Table 13-20 presents Tobit estimates of leavers' earnings in the first year after exiting AFDC. Again, the relationship between preexit welfare receipt and preexit earnings on postexit earnings is of key interest. In these estimates, a measure of average quarterly earnings in the years prior to the exit period are included in this model as an additional measure of prior work history.
|Independent Variable||Model 1||Model 2|
|Intercept||1.85||0.29 (a)||2||0.29 (a)|
|Cycler (reference group is short-termers)||0.03||0.04|
|Average AFDC spell length||0.02||0.003 (a)|
|Average AFDC spell length squared||0||0.000 (a)|
|Total # of spells =1||-0.34||0.08 (a)|
|Total # of spells=2 or 3||-0.36||0.08 (a)|
|Total # of spells=4 or more||-0.26||0.09 (a)|
|Average quarterly earnings before leaving||0.32||0.02 (a)||0.33||0.02 (a)|
|# Quarters with earnings before leaving||0.05||0.00 (a)||0.05||0.002 (a)|
|Age of case head||-0.03||0.02(b)||-0.03||0.02(a)|
|Age of case head squared||-0.001||0.002||0||0|
|Hispanic (reference group is white)||0.27||0.06 (a)||0.27||0.06 (a)|
|No high school diploma||-0.28||0.03 (a)||-0.27||0.03 (a)|
|At least some college (reference group is high school diploma)||0.39||0.03 (a)||0.39||0.03 (a)|
|Age of youngest child||0||0||0.002||0.005|
|# of children under age 5||-0.02||0.02||-0.02||0.02|
|# of children over age 5||0.14||0.02 (a)||0.13||0.02 (a)|
|Legal immigrant||0.51||0.19 (a)||0.53||0.19 (a)|
|Other adult present in household||-0.07||0.03 (a)||-0.07||0.03 (a)|
|Milwaukee County resident||0.45||0.04 (a)||0.43||0.04 (a)|
|Resident of other urban county (reference group is rural county resident)||0.08||0.04 (b)||0.07||0.04 (a)|
|Child receives SSI||-0.18||0.05 (a)||-0.18||0.05 (a)|
|Mother receives SSI||-1.89||0.09 (a)||-1.88||0.09 (a)|
|Sanctioned case||-0.52||0.05 (a)||-0.52||0.05 (a)|
|Unemployment rate in county in 1996||-0.1||0.02 (a)||-0.1||0.02 (a)|
|Left AFDC 4th quarter 1995||0.03||0.03||0.03||0.03|
|Left AFDC 1st quarter 1996||0.01||0.04||0||0|
|Left AFDC 2nd quarter 1996 (reference is left AFDC 3rd quarter 1995)||-0.14||0.03 (a)||-0.15||0.03 (a)|
|Scale parameter||1.64||0.01 (a)||1.64||0.01 (a)|
|Number of censored cases||2,227||2,227|
|(a) Coefficient is statistically significant at the 5-percent level.
b Coefficient is statistically significant at the 10-percent level.
Results show that long-termers have higher earnings than short-termers even after controlling for other demographic, programmatic status, and local economic conditions. Status as a long-term AFDC user is positively associated with earnings after leaving and is statistically significant. This result holds even after controlling for the age of the leaver, prior work experience; and average quarterly earnings prior to leaving welfare, which is surprising because it is contrary to initial predictions that long-termers would have more barriers to self-sufficiency and have lower earnings after leaving. Further explanations for this result should be explored. It is possible that there are compositional differences in the welfare dependency groups that are not observed with these data. Cyclers, however, do not have higher earnings than short-termers. In combination with results from the first model in Table 13-19, although cyclers are more likely to be employed after leaving welfare than short-termers, they do not have earnings that are significantly different from short-termers.
The second model uses ASL and its square as measures of previous welfare benefit receipt history. It also uses the series of dummy variables for the number of prior welfare spells as measures of the degree of cycling on and off welfare. Results show that longer spells of benefit receipt are associated with higher earnings, but that the longer the spells of receipt, the slower the increase in earnings. These results are consistent with the first model. The results are also consistent with findings from the 1995 cohort of leavers in Cancian et al. (1999), but not with earnings of the 1997 cohort of leavers from Cancian et al. (2000b).
Each of the coefficients on the dummy variables for the number of spells of benefit receipt are negative and statistically significant. Those with one spell of benefit receipt, those with two or three spells of benefit receipt, and those with four or more spells have significantly lower earnings than those with no prior spells of benefit receipt. The coefficient is the largest for those with two or three spells (-0.36).
The coefficients on average quarterly earnings and on total number of quarters worked in the years prior to leaving are positive and statistically significant. Those with higher average earnings in the preexit period had higher earnings after leaving. Likewise, those who worked more quarters prior to leaving welfare had higher earnings after leaving welfare, although the size of the coefficient is smaller than the size of the coefficient on average earnings prior to leaving welfare. Both results are as expected and indicate that a key component of labor market success after leaving welfare is work experience prior to leaving welfare.
"01.pdf" (pdf, 472.92Kb)
"02.pdf" (pdf, 395.41Kb)
"03.pdf" (pdf, 379.04Kb)
"04.pdf" (pdf, 381.73Kb)
"05.pdf" (pdf, 393.7Kb)
"06.pdf" (pdf, 415.3Kb)
"07.pdf" (pdf, 375.49Kb)
"08.pdf" (pdf, 475.21Kb)
"09.pdf" (pdf, 425.17Kb)
"10.pdf" (pdf, 424.33Kb)
"11.pdf" (pdf, 392.39Kb)
"12.pdf" (pdf, 386.39Kb)
"13.pdf" (pdf, 449.86Kb)