Poverty rates among female-headed households in the United States are substantially, and persistently, higher than poverty rates for households headed by married persons or unmarried males (see figure 1). This fact is presented as evidence in support of the claim that choices by women to live in nontraditional arrangements rank as an important cause of poverty and that, moreover, an effective anti-poverty policy should focus on redesigning welfare programs and the tax code so as to strengthen incentives to marriage and disincentives to nonmarital births. (1)
The family promotion initiatives favored by the political right are underpinned by viewpoints about the causes of rising female headship and illegitimacy--views articulated by Charles Murray (1984). (2) For Murray, the modern welfare state is supported by an ideology that blames "the system" for society's ills and relieves those dwelling at its fringes of moral responsibility for their own economic fortunes. As a result, behavioral patterns that in a more enlightened society are stigmatized and discouraged have been de-stigmatized and encouraged. (3) The solution is a re-establishment of personal responsibility. The Personal Responsibility and Work Reconciliation Act of 1996 (the welfare reform bill) apparently marks a movement toward this goal.
There remains a perplexing question to which Murray and others on the political right have not given a satisfactory answer. Specifically, if marriage offers a dependable route to a better economic station, why do so many poor females remain single or become divorced?
We argue that the incidence of female headship is to a significant degree the manifestation of fundamental demographic factors, the most important of which is the ratio of marriageable males to females with similar characteristics. Our objective in this article is twofold: (1) to investigate long-term trends in earnings among males ages twenty-two to thirty-four and (2) to assess the implications of these trends for marital prospects facing young females. We argue that two trends of the post-1977 era in the United States have caused a sharp reduction in breadwinner ratios--that is, the number of males per 1,000 females in the same age group that qualify as marriageable based on an income test. These trends are (1) falling average incomes for males ages twenty-two to thirty-four and (2) rising within-group earnings inequality for this population.
Economic Status of Young Males
The term "marriageable" connotes a complex of attributes that vary according to the person asked to define it. (4) But hardly anyone would dispute that earning capacity is an indispensable qualification for males seeking spouses. Thus, a simple approach to sorting out marriageable males is to apply an earnings test. That is, young men are classified as marriageable if their earnings meet or exceed some minimum threshold level deemed sufficient to insure a reasonable chance of success in marriage. So long as the income threshold constitutes an accurate measure of the minimum feasible earning power of marriageable males, counting the number of males above this threshold is a method that cannot underestimate the population of viable marriage candidates.
If a probability density function were formed to describe the earnings of young males, the area under the curve to the right of the marriage-qualifying threshold level of income measures the proportion of this group which meets the earnings test. Assuming no change in the shape of the function, an increase in mean income would increase the area under the curve above the (fixed) income threshold, and a decrease in mean income would diminish it. Changes in the shape of the distribution also affect the proportion of the sample above or below the income threshold. For example, increased skewness to the right of the distribution would cause, ceteris paribus, a decrease in the proportion of …