Child-Care Expenditures and Women' s Employment Turnover

Abstract

Previous studies addressing the issue of child-care constraints on women's employment have drawn on specialized samples or used cross-sectional designs. Using the Census Bureau's Survey of Income and Program Participation, this study draws a 1985 sample of employed women paying for child care. After correcting for selection bias, I used logistic regression analysis to predict the log odds of quitting work by 1986. Weekly child-care payments were a significant predictor of employment turnover. In addition, the effect of child-care expenditures: (1) was stronger for mothers of preschool-aged children, (2) did not vary by the woman's wage, and (3) was not associated with transitions between fulland part-time work statuses. In the postwar period, female labor force participation rates have increased steadily, and the rate of increase has been greatest among married women with young children. In 1960, 19% of women with preschool children were in the labor force; by 1986 this figure was 54% (U.S. Bureau of the Census 1987). The entry of women into paid employment has increased the demand for child-care services while reducing the supply of potential providers, causing severe scheduling and payment problems for many American families (Hofferth & Phillips 1987). The notion that child care inhibits women's employment is common both to the general public and to policymakers.1 Surprisingly, the literature on this topic is sparse and inadequate. Much research has been done on the cognitive and developmental effects of children's placement in day-care centers (for an assessment of the literature, see Belsky 1988); other studies examine the productivity effects of providing child-care benefits to employees (see summary review in Auerbach 1988). Yet the studies on child care's impact on women's employment often draw on unrepresentative i The University of North Carolina Press Social Forces, December 1991, 70(2):495-508 * This research was supported by theNational ScienceFoundation (Grant SES-8912024) and by the Centerfor the Study of Work and Family at the University of Cincinnati. I thank Bill Feinber& Karen Mason, and Rachel Rosenfeld for their comments on earlier drafts of this article; Barbara Arrighi-Merz, Kathleen Fox and Karen Mullinfor their research assistance; and Karen Feinbergfor her editorial advice. Please direct all correspondence to DavidJ. Maume, Jr., Department of Sociology, University of Cincinnati, ML 378, Cincinnati, OH 45221. BITNET: MAUME @ UCBEH.' This content downloaded from 157.55.39.104 on Mon, 20 Jun 2016 05:53:40 UTC All use subject to http://about.jstor.org/terms 496 / Social Forces 70:2, December 1991 samples or use a cross-sectional design to ask women about their intentions to work if child care were available. This study uses a longitudinal design to investigate the impact of child-care expenditures on employment turnover for a nationally representative sample of women. This study's findings should advance our understanding of female employment behavior and has policy implications as well. In a nation that has recently discussed subsidizing the cost of child care (Kosterlitz 1989), determining the effect of reduced expenditures on child care on women's work attachment is important.