As more mothers with young children choose to participate in the labor force, the issue of child care has become increasingly important for both policymakers and academic researchers. Decisions concerning the use of child care are dependent upon many economic and demographic factors, and the process by which these decisions are made has implications for the effectiveness of social policy concerning child care. Efforts to increase the supply of child care must take into account the factors that parents demonstrate as important in their choices of what child care to use.
All of the previous work examining child care has concentrated on just the principal or primary mode of child care that the family uses. For many families, however, there are additional arrangements made for child care. There have been no formal studies of this secondary care, and little is known about the extent of these arrangements and their economic and demographic determinants. An important question is whether this secondary care is used to enhance the care that the child receives, or is required because of the lack of adequate primary sources of child care. In order to fully examine the use of secondary child care, this paper analyzes a variety of different data sources: the National Longitudinal Survey Youth Cohort (1986 and 1988), the June 1982 Current Population Survey (CPS), and the 1984-85 Survey of Income and Program Participation (SIPP).(1)
The extent to which secondary arrangements are used for child care depends heavily on the definition of a "secondary" arrangement. The June 1982 Current Population Survey found that 17 percent of women used more than one child care arrangement for their youngest child under five. In the 1984-85 SIPP data, only about 13 percent of women used more than one child care arrangement for any of their children under five years old. If all arrangements for any children under five are counted, about 16 percent of women in the SIPP sample used more than one type of child care while they worked. When child care arrangements for all children (including those five and over) were reported, 43 percent of all women said they were using more than one form of child care. It is unclear as to whether there is a trend towards less overall use of secondary child care; when the comparable samples of the 1984-85, 1986, and 1987 SIPP are examined, the use of secondary arrangements decreases slightly from 13.1 to 11.8 percent.
The organization of this paper is as follows: the following section reviews the conclusions of some previous empirical work in the economics of child care choice. Section III describes the various sources of data to be used in this paper. Section IV discusses theoretical considerations concerning the analysis of secondary child care arrangements. Section V presents the empirical findings of this paper on both the use and determinants of secondary child care arrangements. Section VI summarizes the paper and adds some concluding remarks.
II. Empirical Studies of Child Care Choice
Although there has been no previous research into secondary child care arrangements, there have been several papers that have examined the process of choice for primary child care arrangements. This section of the paper will summarize significant literature in this area of study, and examine what factors have been important in the choices of primary care, with the purpose of determining if these same factors are prominent in an analysis of secondary child care arrangements.
Duncan and Hill (1977) examine data from the 1974 wave of the Panel Study of Income Dynamics (PSID), and find a statistically significant positive relationship between the mother's wage and the probability of choosing a day care center; they concluded that high-wage women desired a more "reliable" type of child care. Contrary to their expectations, however, Duncan and Hill found no statistically significant coefficients on the variables measuring family income, educational attainment, number of children in the household, or the recent mobility of the family. …