The changing impact of marriage and children on women's labor force participation

Monthly Labor Review

Volumn 132, Issue 2, 2009, Pages 3-14

  Hoffman, Saul D ^a  

^a UNIVERSITY OF DELAWARE   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 69949083470     PISSN: 00981818     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (27)

1. See Sharon R. Cohany and Emy Sok, "Trends in labor force participation of married mothers with infants," Monthly Labor Review, February 2007, pp. 9-16;
2. and Abraham Mosisa and Steven Hipple, "Trends in labor force participation in the United States," Monthly Labor Review, October 2006, pp. 35-57.
3. Heather Boushey, "Are Women Opting Out? Debunking the Myth," briefing paper (Washington, DC, Center for Economic and Policy Research, November 2005);
4. Cohany and Sok, "Trends in labor force participation."
5. The labor force participation rate of married mothers is higher than that of all married women because mothers of children aged 18 years or younger are younger than the population of all married women, a group with no upper age limit
6. Mosisa and Hipple, "Trends in labor force participation," table 10, p. 47
7. See Lisa Belkin, "The Opt-Out Revolution," New York Times Magazine, Oct. 26, 2003;
8. Louis Story, "Many Women at Elite Colleges Set Career Path to Motherhood," The New York Times, Sept. 20, 2005;
9. and Claudia Wallis, "The Case for Staying Home," Time, May 10, 2004
10. Boushey, "Are Women Opting Out?" The years included in her analysis are dictated by the unavailability of information on presence of children by age in the CPS-ORG data between 1993 and 1999
11. Cohany and Sok, "Trends in labor force participation."
12. Data files were obtained from the Center for Economic and Policy Research's data archive at www.ceprdata.org/cps/org_index.php (visited Feb. 27, 2008).
13. "Household Data Annual Averages," table 2, "Employment status of the civilian noninstitutional population 16 years and over by sex, 1973 to date" (Bureau of Labor Statistics, 2007), on the Internet at www.bls.gov/cps/cpsaat2.pdf (visited Mar. 25, 2008).
14. The logit probability is exp(XB)/[1 + exp(XB)], where the B's are the estimated coefficients. The marginal effect in a logit model is B × P × (1 - P), where P is the mean sample proportion.
15. Sok, personal communication, June 2007. Thus, the sample analyzed by Cohany and Sok cannot be replicated here.
16. These proportions are based on information on the presence of a child in given age ranges. Thirty-five percent of the observations have missing data for all child age variables. It is clear that the missing data are actually substantive 0's. With this conversion, BLS distributions of women by age of child may be replicated exactly (see Women in the Labor Force: A Databook, Report 985, May 2005, table 6); without it, the distributions are widely different. The relevant information is shown in the appendix to this article. It appears that some skip sequence triggered the missing data, but the details are not obvious in the CPS-ORG data.
17. Boushey interprets year dummies as business-cycle variables. In modeling women's labor force participation, however, it is problematic to interpret trends or year dummies as due solely to business-cycle effects. Boushey finds that controlling the year has a large effect on the estimated impact of having a child on labor force participation.
18. This calculation reflects the changing estimates of the effects of children, marriage, and marriage × children between 1993 and 2004
19. Just under half of the women with children aged 18 years or younger have a child younger than 6
20. Logit estimates are virtually identical and are available upon request
21. See Welfare Reform: States' Early Experiences with Benefit Termination (General Accounting Office, May 1997)
22. S, 1993), where LFPR is the labor force participation rate and the subscripts M and S denote married and single women, respectively
23. Saul D. Hoffman and Laurence S. Seidman, Helping Working Families: The Earned Income Tax Credit (Kalamazoo, MI, W. E. Upjohn Institute for Employment Research, 2003)
24. For evidence of this effect, see Nada O. Eissa and Hilary Williamson Hoynes, "Behavioral Responses to Taxes: Lessons from the EITC and Labor Supply," NBER Working Paper No. W11729 (Cambridge, MA, November 2005)
25. The actual fertility rates were 205.2 and 198.4 births per thousand for women aged 20-24 years, 98.5 and 118.0 for women aged 30-34 years, and 37.8 and 54.5 for women aged 35-39 years. (See Vital Statistics of the United States, 2002: Volume I, Natality, on the Internet at www.cdc.gov/nchs/datawh/statab/unpubd/natality/natab2002.htm (visited Oct. 20, 2007);
26. Joyce A. Martin, Brady E. Hamilton, Paul D. Sutton, Stephanie J. Ventura, Fay Menacker, and Martha L. Munson, "Births: Final Data for 2003," National Vital Statistics Reports (Hyattsville, MD, National Center for Health Statistics, Sept. 8, 2005)
27. and Joyce A. Martin, Brady E. Hamilton, Paul D. Sutton, Stephanie J. Ventura, Fay Menacker, and Sharon Kirmeyer, "Births: Final Data for 2004," National Vital Statistics Reports (Hyattsville, MD, National Center for Health Statistics, Sept. 29, 2006). Over the entire 1984-2004 period, the fertility rate for 30- to 34-year-old married women increased 43 percent and the rate for 35- to 39-year-olds increased 107 percent.