Political uncertainty and bond defaults: evidence from the Chinese market

ABSTRACT

In this paper, we provide new evidence on how political uncertainty influences bond default, using China’s local officials’ turnover as a source of plausibly exogenous variation in uncertainty. We find that officials’ turnover increases bond defaults when political links are stronger, indicating that political turnover can destabilize government support for politically connected bonds. Additionally, off-budget resources have expanded China’s local government’s ability to bail out local bonds. Further results show that political connections create zombie bonds that ought to be out but instead to be in the market, reducing the allocative efficiency of financial resources. Our study suggests that, at least in some countries, political uncertainty influences financial risk through the mechanism of political connection.

KEYWORDS:

• Political turnover
• political connection
• bond market
• default

JEL CLASSIFICATIONS:

• D22
• H11
• M12

Acknowledgements

We are grateful to the referees and the editors, for valuable comments and suggestions. Finally, the usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 According to the existing literature, it is a major and widely accepted method to take official turnover as a proxy variable of political uncertainty. Many studies consider official turnover in defining political uncertainty. For example, Child, Chung, and Davies (2003) believe that politician turnover was a specific source of political risk; Liu and Zhong (2017) define political uncertainty as the unpredictability of government policy or regulatory changes caused by the turnovers of political leaders. In addition, extensive literature uses official turnover as a proxy variable to measure policy uncertainty (See, for example, Białkowski, Gottschalk, and Wisniewski 2008; Durnev 2010; Julio and Yook 2012; Gao and Qi 2012; An et al. 2016; Lv and Bai 2019). There are still many studies, not enumerated one by one.

2 Many pieces of literature explain corporate default risk based on political uncertain events such as government turnover, but within our reading scope, they basically use credit spreads or credit default swap spreads to represent default risk (e.g., Pástor and Veronesi 2013; Julio and Yook 2012; Waisman, Ye, and Zhu 2015; Wang, Xu, and Zhong 2019). Some literature uses binary dummy variables when measuring bond defaults, such as Moeller and Molina (2003), Brutti (2011), Gennaioli, Martin, and Rossi (2014, 2018), etc., but the discussion of political uncertainty is not involved.

3 www.people.cn.

4 This paper also takes the municipal party committee secretary turnover as the core explanatory variable and the result of regression is not significant.

5 Many studies pointed out that the OLS model, Probit model, and Logit model have their own advantages and disadvantages in handling the regression of 0–1 dependent variable (Noreen 1988; Angrist and Pischke 2008; Mood 2010; Breen, Karlson, and Holm 2013). Many works of literature mainly use OLS estimation when dealing with the explained variables of dichotomies (e.g., Angrist, 2001; Abramitzky, Delavande, and Vasconcelos 2011; Wang and Xie 2015). Referring to the literature, OLS estimation is also used as the main estimation method in this paper, supplemented by Probit estimation and Logit estimation to prove the robustness of the conclusion. At the same time, Tobit estimation is added to eliminate the concern for possible parameter estimation bias caused by asymmetric distribution.

6 The transparency index covers the regional budget and budget implementation, the three public expenses, industrial investment funds, and other dimensions. Because of the difference in the total statistical score each year, this paper will take the total transparency index at the municipal level after the percentage system and then take the average as the basis for the grouping.

7 The general assessment and ranking were given in ‘The Business Environment Index Report of China’s Sub-provinces in 2017' in various provincial administrative regions of China through a comprehensive assessment of the tax burden, financial services, financing costs, infrastructure construction, market environment, and intermediary services of enterprises in 2010, 2012 and 2016.The sample time range selected in this article is2011–2019. Therefore, the sample in 2011–2012, 2013–2016, 2016–2019 will use the values in 2010, 2012, and 2016, respectively. Meanwhile, we select the average ranking of the business environment of enterprises in provincial administrative regions as the basis for the grouping.

8 Municipal ‘two sessions' happen usually later than the national ‘two sessions' time. Therefore, the years 2013 and 2018 are grouped into ‘two sessions' and the rest of the time is grouped into non-‘two sessions' in this paper.

9 In addition, since there may be a lag in policy adjustment after the turnover, and short-term uncertainty should be more reflected as the current impact, to further verify that short-term uncertainty plays a leading role, this paper adds the variable of the previous mayor turnover to the model, the regression results are not statistically significant, and the hypothesis 5 is further verified. Due to space limitations, this result is not presented in the paper.

10 If the two places are consistent, the value is 1; otherwise, the value is 0. The year of turnover is still handled following the general practice. If the turnover of office takes place from January to June of the current year, it is regarded as the change of the current year and the successor official is matched; otherwise, the outgoing official is matched.

11 For example, the source of repayment funds of Sichuan Coal Industry Group Co., Ltd. (‘15 Sichuan Coal CP001’) is the entrusted loans issued by another local state-owned enterprise under the coordination of the local government and dealer associations. ‘Lenders’ decision to issue entrusted loans is not based on their judgment of the issuer's solvency, but on the coordination of the remaining entities', the media report said. (See ‘Economic Reference', May 26, 2017: ‘Sichuan Coal Group is mired in bond default'). Under the situation of AAA local state-owned enterprises defaulted (e.g., Yong Coal Holdings) and coal enterprises’ credit debt plummeted in November 2020, Shanxi State-owned Capital Operations Co., Ltd. which is wholly owned by the Shanxi provincial government, sent a letter ingesting that it would ‘advance intervention in risk disposal intervention', ‘mobilize provincial state-owned enterprises to form a strong pool of funds' and ‘ensure that there is no default on maturing bonds'. (See ‘Economic Report of the 21st Century', 14 November 2020: Shanxi: Ensuring that there is no default on maturing bonds of provincial enterprises).

12 This idea refers to the research on the risk of enterprise exit from Hopenhayn (1992).

13 The same test was done on the non-default bond sample in this paper. It was found that the effect of the official turnover on these key financial indicators was not significant or negative in the non-default sample, which further proved the threshold effect.

14 The estimation factor of the current mayor turnover for the last governor turnover is 0.1052, the t-value is 10.88, and F-value estimated in the first stage is 118.38. Limited to space, the full regression results for the first stage are not listed in Table 8.

Additional information

Funding

We acknowledge financial support from the General Project of National Natural Science Foundation of China under Grant number 71773081; the Sichuan University Education Foundation under Grant number SKSYL201812; the Sichuan University Education Foundation under Grant number 2021CXC11; and the Philosophy and Social Science Foundation of Sichuan Province under Grant number SC22B112; the Philosophy and Social Science Foundation of Sichuan Province under Grant number SC21TJ011.

Notes on contributors

Jin Zou

Jin Zou, Associate Professor, School of Economics, Sichuan University.

Shuxin Li

Shuxin Li, Research Assistant, School of Economics, Sichuan University.

Guoying Deng

Guoying Deng, Professor, School of Economics, Sichuan University.