Estimates of the returns to schooling in Taiwan: evidence from a regression discontinuity design

ABSTRACT

Using micro-data from the Manpower Utilization Survey 1995–2015 of Taiwan, this paper exploits exogenous variation in years of schooling caused by the 9-year compulsory education reform to estimate the economic returns to schooling. The results of fuzzy regression discontinuity design indicate that the causal effect of education on wage is approximately 5.6%, smaller than the corresponding ordinary least squares estimates. Our results are robust to various specifications and placebo tests.

KEYWORDS:

• Returns to education
• regression discontinuity
• Taiwan

JEL CLASSIFICATION:

• I28
• I20

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ 2.30% for senior high school, 3.98% for vocational school, 4.58% for junior college, and 12.20% for university.

² Card (2001) offers a survey of IV literature on returns to education, the IV estimates of returns to education are typically larger than the corresponding OLS estimates.

³ No observations are dropped through procedures (2) and (3).