YOUTH LABOR MARKETS IN THE U.S: Results 7

Tables 4 and 5 turn to results for the number of jobs held during the five-year post-schooling period. For men, the OLS results in column (1) of Table 4 indicate a small (-.014) negative, significant effect of number of jobs held. Using the individual-level unemployment rates to instrument for number of jobs held, in columns (2) and (3), results in a small change in the estimated effect of the number of jobs held, with the estimate falling to zero. However, the F-statistic for the instruments in the first-stage regression is only 1.9, and the overidentifying restrictions are rejected (at the ten-percent level) even when the minimum unemployment rate on the current job is included. Thus, these IV estimates are not reliable.

OLS First stage IV First stage IV First stage IV First stage IV
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Number of jobs held, five-year post-schooling period -.014(.007) .00(.06) -.08(.04) -.03(-04) -.06(04)
Schooling .04 -.18 .04 -.15 .024 -.15 .03 -.19 .03
(.01) (.07) (.02) (.07) (.015) (07) (.01) (.06) (.01)
Experience .06 1.3 .04 1.4 .14 1.5 .08 1.4 .12
(.04) (.23) (.02) (.23) (.07) (.23) (.06) (.22) (.06)
Experience squared x Ю’2 -.12 -8.2 -.00 -9.5 -.62 -9.6 -.23 -9.3 -.47
(.23) (1.3) (.50) (1.3) (.41) (1.3) (.35) (1.2) (.39)
Currently married .17 .05 .17 .07 .17 .08 .17 .06 .17
(.04) (.20) (.04) (.20) (.04) (.20) (.04) (.19) (.04)
Non-white -.07 -.35 -.07 -.35 -.09 -.43 -.08 -.21 -.09
(.05) (.28) (05) (.28) (.06) (.28) (.06) (.26) (.05)
Minimum unemployment rate -.04 -.14 -.04 -.23 -.06 -.15 -.05 -.20 -.05
on current job (.02) (.11) (.02) (-09) (.02) (.11) (.02) (.09) (.02)
Current unemployment rate .016 .11 .01 .08 .02 .11 .02 .12 .015
(.011) (-06) (.01) (.06) (-01) (06) (.01) (06) (011)
uyi .04(.07)
-.11(.06)
U,p -.003(.04)
.08(.05)
Ц5 -.10(.07)
Uj. -.40(.09) -.41(.09)
Uj* -.16(-08) -.17(.08)
U, -1.6(.37)
(Uii.-U.il) .09(.07)
(Uij2-u.j2) -.02(.06)
(Цз-Uj,) -.01(06)
(U.4-U,,) -.08(.07)
(Uij5-Uos) -.03(.07)
(Uj.-U.j.) -.08(-05)
F-statistic for instruments in first stage 1.9 9.4 3.6 10.4
P-value for Hausman exogeneity test .80 .13 .67 .27
P-value for test of overidentifying restrictions .08 .90 .18 .08
P-value for test of overidentifying restrictions when minimum unemploymer rate on current job is excluded it .05 .89 .03 .00

In columns (4) and (5) the cohort average unemployment rates are instead used as instruments. In this case, the F-statistic for the instruments in the first-stage regression jumps to 9.4, and the overidentifying restrictions are not rejected. The first-stage estimates indicate that higher unemployment during the school-to-work transition results in fewer jobs held, consistent with the results in Tables 2 and 3, because the number of jobs is inversely related to stability. Finally, the IV estimates of the wage equation indicate that number of jobs held in the immediate post-schooling period has a sizable negative (-.08) and significant effect on adult wages. cheapest payday loans