YOUTH LABOR MARKETS IN THE U.S: Results

The main results are reported in Tables 2-5, where the estimates are reported first for longest tenure attained in the five-year post-schooling period, for men and women, and then for the number of jobs held during this period, also for men and women. In these tables, contemporaneous tenure is excluded as a control variable in the log wage equations, because it is likely to be strongly linked-and to some extent caused by-early tenure or job attachment; thus, these might be interpreted as “reduced form” estimates of the effects of early job stability. However, the effects of early job stability net of adult tenure are also of interest, in part to test whether early job stability has benefits other than those reflected in the length of attachments between workers and firms. Thus, Table 6 explores the sensitivity of the results to adding current tenure (and its square) to the wage equation, for the key specifications chosen from the other tables. Some caution must be exercised in interpreting these latter estimates, however, given that current tenure, like early tenure, may be endogenous (e.g., Abraham and Farber, 1987; Altonji and Shakotko, 1987); if the “returns” to adult tenure are overstated and adult tenure and early tenure are positively correlated, then including adult tenure will bias downward any estimated beneficial effects of early job stability. best payday loans online

OLS First stage IV First stage IV First stage IV First stage IV
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Longest tenure attained, five-year post-schooling period .01(.01) .08(.05) .08(.04) .07(03) .13(.06)
Schooling .04 .08 .03 .04 .03 .04 .03 .09 .03
(-01) (.03) (01) (.03) (01) (.03) (.01) (.03) (.01)
Experience .06 -.05 .06 -.20 .06 -.20 .06 -.01 .05
(.03) (.10) (04) (.10) (.04) (.10) (.04) (.10) (-04)
Experience squared x 10′2 -.17 2.2 -.25 3.4 -.25 3.4 -.25 2.0 -.31
(.18) (.55) (.19) (.56) (.19) (.56) (.19) (.53) (20)
Currently married .17 -.07 .17 -.10 .17 -.11 .17 -.02 .17
(.03) (.10) (.03) (.10) (.03) (-10) (-03) (.10) (.03)
Non-white -.05 -.07 -.05 -.03 -.05 -.03 -.05 -.03 -.05
(.05) (.14) (.05) (.14) (.05) (.14) (.05) (-13) (.05)
Minimum unemployment rate -.04 .15 -.05 .11 -.05 .13 -.05 .15 -.05
on current job (.01) (.05) (.02) (.04) (.02) (.05) (.02) (.05) (-02)
Current unemployment rate .01 -.04 .02 -.03 .02 -.03 .02 -.06 .02
(.01) (.03) (.01) (.03) (.01) (.03) (.01) (.03) (.01)
-.00(.03)
Ц, .11(.03)
Ц* -.01(.02)
u,4 -.05(.02)
uij5 -.06(.02)
U,i. .30(.05) .30(.05)
Uj3 .04(.04) .04(.04)
UjS -.13(.04) -.13(.05)
u, 1.12(-20)
(U«rUj.) -.04(.03)
(Ulj2-U.j2) .04(.03)
(Цз-Цр) -.00(.03)
(ЦгЦи) -.02(.03)
(Uij5-U.j5) ,00(.03)
(U4.-U.,) -.006(-02)
F-statistic for instruments in first stage 11.4 34.3 13.1 16.1
P-value for Hausman exogeneity test .12 .02 .04 .06
P-value for test of overidentifying restrictions .18 .78 .45 .38
P-value for test of overidentifying restrictions when minimum unemploymen rate on current job is excluded t .02 .80 .09 .01

Column (1) of Table 2 reports OLS estimates of a standard log wage equation supplemented by the longest tenure attained in the initial five-year post-schooling period. Paralleling the results in GN, the estimated coefficient of the longest tenure attained is small (.01) and statistically insignificant. The other estimated coefficients are standard. The low estimated return to schooling (.04) is attributable to the inclusion of the AFQT score; with the test score excluded, the estimated return to schooling is .06. Columns (2) and (3) report results from the first attempt to remove bias from omitted job match quality, by instrumenting for longest tenure attained with early unemployment rates. The first-stage estimates are reported in column (2).