WHEN DO SINGLE MOTHERS WORK - AN ANALYSIS OF THE 1990 CENSUS-DATA

This study analyzes the relative effect of the amount of public assist ance income received one year on the probability that re single mother is employed the following year compared to a variety of other determi nants of employment status. The analysis is based on a national sample which was drawn from the Public Use Micro data 5 percent Sample (PUMS ) of the 1990 U.S. Census. It consists of the 275,744 female household ers who were divorced, separated, widowed or never married, and living with their own children age 18 and under. Logistic regression was uti lized to calculate the probability of being employed in 1990 according to sources and amounts of income in 1989, level of education, age, wo rk experience, number and age of children, race, and marital status. T he results indicate that greater amounts of public assistance income r educed the probability of being employed. However, several other facto rs-including race-ethnicity, family form and size, educational backgro und and previous earnings-were significant, independent determinants o f labor-force status. In particular, African-American women, women wit h children under six, women with relatively low levels of education an d low earnings in the previous year, and never-married women all faced a reduced probability of being employed in 1990 regardless of how muc h public assistance income they received in 1989. The paper concludes with an assessment of the implications of these findings for current d ebates on the relationships among welfare receipt, work incentives, an d employment.