This paper employs data from a merged sample of the National Surveys of Family Growth to examine how female employment status conditions the relationship between education and wanted and unwanted births among African American and white women. A rationale is presented for why a minority group status hypothesis that posits lower fertility among more highly educated African American women as compared to similar white women might find support in the case of wanted births and among certain women, including earlier birth cohorts. Our results provide some evidence for these ideas as well as evidence for a social characteristics hypothesis that predicts convergence of childbearing with rising education. However, persistently higher levels of unwanted births among African American women of all educational levels suggest that the dynamics of racial fertility differences are more complex than either of the hypotheses imply.
PIP: This study examines the impact of female employment on wanted and unwanted births among African-American and White women in the US, the time in the life course when education is measured, and female employment status prior to first birth. The study aims to examine more carefully Johnson's (1979) minority group status hypothesis and the social characteristics hypothesis. Data were obtained from pooled data from the 1973, 1976, 1982, and 1988 National Surveys of Family Growth (NSFG) among a subsample of 34,827 ever married women with at least one child who had completed their wanted childbearing. The three tests of selection bias revealed robust models without substantial selection bias. The sample includes 8515 women who worked prior to first birth (2054 Blacks and 6461 Whites) and 5223 nonworking women prior to first birth (1931 Blacks and 3292 Whites). Findings indicate that nonworking women, compared to working women, had a fertility higher by about 0.33 children. Logistic models indicate that Black women's fertility decreased with increased levels of education. Black women with lower levels of education had higher wanted and unwanted fertility, regardless of employment status or when education was measured. When education was measured late in the life course, working Black women had lower fertility (by 0.1 children) than their White counterparts. The wanted fertility of highly educated, nonworking Black women, regardless of when education was measured, was lower (by 0.33 children) than their White counterparts. When education was measured at first marriage, the differences by race were about 0.4 children. This cross-over pattern for wanted births was not evident in the cohort of women born after 1945. Unwanted childbearing for Black women at all education levels was higher than among Whites. Differences by race decreased with increases in educational level. Findings support both hypotheses but incompletely explain unwanted fertility by race.