LBJ brandishes the freshly signed War on Poverty bill in 1964.
ARNOLD SACHS/CONSOLIDATED NEWS PICTURES/GETTY IMAGESLBJ brandishes the freshly signed War on Poverty bill in 1964.

Nearly half a century ago, I dropped out of graduate school and enlisted as a foot soldier in America’s War on Poverty. Today, I’m still on the front lines, working to move people out of dependency and into employment. But with an important difference: I’ve become fed up with the useless policies that I once supported, and I’m trying to change the strategy of our bogged-down army.

We know for certain that income transfers, the preferred tactic of generations of liberals, have utterly failed to end poverty. My firsthand experience with welfare clients has shown me why: being on the dole encourages dependency. Working at a real job, by contrast, is the surest way for a person to climb out of poverty. Accordingly, the surest way for the government to fight poverty is to eliminate cash assistance almost entirely and offer jobs instead.

During his all-too-brief presidency, John F. Kennedy signaled that he wanted to reform the nation’s Depression-era welfare system by giving “a hand, not a handout” to the poor. As Charles Murray noted in his magisterial study Losing Ground, Kennedy’s small initiative, which “consisted of a few training programs and other rehabilitative efforts amounting to only $59 million in the 1963 budget, . . . represented a major departure nonetheless,” since it shifted welfare policy “away from the dole and toward escape from the dole.” When President Lyndon Johnson expanded Kennedy’s program into the War on Poverty, he likewise wanted not to mire generations in dependency but to free them from it. “The days of the dole in this country are numbered,” Johnson promised at the signing ceremony for the War on Poverty legislation in August 1964.

Listening to his soaring rhetoric, I believed that our nation was on the cusp of one of the great peaceful revolutions of modern times: the elimination not only of welfare but also of poverty and want. After all, by the mid-1960s, America was the world’s most affluent society, and economists predicted that the economic boom and high employment rates would continue for many years to come. The “conquest of poverty,” the 1964 Economic Report of the President explained, was “well within our power. About $11 billion a year would bring all poor families up to the $3,000 income level we have taken to be the minimum for a decent life. The majority of the Nation could simply tax themselves enough to provide the necessary income supplements to their less fortunate citizens.” The following year, the government allocated even more than the report had called for—$14.7 billion—to transfer payments.

That money deployed an arsenal of supposedly innovative weapons, including “community action” organizations requiring “maximum feasible participation” by the poor. Social-policy elites and the media jumped onto the bandwagon, predicting that the war would finally overcome the “structural poverty” imposed by the existing economic order. That understanding of the causes of poverty had been bolstered by the success of Michael Harrington’s 1962 book The Other America, which directly influenced JFK.

It was in this political and ideological atmosphere that I became a young antipoverty warrior, arriving in New York in 1965 to work for the Antipoverty Operations Board, a new agency created by Mayor Robert Wagner to manage the city’s cut of the federal War on Poverty dollars. One of my first assignments was to help write the city’s funding proposal to the federal government. I recall going to City Hall at 3 AM one day, getting a deputy mayor’s signature on that proposal, and then racing to LaGuardia Airport for the Eastern Airlines shuttle to Washington, D.C., so that I could deliver the document to the Office of Economic Opportunity just ahead of the official deadline. Through the first few years of the program, I worked for the city with other young, liberal policy analysts, making decisions about which local antipoverty groups deserved federal funding.

It’s almost impossible to describe the excitement that we felt as we crafted plans for new entitlement programs with few budget constraints. The programs earmarked for federal funding offered health care, education and training, housing assistance, counseling, and other social services meant to prepare the poor for their new responsibilities. Inspired partly by our slain president, who had challenged us to ask what we could do for our country, and partly by our belief that the economic system needed to be humanized by compassionate social-justice policies, we believed that we were part of something great and good.

But the government’s unprecedented expenditures failed to bring about the decline in poverty that Johnson had promised. Instead, they made things worse. Neither city hall nor I comprehended that the “community action” organizations on which we lavished taxpayer dollars would entrench dependency by urging people to get on the welfare rolls. War on Poverty funds paid for social workers, community activists, and lawyers to organize the poor, but these organizers, far from lifting poor people out of dependency, helped them sign up for more—and more expensive—welfare programs. For instance, the National Welfare Rights Organization urged single black mothers to protest the welfare system’s eligibility restrictions, and the organization’s goal was to flood the system with new clients.

The activists succeeded beyond their wildest imagination. By the end of the 1960s, during a period of economic prosperity and low unemployment, one out of every seven New Yorkers was on the dole. By 1975, War on Poverty spending (in inflation-adjusted dollars) had tripled, and the percentage of poor families’ income supplied by welfare had risen from 7.5 percent to 14.1 percent. Under the pressure of the advocates, government programs emphasized “welfare rights,” postponed self-sufficiency, supplied unproven and expensive services, and left most welfare clients out of the workforce. That’s perhaps the main reason that, as some pundits quipped, “in the War on Poverty, poverty won.” Yet my enthusiasm was undiminished; I had become a true believer. Along with my comrades on the left, I continued to think that income transfers were the most effective way to reduce the human pain of poverty.

What opened my eyes was a job I took in the mid-1970s at the Vera Institute in New York, led by the brilliant policy innovator Herbert Sturtz. Sturtz created small demonstration programs; researched them scientifically, sometimes with control and experimental groups; and, if they proved successful, asked governments or foundations to take them on and expand them. One of the demonstration programs, the Wildcat Services Corporation, was running employment programs for the most hard-to-place welfare recipients in all five boroughs, and Vera asked me to head Wildcat’s Manhattan and Bronx offices.

At Wildcat, we showed that the best way to get clients off welfare was to get them paid work immediately, rather than enroll them in training and education programs. I saw with my own eyes the value of work—any kind of paid work—in reducing welfare dependency and attacking poverty. I learned that if we helped welfare clients get jobs, even entry-level jobs, they would then attend to their other needs. By contrast, if the government gave them money and other benefits, they were likely to remain dependent.

The reasons should have been obvious all along. Work maximizes a person’s capacity to achieve economic self-reliance. Work socializes people and instills a sense of personal responsibility in them. Work connects behavior and consequences. And it permits people, especially men, to obtain the admiration and respect of their spouses and children by supporting them.

Another signature achievement of Vera’s, by the way, was its practice of conducting research on its own initiatives, something practically unheard-of at the time. Almost all War on Poverty programs shied away from serious empirical research, since they were mainly run by people with an ideological belief in the redemptive power of their own programs. Measuring the performance of their antipoverty efforts threatened that belief.

I became so convinced that the evidence supported the work-first approach that in 1984, I created a for-profit company called America Works to carry it out. My wife, Lee Bowes, eventually became the CEO and has been responsible for its success. The plan was to offer employment services to state and local welfare agencies with the aim of placing welfare recipients in jobs quickly, with a minimum of time spent on training.

Around the same time, Wisconsin was experimenting with work-first approaches to welfare reform. But the conventional wisdom was that there was no place for a private, for-profit venture in the antipoverty field. Most activists believed that helping welfare recipients was God’s work, that making profits was dirty, and that a private company would inevitably rip off the profits and reduce services to the poor. I got a boost, though, from Ohio governor Richard Celeste, who let me secure a contract from his state’s social-services department to open welfare-to-work programs in Cleveland and Dayton. With that contract in hand, I raised $1 million in start-up money, betting that a for-profit company could do the job better than government welfare agencies could and simultaneously bring accountability to a field that desperately needed it.

America Works staked its survival on the proposition that welfare clients, properly motivated and helped with a limited amount of technical assistance, could be successful at getting and holding jobs. Our typical contract with a welfare department stipulates that we don’t get paid our full fee until we place a client in a job and the client then completes a successful probationary period of four months. This arrangement motivates our trainers and employment specialists to perform well: they understand that if they are unsuccessful with job placements, America Works will fail, and they’ll be out of a job.

Our experience has confirmed that the main obstacles preventing welfare clients from finding and retaining jobs are a lack of connections and gaps in interpersonal skills. Extended education and training programs are unnecessary, time-consuming diversions; clients with shaky self-confidence are best served by early success in getting a job, not by long periods of preparation. Our weeklong training sessions are narrowly focused on the attributes and skills needed to land an entry-level job. Our trainers work with clients on the basics, such as maintaining a businesslike personal appearance, speaking properly, preparing a résumé, and showing up on time. Clients quickly learn that success depends on self-discipline and their own motivation and effort.

In the past 27 years, America Works has placed more than 250,000 poor people, with an average of five to six years on the rolls, in private-sector jobs, with an average starting wage of $10 per hour plus benefits. In our New York program, to take one example, more than half of these new workers were still on the job after 180 days. The employers that we have worked with include prestigious companies, such as Time Warner, Cablevision, Aramark, J. C. Penney, and American Building Maintenance Industries. Most of these employers keep coming back, asking for more of our referrals.

Single parents, drug and alcohol abusers, the mentally handicapped, the homeless, military veterans with posttraumatic stress disorder, and others with disabilities have succeeded admirably in a wide variety of jobs and have lifted themselves out of a lifetime of poverty. Take Ann Marie Lynch, a welfare mother of two who was evicted from her apartment and entered a homeless shelter in 2009. Luckily, she came to America Works, where we helped her compose a résumé, prepared her for interviews, and sent her to meet potential employers. After two months, she began a job as a home health aide; she even began working on an associate’s degree in business administration, which she received this May.

Despite our track record, we were never fully accepted by the welfare-industrial complex. In Ohio, county welfare offices and traditional social-services agencies flexed their political muscles and successfully lobbied the state to discontinue our contract after two years. In New York, during the mayoralty of David Dinkins, city agencies rejected our bids for welfare-to-work contracts four consecutive times. At one point, we were accused of “creaming”—selecting only the ablest clients, who would have landed jobs on their own anyway. That claim was disproved by a study conducted by NYU political scientist Lawrence Mead showing that our clients were representative of the general welfare population.

But we achieved a breakthrough with the election of Rudy Giuliani as New York’s mayor in 1993. Lee and I had first met him after he lost the previous election to Dinkins. Giuliani was initially skeptical that it was possible to move long-term welfare clients so quickly into employment and that there were many companies willing to hire them. But then he made a visit to our offices and became convinced. In his two terms as mayor, Giuliani reduced the welfare rolls by hundreds of thousands. He did this, in large part, by emphasizing work programs. At a celebration for the 25th anniversary of America Works two years ago, Giuliani remembered a Gotham that

was going into a second and third generation, in many cases, of people in families that hadn’t worked, that didn’t know the experience of work. It got me thinking that all the incentives in the welfare programs run by the city, the state, and the federal government were wrong. All the incentives were to put people on welfare. People put more people on welfare, they got bigger offices, they got bigger titles, they got more money, so in essence, failure in the lives of the people and failure in society became success for the welfare worker. I probably never would have thought of our own workfare program if I hadn’t gone to America Works.

The winds of change soon reached Washington. In his first State of the Union address, President Clinton promised to “end welfare as we know it.” His reform plan, he said, would transform welfare from a way of life into a “second chance” on the way to employment. The following year, he signed the Personal Responsibility and Work Opportunity Reconciliation Act, the biggest change in welfare programs since the New Deal. Welfare activists and many Democrats were aghast at the mandatory work requirements in the legislation, as well as its threat that people who didn’t move toward employment could be kicked off the rolls. Even Senator Daniel Patrick Moynihan, no bleeding-heart liberal, famously predicted that “children will be sleeping on the gratings” of sidewalks.

But such dire warnings were off the mark. Clinton’s welfare reform embraced the work-first ideal—at least in part—and the results were nothing short of remarkable. Welfare rolls plummeted from 12 million to 4.5 million nationally within one decade. In 2005, Congress reauthorized the reform by embracing work-first even more enthusiastically (see “Welfare Reform, Phase Two,” Winter 2009).

The success of welfare reform is widely acknowledged. Far less widely acknowledged is that the nation still has a tremendous work problem. In 2010, the latest year for which we have census data, a full 66 percent of poor people older than 15 did not work—a total of 21 million individuals. That isn’t just a consequence of the economic crisis: in 2007, the number was 16 million individuals, or 64 percent of poor people older than 15. Nor does it imply a situation in which one parent works while the other raises the kids, since in 2010, 35 percent of all poor families had no working members at all. Though welfare reform succeeded in getting many poor people into jobs, millions remain on welfare today, to say nothing of such programs as Medicaid, public housing, and food stamps. And welfare reform itself has come under attack recently, with the Obama administration issuing waivers that free states from complying with the law’s work requirements.

My experience with long-term welfare clients has led me to propose a radical solution: that we abolish all cash welfare, as well as food and housing assistance—except for the elderly and the physically and mentally disabled—in order to move from a dependency culture to one of work-first. This recommendation may sound impractical at a time of high unemployment. But the work-first principle can easily be implemented even in a down economy, as America Works proved by getting jobs for more than 500 ex-convicts in Detroit—a local economy with 14 percent unemployment—in the past two years. After all, despite the economic downturn, more than 3 million jobs per year go unfilled in the United States.

The federal government would use the huge savings from eliminating welfare to create or subsidize private-sector jobs, sending money to companies to reduce the cost of hiring and paying new workers. The government could also create programs similar to those run by the Depression-era Works Progress Administration, paying workers to build parks, refurbish bridges, clean streets, and so forth. The workers’ wages would pay for the basics—food, clothing, and shelter.

But once we dismantle cash welfare and other forms of aid and offer paying jobs in their place, what about the children of those few people who simply refuse to work? I think that we should seriously contemplate removing these unfortunate children from their irresponsible parents. Under current child-welfare laws, social-services agencies can already take kids away from their parents if their home environment is unsafe. Is it so extreme to extend that policy to homes ruined by willful poverty and neglect? I concede that the alternatives here are not pretty; government-regulated foster care, in particular, has its own risks of abuse. Adoption, however, works fairly well in most of the country. Another solution would be the establishment of government-funded institutions, operated by voluntary and religious nonprofits, to care for the children.

Jobs can’t replace all welfare and poverty programs. There will always be some people who are emotionally or physically unable to work and who require government assistance. But even the so-called deserving dependents should be more carefully scrutinized. In the last ten years, the number of newly enrolled recipients of SSDI, a federal benefits program that provides aid to disabled people who can’t achieve gainful employment, has risen 44 percent. That suggests that many people are abusing the system.

In public policy, we should deduce our theory from practice. Unfortunately, most people in the business of helping the poor turn that principle upside down, proposing theories first and then basing programs on them. Such people will surely oppose my proposals. Perhaps they should consider a concept from medicine—the word “iatrogenic,” which describes an adverse condition caused by the physician. Doctors take an oath not to contribute to iatrogenic conditions; they may not continue a treatment when that treatment itself causes harm. For the doctors of welfare policy, no such principle exists.

Can we finally confront the problem of entrenched poverty and dependency and make the difficult choices necessary to fix it? The cynic in me sees little chance that the public would seriously consider my proposals. But my optimistic side, the one shaped by finding jobs for supposedly unemployable welfare recipients, holds out hope that the nation’s fiscal crisis will allow serious consideration of policies that were once unthinkable. Stranger things have happened in America.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next