Returns to Education Using a Sample of Twins: Evidence from Japan*

• DOI: http://doi.org/10.1111/asej.12113
• Author: Shinji Yamagata,Tomohiko Inui,Makiko Nakamuro
• Published date: 01 March 2017
• Date: 01 March 2017

Returns to Education Using a Sample of Twins:

Evidence from Japan

*

Makiko Nakamuro, Tomohiko Inui and Shinji Yamagata

Received 12 February 2014; accepted 4 July 2016

The objective of this article is to measure the causal effect of education on earnings

using a large dataset of twins compiled in Japan with wide-ranging socioeconomic

information. Recent research conducted in China showed that, contrary to the case

in Western countries, ability bias is large in estimating the return to education. It is

argued that the low rate of return in China is in part due to the competitive and

exam-oriented education system, which is similar to the system in Japan. The

empirical results in our study show that the estimated rate of return to education

using the sample of identical twins in Japan is 9.3 percent even after omitted ability

bias and measurement errors in self-reported schooling were corrected. Our findings

suggest that the conventional OLS estimate is not significantly contaminated by

potential biases in Japan.

Keywords: economic return to education, endogeneity, identical twins,

measurement error.

JEL classification codes: I21, J30.

doi: 10.1111/asej.12113

I. Introduction

Following the prominent research of Mincer (1974), economists have long sought

unbiased estimates of the rate of return to education. However, unbiased estimates

remain difficult to obtain. One important methodological issue is the potential for

mutual causation between earnings and education. Another important problem is

the possibility that omitted variables that are often unobservable (e.g. differences

in ability and family characteristics) can lead to biased and inconsistent estimates.

An innovative way to address these issues is to estimate returns to education in

samples of identical twins which contain information on earnings and education.

Identical twins (monozygotic (MZ) twins) are produced in the same pregnancy,

when a single zygote splits to result in two separate embryos by chance. The

two are genetically identical, which, hence, enables us to hypothesize that a pair

of MZ twins share the same innate abilities. In addition to genetic endowments,

*Nakamuro (corresponding author), Faculty of Policy Management, Keio University, 5322 Endo

Fujisawa, Kanagawa 252-0882 Japan. Email: makikon@sfc.keio.ac.jp. Inui: Nihon University.

Yamagata: Kyushu University.

© 2017 East Asian Economic Association and John Wiley & Sons Australia, Ltd

Asian Economic Journal 2017, Vol.31 No. 1, 61–81 61

bs_bs_banner

they also share the same family and neighborhood environments. The objective of

using a sample of MZ twins is to control for unobserved ability and family char-

acteristics that influence the incentives for educational investments and reduce the

possibility of omitted variable bias from unobserved heterogeneity, which is often

referred to in the literature as ability bias.

1

The advantages of using a sample of twins include not only isolation of the effects

of education of earnings, holding differences in ability and family endowmentscon-

stant, but also the ability to compare MZ twins with non-identical twins (dizygotic

(DZ) twins). Because DZ twins are produced when two eggs are fertilized to form

two embryos in the uterus at the same time, DZ twins are not genetically identical.

Rather, they are regarded as ordinary siblings of the same age. This setting enables

us to measure the extent of the effect of unobserved ability in the estimated rate of

return to education through the comparison of MZ and DZ twins.

The objective of this paper is to rigorously estimate the causal effect of educa-

tion on earnings using a sample of twins, which may be the largest dataset of

twins complied nationwide in Japan. Our empirical results show that the rate of

return to education is approximately 10.0 percent when we employ the conven-

tional OLS. Once we account for omitted ability bias, the estimated rate of retur n

is dramatically reduced to 4.5 percent. The within-twin pair estimate has, indeed,

biased the conventional OLS estimate upward. Then, we corrected the measure-

ment errors using the instrumental variable method and obtained 9.3 percent as

the estimated rate of return to education, suggesting that measurement errors have

biased within-twin pair estimates downward. Taken as a whole, the bias-corrected

estimate is very close to the conventional estimate, which leads us to conclude the

conventional OLS estimate is not largely contaminated by potential biases. As

compared with estimates from other countries, Japan’s rate of return to education

is more similar to that in Western countries than in China.

The rest of this article is organized as follows. Section II provides a review of

the literature. Section III introduces the empirical models to be estimated. Sections

IV and V introduce data collection strategies and variables defined for empirical

analyses, respectively. Section VI presents the empirical results and identifies

the key empirical issues emerging in the econometric analysis. Section VII pro-

vides conclusions.

II. Literature Review

Pioneering literature in this field of study includes Ashenfelter and Krueger

(1994), Ashenfelter and Rouse (1998) and Rouse (1999). They estimated the rate

1 Note ‘ability bias’here is caused not only by genetic endowmentsbut also by family characteristics

that may be correlated with determinants of the optimal level of schooling. Ashenfelter and Rouse

(1998) defined an omitted ability as ‘unobserved family components’(p. 256), which are a dispropor-

tional combination of inherited ability, family environmentsand other unobserved skills.

ASIAN ECONOMIC JOURNAL 62

© 2017 East Asian Economic Association and John Wiley & Sons Australia, Ltd