The Effect of Household Structure on Women's Economic Activity and Fertility: Evidence from Recent Mothers in Urban Mexico

Abstract

This study examines the effect of the presence of a "mother substitute" in the household on urban womens participation in the labor force and fertility in Mexico in 1982. Data were obtained from the 1982 Mexican National Demographic Survey on reproductive history educational status employment status and household structure for a subsample of 2746 urban women who were married (90%) or in a consensual union (10%) and had had a child within the prior 5 years. Theoretically women with higher value of market time lower household income and lower value of nonmarket time will more likely be working women. Women had 3.4 surviving children 83% used contraceptives and 15% were employed. Analysis involved use of reduced form equations to model the joint determination of labor force participation and fertility; justification of this method is given. Husbands education was a proxy for income level. Cost of child rearing is figured as the mothers number of children >5 years and household structure. "Mother substitute" was an additional unemployed female aged >13 years. An imputed wage for all women in the sample is used to measure the value of time in the labor force. Probit regression estimation methods revealed that mothers were more likely to be working when their wage offers were higher. Those with a caretaker in the household and with a less well-educated husband are also more likely to work. Those more likely to be employed also had few children 6-12 years and more children >13 years. In the fully reduced form of the equation determinants of the wage offer (education age region) were substituted for the instrumental variable; the results were consistent. An additional woman in the household was significantly positively related to the likelihood of women working. In the reduced form fertility equation women who were younger better educated with more children and with better educated partners were less likely to have more than 1 child born within the last 5 years. The implication from the negative coefficient of the household-structure variable is that women with additional caretakers are less likely to have more than 1 young child. The causes are explained. Mother substitutes increase the likelihood of working women and do not increase the likelihood of having more children particularly among nonemployed mothers which differs from US results. Future research should refine the measure of mother substitute and account for her presence in the household.