Family control and corporate risk-taking in China: Does working capital strategy matter?

Figures & data

Table 1. Description of variables.

Variable level	Name of variable	Code	Variable descriptions
Dependent variable	Corporate Risk-Taking	Risk	Three years rolling average of Industry-adjusted Standard deviation of ROA
Independent variable	Family Control	Family	The actual controller's shareholding ratio indicates the strength of family control.
	Working Capital Strategy	WCS	WCS is the ratio of net working capital to total assets, while the net working capital is the difference between current operating assets and operating current liabilities.
Control variable	Firm Age	Age	Age = Ln (Firm’s establishment year +1)
	Return on equity	ROE	ROE = Net profit/ net assets at the end of the year
	Growth rate of sales revenue	Growth	Sales revenue growth rate = (this year's sales revenue amount – the previous year's sales revenue amount)/the previous year's sales revenue amount
	Asset-liability ratio (Leverage)	Lev	Total Liabilities/Total Assets
	Firm Size	Size	Scale = Ln (total assets)
	Industry	Industry	According to the < Guidelines for the Classification of Listed Companies in the Industry > of the SFC in 2013, ‘1’ is the manufacturing industry, and ‘0’ is the non-manufacturing industry.
	Year	Year	Dummy variable, 2010–2018

Source: CSMAR database.

Table 2. Descriptive statistics.

Variable	N	Mean	Standard deviation	Minimum	Maximum
Risk	7551	0.0236	0.0268	0.0001	0.4981
Family	7551	0.3428	0.1518	0.1000	0.8999
WCS	7551	0.3691	0.1873	−0.5070	0.9590
Age	7551	1.9742	0.6223	0.6931	3.3322
ROE	7551	0.0637	0.1540	−7.5876	0.9900
Growth	7551	0.4008	3.0627	−0.9532	140.2414
Lev	7551	0.3857	0.1982	0.0071	0.9890
Size	7551	21.7914	1.0488	17.8035	26.6525

Source: CSMAR database.

Table 3. Pairwise correlation.

Variables	Risk	Family	WCS	Age	ROE	Growth	Lev	Size	Industry
Risk	1
Family	−0.051***	1
WCS	−0.050***	0.135***	1
Age	0.118***	−0.338***	−0.295***	1
ROE	−0.195***	0.091***	0.101***	−0.026**	1
Growth	0.130**	0.012	−0.015	0.060***	0.025**	1
Lev	−0.028**	−0.070***	−0.436***	0.358***	−0.081***	0.057***	1
Size	−0.124***	0.027**	−0.245***	0.300***	0.156***	0.049***	0.483***	1
Industry	−0.075***	0.038***	0.051***	−0.189***	−0.033***	−0.035***	−0.178***	−0.127***	1

Note: Risk is three years rolling average of Industry-adjusted Standard deviation of ROA; Family is the actual controller's shareholding ratio indicates the strength of family control; WCS is the ratio of net working capital to total assets while, the net working capital is the difference between current operating assets and operating current liabilities; Age is Ln (Firm's establishment year +1); ROE is Net profit/ net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount - the previous year's sales revenue amount) / the previous year's sales revenue amount; Lev is Total Liabilities/Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Table 4. The impact of family control on corporate risk-taking.

Variables	Corporate risk-taking
	Coefficient	T-Value
Family	0.004**	1.981
Age	0.006***	11.188
ROE	−0.032***	−16.207
Growth	0.001***	11.892
Lev	−0.008***	−4.064
Size	−0.003***	−9.701
Constant	0.092***	12.257
Industry	Control
Year	Control
Observations	7551
R-Square	0.098
Adjusted R-Square	0.096
F-Value	54.333***

Note: The table presents the results of regression analysis for family control effect on corporate risk-taking. Risk is three years rolling average of Industry-adjusted Standard deviation of ROA; Family is the actual controller's shareholding ratio indicates the strength of family control; Age is Ln (Firm's establishment year +1); ROE is Net profit/ net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount – the previous year's sales revenue amount) / the previous year's sales revenue amount; Lev is Total Liabilities/Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Table 5. The impact of working capital strategy on corporate risk-taking.

Variables	Corporate risk-taking
	Coefficient	T-Value
WCS	−0.007***	−3.866
Age	0.006***	10.598
ROE	−0.031***	−15.843
Growth	0.001***	12.035
Lev	−0.010***	−5.117
Size	−0.003***	−9.549
Constant	0.099***	12.966
Industry	Control
Year	Control
Observations	7551
R-Square	0.099
Adjusted R-Square	0.097
F-Value	55.147***

Note: The table presents the results of regression analysis for working capital stability effect on corporate risk-taking. Corporate Risk-Taking is three years rolling average of Industry-adjusted Standard deviation of ROA; WCS is the ratio of net working capital to total assets while, the net working capital is the difference between current operating assets and operating current liabilities; Age is Ln (Firm's establishment year +1); ROE is Net profit/ net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount – the previous year's sales revenue amount) / the previous year's sales revenue amount; Lev is Total Liabilities/Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Table 6. The impact of family control and working capital strategy on corporate risk-taking.

	Corporate risk-taking
Family control	Weak		Strong
	Coefficient	T-value	Coefficient	T-value
WCS	−0.008***	−3.060	−0.007***	−2.908
Age	0.007***	8.692	0.005***	5.713
ROE	−0.025***	−10.455	−0.054***	−13.631
Growth	0.002***	9.860	0.001***	7.980
Lev	−0.009***	−3.183	−0.016***	−5.488
Size	−0.005***	−9.945	−0.001*	−1.920
Constant	0.126***	11.962	0.052***	4.695
Industry	Control		Control
Year	Control		Control
Observations	4019		3532
R-Square	0.116		0.092
Adjusted R-Square	0.113		0.089
F-Value	35.131***		23.857***

Note: The table presents the results of regression analysis for family control and working capital stability effect on corporate risk-taking. Corporate Risk-Taking is three years rolling average of Industry-adjusted Standard deviation of ROA; Family control is a dummy variable that takes a value of ‘0’ for weak family control and ‘1’ for string family control; WCS is the ratio of net working capital to total assets while, the net working capital is the difference between current operating assets and operating current liabilities; Age is Ln (Firm's establishment year +1); ROE is Net profit/ net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount - the previous year's sales revenue amount)/the previous year's sales revenue amount; Lev is Total Liabilities/Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Table 7. The impact of family control and working capital strategy on corporate risk-taking (robustness test-alternate proxy for corporate risk-taking).

	Corporate risk-taking (alternate proxy)
Family control	Weak		Strong
	Coefficient	T-value	Coefficient	T-value
WCS	−0.018**	−3.029	−0.016**	−2.967
Age	0.016***	8.862	0.011***	5.884
ROE	−0.056***	−10.357	−0.126***	−13.766
Growth	0.004***	9.989	0.002***	7.896
Lev	−0.019**	−3.137	−0.037***	−5.662
Size	−0.011***	−10.054	−0.002*	−1.855
Constant	0.289***	12.065	0.117***	4.663
Industry	Control		Control
Year	Control		Control
Observations	4019		3532
R-Square	0.094		0.117
Adjusted R-Square	0.090		0.114
F-Value	24.239***		35.494***

Note: The table presents the results of regression analysis for family control and working capital stability effect on corporate risk-taking. Corporate Risk-Taking is an alternate proxy of risk following the model of (Faccio et al., 2011); Family control is a dummy variable that takes a value of ‘0’ for weak family control and ‘1’ for string family control; WCS is the ratio of net working capital to total assets while, the net working capital is the difference between current operating assets and operating current liabilities; Age is Ln (Firm's establishment year +1); ROE is Net profit/ net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount - the previous year's sales revenue amount)/the previous year's sales revenue amount; Lev is Total Liabilities/Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Table 8. The impact of family control and working capital strategy on corporate risk-taking (robustness test-alternate proxy for working capital strategy).

	Corporate risk-taking
Family control	Weak		Strong
	Coefficient	T-value	Coefficient	T-value
WCS (Alternate Proxy)	−0.0003**	−2.097	−0.0001**	−2.128
Age	0.005***	5.938	0.007***	9.384
ROE	−0.055***	−13.832	−0.025***	−10.741
Growth	0.001***	7.991	0.002***	9.704
Lev	−0.014***	−5.003	−0.006**	−2.309
Size	−0.001*	−1.951	−0.005***	−9.923
Constant	0.048***	4.429	0.121***	11.658
Industry	Control		Control
Year	Control		Control
Observations	3532		4019
R-Square	0.091		0.115
Adjusted R-Square	0.087		0.112
F-Value	23.559***		34.767***

Note: The table presents the results of regression analysis for family control and working capital stability effect on corporate risk-taking. Corporate Risk-Taking is three years rolling average of Industry-adjusted Standard deviation of ROA; Family control is a dummy variable that takes a value of ‘0’ for weak family control and ‘1’ for string family control; WCS is an alternate proxy for working capital stability, i.e., the ratio of net working capital to operating income; Age is Ln (Firm's establishment year +1); ROE is Net profit/net assets at the end of the year; Growth is Sales revenue growth rate = (this year's sales revenue amount – the previous year's sales revenue amount) / the previous year's sales revenue amount; Lev is Total Liabilities / Total Assets; Size is Ln (total assets); Industry is a dummy variable that takes the value of ‘1’ for the manufacturing industries, and ‘0’ for non-manufacturing industries. ***, **, and * indicate that the significance level is 1%, 5%, and 10%, respectively.

Source: CSMAR database.

Faccio, M., Marchica, M.-T., & Mura, R. (2011). Large shareholder diversification and corporate risk-taking. Review of Financial Studies, 24(11), 3601–3641. https://doi.org/10.1093/rfs/hhr065

 Web of Science ®Google Scholar