Religiosity and Political Connections of Private Firms in China

ABSTRACT

The extant literature on political connections has paid little attention to the role of entrepreneurial traits in firms’ pursuit of connections with government agencies or bureaucrats. Using a nationally representative survey of private enterprises in China, we investigate whether and how religious beliefs of entrepreneurs affect their firms’ reliance on political connections. We find that firms founded by religious entrepreneurs are significantly more likely to establish political connections compared to firms founded by nonreligious entrepreneurs. The positive relation between religiosity and political connections, however, is found to exist only in regions with weak market-supporting institutions. Considering the link between religiosity and risk aversion, our findings suggest that religious entrepreneurs, especially those in regions with underdeveloped institutions, may establish political connections so as to reduce the institutional risks in their business operations.

KEY WORDS:

• political connections
• religious belief
• risk aversion

JEL:

• H70
• P26
• P30
• Z12

Supplemental material

Supplemental data for this article can be accessed here.

Notes

1. Although political connections generally bring returns to firms, it has also been found that politically connected firms tend to be less efficient than unconnected firms (Faccio 2010; Fan, Wong, and Zhang 2007).

2. It has been found in several surveys that between 10% and 15% of the Chinese population adhere to a religion (Lu 2012). Furthermore, based on the Spatial Explorer of Religion jointly developed by the Center on Religion and Chinese Society at Purdue University, the China Data Center at the University of Michigan, and the State Key Laboratory for Information Engineering in Surveying, Mapping and Remote Sensing of Wuhan University, we find that on average the amount of venues for religious activities per ten thousand people at the province level increased from 0.12 in 1978 to 0.47 in 2004; and the mean values of venues for religious activities per ten thousand square kilometers at the province level increased from 21.76 in 1978 to 149.81 in 2004.

3. These mixed results might be partly explained by international differences in religious doctrines and historical evolutions, particularly within the Protestant segments (Noussair et al. 2013).

4. http://news.xinhuanet.com/mrdx/2007-06/10/content_6222271.htm.

5. http://miit.gov.cn/n1146295/n1652858/n1652930/n3757016/c3762062/content.html.

6. The data can be accessed at http://www.enterprisesurveys.org/. Figures are calculated by authors based on the data.

7. The survey was conducted every two years since the early 1990s. Firms are re-sampled for each survey, which makes the data a repeated cross-sectional in nature. We use the surveys in 2008 and 2010 as survey questions with respect to entrepreneurs’ religious beliefs are contained in questionnaires only for years 1993, 2008, and 2010 (not in 2012 and thereafter), and data in the year 1993 has a long time interval with the other two surveys.

8. We carefully check the distribution of the sample across regions and industries before and after the data cleaning. We find that there is almost no difference in the distribution of religious entrepreneurs (the focus of this study) across regions and industries between the initial and the final sample. Details about the distribution of the sample across regions and industries are available upon request.

9. There are 25 entrepreneurs who chose “other” religion and we could not classify their religious beliefs to eastern or western religion.

10. Marginal effects are evaluated at the means of the independent variables, and the marginal effect of a dummy variable is calculated as the discrete change in the expected value of the dependent variable when the dummy variable changes from zero to one.

11. We compare the absolute size of the marginal effects of other entrepreneurial variables and that of the religion dummy.

12. There are 379 entrepreneurs who are deputies of both the CPPCC and the PC, accounting for 23.1% and 32.8% of the total deputies of the CPPCC and the PC, respectively.

13. We also investigate the relationship between religiosity and political participation in the PC for firms located in provinces with different institutional risks. Similar to the results in Table 3, the coefficients of religion are statistically insignificant in both subsamples and available upon request.

14. We exclude firms headquartered in the following provinces: Qinghai, Ningxia, Xinjiang, and Tibet.

15. The multiplier, 1.3, is identified in Oster (forthcoming) in a sensitivity analysis as the cutoff value which would allow at least 90% of results to survive assuming δ = 1 for articles based on randomized experiments published in several reputable economics journals between 2008 and 2013.

16. Here we use the absolute number of religious sites as the proxy of religiosity at the province level. We also experiment with the amount of venues for religious activities per ten thousand people at the province level in 2004 and the amount of venues for religious activities per ten thousand square kilometers at the province level in 2004, taken from Spatial Explorer of Religion, for robustness checks. Results are consistent with those in Table 5 and available upon request.

17. Data on GDP, population, and total public officials are taken from various years of China Statistical Yearbook. Data on corruption and bribery public officials are from various years of Procuratorial Yearbooks of China.

18. The division of the nine economic regions takes into account not only the level of economic development, but also factors such as population size, geographical area, and major geomorphological demarcations (Deng 2010). The nine economic regions are North East (Heilongjiang, Jilin and Liaoning), Bohai Rim (Beijing, Tianjin, Hebei and Shandong), Pan-Yangtze River Delta (Shanghai, Jiangsu, Zhejiang, Anhui and Henan), Southern Coastal (Guangdong, Guangxi, Fujian and Hainan), Xiang E Gan (Hunan, Hubei and Jiangxi), South West (Sichuan, Chongqing, Yunnan and Guizhou), Northern Plateau (Shaanxi, Gansu, Ningxia, Shanxi and Inner Mongolia), Xinjiang (Xinjiang), Qinghai–Tibet Plateau (Qinghai and Tibet).