The Demand for Child Care and Child Care Costs: Should We Ignore Families with Non-Working Mothers

This paper investigates the patterns and determinants of how families care for their pre-school age children. In contrast to all existing studies, we analyze the child care decisions of households with non-working mothers as well as those with working mothers or mothers in job training or educational programs. Our source of data is the Fifth Follow-Up of the National Longitudinal Survey of the High School Class of 1972 (NLS-72). In the first part of the paper, we present a statistical portrait of the utilization of non-parental forms of child care, as well as the parental expenditures on and costs associated with such care. By and large, our estimates of the incidence of non-parental child care utilization, the hours used, the distribution of types of care chosen, and the expenditures on such care by households with working mothers corresponds with estimates derived from previous surveys. We also find that households with non-working mothers, make substantial use of non-parental forms of child care. Such households disproportionately use center-based care (e.g., child care centers and nursery schools) and spend non-negligible shares of total family income on these services for their pre-school children. We also find that the utilization of non-parental care among households with non-working mothers is concentrated among households which are black, have lower levels of income, and are female-headed. In the second half of the paper, we examine the sensitivity of the inferences drawn for estimates of parental demand functions for child care, of models of child care mode choice, and of equations characterizing the determination of the hourly cost (or price) of non-parental child care to the sample restrictions present in previous studies. We examine whether the estimates from those studies generalize to all households by conducting a systematic verification analysis of the estimates found in the previous literature compared to the population of all households. Based on these comparisons, we find clear evidence that the child care demand of households with working mothers are selectively different from the population of all households with pre-school age children. Moreover, we find important differences in the inferences about the effects of various variables on parental child care decisions based on an analysis of all households versus samples of households with working mothers only. In contrast to previous studies, we find significant effect of state regulations governing child care providers on the utilization of non-parental child care. States which limit the child-to-staff ratios of licensed day care centers and/or impose minimum educational requirements on providers of less-formal family day care centers are found to have substantially lower rates of non-parental care utilization and the presence of such regulations significantly increase the cost of non-parental care. In addition, while previous estimates suggest that the demand for non-parental child care services increases with the level of income from sources other than the mother’s labor earnings, among all households, increases in income lead to increases in the demand for parental care. We also examine whether the structure of child care decisions systematically differs if a mother works as compared to when she does not, a question which could not be addressed with previously available data sets. We present estimates of child care demand equations conditional on the mother’s work decision and which allow for the endogeneity of the mother’s work choice. We find evidence consistent with structural differences in child care demand by the mother’s work status. Moreover, the results indicate that the child care demand of households with non-working mothers is more price sensitive than when the mother works and that the demand-deterring effects of state child care regulations we find in the analysis of all households is primarily due to the adverse effects such regulations have on households with non-working mothers.