Mickey Kaus, author of the provocative book The End of Equality, has emerged as a leading voice in the debate over welfare reform. Kaus, a senior editor of The New Republic, proposes wide-ranging policy reforms aimed at encouraging social equality and rebuilding the work ethic in the American underclass. Michael Horowitz, a senior fellow at the Manhattan Institute and former general counsel in the federal Office of Management and Budget, is critical of many of Kaus’s proposals. The Manhattan Institute recently hosted a debate between Kaus and Horowitz.


What do we mean by “equality”? Liberals have often pursued equality of income, or what I call “money equality.” But I believe that what lies behind this pursuit is a goal that liberals and many conservatives could agree on: social equality. The best definition of social equality I have heard was offered by Ronald Reagan at the 1992 Republican Convention. “We are all equal in the eyes of God,” he said. “But as Americans that is not enough-we must be equal in the eyes of each other.” How can we achieve this goal?

In my book, I discuss three threats to social equality. The first is the growing inequality of income. I argue that it’s growing so fast that the trend cannot be reversed. The second threat is meritocracy. If “what you earn depends on what you learn,” then there’s a dispiriting possibility that those on top will think not just that they have more skills, but that they’re better than those on the bottom. The third threat is the destruction of the public sphere of life in which everybody participates as an equal. The classic example is public parks, once havens of social equality where everybody treated each other with respect and civility. A second example is the draft, which once-but no longer-combined Americans of all classes in a common endeavor.

How are we going to restore social equality in the face of these threats? One alternative is the traditional liberal approach of trying to equalize incomes. I argue in the book that this effort is futile, not only because the inequality trend is so strong, but also because you can’t buy much social equality by equalizing money anyway.

A far more productive strategy is what I call civic liberalism. There are three elements to this approach. First, rebuild the public sphere as much and as quickly as possible. Reinstitute the draft, start a national service program, create a national health system that mixes all the classes in a single system of care, take the money out of politics so that our political life becomes more egalitarian, have a system of day-care centers where the children of rich and poor use the same facilities, and restore some meaning to our public holidays so that they become more than three-day weekends.

The second element is to assimilate the underclass. The very existence of the underclass is a violation of social equality. People who are trapped in long-term poverty, resented and resentful, can’t possibly think of themselves as equals. And it’s the crime generated by the underclass that most obviously destroys the public sphere and sends middleclass people scurrying for the safety of the suburbs and the private sphere.

The third element of this strategy, once the underclass has been assimilated, is to integrate the social classes in neighborhoods and schools. As long as there is an underclass, integrated housing is impossible-nobody wants to live next to people in the welfare culture. But once the underclass has been assimilated, we will have a shot at getting working-class and upper-middle-class people to live close enough to each other to use many of the same schools and other public facilities.

How do we assimilate the underclass? In my book I attack what I call the “give them cash” consensus of the late 1970s. This was not merely a leftwing consensus; the notion originated with Milton Friedman. He argued that antipoverty programs spend a lot of money trying to help the indigent, but end up supporting bureaucrats and poverty pimps. Instead, he said, let’s be efficient: the one thing the government can do efficiently is send out checks, so let’s send the checks directly to the poor and get rid of the middleman. The left embraced this idea and produced a variety of guaranteed income plans. The problem, unfortunately, is that sending checks to poor people sustains the current culture of single-parent families dependent on welfare rather than work.

If welfare enables the culture of the underclass, then ending welfare can disable that culture. But we have to offer something better; it’s immoral, I think, to end welfare unless we can ensure that there is a decent job available for anyone who needs one. So I propose ending all cash assistance to able-bodied Americans, including single mothers, and replacing it with a program similar to the New Deal’s Works Progress Administration (WPA), guaranteeing a job to any American over 18. In 1935, when Franklin Roosevelt ended cash “relief” and initiated the WPA, he failed to apply this decision to single mothers. Aid to Families with Dependent Children (AFDC) is the one loophole Roosevelt left in his welfare state, and it has grown into a huge problem.

Ending welfare entails doing a lot of very expensive things. We have to provide day care for all single mothers who need it. We have to offer income supplements to all low-wage workers: You can’t have a WPA that pays more than McDonald’s, because everybody would leave McDonald’s and go to work for the WPA. The WPA therefore has to pay a subminimum wage, and income supplements are necessary so that somebody in a WPA job can live a minimally decent life.

The payoff for these expenses is that the underclass welfare culture would be unable to sustain itself. Today a girl growing up in the ghetto knows that if she has a child out of wedlock, welfare will sustain her. Under my proposal she would not have that assurance. If she had a child, she would have to provide for herself, juggling motherhood with a job-possibly a not-very-lucrative job. Given this choice, she would tend to think twice; she would more likely wait to get married or finish school and start working before having children.

In what is probably the first shot of the impending attack on Bill Clinton’s welfare plan, Christopher Jencks has argued that requiring welfare mothers to work does not make them richer; it just makes them busier. He is right, but that is the point: to force welfare mothers to confront the full consequences of their decision to have a child out of wedlock. The prospect of being busier will deter some women from making the disastrous decision to go on welfare in the first place. Remember that the goal of welfare reform is not simply to help those people currently on welfare. It’s also to keep people who aren’t on welfare from going on the rolls.

Such a plan of guaranteed employment would promote social equality by vindicating the value of work. The work ethic, as Lawrence Mead has pointed out, is one value that can unify all income classes. Work is one of the few tests that both rich and poor can pass. We need to create a society in which everybody who works can hold his head up and live with dignity. That’s why I think that all government cash benefits-not just welfare-should be work-tested.

A WPA program would be expensive. The people I’ve talked to estimate that it would cost about $50 billion a year to do everything right. That includes the cost of day care and income supplements. I think it’s well worth spending that money. The economic payoff would be immeasurable if you could take all the potential doctors and physicists who are now wasting their lives in the ghetto and bring them into the mainstream economy.


Mickey Kaus has written an extraordinarily brave and honest book. He understands the depredations that the welfare system has committed on the underclass. He not only talks about AFDC having caused an explosion in the numbers of people on welfare, but also criticizes Medicaid and Food Stamps for having caused millions of people to “go on the dole.” He identifies a decline of civility and shows that government programs are the primary cause, making this case far more convincingly than the Reagan administration ever could.

But what is his solution? He doesn’t want big government, which he says has caused these problems-he wants really big government. He says, in effect, give government another chance, a lot more money, and a lot more coercive authority, and my tough-love regime will work in ways all of the other programs haven’t.

My six years in government, however, showed me the ease with which government programs are captured by special interests who invariably sell out the poor. I agree that it would be well worth spending $50 billion on Mickey Kaus’s program-or twice that-if it would work. I’d even be inclined to do so if our only risk were alienating the suburban taxpayers Kaus wants to soak in order to pay for it. The most troubling aspect of his proposal, however, is the grave risk-which Kaus fails even to acknowledge-that it could make matters worse.

In my view, the Kaus plan is almost certain to become another entitlement. Decades down the road, the next generation of Mickey Kauses will talk about how the Kaus government-guaranteed job program created yet more indifference to the work ethic, yet more underclass dependence and hopelessness, yet more social division. And they won’t even be able to get rid of the program.

In The End of Equality, Kaus discusses the infamous CETA program, through which 650,000 welfare recipients were supposed to receive job training. The public-employee unions gutted the program, fearful that their members would lose their jobs. The Kaus program would likewise be undermined. It would turn three or four million welfare recipients into union members with civil service tenure in make-work jobs. Nor would Kaus’s notion of a subminimum wage rate last very long. Congressional hearings raising the specter of allegedly hard-working poor people employed by the government at “less than living wages” would have their predictable effects. Kaus can’t name many government employees who don’t make out well, and he won’t be able to give examples of federal entitlement programs whose costs haven’t gone up sharply over time. It’s no accident.

How about the idea of firing people who don’t show up for work? There will be hundreds of pages of regulations dealing with who can be fired, when, how, and under what conditions. Legal-aid lawyers will create endless litigation over job ratings, permissible work assignments, and so forth-and they will, as they’ve always done, beat the government agencies’ lawyers hands down. As Kaus himself noted, the Federal Government had 2.2 million employees in 1990, and only 403 were fired for incompetence. Can we expect judges to uphold the firing of WPA workers, particularly if it means they will be thrown out into the cold, welfareless world?

Liberal judges may, moreover, effectively invalidate work requirements through their construction of constitutional rights. Justice William Brennan has said that his favorite decision in all his years on the Supreme Court was Goldberg v. Kelly, a ruling that said welfare benefits could not be taken away without so-called due process hearings. Other decisions, some of constitutional dimensions, have made it difficult to dismiss government employees. Two Clinton appointments to the Supreme Court, and Mickey Kaus’s program could become precisely what he doesn’t want-a constitutional entitlement program which makes it prohibitively difficult to apply sanctions for refusal to work.

Even if the program survived the courts, there are other reasons it would not work. Mickey Kaus talks about the WPA as if it were a successful program. In fact, the term “boondoggle” came into being with the WPA. The unpopularity of makework jobs, leaf raking and the like, almost sank the New Deal. The New Deal programs that worked and were popular offered cash assistance-deliberately modest cash assistance-to people who were able-bodied but could not find work, or were not able-bodied and therefore could not work.

The great battle over social welfare programs in this country has been between the ideas and design of Roosevelt’s New Deal and those of Lyndon Johnson and his Great Society. The New Deal sought to provide a safety net for people without jobs so that they would not need to resort to private charity for their bare necessities. The grandiose Great Society had as its objective the abolition of poverty, not “merely” the amelioration of its immediate effects. Grand schemes with utopian objectives were hallmark Great Society approaches, precisely on the order of Mickey Kaus’s guaranteed jobs program.

One benefit of the New Deal design-and this should provide a profound caution to Mickey Kaus-was that its principal programs involved little more than check writing. There is a good reason why this is one of the few things government does efficiently. Such ministerial programs are inherently difficult for administrators to capture for their own purposes, so they have a lesser stake in expanding the programs. This virtue will, of course, not be present in a $50 billion regime of job supervisors, training areas, work orders, and high overhead costs. Those “payments for services” would mostly go to people like us, leaving precious little money trickling down to the poor.

The intent of the New Deal, moreover, was to provide assistance only to the “deserving poor”-those unable to work through no fault of their own. Thus, Social Security for the elderly, disability assistance for those who could not work, and unemployment insurance for those who had lost jobs were hallmark New Deal programs. The only major break in the pattern occurred on an unauthorized basis, and the story is revealing. Frances Perkins, perhaps the most liberal member of the Roosevelt Cabinet, broke sharply with Katherine Lenroot, the head of her Children’s Bureau, because when the Social Security Act was being written, Lenroot, unbeknownst to Perkins, had made unwed mothers eligible for AFDC assistance. Perkins had worked in homes for unwed mothers and would have restricted AFDC to widowed and divorced mothers. She was prepared to give no government money at all to impoverished mothers of illegitimate children, rather than create the irresponsibility and social pathology she believed would result from a federal support system.

How can we reform welfare? Not with bigger government, or with Frances Perkins’s draconian approach of none at all. Less government is the answer. We can scale back welfare payments to New Deal dimensions. We can also deregulate the bottom rung of the economic ladder, repealing minimum-wage laws that have priced underclass people out of the labor market; the Davis-Bacon Act, which requires union wages for government construction jobs and was originally passed in part to keep blacks out of the construction business; licensing laws that make it difficult to enter such high-tech fields as cosmetology and taxi driving; and rent-control laws, which discourage the development of low-income housing.

Another important element of reform is federalism. Instead of imposing a “one size fits all” solution, allow states to experiment, to compete with one another to see who does better. Resistance to federalism has been based on the notion that states are looking to push poor people away, and therefore the Federal Government has to act as the tribune of the poor. But it is precisely this tribune, as Kaus himself points out, that has been a prime source of our current problems.

Money isn’t the answer, and Mickey Kaus doesn’t need to feed the bureaucratic hand he otherwise fiercely bites in order to insist on the standards of behavior he knows are needed. In fact, we can flatly insist on good conduct without having to create massive programs to entice people to live up to appropriate rules. When I was in the government, I was visited by a housing official who had come to protest the Reagan cuts in the housing budget. I asked him, “Suppose we offered to double your budget, but we gave you an alternative: the right to summarily evict no more than 10 percent of your tenant population. Which would you choose?” He looked around theatrically, as if he wanted to make sure no one else was in the room, and said, “Are you kidding? Give us a chance to run our projects and to enforce our rules, and we’ll do wonders.”

Another time, I set up a meeting between Reagan and twenty ghetto high school principals who had turned their schools into places with long waiting lists, high college entrance numbers, and other sadly rare measures of success. One of them said, “You could have put a billion dollars into my school, and it would have been pouring it down a rat hole. But when I came into the school, I instituted dress codes and I suspended the bad kids-who, by the way, often came back and became my best students.” The other principals nodded their heads as he continued: “The reason I made it was that when I took the steps I did, neither my school board nor the ACLU had a clue of what I was doing. If they had, they’d have stopped me cold.”

Just as he thinks we need bigger government to cure our underclass problems, Kaus also thinks we need more powerful government to create the social integration he correctly says we need. He’d do it by forcing us all into the same medical waiting rooms and day-care centers, by reinstating the draft (irrespective of its effect on military efficiency), and by throwing monumental sums of fresh federal funds at city governments and school systems. It doesn’t occur to Kaus that curbing political machines and their union partners, another no-cost reform, would work wonders in making cities more livable and viable. It also doesn’t occur to him that even if we could afford to give current urban regimes more money, doing so could exacerbate our cities’ problems. In sum, Kaus has no faith that people will accept the richness of diversity unless forced to do so by federal bribes and coercion. My faith is different: I believe that people will wish to enjoy the lively interchange of communities like New York once we empower them to create incentives for responsible behavior.


Michael Horowitz has an idiosyncratic idea of the New Deal: that its genius lay in giving cash to poor people. In fact, his own story about Frances Perkins proves him wrong. She was furious that the AFDC program had been created, precisely because it gave assistance in cash. The WPA was not some sort of aberration in an otherwise brilliant New Deal scheme of dispensing cash to the poor. In his 1935 State of the Union address, Roosevelt called the cash dole “a narcotic, a subtle destroyer of the human spirit” and declared, “I am not willing that the vitality of our people be further sapped.” He called for ending the dole and replacing it with the WPA. The only exceptions were for people in the categorical aid programs, the most important of which is now AFDC for single mothers.

A couple of years after the WPA was created, Harry Hopkins, a Roosevelt aide, testified before Congress on the question of a work program versus direct cash relief. “It is my conviction,” Hopkins said, “and one of the strongest convictions I hold, that the Federal Government should never return to a direct relief program. It is degrading to the individual, it destroys morale and self-respect, it results in no increase in the wealth of the community, it tends to destroy the ability of individuals to perform useful work in the future, and it tends to establish a permanent body of dependents. We should do away with direct relief for the unemployed in the United States.” The New Deal was not a “give them cash” program.

Would the WPA work today? All the problems Michael Horowitz points out are real, but there are some countervailing tendencies that he ignores. It is true that all entitlements tend to grow, but taxpayers will object to paying a lot of money for a useless WPA. On the other hand, taxpayers do like receiving services, so they will demand that WPA not give people make-work jobs but jobs that actually produce an improvement in the quality of life. Much of the $50 billion cost of the program, incidentally, would go toward the check mailing Horowitz prefers, namely to provide a bigger earned income tax credit.

I agree that unions pose a serious threat to my proposal, but not an insurmountable one. The unions tried to destroy the original WPA, and in 1939 Roosevelt broke a strike over the “prevailing wage” issue. The presence of a Democratic president gives reason for optimism. Just as Nixon could go to China, Clinton could break a strike against a WPA program.

Would liberal judges prevent the imposition of work requirements? Probably not. Justice Brennan’s decision in Goldberg has been undermined by a series of Burger and Rehnquist Court decisions. Moreover, its reasoning is circular. It says that if one has the expectation of receiving welfare, the government can’t take away those benefits without a due-process hearing. But you only have expectations of a hearing if the government gives you those expectations. If we passed a WPA that did not create such expectations, but instead simply promised a job from which one could be fired, the courts would have no logical basis for saying there’s a legitimate expectation of elaborate hearings. Obviously there has to be some due process, but I think judges could be stopped from creating so many procedural protections that the program would be hobbled.

Horowitz’s position boils down to the proposition that because liberals have screwed up this sort of program before, it can never be made to work. This assumes that they haven’t learned any lessons from their previous failures. Horowitz’s knee-jerk pessimism would be more palatable if he had a better solution. But he doesn’t. For example, the idea that eliminating the minimum wage will solve the problem of the underclass ignores Charles Murray’s observation: The problem is not that people have no incentive to get off welfare, but that welfare enables the underclass. As long as we have welfare, many people will not take advantage of a lower minimum wage. People don’t show up now for $6-an-hour jobs; how can anyone think that welfare recipients would leave the underclass to take $3-an-hour jobs?

Ultimately, I do not quite understand Michael Horowitz’s position on welfare. Does he think single mothers who do not enter the labor market should continue to get cash welfare checks? First he praises Frances Perkins for wanting to “give no government money at all,” explaining that the brilliance of the New Deal was that it helped only the “deserving” poor. Then he calls the Perkins solution “draconian” and implies that nonworking welfare mothers-”undeserving poor”-should get checks. He only insists, apparently, that they not be very big checks. Welfare checks aren’t very big now-how much smaller should we make them? Would it transform the underclass if we cut welfare payments from $450 to $350, or from $600 to $500? I simply do not think this is the solution to the welfare problem.


It is not true that our current welfare system offers only modest benefits. If you add up the welfare check, the housing benefit, and the Food Stamp allowance, it comes to some $15,000 to $18,000 a year being paid to people who are not working. It is no accident that high-welfare states like New York have the biggest underclass. Lower benefits would do far more for the work ethic than Kaus supposes.

Kaus acknowledges that the courts or the unions could sabotage his program, but his discussion of why this would not happen is naïve. The fact that his workers would be told before they came on their jobs that they had no right to contest their dismissals would have little effect on the disposition of their due process claims. Expecting that activist judges would not severely curb the management discretion of bureaucrats seeking to fire underclass beneficiaries is as realistic as expecting President Clinton to break a strike of unions and their liberal allies against the use of unorganized, low-wage “scab” workers.

Kaus is wrong about the New Deal WPA and its present-day relevance. He quotes Roosevelt’s attack on cash doles, without acknowledging that Roosevelt laid equal attack on “the giving of cash, of market baskets, or of a few hours of weekly work-cutting grass, raking leaves, or picking up papers in the public parks.” WPA was only successful when it hired unemployed workers with high job skills and an intense desire to work, people in large supply during the Depression. Today’s problem is an underclass without work habits or work experience. We need a way to create a work ethic, not a system in which it is a prerequisite.

It’s never glamorous to propose “less” and always romantic to offer massive government initiatives-even to solve problems which, as Kaus points out, were largely caused by massive government initiatives. In the end, however, a combination of more-modest government and, as Myron Magnet makes clear, assaults on the prevailing culture represents the best hope for progress. Reduced welfare benefits and markets able to pay unskilled workers at wages approximating their value will achieve real reform. An understanding of how the poor have suffered from the cultural elite’s dismissive attitude towards self-discipline and personal accountability would also be of great value. In a like vein, this same elite should recognize that America is a blessed, opportunity-rich society, not a racist order that produces hapless victims. Such an acknowledgment would go far-much further than all the dollars Kaus would spend-to promote his very notion that a no-nonsense, no-excuse community response to underclass pathology is morally appropriate for caring people to advocate.

A renaissance of federalism could also free us from the need to have unitary solutions which have in the past proven impossible to modify once put in place. (Kaus is at his naive worst in believing that “taxpayer demands” will cause bureaucrats to administer his program efficiently and with proper regard for enforcing the work ethic.)

The tragic flaw of Mickey Kaus’s approach to welfare policy is his faith that more government will mean more, rather than less, commitment to the work ethic, which he rightly places at the center of the welfare debate.


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