Changing Work Behavior of Married Women

Featured in print Digest

By 2000, a wage increase had only about half as much effect on women's decision about how many hours to work during the year than in 1980. Further, married women's work hours became less responsive to their husbands' wages as well.

Married women apparently are becoming accustomed to working outside the home. Maybe they like having their own careers. Or maybe they worry that, with a high divorce rate, they might split from their husband and need a separate income and career. Whatever the cause, women in the 1990s were less likely to leave employment or to reduce their annual work hours if their husbands got a pay hike than was the case in the 1980s. And, women were less likely to become employed or work more hours if their spouse got a pay cut. In other words, their labor supply decisions were less sensitive to their husband's income than they had been in the past. Equally interesting is that their work decisions also became much less sensitive to their own pay - whether it was high or mediocre - than before.

These are among the findings of Francine Blau and Lawrence Kahn in Changes in the Labor Supply Behavior of Married Women: 1980-2000 (NBER Working Paper No. 11230). The dramatic increase in the participation of women in the paid labor force since World War II is well known. In 1947, 31.5 percent of women and 86.8 percent of men were in the labor force. By 1999, women's labor force participation had roughly doubled to 60 percent, while men's had fallen moderately to 74.7 percent. What was a comparatively rare event in the late 1940s - women working outside the home - had become the mode by the 1990s. Women were only 15 percentage points less likely to work for money than men.

Beginning in the late 1970s or early 1980s, the authors further note, women's wages relative to those of men also rose: the female/male ratio of annual earnings of full-time, full-year workers increased from 60.2 percent in 1980 to 72.2 percent in 1999. That change partially reflected a slight decline in men's real wages in the 1980s, followed by some recovery in the 1990s. Moreover, after 1970, more and more women proportionately got high paying professional and managerial jobs.

Since 1990, though, the increase in the participation of women in the labor force and the gain in their wages relative to men have both slowed from previous decades. Women's labor force participation rose only from 57.5 percent to 60 percent between 1990 and 1999. The female/male annual earnings ratio barely increased, from 71.6 percent in 1990 to 72.2 percent in 1999.

Using Current Population Surveys done each March by the Census Bureau, Blau and Kahn examine the 1980-2000 changes in married women's labor supply, taking account of demographic factors, such as the number and age of their children, and other factors, including non-labor income. As women's relative wages increased, more and more went to work - but less so in the 1990s than in the 1980s. "This is a significant new development," the authors write.

What economists call the "labor supply elasticity" of women declined from a range of about .8 to .9 in 1980 to about .6 in 1990 and .4 by the year 2000. In other words, by 2000, a wage increase had only about half as much effect on women's decision about how many hours to work during the year than in 1980. Further, married women's work hours became less responsive to their husbands' wages as well - the effect of, say, an increase in their spouses' wages on married women's work hours fell by between 38 and 47 percent. "Fewer and fewer women are on the margin between participating and not participating in the labor force," Blau and Kahn note.

In contrast, the "labor supply elasticity" of husbands has always been very low (0 to .1) and has changed little in the two decades. Nor does a change in the pay of their spouses have much Influence on married men's labor supply.

Going back decades, women were perceived as secondary earners within the family, more likely to be affected by their spouses' wages. Now the traditional division of labor between men and women is breaking down. Men and women are more equally sharing home and market responsibilities, although women still bear a larger share of housework and child care than men do. These changes in gender roles are likely to help explain the authors' finding that the responsiveness of women's labor supply to economic incentives is becoming more like men's. That result holds up, the authors find, if tested under a variety of alternative specifications for married women and their mates. It holds under various tax regimes, say if the couple is eligible for the earned income tax credit, or if the couple is not legally married but rather cohabiting. It holds for different education levels and for mothers of small children. It also holds when some of the income of married women is unearned, that is, from investments. And it holds under a variety of assumptions about the wages women who are currently out of the labor market were offered.

These findings have implications for the national debate over supply-side tax policy - tax cuts aimed at encouraging more work. If married women's labor-supply elasticity has declined, then "the potential for marginal tax rate cuts to increase the labor supply is much smaller now than 20 years ago, since tax rates were much higher then as was married women's labor supply responsiveness," the authors write

-- David R. Francis