The data set used for the exercise is the 1979 National Longitudinal Survey of Youth, which is a nationally representative set of individuals who were 14 to 21 in 1979 and who were interviewed annually up to 1994 and biannually since then. We utilize the survey through the 1996 interview, which gives us a maximum of 18 years of data.(14) We select only women, given our interest in welfare participation. We focus on a 10-year period of each woman's life, from ages 20 to 29. We could examine the entire 18 years, but such a time period is so long that women might not be easily characterized as long-termers, short-termers, or cyclers for the entire period; they easily could have been long-termers for the first 10 years and cyclers for the second 8 years, for example. The shorter, 10-year time period is less likely to capture multiple stages in the life cycle where welfare participation behavior is markedly different.(15)The data give us monthly AFDC participation information (only receipt of AFDC benefits is examined, not other welfare programs), thereby providing us with 10*12=120 months of observations with which to construct our measures of welfare experience. We have 2,763 women in our sample, 514 of whom experienced at least one month of AFDC receipt from ages 20 to 29.(16)
Figure 14-1 shows the distribution of T in the sample for the 514 women with at least one month of receipt. About 81 percent of the population had no months of receipt--the exact numbers are given in Table 14-A1.(17) The distribution of total months of receipt declines with the number of months on welfare, on average, as is typical of these types of distributions. About one fifth (19 percent) received AFDC for only 1 to 6 months in the 10-year period; but, on the other hand, 9 percent received benefits for 8 or more years. The mean and median number of months of benefit receipt are 39 and 28, respectively.
Figure 14-2 shows how these T distributions differ for those with different numbers of spells over the period.(18) Not surprisingly, the distributions are shifted to the right for those with larger numbers of spells. The median T for those with one, two, or three spells are 12, 47, and 47, respectively.
Figures 14-3 and 14-4 show the distributions of N and L, respectively.(19) Of those ever on AFDC over the 10 years, about 48 percent had only one spell of receipt. The distribution rapidly declines and, in fact, there are relatively few women who had large numbers of spells in these data: Slightly more than 8 percent had four or more spells and only 2 percent had five or more spells. A 10-year period is a long time and offers the possibility of many more spells than this. Thus, even at this relatively young age, the sample shows relatively little turnover.(20)
Figure 14-4 shows that most recipients have relatively short spells: One-quarter had average spells on AFDC of shorter than 6 months, and 57 percent had average spells of 18 months or shorter. However, this distribution has a long right-hand tail, and a significant number of women have long average spell lengths. Nearly another 20 percent of the sample, for example, had average spells of 3 or more years in length (37 months and over). This skewness is reflected in the marked difference in the mean and median spell length (24 versus 15). Table 14-A3 shows how these average spell lengths differ by the number of spells in the period. Interestingly, those with larger numbers of spells tend to have larger numbers of medium spell lengths, but smaller numbers of both long and short spells. That those with many spells are less likely to have long spells is expected, but that they also have fewer shorter spells (1-6 months) is not. This suggests that those with greater number of spells--cyclers--may have greater welfare dependency than what might have been thought otherwise, a suggestion that will come up again in subsequent tables.
Table 14-1 shows the characteristics of those women never on AFDC and those ever on in the 10-year period. We focus, in this table and all subsequent tables, on only a few fixed, unchanging background characteristics of individuals. These include education and race, but also employment status (whether worked at all during a year), annual earnings, weekly wages, and hourly wage rates, all measured only over periods off AFDC. The earnings and wages are measured only over periods off AFDC because they are intended to represent earning capability; including the AFDC periods would bias the measures in this sense, for earnings and wages are always lower on welfare than off.(21)In traditional economic analyses, an individual's potential hourly wage and weekly wage are usually considered to be the best indicators of labor market skill. Table 14-1 shows, as expected, that those ever on welfare have lower levels of education, are more likely black or Hispanic, and have lower mean earnings and wages than those never on welfare. Mean earnings and wages are, in addition, extremely low in absolute terms for those women who have been on welfare in other periods. On the other hand, annual employment rates are moderately high (66 percent), on average--women work about two-thirds of the time that they are off AFDC.
|Education||Never On||Ever On|
|Employment rate off AFDC||0.98||0.86|
|Average annual earnings off AFDC(including zeroes)||$11,698||$5,179|
|Average annual earnings off AFDC (excluding zeroes)||$11,940||$5,997|
|Average weekly earnings (excluding zeroes)||$293||$193|
|Average hourly wage (workers only)||$7.95||$5.50|
|Note: Monetary figures are in real 1992 dollars.|
Table 14-2 shows how these characteristics vary with T, N, and L. The variation in the characteristics with all three variables goes in the expected direction: Those with greater dependence, more spells, and longer average spell lengths tend to have lower levels of education, are more likely black (but not Hispanic), and have lower employment rates, earnings, and wages. A key additional question is whether there is any variation in characteristics by N and L, holding T fixed. We will consider this question using regression analysis.
|Number of Spells
|Average Spell Lengths
|Employment rate off AFDC||0.74||0.65||0.55||.65||.65||.65||.76||.67||.57|
|Average annual earnings off AFDC (including zeroes)||$6,672||$5,125||$2,911||$5,795||$5,066||$3,834||$6,063||$5,386||$3,811|
|Average annual earnings off AFDC (excluding zeroes)||$7,010||$5,794||$4,942||$6,484||$5,999||$4,641||$6,616||$$6,156||$4,998|
|Average weekly earnings (excluding zeroes)||$210||$193||$155||$209||$186||$160||$205||$192||$179|
|Average hourly wage (workers only)||$5.80||$5.65||$4.41||$5.92||$5.43||$4.42||$5.52||$5.59||$5.30|
"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)