This paper examines the properties and prevalence of measurement
error in longitudinal earnings data. The analysis compares Current
Population Survey data to administrative Social Security payroll tax
records for a sample of heads of households over two years. In contrast
to the typically assumed properties of measurement error, the results
indicate that errors are serially correlated over two years and
negatively correlated with true earnings (i.e., mean reverting).
Moreover, reported earnings are more reliable for females than males.
Overall, the ratio of the variance of the signal to the total variance is
.82 for men and .92 for women. These ratios fall to .65 and .81 when the
data are specified in first-differences. The estimates suggest that
longitudinal earnings data may be more reliable than previously believed.
We present a survey of the literature on the economic returns to school quality. A
dozen studies conducted over the past 20 years show remarkably consistent estimates of the
effect of school quality on students’ subsequent earnings. A 10 percent increase in school
spending is associated with 1 to 2 percent higher annual earnings for students later in life.
We argue that the similarity of the ﬁndings across data sources and research methods suggests
that school quality has a true causal effect on student earnings. Increases in school resources
are also associated with signiﬁcantly higher educational attainment, although the range of
estimates of the effect is relatively wide.
This paper presents an overview and interpretation of the literature relating school quality to
students‘ subsequent labor market success. We begin with a simple theoretical model that
describes the determination of schooling and earnings with varying school quality. A key insight
of the model is that changes in school quality may affect the characteristics of individuals who
choose each level of schooling, imparting a potential selection bias to comparisons of earnings
conditional on education. We then summarize the literature that relates school resources to
students’ earnings and educational attainment. A variety of evidence suggests that students who
were educated in schools with more resources tend to earn more and have higher schooling. We
also discuss two important issues in the literature: the tradeoffs involved in using school-level
versus more aggregated (district or state-level) quality measures; and the evidence on school
quality effects for African Americans educated in the segregated school systems of the South.
We estimate the post-release economic effects of participation in prison-based General
Educational Development (GED) programs using a panel of earnings records and a rich set of
individual information from administrative data in the state of Florida. Fixed effects estimates of
the impact of participating in the GED education program show post-release quarterly earnings
gains of about 15 percent for program participants relative to observationally similar nonparticipants.
We also show, however, that these earnings gains accrue only to racial/ethnic
minority offenders and any GED-related earnings gains for this group seem to fade in the third
year after release from prison. Estimates comparing offenders who obtained a GED to those who
participated in GED-related prison education programs but left prison without a GED show no
systematic evidence of an independent impact of the credential itself on post-release quarterly
This paper quantities the extent to which the rise in the measured
return to education between I979 and 2000 is reﬂecting a change in the
causal effect of education on labor market eamings. The conceptual issues
are formalized in a two-factor model of ability. schooling and eamings that
allows heterogeneity in absolute and comparative advantage across the
population. ln particular, the framework implies that a rise in the true return
to education will increase the degree of convexity of the relationship
between eamings and years of education for a ﬁxed cohort of individuals.
Permanent differences in the levels of the eamings-schooling relationship
across cohorts will arise if the mapping between schooling and ability differs
across cohorts. These implications of the two-factor model allow the
identiﬁcation of changes in the causal effect of education over time and
During the Vietnam draft priority for military service was randomly
assigned to draft-age men in a series of lotteries. However, many men
managed to avoid military service by enrolling in school and obtaining an
educational deferment. This paper uses the draft lottery as a natural
experiment to estimate the return to education and the veteran premium.
Estimates are based on special extracts of the Current Population Survey
that the Census Bureau assembled for 1979 and 1981-85. The results suggest
that an extra year of schooling acquired in response to the lottery is
associated with 6.6 percent higher weekly earnings. This figure is about
10 percent higher the OLS estimate of the return to education for this
sample, which suggests there is little ability bias in conventional
estimates of the return to education. Our findings are robust to a variety
of "alternative assumptions about the effect of veteran status on earnings.