Jason DeParle

By Jason DeParle
The New York Times Magazine
December 8, 1996

NOT LONG AGO, A CLASS OF HARVARD GRADUATE students joined their distinguished professor in a study of public attitudes toward welfare. That is, they opened their notebooks and switched on a tape of "Oprah."

Things got nasty fast.

A white guy in polyester sprang from the studio audience to condemn his sister-in-law, the "welfare lout" who "bilked" taxpayers for "new teeth." A white woman with beehive hair griped about food-stamp shoppers in "an Eldorado Cadillac." A black single mother with a low-paying job screamed at a black single mother with no job at all. "You have to show your children something better!"

For 60 long minutes, the halls echoed with working-class complaint.

"Why have all the babies?!"

"Get a trade!"

"Don't sit on your butt!"

As the insults mounted, the welfare mothers who had been invited on the show looked ever more angry and confused. One appealed her case to Oprah, complaining that the griping audience wanted her children to "to look nasty and have dirty faces." But the doyenne of daytime TV, herself a working woman, did not seem sympathetic. She faced the camera, locked eyes with America and asked the country this question: "Should you be going to work while someone on welfare's sitting at home with their feet up?"

The students scribbled, giggled, gasped and frowned, but none seemed more captivated than the man in the back row, a 43-year-old economist slouched low in his seat and nervously bobbing his knee. David T. Ellwood, the Malcolm Wiener Professor of Public Policy and the academic dean of the John F. Kennedy School of Government, was studying this particular "Oprah" broadcast for the 10th time. "Each time it's more painful," he whispered, jotting notes in the dark. "Why so much anger? What's going on here?"'

Perhaps no one has pondered those questions with more boldness, creativity and good intentions than Ellwood himself. And perhaps no one feels quite as unsettled at the nation's latest answers. To an uncanny degree, the story of the recent welfare revolution is a story of David Ellwood's ideas. Or, as he views it, a corruption of his ideas. More specifically, a corruption of his idea that cash assistance to poor women and their children should come with time limits. Now that those limits are the law of the land, Ellwood says, "I do lie awake at nights, worried about what's going to happen to our children."

The tale of Ellwood's journey has all the trappings of a Washington thriller -- tense meetings in the Oval Office and a triumphant ride on Air Force One -- yet the arc would be familiar to the ancient Greeks. The professor had an idea. But the idea gained a life of its own, to the profound dismay of the professor.

As a young academic, Ellwood wrote "Poor Support" (Basic Books, 1988), an instant anti-poverty classic that called for time limits on welfare, but only as part of a much broader plan to shore up the lives of the poor. Ellwood envisioned universal health care, expanded training programs, wage supplements, guaranteed child support and last-resort Government jobs for those who could not find them. He pictured a system that "ensures that everyone who exercises reasonable responsibility can make it without welfare."

For a while, the "Ellwood plan" shimmered only as a distant nirvana for policy wonks. But by the 1992 Presidential campaign, a wonk none other than Bill Clinton was stealing Ellwood's lines, promising to "make work pay" and to defend the "people who play by the rules." Soon he stole Ellwood too, bringing him to Washington to help fulfill Clinton's pledge to "end welfare as we know it."

But by the time Ellwood resigned from the Government last year, his proposal was in tatters, and Washington was converting the welfare debate into its own version of the "Oprah" show. Taking Clinton at his end-welfare word, the new Republican Congress pushed through unprecedented cuts in anti-poverty spending, and invited states to weave -- or to shred -- whatever safety nets they chose. Instead of time limits followed by a work program, Congress created time limits followed by . . . nothing. As Clinton went along, Ellwood, back in Cambridge, was fruitlessly urging him to resist.

Like Ulysses on the isle of Corfu, Ellwood now finds himself marooned, on the foreign shores of the "Personal Responsibility and Work Opportunities Reconciliation Act of 1996," and he is picking his way through its 400 pages, searching for hope in the wilderness. But Ellwood also can't help looking back. When talking about the shipwreck that landed him here, he sounds more numb than angry -- as though he's studying his own emotions from the same perplexed distance that he studies the "Oprah" show. "You do go over in your head all the things you might have done differently," he says. "You can't help thinking, 'Oh my God, all that work, and this!" He hesitates for half a second, then finishes the thought. "And I do have some frustration and some anger and some guilt -- all those things -- all tied up about it."

WALKING INTO A ROXBURY WELFARE OFFICE ONE DAY LAST month, Ellwood had his own chance to play Oprah. Local officials had assembled a half-dozen recipients to chat about the system, and they were gathered around a basement conference table, eyeing him warily. He was not quite sure where to start.

"Let me just say there are lots of people in Boston and Washington, trying to make changes, who've never been to a welfare office, and so forth," he said. "You meet with a governor and a President or senator, and they don't know what they're talking about. As far as I'm concerned, you guys are the real experts."

The room was silent. Ellwood tried again. "Here I am, a guy at Harvard," he said . "Not exactly connected to the system. And yet I'm involved in making decisions. I feel like I gotta spend time, talking and so forth."

It was as awkward as the start of a junior-high-school dance, but it worked. Ten minutes later, the women couldn't stop interrupting one another. "Once you get on, you tend to get a little lazy," a mother of four who has collected a check for five years said. A woman in a training program agreed. "People think they can sit and have a football team full of kids, and the state's gonna take care of them," she said. Few subjects are as close to Ellwood's heart as the need for child support enforcement, and he beamed as he discovered an ally. "Momma's home with the babies, while Daddio's out at Wonderland," she said. Aim for his wallet, and "jack-his-pants-up!"

As he pledged thank you notes to all involved, Ellwood couldn't have been more pleased. The morning-long session confirmed his three central beliefs: 1) the current welfare system is a disaster; 2) the new law will make it worse; 3) his original plan wasn't just a better solution, but one the recipients themselves would support. "They all said that time limits of some sort make sense," he said , the professorial authority returning to his voice as heads back to Cambridge. "I emphasize: of some sort. Once you asked, 'What if they just cut people off after two years?' they all said that'd be a disaster."

It was a subjective read of a highly unscientific sample -- but plausible. Most of the women did confess to feeling trapped by a system that gives something for nothing, and they at least professed to find time limits reasonable. ("If you can't get it together in two to three years, there's something wrong with you," one had said.) At the same time, the details of their misfortune were complicated, and not necessarily set right by a kick in the pants. Two of the five women had asthmatic children, whose care sometimes made it difficult to work. While one had a mother who can baby-sit, the others called reliable child care as elusive as a reliable man.

The answer, in Ellwood's view, is as clear as always -- a mutual responsibility pact in which poor women agree to work, and taxpayers pledge the health care, child care (and, if necessary, the jobs) they need. As it happens, there's nothing in the new welfare law to prevent states from concocting their own version of the Ellwood plan. But there's also nothing to keep them from offering a 30-day time limit and a bus ticket out of town. "States can do the right thing in this bill," he says. "But the incentives are to do the wrong thing, to just cut people off the rolls."

The law is really two separate initiatives, joined at the hip. The first is a set of Federal budget cuts, designed to save $55 billion over the next six years. Of that, $24 billion will come from programs for legal immigrants, primarily Medicaid, food stamps and Supplemental Security Income. Another $23 billion will come from other food stamp recipients, including families with children. When the cuts are fully implemented, the food stamp program will have shrunk by nearly 25 percent.

The second part of the bill covers what people traditionally consider "welfare": payments to single mothers with children. Here the changes are more significant still, but for altogether different reasons. In the short run, the Federal money actually grows. But the old program (Aid to Families with Dependent Children) has been abolished, and the new one (Temporary Assistance to Needy Families) shifts autonomy almost entirely to the states, with long-range incentives to spend ever less.

States can immediately cut their own spending by 20 percent. They can shift 30 percent of the Federal money to child care programs for other constituents. They must limit any family's cumulative, lifelong benefits to less than five years (though 20 percent can be exempted for hardships). And while states are supposed to put their recipients to work by the year 2002, the fine print tells another story. States can meet this standard either by placing families in a work program (an expensive option) or finding ways to shear them from the rolls (a cheap one). "Not only do we allow states to cut people off -- we encourage them," Ellwood says.

Between the benefit cuts and the time limits, the Urban Institute, a nonpartisan research group in Washington, has predicted that as many as 2.6 million additional Americans will fall into poverty. The key word, of course, is "predicts." No one knows for sure what will happen, and not every advocate is predicting disaster. "I think there are more opportunities to do good things with this bill than Davis realizes," says Toby Herr of Project Match, a training program for long-term welfare recipients in a Chicago Housing project. Even Ellwood thinks that a few states will design better systems. And where they don't, some of the poor may prove more resourceful than their advocates think. "There's a lot of people out there that don't like cleaning nursing homes," one of the Roxbury women had said. "Let me tell you, if it's two years and you get nothing, they'll start liking that nursing home."

But as Ellwood notes, the long-range logic suggests even deeper subsequent cuts. As an "entitlement," welfare spending used to rise with caseloads; as a "block grant" it will have to compete with other, more popular programs like TK [education grants, etc.]. "Who's going to stand up and say, 'Cut Medicare instead?' " Ellwood says. By the time he reaches Cambridge, his morning bounce has faded to a mid-afternoon mournfulness. "The welfare system is in a death spiral."

"I CAN'T SAY I'M A RAGS-TO-RICHES SORT OF STORY," ELLWOOD BEGAN one afternoon, his feet up on a coffee table in a fourth-floor office that looks out at the resplendent Cambridge hills. "I grew up in an upper-middle-class family where everyone believed in a form of public service. Interestingly, it wasn't a family where everyone was volunteering all the time, though we did some volunteer work. It was a family where everyone was focused on systemic change."

His characterization underscores the struggle Ellwood faces, as a big-thinking poverty fighter from a privileged Minneapolis suburb. The notable thing about his childhood home wasn't just its affluence, but its splendid, benevolent isolation. The Ellwoods had their own wooded acre on Christmas Lake, a half-hour from downtown but a world away from its concerns. "I now prefer skiing in 'deep bottomless powder,' " he wrote to his grandparents when he was 13. "This weekend I debated on the question of 'compulsory service for all citizens.' " Even when he ventured overseas as a high-school exchange student, Ellwood wound up in Sweden. It's possible to read his subsequent career as an effort -- noble or naive, incisive or quixotic -- to understand the murky realms of race, class and human need from which his own upbringing protected him. "It was a great life, an idyllic life, one I can't reproduce for my kids," he says.

What truly set the Ellwoods apart wasn't just the neighborhood, but an ethic of public service that dates back generations. His father's father was a physician who frequently treated the poor, and his mother's father was a preacher. But the ancestor with the truly metaphorical occupation is his great-grandfather, Willard Ellwood, a church architect. In sacred form or secular, the Ellwoods have been sketching blueprints for public virtue for the past four generations. The oldest of three children and the only son, David was saddled from the start with the mixed blessings of his early intellect. Though he can now appear confident bordering on brash (his detractors often say "arrogant"), he spent his childhood, in his mother's words, as "a bit of a loner." His sisters, Cynthia and Deborah, smart achievers themselves, prospered with public educations. But by the sixth grade, David was eager to attend the elite Blake School, an all-male academy. He more or less kept to himself, and excelled in math and science. "I definitely wasn't very good with girls," he sighs. "I guess I was kind of a nerd."

And one with a formidable father, obsessed with changing the world. David Ellwood was entering his adolescence just as Paul Ellwood was firming his conviction that he held the cure to the nation's health care system, with something he dubbed the "health maintenance organization." The dinner hour became a lecture series on his conversations with the Nixon Administration, which was giving his cause a hearing. "He would talk about it constantly, just constantly," Ellwood says. Among the characters in the story was someone at the Department of Health, Education and Welfare called "the assistant secretary for planning and evaluation." Seeing the power the crucial gatekeeping post held over his father, Ellwood aspired to hold it one day. (He succeeded at the age of 39.)

When he entered Harvard in 1971, Ellwood was still a chemistry buff who had stunk up his parents' home with mad-scientist experiments. But as college chemistry grew increasingly theoretical, he found a more useful tool for decoding the complex world. "What I like about economics is that it starts with the basic incentives facing individuals -- then what's neat is it aggregates those choices, up to the systemic level," he says. He emerged in 1975 as a newly minted summa graduate, less socially awkward than when he entered, but scarcely a portrait of sophistication.

He had a reputation as a computer whiz, and a consulting firm with a Medicaid contract invited him to apply for a job. He showed up for his interview in a plain white T-shirt. "He seemed incredibly bright, but a little naive about what this world involved," recalls his supervisor, Marilyn Rymer. Shortly after their work ended, she discovered his interests had expanded beyond computer models; at one point, Ellwood asked her to estimate the probability that "something would work out" between them. As a 29-year-old professional woman looking at a doughy 21-year-old computer buff, Rymer, 51, thought the probability close to zero. But to let him down easily, she said 10 percent. They married four years later. They have two daughters, Malinda, 17, and Andrea 12, and Ellwood is still known, in affectionate moments, as "Mr. 10 Percent."

The bond with Marilyn gave Ellwood not just his family but also a guide to anti-poverty policy. She had begun her career in a Georgia welfare office, and a subsequent consulting project took the two of them to welfare outposts across the south. Ellwood enrolled at Harvard as a graduate student in economics, and wrote his doctoral dissertation about jobless black teen-agers. He began a teaching job at the Kennedy School in the fall of 1980, just shy of his 27th birthday.

At the time, the welfare debate was deadlocked. Conservatives complained about long-term dependency, while liberals insisted that most recipients left the rolls in two years. In a groundbreaking study with Mary Jo Bane, a Kennedy School colleague, Ellwood returned to his computer and proved them both right. Most of the people who went on the rolls did leave in two years. But a substantial minority stayed. That meant the majority on at any given time were in the midst of a long-term spell, one that averaged eight years. "Welfare had two faces," he says.

The "dynamics study" had all the hallmarks of Ellwood's subsequent work: it was empirically grounded, nuanced, accessible and relevant. And it allowed readers of varying ideologies to read into it what they would. Though it was never published in an academic journal, the study went on to a samizdat life that changed the welfare debate of the 1980's. Though liberals still emphasized the considerable turnover, it became increasingly difficult, in light of Ellwood's data, to dismiss dependency as a figment of conservative bias.

Politically, however, Ellwood remained a liberal, defending the system in public talks. As he did, he says, "I got creamed," by social workers and recipients themselves. "I realized I had to do some basic rethinking." The result is Ellwood's landmark text, "Poor Support," whose double-edged title delivers his guilty verdict on the welfare system. While it's an impressive book, its appeal resides less in policy esoterica than in its measured common sense. "We want to help people who are not able to help themselves, but then we worry that people will not bother to help themselves," he writes. His solution isn't to reform welfare -- and inevitably bog down in that tradeoff -- but to alleviate the need for it by shoring up all low-wage workers and single parents.

Ellwood suggested wage supplements, health insurance, job training, child care, better child support collection and public-service jobs -- policies he hoped would raise incomes without removing the incentive to work. With those "poor supports" in place, he suggested, the country should replace an open-ended welfare system with time limits of two or three years. The idea of time limits was not a new one. But with the endorsement of a prominent, center-left Harvard academic it quickly gained new credibility. "Some of this is just Minnesota, O.K.?" he says. "Minnesotans basically believe you oughta help your neighbor help himself or herself. I just think that's as American as apple pie."

Soon Ellwood himself was making an appearance on "Oprah." Knowing that his time on the air would be minimal, he shortened his message to bumper-sticker length. "If you work, you shouldn't be poor," he said. (That is, raise the minimum wage and tax credits.) And "one parent shouldn't be expected to do the work of two." (Get tough with child support collection.) In truth, many people had a surer knowledge of the welfare program's intricacies. But Ellwood could articulate a plausible solution in crisp, appealing terms. He had a winning rap.

By the 1992 campaign, it wasn't just the broad outline of Ellwood's plan that was coming out of Clinton's mouth, but the aphorisms too. Yet the two had met only in passing, years earlier, and they never spoke during the campaign. Ellwood was thrilled, and a bit alarmed -- especially by the most potent Clinton sound bite of all, the pledge "to end welfare as we know it." Ellwood feared this was so vague it could be taken to mean anything, including just literally "end it." (In a recent speech he called the Clinton phrase "vacuous and incendiary.")

Many liberals had harbored similar anxieties about misinterpretations of Ellwood's work. They welcomed his calls for increased spending, but worried that conservatives would hijack the time limits while forgetting everything else. Now Ellwood was fretting, too. "I don't think these are issues that are best discussed under the klieg lights and sound bites of a Presidential campaign," he told me at the klieg lights and sound bites of a Presidential campaign," he told me at the time. A month after the election, he was introduced at a meeting as the godfather of time-limited welfare. "I deny paternity," he said.

But as he was leaving his office a few weeks later, a call arrived from Washington. It was Donna E. Shalala, the incoming Secretary of Health and Human Services. She had spoken to the President, and he had approved her suggestion that Ellwood join the Administration to help craft the welfare plan. The title: assistant secretary for planning and evaluation. Reminded a few months later of his previous doubts, Ellwood's new enthusiasms were on full display. "Now I work for the President," he said.

BUT HE WAS farther from Minnesota than he knew. Shortly after arriving in Washington, Ellwood was escorted into the inner office of Senator Daniel Patrick Moynihan, where he was greeted with a sour prediction. "I'll look forward to reading your book about why it reform failed this time," Moynihan said, referring to welfare reform. Clinton's promise -- two years of training, followed by a work program -- would cost billions. And in Moynihan's view, the public was looking to spend less on welfare, not more. "There simply is no money," Moynihan said. "None! So what are you going to do?"

Ellwood remained optimistic. From his perspective, the Federal budget seemed vast, and surely with Clinton's commitment such details could be worked out. In the spring of 1993, Clinton named Ellwood as one of three leaders of the task force that would draft his bill. (The other two were Bruce Reed, a White House aide, and Mary Jo Bane, who had also become an assistant secretary of health and human services.) But the task force soon grew unwieldy, with three dozen members from across the Government. And suffering from high-level neglect, the task force took a year to finish its work.

Moynihan's gloomy forecast proved prescient. The search for the $10 billion to $15 billion Ellwood needed for training, jobs and child care bogged him down in months of bureaucratic wrangling. The enemies of the plan repeatedly leaked his lists of possible tax hikes and spending cuts, leaving him embarrassed and preoccupied with a secrecy he could not preserve. And his relationships quickly frayed on Capitol Hill, where Ellwood's professorial air was often interpreted as arrogance. Unlike some Washington officials, who use their egos as weapons, Ellwood seemed disturbed to discover his condescending side, and he sought colleagues' advice about setting things right.

Ellwood also found himself fighting an ideological battle with Reed, whom he once jokingly called "a right-wing hack." Though their relationship was friendly and respectful, Reed was, in fact, considerably to Ellwood's right. When he joined the task force, Ellwood was pushing hard for a system of "child support assurance," in which the Government would make the payments itself if it failed to collect from absent fathers. The idea was a centerpiece of "Poor Support," and in Ellwood's mind far more important than time limits. But Reed nipped the notion in the bud. He called it welfare by another name, and stressed that Clinton had never endorsed it.

As the process progressed, Ellwood discovered that Reed had a much different vision of time limits in mind. To Ellwood, the idea meant that after two years, recipients would have to work; but the Government would insure that the work existed, and that the checks would continue to flow. To Reed, "ending welfare" ultimately meant just that: ending welfare, not continuing it through some make-work scheme. At some point, he argued, eligibility for the work program itself should end. The details sound trivial, but they cut to the philosophical core: Is there a safety net or not? The two ultimately appealed to Clinton himself, who in a final Oval Office meeting essentially split the difference.

The process was long and harrowing and sometimes embarrassingly public, but in the end it produced a plan. There was less money for child care than he had hoped, and guaranteed child support survived only as a small experiment. But Ellwood still glows when he recalls the final product. "I'm incredibly proud about the welfare bill," he says. "We all liked it. The Cabinet liked it. And I think it's where the public was at." Clinton flew to Kansas City to unveil the plan in June 1994, and Ellwood went along to brief him aboard Air Force One. He split off in Missouri for an appearance on "McNeil-Lehrer." "And I fly home alone in a middle seat on an absolutely packed airplane -- and I have to say, I just loved it," he says. "There was some Calvinist side of me, feeling like, "O.K., now come back down to earth. This is only stage 1.' "

Indeed, it was. The political landscape had been transformed in the two and a half years since Clinton had first pledged to "end welfare" -- transformed, in no small part, by the pledge itself. Clinton's bold oratory, followed by years of delay, had created expectations that his plan could not fulfill. Never mind that it was far tougher than anything a President had ever proposed, including Ronald Reagan. The Republicans were drafting their Contract With America, which went beyond work programs to endorse absolute bans on cash aid to young mothers.

Liberal House Democrats feared that an election-year welfare debate would be a disaster for them.. Eager to defer consideration of the bill, they needed a target, and they found one in Ellwood, the Harvard man with the plan. In four days of hearings in July 1994, Representative Robert Matsui of California leapt on Ellwood's every word. Some people might even question Ellwood's motives, he said, and assume he was "pushing welfare this year to enhance his resume." The message was clear: the bill wasn't moving.

It died a few months later when the Republicans captured Congress and started laying their own plans to "end welfare." The Democrats could no longer indulge their internecine disputes, and Ellwood and Matsui, along with many others, united to play months of defense. But Clinton himself showed little stomach for the fight. At the nadir of his political fortunes, he was wary of struggles that might further tag him as liberal. And as a former governor, he had more trust than his aides in unfettered state choice. His moment having passed, Ellwood resigned in July 1995. A year later, six decades of welfare policy was swept away, while Ellwood, vacationing on his father's Wyoming ranch, penned a lonely op-ed to urge a Presidential veto.

What, he now asks, went wrong?

It's a more complicated question than it may seem, and inevitably, some fingers already point Ellwood's way. By polishing the notion of time limits with his liberal academic gloss, they say, Ellwood was asking for it. "I remember saying to him at the time that in the real political world, the time limits would survive, and the rest would be brushed aside," wrote Frances Fox Piven, a Columbia University sociologist, in a recent issue of The American Prospect, the liberal journal. In a subsequent interview Piven went on to say that Ellwood was "in a wonderland" to think that policy experts could keep control of a national welfare debate. "This is a pretty nasty and ferocious politics," she said. "Welfare has always been a scapegoat."

Ellwood, for his part, is willing to say, "I was incredibly naive about a number of things," but whether he means it is another question. He says the public's disenchantment with welfare was so severe that some upheaval was bound to occur. While he sees mistakes, they are mostly at the level of strategy and tactics. He faults himself for not building better relationships with legislators and reporters. And he faults the senior-most White House officials for not lending the effort more support. Had the Clinton bill been introduced a year earlier, Ellwood says he believes it would have passed. "The values and the direction made sense at many different levels," he said. "It really was the proverbial hand up, not a hand out."

To believe, as Ellwood does, that he got agonizingly close is to believe several things. One is that the country is willing to spend more to make welfare reform work. Another is that the public will tolerate the inefficiencies of a Government jobs program. A third is that the country can debate welfare without unleashing the demons of the "Oprah" show, especially the unspoken racial animosity that inevitably comes into play. It's an uphill struggle to believe those things, but not, in the end, impossible. Though the taxpaying public feels angry at the welfare poor, it feels guilty about them as well. It wants to spend less; it wants to help more. Its attitude is volatile and contradictory and hard to predict.

But susceptible to leadership, especially in the hands of a politician as skillful as Clinton. Sure, Ellwood had a streak of naivete and a weakness for optimistic forecasts. But he was hired to be an assistant secretary for planning and evaluation, not a political consultant. Clinton's words radicalized the debate, not Ellwood's, and the responsibility for the subsequent politics resided with the President himself. The Ellwood plan always had a chameleon quality, and in Clinton it found a chameleonlike champion, whose true coloration remained impossible to fix. Did he abandon his original vision in favor of election-year votes? Or does he really believe the bill he signed will do right by the poorest Americans?

"Clinton is a complicated man, with some wonderful good qualities and some very troubling ones," Ellwood says with a sigh. "I really believe there are some core convictions there, but I also believe, obviously, those are more flexible than I wish they were. I think that Bill Clinton honestly believes he did the right thing in signing the welfare bill. What I don't know is: 'If Bill Clinton were not President of the United States running for office, would he still feel that way?' "

IT WAS COLD IN Cambridge on the night Clinton captured 379 electoral votes and a coveted second term. While Clinton was in a Little Rock holding room, contemplating his place in history, Ellwood was walking across the Harvard campus, with his hands jammed in his pocket.

He has spent much time recently trying to conceive viable "fixes" to the bill. The problem is he can't think of too many, especially when he applies a political reality test. "This bill is not going to get fixed," he says.

But Ellwood's spirits brighten a bit when he turns away from the law and back toward the country that produced it. He says he believes that taxpayers have a legitimate gripe: they want the poor to work.

Of course, as Ellwood well knows, that begs the bigger question: how to make work work? How to move four million low-skilled women, many of them borderline disabled, into a labor market where entry-level wages have been declining for two decades? How to lift their households out of poverty?

The more Ellwood mulls those questions, the more he returns to the blueprint in "Poor Support": tax credits, child care, child support, medical care -- in exchange for a commitment to work. For all his disappointments, some of his plans did make it. The new welfare law seeks to improve child support collection, and many of its provisions were borrowed straight from Ellwood's bill. ("One parent shouldn't be expected to do the work of two.") And for all the spending cuts, the country has significantly bolstered the wages of workers; it has raised the minimum wage and expanded a tax credit program that now offers payments of up to $3,500 a year. ("If you work, you shouldn't be poor.")

With those contrasting legacies in mind, there's a way to start interpreting Ellwood's tumultuous ride through the "Oprah" studios of life. Haltingly, imperfectly, the public may still be willing to help those who help themselves. To support programs for workers, that is. But the country does not seem as willing as he had hoped to make sure that jobs exist. And it seems especially unwilling to fret over the fortunes of those who do not or cannot work. Even if they have children. Even, for that matter, if they wind up on the streets. And that's just not what an optimistic reformer from the Minneapolis suburbs counted on a decade ago, when he bet on poor support.