misunderstood the essence of what you said. But I think you said that accountability will not be overcome through federal oversight, but through better performance at the state and local level. I think that's what you said.

I don't understand whether you meant performance standards aren't necessary or what you meant by federal oversight, or whether there's something about devolution that there's more competition at the local level or there's a role for the private sector and you meant that there's some pull toward efficiencies from the grassroots up.

MR. GAIS: What I said was that the people at the state and local levels, that these have become very complex programs and quite different from one state to the other, and to the extent that there is going to be strong accountability to [inaudible] state administrative system standards, I think it really requires much stronger, more open, more vigorous state and local level politics. These are becoming in some ways very hard to understand systems and in some ways almost opaque systems at the state and local level.

I don't think that the federal government has a whole lot at stake in being sure that states are running the most efficient child care outreach programs. I do think there's a lot of people at the state level who care deeply about whether those programs are reaching the intended target population.

So what I think is really important is to be sure that we have better information at the state and local level about the program, about who's being served, who's not being, about changes in poverty level, about changes in participation in various programs. It would also include understanding what's happening to people with respect to housing.

I do think we've devolved responsibility and control down to the states under TANF yet we have not devolved the availability of basic programs and information about families to the state and local level. I think we have to do that.

The federal government has some oversight and accountability mechanisms under TANF. I do not think they've been particularly effective. I don't think the government really has a lot at stake in enforcing those rules very well.

I do think people at the state level do care, but in order to make those political and accountability mechanisms effective, we just need better routine basic information about who's being served, who's not being, how these programs are being implemented, where the money is going, and also more broadly what's happening to all these families?

I think the gentleman spoke earlier about work participation rates so why aren't we getting good outcomes, why don't we get better outcome information?

I don't know whether they should be enforced by the federal government but I do think we need better outcome information at the local level and I think one of the consequences if we were to do that would be to understand, for example, how important housing is to a lot of low income families and how far up the income scale those housing problems are really biting into family income.

But we really don't. Most of the programs and most of the oversight and most of the surveys that were enacted under TANF were really national level surveys and don't provide good state level information except for the very largest states.

SENATOR CARPER: Can I make a quick followup?

Housing. I'm on the Housing Subcommittee and the Banking Committee and I have a real strong interest, and I'm sure you do too, probably a lot of you do on affordable housing. We're on recess next week. I'm going to use my recess next week back in Delaware to hold a series of housing summits. At least one of those housing summits is going to getting a report from the Delaware Housing Authority which has been experimenting rather boldly with respect to subsidized housing. What we're trying to do is use the TANF model where we actually time limit the amount of time that people can be in public housing. And during the time that they're in public housing we do a number of things to enable them to increase their earning power. Education, job placements, job support, that kind of thing. The idea is to — We even set up these accounts for people so they can, instead of as their income goes up instead of paying more rent as you normally do, your extra marginal income goes into this account and you can draw on that account when you leave public housing for closing costs on a house, some use it for a security deposit on your own apartment on the outside.

But the idea there is to try to move back closer to the model we had for public housing back immediately after World War II. The GIs were coming home from the war, set up public housing, they're there for a period of time, transition into housing that they would own.

We've been doing this for a couple of years in Delaware and our housing summit is designed to let us learn how it's working. Again, laboratory of democracy. We'll see how it's doing, what lessons we can learn and go from there.

MR. REISCHAUER: Before going to the next question, I'm interested that all of the panelists have mentioned the need for more data, information on outcomes, performance. I wonder how realistic this is. Tracking the outcomes of low income people, particularly when they have left the income support component and when many of them are mobile, particularly in a small state like Delaware and might be in Philadelphia or Maryland or Washington. One wonders whether we should rest so heavily on procedures that really require information that never will be up to the standards that we might like.

QUESTION: Al Millikan, Washington Independent Writers.

One of Professor Mead's previous books was the new paternalism. How paternalistic is current welfare reform with previous welfare reform? Is there any possibility of real fathers being able to show their own paternalism? [Laughter]

MR. REISCHAUER: Do you want an unbiased answer from Larry? Or should I turn to —

QUESTION: [Inaudible]

MR. MEAD: I'm disappointed that the reauthorization has not paid more attention to the buildup of fathers programs like Parents Fair Share or Children First in Wisconsin. These are child support enforcement programs that basically say to fathers you have to pay your judgment or you have to go to jail or we offer you a third option which is participate in this work program.

There's some evidence so far that these programs do succeed in increasing child support payments. It's not clear they increase employment. We're clearly not doing as well with the fathers as we have done with the mothers. But I would have liked to see more money invested in these programs at least at a developmental stage as we are proposing to do for marriage programs. So this is a disappointment. In that sense I agree with you.

MR. GREENSTEIN: To me this is another example of an area where actually the House bill is too inflexible. There is a fund, is it $300 million a year, Belle? For marriage promotion. And as I understand it, it's designed —

MS. SAWHILL: [Inaudible]

MR. GREENSTEIN: It's designed in an inflexible way such that a state could not elect to use the money for things ranging from trying to combat teenage pregnancy to trying to do things with non-custodial fathers, ranging from improving their employment, thereby trying to collect more child support and the like.

This is a perfect example where there's a fund in the bill but there are some other uses that might yield better results, at least which some states believe might yield better results. People like Belle, I think, would agree, than simply mandating that the money only could be used for a narrow range of marriage promotion activities. Why shouldn't we give states more flexibility on the use of those funds? I think we should.

SENATOR CARPER: The '96 law that was passed really has ensured that we expect a lot of moms and we say you've got to, we're going to time limit your benefits, we're going to make sure you get into job training, you've got to go interview for a job, if it's offered to you you've got to take the job, if you quit your job we'll sanction you, and eventually sanction you off of welfare altogether. We don't do a whole lot with the guys.

One of the things we did with the '96 law, we tried to fix it and say heretofore if you were a father and your family needed welfare benefits, cash assistance, the only way you could really get it for your family is to desert your family. We've fixed that so that you can receive cash assistance with a two parent family. And we've done some other things that actually make sense to try and encourage dads to be a part of the life of their families.

In my state, one of the things that we've done that I like — It used to be if you had a child out of wedlock in order to be able to acknowledge paternity you had to go to court. A lot of these guys [inaudible], they don't want to acknowledge paternity anyway. And if you want to acknowledge paternity you've got to go to court. Forget that. Now we let people acknowledge paternity in a hospital setting. We've got thousands of dads who acknowledge paternity for their children. At one time a lot of guys wouldn't show up, if they're not married to the mother of their children. They'll show up when the baby's born.

What we do in my state and what we do in a lot of states is say at that time, let's give you a chance to acknowledge the paternity of your child that you fathered.

The other thing, Evan Bayh, Evan's not here so I'll speak for him. He's been a huge champion of this fatherhood stuff. I won't go into all the details, but as a father myself, I find one of the great joys in life, maybe the greatest joy that's in my life, is what I get from my kids. They're 12 and 13. They're getting older now, but still a huge joy that they give to me. A lot of times we say to the dads we're going to be punitive. If you don't show up, you don't pay your child support, let's put you in jail, we'll do this, we'll do that. Part of what we need to do is to show guys what they're missing by missing out on their children's lives.

One of the neat programs we do in Delaware, we do parenting training in all of our prisons. All of our prisons. Most of the guys that are in prison are fathers. Most of them are not very good fathers. We try to make sure that when they leave prison they don't just leave as better criminals, they leave as better fathers.

The last thing I would say, and the welfare bill, we tried to fix this with the bill that Evan and I have coauthored, deals with we'll say Bob's my child here, he's on welfare over here, and I have an obligation to pay child support but I don't. And if I did, the money would go to the state, not to your family. We changed that. We say that you pay your child support, your family's on welfare, the money goes to the family. And that's actually, I think that [inaudible] whole lot more [inaudible] benefits. We hope so.

MR. REISCHAUER: Thank you.

QUESTION: Alan Weil, The Urban Institute.

I'm interested in the back end of the experimentation metaphor. If we're giving states flexibility so they can be laboratories, then we conduct experiments and presumably we learn some things. And there are two models. One is you leave it to the governors to steal each other's ideas, which they do a lot of; and the other is the federal government steps in and says we've learned something and we expect states to come up to this standard or take this approach because we now have evidence that it works.

Bob gave an example of Minnesota where it looks like we've learned something and there's a risk. We're not going to let the state do it. But I'm interested in the other option. Is there anything we've learned in the last few years of welfare reform that we could feel strongly enough that we would actually go to the states and say you need to do this because it works in welfare policy.

MR. GAIS: I think a lot of the states are already doing this, and also some of the reauthorization bills are increasing spending for child care. I think a lot of the states have been finding that is the root of a major problem for moving people into work and keeping people into work. A good many of the bills that are being considered have increased the spending for child care. So there's no doubt that that is being learned and the federal government is responding to that.

A lot of the money is in fact going into the mandatory child care component.

Beyond that, I think it's more complicated.

MR. REISCHAUER: There's also a very real possibility that what you're learning is it doesn't work so don't waste your money on this. Or it works, but Delaware isn't the same as Montana.

MR. GREENSTEIN: When Alan finished his question there was a second of silence on the podium. I think part of what is reflected, in addition to the fact that, you know Alan better than any of us, there are a range of different things states have done, there are a range of different evaluations and there might be some controversy in saying we know the following three things worked so well that they should be required.

But the larger plane, I think, is that in the current political environment in the area of TANF, even if one had definitive evidence from a range of evaluations, it is simply against the current [inaudible] to mandate or require anything like that.

There's an interesting test, in a sense, coming up perhaps in the Finance committee today. Here's an interesting dilemma. I forget which Senator's going to offer this. There's an amendment listed to be offered that would require that states not discriminate, if you will, against two-parent families. That they not have more restrictive eligibility criteria for two-parent families than for single parent families.

Now as a general rule I think an overwhelming array of policymakers would agree we shouldn't do that, but should one make that a requirement in federal TANF law? It will be interesting to see how the vote goes. I suspect that something like that will not make it into the final law.

SENATOR CARPER: Let me take sort of a left-handed approach in answering your question. I was elected governor in '92. I had been in the House for ten years before that. I couldn't wait to be part of the NGA, the National Governors Association. I went to my first NGA meeting two months before I ever got to actually be a governor. I just couldn't wait to be a part of it.

I went to New Governors School hosted by Roy Roemer out in Colorado in November of '92. There were about 10 or 12 of us, new governors, and had a faculty of old governors, grizzled veterans and their spouses to teach us newcomers the ropes. I could not get over how, and we had a lot of meetings in rooms much smaller than this. No staff, in some cases no media. Just governors trying to help each other. The old guys and gals saying this is where I screwed up, these are the mistakes I made, trying to help the new kids on the block not make the same mistakes.

I was struck by how that goes across party lines and across geographical lines.

The other thing that struck me is the healthy spirit of competition between the states. Not competition in a negative sense, but in a very good sense. We actually really do compete, try to make sure our test scores are going up, make sure that our welfare rolls are coming down, our teen pregnancy numbers are coming down, to make sure that our recidivism rates in our prisons are coming down. We like to steal ideas from each other, borrow ideas, but there's really a good focus on what works, a willingness to share, but also this great sense of competition. A wonderful dynamic. Better than anything I ever expected.

MR. REISCHAUER: On that positive note, let me bring this to a close. I want to thank the panel. It's been very informative. And thank you, audience, both those in the room and those who are watching on Webcast. There undoubtedly will be further events in this series.

Thank you.