The purpose of this paper is to illustrate the importance of characterizing the composition of the caseload at the time the welfare leavers sample is drawn. The paper also aims to exemplify one method of standardizing results across different types of leavers with different benefit receipt and work histories in order to make the studies more comparable across time and across areas. In general, we find that past welfare receipt history matters a great deal for outcomes, but not always as expected. We also find that those with more work experience prior to leaving were more likely to leave welfare and were much more successful in gaining employment and earnings after leaving welfare.
We described the composition of the caseload during the time the leavers sample was drawn according to their prior work and benefit receipt. Results presented in that section show that a significant portion of the caseload received AFDC benefits for at least 5 of the 6 years in the preobservation period. Most of the cases on AFDC in 1995 had fewer than two spells of benefit receipt in the preexit period. Only 14 percent had three or more spells of receipt. The caseload was divided into three groups: long-termers, short-termers, and cyclers. Under these definitions, 55 percent of the caseload were long-termers, 31 percent were short-termers, and 14 percent were cyclers. The caseload was also broken down by past work experience, as measured by the percentage of quarters in the preexit period with UI earnings. Twenty percent of the caseload did not work at all in the preexit period, 60 percent worked at least one quarter but no more than half the quarters, and 25 percent worked for more than half the quarters. Crossing work history with welfare receipt history, we found that those who had received benefits the longest had the least amount of work experience. Short-termers had the most work experience. Cyclers had the least amount of work experience.
We also showed outcomes by past benefit receipt and work experience. The first outcome examined was the proportion of cases that left welfare. Results showed that higher percentages of cyclers and short-termers left welfare than long-termers. Results also showed that higher portions of leavers were found in the groups with the most work experience. For those who left welfare, two sets of outcomes were examined: benefit receipt after exit (return to AFDC, food stamps, or Medicaid) and employment status and earnings after exit. Results show that the cycler, short-termer, and long-termer distinction is an important distinction for benefit receipt outcomes. Long-termers were much more likely than short-termers and cyclers to return to welfare, and a higher proportion of long-termers continued to receive food stamps and Medicaid after leaving than short-termers. Benefit receipt outcomes after leaving did vary by work experience prior to leaving welfare, but the differences were not large. On the other hand, employment and earnings outcomes after leaving varied substantially across prior work experience strata. As expected, those who had worked more prior to leaving welfare had higher employment rates and higher earnings after leaving. Employment and earnings outcomes also varied by prior AFDC benefit receipt, but not as drastically. Surprisingly, long-termers had better employment outcomes than short-termers. Long-termers were more likely to be employed after leaving and their earnings were higher after leaving than short-termers. Cyclers' employment rates and earnings did not differ greatly from those of long-termers.
The final part of the paper examines how important past benefit receipt distinctions and work experience distinctions are for these outcomes when other background characteristics of the cases are controlled. The probability of leaving welfare and the probability of ever being employed in the year after leaving welfare were estimated. Earnings after leaving were also predicted for welfare leavers. The primary finding in this section is that prior work experience was a consistently strong predictor of success. The percentage of quarters worked in the preexit period was positively associated with the probability of leaving welfare and the probability of employment after leaving. Quarters worked and average wages in the preexit period were both positive and strong predictors of quarterly earnings after leaving welfare.
We also found that past welfare receipt distinctions were important predictors of the probability of leaving welfare. Short-termers were significantly more likely to leave welfare than long-termers and in general, results consistently show that those who had received AFDC longer were less likely to leave AFDC. The cycler distinction was not a strong predictor of the probability of leaving welfare, although there is some evidence that those with one spell of benefit receipt were less likely to leave welfare than those with no prior spells of receipt.
The probability of being employed after leaving is, surprisingly, positively related to the length of time spent on welfare prior to the preexit period. Average spell length and long-termer status were both positive and strong predictors of the probability of employment after leaving welfare. For this outcome, the cycler distinction was an important predictor of employment as cyclers were significantly more likely to be employed than short-termers.
Spell length is positively associated with earnings after leaving as well. Long-termer status is associated with higher earnings after leaving. Furthermore, average spell length is positively associated with earnings after leaving. The number of welfare receipt spells were significant predictors of earnings after leaving. The coefficient for each category of number of spells (one spell, two or three spells, or four or more spells) is negative and statistically significant compared to those with no prior spells.
The results that long-termers worked more quarters and had higher earnings after leaving than short-termers and cyclers is contrary to expectations that previous dependency levels would be negatively correlated with employment outcomes. A good explanation for these results is not clear.
In summary, we conclude that in examining the outcomes of welfare leavers, it is important to characterize the caseload by their past work experience and by their past benefit receipt history because outcomes vary widely across different work experience and benefit receipt backgrounds. Work history background is especially important, we find, as the outcomes vary greatly according to different work experience groups. In terms of past benefit receipt history, the long-term versus short-term distinction is an important one. Distinctions by the number of spells of receipt show mixed results--sometimes this distinction matters, sometimes it does not.
"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)