Profit or legitimacy? What drives firms to prioritize social stakeholders?

The definition of a private firm in China is different from that of most developed countries. In most developed countries, private firms refer to unlisted firms and public firms are publicly listed firms. In China, private firms are in contrast of state-owned enterprises. Private firms refer to firms whose largest shareholders are individuals or private legal entities. They can be publicly listed. In 2008, for example, 40% of domestic publicly listed firms were private firms and 60% of them were state-owned enterprises. Studies of Chinese private firms using publicly listed firms’ data are quite common (e.g., Li and Liang 2015; Marquis and Qian 2014).

The Chinese Securities Regulatory Commission (CSRC) requires firms to complete a prelisting tutoring period before they apply for initial public offering (IPO) with the CSRC. During the pre-IPO tutoring period, certified securities services institutions provide tutorials to firms to ensure they meet all IPO requirements. The process typically takes about 3 months, but occasionally can be longer. Tutorials may include but are not limited to (a) training owners and executives on regulations related to listing on the stock market, the Company Law, and the Security Law; (b) helping firms build a complete structure and process for shareholder meetings, boards of directors, and supervisory boards; (c) helping firms establish their financial reporting system, governance structure, and other internal control mechanisms; (d) helping firms establish a system for information disclosure required for publicly listed firms; and (e) advising firms on how to regulate their relationships with shareholders and related stakeholders.

We also collected data on firm age and founder’s management experience. However, these two variables are highly correlated with other variables. Firm age is highly correlated with firm size (0.42), and founder’s management experience is highly correlated with firm size (0.36), firm age (0.56), and founder’s age (0.62). To avoid multicollinearity problem, we did not use firm age and founder’s management experience as control variables. We also conducted analysis including these two variables in all regression models. The results are similar.