Welfare-bashing redux: the return of Charles Murray.

In 1984, a little-known social theorist named Charles Murray published a book entitled Losing Ground: American Social Policy, 1950--80. Murray argued that our policy solutions to poverty were actually exacerbating social problems. Above all, he blamed the welfare system for the increase in out-of-wedlock births and families headed by single mothers, which in turn led to increased poverty, violence, crime, illiteracy, and a host of other social ills. It was a provocative argument, and one that gained very sympathetic attention from the Reagan administration. It was also unsupported by the facts.

The book unleashed a vigorous response by social scientists who objected not necessarily to Murray's arguments but to his misuse and abuse of data. Like many theorists, Murray is on much safer ground when he's broadly speculating about the causes of social ills (he happens to be a fine writer) rather than when he's trying to measure them. The critics of Losing Ground pointed out that social problems are complex constructs with multivaried causes and solutions. To attribute a major demographic shift--that is, the long-term increase in families headed by single mothers--to a single factor like welfare programs is egregious reductionism.

In spite of the criticism, Murray's early work gave the Reagan welfare-bashers the "scientific" authority to do their worst. They sharply curtailed access to the welfare system by changing the laws regarding eligibility and setting much tougher standards. Yet, rather than reducing out-of-wedlock births, these measures had the opposite effect: poverty deepened, homelessness increased, and cities became more impoverished. Murray seemed to be discredited and disappeared from the scene.

Unfortunately, he's back. In addition to his controversial new book, The Bell Curve (written with Richard Hernnstein), Murray has a bully pulpit at the corporate-backed American Enterprise Institute, where he has renewed his anti-welfare diatribe with a vengeance. That's not too surprising; with "welfare reform" once again on the national agenda, the far right has chosen Murray to represent them. What's surprising--and alarming--is how seriously he's being taken.

Murray's current arguments are the same as they were a decade ago: the welfare system, instead of relieving poverty, is causing it by creating incentives to form "nontraditional" families. Since the most valuable benefits package (including the value of Aid to Families with Dependent Children {AFDC}, food stamps, and the relevant portion of Medicaid) is available to families headed by a single mother, there exists, according to Murray, a positive incentive to form such families--either by delaying or avoiding marriage, dissolving existing unions, or having children out of wedlock.

The evidence does show that the proportion of families with children headed by females has grown over time, as has the benefit package; however, the correlation between benefits and family structure holds up only through the early 1970s. Beginning around 1972, the benefit sum began to fall, which was primarily due to the declining real value of AFDC benefits. Yet none of the demographic indicators of single parenthood followed the plunging benefit package. If growing benefits led to increased formation of single-headed families, then falling benefits should have led to a decrease in the formation of these family types. The independence of these demographic indicators and the benefit sum challenges Murray's reductionist theory that welfare causes the rising proportion of female-headed families.

To correctly analyze the impact of benefit trends on family formation, one needs to examine the value of the welfare package relative to other economic opportunities, primarily earnings. To the extent that welfare offers greater financial resources than marriage, an economic incentive to form welfare-eligible families is strengthened. But to argue that this is a cause for the increase in single-parent families, the focus would need to shift from welfare benefits to male and female wage and employment trends--an analysis Murray never undertakes.

Another challenge to Murray's hypothesis comes from the independence of state-level welfare payments and state-level demographic trends. Large differences in state benefit levels offer a natural test of the Murray hypothesis. If he's correct, states with high benefit levels should have high unmarried birth rates and high proportions of single-parent families (and vice versa). Once again, the data reveal no such pattern. For example, the benefit package is 58 percent higher in New Jersey than in Mississippi, yet their rates of single parenthood are the same (about 50 percent of black children and 14 percent of white children live in single-parent families in both states). In fact, Mississippi and Alabama have the lowest AFDC benefits levels in the nation, yet they rank among the highest in teen birth rates.

More recently, it has been shown that the share of single mothers with some college education (who are less likely to be poor) has risen while the share with less than a high-school education has declined. Since the probability of seeking welfare decreases as the mother's education level rises, this pattern flies in the face of "the Murray connection." Similarly, if welfare payments were driving single motherhood trends, we would expect to find a higher proportion of single mothers in Western industrialized countries with more generous welfare systems than our own. Once again, this is not the case, as single motherhood is more prevalent here.

Despite the wealth of evidence, Murray continues to attract an audience, and it is worth thinking aobut why. First, his connection between welfare and single motherhood is an intuitive, albeit simplistic, one. It is easy to understand and has a widespread emotional appeal; were it not for data that show the opposite, Murray's stories might be plausible. Second, and more importantly, there is no question that the rate of out-of-wedlock childbearing and the consequent increase in single parenthood are serious societal challenges. The increased share of families headed by single parents, coupled with the fact that such families have relatively high probabilities of poverty, constitutes a national problem. Murray offers an intuitively satisfying, yet dangerously misguided, solution: that is, to end welfare.

Ending welfare is a very dangerous prescription indeed, and one that even conservative analysts agree would lead to intense deprivation and hardship for single mothers and their children. What is called for instead is a more complex paradigm within which to understand these trends. Maybe nothing will make Charles Murray go away, but if understanding is an antidote to obfuscation, then a genuine attempt to explain the increase in single parenthood may at least reduce his influence.

There is no denying that the share of unmarried births as a proportion of total births has grown over time, particularly for blacks. However, the main reason for this trend results from the fact that a smaller proportion of women of childbearing age are married than in the past, and those that are married are having fewer children than before.

Both table 1 and figure 1 present the relevant data using the concept of "expected lifetime births," or the number of in-and out-of-wedlock children a woman could expect to have throughout her lifetime. For example, based on prevailing fertility rates at the time, a married white woman in 1960 would have been expected to have an average of 3.45 children over her lifetime.

[CHART OMITTED]

The trends depicted in figure 1 show that, while blacks have a higher level of expected unmarried births than whites, both groups show a slight upward trend (though blacks show a steeper rise at the end of the period). While a rise in the number of unmarried births among blacks--1.05 in 1960 to 1.64 in 1989--is not a trivial development, an increase of this magnitude would have been unlikely to generate the controversy surrounding this issue. The significant trend is the steep drop in the expected births to unmarried women of both races, as the average married white woman in 1960 went from having 3.45 children in her lifetime to having 1.53 children in 1989. The expected child-bearing behavior of her black counterpart changed even more dramatically, with her expected births falling from 3.49 to 0.90 over the past 30 years. The ensuing rise in the ratio of unmarried to all births (as shown in the last panel of table 1) has been driven to a greater degree by the falling birth rates of married women than by the rising birth rates of unmarried women.

TABLE 1 Expected Lifetime Births by Marital Status and Race, 1960-1989

The numbers in the following table refer to the expected number of lifetime births per woman of child-bearing age given age-weighted fertility rates in each year.

Race   Marital Status      1960   1965   1970   1975   1980   1989
White   Unmarried          0.08   0.11   0.14   0.12   0.18   0.36
Black   Unmarried          1.05   1.08   1.16   1.09   1.25   1.64
White    Married           3.45   2.67   2.25   1.56   1.57   1.53
Black    Married           3.49   2.75   1.93   1.15   1.01   0.90
White  Married
& Unmarried                3.53   2.78   2.39   1.69   1.75   1.89
Black  Married
& Unmarried                4.54   3.83   3.10   2.24   2.27   2.54
Unmarried births as a percent of total births
White                      2.3%   4.0%   5.7%   7.3%  10.2%  19.0%
Black                     23.2%  28.2%  37.6%  48.8%  55.5%  64.5%

Source: State of Working America, 1994-1995 (forthcoming)

Thus, a complete understanding of the problem demands that we explain the decreased propensity of married women to have children, since that is driving the increase in the share of all births that are unmarried. Unfortunately, this question has yet to be answered, but we do know some interesting corollaries. First, we know that these trends are society-wide, with marital birth rates falling for women throughout the income spectrum. Second, despite the fact that poor women still have higher out-of-wedlock birth rates, the rate for more affluent women is rising rapidly. For all women, rich and poor, not being married has become less of a constraint against child birth. Murray's argument that this broad societal trend is a result of welfare benefits that are actually declining is simply not credible.

An unfortunate fallout of the Murray debate is that everyone is so busy arguing about the role of welfare that little attention is paid to the issue of why we should be so concerned about single motherhood. The most detailed work on this question has recently been completed by sociologist Sara McLanahan, who presented her findings in a recent issue of American Prospect (#18). She finds that children who grow up in singleparent households "are disadvantaged across a broad array of outcomes," including low levels of education, teen parenthood (for women), and labor-force nonparticipation. Interestingly, these findings hold even when controlling for socioeconomic status, once again reinforcing the problem of Murray's narrow scope. Single parenthood and its consequences are not the problems solely of the poor and disenfranchised. This does not negate the fact that single-parent families often lack the economic resources needed to thrive. In fact, McLanahan shows that the loss of income to divorced families accounts for half of the disadvantages associated with single parenthood.

Where Do We Go from Here?

Despite Murray's efforts, several facts appear through the smoke:

* Unmarried birth rates and the share of families headed by a single parent have been rising for a long time now, and their growth is an important sociodemographic phenomenon.

* The claim that the availability of welfare benefits has caused these trends is not supported by the data. Trends over time and across states (and even countries) also challenge this reductionist explanation.

* The sharp growth in the share of all births to unmarried women has been driven much more by the decrease in married births than by the increase in unmarried births.

* Rates of single motherhood increased sharply over the last decade among highly educated women, while the proportion of single mothers with less than a high-school degree declined. Single motherhood, while still most common among poor women, is becoming more common throughout the socioeconomic spectrum.

* Because Murray is wrong about the causes, his draconian solution--ending welfare--won't help. Nevertheless, he is correct to be alarmed about this demographic trend. Children who grow up with one parent tend to have significant disadvantages, whatever their socioeconomic status.

At present, it's fair to say that social science does not know how to counteract the increase of single-parent families. It seems obvious that at least three strong social forces are at work here: the greater economic independence of women generated by their increased labor-force participation; the declining economic prospects of males, particularly those without a college education; and a shift in social mores such that out-of-wedlock childbearing is less stigmatized. Women's increasing economic independence is an important gain, and if that independence has enabled women to leave bad marriages or choose marriage partners more carefully, so much the better. Reversing falling male wages (and those at the low end of the earnings scale have lost the most) should be a national priority and would likely reduce single parenthood. Such policy fixes are beyond the scope of this piece, but raising the minimum wage and strengthening labor law (thus bolstering unions) would help restore wage growth to those who have lost the most. Social mores, we can probably agree, are beyond the scope of policy. Certainly cancelling "Murphy Brown" won't help.

Perhaps the most effective public policy to reduce the increase in out-of-wedlock births by young, poor women would be to ensure that other avenues of opportunity are open to them. If young women perceive clear access to higher education and thereby to decent (not dead-end) jobs, they may be less likely to undertake childbirth prior to establishing a partnership. In addition, if young men are able to earn a wage that makes them suitable marriage partners (at least from an economic standpoint), marriage will be a more attractive option to women contemplating single parenthood.

In the meantime, we can fight the fire at our backs by trying to make sure that unmarried mothers have the resources they need at their disposal. This means strengthening our weak child-support system so that absent fathers are forced to contribute their fair share to their children's well-being. Investing in the skills of those persons with limited access to labor markets, including welfare mothers, is part of a program to help them provide for their families.

While these ideas are conventional wisdom to progressive social scientists, you won't find them promulgated by Murray. He will continue to stress his reductionist solution and ignore the evidence that contradicts him. And until we are ready to reject simple answers to complex problems, his influence will be felt.