Female employment and fertility in developing countries
Simple no-work/work distinctions are an unreliable basis for estimating causal linkages connecting female employment/work-status patterns to fertility. World Fertility Survey (WFS) data show about 3/4, 1/2, and 1/4 child differentials for over 20, 10-19, and under 10 years marital duration grouss respectively, for women employed since marriage. Effects on marriage seem strongest in Latin America and weakest in Asia. Controlling for age, marital duration, urban-rural residence, education, and husband's work status. But from the results of a number of WFS and other studies, it seems relationships of work status and fertility are difficult to confirm beyond directional indications, even in Latin America. A UN study using proximate determinants such as contraception and work status including a housework category indicated differentials in contraceptive practice were not significant net of control for education. Philippine data indicates low-income employment might increase fertility by decreasing breastfeeding, while WFS data from 5 Asian countries indicated pre-marital work encourages increased marriage age, without being specific about effects. Also, female employment must affect a large population to have a real impact on aggregate fertility, since female labor force activity is likely to change slowly if at all. Data presently available do not cover micro-level factors that may be important, such as effects of work on breastfeeding, nor do they lend themselves to examination by multi-equation analysis. More work is needed to isolate effects of work-status attributes like male employment, and to analyze intra-cohort mid-course fertility objective changes, as well as new theoretical process models such as competing time use and maternal role incompatibility.