(The following appeared in the Congressional Record on December 12, 1995.)

We are now in the final days of the first session of the 104th Congress. In a short while we will have worked out some accommodations on the budget. We must do this, for we will now be engaged in the establishment of some measure of peace and lawful conduct in the Balkans. It would be unforgivable if we put our military in harm’s way abroad without first getting our affairs in some minimal order here at home.

I am fearful, however, that as we close out this session we will also close down the provision for aid to dependent children that dates back sixty years to the Social Security Act of 1935. If this should happen, and it very likely will, the first and foremost reason will be the monstrous political deception embodied in the term “Welfare reform.”

In my lifetime there has been no such Orwellian inversion of truth in the course of a domestic debate. “Welfare reform” in fact means welfare repeal. The repeal, that is, of Title IV-A of the Social Security Act. Everyone is to blame for this duplicity, everyone is an accomplice.

For practical purposes, we can begin with the celebrated “Contract with America,” which pledged that within one hundred days, a “Republican House” would vote on ten bills, including

  1. Welfare reform. The government should encourage people to work, not to have children out of wedlock.

This in itself was unexceptional, especially the second clause. By 1994, the nation had become alarmed by an unprecedented rise in illegitimacy, to ratios altogether ahistorical—from practically nil to almost one third in the course of a half-century. Since illegitimate children commonly end up supported by Aid to Families with Dependent Children (AFDC), a causal connection was inferred. Not proven. We know desperately little about this great transformation, except that it is happening in all the industrial nations of the North Atlantic.

Undeterred, the new House majority promptly passed a bill which repealed AFDC. Such an act would have been unthinkable a year earlier, just as repealing Old Age pensions or Unemployment Compensation, other titles of the Social Security Act, would be today. At a minimum, it would have seemed cruel to children. But the new Republicans succeeded in entirely reversing the terms of the debate. Instead of aiding children, AFDC was said to harm them. Last month, a Republican member of the House remarked on the importance of child care:

Because our welfare reform package is going to remove people from welfare and get them to work. We understand that child care is a critical step to ending the cruelty of welfare dependency.

What once was seen as charity, or even social insurance, is redefined as cruelty.

This happens. Social problems are continuously redefined. Malcolm Gladwell of The Washington Post has noted that,

In the nineteenth century, the assumption had almost always been that a man without a job was either lazy or immoral. But following the depression of the 1890s, the Progressives “discovered” unemployment.

Which is to say, a personal failing became a societal failing instead. This redefinition has wrought what would once have seemed miracles in the stabilization of our economy. Mass unemployment is now history. On the other hand, such cannot be said for the attempt to dissociate welfare dependency from personal attributes, including moral conduct. As we would say in the old Navy, I am something of a plank owner in this regard.

It is just thirty years since I and my associates on the policy planning staff of the Department of Labor picked up the onset of family instability in the nation, in this case among African Americans. Interestingly, this followed our having failed to establish that macroeconomic problems were the source of the trouble. In the event, I was promptly accused of “Blaming the Victim.” For the thirty years that followed there was an awful tyranny of guilt mongering and accusation that all but strangled liberal debate. One consequence was that when a political force appeared that wished to change the terms of debate altogether, “established” opinion was effortlessly silenced and displaced. Again, Gladwell:

But if anything is obvious from the current budget fight and Capitol Hill’s commitment to scaling back welfare and Medicaid while lavishing extra billions on the Pentagon, it is that this once formidable confidence has now almost entirely slipped away. This is what has given Washington’s current re-examination of the size and shape of government its strange ambivalence. In most revolutions the defenders of the status quo have to be dragged from power, kicking and screaming. In this revolution, the defenders of the old activism toward the poor surrendered willingly, with the shrugs and indifference of those who no longer believed in what they stood for either.

This was painfully evident in the Senate. On August 3, 1995, the Republican majority introduced a “Welfare reform” bill which abolished AFDC. That same day, the Democratic minority introduced a competing “Welfare reform” bill—which also abolished AFDC. On the minority side an enormous fuss is now being made over adding a little extra child care, some odd bits of child nutrition aid, perhaps a little foster care: literally arranging flowers on the coffin of the provision for children in the Social Security Act. Coming from devious persons this would have been a conscious strategy—distracting attention from what was really going on. But these were not, are not, devious persons. Sixty years of program liberalism—a bill for you, a bill for me—have made this legislative behavior seem normal. The enormity of the event has been altogether missed.


I hope this is not mere innocence on my part. The Washington Post editorial page has been unblinking on this subject. An editorial of September 14 described the bill on the Senate floor as “reckless,” adding with a measure of disdain: “Some new money for child care may…be sprinkled onto this confection.” Those seeking to define welfare repeal as welfare reform by “improving” the Republican measure should have known better; but I truly think they did not. In recent years, “child care” has been something of a mantra among liberal “advocates” for the poor. For all its merits, it has awesome defects, which are the defects of American social policy. The most important is that it creates two classes of working mothers: one that gets free government-provided child care; another that does not.

The Clinton administration arrived in Washington sparkling with such enthusiasms. At this time, I was chairman of the Committee on Finance, charged with producing $500 billion in deficit reduction, half through tax increases, half through program cuts. I thought deficit reduction a matter of the first priority, as did my fabled counterpart in the House, Dan Rostenkowski, chairman of Ways and Means. In the end, we got the votes. Barely. Fifty, plus the Vice-President’s in the Senate. But all the while we were taking on this large—and as we can now say hugely successful—effort, we were constantly besieged by administration officials wanting us to add money for this social program or that social program. Immunization was a favorite. Rosty and I were baffled. Our cities had had free immunization for the better part of a century. All children are vaccinated by the time they enter school. If they aren’t vaccinated at earlier ages, it is surely the negligence or ignorance of the parents that can best explain it. But nothing would do: we had to add whatever billion dollars for yet a new government service.

My favorite in this miscellany was something called “family preservation,” yet another categorical aid program (there were a dozen in place already) which amounted to a dollop of social services and a press release for some subcommittee chairman. The program was to cost $930 million over five years, starting at $60 million in Fiscal Year 1994. For three decades I had been watching families come apart in our society; now I was being told by seemingly everyone on the new team that one more program would do the trick. The new Family Preservation program was included in the President’s first budget, but welfare reform was not. In fact, the Administration presented no welfare plan until June of 1994, a year and a half after the President took office. At the risk of indiscretion, let me include in the record at this point a letter I wrote Dr. Laura D’Andrea Tyson, then the distinguished chairman of the Council of Economic Advisors, regarding the Family Preservation program:

July 28, 1993

Dear Dr. Tyson:

You will recall that last Thursday when you so kindly joined us at a meeting of the Democratic Policy Committee you and I discussed the President’s family preservation proposal. You indicated how much he supports the measure. I assured you I, too, support it, but went on to ask what evidence was there that it would have any effect. You assured me there was such data. Just for fun, I asked for two citations.

The next day we received a fax from Sharon Glied of your staff with a number of citations and a paper, “Evaluating the Results,” that appears to have been written by Frank Farrow of the Center for the Study of Social Policy here in Washington and Harold Richman at the Chapin Hall Center at the University of Chicago. The paper is quite direct: “…solid proof that family preservation services can effect a state’s overall placement rates is still lacking.”

Just yesterday, the same Chapin Hall Center released an “Evaluation of the Illinois Family First Placement Prevention Program: Final Report.” This was a largescale study of the Illinois Family First initiative authorized by the Illinois Family Preservation Act of 1987. It was “designed to test effects of this program on out-of-home placement of children and other outcomes, such as subsequent child maltreatment.” Data on case and service characteristics were provided by Family First caseworkers on approximately 4,500 cases; approximately 1,600 families participated in the randomized experiment. The findings are clear enough. “Overall, the Family First placement prevention program results in a slight increase in placement rates (when data from all experimental sites are combined). This effect disappears once case and site variations are taken into account.”

In other words, there are either negative effects or no effects.

This is nothing new. Here is Peter Rossi’s conclusion in his 1992 paper, “Assessing Family Preservation Programs.” Evaluations conducted to date “do not form a sufficient basis upon which to firmly decide whether family preservation programs are either effective or not.”

May I say to you that there is nothing the least surprising in either of these findings? From the mid-’60s on this has been the repeated, I almost want to say consistent pattern of evaluation studies. Either few effects or negative effects. Thus, the negative income tax experiments of the 1970s appeared to produce an increase in family break-up.

This pattern of “counterintuitive” findings first appeared in the ’60s. Greeley and Rossi, some of my work, Coleman’s. To this day I can’t decide whether we are dealing here with an artifact of methodology or a much larger and more intractable fact of social programs. In any event, by 1978 we had Rossi’s Iron Law. To wit: “If there is any empirical law that is emerging from the past decade of widespread evaluation research activities, it is that the expected value for any measured effect of a social program is zero.”

I write you at such length for what I believe to be an important purpose. In the last six months I have been repeatedly impressed by the number of members of the Clinton Administration who have assured me with great vigor that something or other is known in an area of social policy which, to the best of my understanding, is not known at all. This seems to me perilous. It is quite possible to live with uncertainty; with the possibility, even the likelihood that one is wrong. But beware of certainty where none exists. Ideological certainty easily degenerates into an insistence upon ignorance.

The great strength of political conservatives at this time (and for a generation) is that they are open to the thought that matters are complex. Liberals have got into a reflexive pattern of denying this. I had hoped twelve years in the wilderness might have changed this; it may be it has only reinforced it. If this is so, the current revival of liberalism will be brief and inconsequential.

Daniel Patrick Moynihan

Note that concluding paragraph: If we don’t get as good at asking questions as conservatives have become, “the current revival of liberalism will be brief and inconsequential.”


In the course of the recent debate on “Welfare reform,” specifically on September 14, I took occasion to note that almost the only serious critique of the Republican proposal, and its Democratic variant, was coming from conservative social analysts and social scientists. Let me cite three such criticisms which in sum, or so I would argue, make a devastating case against what Congress and the administration seem bent on doing.

First George Will, who in the high tradition of conservative thought, asks us to consider the unanticipated consequences of what we are about to do to children in the course of disciplining their parents. He wrote in September:

As the welfare reform debate begins to boil, the place to begin is with an elemental fact: No child in America asked to be here. No child is going to be spiritually improved by being collateral damage in a bombardment of severities targeted at adults who may or may not deserve more severe treatment from the welfare system.

Let me attach numbers to this statement. In 1968, as part of the social science undertakings associated with the Economic Opportunity Act of 1965, the Federal government helped establish the Panel Study of Income Dynamics at the Survey Research Center of the University of Michigan. The thought was to follow cohorts of real, named individuals over the years to see how income rose and fell over time. Earlier this year, using this data, Greg J. Duncan and Wei-Jun J. Yeung calculated that of children born between 1973 and 1975, some 24 percent received AFDC at some point before turning eighteen. Among African Americans this proportion was 66 percent, while for whites it was 19 percent. All told some 39 percent of this cohort received AFDC, Food Stamps, or Supplementary Security Income.*

And so we know what we are talking about. A quarter of our children.

A year ago November, James Q. Wilson gave the Wriston lecture at the Manhattan Institute, entitled “From Welfare Reform to Character Development.” He began by insisting on how little we know:

Let me confess at the outset that I do not know what ought to be done and assert that I do not think anyone else knows either. But I think that we can find out, at least to the degree that feeble human reason is capable of understanding some of the most profound features of the human condition. What we may find out, of course, is that we have created a society that can no longer sustain a strong family life no matter what steps we take. I am not convinced of that, for the very people who express the deepest pessimism are themselves leading, in most cases, decent lives amid strong human attachments and competent and caring families.

What we worry about is the underclass. There has always been an underclass and always will be one. But of late its ranks have grown, and its members have acquired greater power to destroy their own children and inflict harm beyond their own ranks. The means for doing so—guns, drugs, and automobiles—were supplied to them by our inventive and prosperous economy. We must either control more rigorously those means or alter more powerfully the lives of those who possess them. I wish to discuss the latter, because the public is rightly dubious about how great a gain in public safety can be achieved by the legal methods at our disposal and is properly indignant about the harm to innocent children that will result from neglecting the processes by which underclass reproduces itself.

The great debate is whether, how, and at what cost we can change lives—if not the lives of this generation then those of the next.

He then set forth three precepts. Note that the first is precisely where Will began.

First precept: Our overriding goal ought to be to save the children. Other goals—reducing the cost of welfare, discouraging illegitimacy, and preventing long-term welfare dependency—are all worthy. But they should be secondary to the goal of improving the life prospects of the next generation.

Second precept: Nobody knows how to achieve this goal on a large scale. The debate that has begun about welfare reform is largely based on untested assumptions, ideological posturing, and perverse priorities. We are told that worker training and job placement will reduce the welfare rolls, but we know that worker training and job placement have so far had at best very modest effects on welfare rolls. And few advocates of worker training tell us what happens to children whose mothers are induced or compelled to work, other than to assure us that somebody will supply day care. We are told by others that a mandatory work requirement, whether or not it leads to more mothers working, will end the cycle of dependency. We don’t know that it will. Moreover, it is fathers whose behavior we most want to change, and nobody has explained how cutting off welfare to mothers will make biological fathers act like real fathers. We are told that ending AFDC will reduce illegitimacy, but that is, at best, an informed guess. Some people produced many illegitimate children long before welfare existed, and others in similar circumstances now produce none, even though welfare has become quite generous. I have pointed out that group homes and boarding schools once provided decent lives for the children of stable, working-class parents who faced unexpected adversity, but I do not know whether such institutions will work for the children of underclass parents enmeshed in a cycle of dependency and despair.

Third precept: The federal government cannot have a meaningful family policy for the nation, and it ought not to try. Not only does it not know and cannot learn from “experts” what to do; whatever it thinks it ought to do, it will try to do in the worst possible way: uniformly, systematically, politically, and ignorantly. Today official Washington rarely bothers even to give lip service to the tattered principle of states’ rights. Even when it allows the states some freedom, it does so only at its own pleasure, reserving the right to set terms, issue waivers, and attach conditions. Welfare politics in Washington is driven by national advocacy groups that often derive their energy from the ideological message on which they rely to attract money and supporters. And Washington will find ways either to deny public money to churches (even though they are more deeply engaged in human redemption than any state department of social welfare) or to enshroud those churches that do get public money with constraints that vitiate the essential mission of a church.

Finally, to address Wilson’s point that any welfare program significantly funded from Washington will be run “uniformly, systematically, politically, and ignorantly”: I don’t disagree. The Family Support Act of 1988 had two basic premises. The first was that welfare could not be a way of life; that it had to be an interlude in which mothers learn self-sufficiency and fathers learn child support, and also that this goal was to be pursued in as many different ways as state and local governments could contrive. I would like to think that I am not the only person still in Washington who recalls that in debate we would continually refer to the experiments being carried out by a liberal Democratic governor in Massachusetts, Michael Dukakis, and a conservative Republican governor of California, George Deukmejian. Our expectations, very much under control I should say, were based on the careful research of such programs by the Manpower Demonstration Research Corporation based in New York.

On December 3, Douglas J. Besharov of the American Enterprise Institute, the third of the conservative analysts I will cite, wrote in support of the welfare measure now in conference, stating that the experience of the JOBS program under the Family Support Act showed just how innovative and responsible states can be. He said:

Since 1992, the federal government has allowed states almost total freedom to reshape their welfare systems through the waiver process. According to the Center for Law and Social Policy (CLASP), as of last week, 42 states had requested waivers and well over half had already been granted.

As some will know, earlier this year I introduced the Family Support Act of 1995, seeking to update the earlier legislation, given seven years’ experience. In the current issue of The National Journal, in which I am referred to as the “champion” of “left-of-center advocacy groups,” this measure, which got forty-one votes on the Senate floor, is simply dismissed: “…Moynihan’s bill is principally a vehicle for defending the status quo….” Dreadful charge, but not unwarranted. The status quo is meant to be one of experiment and change. And it is. I so state: the idea of changing welfare has even taken hold in New York City!

Now to what I think of as a constitutional question, the source of my greatest concern.

I have several times now related an event which took place in the course of a “retreat” which the Finance Committee held last March 18 at the Wye Plantation on Maryland’s Eastern Shore. Our chairman, Senator Packwood, asked me to lead a discussion of welfare legislation, the House bill, H.R.4, having by then come over to the Senate where it was referred to our committee. I went through the House bill, and called particular attention to the provision denying AFDC benefits to families headed by an unwed female under eighteen years of age. I said that these were precisely the families we had been most concerned about in the Family Support Act. The welfare population is roughly bi-modal. About half the families are headed by mature women who for one reason or another find themselves alone with children and without income. AFDC is income insurance, just as unemployment compensation is income insurance. Or, if you like, social insurance, which is why we call it Social Security. These persons are typically in and out of the system within two years. The other AFDC families, rather more than half, begin as AFDC families. Young women with children typically born out of wedlock. These are the families the Family Support Act was concerned with. There are millions of families in just this circumstance.

A few days later, a colleague on the Finance Committee came up to say that he had checked on this matter at home. In his state there were four such families; two had just moved in from out of state. I can imagine the state welfare commissioner asking if the Senator wanted to know their names!

Here is the point as I see it. Welfare dependency is huge, but it is also concentrated. That portion of the caseload that is on welfare for two years or less is more or less evenly distributed across the land. But three quarters of children who are on AFDC at a point in time will be on for more than five years. They are concentrated in cities. In Atlanta, 59 percent of all children received AFDC benefits in the course of the year 1993; in Cleveland, 66 percent; in Miami, 55 percent; in Oakland, 51 percent; in Newark, 66 percent; in Philadelphia, 57 percent.

By contrast there are many states that do not have large cities and do not have such concentrations. The Department of Health and Human Services has estimated the number of children who would be denied benefits under the sixty-month time limit contained in both the House and Senate welfare bills, now in conference. For California, 849,300. For neighboring Nevada, 8,134. For New York, 300,527. For neighboring Vermont, 6,563.

If welfare were a smallish problem—if this were 1955, or even 1965—an argument could be made for turning the matter back to state government. But it is now so large a problem that governments of the states in which it is most concentrated simply will not be able to handle it. On December 3, Lawrence Mead had an excellent article in The Washington Post in which he described the recent innovations in welfare policy, all provided under the Family Support Act, in Wisconsin. His article is entitled: “Growing a Smaller Welfare State: Wisconsin’s Reforms Show That to Cut the Rolls, You Need More Bureaucrats.” It begins:

The politicians debating welfare reform would have us believe that their efforts will greatly streamline the current system, help balance the nation’s books and reverse the growing tide of unwed pregnancy among the poor. What they aren’t telling us is that, at the state and local level, the federal cuts in the offing are apt to increase—not shrink—the size of the welfare bureaucracy.

Mead’s point is one we understood perfectly at the time we enacted the Family Support Act. The cheapest thing to do with chronic welfare dependent families is simply to leave them as they are. Changing them in ways that Wilson speaks of is labor intensive, costly, and problematic. A nice quality of the Wisconsin experiments is that job search begins the day an adult applies for welfare. But this takes supervision. Mead notes that high performing areas of the state “feature relentless follow-up of clients to see that they stay on track.” The term “client” is important; it is a term of professional social work. This sort of thing is not for amateurs. Most importantly, he concludes:

Even with Wisconsin’s successes so far, important questions remain unanswered: What happens to the people who were formerly on the welfare rolls? Are they better or worse off than before? Can they sustain themselves long term? Anecdotes don’t suggest great hardship, but nobody knows for sure. And what evidence is there that this approach can flourish in inner cities where the social problems are far more serious? In Milwaukee, which has half the state’s welfare caseload, the success has been far more modest than in the rest of the state.

These questions need answers before a case can be made that Wisconsin is the model on which other states should base their reforms. But this much is clear; Wisconsin’s fusion of generosity and stringency does represent what the voters say they’re looking for.

In Milwaukee, 53 percent of children are on AFDC in the course of a year.

I have been taken to task for suggesting that the time limits in the House and Senate bills will produce a surge in the number of homeless children such that the current problem of “the homeless” will seem inconsequential. So be it; that is my view. I believe our present social welfare system is all but overwhelmed. Witness the death of Elisa Izquierdo in Brooklyn. If 39 percent of all children in New York City were on AFDC at some point in 1993, I would estimate that the proportion for Brooklyn would have been at least 50 percent, probably higher. Hundreds of thousands—I said hundreds of thousands—of these children live in households that are held together primarily by the fact of welfare assistance. Take that away and the children are blown to the winds. A December 6th administration analysis concludes that the welfare conference agreement will force 1.5 million children into poverty. To say what I have said before here in the Senate: young males can be horrid to themselves, horrid to one another, horrid to the rest of us.

By way of example, or analog, or what you will, I have frequently referred to the Federal legislation that commenced the “deinstitutionalization” of mental patients. I was present at the creation of this movement. Early in 1955, our former esteemed House colleague Jonathan B. Bingham, at that time Secretary to Governor Averell Harriman of New York, brought Dr. Paul Hoch, the new Commissioner of Mental Health, in to meet the Governor. I was present, along with Paul H. Appleby, the new Budget Director. Dr. Hoch, a wonderful, humane man of science, told of a new chemical treatment for mental illness which had been developed by Dr. Nathan Kline at Rockland State Hospital in the lower Hudson Valley. It had been tested clinically. Hoch proposed that it be given to all patients, throughout the New York mental hospital system, which then held some 94,000 patients. (Today there are 8,000). Harriman asked what the program would cost. Hoch mentioned a sum in the neighborhood, as I recall, of $1.5 million. Harriman asked Appleby if he could find the money. Appleby, I cannot doubt having been cued by Bingham, replied that he could find it. Done! said Harriman. I am an investment banker and believe in investment. And so reserpine medication commenced.

Eight years later, on October 31, 1963, in his last public bill-signing ceremony at the White House, President John F. Kennedy signed the Mental Retardation Facilities and Community Health Center Construction Act of 1963. I was present, since I had worked on the legislation, and the President gave me a pen which I have in my hand here. We were going to empty out our great mental hospitals and treat patients in local community centers. We would build two thousand by the year 1980, and thereafter one for each additional 100,000 persons in the population. Alas, we built some six hundred centers or so, and then just forgot about our earlier plans. But we emptied out the hospitals. A decade or so later, the problem of the homeless appeared, to our general bafflement. I have earlier commented that in New York, with our singular ability for getting problems wrong, homelessness has been defined as a problem of lack of affordable housing. We will very likely think up some equally misleading explanation for the growing numbers of homeless children when they appear, and so I would like to put this on the record now.

On December 3, a newspaper of considerable circulation did just this, however inadvertently. A long article on “welfare reform” was accompanied by a photograph of an overstuffed chair on which a broken or battered doll had been placed. The caption read: “Republicans blame failed welfare policies for today’s problems. Above, an easy chair at a Philadelphia homeless encampment.” A photograph, I dread to say, of things to come.

Republicans must look to their own consciences. I would appeal to that of my own party. A few weeks ago our distinguished Majority Leader, Senator Bob Dole, stated that he hoped to bring “welfare reform” to the floor within a week.

It is very likely next week there will also be a conference report on welfare reform. I think we have about concluded the conference. [T]he original bill passed in the Senate by a vote of 87 to 12. We believe we have retained most of the Senate provisions in the conference, and I ask my colleagues on both sides—this bill had strong bipartisan support—to take a close look.

Eighty-eight percent of the American people want welfare reform. We will have it on the floor, we hope, next week. We hope the President of the United States will sign it. In my view, it is a good resolution of differences between the House and the Senate. We still have one or two minor—well not minor—issues in disagreement we hope to resolve tomorrow, and then we hope to bring it up by midweek next week.

What is one to say? The Senate bill did indeed have “strong bipartisan support.” If we do get a conference committee report, it will pass and will, I am confident, be voted. What I fear is that the repeal of the Social Security Act provision will return as part of a general budget reconciliation, and that bill will be signed into law. Should it do so, the Democratic Party will be to blame, and blamed it will be. It will never again be able to speak with any credibility to the central social issue of our age.

We will have fashioned our own coffin. There will be no flowers.

December 14, 1995

This Issue

January 11, 1996