THE CENTER FOR NEW YORK CITY AFFAIRS

NEXT STEPS:
New York City Welfare Policy After Giuliani

Transcripts:
DAY 3: Politics, Public Opinion and Leadership for Change
A Roundtable Discussion
Tuesday, June 12, 2001 9:00 a.m. to 12:00 p.m.

This is part one of a three-part series of miniconferences for practitioners, policymakers, researchers and communities to share their visions for the future of social welfare policy in New York. Presented by The Center for New York City Affairs and The Community Development Research Center, MilanoThe New School for Management and Urban Policy

Presenter's biography:

Gordon L. Berlin is currently Senior Vice President for Work, Community, and Economic Security at the Manpower Demonstration Research Corporation (MDRC), a research and demonstration intermediary organization which tests new approaches to the nation's social welfare problems. In addition, he currently serves as Project Director for MDRC's study of welfare reform in urban areas (Urban Change), the multi-site test of employment programs for non-custodial parents (PFS), and the Canadian Self-Sufficiency Project. Before joining MDRC in 1990, Mr. Berlin was Executive Deputy Administrator for Management, Budget and Policy at the New York City Human Resources Administration.

---

EDWIN MELENDEZ, DIRECTOR, COMMUNITY DEVELOPMENT RESEARCH CENTER: I would like to welcome you on behalf of Milano, the Community Development Research Center and the Task Force for Sensible Welfare Reform who are co-sponsoring this event this morning. This is the third conference that we sponsored on the future of welfare policy in New York City, which started by reviewing what we thought was the status quo, what has happened in the last few years and from that we moved to models for the hard-to-serve service interventions.

Today, we would like to focus on what will happen after Giuliani. What is ahead for the City? I would like to say just two words about The New School, more precisely about Milano. We are a management school with programs in nonprofit management, health and human resource management. We also have a program in urban policy, both on a Master and Ph.D. level. And we are trying to forge links with all the community professions in NYC. This is one of many events that the Center for Urban Affairs in New York City has sponsored this semester. We are making an effort to develop professional development programs in a number of areas like workforce development and nonprofit management and we will welcome your suggestions as to how to continue these activities. It is my pleasure to introduce to you Andrew White, who is going to explain the program for today. And introduce our first speakers. Thank you very much.

ANDRW WHITE, DIRECTOR, CENTER FOR NEW YORK CITY AFFAIRS: Thanks to all of you for coming. This is unusual in that this was all by word of mouth and invitations as opposed to a wide broadcast invitation so I was a little worried that folks, who said they were coming, wouldn't. But it looks great.

The Center for New York City Affairs exists to promote civil, meaningful dialogue on important issues facing the City and in particular in issues affecting low income neighborhoods in New York City. And the Task Force for Sensible Welfare Reform, which this project is a part of, was convened at Milano several years ago and published several reports. This is in many ways the follow up on those reports, which will all be posted on the Web later this summer, since we have run out of printed copies.

I want to thank Isabel Pradas for doing most of the coordination of this and the other sessions and I want to thank the New York Community Trust for making it possible.

So this is the third session dedicated to discussing welfare policy after Giuliani. In April and May we held much larger conferences which were devoted to the current status of welfare reform in New York and its impact. We did another session just two weeks ago on supports for people in the low wage workforce. And strategies for more effectively supporting them.

I wanted to convene this group to look at what is probably the hardest nut to crack of all the issues we face with creating a creative welfare or ambitious welfare reform program, and that is public opinion and politics and leadership.

Even the most conservative speakers that we've heard in the last couple of sessions supported the notion that working people, or at least working people who have been on welfare, should have access to subsidized supports of some kind that will help lift them out of poverty. Jason Turner said it. Larry Mead said it. And many of the more progressive folks that have been speaking in these sessions have said so as well. So how do we get at least that far for the working poor, ensuring that people can have reasonably straight-forward access to health insurance, childcare, social services and even some cash assistance, if they quality? And at the same time, how do we ensure a safety net that prevents families from falling into destitution, whether they are working or not?

The next mayor will be investing a great deal of political will and political capital, I should say, on education and possibly even housing. Probably police. Anti-poverty policy does not rank, as Jen Marsh was just saying a couple of seconds ago, the City Council didn't really talk a whole lot about anti-poverty policy in passing the budget a few days ago.

So I would like to use today's session to determine whether or not this community, which includes people from two very different sectors, the workforce sector and welfare policy sector, we can offer insights that might help the next mayor and the next City Council understand the importance of taking new leadership in the fight against poverty in New York City. I know there are many different opinions in this room but I would like to see if perhaps this group can begin to draft something like a win/win argument. A perspective that might help the next mayor champion a more expansive economic opportunity policy that will truly benefit the City's low income people.

Later in the summer or early in the fall, we will alert all of you to the materials that we're creating out of this series and they will all be posted on the Web, including transcripts and charts and summaries and so on and also the previous publications.

After the opening presentation, we'll go straight into a roundtable discussion and everyone will be welcome to participate but I'll be facilitating it, calling on some people who don't even raise their hands, so watch out. But also feel free to cut off. I'm also going to feel free to cut off people's comments if it's not part of the flow because I really do want to cover some ground in the two hours or less that we're going to have for that discussion.

So to start, I want to introduce Gordon Berlin, whose biography you all have, but I don't have in my hand. Gordon has been Senior Vice President for Work, Community and Economic Security at MDRC and several months ago he presented a piece for John Mulenkoft and Ken Emerson on a project they are working on, on new directions for New York City government. His piece was about redesigning the safety net. And it presented a lot of the issues that we all have to face so I wanted to invite him here to give sort of an update on that. And some of his other thinking. Gordon:

GORDON BERLIN: Thank you all very much for coming. What I'm going to do is three things. First, I want to talk a little bit about what I think the political context is for reform and the way one ought to think about it and I'm going to try to be, given my sense of who this audience is, as controversial as possible since some segments don't seem that well represented. Then, I really personally think that evidence matters a lot because the history of welfare and poverty policy is a history of unintended consequences and I think that it's really a mistake to redesign these systems without having a good understanding of what we know about the effects of different programs and policies. We have a lot of strong, reliable information about that. And then I want to kind of lay out the framework of the issues that you might think about and discuss.

So first, for the political introduction. It seems to me that for the first time in modern history, New York City can address its social welfare needs from a position of strength. And that is just a completely different world to be in. The economy is strong. Unemployment is relatively low. Poverty rates are falling, including child poverty. Labor force participation rates are rising. The City budget is relatively healthy. The crack epidemic is receding and possibly most remarkable of all, contrary to everyone's predictions, the welfare caseloads have fallen by half.

Now, there's no question that substantial problems remain. Relative to other big cities, poverty in New York is more persistent. Many New Yorkers remain poor. Employment rates are relatively lower than in other big cities. The cost of living is obviously a lot higher. The extraordinary diversity of the City is really only matched maybe in Los Angeles and Miami, which adds to the complexity of addressing the needs of the poor. And most obvious of all, there are still a very large number of people left behind on the welfare rolls who are poor.

Moreover, it seems to me there's a lot to worry about on the horizon. The simple fact is the economy that we've been experiencing can't last. It simply cannot last. And that also frames what we have to think about for the future.

On balance though, I think we have to conclude that we're in a very different place today than we were just five years ago. And what worries me is that if you stop to listen to the current debate, it sounds like the same debate we had five years ago, ten years ago, and fifteen years ago. And it seems to me that's a mistake. I think, in fact, it is a trap for reformers. And that trap is that we're going to try to design pollicies based on our past and now outdated understanding of what the current welfare and poverty problems really are about. In other words, I'm fearful that the response is going to be to roll back the clock and try to recreate the welfare system that existed in the past.

Now why do I think that's a trap? First, it's a political trap because no matter what the policy is, the welfare rolls are going to rise and the next mayor is going to get blamed for that increase. The next mayor has the same problem with regard to as well. It simply can't stay this low. And the same with employment rates. And the policies that one follows could exacerbate that problem. After all, that's really what Dinkins' problem was. He had extraordinarily bad luck. He relaxed the welfare rolls and rules and eligibility requirements at the same time that the economy tanked and the welfare rolls naturally would have risen. So they rose more than they probably would have otherwise.

Of course Giuliani on the other hand had extraordinary good luck. He closed the front door and just happened to do it at the same time that the economy was really taking off. I think the next mayor's risk is the David Dinkins' risk and it seems to me that you want to plan ahead for that risk.

Second, if you were going to build a new safety net from scratch, would you build the AFDC system again? I don't think that you would. It was a system built for a different era. For widows so they would not have to work and could stay home and care for their children. Most women with children work today, even very young children and the simple truth is the public, even the New York public, isn't going to tolerate a system like the old one that both discourages work and marriage. The evidence is clear that the welfare system both discourages work and it discourages marriage.

Third, government money alone will never be adequate to lift families out of poverty. We simply won't have enough money. And it's going to take a mix of earnings and support from both parents, that means father involvement and government help. In short, we need to develop a new safety net without sacrificing the welfare caseload reductions of the last five years, if we can. I would argue that we ought to build on where we are and not try to dismantle it and restructure it.

Now, this diagnosis poses three key challenges. The first is what kind of temporary welfare system should we build for those who are down on their luck? How should we change the current structure with its strict eligibility verification at the front door, its emphasis on workfare, etc…?

Second, how might we build a safety net around work that will help the working poor further secure their tenuous foothold in the labor market? Here I'm talking about the safety net of work supports that recognizes that many are going to mired in low wage work with little opportunity for advancement, while we also try to build a structure that encourages retention and advancement for people, but recognizes that lots of us will be stuck in low wage jobs and will need supports in that structure. Miraculously we have all the building blocks in place to build that work support system.

And last, how might we design the next generation of programs for the hard-to-employ, with multiple barriers? Programs that will require new relationships with service providers, new integration between employment and treatment programs and the like. Those are set of challenges that I think we haven't adequately addressed yet.

Inherent in these decisions are questions about what the structure of the current fragmented service delivery system ought to be. Should we continue to have separate job training, welfare, child welfare, childcare systems and if not, how ought we realign them? Certainly, we ought to rethink, I think, where the childcare system is currently. I don't think it belongs within the child welfare system.

Okay, if that's the challenge, what evidence do we have to go on and that's what I want to try to walk you though now and I'll try to do it quickly. I've already gone through this, we're in a new context, you have to decide whether you buy that or not, but it seems to me it's fundamental to thinking about where you want to go.

Historically, it's important to remember what the dilemma is. We all agree on the goals for welfare reform: to increase work, reduce welfare dependency, increase income, reduce poverty, improve child outcomes and most of us, except maybe my colleague Amy Brown, think that we should contain costs, at least to some extent. And of course Liz might have that problem too. But then, they have been working together for a long time.

The problem is these goals are often mutually exclusive. If you achieve one or two of them, it often comes at the cost of another one. That's the dilemma that has made welfare so impossible to really successfully reform.

Now, what are the four purposes of the current welfare law that we are going to have to live within? Because the TANF reauthorization is not going to radically change the current welfare structure and the four goals were essentially to assist needy families with children and dependents by promoting work, reduce out-of-wedlock parenting and to promote family formation and maintenance of two-parent families. And what did states and localities implement, including New York? Almost everyone in the country is running a work first program with mandatory participation requirements and much stricter sanctions. They are offering some education and training after the front end job search and probably the biggest surprise of all in welfare reform and probably the least heralded development, virtually every state in the nation dramatically changed their earned income disregard structure so that when you go to work, you can now mix work and welfare. Under the old structure, when you went to work, you lost a dollar of benefits for every dollar of earnings and as you'll see, it ended up not coming out ahead.

The third thing that everybody did was to impose a time limit. The feds imposed a time limit and most of the states implemented a variety of different things under the guise of a time limit so that, in fact, most of the welfare population, to everyone's surprise, isn't really subject to a strict time limit. California exempts children from the time limit, so it becomes a child-only case at the end of five years. Michigan, Governor Angler, Republican, has no time limit. Rhode Island has no time limit. Texas has a weird time limit that probably won't affect that many people. And New York doesn't have a time limit, although my friend and colleague Jason Turner doesn't seem to have heard that yet. And then many states are imposing diversion.

What I want to talk to you about is what do we know about the effects of mandatory welfare to work programs, work incentive programs and time limits? And I'm going to be reviewing evidence from 20 different random assignment research studies that are very reliable evidence on these three different policies. And the way to think about these studies is we took everybody in this room when they came through the door and applied for welfare and flipped a coin and Andrew got a heads and Ronny got a tails and they both ended up in different groups. And we did that for every one of you and when we're done, we're going to have the same percentage in the program group and in the control group, that have a high school education or don't, that hate life, that love life, you name the variable, you're going to have it represented in both groups equally. The only difference between those two groups is that one of them is going to be offered the program that I'll be describing. So if there's a difference later, the two groups are identical at the point of random assignment, if a difference emerges down the road, it has to be because of the program, because everything else is the same. The groups look the same in every way that you can measure and you can't measure. They are subject to the same economy and all the same external events. So it is very reliable, easy to understand. Everything that I'm going to show you is the difference between those two lines. So in this case, the difference is about 15 percentage points. The control group went to work after random assignment at about this rate, around 15 or so percentage points. And the program group was employed at about 30-35%.

Now, I'm also going to talk to you about how welfare policies might also affect families and children. Remember, three treatments, that is mandatory employment services, financial work incentives and time limits. Those programs should affect parents' employment, their welfare receipt and their income. Those three variables should, in turn, affect children's daily experiences. That is, how many books and toys and clothes you buy. The stress and levels of depression that your parents feel. The way they parent. How much childcare you can get. And the result ought to be some effect on child well-being and I want to talk about in each of these areas, mandatory work programs, financial incentives and time limits, what difference do these programs make for families, the adults and for the children. Because after all, it used to be welfare's purpose was actually to help children.

Alright, let's look first at mandatory employment services. And what I'm doing here is I'm looking at 20 different welfare to work programs and I'm looking at the difference each of them made on annual earnings. So if you take the top bar, here, it made about a $1200-1300 difference in earnings relative to the control group at the end of the year. The program group earned about $1300 more on average than the control group. Now there are three different programs here. The program with the blue bars on the left said, "Go to work." They are work first programs. They emphasize job search first. The programs on the right are education first programs. They basically said, "The best route to a job is to get an education first." Usually remedial education or GED. Very few people went to college. Only a few went to vocational training.

If you look at these bars, you can see that the job search programs, on average, increased employment or earnings by about $600 and the education oriented program increased earnings by only about $200 or $300.

The group in the middle seems to have done the best and they were offered a mix of activities. That is they had a very strong work focus but they offered a range of employment, job search and education and training services, depending on the individual's needs but they were heavily employment focused programs. In an earlier life, Amy Brown called them work first programs but, we've increasingly begun to call them employment focused mixed activity kinds of programs. And they tend to get much bigger gains for people. So the first thing I want you to take away from this is that virtually anything you can think of will make a difference in earnings, employment and earnings. Okay? The mix of activities seemed to work the best. If you had to choose basic education didn't do better than job search. That's a blow to many of you, I know but cumulatively the education programs simply aren't making a big enough difference to catch up to the job search programs. At the end of three years, the earnings levels are about the same but the cumulative gain is not. And that's why work first has been such a dominant force currently and maybe in the questions and answers we can come back and talk about why it may be that these education-oriented programs haven't produced bigger gains for people. On the other hand, these employment focused programs that included education did produce big gains for people.

Now, I want to use an example, the Portland result, which was the bar that was about $1200 more in earnings and I'm going to show you the results on earnings, welfare and income. There are three key outcomes. And then we'll look at the child outcomes and we use Portland as an example of a mandatory job search program. It was really work first oriented. It didn't take any efforts to try to make work pay. In the first year of follow-up, the program group earned about $500, $600, $700, maybe $750 more in earnings than the control group but because the government took away a dollar of welfare benefits for every dollar of earnings, this mandatory work first oriented program actually saved the government money because it reduced welfare benefits. When you add together earning sand welfare benefits, you can see that the program in year one made little to no difference in income. So people traded a welfare check for a paycheck but they didn't benefit financially. What happened after three years, the earnings continued to grow but so did the welfare savings and at the end of three years, the Portland program still hadn't made much of a difference in income. But it was saving taxpayers a lot of money and it was shifting people from welfare to work.

What difference did these programs make for children? I'm sorry about the colors here, I haven't mastered my Power Point skills very well yet. In this, we are looking at what was the effect of the mandatory employment programs on young children and we gave them some tests that measured school readiness and these were kids, I think 3-7 years old or something like that, in this particular set of studies. The group on the left was the education first programs and the group on the right was the job search first oriented programs. And these are the names of the different sites. And basically the bottom line is that none of these effects are really statistically significant, except for one. There was no consistent pattern. They were as likely to be moderately negative as moderately positive and the bottom line conclusion that we reached was that there was little consistent evidence of harm or benefit to the children. And remember, when we required welfare recipients to go to work, it meant moms had to leave their children and there was a lot of fear that kids might be harmed. But in all of the different studies that we've done where we've looked at outcomes for children, we haven't seen any systematic harm but we haven't seen any systematic benefit either something that conservatives promised. So those are the results for the mandatory employment programs for people. Increases in employment and earnings but no change in income, no benefit for children.

Now that led a number of places to be concerned about what could we do to increase income for people? In the past we've tried to increase income by increasing the welfare grant. The problem with that was that we then had less work effort and there was some evidence that we had less likelihood of getting married, as well. So in three places, Minnesota, a couple of provinces in Canada, Milwaukee, governments and community groups set out to design a system that would be different. A system that would be more in tune with today's values and norms. And that was a system that paid monthly cash supplements tied to earnings. That is the benefits were really focused on work. When you went to work, you got more money. And they tried to tie payment to full-time work in some form or another. They aggressively explained, marketed and outreached, and they offered a range of other supportive services, depending on the program. So what difference did these programs make?

Here, I'm going to look at the Minnesota program, which offered a combination of financial work incentives and a participation requirement and it did some other things but basically the program supplemented someone's earnings when they went to work up until their income reached 140% of the poverty line. And I can look later for you, if you're interested in how much more per month that actually was. It was in the couple of hundred dollars more per month for the average worker. So what difference did the Minnesota program make? Then look at the same three outcomes, employment and earnings, welfare and income.

And in year one, the Minnesota program got about the same gain as the Portland program in earnings but remember in Minnesota if you went to work, they supplemented your work. That is they didn't count a good chunk of your earnings so you were able to continue to mix work and welfare. So welfare costs went up relative to the control group. And when you combine earnings with continued and increased welfare receipt together, your income goes up. So the Minnesota program got more work and it got more income for people but it cost the government more. And that's the key tradeoff between the Portland model, which saved the government money but didn't benefit recipients with increased income. And the Minnesota program which set out to make work pay, it got people more work just like Portland but it cost the government more money, in order to also get them more income. And that's the fundamental trade-off that I think faces these welfare systems.

What happened in year three, more of the same, basically. So financial work incentives increase earnings and income.

Now, what difference did these programs make for families? In Minnesota there was some evidence that single parents were more likely to get married but the effect was very tiny. 7% of the control group got married, was married at the follow-up point, versus 11% of the program group. We haven't seen those results replicated anywhere in a reliable way and we don't really put much confidence in them. But one thing that Minnesota did that has really peaked our interest is it dramatically decreased divorce and separation among two-parent families and that is something that we ought to probably come back to, that is not showing up on this slide.

It also had a pretty big decrease in domestic violence. I think we're honestly not exactly sure why. Probably the best explanation is that women could make other choices. They had more income. And they were more likely to work. And they were also more likely to be out of the house more, because they were working so the opportunity for these incidents to occur was lower.

It increased the use of stable formal childcare and because you didn't actually leave welfare in the Minnesota program, you were also more likely to retain and maintain your health insurance benefits.

What difference do earning supplement programs make for children? Here, we looked at four different programs, two in Minnesota and one in Canada and one in Florida. And these programs consistently improved children's performance and it's a very different picture, if you can see it, from the picture that I showed you as a result of the mandatory employment programs. These programs generally improve kids' performance in school according to self-reports by their parents. We asked the program and control group parents a set of questions about how their children were doing and the parents of the program group kids, said their kids were doing a lot better. Now, you might question that because if I'm a mother who just went to work, I might be feeling a little guilty about that and I am more likely to say maybe that my kids are doing well. You could argue that. So we also, in a couple of the sites, gave a test where we asked teachers whether or not the program group and the control group, how they were doing? Without telling the teachers which group the kids were in. And as you can see, the teacher reports and the test that we gave also confirmed what the parents said, that their kids were performing better in school. They did better on achievement tests and teachers reported that they were doing somewhat better. So the full story on incentive programs versus mandatory programs includes benefits for children. Employment goes up, income goes up, poverty goes down, there are some improvements in family well-being and improvements for children relative to those mandatory employment programs only.

Now Minnesota was also a mandatory program but it combined that mandate with making work pay. But it cost the government more. It raises the question of how much more you are willing to spend.

What was driving these improvements in the Minnesota program, which included both incentives and mandates? The incentives accounted for the increases in part-time employment, income, reduction in poverty but they also raised welfare uses and costs. They were also responsible for the continuous health insurance and all of the other kind of social and family and children outcomes that were positive. The mandate, that is Minnesota said, if you want to get this incentive you have to work and you have to really work at least 30 hours a week, or else you would have to participate in a mandatory program, that is come in every day and meet with your caseworker and go to some other program. The mandate increased the full-time employment and earnings. So it was the two working together that really drove the Minnesota results.

Now Minnesota, who lives there? Why is that relevant? As we all know how that goes, "All the men are good looking and all the women are hard working."

Audience member: And all the kids are above average. (Laugh)

GORDON BERLIN: And all the kids are above average.

Well, a different kind of incentive program was operated in New Brunswick, Canada, and British Columbia, Canada, which is a lot like Maine and New Hampshire and Portland and Seattle and San Francisco and produced almost identical results to the Minnesota results. And there's a third program in Milwaukee that produced similar results. So we have growing confidence in these results because we are seeing them in a lot of different places and we're hoping to run some tests in the south and in some other big cities.

I've already, I think, impressed upon you how impressive these findings are and I won't spend any more time on that for now.

Now I want to turn to time limits and essentially, as I indicated earlier, the policies vary dramatically. Places without termination time limits actually account for 40% of the national caseload. A little known fact. Very few people nationally have reached the time limits yet. The places that they have are Connecticut, Massachusetts and Louisiana. Connecticut, I think, accounts for most of them. They have a 21-month time limit and we are going to talk about Connecticut in just a second. But Connecticut, if you aren't earning more than the poverty line when you hit the time limit, then they automatically grant you an extension. If you leave the rolls because you're working and earning more than welfare and then lose your job, you can come back on the rolls. So what looks like one of the toughest time limits in the country, has actually been implemented with some of the fewest teeth of any of those programs in the country. So again, this idea about welfare leavers and time limits, I think is at this point probably the most important policy that was never implemented.

What was implemented was the message. Everybody thinks that time limits apply, even in New York, where they don't. There is not a lot of evidence that people are banking months and let me go now to the Connecticut results. The same three outcomes. But this time instead of looking at year one and year three, we are going to look at before the time limit and after the time limit. Connecticut has the most generous work incentive in the nation. They disregard 100% of the your earnings when you go to work for that first 21 month period. You can keep all of your welfare and all of your earnings in Connecticut. Technically, they are supposed to cut you off if your earnings rise about a certain level but because they don't collect any information and they never ask any questions, virtually everybody gets to keep their entire welfare benefit. So for the beginning, before the time limit, Connecticut ought to look a lot like Minnesota. Right? It's a work incentive program. And that's exactly what it looks like. It looks just like Minnesota. $500 or so earning gains an increase in welfare. Add the two together and you get a really big jump in income for people.

Now, what happens after the time limit hits in Connecticut? People who are working have their benefits taken away. Now the results at post-time limit period look exactly like Portland. They continue to go to work at greater levels, their earnings went up but now they lost their welfare benefits, the government saved money. When you add together earnings with the less welfare income, there is no change really in income. The key tradeoff being posed here, when you combine mandates, incentives and time limits, is the tradeoff between incentives and time limits and that's the point that we are going to want to come back to.

And what about the effects of time limits on young children. The truth is I don't have any idea. We have limited evidence from one program and we're not seeing really any benefit or harm so that suggests that maybe if the income gain doesn't last more than that 21 months in Connecticut, you're not going to get much of a gain for kids. The income gain has to stay there longer. And I hope that you're seeing that we're really setting up a set of choices.

So just to summarize the adult findings, welfare to work increase earnings and reduce welfare but they don't affect income. Mixed strategies produced consistently larger gains than either employment or education focused programs. You want to think about New York where we're being primarily work first and job search oriented and workfare oriented. Work incentives increase employment and income but they definitely cost more. Time limit to welfare aids the employment gains but not the income gains and the long-term effects on children aren't clear yet.

And the child conclusions, programs that increase parents' employment but not income neither help nor hurt children. But programs benefit children if they increase parents' employment and income together and one thing I haven't talked about yet, we're seeing increasing evidence for adolescents, that when their parents go to work, there is some hint that in fact the less supervision that results, results in more experimental behavior. These are barely statistically significant. They are differences that occurred in Canada and look like they are going to occur in several other sites. They are off very small bases. That is 5% of the kids were drinking and now about 7-8% of the kids are drinking. We didn't try to measure more positive benefits that might have occurred for children because we really didn't anticipate these findings, although any of you who know adolescents, which I'm sure all of you do, know that this is a problem we probably should have anticipated.

On the other hand, we are basing these results on adolescent self-reports for the most part. And any of you who know adolescents, wouldn't have the foggiest idea what they meant when they answered those questions anyway. So I think at this point, it's a yellow caution flag for everybody and it's happening whether it's an incentive program or an employment oriented program. It's simply seems to occur when mothers go to work for adolescents. But again, it's a small effect and we're not really convinced that it is real yet.

So what should we do in New York? I think the first question is what kind of welfare to work system do you want to have. I said at the beginning that I think you want to continue the idea that welfare ought to be a temporary system for people. If you can figure out how to do that. One of the first things you have to do is decide what you want to do with the new eligibility verification system that has been put in place. If you pull it back, then you're going to have some increases in the rolls and that's a choice that one has to think through.

Even John Lindsay tightened eligibility verification in his second term. That's how Charlie Morris actually got a hold of the rapid welfare increase that was occurring in New York City in 1971 and 1972. They put in place a very strong eligibility verification structure and tried to get control of the front door of the system.

The second big question is whether to continue the workfare emphasis. I didn't put up any results for workfare programs but in the studies we did in the 1980s, the results were a little surprising. Virtually everyone who participated thought they were a good, fair thing to do. The recipients themselves thought it was fair. They also thought, and their supervisors thought that they did meaningful, valuable, important work. At the same level as other people there. But the recipients also said that they weren't learning anything new that would help them get a job. And in fact we saw no evidence that workfare led to increases in employment getting. However, we didn't measure deterrent effects and that's obviously one of the things that they are trying to do at workfare programs, Discourage people from actually coming on the rolls to begin with.

The evidence I've showed you suggests that you'd like to run a more of a mixed strategy kind of program. You'd probably get bigger gains than the current system is getting. And the last question that we're all waiting to hear is what will the new safety net voucher system look like? And I think there's a big question out there about whether you want that system to be readily available and known and transparent so that everybody understands that they can, in fact, stay on the rolls. And I'm just raising this question of how much do you want to encourage people to leave, given what I said earlier that welfare is not going to provide for them and their families and children adequately. And it never will. And what will is a system of work supports.

Now, what could we do on supporting the working poor? The first big issue is that there's this inherent conflict that I showed you between time limits and incentives and one way to get around that inherent conflict is to stop the time limit clock when someone goes to work full time. After all, they are doing the right thing. Why would you want to kick them off the rolls and get the result that you saw in Connecticut?

In order to do that, without incurring this welfare increase problem that is looming out there for the next mayor, I think you want to do it by creating a separate and distinct program. That you leave welfare and you join a new program that supports the working poor, that you're more likely to be able to build public support for . Because if you do it through welfare, you're not going to get the public support that you need. Critics will be out there claiming that the rolls are skyrocketing and the fact that a higher percentage of those on the rolls are working won't be enough to secure you politically against those attacks, in my judgement. I think that's a risk. But you could do it by creating a separate and distinct program.

Now that raises the question of whether you ought to do it through the state and City EITC, Earned Income Tax Credit, structure that is universal. Or whether through a more targeted structure like the Minnesota program or the Canadian program. And that is something that targeted welfare recipients primarily when they were leaving. And the tradeoff in the results is that you get the biggest effects from the incentive programs for the most disadvantaged. As you move up the employability ladder to applicants to welfare who are more employable, to two-parent households, to the working poor as in New Hope demonstration, the effects decline substantially. In the New Hope program, for families that were already working a lot, the incentive structure actually resulted in a decline in employment and earnings for people. So there is an increase in efficiency if you target. You get better equity if you spread it out universally but you don't have the money to pay as big incentives. So that's another big choice that people have. And it gets at this issue of how generous can a subsidy be? Who should it be targeted to?

You have to market it. If people don't know about incentives they can't act on them. And at least in New York, I have never seen anybody told about the 49%, the $90 plus 49% disregard.

At the same time, we have to come to grips with the fact that low wage work is here to stay and we've got to find a way to move people up into better jobs. And I think we just don't really have a clue how to do that effectively. And we're starting a whole set of experiments like the ones that I've already put up on the board here, to try to get a handle on retention and advancement models that might be more effective for people. Things that buy back time from employers so if you hold onto your job, you get to go to school. And other structures like that. Changing the way community colleges support people, while they're working.

Lastly, if you think about it, we have a more generous income disregard, we have a state, a federal and maybe even eventually a city EITC, that takes a $6.00 an hour job and turns it into a $9.00 an hour job, which is a pretty substantial work support system. The federal EITC and the state EITC together do that, move you from $6.00 to $9.00. We have the food stamp system. If you put all those systems together, we have all the elements of a reasonable support system for the working poor.

What's the problem with that? The problem is that no one is accountable for delivering those services. There is no agency that is actually on the line to make sure that the working poor know about and get those services and benefits for people. And furthermore, this message conflict between incentives and time limits mean that the welfare system is in a schizophrenic state. Right, I'm a caseworker. You come in and meet me and the first thing I say is there's a clock ticking so you need to get off of rolls and bank your months. Oh, but by the way, if you do that, you don't actually have to leave because if you stay, we'll match your earnings and let you keep getting some welfare benefits." Those two messages are inherently contradictory. No caseworker can possibly deliver them clearly or meaningfully and no recipient can possibly understand them and that's a major problem that is occurring currently in every state in the nation. It's not really occurring in NY and in many places, because the time limit message is the dominant message. That is the message that everyone is hearing. They don't even know about the incentive structure.

Then the last big area is the hard-to-employ. A large number of people are left on the rolls and they are more likely to have multi-barriers to employment that include mental health issues and depression especially is surprisingly prevalent or maybe one should say not very surprising. And it goes pretty much untreated. Substance abuse remains an issue. It goes pretty much untreated. The simple truth is the welfare systems don't know how to deal with these problems. They really haven't built the kinds of relationships that would be required to integrate treatment with employment programming.

In New York actually, the PRIDE program and a new case management program for substance abuse programs actually are starting down that road and I think it's a healthy development and one that really requires and warrants careful study and experimentation as a framework for beginning to figure out what really would work more effectively in this area. We probably have to create a kind of sheltered workshop structure for a group of people who really have long-term issues.

And lastly, there's a group of people that fall between welfare and SSI. People who are being treated for chemotherapy for cancer, for example. We've turned up a number of these surprising cases in our ethnographic work in Philadelphia, Los Angeles, Miami and Cleveland. And the welfare system's rules keep applying, even though for a good chunks of time, these are folks that can't go to work and people really need to begin defining better what to do for that group of people. The striking thing is the Congress said 20% of the welfare population could be exempted from the time limit structure federally but there's not a state in the nation that actually is granting anyone an exemption. Everyone is granting extensions on a wholesale basis but almost no one is granting exemptions and there's a group of people that we really ought to be thinking about those issues with.

I have lots more slides there. I would be happy to answer any questions you have about them. But you all actually know more about this than I do and you're very kind to sit here listening to me tell you everything you already know. Thank you.

APPLAUSE

ANDREW WHITE: We'll take a few minutes for comments on this before we go into direct conversation. Is there anyone who wants to pose a question of Gordon? Bill.

BILL AIRES FROM THE WORLD HUNGER ORGANIZATION: My first question is, when you're judging the effect of work programs as opposed to study programs, how can you do that when study programs don't really pay off for two, three four years.

GORDON BERLIN: That's a good question. In the study we randomly assigned people to three groups in three cities. The first group was a work first program, the second group was an education first program and the third group was the control group that got the old welfare system and we followed them for five years. And the five years results will be out soon and HHS is forbidding us to talk about them but I haven't said anything that would contradict them. In other words, there's no evidence that the education programs are going to go ahead. At the end of three, four, five years, they catch up but they don't actually go ahead. So the cumulative gains for someone who went to work first are higher and I think there are several reasons for this. The first reason is the dropout rates in the remedial education programs we are sending people to are extraordinarily high. When I was in NYC government we tracked 100 people that we referred for remedial education and only 10 of them ever completed. So if only 10 are completing, you're not going to see in the averages, any real gain.

The second problem could be that education is necessary but it's not sufficient. It's a gatekeeper, the people need to move from remedial education and GED getting to educational training and most of these systems aren't really built that way. And they don't encourage that kind of long-term investment. But again, if people drop out at the rates they are dropping out, you're not going to get there.

BILL AIRES: I have some questions about the package of work support for working poor people which I think is a great way to go. I think most of us would support that. Can you say more what they are and would you suggest something like child support and enforcement insurance, which was an old idea from years ago, seems to have fallen out of favor but do you see that as part of the package?

GORDON BERLIN: Well, the work support system is basically the federal and state EITC, the welfare earned income disregard, including the food stamp package. And then the childcare and the continued Medicaid benefits. That's the structure that is out there now.

The problem with running it through a welfare structure is that in many states, and I forget what we do in NY, you have to spend your way down to meet the income eligibility levels and then you can work your way back up. In some places they've said, "If you have earnings of this level, which is where you would be working your way back up, you don't have to spend down in order to get into the system. You just come in on a lateral." And I am proposing that you build work support systems that was for the broader working poor population and not just for people who have been on welfare previously.

Child support insurance is an interesting and very good idea and at one time I was a big supporter of it but I think the political problems are just too big. Irv Garfinkel proposed this 20 years ago. It's never been able to get a big enough following. I think politically it's just not going to get us anywhere.

Q: I'm Briana Sanger, I'm a faculty member at Milano and I'm at Bookings, as well. I wonder whether it matters who provides the services and whether or not in any experimental sites there were variations of the character of providing the mix or were these all publicly provided services from official welfare departments?

GORDON BERLIN: I think it was a wide range of providers. We did, in one of our studies, Karn Martinson did a look at the quality of remedial education programs and it was in San Diego where they systematically try to build really high quality programs, in a range of for-profit and nonprofit providers. We thought the quality was really good but I don't recall the outcomes being particularly different. It's not to say that you couldn't do better and I'm an advocate for trying to do better but I am very cautious about the likely outcome.

BRYNA SANGER: Is that going to figure in here? I would imagine in New York and certainly elsewhere in thinking about restructuring what the kind of implementation vehicle is of any of these.

GORDON BERLIN: Right. Implementation matters a lot. It's not just what you do, which is what I've been talking about. But how you do it.

Q: I'm Russ Sykes with SCAA in Albany. I'm just curious, you mentioned something about the no accountability for delivery when you talked about EITC and food stamps and while I agree with you on the food stamps to a great degree and a lot of improvements could be made. I wonder if you'd comment, from my understanding, at least from everything I've read, the take-up rate on EITC is extremely high. Federally and at the state levels. And basically, because it's very simple, very easy to access, there's not a lot of bureaucracy, as we know.

ANDREW WHITE: Explain what you mean. Take-up rate.

RUSS SYKES: Basically the penetration rate of eligibles using the earned income tax credit is as high as 80-85% and in New York almost 90% of the federal eligibles get the EITC and so I think the delivery systems aren't as profound there. And I wonder if you'd comment on that.

And I had a second comment. It's the fact that someone asked about child support assurance but what about child support in general. Outside of the concept of the assurance. For states to really focus their energies on building the systems and make them structurally more sound and getting more child support benefits to families directly from the noncustodial parent.

GORDON BERLIN: I'll take the second one first. We've done some work in project called Parents Fair Share. We are about to issue the final report, which is much more accessible and readable than the traditional MBRC final report. We really tried to combine a set of services that would increase the father's employment and earnings so they would actually have some money to pay the child support. At the same time that we made the child support system more responsive to the ability of the fathers to pay and we also have an initiative that tried to increase father involvement on the theory that would be good for kids and it would increase the likelihood that fathers would pay.

And we had limited success, as you'll see. Some positive success but it turned out to be hard to implement. I think that it is clear that if we're going to improve the well being of children, it's going to take two parents' income. And I think you can't ignore the role that fathers have to play, both emotionally and financially. And I think the question is whether you can design a child support system that makes it better as opposed to makes it worse. And the problem with child support is that it works on a kind of middle class father's paradigm. Mostly, until very recently, these systems had basically decided that all these fathers had income and they were just lazy and deadbeat and just go after them and lock them up if they won't pay. But for poor fathers that just drives them farther underground because the fact of the matter is they often don't have the employment and earnings and the income required to make those payments and they end up incredibly in debt and always have the child support system and thus the courts looking for them. And you have to design a system that works better.

I actually think we've made some progress in that regard nationally. I'm a little less clear how much we've made. We've certainly increased collections in New York but I don't know how much we've made the system more responsive to the actual ability of fathers to pay.

On the EITC, you're right. The take-up rate is very high. It's a non-system though. I personally don't know anyone who works with low-income people who actually believes that the majority of poor people know about, understand and get the EITC. So how do you reconcile that kind of on-the-ground sense that it's not really all that well understood with these national statistics that show 85-90% getting EITC? Well, the problem is that the way they are reaching those numbers is they've taken estimates from the census about who is eligible and they've divided it by the amount paid out in EITC and the result is that more than are eligible are getting, more money is paid out that should be paid out under this structure. No one wants to talk about this particularly, because it suggests that there is a hell of a lot of fraud going on in the structure and the Republicans at one point were really focused on that issue. So I actually don't really know where the truth lies. In our surveys and our various studies, about 60% of people say they get EITC and we go back and forth about how to value the EITC. Should we assume they are all right and they all know? Or should we assume the national numbers are right, 80-85%? But I think you're right, most people are probably getting it. You could say I'm overstating the case.

RUSS SYKES: They may not be aware it's EITC either.

GORDON BERLIN: If you go hear somebody like Ron Haskins talk, who wrote the original welfare bill, he will say that the support for working poor system is in place already. In fact, the average person is entitled to $16,000 to $18,000 a year including their earnings. And what he does is he takes their earnings, he adds the EITC and then he adds the welfare, food stamp, and earned income disregard structure. If you add the welfare and the food stamp amount together, that equals more than the EITC. And interestingly, liberals, David Ellwood and others, Irv Garfinkel, are spending a lot of time talking about go to a universal system. Don't do it through welfare. But they are missing the point that it takes both of those to actually get people up to the $16,000-$18,000 a year range that we're trying to get them to. But figuring out how to deal with this incentive/time limit conflict and making sure that the earned income disregard structure and the food stamp benefits actually get paid, is critical to getting there. The EITC is not enough.

ANDREW WHITE: We're obviously running late and we're going to go a little bit later than noon. The single premise that has come out of both of the sessions prior to this seemed to have some sort of universal appeal on the size of the public assistance caseload should not be the defining measure of success in welfare reform. Should reduced caseloads be what we use to measure success? So let's say the measure of success is poverty reduction and increased income and increased well being, like much of Gordon has just spelled out.

Jason Turner, when he spoke, gave an example of how a single parent with two kids can have a higher income working at minimum wage than just through their welfare benefits. Of course the example he used was heavily dependent on public subsidies. But the example of the family that has double the income but since the mother was working she also had the income disregard, which means she was still getting cash benefits. She had food stamps worth $200 something and she had earned income tax credit, worth $300 something dollars. And so on. So that family's income was $1512 a month on minimum wage, so long as she is getting all the subsidies she's eligible for. So if we take Turner at face value then he is saying we should encourage people to work and receive benefits for which they are eligible, whether that's Medicaid or food stamps or job training or whatever. But in New York City there is plenty of data that shows that at least 20% of the City's working families with children have incomes well below the poverty line. So while we've had a lot of job growth in the low wage sector, the wages in the low wage sector are going down.

How do you square these two? And we also have the issue of the plenty of people who are eligible for various benefits who aren't getting them. Which comes back to the whole marketing issue and diversion, as well. So in this roundtable there are a few core areas for discussion and some ways it breakdown just like one of those slides broke it down. What about accessibility to services and benefits? Mayor Giuliani has made a lot of political…or gained a lot of political value, a lot of political distance, by talking about the reduced welfare caseload. That has meant that because of the divergence strategies, basic benefits, whether you're eligible or not, are often very hard to obtain. Many of us have worked with people on welfare know how incredibly hard it is to obtain, and especially if you are a working person. Go down to the welfare office repeatedly and do whatever you have to be because you are going to get Medicaid can be extraordinarily difficult.

Giuliani has also used very popular language to claim that his system catches welfare cheats and that people can't be freeloaders any longer. That is very popular even in New York City for him to be able to say that, he can wins a lot of positive attention, not just from the press and the middle class, wealthy white folks. It has a pretty broad appeal.

So how do we propose that the next administration makes those kinds of supports available to people so fewer people live in poverty? How will the next mayor do it, while watching the caseloads go up? How will the next mayor do it while being accused of allowing more cheating or whatever, just because the numbers are going up? And the attacks will be coming in so many directions including editorial boards and former Mayor Giuliani but also from the general public who is attuned to the notion that the caseloads have to go down if we are going to be measuring success. If this is a successful policy, then caseloads will go down. That's the general perception.

Secondly, what about the nature of the support programs themselves. How generous should they be? All the things that Gordon spelled out. And finally what about those who aren't working? What kind of benefits need to be available for people who are out of work or unable to work Do we allow them to become second class citizens who are simply surviving on vouchers?

Obviously, this doesn't cover everything. The whole issue of education versus work versus education is something that has been fought over by everybody in this room much less in a much broader political spectrum.

So in summary, many of the things we've discussed all along require that the front door be open wider and they require that making support services of many kinds become more accessible and more available. Lindsay, in 1969, the height of liberal New York was hammered for tripling of the caseloads. And we're talking about something very different today. But even today, politics and public opinion oversimplifies everything. So how will we define caseloads? We have to change the definition of caseload. Do we have to somehow market to the City that there are other ways to measure success such as decline in poverty rate? It's a little more complicated. Do you count people in the caseload if they are working, even though they are still receiving some benefits and so on.

And how might we push the political leaders or help the political leaders to spend capital on anti-poverty programs? So we are going to start with less than 5 minutes from a few people who I asked to open things up and then we'll discuss. So Liz is first.

LIZ KRUEGER: Less than 5 minutes, let's see. Hi, I'm Liz Krueger. I've worked on anti-poverty issues for the last 20 years. I agree with much of what Gordon said in his presentation. I think that Andrew sort of pulled us back to some of the questions from the previous conferences. What do we do with the next administration and I think a big problem facing whoever is the next mayor is the fact that the current mayor lied to the public for 8 years about reality. And defined the assignment to the assignment is if you reduce caseload you win. And of course, I think everyone in this room knows that there's lots of easy ways to reduce caseloads that do nothing to improve the economic situation of poor New Yorkers and we've pretty much tried all of those in this administration. But it doesn't change the reality. The problem is public perception and politics often play a bigger role than reality and Gordon's point about David Dinkins had the bad luck to be mayor when the economy was bad and Rudy Giuliani had the good luck to be mayor when the economy was good, is absolutely true. But the fact that he then lied about taking credit for the caseload reduction being an impact of his policies is really a major problem for the future mayor.

Where I don't agree with Gordon is the perception of these studies that were done by MDRC having shown that there are models that work and models that don't. If the best case scenario after five years of federal welfare reform in this country are programs that move people's income up at best case $1300-$1500 a year, for an average size family of three, which works out to be about $45 a month win per person, we didn't go down the right road yet.

I'm not suggesting we should go back to what we did pre-TANF because there wasn't anybody in this room who would have been here 5 years ago who would have said, "We love welfare as it is." So I don't see any point of having the discussion of having to go backwards. The question is what did we learn and Gordon's information was helpful there. But the fact that even Minnesota and the other projects that are talked about having the best outcomes after three years of study, that the best case is $1300-$1500 per year annual increase. Let's be reality based here. That's not going to address the problems in New York.

I do completely agree with the issue of doing coordinated service integration and looking at how we use integrated programs to maximize income for people who have to combine work and welfare. I have no problem with suggesting to the next mayor, change the name of a lot of these programs. If the programs got the bias against it and you have to deal with the issue of public perception and how they are going to interpret what you are doing, fine. Nobody likes the name welfare. We actually blocked it out with anything in a government agency. We don't have any welfare department. We don't have…I don't even know all the nicknames inside of HRA anymore because they change so frequently. So change the name of the programs so that it does reflect combinations of work and income supports.

We did a horrendous job in this city of assuring that anybody got the benefits that they were eligible for, even under the existing laws. So things that we can do to help maximize food stamps for working people, childcare for people who are eligible but not receiving it, the earned income disregards. I think when you look at the statistics at HRA, although granted it is very hard to find statistics from HRA, the participation rate of people who are actually getting the income incentives along with welfare is tiny because the assignment at HRA was not to help people maximize their income through existing programs. The assignment was diversion, because that's the easiest way to do caseload reduction. If you never let somebody on in the first place, you never have to worry about getting them off later or implementing programs. So that is the great model of this administration.

So I do agree with Gordon about the idea that the next administration focusing on making sure that we can maximize income disregards, earned income tax credit, childcare, food stamps, health care. There is a lot to be learned about, even looking at what has happened with Child Health Plus versus Medicaid in this city. All the studies showing that kids are on Child Health Plus even though they are eligible for Medicaid. For several reasons, apparently, but one being we stigmatize Medicaid so families who are working prefer the thought of having their kids on Child Health Plus even though, of course, it's a worse package of healthcare than Medicaid, for kids. So if you're eligible for Medicaid and you know the facts, you should want your child on Medicaid not Child Health Plus. But facts and perception are different and the direction has clearly been anything that is stigmatized is something that people don't want to participate in and then that translates into increased public perception of hostility towards those programs.

A lot of us who worked on food stamps talked about when EBT came in, Electronic Benefit Transfer and we were getting rid of food stamp coupons, change the name of the program. There is so much stigma associated now with the food stamp program, it's not stamps anymore anyway, let's call it something else that won't have the stigma associated with it.

A couple of things that Gordon didn't bring up that I think are fundamental for us here in NYC to deal with, that are perhaps not the pattern nationally and so MDRC research doesn't necessarily look into it that much. Two major factors for NYC, our housing costs are out of the world in comparison to almost anywhere else. And when you're talking about income issues and poverty, if you don't have an affordable housing strategy attached to anything else you're doing for poor people, including trying to increase their wages and combining it with better benefits, you sort of lost the game anyway because you can't afford the housing. It just keeps putting you back and back and back.

And second, the impact of having special needs populations that are not just the categories that Gordon was listing. Although I also think the idea is very interesting of pulling out specifically programs for people with mental health problems, with substance abuse problems, with other special needs and perhaps defining them also as new programs that we don't call welfare, whatever the funding streams for them may be.

We have an enormous immigrant population in NYC and it has special needs, particularly for English as a second language programs and special assistance to people who are coming in to our country for the first time through our city and what we are going to do or what we ought to be doing to help maximize their ability to move up and out of poverty as quickly as possible as new immigrants.

And finally, just one other issue that I don't think I've mentioned that I think we should put on the table is the need to reform the unemployment benefit system. In today's economy everyone likes to talk about full-time work. I actually don't think that it is an ideal model for single mothers raising kids. I think we should be talking about less than 40 hours a week jobs as realistically 35 hour a week jobs. But no matter what I think the reality of our economy is that it's temporary and transitional and part-time job availability. Yet people who are in those categories of jobs, which are primarily low income workers, don't find themselves eligible for unemployment benefits during the time periods that they lose those jobs. And as our economy doesn't continue to grow but hopefully doesn't continue to plotz as badly as I'm afraid it will, we will have to deal with a growing unemployment rate and the fact that we didn't do anything during these good times to address and reform our unemployment system in relationship to the changes in our labor market economy, are a real problem for us. So I think we squandered a lot of opportunity over the last five years which were good economic times. And it's absolutely the right time to be making fundamental real reforms. We squandered this administration's time to do that. We squandered an immense amount of money. NYS took our TANF dollars and built roads and gave us tax cuts instead of investing it in programs for poor people. And the next mayor is stuck, frankly, inheriting the post-five year federal money, probably post-good economic times and eight years of lying to the public about what has been happening and that's a real problem for him or her. I can't say her. Him.

MICKEY BLUM, POLLSTER, BLUM & WEPRIN ASSOCIATES: Hi there. I'm Mickey Blum from Blum & Weprin Associates and we are an independent polling firm and I probably know less about poverty and welfare than anyone in this room but what I do know something about is public opinion. I guess I want to talk a little bit and definitely less than five minutes about is public opinion. Public opinion in the country, and public opinion in NY on these issues. And how it sort of relates to what we have been hearing. And in fact, it does relate very directly.

To begin with there haven't really been a whole lot of polls in NY, at least recently, about welfare. I do a lot of polling for NY One, for the Daily News, for the Urban League, for Hispanic Federation. A lot of groups that maybe we would think would poll about these issues or try to poll about the issues they think are uppermost in the minds of New Yorkers and none of them have chosen to poll about this. And that, I think, tells us something right here. That doesn't mean that some of it isn't in the back of people's minds when we do ask people about what issues matter to you right now and what issues do you think are most important facing New York right now. And certainly, when we ask on our Urban League poll in New York what issue they thought and problem they thought was the most important facing blacks in New York right now, as a poll that was called State of Black New York, people felt that economic issues and issues of jobs, job training and poverty were crucial right now and there is certainly some fear of the economy tanking so that it shouldn't be said that is somehow not there at all. It's just people aren't necessarily saying welfare. They aren't necessarily saying anything about government assistance but they are concerned about poverty. It hasn't entirely disappeared from the consciousness.

There are some contradictions in what people believe and in their attitudes and beliefs about the welfare system past and present and much of it is expressed in the differences people see in the language. Liz is absolutely right about this, about language and terminology. If you use the word "welfare" they hate it. I don't care what you say about it, they hate it. Everybody hates it. If you say "assistance," they don't hate it. Assistance is a good thing. Welfare is a bad thing. Assistance means something that is temporary. Assistance means that you are trying to help yourself, we are just helping you do that and that's a good thing. Welfare means you're a cheat, you're a bum, you're trying to do this for life, you want to be supported by someone else. It does matter the language that is used. It will matter in the next administration, the language that is used. And maybe it will require programs that have different terminology.

There are some probably not surprising differences by income and by race. What might be more surprising to people is that there's a lot of very basic agreement across the U.S., across the state, across the City, black and white, about welfare, about welfare reform and that has held true over time. Everybody thought that the old system was bad. Everybody thought it was bad for everyone. Everybody thought that it was destructive mostly to the people who were on it. That as much as they resented it as taxpayers, they thought it was even worse for the people who were actually receiving welfare. They thought it really destroyed them. And that it caused poverty.

There is some awareness of welfare reform in the sense that people know that some laws have changed. They just don't know the details. They don't really know what it means. They know there's been some kind of change. Most people are aware that there's now some sort of work requirement. People generally approve of reform. Again, black, white, across the country, across the city, they approve of reform. They approve, like it or not, of caseload reduction. And like it or not, they credit that caseload reduction to the reform in the law, much more than they do to the good economy. They really do believe that welfare reform has caused this great drop and that that's been a wonderful thing.

There are certain things that people in general consider absolutely crucial to the reforms that they wanted before there was reform and that they still want. And basically there are a couple of things that are very intertwined. They are competing and sometimes not competing. And that is work and job training and education. And what we heard about the mix being perhaps the best system, is actually in some ways good news for anyone who is trying to appeal to the public because, in fact, that's exactly what the public wants. The public wants a mix of both. They very much approve of work first and they approve of job training. They have some differences as to when they think one is appropriate or the other is appropriate. They want everybody to go to work and even if that work doesn't fully support a family. Even if that work produces an income that's not quite sufficient and needs some additional help. They want people to go to work. But mostly, they want people who are long-term recipients to go to work because they think the most important training that those people need is at work habits, is the work ethic, is actually learning what it is like to go to work every day.

Where people seemed most interested, again across polls, in job training and in education, was more with people who had recently lost jobs because there was some sense that some of those people may have lost jobs in industries that were no longer going to be there….be viable and that some of these people needed some retraining, needed education for new or different kinds of jobs.

They are very happy to support a combination of work and training for basically anyone and everybody who goes on…who is trying to get off of assistance. They do want mothers to go to work. And actually when people are asked about it, they even tend to say things like, "Well, I work. All moms work now. Why shouldn't they have to work?" They do want moms to go to work. They may be happy, again, to have some assistance when somebody has an infant. They want childcare support. But they do…and in fact they want a lot of childcare assistance, but they do want moms to go to work even if they had young children. And there has been some different findings on that, sort of depending on how you ask the question about whether they should have the right to stay home with young children for a certain amount of time but across most polls that seemed pretty clear.

There is support for lots of benefits that would help people go to work and help people get education in that there is support for childcare. There is support for health insurance. If you get a job that doesn't have health insurance, should we be helping people get that health insurance? Should the government be doing that? Yes. Certainly in New York that is the feeling. It is across the country but even more so here.

In New York there is a belief in the help with housing costs as well, since we were just hearing about the issue that housing is a huge problem here. People really know that. It does come up as a huge problem and there is some sense that people who are at low-income levels need housing subsidies.

As for time limits, time limits were something that has much less support. It does have support but not as much as the sort of general work first concept. And again, even with time limits, people wanted to see some kinds of exceptions to time limits and were aware that there might be some situations where they needed to be extended, where they couldn't really be applied. They didn't ever want them etched in stone. So the fact that none of the states are really doing them the way they were supposed to have been doing them, is probably just fine with most people. They would be happy to know that they are not.

They kind of wanted to know that they were sort of hanging over people and pushing them a little bit but in fact, they didn't necessarily want them enforced in certain situations. They certainly didn't want it to cause suddenly homelessness and all sorts of problems when they were suddenly withdrawn.

There is a lot of awareness that people, even when they go to work, they maybe get off of traditional welfare or they are put into jobs and go to work, are still poor. And are still really not earning enough to support their family. There is an awareness of that and people aren't really happy about that. They do want them to work, even if those jobs don't make enough but the ultimate goal is in fact to help lift people out of poverty and there is a lot of support out there for the idea of assistance to the working poor. It is, in fact, exactly what people would rather see. They would rather see us helping people they think of as perhaps deserving in that they are trying to help themselves and they are the working poor. So I think ultimately saying that we might need to have programs which now support the poor, not just because they are not working but simply because their income is insufficient, is something that I think could be popular in this city.

I'm sorry, I will allow myself to be cut off. That's fine.

KEVIN FINNEGAN, POLITICAL CONSULTANT: My name is Kevin Finnegan and I'm here today in a capacity that I'm always trying to avoid being called which is a political consultant but I actually do a lot of politics and that what I'm going to talk about today. .

The role of really defining your argument and really defining your debate and how that works in politics so you get to where you need or you want it to be. I am certainly not an expert on welfare but I'm comfortable with turning to a lot of people I know in this room and asking them questions, "what do we do on this?" They will give me an answer and we'll figure this out together and get what you want. That's how Liz and I know each other and a bunch of other people in this room.

It's important that the wanks of this room tell us what the answer is to this problem and then we'll kind of figure it out and Mickey just sort of ran through exactly what I think people in the political world know intuitively in a certain sense. The first thing you do obviously, you hear from the wanks what the answer to the problem is. The politician or the political group that you're working with accepts that argument or they go to the wanks anyway. And then you go to a pollster and find out how to craft the message. And it's an extraordinary failure on behalf of welfare advocates, certainly since the mid-60s to define the arguments in a way that serves their interests and services their clients' interests. It's clear it's a complex issue from the public's point of view. People support children. That's it. If the children are victimized in some way or form, this society will make the expenditures necessary to fix that problem. A Child Health Care Plus program was so successful a year, year and a half ago was based solely on that argument and we all know what the program is. It's straight state subsidies for poor people, which used to be called welfare.

Health care in general is something that is difficult and the state does need to intervene in some way or form. The Clintons failed miserably in that but only politically. Whenever you broke down those issues, no one disagreed with Hilary Clinton on that stuff. They lost the political argument in a huge way with the nasty Congress.

Job training is always popular. People understand fundamentally that we all go to work, we work too hard and we get paid too little, all the way up to somewhere below the financiers on Wall Street. That's the way most people feel. And they see that other people are in worse shape.

At the same time people are not really concerned with taxes at the moment. That's just not high on people's list, though it is often the way people express themselves in hostility towards state subsidies. I don't want to pay my dollars to someone who is undeserving and that certainly involves a work ethic, which dominates this society like probably no other in the world. There's a real concern with giving money to people who don't sort of pull themselves up by their bootstrap for whatever reason that is. We just have to live with that and can never define ourselves in that way.

When you look at how the debate has been defined from the point really in the early 60s when there was popular war on poverty program that is associated with Johnson. And increasing caseloads was a benefit. That was a good thing. And we wanted the government to solve some of these problems. To the late 60s, sort of breaking this cycle on poverty. A sense that there were people out there sort of trying to take advantage of the system and then sort of ending in this debate that we have now which really kind of drives a lot of what I'm hearing and reading about this conference, which is ending dependence. And when you look at how each one of those values play out in the minds of the public, one there's a racial aspect that we have to acknowledge if we are going to be successful in the political area. The War on Poverty is fascinating to look at how that was marketed in the early 60s. The focus was on the white Appalachian poor. The ending dependence is from the conservative, you never say the wrong words, but your message is clear, we're talking about African Americans who are in the inner cities. With code words or whatever.

When you're marketing yourself and your own message, you can't one, obviously play into that, which I think all of us sort of know intuitively but two, try to correct it. It's totally foolish to talk about solving a problem that no one wants to solve and someone is hostile to. Which is the perception that the people on welfare are black, married to a drug addict or not married, had a baby with a drug addict black man and now expects the government to pay for them. You can't….I'm not talking about reality, but obviously about perceptions, it's so important that you stay away from those and anything that sort of plays to that. It's really kind of the most difficult thing I think for people to deal with, is that you're in the business of winning. You're not in the business of changing public perceptions when you're in the middle of a battle. You do that later. You need to not play into that stuff. You see it go on all the time. The City Council hearings have the wrong people testify. You don't see it in a certain commercial level but people are constantly not being aware of the message that they need to create.

And while we're all trashing Giuliani for lying and stuff, lets not forget this guy won small, 40,000 margin originally and was hugely popular shortly thereafter. That's real success and he was playing on popular issues. Reducing caseloads on welfare is a real easy sell. It's a real easy sell. These are bad people, who take government money and I'm cutting them off. That was the message that was out there, whatever the reality is. And of course Giuliani barely hid the racial aspects. He still barely does too. It's easy because of Dinkins' racial background and Giuliani's. He doesn't even have to use all the code words that are there but that's part of the argument and sometimes you have to call it what it is and sometimes you don't. It's difficult.

Finally, that's just the kind of give back, here's the world, what is work? What do I think is working now? Economic development. Job creation on the one hand, hugely popular to justify corporate subsidies and a great argument for pumping money into areas that need money. Right now we're developing over the last couple of weeks a living wage campaign and what our message is for the City of New York to have all its contractors pay $10 an hour to anyone who's getting city money, or more if they are not providing benefits. There's a very good argument to a lot of the council members who represent poorer areas of the city that this is an economic development package because people who are getting $7 an hour, and then $10 an hour will spend money in their district. It's absolutely true. It's a more efficient way of doing an economic development package. And we have to address costs and what are the real costs. I'm not sure that it's fair to say you need to spend more money to get the right result with respect to welfare because the costs are so dramatically reduced in different areas. Certainly in policing, in crime. When the community is better off economically you have a lot of reduced costs in crime, in drug addiction and everything else and that needs to be factored in.

When you take away the popular pieces of public assistance, it's pennies on the dollar in terms of taxes. If you take out everyone who agrees health care subsidies and childcare and take those out, you need to be able to present yourself as very cheap. That's the bottom line. And I think finally, it's really difficult for all of us who are experts in a certain area to not see the forest for the trees. But as Nickey just pointed out, people don't, as a general rule, understand any of this. They don't even know, which is actually quite surprising that there is this five year time cut off, sort of in New York, sort of not. But that is kind of astounding. President Clinton really understood how to do welfare and he took the pieces of the Republican argument on ending dependency and forcing people to work and Congress did go overboard. Went much further than he wanted with respect to food stamps and stuff. But marketed it still in a way, with the Democratic sort of kind hearted, decent, probably nicer, I don't know about nicer but less severe than it would have been with the Republicans. Certainly the time had come for that stuff, for welfare reform.

We just have to keep that in mind and you can't be on your white horse and doing the right thing and talking about the complexities of the issue. You just need simple straight messages that people understand and they really aren't about complexities.

ANDREW WHITE: So we've heard the conventional wisdom side. We hear the numbers side next. Ronnie Lowenstein is from the Independent Budget Office.

RONNIE LOWENSTEIN, DIRECTOR, NYC INDEPENDENT BUDGET OFFICE: What you're going to get from me is the really dismal side. I was struck by Gordon's first statement, which was if you want to act while we're in a position of strength, you need to act fast. From where we're sitting, looking at the City's budget, that time has passed. We have a U.S. economy that at this point is growing at well under half the rate that we have enjoyed in recent years. New York City's economy hasn't been as affected as quickly as a lot of other places but we are starting to see real slowdowns. Yeah, we're still growing but nowheres near we were growing in the last couple of years. And moreover, over the last few months, we've also begun to see the impact on the City, which is a lot of the money that drives these sorts of programs.

So first we saw some impact on the City's business taxes, and that of course makes sense because those are particularly sensitive to business conditions across the country. As those conditions weakened, our tax take from our corporation tax and our unincorporated business tax and our bank tax, all started to decline. Not just slow but actually declined from the levels we've enjoyed for the last couple of years.

Then over the last couple of months, we've begun to see a weakness in city withholding, on the personal income tax. For the first half of this fiscal year that's about to end and our fiscal year ends in a couple of weeks, money was just booming. Collections were very, very strong. Then as the U.S. economy started to slow we said, "Well, gee, maybe the City's revenue sources will begin to weaken." But then January and February were just stupendous and we could barely put this stuff in the bank as fast as we were getting it. And that was largely because of the bonuses that were being paid out on Wall Street, which were all a function of the tremendous economy that everybody enjoyed back in 2000. So for the first couple of months of this fiscal year, we were still rolling in it and in fact, that will drive a significant surplus for the current fiscal year.

But more recently our early warning systems on the City's personal income tax looked bad. Withholding is genuinely down. What that is going to mean in terms of total city revenues for the next year is unclear. But unless things get a lot better very quickly, the city budget is going to be tighter over the next year and particularly the year after, than it's been in four or five. And that puts us in a very different fiscal position than what we've enjoyed in recent years.

Moreover, of course, and all the people in the room know this far better than I do, we're looking at uncertainty in terms of TANF reauthorization. It's very likely that the city will have less funding from TANF than we've had in the past. The city and state collectively will have less money from this source than we've had in the past. At the same time that caseloads will either stabilize or perhaps more likely increase. And that makes the choices that you're facing tremendously harder. Doug Turetsky jokes that I can't say anything without using the word tradeoffs so yeah, the tradeoffs are becoming harder.

IBO has been engaged in the last couple of months in a classic follow the money exercise. We've been looking at what the state has done with the TANF funds over the last several years. How much true TANF surplus has existed. What the state has done with those surplus funds. How much of those funds have come to the City and what the City has done with them. I only, at this point, understand a fraction of those dollars and I think the numbers are pinned, it's just being able to tell the story. But basically for the state fiscal year that has just ended, the City received nearly a half billion dollars in TANF surplus funds. And we are defining those funds as monies we get from the state that we don't have to write checks for in terms of individual grants. So there's about a half a billion dollars there. And the striking thing from the exercise that comes out even to us neophytes, is what the state and the city have done with those funds are very different.

At the city level, very little of that half billion dollars has actually gone for anything that most of us would consider to be general fiscal relief. Unlike the state level, where some of these funds have been used for tax cuts. Some of them have been used for increased spending of other sorts. A wide variety of fiscal purposes other than public assistance. That's not the case in the City. So for the last fiscal year only a very small fraction, just a couple of percentage points of that nearly half billion dollars has in fact gone to purposes other than things that we can broadly define as public assistance. A lot of the funding has gone for things like that childcare block grant. As the state has increased its funding, the city has followed suit and we've spent a lot more on that.

Similarly, the City has used a lot of the funds, nearly $100 million of the half billion, just over the last fiscal year, to increase spending for Title 20 programs. As federal monies have declined, the City has moved in and increased its spending there.

All of which is to say that as far as we can tell these TANF surplus funds, which may or may not exist in a year or two, at the City level, have been used for things that I think we can broadly characterize as public assistance and quite honestly from where you're sitting that's good news and bad news.

It's good news because it presumably moves you closer to where you want to be in terms of an efficient and equitable system. But it is also bad news because those funds are genuinely in jeopardy. In short, the tradeoffs that the City is facing are going to get harder and as we look towards ways to develop a more equitable and I think equally important, more efficient system of public assistance here in the City. We're going to have to do so with fewer resources than we've had in the past.

ANDREW WHITE: Diane Servino is with the Social Service Employees Union, bringing that last bit of financial gloom into the world of the welfare office, the job center.

DIANE SERVINO: Thanks, Andrew. As he said, I'm Diane Servino. I'm with the Social Service Employees Union. Currently I handle the union's political structure so I've been out of the welfare centers for quite a while now. Most of my interactions with it is peripheral. Actually, many of you are experts on welfare and even out members are in the social service delivery system. I never myself worked for welfare. I was a child welfare worker. I worked specifically with kinship cases in the mid-90s. However anyone who knows child welfare, knows it's the same caseload. Dealing with the same people on welfare that you're dealing with in child welfare, that you're dealing with in homeless services, that you're dealing with in Division of AIDS or protective services for adults. Our members work in all of those programs. And anyone who knows the history of Local 371, knows that we are a socialist, activist union. Our strike in 1965 was not just around our workers' issues but it was also around caseload reductions, management of caseloads and welfare so we could provide quality services to clients. And over the many administrations that have been there and the many commissioners and the many name changes, it's interesting that Mickey raised that issue, every new commissioner, every new mayor who comes in, looks around and decides how to handle this agency from a public relations perspective. So let's change the name. Let's change the names of the programs. Right now it's still HRA however it's no longer welfare. When I came in it was income maintenance, it became income support, now it's called Family Independence Administration and you don't have caseworkers, you have job opportunity specialists/employment counselors, etc. And you are no longer a client of the agency, you're a customer. Although what we're selling, I do not know. (laughter)

Sometimes we've been policy shapers with the mayor and commissioner and their deputy commissioners but under the current administration, we've unfortunately been relegated to policy implementers. Our members are out there working in these centers and the push for the past 8 years has been simply caseload reduction. Not poverty reduction, simply caseload reduction. And I'm not sure whether the man is lying to the public about how they reduce cases but I can tell you one of the primary reasons that caseloads have dropped is because of the increase of the fraud investigation program. So when the clients come in and they apply for benefits first, everyone reads the paper so they know that people have been stopped at the door. You have to come back within 24 hours. You can't apply today. So if you send people back eventually they will stop coming back.

But then we have this expanded fraud investigation program where we went out there and we investigate people and the primary goal of that, of course, is to disqualify people for benefits. So if you shut them down at the very beginning and you weed out the people who really are working but haven't been reporting to their welfare worker, you are eventually going to lower caseloads. But that number seems to be static.

More importantly to us right now is in their continuing effort to change the public perception of what welfare is, there is this attempt to reorganize public assistance right now. Over the years we've had eligibility done by one worker, you've had case work done by another worker, you might have a protective service worker, you might have a child welfare worker, you could have conceivably six workers, working on your case at any given time, working at cross purposes. Not knowing what the other one was doing and not having access to information from each other. And that has always been a problem for those of us who do this kind of work.

One of the things Turner would like to do before he leaves office is to create this single case management approach to welfare and certainly we approve of that. We've always believed in the single case management approach but our concern right now is who is going to do that work and by the end of this year? How is that new approach in welfare going to deal with the 65,000 people that they claim, this is their information, 65,000 clients…customers, are going to drop off of welfare on December 31st in the year 2001? That is going to be the real challenge not just for us who actually provide the services but for all of us in this room and the next mayor. What is going to happen to that caseload population? If you guys can help me out with that, I would certainly appreciate it.

AUDIENCE: Well, the state safety net.

DIANE SERVINO: Yes, that's technically what is supposed to happen to them, however there's not going to be any cash assistance. I don't think there's been an increase in the level of benefits since the 1970s. When I came in in 1990 as a caseworker, I had clients who were receiving $342 a month shelter allowance for a family of four. Well, that number really has not changed. And you can't rent a bathroom in New York City for $342 a month. So I don't see how the state safety net is really going to address the affordable housing problem. The fact that we are not really creating jobs in the workfare program. Where our workers are assigning people to do things like clean bathrooms. Well, it takes you three weeks to learn how to clean a bathroom. Then what do you do? Where do they move from there? They're not. So what is really going to happen to these people next year and the year after? And I think that's the biggest challenge for the next administration.

ANDREW WHITE: I want to start with diversion. Which is the very front end of the system. We continue to divert people in a number of different ways, whether it's just through frustrating them or legitimately finding people who aren't reporting income and there's not any adequate research into how many fall into what category. But the fact is that diversion is a huge force in driving down caseloads. I wanted to ask Bob Doar, who is here from the state, to give us in real practical terms, what is the value of diversion, how it's done well in your perception or in the way the state sees counties doing it. Where is it done well? Where does it have real value in the system?

ROBERT DOAR, EXECUTIVE DEPUTY COMMISSIONER for NYS Office of Temporary and Disability Assistance OTDA: I'm going to start back a little bit and give a general overview of how we see, in answer to your question about what the City should do to transition the next administration. So the first thing I want to say is I do think that what Gordon said in the beginning, that we are in more of a position of strength than we realize. That should be recognized. There have been dramatic caseload reductions, no question about it. But there also have been poverty rate reductions, as well, and I think there's some evidence