Economic Returns to Community College

ln CPS data, the 20% of the civilian labor force with 1-3 years of college earn 15% more
than high school graduates. We use data from the National Longitudinal Study of the High School
Class of I972 which includes postsecondary transcript data and the NLS Y to study the distinct returns
to 2-year and 4-year college attendance and degree completion. Controlling for family income and
measured ability, wage differentials for both 2-year and 4-year college credits are positive and
similar. We find that the average 2-year and 4-year college student earned roughly 5% more than
similar high school graduates for every year of credits completed. Second, average bachelor and
associate degree recipients did not earn significantly more than those with similar numbers of college
credits and no degree, suggesting that the credentialling effects of these degree are small. We report
similar results from the NLSY and the CPS.
In addition to controlling for family background and ability measures, we pursue two IV
strategies to identify measurement error and selection bias. First, we use self-reported education as
an instrument for transcript reported education. Second, we use public tuition and distance from the
closest 2-year and 4-year colleges as instruments, which we take as orthogonal to schooling
measurement error and other unobserved characteristics of college students. Although research over
the past decade has been preoccupied with selection bias, the two biases roughly cancel each other,
suggesting that the results above are, if anything, understated.

Milwaukee Parental Choice Program Evaluation

In 1990, Wisconsin became the first state in the country to provide vouchers to low income
students to attend non-sectarian private schools. In this paper, I use a variety of estimation strategies
and samples to estimate the effect of the program on math and reading scores. First, since schools
selected students randomly from among their applicants if the school was oversubscribed, I compare
the academic achievement of students who were selected to those who were not selected. Second,
I present instrumental variables estimates of the effectiveness of private schools (relative to public
schools) using the initial selection as an instrumental variable for attendance at a private school.
Finally, I used a fixed-effects strategy to compare students enrolled in the private schools to a sample
of students from the Milwaukee public schools. I find that the Milwaukee Parental Choice Program
appears to have had a positive effect on the math achievement of those who attended a private
school; but had no benefits for reading scores. I have found the results to be fairly robust to data
imputations and sample attrition, however these limitations should be kept in mind when
interpreting the results.

Data from "Wage Effects of Unions and Industrial Councils in South Africa"

We propose a general method of moments technique to identify measurement error in self-reported
and transcript-reported schooling using differences in wages, test scores and other covariates to discern the
relative verity of each measure. We also explore the implications of such reporting errors for both OLS and
IV estimates of the returns to schooling. The results cast a new light on two common findings in the
extensive literature on the retums to schooling: “sheepskin effects” and the recent IV estimates, relying on
“natural experiments” to identify the payoff to schooling. First, respondents tend to self-report degree
attainment much more accurately than they report educational attainment not corresponding with degree
attainment. For instance, we estimate that more than 90 percent of those with associate’s or bachelor’s
degrees accurately report degree attainment, while only slightly over half of those with l or 2 years of college
credits accurately report their educational attainment. As a result, OLS estimates tend to understate returns
per year of schooling and overstate degree effects. Second, because the measurement error in educational
attainment is non-classical, IV estimates also tend to be biased, although the magnitude of the bias depends
upon the nature of the measurement error in the region of educational attainment affected by the instrument.

Ashenfelter and Rouse Twinsburg Data

In "The Varied Economic Returns to Postsecondary Education: New Evidence from the
Class of 1972", an article recently published in the Journal of Human Resources (Volume 28,
no. 2, pp. 365-382), Norton Grubb reaches two main conclusions: (1) students who enroll in
two-year colleges without completing degrees earn no more than comparable high school
graduates; and (2) degrees from two-year colleges and vocational and technical institutes only
indirectly lead to higher earnings by providing students with access to jobs in which they can
accumulate experience and on-the-job training (i.e., access to "careers" instead of "jobs"). Given
that roughly half of those entering college today do so at community colleges and that roughly
a fifth of federal Pell Grant subsidies are spent at these institutions, such results are quite
provocative.
However, in this comment we show that several of the variables used in Grubb’s paper
are severely mismeasured and that, when they are corrected with reasonable alternatives, his
conclusions no longer receive empirical support. On the contrary, even those who enter but fail
to complete degrees at community colleges do seem to earn significantly more than similar high
school graduates. Further, controlling for work experience has relatively little effect on the
estimated returns.

Many argue schools that serve inner-city and rural children are in “crisis.” This paper
reviews the best available evidence on the effects of class size and school vouchers. Results from
the Tennessee STAR experiment suggest smaller class sizes improve achievement, particularly for
inner-city and minority children; results from the New York City voucher experiment and the
Milwaukee Parental Choice program suggest there may be small achievement gains in mathematics
for the African-American and Hispanic children who use vouchers. Although the reason of the
achievement gains is unknown, one candidate is the smaller class sizes in the private schools.

In a recent, and widely cited, paper, Ashenfelter and Krueger (1994) use a new sample of identical
twins to investigate the contribution of genetic ability to the observed cross-sectional return to schooling.
This paper re-examines Ashenfelter and Krueger’s estimates using three additional years of the same twins
survey. I find that the return to schooling among identical twins is about 10 percent per year of schooling
completed. Most importantly, unlike the results reported in Ashenfelter and Krueger, I find that the within-
twin regression estimate of the effect of schooling on the log wage is smaller than the cross-sectional
estimate, implying a small upward bias in the cross-sectional estimate. Ashenfelter and Krueger’s
measurement error corrected estimates are insignificantly different from those presented here, however.
Finally, there is evidence of an important individual-specific component to the measurement error in
schooling reports.