Are women more likely than men to oppose corruption in China? Not yet

ABSTRACT

We examine the relationship between the ratio of females in the population and the level of corruption in China using annual panel data from 29 provinces during the period 1995–2012. We find that the influences of women on corruption are ambiguous and vary across regions.

KEYWORDS:

• Women
• corruption
• China
• panel model

JEL CLASSIFICATION:

• H30
• H60

Acknowledgements

The authors acknowledge financial support from Beijing Natural Science Foundation (9162013), the National Key Research and Development Program of China (2016YFA0602801) and Joint Development Program of Beijing Municipal Commission of Education. Chun-Ping Chang is also grateful to the eighth ‘Hundred Talents Program’ in Shaanxi province for financial support. All remaining errors are our own.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Goetz (2007) finds no evidence that women are less corrupt than men in the public sector in Ghana.

² As an anonymous reviewer pointed out, the registered number of cases of corruption may not be equal to the actual number of corruption cases. However, due to data accessibility, the actual number of cases in corruption in China is unavailable. Moreover, the majority of the recent studies about corruption in China utilized the registered number of cases of corruption as the measurement of the level of corruption in China, including Liu and Lin (2012) and Jin and Wu (2014).

³ Because media in China is tightly controlled by the government, it is interesting to examine whether and to what degree the controlled media power could still suppress corruption (Treisman 2007; Pellegrini and Gerlagh 2008).

⁴ Chinese provinces are categorized into six groups according to their geographic similarity, including (1) Northern China: Beijing, Tianjin, Hebei, Shanxi, Inner Mongolia; (2) Northeastern China: Liaoning, Jilin, Heilongjiang; (3) Eastern China: Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Shandong; (4) South central China: Henan, Hubei, Hunan, Guangdong, Guangxi, Hainan; (5) Southwestern China: Sichuan, Guizhou, Yunnan; and (6) Northwestern China: Shannxi, Gansu, Qinghai, Ningxia, Xinjiang.

⁵ According to the results of the Hausman test at the bottom of Table 2, the random effect and the OLS methods are proved to be inferior to FE estimators (Chang, Lee, and Weng 2011).

⁶ As a high-flying Communist official, Bo Xilai was a member of the political bureau of the powerful Communist Party Central Committee and the Party secretary of southwestern Chongqing city. In 2012, Bo was dismissed for a series of political misbehaviours and corruption allegations. Numerous officials have been taken into custody or investigated in the wake of the scandal. For more information, refer to http://www.bbc.com/news/world-asia-china-17673505.

⁷ The lagged levels of both dependent and explanatory variables are used as instruments to ensure maximum efficiency. According to the Sargan test of over-identifying restrictions and second-order serial correlation, the results verify the consistency and efficiency of the GMM estimators.

⁸ This effect is anticipated, because corruption is a ‘highly persistent phenomenon and contains strong forces that tend to perpetuate a given level of corruption’ and the ‘history of corruption is important in explaining current corruption levels’ (Herzfeld and Weiss 2003).

Additional information

Funding

The authors acknowledge financial support from Beijing Natural Science Foundation [9162013], the National Key Research and Development Program of China [2016YFA0602801] and Joint Development Program of Beijing Municipal Commission of Education. Chun-Ping Chang is also grateful to the eighth ‘Hundred Talents Program’ in Shaanxi province for financial support.