LABOR MARKET INFORMATION: Results 9

Results for the non-probationary and probationary subsamples are reported in Table 4.

Columns (1) and (2) present results for men and women combined, using age and education as instruments, and columns (3) and (4) present results for men only, using age as the instrument. The proportion subject to probation is testing is .79, with the proportions very close among men and women.34 The results in Table 4 suggest that labor market information drives the differences between the OLS and IV estimates of a. For both samples, the ratio aoLS/aIV is considerably higher for the non-probationary sample for which initial labor market information should be better, although this ratio is estimated imprecisely for the smaller non-probationary subsamples.

OLS OLS OLS IV OLS IV OLS IV IV
(1) (2) (3) (4) (5) (6) (7) (8) (9)
A. Usine linear performance ratine
Performance rating .0014

(.0011)

-.0001

(.0010)

.027

(.009)

.0004

(.0010)

.034

(.010)

.0010

(.0010)

.041

(.010)

.047

(.011)

Black -.136

(.036)

-.156

(.036)

-.101

(.057)

-.172

(.037)

-.089

(.066)

-.185

(.039)

-.081

(.075)

-.065

(.083)

Hispanic -.010

(037)

-.042

(.037)

-.026

(.056)

-.062

(.038)

-.026

(.065)

-.041

(.040)

-.004

(.073)

.001

(.081)

Schoolings 12 .090

(.056)

Schooling = 13-15 .173

(.059)

Schooling – 16+ .227

(.066)

Age .035

(.009)

.039

(.009)

.037

(.013)

Age2 x 10‘2 -.040

(.012)

-.044

(.012)

-.047

(.019)

General experience required -.036

(.039)

-.033

(.039)

-.015

(.058)

-.030

(.040)

-.006

(.068)

Specific experience required .132

(.042)

.135

(.043)

.091

(.066)

.181

(.043)

.103

(.077)

Vocational education/training required .233

(.044)

.247

(.044)

.207

(.068)

.255

(.046)

.205

(.079)

R2 .228 .003 .204 .148 .037
P-value from F-test of instalments in first-stage regression:
Age variables only .17
Education variables only .00
Job requirement variables only .69
Instruments Educ. Educ., age Educ., age Educ., age, job requirements
F-statistic on instruments in
first-stage regression: 5.00 3.98 4.44 2.95
Overidentifying restrictions, p-value: .75 .95 .20
Bias in OLS estimates, p-value from Hausman test:
Performance rating .00 .00 .00 .00
Black .21 .13 .11 .10
Hispanic .70 .50 .55 .55

Thus, based on the distinctions between probationary and non-probationary workers, the OLS and IV estimates of a correspond to what would be expected if there is imperfect information in labor markets. Although the evidence is not statistically strong, it suggests that the changes in estimates of a that result from instrumenting reflect errors of measurement on the part of employers in the reported performance rating, relative to initial expected productivity, rather than perfect information on the part of employers, with the measurement error reflecting only errors in the reported (to the researcher) performance rating relative to true, known productivity./spantdtd