1. We’re debating whether signing the 1996 welfare law was, as the title of your Atlantic piece claims, “The Worst Thing Bill Clinton Has Done.” Your recent posting seems to suggest that welfare reform is merely “part”–or perhaps merely “a vital part”–of a “complete strategy” to end ghetto poverty. But your Atlantic piece didn’t fault the 1996 reform for being only a partial solution; you argued that it was no solution at all, that it was “awful … so bad … a terrible mistake … very bad … awful … terrible legislation” that would cause “serious injury to American children.” If you now say that welfare reform is only one component of a strategy that should include other laudable enterprises such as breaking down the city/suburb divide, ending job discrimination against “people of color,” improving schools and “community and economic development”–well, I’ll settle for that. So President Clinton is guilty of doing the welfare part first. Going partway toward solving “our nation’s most shameful poverty problem” is not such a meager accomplishment. It’s certainly not “the worst thing Bill Clinton has done.”
       2. But, I admit it, I am welfare-centric. There are good reasons to do the welfare part first. You say we should look at how ghetto-poor neighborhoods came to be. Fine. William Julius Wilson has told the story. First, African-Americans of all classes, victims of housing discrimination, lived in central-city ghettos. Then good unskilled jobs started leaving the cities for the suburbs–here Wilson relies on the “job flight” research of John Kasarda. Then middle- and upper-class blacks started leaving the central cities too. That left the poorest elements of African-American society behind, isolated from the job market. But why did they stay–how did they survive? Kasarda asks this question, and comes up with an answer: “welfare programs.” Without welfare, people would have had to move to where the jobs were, as they had done in the past (Northern blacks having just completed the great migration from the South). So welfare was a key element in the formation of ghetto-poor neighborhoods.
       3. And, yes, I think reforming welfare is the key to solving the problem. Retain the old AFDC system, and young unmarried women will still be able to have out-of-wedlock children and raise them without having a breadwinner in the house. We can try for a few hours in school each day to compensate for the handicap of growing up in these fatherless, jobless homes and neighborhoods, but the results haven’t been encouraging–at least not encouraging enough. Replace welfare with work, though, and all sorts of good things may happen. Women, realizing welfare won’t support them, may begin to make better choices: demanding more from the men in their lives, delaying childbirth, teaming up with breadwinners. Even single mothers will be working mothers. Not only will they do better themselves in the long run, but their children will grow up in homes conditioned by the discipline of work. Such children (as studies by Dolores Norton at the University of Chicago, among others, have shown) do measurably better in school. Your education initiatives would have a chance to work.
       4. After welfare is replaced, there’s also a chance to break down the city/suburb wall. The Gautreaux program, which you cite, makes my point. Gautreaux has placed several thousand black, inner-city families in the Chicago suburbs, with much success. But when the Clinton administration tried to expand the idea to five other cities, “the result was disaster,” as S



’s Jacob Weisberg reports in his book, In Defense of Government. The program was simply killed when suburban whites revolted. They didn’t want to live next door to welfare recipients, and I don’t blame them. Nobody wants to live next door to the welfare culture. Without welfare reform, improving what you call “the relationship of the inner city to the larger urban and metropolitan context” means trying to cram welfare recipients down the throats of reluctant suburbs. The suburbs will win. But if you first change the welfare culture to a working culture–even a poor working culture–the basis for suburban resistance grows much weaker. We then have some hope of replacing the vicious cycle of race discrimination and resentment with a virtuous cycle, in which suburbanites and employers learn that their stereotypes are inaccurate.
       5. I don’t think our “back-and-forth about whether the states are being mean or not” is somehow fruitless or uninteresting. The claim that states will be mean is the essence of your Atlantic argument against the welfare bill. The “fundamental problems” of the bill, you say, come from replacing the federal welfare entitlement with block grants to the states, which “can now do almost anything they want.” You argue that states will engage in a “race to the bottom,” encouraged by federal time limits and work requirements. Yet virtually all the examples you have offered of state meanness (except in New Mexico) proved, on closer inspection, to be not so mean after all. No wonder you want to change the subject. You may turn out to be right, if states are driven to meanness in the future. But there’s little evidence of this now.
       6. As for the famous five-year time limit in the bill, you claim to have pointed out, in the Atlantic, that it only prevents states from using federal money beyond five years–in other words, states can keep families on the rolls simply by using state funds. You did sneak in a mention that the five-year limit applies to “benefits paid for with federal money.” But that was after you’d talked repeatedly about “the absolute time limit,” and “the absolute lifetime limit of five years, cumulatively, that a family can be on welfare.” You argued that the “big hit, which could be very big, will come when the time limits go into effect–in five years, or less if the state so chooses” at which point “a large group of people in each state will fall into the abyss all at once.” A reader could be forgiven for getting the impression that the five-year limit means a big cutoff must come in five years, or sooner. That impression is false. “Very few states,” you say, “contemplate offering assistance financed with state money beyond the five-year time limit.” Well, all of the competing reform plans in California–one of your “macho race to the bottom” states–contemplate open-ended, state-funded aid, as does Republican Governor George Pataki’s plan in New York. In Massachusetts, another allegedly “macho” jurisdiction, the state plan explicitly provides “[b]enefits to recipients who have exceeded the … sixty month limit … funded through the State …” I called up Gary Stangler, Missouri social services director and chair of the National Council of State Human Service Administrators. He told me that while states are not yet focusing on what will happen five years down the road, “most of my counterparts are contemplating some state assistance after five years, at least for certain populations.” It’s not as if states actually have to ante up extra dollars to keep recipients on the rolls. All that’s required is a bookkeeping shuffle–pretending that the money used before five years is “federal” and the money after five years is “state.” States can also use various tricks (like classifying cases as “child-only”) that will allow them to continue aid without even pretending they aren’t using federal money.
       7. You are right, I think, when you say that in many states the burden will be on recipients to ask for extensions, and caseworkers will have a lot of discretion in deciding whether, for example, a recipient is really trying to find a private-sector job. That’s an important change, and not necessarily a good one–even if caseworkers take a paternalistic interest in their “clients,” paternalism may not be so desirable. We’ll see. I tend to favor a nonpaternalistic system that, wherever possible, says, “Here’s a job,” and leaves it up to the “client” to either show up for work or not. (You are the one who argues that elaborate, paternalistic government “coaching or case management” is needed.) But it’s one thing to say that caseworkers may have too much power in individual cases; it’s another to suggest, as you do in the Atlantic, that large groups of children will “fall into the abyss” due to mass cutoffs.
       8. I still think it was ridiculous for the federal Department of Health and Human Services (with your support) to deny Wisconsin permission to try its welfare plan–which features thousands of expensive public jobs and subsidized health care and child care–because HHS objected to the shape of the child-care copayment schedule. But there is a larger point I failed to articulate, about the advantage of letting states make their own mistakes (as the new law allows them to do). Let’s assume you and HHS are right about the child-care payments. If HHS can insist that states conform to its views on all such details, then it prevents Wisconsin from trying a bad idea and discovering it doesn’t work. The bad idea lives on. State officials can say “if only HHS had let us do what we wanted to do …” But if Wisconsin is allowed to try its idea, discovers it has flaws, and fixes it (as Wisconsin, left to itself after the 1996 reform, in fact did on child care), then everybody has to concede the bad idea didn’t work. It’s dead. The same principle applies to bigger issues: Let conservative states try to require work without providing backup public jobs. They will show everyone it can’t be done.
       9. You’ve reformulated several of your other objections to the Wisconsin plan. You say that in Wisconsin “when people go to work, they get no earnings supplementation for being in a low-paying job, even one in the private sector.” You propose something like “a state earned-income tax credit.” I hate to break it to you, but Wisconsin already has a state earned-income tax credit, which will be available to low-paid private-sector workers under the state’s reform. You also say you objected because the original pay for Wisconsin’s community-service jobs was lower than “average welfare payments.” It wasn’t. (The average welfare payment was about $512 monthly; the jobs paid $555). The larger point here is that while you say you aren’t concerned only with money–that you care about work and “personal responsibility”–you still assume it’s a horrible, fatal problem if going to work means that some welfare mothers get a few dollars less than they were previously getting on welfare. Yes, the designers of the Wisconsin plan care more about getting people working than getting them more money. They are right to do so. It’s better for children, and for communities, to have intact working families making 95 percent of the poverty line than nonworking non-families on the dole at 105 percent of poverty. (Susan Mayer’s new book, What Money Can’t Buy, apparently confirms that culture, not income, is what matters for childrens’ success.)
       10. Homelessness in Wisconsin is up 30 percent, you say. But you don’t deny this amounts to about 26 homeless families out of more than 7,700 who have left welfare. I agree, though, that the jury is still out; I would certainly expect homelessness to rise a bit more when Wisconsin fully replaces welfare with public-service jobs. That is a price I think we should be willing to pay, temporarily, to end the welfare culture.
       11. You are “willing to use compulsion” against women who refuse work, but want to debate the size of the “sanctions we would impose.” I had actually asked if you would keep sending checks to such a woman. I say if she doesn’t work, she doesn’t get a check. (I’m not sure that is “compulsion.”) You seem to be saying that you might “sanction” her by taking away only part of the check while still sending her the rest. But if the remaining part of the check is enough for her to live on–in California, under the pre-1996 law, being “sanctioned” meant losing only about $90 a month, out of a combined welfare and food-stamp benefit of $850–then it will be enough to sustain the welfare culture.
       12. It isn’t enough to be “work-oriented.” We have to actually require work. When liberals fudge this issue it infuriates the electorate. But when liberals advocate programs that apply only to workers, they tap into the voters’ generosity. The minimum wage, the earned-income tax credit, Social Security–all are examples of strict worker-only programs, all benefit the poor, and all are very popular. Clinton seems to have learned this obvious lesson. When will the rest of the Democrats?
       13. Public-jobs programs are another “worker-only” effort voters might support. Initially you chastised me for ignoring public jobs. Now you complain of my “overemphasis on public jobs.” Suffice it to say that, for reasons already discussed, we agree that public jobs are the key to successful welfare reform. And I agree with you that eventually they should be available on a permanent basis–and available to single men and women–in a sort of neo-WPA. But before we create a big public-jobs program for the whole low-income population of the entire country, let’s show, in one or two states, that the idea can work. By the way, another virtue of the welfare-reform plan in, yes, Wisconsin is that it takes a small step toward a broader jobs program by offering community-service employment to the fathers of poor children, as well as to the mothers (even though the fathers might not have been eligible for welfare).
       14. You criticize me–fairly, I think–for glossing over the “pyramiding effect of the cuts in aid to immigrants” in the 1996 bill. If states have to use their own money to make up for lost federal aid to immigrants, they will have less to spend on getting long-term welfare recipients into jobs. But the news, as of this writing, is that the worst of the immigrant cuts–about $9 billion or $10 billion worth of aid for poor, disabled legal immigrants–will be restored in the Clinton-Lott budget deal. Liberal editorialists (not you) ridiculed Clinton for pledging to undo some of what he himself did by signing the welfare bill. But now he appears, in large part, to be fulfilling his pledge. That’s another reason his decision to sign the bill looks like the right one.