Commentary Magazine

Helping the Poor: A Few Modest Proposals

Last fall I published a book entitled Losing Ground.1 It called attention to the fact that on several of the dimensions we ordinarily use to measure quality of life—unemployment, education, crime, family structure, economic dependence—things have gotten worse, not better, for the poor since the mid-1960's. I blamed the reforms of the 1960's for some large proportion of this decline in the fortunes of the poor.

Many readers, liberal and conservative, were prepared to accept Losing Ground's account of the problem itself. Some were even prepared to acknowledge that the reforms should be assigned a large share of the blame. But then I had gone on to argue that a federal system must of necessity have such effects. A federal system cannot do much to entice people into behaving in ways that society thinks will be better for them. It cannot even harmlessly relieve the problems created by the unwanted behaviors. All it can do is to stop subsidizing the behaviors it wants to prevent. What would the United States look like, I asked, if we eradicated the entire federal social-welfare system for the healthy working-age population? Better, I answered.

No one except libertarians could accept that conclusion. Liberals who otherwise thought Losing Ground had something to offer were exasperated. If social programs have been poorly designed, they said, don't throw them out; improve the designs. Conservatives who otherwise thought there should be cutbacks in social programs were dismayed. It was not serious to say we ought to scrap welfare—conservatives should be playing a constructive role in fine-tuning the system.

I sympathize with both of these reactions, but I submit that liberals and conservatives alike are indulging themselves when it comes to the problems of poverty. The nation is faced with some critical social pathologies that are not going to be solved or even much changed by a continuation of present policies. The moderate fixes and the fine-tuning will make us feel we are doing something. They will provide work for the program evaluators for a few more years. And each time we stop to take measure, it will be found that the problems are just about as bad as they were a few years earlier.

It is incorrect, however, to say that dismantling the federal system is the only alternative. Policies can be devised that plausibly will have large positive effects. But each of them entails a major departure from business as usual, and each has a dark side that must be accepted as part of the price of the improvement that might occur.

It is easier to think in terms of specifics than of the “poverty problem” as a whole. I therefore begin by considering perhaps the single most important element of the “poverty problem”—the case of poor single young women, often still in their teens, who have babies and keep them.


Babies Having Babies

Hardly anyone needs to be convinced that it is bad for poor single women to be having babies, and I will not labor the point. Let it be said very briefly that the reality is even worse than most people realize, most depressingly so for the children of single teenagers. Go to hospitals in low-income neighborhoods and talk to the pediatricians about the physical care of these children. Go to Head Start programs and ask the teachers about the emotional and intellectual development of the children of unmarried teenagers. Go to the police and social workers who visit these children in their homes, and ask about the nature of the environments in which they are growing up. It is a disabled population.

As the babies grow up, they not only continue to have problems; they also add to other people's problems. They tend to make schools hard to learn in and neighborhoods hard to live in. They have abnormally high incidences of mental and physical handicaps and abnormally bad employment records. They commit a disproportionate share of crimes and are disproportionately addicted to drugs. In all these ways, they are threats not just to themselves, not even just to other people, but to the web of bonds and social institutions that enables communities to function.

Twenty years ago, in 1965, the nation absorbed about 155,000 new adolescents (reaching their thirteenth birthday) who were children of single mothers.2 Ten years ago, in 1975, the nation absorbed about 240,000. This year, 1985, the nation will absorb approximately 394,000. In 1995 the nation will have to absorb about 700,000. If these numbers do not seem worrisome enough, try adding some cumulative totals. Right now, in 1985, the population contains roughly 2.5 million teenagers who were born into single-parent homes. In 1995, it will contain about 4.1 million. In 1985, they represent about 10 percent of the teenage population. In 1995, they will represent about 17 percent of the teenage population.

If these children (and their mothers) were scattered randomly throughout the nation, absorbing them would not be such a daunting task. But they are concentrated in the poorest, most vulnerable communities, and especially in poor black communities. In central Harlem, the most recent figures from New York's Department of Health reveal, 79.9 percent of all births are to single women.3 That would indicate that almost a third of all live births in central Harlem are to single teenagers. There is no reason to think that the figures for Harlem are substantially different from those in other inner cities. Overall, as of the most recent (1982) figures, 56.7 percent of all live births to blacks were to single women. If we assume, as collateral evidence suggests we should, that the illegitimacy rate for the black middle class is much closer to the white rate (12.1 percent in 1982), then the illegitimacy rate for poor blacks nationwide is perforce near the 80-percent figure reported for Harlem. Nor is there any reason to think that the problem will get better on its own. The proportion of live births to single women has continued to increase among whites. Among blacks, the increase has slowed (but not stopped) as it approaches what must be its asymptote.

The current state of affairs is producing enormous misery and suffering that no increases in food stamps or Medicaid or payments for Aid to Families with Dependent Children (AFDC) or educational programs are going to cure. It follows that something should be done to reduce the number of children being born to and kept by poor single women. But what?


Are Solutions Possible?

The first question to ask is whether, in fact, the nation is prepared to take steps to reduce the births. There is reason to think that the answer is no. One of the peculiarities of the way the illegitimacy problem has been discussed is the inevitability that people seem to attach to it. It is treated as an ineluctable physical law: poor single women will get pregnant and keep their babies, no matter what. The explanation goes like this: poor single women do not “decide” to have babies as a rational process. Peer pressures and social norms of poor communities are at work. The future of a young poor girl is so bleak that a baby represents the only accomplishment she can hope for. A baby is someone she can love, and someone who will love her. The man who fathers the child is kept from taking responsibility for it by his own hopeless situation. Poverty and despair are ultimately to blame. In sum, it is impossible to reduce births to single women without changing the nature of these larger forces.

Five years into the Reagan era, the “larger-forces” view is still nearly a consensus. It has, to my way of thinking, two flaws. In principle it is objectionable. It implies that poor people are less human than not-poor people. We, the not-poor, make choices about such matters and expect our own teenage sons and daughters to do so as well. But we exempt poor people, above all black poor people, from the same responsibility. In addition to being thus morally objectionable, the “larger-forces” explanation also runs up empirically against inconsistencies in the American experience (in the past, desperately poor immigrants and ex-slaves did not have 80-percent illegitimacy rates). And why should the larger forces of poverty tend elsewhere in the world to produce extremely low birth rates among single women?

Before anything else can be done to deal with the problem, this “larger-forces” consensus has to be changed. It is not a question of denying that environment affects behavior. Of course it does. But people still make choices. Large numbers of poor men impregnate women and choose not to take responsibility for the resulting child. Knowing their men may well behave irresponsibly, large numbers of poor women who cannot afford to take care of a child on their own nonetheless choose not to practice birth control, choose not to have abortions, and choose not to put the child up for adoption. How might these men and women be led to make other choices?

In presenting possible strategies, I omit any that are plainly unconstitutional (e.g., requiring licenses to have babies). I also omit discussion of strategies such as sex-education programs that have already been widely tried, without success. Apart from those restrictions, anything I can think of or which has been at least semi-seriously suggested to me is part of the roster.


The Choice Among Carrots

The first, rather small, set of available strategies consists of positive inducements—carrots rather than sticks. Here the solution devised must be one that does not punish people. The rationale is that, even though the “larger forces” can be overridden, they still explain why the mother and father of the child behaved as they did, and it is not appropriate to blame them. They too are victims, not morally culpable.


1. Child allowances: The “carrot” approach with the most widespread current support seems to be the child allowance, sometimes known also as the family allowance. Child allowances appeal to the Left by extending the principles of entitlement and a guaranteed income; they appeal to the Right as a means of encouraging the formation of two-parent families. Thus Robert Kuttner and George Gilder in their recent writings agree on at least something.

Proposals vary, but in most versions the child allowance is a monthly payment to all families with children. Why would this reduce births to poor single women? The logic is that child allowances at least reduce the current incentives not to get married. As things stand now, there are many concrete economic reasons (albeit valid only in the short run) why a woman who is both pregnant and poor would choose to remain unmarried even given a choice. With child allowances, the short-run tradeoff between being married and unmarried would balance out.

I raise the child-allowance option because it is the only major policy change being seriously discussed. But it is doubtful that child allowances would do much to reduce the number of children born into single-parent households (I will not argue the other merits and drawbacks of child allowances). One reason is that the size of any feasible allowance is going to be too small to be much of an inducement—in most Western countries that use child allowances, the average payment is only about $500 per child per year.4 But if the size of the allowance is increased to a point where it becomes competitive with the money now provided under the AFDC program—let alone the Medicaid, food-stamp, and housing components of a welfare package—it quickly becomes unaffordable. In 1982, the United States had 31 million families with a child under eighteen. The average payment per AFDC family that year was about $3,700 per year. Even after adjusting for the differences in the size of AFDC families and other families with a child under eighteen, the cost of child allowances would be prohibitive, probably on the order of $60 to $80 billion more than state and federal governments now spend on AFDC.5

Let us assume for purposes of argument, however, that methods can be devised to provide a significant allowance at an affordable overall cost—for example, by limiting the child allowance to persons below a certain income. The salient question then is: which women would behave differently?

The answer seems to be those single women who, being pregnant, would like to be married and have a willing male partner, but are dissuaded from marriage because of the welfare alternative. There is no empirical basis for estimating how many of the 715,000 women who bore illegitimate children in 1982 might fall into this category. The number would seem to be small. Anecdotal accounts do not give the impression that large numbers of unwed fathers are pining for the responsibilities of parenthood. Even if the man is willing, the woman does not necessarily gain from marriage. If his employment record is bad, she might well be better off avoiding marriage. The child allowance unquestionably has a definite advantage over the current system: the woman does not lose the child allowance if she marries, whereas now she does lose her AFDC benefits. But the child allowance does no more than remove an obstacle.

A more important objection is this: if the payment is large enough to replace AFDC, and if it is made available as an entitlement rather than as a welfare program subject to regulations, restrictions, and general hassle, then it is likely to be more attractive to a poor single woman than even the current system. In trying to project the ultimate effects on illegitimate births, I am reminded of the the attempts of the late 60's to provide AFDC mothers with incentives to get a job, and that turned out instead to provide a greater incentive for single mothers with jobs to get on AFDC. Child allowances could easily backfire.


2. Bribes: The most outrageous and yet plausibly the most effective of options is to pay single women not to have children. Beginning at the age of, say, fifteen, give each single female in the country a birthday present ($1,000, perhaps) if she has not had a baby during the preceding year.

Outrageous, but also difficult to discredit. A thousand dollars in cash at one time is a lot of money to a teenager, especially a poor one. It is a highly tangible incentive that does not require the recipient to think very far ahead (throughout the school year, a girl would be observing her friends receiving their checks and spending them conspicuously). The annual cost of paying $1,000 a year to each single woman aged fifteen to thirty-four not to have babies is affordable. In 1982, the cost would have been less than $18 billion.6 If it reduced the number of illegitimate children by even 20 or 30 percent, it could probably be shown to be a highly cost-effective investment. Once the right size of the bribe is established, the reductions might be much larger.

The scheme even has the virtues of being testable (it lends itself to a one- or two-year pilot version in a small city) and administratively feasible. To advocate that such a test be conducted, one simply has to repress the thought that the solution is tantamount to paying young males not to commit armed robbery or giving a rebate to citizens who can prove they have not cheated on their tax returns.


The Choice Among Sticks

An alternative way of approaching solutions is to assume that having babies is infused with moral content. To have a baby and keep it, knowing in advance that one cannot provide for it, is irresponsible, wrong, to be condemned. Although society should take steps to protect the welfare of the child, such steps not only may penalize the mother and father but in some sense should.


1. Residential facilities: In the 19th century a lively debate took place over the merits of payments to poor persons living in their own homes versus various forms of the residential facility—almshouses and work houses. The debate has disappeared almost altogether in the 20th century. We have come to assume that herding people together into a communal-living facility is uncivilized.

Well, is it? First, we must disentangle the idea of residential facilities from our reflexive assumptions about what they must necessarily be like. Granted, Dickensian horror stories about almshouses plus more recent ones about prisons and mental institutions are relevant. But there are “good” prisons and mental institutions—clean, well-run, humane—and, for that matter, there were good almshouses.

What would happen if there were no provision for AFDC, food stamps, Medicaid, or housing subsidies, but any mother could, if she chose, live with her child in a residential facility? The facility would provide the standard of living and regulation of a good correctional “halfway house” (except that a mother would be free to quit the facility at any time) plus, let us assume, a generously-funded and professionally-staffed program for the children—daylong Head Start, as it were.

It is reasonable to predict at least a few good outcomes. For many of the children, physical well-being would increase dramatically—cleaner and safer housing; less vulnerability to abuse, drugs, and crime; better nutrition and accelerated intellectual development. There would also be numerous bad effects, even assuming that the facilities were properly run. Even under ideal circumstances, institutions are not the best places for children to grow up in. But if one is primarily concerned about the net well-being of a large population of children, one must balance the disadvantages of a residential system against the disadvantages of the environments where the children live now.

The system presumably would have a strong deterrent effect as well. Few adults, especially young ones, want to subject themselves to life in a halfway house. When a young mother, once pregnant, is faced with the choice of raising a child without any outside support whatsoever or living in a residential facility, adoption (and abortion) can reasonably be expected to rise. Obviously, too, some mothers without adequate resources would probably refuse either to give up the child or to live in the facility, thereby putting the child in jeopardy. The laws protecting children from neglect (through foster care if necessary) would have to be strictly enforced.

It is a chilling prospect. But it would probably work, in a way. It might make us, the sponsors, feel bad, but it would also mean fewer malnourished children, fewer neglected children, fewer children coming into kindergarten already lost causes, than we tolerate now. Given such a balance, what would constitute kindness?


2. Age discrimination: A less extreme variation of the residential solution, but also much less effective, would be to keep the present social programs intact except for one change: prohibit any kind of direct welfare assistance to people below an age limit—twenty-one, for example. The AFDC money, food stamps, and any other benefits would go to a parent or guardian. For teenagers who use the welfare system as a means of setting up their own household, the deterrent effect might be significant. Prospective grandparents appalled at the prospect of having to keep a baby in the house could also be expected to exert their influence on behalf of adoption or abortion. Whether the total effect would be large or small is impossible to assess.

This strategy, like child allowances, has considerable potential for backfiring. Some prospective grandparents will think that the idea of a new baby in the house is attractive and even economically advantageous. Probably the only way to make age discrimination in welfare truly effective is simply to prohibit altogether any welfare assistance for a child born to a woman under the age limit—which in effect is a limited version of the dismantling proposal for which I am trying to find alternatives.


3. Making the Father Take Responsibility: Even now, the fathers of children born to single mothers theoretically are responsible for the support of the child. But the existing child-support laws have been frustrated by mothers who do not try to get support, by poor enforcement of the court's decrees when support has been mandated, and, most importantly, by the difficulty of getting money from a man who has no income. But if the goal is primarily to deter births, none of these problems is insurmountable.

Suppose that as a condition of qualifying for existing welfare benefits, the mother had to identify the father of the child (not dissimilar to some current regulations). The father would then be required to provide a sum of money equal to the entire package of benefits the mother would receive. This action would be taken at the initiative of the government; it would not require the support or even consent of the mother. If the father were unable to pay, he would be put in jail. He would have the option of marrying the mother instead, in which case he would subsequently be subject to the existing abandonment laws, strictly enforced.

It would not work in practice, because the justice system would never behave that way—could not so behave even if it wanted to, given the strains on its correctional facilities. But as with the other alternatives, we are asking only, “what if?” And if it were truly the case that a man who fathered a child either had to support it or go to jail, no exceptions, it is reasonable to expect substantial changes in male behavior prior to the birth of the child.


4. Making the Mother Get a Job: Much the same logic applies to a workfare program for mothers. For example, we may envision a system in which a single mother with a baby could get the existing welfare support if and only if she were working a forty-hour week at a job in the private sector or, if one could not be found, at a job created by the government. The deterrent effect might well be large in the ideal case, especially among teenagers. If a woman knew that having a child and keeping it automatically meant that she would have to work full-time or forfeit any welfare support, then she might decide to delay having a baby. But the experience of CETA, the work-incentive program (WIN), and a variety of workfare programs indicates that such plans are very difficult to carry out. The effectiveness of the program depends critically on the credibility of the threat, and even if the necessary laws were on the books, there are too many ways in which that credibility could not be maintained.


5. Stigma: The stigma attached to welfare has been one of the principal topics of liberal discourse about welfare for many years. The draft of the Catholic bishops' “Pastoral Letter on Social Teaching and the U.S. Economy” is only the most recent reprise of the 1960's theme that accepting welfare should not cause feelings of shame or guilt. If it is not a person's fault that he requires help (the economy is to blame, environmental factors are to blame, uncontrollable circumstances are to blame), then to stigmatize the recipient is unfair and cruel. But if it is the mother's fault that she has the baby, and stigma does have powerful effects, why not try to use stigma as a remedy?

This strategy accepts the argument (commonly made by the same people who have wanted to rid welfare of its stigma) that the epidemic of illegitimate children among teenagers, far from being the product of the welfare system, is caused by a mysterious process whereby in the inner city it has become a cultural norm for poor teenage girls to have babies. Because it is part of the culture, changes in the welfare system will have no impact. If indeed that is the case, then clearly it is necessary to change the norm.

Under this proposal, the current social programs would remain intact. Eligibility rules would remain intact. But the federal government would commission advertising campaigns from the best advertising firms whose mission would be to make unmarried pregnancy disreputable; to ruin the image of the single woman with a baby. I am not thinking of a sympathetic campaign of public-service announcements conveying the facts about life on welfare and what it does to one's future. Rather, the notion would be to put the impressive selling talents of Madison Avenue behind making unwed parenthood contemptible, unfashionable, dowdy, stupid—whatever would work.

For adults, the stigma would be institutionalized by taking away the right to vote from anyone who had no source of income except welfare—a suggestion that has been put forward by William F. Buckley, Jr. Disenfranchisement would be an official stamp of second-class status: people who live off the largesse of the state should not have a role in determining the rules for dispensing it. Stigma would be reinforced by the way that welfare is administered. Instead of trying to reduce the rudeness of welfare workers and eliminating embarrassing eligibility investigations (as was the objective, only partially achieved, in past reforms), such practices would be encouraged.

Based on what is known about the growth of the AFDC rolls in the 1960's, a major side-effect of a successful campaign to increase stigma would be to discourage some eligible women from applying for welfare benefits. This would be bad insofar as they would be left destitute, good insofar as it could lead some women who otherwise would have become trapped on AFDC to make a living on their own. Given the fact that it is extremely difficult for a single, uneducated, poor woman with a baby to make a decent living, the balance would probably be negative.

Further, a successful campaign would increase public opprobrium toward single women already on welfare, a result which seems patently unfair. For the last twenty years, the federal government has studiously avoided making moral judgments about the phenomenon of illegitimacy. Indeed, it has used government funds for publicity campaigns to convince women that they should not be reluctant to accept welfare assistance. To do a sudden about-face and acknowledge that having a baby one cannot care for is contemptible would be a contemptible thing in itself. It would be particularly so in view of the widespread hypocrisy among policy-makers that characterized the earlier posturing on the subject.


The Dismantling Option Reconsidered

However frivolous some of these proposals may be, they were not concocted in an effort to be outrageous. Rather, they represent just about the only strategies available. Anyone who doubts this is invited to try to think up some other, less extreme, measures that will plausibly have a major effect.

This is why in Losing Ground I asked the reader to imagine dismantling the entire system. It is beyond dispute that the human and social consequences of the current state of affairs constitute a major tragedy. Business as usual cannot be expected to produce much improvement. I could think of nothing better than getting rid of the whole system. Everywhere in the world, including this country in the past, a “natural state” produces very few children born into single-parent families. Why this should be so is no mystery. It is an entirely understandable consequence of human beings recognizing that babies must be cared for and responding to that imperative by arranging matters so that care is available. Usually this takes the form of extreme social pressure on the male to marry the female (legally, or in common law). When this does not work, the child usually remains in an extended family and is cared for by relatives. In a small proportion of cases, the child becomes an orphan.

This natural system has been far from perfect. The main source of breakdown has been the lack of human control (short of abstinence) over whether a child will be conceived. But cheap and effective contraception means this no longer needs to be an issue. Add to contraception the availability of adoption and abortion, and there is very little reason why, in the United States of the late 20th century, a baby has to be born into a family that has inadequate means to care for it. Take away all governmentally-sponsored subsidies for irresponsible behavior, and the natural system will produce the historically natural results. The residual problem, infinitesimal compared to the present one, can safely be left to the equally natural responses of relatives helping relatives, friends helping friends, and communities protecting communities. Such is the logic of the “no-system” option. I am no more satisfied with it now than I was then, if only because it is politically impossible. But this survey of the alternatives has failed to persuade me that any of them is any better.

Persuading others of that conclusion, however, is not the central point of the discussion. The central point, or pair of points, is that (1) the present state of affairs is not acceptable, and (2) there are no painless ways to rectify it. To that, the standard response goes something like this: “This is a rich country. Let's stop worrying about the morals of the thing and do whatever is necessary to give everyone a decent living. We can afford it.”


Washing Our Hands of It

I therefore no longer confine the discussion to poor single mothers; they are just one more type of poor person, and in this scenario the nation has decided to eradicate poverty altogether.

It is certainly within our power to do so, in a technical sense. In 1982, the federal government spent roughly $55 billion7 on aid programs for the working-age needy. This figure excludes state expenditures. It excludes federal expenditures on unemployment insurance, veterans' programs, workers' compensation, and educational programs. The $55-billion figure includes just the programs that most people think of when they think of “welfare”—food stamps, AFDC, child nutrition, Medicaid, public housing, and the like. Add in the amounts spent on the elderly poor through Social Security and Medicare, and the nation is already spending enough to eradicate poverty.

To borrow a familiar way of illustrating the size of the existing system: if the federal government took all the money budgeted for the poor and used it for cash grants, it could, without increasing total expenditures, bring every poor person in the country above the official poverty line. The assertion is statistically shady for a number of reasons (for one thing, it assumes that the earned income of all poor people would be unaffected by the availability of the supplemental grant), but the underlying point is valid. Lack of money is not really the barrier to solving the poverty problem.

The cleanest of the systems for eradicating poverty is the negative income tax as advocated in its pure form by Milton Friedman. Let us suppose that it has been adopted. A floor has been set under the income of everyone in the country, and those whose income is beneath that floor receive a cash grant that brings them up to the mark. To avoid straying from the point 1 wish to make, let us also assume away a few practical difficulties. The system has managed to solve the problem of defining the right level for the floor. The sum is that amount sufficient to purchase, as Adam Smith put it, “whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without.” The floor has been adjusted to take into account differences in family size, inflation, size of community, and geographic location. Finally, the system is designed so that it is instantly responsive to changes in employment and family status—that is, the negative income tax is paid on a monthly basis, and each month's check to people below the income floor is adequate to assure them of the basic income for the coming month.

The result is that no one in the entire United States, of whatever age, sex, race, family situation, or education need live in poverty.

The first month's checks go out. By the end of the first month, it is apparent that suffering goes on. For most of the people in one large population, those formerly poor people with expensive drug and alcohol habits, the additional cash they receive under the new system is absorbed by their habit. Some have spouses and children. All are as deprived as before.

For a second group of poor people, that first month's check is immediately used up)—sometimes foolishly, sometimes for things that are perhaps reasonable but should not be purchased all at once, sometimes for emergency needs. Whatever the reason, the result is that before the end of the month, there is no money for food. Or for rent. Or for the bus fare to get to the job. And a secondary result is hungry children. Or eviction. Or a lost job. Many in this population fall deeply into debt and soon will have signed away their income for the foreseeable future.


The upshot is that, within some months after the new system has gone into effect, a substantial number of poor people will be observed to be in the same shape they were in under the old system, even though not one single person in that population “needs” to be experiencing these problems.

What is to be done? If food stamps or subsidized housing or any of the rest of the in-kind services are reestablished, the system begins sliding down a slippery slope. As these supplemental services are made available, the number of people who need them rapidly increases. Consider, for example, the situation of someone whose income is precisely at the minimum. If he spends the money so that it lasts for a month, he gets no food stamps (or any other of the supplemental services). If he spends the money so that it runs out, he will become eligible for food stamps. If the food-stamp allotment is $50, the only reasonable course of action is to reach the end of the month $50 dollars short of food money. It soon becomes apparent that the main function of the food-stamp allotment is to increase the effective minimum income, not to relieve hunger. But there are still truly hungry people (because some people will be coming up $100 short).

Up to this point I have been speculating on what would happen to those who were already poor at the time the guaranteed-income system was installed. Now, let us introduce what was learned from the negative-income-tax experiment conducted by the federal government in the late 1960's and early 1970's.8 The work disincentives of the system were unmistakable and quite large. They were especially large among wives (20-percent reduction in desired hours of work in the Denver and Seattle sites) and among males who came of working age during the course of the experiment (33-percent reduction among those who became heads of household, 43-percent reduction among those who did not).

We must therefore expect that under a negative income tax, many formerly self-reliant people will to some extent become dependent on the government payment. Of those who become dependent, it must be assumed that some number will also join the ranks of those who cannot manage to provide themselves with a decent existence. These persons must be added to the accounting of gains and losses that a negative income tax would yield.


There is a way out of this bind that makes a negative income tax a desirable alternative after all. It requires only that, once the system is installed, the government say. “That's it. If you run out of food money, try the Salvation Army. If you get evicted because you can't pay the rent, get a friend to take you in.” The next month, somewhat fewer people run out of food money and somewhat fewer forget to set aside the rent money. Over the long run, some sort of equilibrium is reached in which, I am willing to believe, the net physical wants of the bottom income level of society are smaller than they were before. In that sense, the negative income tax represents an improvement.

But certain other kinds of suffering that we associate with the “poverty problem” are little changed:

Family life? The chronically unemployed man who beat his wife as a way of working out his anger at being a failure is still a failure, still knows it, and still beats his wife. For that matter, such problems might well get worse, for a large new population of men who formerly had a function in life—they made a living for their families, even if a poor one—will have lost their role.

Drug addiction? It is more likely to increase rather than decrease, for analogous reasons.

Crime? Perhaps the youngster who robbed liquor stores so he could buy food for his sick mother now desists, but, from everything we know, most youngsters rob liquor stores to satisfy other needs that would not be met by a minimum family income. Street crime continues largely unaffected.

Ignorance and illiteracy? The only effect of the new system is to diminish still further whatever pressure to learn that a youngster had formerly felt. A living is going to be there, regardless of whether he can read or calculate.

Deprived children? The children who are now malnourished because the parent has no resources to buy an adequate diet will be better fed. The ones who are malnourished because the parent is too ignorant or indifferent to feed a small child properly will continue to be malnourished. The children who have birth defects because the mother abused herself, and her child, during pregnancy will still have birth defects. The children who come to kindergarten never having seen a book will still not have seen a book.

The nation will have eradicated any need to be poor. It is not at all clear to me that, after all the accounting is done, we will have reduced the net misery by one iota.


Does this pessimistic view square with what we know from experience elsewhere? What about Sweden and Denmark? Finding out more about what we really “know” is one of the urgent research tasks for social planners, but let me offer a preliminary hypothesis. The lessons from Sweden and Denmark tell us nothing about what would happen in Appalachia and the South Bronx. They do, however, tell us what eventually happens, under the most ideal circumstances, to an ethnically homogeneous population of a few million people with several hundred years of Lutheran socialization behind them—they tell us what will happen to the populations of Minnesota and Iowa a few generations down the road, let us say—and the news is increasingly bad.

If we are truly going to do something about poverty in this country—remembering that “poverty” is shorthand for a set of problems, not a condition in itself—then it will happen only after the nation reaches a consensus on what it is trying to accomplish. If it wants to wash its hands of those problems, then by all means let us install a guaranteed income and be done with it. The problems won't be our fault anymore. If, however, we want to do good; if we want to arrange society so that more people are living lives that provide them with satisfaction, achievement, and a measure of happiness no matter what their income level; if we want fewer hungry children and more loving families—then our course is tougher. Ultimately, we must come to terms with the limits of government in promoting these ends and we must begin taking a much larger measure of personal responsibility for ministering to the needs that remain.


1 See the review by Brigitte Berger, COMMENTARY, January 1985.—ED.

2 The source for all the data on births is Vital Statistics of the United States (National Center for Health Statistics), various editions. The data for 1982 are taken from Monthly Vital Statistics Report, vol. 33, no. 6, September 28, 1984.

3 Cited in Ann Hulbert, “Children as Parents,” New Republic, September 10, 1984, p. 18.

4 Robert Kuttner, The Economic Illusion (Houghton Mifflin, 1984), p. 244.

5 The number of families with a child under eighteen is taken from U.S. Bureau of the Census, Statistical Abstract of the United States: 1984 (104th edition), 1984, table 60, p. 47; the average AFDC payment is computed from the same source, table 651, p. 393; the total current costs of AFDC are estimated from table 608, p. 371.

6 Statistical Abstract of the United States: 1984, estimated from population figures in table 33, p. 33 and table 51, p. 44.

7 Social Security Bulletin, December 1984, vol. 47, no. 12, table 2, p. 16.

8 See Losing Ground, chapter 11, for a discussion of the negative-income-tax experiment.

About the Author

Charles Murray is the W.H. Brady scholar at the American Enterprise Institute and the author most recently of In Our Hands: A Plan to Replace the Welfare State (2006). This article has been adapted from a presentation at the annual Herzliya Conference in Israel in January.

Pin It on Pinterest

Welcome to Commentary Magazine.
We hope you enjoy your visit.
As a visitor to our site, you are allowed 8 free articles this month.
This is your first of 8 free articles.

If you are already a digital subscriber, log in here »

Print subscriber? For free access to the website and iPad, register here »

To subscribe, click here to see our subscription offers »

Please note this is an advertisement skip this ad
Clearly, you have a passion for ideas.
Subscribe today for unlimited digital access to the publication that shapes the minds of the people who shape our world.
Get for just
Welcome to Commentary Magazine.
We hope you enjoy your visit.
As a visitor, you are allowed 8 free articles.
This is your first article.
You have read of 8 free articles this month.
for full access to
Digital subscriber?
Print subscriber? Get free access »
Call to subscribe: 1-800-829-6270
You can also subscribe
on your computer at
Don't have a log in?
Enter you email address and password below. A confirmation email will be sent to the email address that you provide.