The term social network refers to a collection of actors and their relationships (e.g., friendship and communication; Kilduff & Tsai, 2003). According to Fracassi and Tate (2012), directors’ social network in enterprises is one of the most common social networks, and the behaviors and systems of individual directors are embedded in the social network. Further, the behaviors of directors are impacted by the social network they are in (Granovetter, 1985). Directors who serve on the board of two or more companies at the same time are called connected directors (Macaulay, Richard, Peng, & Hasenhuttl, 2017). Among these directors there are formal or informal connected relationships that form a network.

There is an ongoing need to examine the association of corporate social responsibility with the leaders of the enterprise (Choi, Ullah, & Kwak, 2015). However, the focus of researchers in previous studies has been on how a network of connected directors influences corporate decisions and firm values (Larcker, So, & Wang, 2013). The question of how a network of connected directors influences corporate social performance is still largely an underexplored area (Macaulay et al., 2017). Shan, Gan, and Zheng (2008) suggested that philanthropy should attract more attention from society and scholars. Thus, in the current study we analyzed the effects of connected directors’ social network centrality on the charitable donation of the enterprises, taking into account the properties of the enterprise and market competition. We expected that our study would contribute to a more accurate understanding of how directors’ resources derived from their social network centrality impact organizational behavior in China.

Literature Review and Hypotheses

Directors’ Social Network and Philanthropy

According to stakeholder theory, the purpose of a firm is to create as much value as possible for stakeholders (Freeman, 1979), including employees, investors, and communities (Kreng & Huang, 2011). Firms with effective management of stakeholder demands obtain some advantages over competitors (Macaulay et al., 2017). However, philanthropy does not mean altruism (Gautier & Pache, 2015). In the Chinese business context, philanthropy is a strategic behavior performed by companies in order to obtain financial and political resources from the government (Shan et al., 2008). The Chinese government hopes to ease pressure on its finances through the donations made to charities by enterprises and, thus, there are incentives (such as tax incentives) to encourage enterprises to donate. 

According to political cost theory advanced by Jensen and Meckling (1976), larger firms face more government scrutiny than do smaller firms. The higher visibility of firms causes them to become targets of greater regulatory actions by government and vehicles for wealth transfer (Zimmerman, 1983). Thus, corporates with higher visibility incur higher political costs. As corporate philanthropy is not for the purpose of altruism and the level of donation is largely influenced by the relevant government regulations and concessions, donations become part of the total political cost borne by firms. Connected directors bring real or potential resources to an enterprise through their network of social relationships (Nahapiet & Ghoshal, 1998). The more connected the directors’ network, the faster the corporate achieves acquisition and transmission of resources (Freeman, 1979), and the higher the visibility of the corporate. It is easier for the government to lobby those corporates with higher visibility for donations, because in China, due to taxation and business management relationships, there are business channels between government workers and business managers. Accordingly, we formed the following hypothesis:
Hypothesis 1: Chinese corporates whose directors have denser networks than others have will be more likely to donate to charity. 

Properties of Enterprises and Philanthropy

In China, state-owned enterprises may make fewer charitable donations because they are often expected by the government to undertake responsibility for social stability, such as curbing inflation or reducing high unemployment rates, and  a lower level of charitable donation is regarded as a compensation for their contribution to social welfare (Wu, Wang, Luo, & Gillis, 2012). Therefore, Chinese state-owned enterprises are less politically pressured and more politically protected than are non-state-owned enterprises, so that they are not sensitive to political costs. On the contrary, non-state-owned enterprises in China may face more government scrutiny compared to the state-owned enterprises and, as a result, they tend to make more charitable donations in order to avoid being regulated. Thus, we formed the following hypothesis:
Hypothesis 2: Chinese enterprises that are not state-owned whose directors have denser networks than others have will be more likely to donate to charity, but this will not be the case in state-owned enterprises.

Moderating Effect of Market Competition

Donations to charity made by Chinese non-state-owned enterprises are affected by the economic environment. Enterprises in monopolistic industries have been protected by government regulations that apply to their entry into the industry, whereas for enterprises in competitive industries, it is important to win government support by maintaining a good relationship with the government. When the enterprise faces fierce competition for its product in the marketplace, managers would be happy to pay when they are lobbied for donations to charity, and the government would then give them priority for financing, project approval, and tax incentives. Therefore, those enterprises that face fierce market competition tend to incur more political costs and make more “voluntary” donations. Thus, we proposed the following hypothesis:
Hypothesis 3: For non-state-owned enterprises in China, market competition will positively moderate the relationship that we have hypothesized above between directors’ centrality in a network and the enterprise’s level of charitable donation.

Method

Procedure

We obtained data from Shanghai and Shenzhen A-share listed companies in China from 2003 to 2015. We selected data according to the following criteria: (a) exclude financial companies, (b) eliminate listed companies that are specially treated or warned of delisting risk by stock exchanges, and (c) eliminate listed companies that lack data. The data were mainly from the Guotaian and Wind Financial Terminal databases, which had been 1% winsorized, that is, for a given quantile limit (such as 1%), the part that exceeds the upper and lower bounds is replaced by a quantile. We used Pajek to calculate centrality and Stata14 to do other data processing. Part of the directors’ social network is shown in Figure 1. The entirety of the directors’ social network is too dense to show clearly here.

Measures

Dependent variables. The dependent variable of this paper was charitable donation, and the data came from the information on the listed company’s financial statements. In order to eliminate the influence of the outliers on the estimation, we added 1 to the charity donation data of the listed company and then took the logarithm to reflect charitable donation of enterprises. Thus, the greater the value, the higher the level of charitable donation.

Independent variables. Closeness centrality measures the distance between the directors of a company; the more central the director is, the closer the centrality (Freeman, 1979). Therefore, we used closeness as the variable to assess centrality in the network. Using the formula shown below (Freeman, 1979), we calculated the closeness centrality of all the directors in each company, taking the average value as the measurement index of the directors’ social network centrality in the company where i is a director, j is a director other than i in one year, and g is the total number of directors.

Moderator. We selected state-owned enterprises (SOEs) and market competition as moderators. Market competition was measured using the Herfindahl-Hirschman Index (HHI), which is the sum of the squares of the percentage of sales of the industry’s competitors (Tong, Feng, & Lu, 2012). The smaller the sum of the squares for this index, the fiercer the market competition.

Control variables. We controlled for the follow variables: top three executives’ total compensation (executive compensation), market-oriented degree in different regions (free market), political connections, number of executives, whether the positions of chairman and general manager were concurrently held by one person (chief executive officer; CEO), total asset net profit rate (return on assets; ROA), asset–liability ratio (leverage), total number of assets (firm size), sum of the top 10 shareholders (Top 10), listing period (firm age), annual impact, and industry impact.  

Models. Our models are as follows:
Donation = α + β₁ × Closeness centrality + β₂ × SOE
+ β₃ × Executive compensation + β₄ × Executive number
+ β₅ × CEO + β₆ × ROA + β₇ × Leverage + β₈ × Firm size
+ β₉ × Top 10 + β₁₀ × Firm age + β₁₁ × Year + β₁₂ × Industry + ε

Model 1
Among them, α is the constant term, βi(i = 1, 2, 3, ... 12) is the coefficient of variables, and ε is the random error term.
Donation = α + β₁ × Closeness centrality + β₂ × Executive compensation
+ β₃ × Executive number + β₄ × CEO + β₅ × ROA +  β₆ × Leverage
+ β₇ × Firm size + β₈ × Top 10 + β₉ × Firm age +  β10 × Year
+ β11 × Industry + ε

Model 2
Among them, α is the constant term, βi(i = 1, 2, 3, ... 11) is the coefficient of variables, and ε is the random error term.
Donation = α + β₁ × Closeness centrality + β₂ × HHI
+ β₃ × Closeness centrality × HHI + β₄ × Free market
+ β5 × Political connection + β₆ × Executive number
+ β7 × CEO + β₈ × ROA + β₉ × Leverage + β10 × Firm size
+ β₁₁ × Top 10 + β₁₂ × Firm age +  β13 × Year + β14 × Industry + ε

Model 3
Among them, α is the constant term, βi(i = 1, 2, 3, ... 14) is the coefficient of variables, ε is the random error term.

Results

The means, standard deviations, minimums, medians, and maximums are shown in Table 1.

As shown in Table 2, closeness centrality had a positive impact on charitable donation; however, the relationship between the two variables remained to be further verified by regressions. The maximum variance inflation factor was 1.60, which excluded the possibility of multicollinearity.

For all regressions in this study we used cluster-robust standard errors. In Hypothesis 1, we predicted that corporates with directors whose networks were denser would be more likely to donate to charities. Results in Model 1 of Table 3 show that the relationship between closeness centrality and charitable donation was positive. Therefore, Hypothesis 1 was supported.

To test Hypothesis 2, we classified the enterprises into two groups: state-owned and non-state-owned. As shown in Model 2 of Table 3, we found that the effect of closeness centrality on charitable donation in state-owned enterprises was nonsignificant, whereas in non-state-owned enterprises, closeness centrality positively predicted charitable donation. Thus, Hypothesis 2 was verified.

To test Hypothesis 3, we took non-state-owned enterprises as the sample and added interactive items to the model. As shown in Model 3 of Table 3, the interaction term was negative. The smaller the Herfindahl-Hirschman Index, the fiercer the market competition was, so that market competition positively moderated the relationship between closeness centrality and charitable donation. Thus, Hypothesis 3 was fully supported.

The results of robustness analysis are shown in Table 4. We used donation plus 1 divided by total sales, and then took the logarithm as the donation level. The results show that our findings remained unchanged.

Discussion

We examined the influence of directors’ social network centrality on corporate charitable donation. As predicted, we found that in non-state-owned enterprises corporates with denser networked directors would be more likely to donate, but this phenomenon was not evident in state-owned enterprises. A possible reason for this difference might be that state-owned enterprises are less politically pressured and more politically protected than are non-state-owned enterprises. We also found that those non-state-owned enterprises in an industry where there is fierce market competition were inclined to make more donations with the aim of setting up a good relationship with the government (Shan et al., 2008). Thus, we believe that the Chinese government should not only lobby resource-rich enterprises to exercise social responsibility, but should also break the monopoly power of the market and promote effective competition.

Our study contributes to the current literature in several ways. First, our findings provide a new perspective on social networks. It has long been considered that directors’ social network is associated with firm governance (Fracassi & Tate, 2012). However, to our knowledge, the relationship between the directors’ network and corporate charitable donation had not been examined. Our research findings are the first to provide concrete evidence that directors’ network centrality positively predicts corporate donation to charity.

Second, by introducing the concept of political cost, we unveiled the mechanism of corporate charitable donation. Although researchers have pointed out that philanthropy does not mean altruism (Gautier & Pache, 2015), our study is the first in which donation to charity has been regarded as a political cost. Our findings show that in China being a state-owned enterprise and market competition both acted as mediators in the directors’ social network centrality–charitable donation relationship, which implies that making donations to charity is just a political cost that companies pay to get more government support. The results further confirm that philanthropy is self-interested.

There are still several limitations in our study. The first is that we used only two formulas for the calculation of donations. Second, we found that the amount of donations made by state-owned enterprises was not affected by the directors’ social network centrality, and may be affected by other factors; for example, sometimes the government may directly designate donations for state-owned enterprises. We did not offer detailed evidence to support this conjecture and we have only suggested possible reasons according to findings in the literature. A more specific investigation could be made in future research.

Choi, S. B., Ullah, S. M. E., & Kwak, W. J. (2015). Ethical leadership and followers’ attitudes toward corporate social responsibility: The role of perceived ethical work climate. Social Behavior and Personality: An international journal, 43, 353–365. https://doi.org/f69h8x

Fracassi, C., & Tate, G. (2012). External networking and internal firm governance. The Journal of Finance, 67, 153–194. https://doi.org/fzg8jb

Freeman, L. C. (1979). Centrality in social networks conceptual clarification. Social Networks, 1, 215–239. https://doi.org/bx3m36

Gautier, A., & Pache, A. C. (2015). Research on corporate philanthropy: A review and assessment. Journal of Business Ethics, 126, 343–369. https://doi.org/gc8swc

Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91, 481–510. https://doi.org/dbkm9k

Jensen, M. C., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3, 305–360. https://doi.org/hjg

Kilduff, M., & Tsai, W. (2003). Social networks and organizations. London, UK: Sage.

Kreng, V. B., & Huang, M.-Y. (2011). Corporate social responsibility: Consumer behavior, corporate strategy, and public policy. Social Behavior and Personality: An international journal, 39, 529–541. https://doi.org/bv9bzq

Larcker, D. F., So, E. C., & Wang, C. C. Y. (2013). Boardroom centrality and firm performance. Journal of Accounting and Economics, 55, 225–250. https://doi.org/csg2

Macaulay, C. D., Richard, O. C., Peng, M. W., & Hasenhuttl, M. (2017). Alliance network centrality, board composition, and corporate social performance. Journal of Business Ethics. Advance online publication. https://doi.org/csg3

Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. The Academy of Management Review, 23, 242–266. https://doi.org/bxkbk8

Shan, L., Gan, L., & Zheng, T. (2008). Corporate contributions and economic motivation [In Chinese]. Economic Research, 43, 51–61.

Tong, Y., Feng, H., & Lu, D. (2012). Market concentration, control characteristics, and disclosure of internal control [In Chinese]. Accounting Research, 6, 63–68+95. https://doi.org/csg4

Wu, L., Wang, Y., Luo, W., & Gillis, P. (2012). State ownership, tax status and size effect of effective tax rate in China. Accounting and Business Research, 42, 97–114. https://doi.org/csg5

Zimmerman, J. L. (1983). Taxes and firm size. Journal of Accounting & Economics, 5, 119–149. https://doi.org/ch3kq8

This work was supported by the Project of the National Natural Science Foundation of China (71472088) 

and the Jiangsu Applied Social Science Research Project &ldquo

Transformation of Accounting Talent in the Background of Internet&rdquo

(17SCQ-04).

Guanghua Xu, School of Economics and Management, Nanjing University of Science and Technology, Xiaolingwei 200, Nanjing, Jiangsu, 210094, People’s Republic of China. Email: [email protected]