We have started with the important economic proposition
that investment in human capital competes with investment in
physical capital by using scarce resources. Thus, the purpose if
our study was to estimate the "social" marginal rates of return
on investment in different levels of education within the context
of benefit-cost analysis, and by doing so to help the policy
maker in deciding the desirable level of investment in education.

One of the major findings of our study is that social
marginal rates of return on investment in education is relatively
lower than the rates of return to physical capital. Since the
average rates of return on investment in physical capital are
believed to be about 20 percent in Korea, rates of return for all
levels of education are lower than those for physical capital.
This means that there has been overinvestment in human capital
relative to physical capital, and from the purely economic point
of view, misallocations of resources between human capital and
physical has occurred in the Korean economy.

Reflecting the overinvestment in human capital, the annual
rate of increase in stocks of human capital exceeds that of
physical capital throughout the period between 1960 and 1971.
The ratio of human capital stocks to physical capital stocks is
124.79 percent in 1971, which is quite opposite to the case of
developed countries. Annual aggregate investment in education is
thus also very high, and for 1971 the ratio of aggregate
educational investment to GNP reaches 18.2 percent.

The second point to be mentioned is that social rates of
return first increase in the interval of schooling years 9 and 12,
and then decrease in the range 12 and 16. This shows that
additional social rates of return sharply decrease for high school
graduates when he further goes to finish college. This may be
interpreted as a supporting evidence to the common observation
in Korea that we have too many colleges over our need.
Therefore, the present heavy subsidization for national and public
colleges should be reexamined.

There are, for sure, numerous difficulties in applying the
rate-of-return approach to developing nations. Expecially,
earnings differentials does not reflect true marginal social
products due to labor market imperfections. However, even if we
take into considerations all the necessary qualifications, our
conclusion that relatively overinvestment in human capital has
occurred in the Korean economy does not seem likely to be
changed.

Therefore, the right direction for further study to pursue is
for example to divide colleges into their major departments and
to calculate rates of return for each field. Also, the distinction
between private and social rates of return is necessary to find
out the various levels of government subsidization for different
kinds of schooling. These are a couple of example for further
study to follow.

Ⅰ. Introduction
Ⅱ. The Theoretical Framework
Ⅲ. Sources of Data
1. Costs of Education
1) In-School Expenditures
2) Out-of-School Expenditures
3) Total Direct Educational Expenditures
2. Average Incomes Streams
Ⅳ. Social Marginal Rates of Return on Investment in Education.
1. Social Marginal Rates of Return on Investment in
Education.
2. Stocks of Human Capital.
Ⅴ. Summary and Conclusion