Gender quotas: A key to equality? A look at South Korea and Japan

            South Korea and Japan both face a huge, imminent economic and social problem: rapidly shrinking birthrates and slowing economies. To boot, the two countries are often at the bottom of the OECD list for gender equality, each ranking in the low 100's on the World Economic Forum list.[1] Measures have been put into place by both federal governments to create more equality for women in the hopes of eventually boosting the economy. One method used by states worldwide is gender quotas. South Korea has federally legislated gender quotas, and Japan does not. This post seeks to find the effect of gender quotas in the social sphere by looking at the effect of gender quotas on percent female enrollment in tertiary education.

Gender quotas are a controversial topic in places like the U.S. Around the world however, more countries have some type of gender quota than do not. The most important question is whether these top-down policies actually have an effect on the ground. In South Korea, this seems possible. Once more women were in South Korea’s legislature, they increased the number of women-friendly policies[2] Paid family leave increased from 8 weeks in 1998 to 65 weeks by 2012.[3] These policies, and social policies in general, go hand in hand with economic reform meant to stimulate an economy.[4] If gender quotas do increase gender equality in a state with women-friendly policies, and more women can help speed a slowing economy, gender quotas have potentially huge implications for the future by way of increasing female presence in the workforce and offering plenty of benefits to encourage them to also have children.[5]

H0: Gender quotas have no effect on female enrollment in tertiary education in South Korea.

HA: Gender quotas will have a positive effect on the enrollment of women in tertiary education in South Korea.

 A difference-in-difference test will be used to test the above hypothesis. This method is especially good for measuring the effect of a policy change by selecting two similar cases and applying a treatment. Difference-in-difference also allows for a temporal element which helps lessen the risk of reverse causality. In this study, Japan and South Korea will be compared and the treatment will be gender quotas. Japan and South Korea are apt comparisons because both are in the same region, have shared history and are wealthy OECD countries. Since the legislation for quotas in South Korea was put into law in 2000, this study focuses on the years 1998-2012. Difference-in-difference is better than a cross-sectional time series analysis because it allows us to see what South Korea may have looked like without gender quotas. While not perfect, this still provides very nice answers to “what if” questions.

Table 1. Summary Statistics

	Mean	Std. Dev.	Min.	Max.	Description
% Women in Japanese Parliament	8.25	2.1	4.6	11.3	Percent women in Japanese national legislature
% Women in Korean Parliament	10.58	4.67	3.7	15.7	Percent women in Korean national legislature
Japanese Female Tertiary Enrollment	50.47	5.22	41.19	58.23	Percent gross women enrolled in Japanese tertiary education
Korean Female Tertiary Enrollment	71.96	11.76	50.07	85.32	Percent gross women enrolled in Korean tertiary education
Observations	15

The main variables will be gender quotas as the dependent variable and the female percentage enrollment in tertiary education as the independent variable.[6] The percentage of tertiary enrollment will be regressed while controlling for labor force participation. Enrollment in higher education was chosen because it can show whether these top-down policies trickle down to the societal level by breaking down cultural norms that might have otherwise prevented women from pursuing higher education. Tertiary education is also a key element women need to compete in the workforce.[7]
It is hard to quantify gender inequity in a state, but looking at social institutions is a good place to start.
 

The common trend assumption can be justified by looking at Figure 1. Both countries had only about a 10% difference in the percentage of women enrolled in higher education, and both were steadily increasing. However, in 2000 when gender quotas were legislated into law, South Korea begins to see a sharp spike in female enrollment, eventually far eclipsing Japan, which still continues to increase, as well. There appears to also be a clear correlation.[8]

Table 2. Average Effect of Gender Quotas on Female Enrollment in Tertiary Education

DV: % Enrollment	Pre-Quota	Post-Quota	Difference
South Korea	65.1	75.14***	13.5***
Japan	49.5***	47.34	1.77
Difference	15.6***	27.8***	11.6***

Note: Differences-in-differences estimates. *, **, *** indicate statistical significance at the 10%, 5%, and 1% level respectively.

As seen above, the results are very significant for tertiary enrollment.[9] The difference-in-difference is a positive 11.6. This can be translated as gender quotas led to an 11.6% increase in women enrolled in tertiary education. In Japan, there was only a 1.77% increase, while in South Korea there was a 13.5% increase-- a marked difference. While it is always difficult to prove causality, there does appear to be a clear correlation. With all this information, I can reject the null hypothesis. Unfortunately, gender quotas were insignificant when regressed with female labor participation rate as the outcome variable.[10] This is interesting, that even though Korean women appear to be excelling in higher education, it appears their fortunes are not improving upon entering the labor force.

This has exciting possibilities for countries like Japan, which is hoping to level the playing field for women. While there were perhaps other policies in South Korea that led to an increase in women in higher education, there is a strong enough correlation to presume that gender quotas did indeed have an effect. However, gender quotas do not appear to be enough. Even when women are excelling in higher education and gaining the credentials to theoretically compete equally with men in the labor market, and with all the legislation in place designed to encourage women in the workforce, it appears that top-down policies are helpful, but not the final answer.

---

[1] World Economic Forum. "The Global Gender Gap 2014." http://www3.weforum.org/docs/GGGR14/GGGR_CompleteReport_2014.pdf.

[2] Peng, Ito. "Social Investment Policies in Canada, Australia, Japan, and South Korea." International Journal of Child Care and Education Policy 5, no. 1 (02, 2015): 41-53, http://search.proquest.com/docview/1656262545?accountid=8285.

[3] Ibid, pp 48.

[4] Ibid.

[5] Subacchi, Paolo. "The G20 Are Finally Taking Gender Equality Seriously."  (2015).  [accessed April 23, 2015].

[6] There are three main types of gender quotas: reserved seats, legal candidate quotas and political party quotas. The first two are legislated and enforced by federal law. The first regulates the number of women elected, and the other two forms set a minimum number of women candidates on the lists. The last is voluntary on behalf of the political party. South Korea has legal candidate quotas. (See: Gender Quota Database) 

[7] David, Miriam E. "Women and Gender Equality in Higher Education?" Education Sciences 5, no. 1 (2015): 10-25, http://search.proquest.com/docview/1672897504?accountid=8285.

[8] 0.983 is the correlation between the two, to be exact.

[9] All p-levels are well below the .05 level, almost all below the .01 threshold.

[10] I also ran a difference-in-difference test for just labor force as the outcome variable, which interestingly, but perhaps logically, yielded insignificant results.