Afterschool Child Care Subsidies and Maternal Employment Stability among Low-Income Populations

Purpose: Empirical studies have shown that the high cost of substitute child care pressures parents (especially mothers) to accommodate work arrangements to child afterschool time either by reducing work hours or adjusting work shifts, activities regarded as harmful to employment stability (Kimmel & Powell, 2007). Stable employment is crucial for wage growth for low-income populations (Blank, 2007). Childcare subsidy policies are designed to promote employment. Changes in the spending of child care subsidy explained an estimated 7 percent of the increase in employment among single mothers with children under 13 (Bainbridge, Meyers & Waldfogel, 2003). However, only about 15 percent of eligible families did use child care subsidy (Mezey, Greenberg, & Schumacher, 2003). Insufficient funding, fragmentary administration structure, as well as insufficient regulation of child care providers, weaken the effects of child care subsidy policies. Also, state governments have substantial discretion in formulating child care subsidy policies and vary on important policy characteristics. This study decomposes the components of child care subsidy policies. Specifically, it hypothesizes that the income eligibility level set by states (the higher the level means the wider the coverage range of eligible recipients) influences the likelihood of full time work (an indicator of employment stability) among low-income mothers through affecting their use of center based afterschool child care.

Method: Using individual data from National Survey of American Families 2002 and child care policy data from 13 states provided by the Child Care Bureau, 4,401 working mothers who had at least one biological child aged 6 to 11 years old (eligible for child care subsidy) are selected into the sample. In order to correct the sample selection bias caused by the unmeasured factors, such as the employment motif that may explain subsidy application and employment participation, this study uses Bivariate Probit Model (Instrumental Variable method) to specify the proposed model. The key exogenous predictor is the income eligibility level set by states (as an instrument variable). Other state level control variables are 85 percent of state median income, 85 percent of state median income as a percent of 1999 federal poverty, and percent of children served at child's home, or family home, or childcare center across 13 states. For individual level data, this study controls for mothers demographic characteristics such as age, race, education, health condition, immigration status, marital status, number of children, as well as child characteristics such as age, health condition, behavior problem index, and number of relatives (including the biological father) at home. The outcome variable is whether mother worked full time in 2002.

Results and Implications: The analysis shows that the higher the level of income eligibility set by the state, the higher likelihood of center-based child care use (probit coefficient is .113), as well as the stronger effect of center-based child care on full time work decision (probit coefficient increases from .583 to .864). This study provides policy makers, social workers, and policy advocacy group empirical evidence on what specific aspects of policy programs can be improved to be more effective.