Dear Mickey,

       Thank you for your opening missive. I count you among those supporters of this legislation who genuinely believe it will reduce poverty, and I note our areas of agreement. Beyond that, however, I think you are mostly mistaken in your analysis.
       1) Yes, the caseload is dropping, and that is a good thing, and I hope it continues. But it doesn’t mean what you say it means. You should tell the whole story. In 1989 there were 10.8 million people on Aid to Families with Dependent Children. During the recession of the early 1990s the rolls shot up to 14.3 million people. In the last two years 2.5 million people have left the rolls. Even with whatever additional numbers have left since the last count was released, there are still around 11.5 million people on the rolls. All that has happened is that the bubble of the early ‘90s has disappeared, and we haven’t even gotten back down to where we were in 1989. Why have people left the rolls? This is not a big mystery. We have had sustained, relatively low unemployment for quite a while now. I would have expected the decrease to have begun earlier and gone further, actually. It was late in coming because there is such sustained structural weakness in the low-wage end of the labor market (which is a problem we have had for 25 years, actually). Who has gone off? As you know, the welfare caseload is composed of two big groups: those who go on and off, and those who stay on for long periods of time. There are, of course, some who come on only once, but 70 percent of those who go off within two years come back on again later. (This makes cumulative lifetime time limits a problem even for people who use the welfare system only as a safety net between jobs.) Common sense makes me suspect (although I cannot prove, of course, and no one can prove the contrary, either) that those going off are mainly the people who go on and off. They are getting jobs (and keeping jobs longer). There are still nearly 4 million adults on welfare (now TANF, Temporary Assistance to Needy Families, instead of AFDC). About half of those are people who are long-term recipients. They are the people who are the target of this legislation. We have not even begun to reach them in any significant measure yet with private-sector work. The tough part of the challenge lies ahead. It is premature to pronounce success. And such pronouncements are not very helpful, because it makes it seem as though the longer-range task is going to be easier than it in fact will be. Why did the rolls drop an extra amount in states with highly publicized, new, tough approaches? The truth is, neither you nor I know exactly. But reports that I get from Wisconsin tell me that the bureaucratic culture discourages people from applying and is very tough on sanctioning people who don’t meet the various reporting and behavioral requirements of the new system. Homeless shelters in Milwaukee report increases in demand. Those are unlikely to be people who have found work because of the new approach. The Minnesota-Wisconsin comparison is interesting. Minnesota has a constructive approach to welfare reform (with a Republican governor, by the way). They let people who go to work in low-paying jobs keep a part of their welfare payment, so they can get out of poverty. They have studied this carefully, and found that the rate of job-taking has been better and the rate of job-keeping has been much, much better. But, people who get an income supplement from welfare to fill out a living income are counted as still being on the welfare rolls. So Minnesota’s rolls are not going down at the same rate as Wisconsin’s.
       2) You point to extra money that is there as a result of the welfare rolls going down. There is some, to be sure, and that is good if the states use it to help people go to work, although it is still true that because the funds are fixed at $16.4 billion, their value loses ground to inflation and to population increases as time passes. Plus, this is really a prosperity-based bill. When a recession hits, all the wonderful assumptions underlying the new law will be thrown out the window. I heard you say in our radio debate that we have come to a time when we won’t have any more recessions. Did you really say that? I’m not going to take up space to answer that one. The further problem is that the states are not exactly leaping into the fray to invest their own funds in this endeavor, or even to use all of the federal money for this purpose.
       The governor of Connecticut is proposing to use about 40 percent of his so-called “windfall” to refinance existing state-funded programs for low-income people, so he can cut taxes. Gov. Pete Wilson of California wants to cut $274 million of state money from programs for low-income people. This is an effort that will cost money, Mickey. There has to be child care available and affordable in sufficient measure if people are going to be able to go to work. In Minnesota they are adding $50 million for child care. In New York, which is six times the size, the governor is proposing to add $54 million. In New Jersey the governor announced a few weeks back that she was going to have to cut child-care subsidies for people already working in order to be able to provide help to welfare recipients now entering the job market. Health coverage is a huge issue. The states are all over the lot on this one. Vermont covers children up to 225 percent of the poverty line and adults up to 150 percent of the poverty line. Other states–too many–provide the minimum one year of transitional Medicaid coverage. If people don’t have health coverage, especially for their children, it puts them between a rock and a hard place. The welfare supplementation that I mentioned is another cost item. Transportation help, literacy training, alcohol and drug treatment–the list goes on and on. Of special importance in the cost department is the support that people with little work experience need to make it in the job market. Call it coaching or case management. But if we want long-term public-assistance recipients to succeed in the job market, there has to be this kind of help. Seventy-one percent of the Project Match participants in Cabrini Green in Chicago lost their first job by the end of the first year. But they had help and support, and by the fifth year 54 percent were working all year. It can be done, but it takes an investment. And all of this assumes that relevant, geographically accessible private-sector jobs are available. That is in fact far from obvious, to understate the matter, and again, that is now, when we are at the top of the business cycle. I suggest we need a serious discussion of how to use some of the money–federal or state, or both–to fill in the gaps with public jobs if it turns out that private-sector work is in unduly short supply. (And we should be prepared to be honest: If there aren’t enough jobs and we aren’t willing to create real public-sector jobs at tasks we want done, we shouldn’t have time limits.)
       3) You say the bill’s “draconian” (your quotation marks) provisions are phony. That’s a huge oversimplification. It depends on state behavior that in fact is turning out to be more negative than positive around the country. The No. 1 bumper-sticker assumption of the bill is that there are enough jobs out there. You think there are. I think there aren’t. We’ll see. The No. 2 bumper-sticker assumption is that all of the people on welfare can just walk out of their houses and take the jobs. That’s not true either, as I said above, and I’m not sure where you stand on that one. The bill can turn out not to be draconian if a state chooses to do things right (which I believe will cost the state at least as much as it was previously spending). I would list Vermont, Minnesota, and Oregon as states that are basically doing it right, and maybe Iowa and Missouri as well. (I know there are local advocates in all of those states who would dispute my positive assertions even as to those places.) But the problem is all of the states, including most of the big ones, where there is what I call a macho race to the bottom about the time limits, and about other things as well. Some have lifetime time limits of less than the federal five years. Florida has four, for example, and Connecticut has 21 months (albeit with exceptions that can extend it for some people). The governor of New Mexico has proposed a rigid two-year time limit, and he proposed that even a month of welfare for a seasonal worker in a given year would use up a year of eligibility. Lots of states have time limits within the time limits. In Massachusetts a person can be on for two years out of five; Florida, two years out of four; and so on. Gov. Wilson is proposing that new recipients can only be on for a year out of each two. I’m just using time limits as examples. There are lots of illustrations of grossly inadequate or unrealistically tough provisions in the various states. All of that is the part of the answer that makes the bill turn out to be draconian.
       We could have experimentation within a structure of basic national definitions of eligibility and procedural rules. That’s what Clinton’s waiver strategy was about. I didn’t agree with all of his waivers, but that was a lot better than this bill, which allows nasty “experiments” as well as good ones. As far as the Health and Human Services delay of the Wisconsin waiver goes, it was because it was troublesome in a number of ways. It promised child care up to 165 percent of the poverty line and then proposed copayment fees that were impossible for people to afford. Those have been somewhat lessened recently, but they were unworkable at the time. The “community service” jobs offered “pay” that was less than the welfare payment and way less than the minimum wage. Again, the pay has been raised somewhat in recent weeks, but it is still the case that the jobs are not defined as jobs, so the earned-income tax credit doesn’t attach to them. The Wisconsin plan is interesting in its rhetoric and premises. As is usual, the devil is in the details. I think you should look at it a little more closely before you pronounce it the best in the nation. It is the best publicized. That’s a different point.