Featured White Papers
Content provided in partnership with
Business Services Industry
Single mothers in various living arrangements: differences in economic and time resources
American Journal of Economics and Sociology, The, July, 1996 by Karen Fox Folk
In addition to requiring teens to live with parents, proposed welfare reforms would promote employment and some educational training. The results in Table 4 show that the marginal effect of living in the parent's home on public aid receipt is similar in size to the marginal effect of being employed for both white and black mothers (comparing standardized regression coefficients) with a much smaller and nonsignificant effect of education. However, education has significant effects on increased income adequacy among single mothers in the NSFH, with greater returns for black than for white mothers. An added year of education for blacks increases income per adult equivalent by $1,340, a 26% increase over mean black income per adult equivalent (see Table 4). For white mothers, an added year of education increases income per adult equivalent by a similar amount ($1,270), but, since white mothers have much higher average levels of income adequacy, this is only a 13.5% increase in overall income per adult equivalent.
Given these results, focusing on increasing education and employment of all single mothers would have longer-lasting effects for both mothers and their children than mandating coresidence with parents. A recent longitudinal study of AFDC recipients also found that education and earnings were the most important factors reducing welfare receipt. Regardless of race or ethnicity, having a high school diploma was the most powerful influence increasing the probability of living above the poverty line (Spalter-Roth, Hartmann, and Andrews, 1992). Both these results support Bane's (1986) earlier analysis of household composition and poverty, which concluded that the problem of poverty would be most effectively alleviated by focusing on improving employment skills that would increase the labor force participation and/or the wages of single mothers.
Notes
1. Unless otherwise noted, all data are weighted to adjust for selection of individuals from households of varying sizes and nonresponse and for the oversampling of minority and particular family types, such as single-parents and cohabiting couples. Weights used also adjust the sample to match the current U.S. population profile for age, race, and sex (Sweet, Bumpass, and Call, 1988).
2. The Current Population Survey (CPS) does not ask directly about cohabitation, which must be inferred from household structure, that is, two unrelated adults of opposite sex living together. Another difference between data from the NSFH and the CPS is that the NSFH includes only respondents 19 or older while the CPS includes single teenage mothers 15 years or older (Winkler, 1993). However, these younger teenage mothers make up only 1.1% of all single mothers in the CPS sample. They do differ from older mothers in living arrangement, with almost all living in a subfamily arrangement rather than heading an independent household.
3. Respondents were asked to estimate number of hours per week spent in nine household tasks: (1) preparing meals, (2) washing dishes and meal cleanup, (3) cleaning house, (4) outdoor and other household maintenance, (5) shopping, (6) washing, ironing and mending clothes, (7) paying bills and recordkeeping, (8) automobile maintenance and repair, and (9) driving other household members to school, work, or other activities. These estimates were summed for total time spent in household work. This is an admittedly crude measure of time use which probably overestimates time spent. An analysis of single-parent's time-diary data from the 1985 survey of Americans' use of time finds total household work time averaging 10 hours per week less than that reported by single mothers in the NSFH (Huang, 1994). However, this overestimation is likely a simple inflation since studies using the NSFH time-use data have obtained results consistent with other studies using time-diary data (see Perry-Jenkins and Folk, 1994 and studies they cite).
