1. Political uncertainty and cash holdings: Evidence from China
Nianhang Xu, Qinyuan Chen, Yan Xu , Kam C. Chan
《 Journal of Corporate Finance》
pp , OCT. 2016
By: 张莉

2. Contents Introduction Research Objectives & Hypothesis1
Introduction 2
Research Objectives & Hypothesis 3
Research Methodology 4
Empirical Analysis 5
Summary & Conclusion

3. 01 / Introduction Political connections
Political connections play an important role in many economic activities
They are sometimes very good for a firm
Political connections can sometimes harm a firm, especially in emerging markets with weak formal institutions and high levels of corruption

4. 02 / Research Objectives & Hypothesis
Main objective: to examine the relation between cash holdings and political uncertainty in China
Hypotheses
H1A. : During a period of political uncertainty, a firm holds more cash.
H1B. : During a period of political uncertainty, a firm holds less cash.

5. 03 / Research Methodology
Data
CSMAR
Ranging 1998 to 2014.
23,955 firm-year-official observations
Variables
Dependent Variables:
Cash holdings: the amount of cash and cash equivalents divided by total assets net of cash and cash equivalents
Excess return: the firm's stock return minus a benchmark return

6. 03 / Research Methodology
Variables
Independent Variables:
Political uncertainty: a dummy variable, INDUCTION, For each firm-year-official observation, INDUCTION takes a value of one when the city where the firm's headquarters are located experiences a government official turnover (i.e., a mayor or community party secretary is newly appointed)and zero otherwise
Moderating factors: politically connected executives、SCALE to capture a firm‘s importance in the local economy、a firm’s change in corporate debt(DD) to capture the effect of its leverage on political extraction risk、city‘s fiscal situation to proxy for expropriation risk。
Variables that proxy for twin agency conflicts: the control-ownership wedge, analyst coverage, and auditor quality

7. 03 / Research Methodology
Model for Analysis

8. 04 / Empirical Analysis
Regression Analysis---Cash holdings & political uncertainty
For robustness, the paper uses six different empirical models to study the impact of political uncertainty on cash holdings
Consistently across all six models, the coefficients of INDUCTION are negative and significant at the 1%, 5%, or 10% level, suggesting that firms hold less cash during periods of political uncertainty. These findings support H1B, not H1A

9. 04 / Empirical Analysis
Regression Analysis---Expropriation risk, political uncertainty, and cash holdings
(a)firms' executives are politically connected (PC = 1), (b) firms are larger (SCALE = 1), (c) firms have larger changes in debt level(DD = 1); or (d) firms are located in cities with better fiscal situations (DEFICIT = 1)

10. 04 / Empirical Analysis
Regression Analysis---Expropriation risk, political uncertainty, and cash holdings
The coefficients of PC ∗ INDUCTION, SCALE ∗ INDUCTION, DD ∗ INDUCTION, and DEFICIT ∗ INDUCTION are positive and significant at the 5% or 10% level ; suggesting that low expropriation risk firms hold more cash than those of high expropriation risk firms.
These expropriation risk factors moderate the relation between political uncertainty and cash holdings.

11. 04 / Empirical Analysis
Regression Analysis---Cash holdings & origin of the new appointee
The paper expects firms to face less political uncertainty if the new appointee is from the same city rather than from a different one.
Add ORIGIN as a dummy variable with a value of one if the new appointee is from the same city to examine the prediction. ORIGIN ∗ INDUCTION captures the incremental impact of having a same city appointee on a firm's cash holdings.
The coefficients of ORIGIN*INDUCTION are positive and significant at the 5% or 10% level. That is, when the new appointee is from the same city, the adverse impact of political uncertainty on cash holders is alleviated.

12. 04 / Empirical Analysis
Regression Analysis---Cash holdings & Twin agency problem
Use the control-ownership wedge (WEDGE), analyst coverage (AC), and auditor quality (BIG4) to examine the moderating effect of twin agency problem.
The coefficient of WEDGE ∗ INDUCTION is negative and significant at the 10% level . The coefficients AC ∗ INDUCTION, and BIG4 ∗ INDUCTION in are positive and significant at the 5% and 10% level.
Overall, when a firm has large twin agency problems, they reduce more cash than those of having small twin agency problems.

13. 04 / Empirical Analysis
Regression Analysis---Market value of cash holdings
Four different models using various combinations of year and industry effects
The coefficients of the interaction variable (INDUCTION ∗ ΔCASH t ) are
negative and significant at the 10% level in three models, suggesting changes in cash holdings during periods of political uncertainty are negatively correlated with firm excess stock returns. That is, the market value of cash decreases with political uncertainty,

14. 05 / Summary & Conclusion
During a new city government official's first year, a firm holds less cash, which is consistent with the grabbing hand hypothesis of politician.
A firm facing a higher political extraction risk keeps significantly less cash in a period of political uncertainty if (a) it does not have politically connected executives, (b) it is smaller, (c) it has less debt, or (d) it is located in a city with high fiscal deficit.
The appointment of a new official from a different (the same) city means greater (less) political uncertainty for a firm and, hence, there is a need (no need) for the firm to decrease its cash holdings.
The market value of cash holdings is significantly negative during periods of political uncertainty, suggesting that the market value of cash to a firm is lower due to the potential loss of cash to political extraction.

15. Thanks!