Sunday, June 2, 2002

A story published today by the New York Times reports that for years, work and marriage have been seen as twin pillars of welfare reform. But just as President Bush is seeking welfare legislation with more stringent work requirements and more support for marriage, an unexpected contradiction is emerging. New research findings in Connecticut and Iowa show that the stricter work requirements of contemporary welfare policy significantly reduce the chances that a single mother will wed.

Researchers suggest two main reasons. Like middle-class married women whose divorce rates spiked when jobs and rising wages made them more self-reliant, some women who moved from welfare to work may have become less willing to settle for the wrong man, they say. At the same time, strict work requirements and low wages may have left some mothers with less time, energy and income to attract a partner or nourish a relationship.

"Tough love, less romance," summarized Bruce Fuller, a social scientist at the University of California and an author of the Connecticut study. He privately discussed similar findings with Thomas M. Fraker, the author of the unpublished Iowa evaluation, which is awaiting state approval. "If tough-love work policies suppressed marriage at this magnitude nationwide, just under a quarter-million women would not be getting married in any one year," Professor Fuller said.

The effect of changes in welfare on marriage surfaced first in an evaluation of preschool children whose mothers were randomly assigned to Connecticut's program, known as Jobs First.

Researchers at the University of California, Columbia, Yale and Stanford were startled to discover that three years later, only 7 percent of the mothers who had gone through Jobs First were married and living with a spouse, compared with 15 percent of those randomly assigned to receive traditional welfare grants under the old program, Aid to Families with Dependent Children. Among women in the two groups with stronger employment histories, the gap was even greater: 6 percent versus 18 percent.

Now, researchers say, the same wedding gap has shown up in Iowa, in the results of a much larger welfare evaluation. It was conducted by Mathematica, a research corporation, under a state contract that forbids public release of the findings without approval by state officials. At the close of the study, which followed 4,000 women for as long as six years, 32 percent of the old-style welfare group were married, but only 24 percent of the welfare applicants assigned to Iowa's new program, a mix of work requirements, sanctions and incentives typical of many states' new welfare programs.

In interviews, women who were subjects of the Connecticut study echoed theories about new self-reliance and greater stress, even as they described more complicated reasons for broken engagements and failed relationships.

The new findings might seem inconsistent with an earlier study that found a small decline in the proportion of low-income children living in one-parent households without other adults, and another that found a two-percentage-point increase in marriage among 2,000 low-income women in Boston, Chicago and San Antonio. But those studies did not focus on welfare applicants, their authors noted. Unlike the Connecticut and Iowa evaluations, they were not set up to separate the impact of welfare policy from the strong economy and other factors. And the results are quite compatible, said Andrew Cherlin, a professor of public policy at Johns Hopkins University and an author of the three-city study.