Changhui Kang

Does education induce healthy lifestyle

We investigate whether education induces individuals to have healthy lifestyle. To test for causality we instrument education by high school availability and birth order using data of Korean men; to account for correlations among health behaviors, we estimate a quadvariate probit model using simulated MLE. Our results indicate that an increase in education induces individuals to exercise regularly, and to get regular health checkups. We find, however, that education has little effect on smoking or drinking. We also discover that unobserved determinants of health behaviors are correlated, especially between smoking and drinking, and between exercising and getting health checkups.

Family Size and Educational Investments in Children: Evidence from Private Tutoring Expenditures in South Korea*

Relying on private tutoring expenditures in South Korea, this paper examines whether large family size has a strong negative impact on educational investments in children. To deal with endogeneity of family size, the paper employs a non-parametric bounding method as well as an instrumental variables method. Our primary finding is that quantity–quality trade-offs in educational investments function in a way that varies by the sex of the child. While there is a non-negligible negative effect of large family size on educational investments for girls, there is little if any impact on those for boys. Son preferences traditionally shown by Korean parents seem to underlie such empirical findings.

Academic interactions among classroom peers: a cross-country comparison using TIMSS

Using an international data set from the Third International Mathematics and Science Study (TIMSS), we examine academic interactions among classroom peers for each country, and compare them across different countries. To minimize the bias that usually plagues peer effects studies, we take within-student differences between mathematics and science test scores. The results show a significantly positive association between peers’ performance and own achievement for most of the TIMSS countries. Moreover, the degree of mutual peer interactions within classroom is found to be surprisingly close across different countries, even if there exists a wide range of institutional differences in middle-school education (e.g. degree of ability mixing).

The Effect of Protection of Temporary Workers on Employment Levels: Evidence from the 2007 Reform of South Korea

The authors investigate the effects of recent South Korean labor reform, which was designed to protect temporary workers, on the level of temporary, permanent, and overall employment in that country. Because the effects of employment protection legislation (EPL) on employment levels remain theoretically ambiguous, they must be determined by empirical analysis. The impacts of the reform on both temporary and overall employment are negative at first and fade away in two years after the reform, hence taking a U-shape in the post-reform period. The impact of the reform on permanent employment is, however, positive two years after the reform. At least in the short run, an increase in protection for workers comes at a cost of reduced levels of employment.

PERFORMANCE OF VARIOUS ESTIMATORS FOR CENSORED RESPONSE MODELS WITH ENDOGENOUS REGRESSORS

The paper reviews four approaches (substitution (SUB), control function (CF), system reduced form (SRF) and artificial instrumental regressor (AIR)) dealing with endogenous regressors in censored response models, and compares them through a simulation. Based on mean-squared-error type criteria, CF and AIR perform better than SUB and SRF; in terms of computation, however, SUB and CF are the easiest, closely followed by SRF. Although CF does well in both accounts, its assumptions are restrictive, and CF provides very different results from the other estimators when a real data set is used. Therefore, although the choice of an estimator among the four should be case-specific, for practitioners, we would recommend SUB.

Estimation of Binary Response Models with Endogenous Regressors

This paper reviews six approaches to binary response (y 1 ) structural forms with an endogenous regressor y 2 : (i) the two-stage least squares estimator-like substitution approach, (ii) the control function approach, (iii) the system reduced-form approach, (iv) the artificial instrumental regressor approach, (v) the transformed-response instrumental variable estimator approach and (vi) the classical maximum likelihood estimator approach. The applicability of the six methods differs greatly, depending on whether y 2 is a continuously distributed random variable or a discrete transformation of a latent y 2 . We conduct a real-data-based simulation study, and provide an empirical illustration. Our overall recommendation is using (i) and (ii), as the others have undesirable features such as analytic complexity in (iii), computational difficulty in (iv) and (vi), and poor finite-sample performance in (v).

Effects of a holiday trip on health and quality of life: evidence from a natural experiment in South Korea

ABSTRACT We examine whether having a holiday trip affects an individual’s well-being, namely quality of life, health status, stress level and health behaviours. We use the two-stage estimation method to control for endogeneity of a travel experience, exploiting a natural experiment of distributing Travel Vouchers at random among qualified applicants in South Korea in 2012. We find that, for applicants whose decision to travel is influenced by receiving a voucher, a travel experience has no significant effects on the traveller’s well-being measured 3–12 months later. We also find that the OLS estimates overstate benefits of a travel.