1 Introduction

Privatization refers as the deliberate transfer of state ownership or government assets to private investors (Megginson and Netter 2001) has always been the subject of scholars’ interest. Prior research points out that privatization resulted in a reduction in state owned enterprises (SOEs) by selling state ownership, and also provide evidence regarding consequences of post-privatization (Jiang and Kim 2015; Cuervo-Cazurra et al. 2014; Guthrie 2012; Borisova and Megginson 2011; Boubakri et al. 2011). However, these studies on the impacts of privatization largely focus on its effects related to particular stakeholders in privatized firms—such as the state being a residual owner, creditors and shareholders—rather than focus on its broader societal welfare perspective. This is interesting given that the state, the ultimate shareholder of SOEs, is hypothetically the guardian of social welfare, and privatization of government/state assets is generally considered to be very costly to society.Footnote 1 Considering this interesting note, we aim to extend the existing privatization literature by examining the social performance of Chinese firms in context of secondary privatization.

There are a number of studies exploring CSR in the Chinese firms, which is typically characterized by the dominant and strong role of private and state ownership in general context (Bai et al. 2006; Li and Zhang 2010; Marquis and Qian 2014; Cumming et al. 2016). In our setting, we exploit the secondary privatization reform to investigate whether government withdrawal affects Chinese firms’ post-privatization CSR performance. We consider China because as the leading and influential emerging economy—China has been undergoing gradual privatization since 1990s (Peng 2000; Fan et al. 2007), though large scale secondary privatization took place after split share structure reform that was initiated in 2005 (Jiang and Kim 2015; Guthrie 2012). Prior evidences show that secondary privatization has reduced the ratio of state-owned firms (Usman et al. 2018; Yu 2013; Peng et al. 2016), introduced competitive environment (Liao et al. 2014) decreased direct government involvement (Mutlu et al. 2015) and changed firms’ priorities more closely linked with economic objectives (Naughton 2006; Peng et al. 2016). Based on prior evidences, we assume that privatization constrain government from making direct interference in firms’ important decisions, which may negatively affect CSR related objectives and performance. Alternatively, we make prediction based on the shareholder theory of CSR proposed by Friedman (1970), which predicts that privatized firms may exhibit lower CSR performance, as their prime social responsibility is to focus on shareholders’ value.

We further extend the debate on privatization and CSR performance by examining the interaction effect of firms’ political connection. Prior privatization studies suggest that other than to use the direct ways of maintaining control (i.e., government ownership), there are other indirect modes (i.e., political connections) allowing government to hold a tight grip on privatized firms’ strategic decisions (Boubakri et al. 2013; Fan et al. 2007; Chen et al. 2018). Privatization literature highlights that political connections condition the privatized firms’ performance outcomes (Boubakri et al. 2013; Borisova et al. 2012; Guedhami et al. 2014). Motivated by these studies, we examine that whether board’s political connection moderates the link between privatization and CSR performance. Studies point out that as connected firms receive preferential treatment from government, they need to comply with government policies (Liu et al. 2011; Ma and Parish 2006; Peng 2003). We therefore expect that loyal politically connected bureaucrats appointed in privatized firms’ board would facilitate residual state owners in making important decisions regarding CSR.

To investigate these predictions, our study uses large sample of Chinese listed companies—with data on CSR-ratings from Rankins agency. Using sample of 10804 firm-year observations over the period 2010–2015, we find that privatization of state ownership significantly and negatively affects firms’ CSR performance, as consistent with the notion of shareholder theory of CSR. In addition, because network-connection is one of significant driver of strategic options, we find that firms’ board member’s political connections weaken the negative relationship between privatization and CSR performance. Our findings contribute to the debate on privatization, suggesting that when government reduces its ownership in a firm, then it relinquishes its direct strong hold over such firms, which may affect their likelihood toward CSR engagement. We further contribute to the privatization studies by finding that political connections provide indirect ways to government involvement for making and implementation of social policies in privatized firms, and therefore, it affects the negative link between privatization and CSR.

Our study’s findings have some useful implications. For instance, our study may instruct academic scholars and policy makers about the social consequences of privatization and importance of government control through ownership as an important determinant of CSR. More specifically, our findings provide guidance that privatization is not beneficial from social perspective in emerging economies (i.e., China), it may undermine the firm’s social objectives. Taken together, this implies that the social outcomes expected to get from secondary privatization may not materialize under reduced government ownership. Considering that privatized firms show less likelihood toward social responsibilities, government being the guardian of society should enhance information transparency and treat all firms equally to get favorable social consequences. In addition, our study findings also provide insights for policy makers regarding the usefulness of political connection in context of privatization’s social effects, and can be used as a reference for other emerging economies.

The remainder of the paper is organized as follows. Section 2 provides the related literature, and Sect. 3 covers the sample, the variables and their measurement, and the model selection. Section 4 provides the results and a discussion, while Sect. 5 concludes.

2 Literature and hypotheses

2.1 Privatization and CSR

Drawing on the shareholder theory of CSR put forward by Friedman (1970), privatized firms may not show higher CSR performance than their counterparts, because the firms’ only social responsibility is generating revenues for its shareholders (Sternberg 2004; Marcoux 2003). In a similar context, Mansell (2013) points out, that the privatized firms’ directors are just employees, who have direct responsibilities of running business according to owners’ desire of promoting shareholders’ private objectives along with raising revenues. In this case, large investment in CSR is a misuse of financial resources, which should be utilized in profitable NPV projects (Friedman 1970). Given the opposing role of privatization in CSR, this theory implies that privatized firms may be expected to make less CSR investment and show poor performance.

Another reason that privatized firms might show weak CSR performance relates to relinquishment of government ownership and its reduced interference. According to prior literature, government’s presence on firm’s board is very necessary for promoting citizens’ welfare, especially in emerging countries like Brazil, China, Russia, India etc. (Naughton 2006; Marquis et al. 2011; Faccio and Lang 2002; Cuervo-Cazurra et al. 2014). It is well documented that firms with significant government ownership were more socially oriented before privatization in emerging economies (Bai et al. 2006; Li and Zhang 2010; Stan et al. 2014). However, under post-privatization view, privatized companies became more independent of state control, in which case their private objectives started to prevail (Liao et al. 2014; Wu et al. 2014; Zhang et al. 2010; Li et al. 2015; Inoue et al. 2013). As the anecdotal evidence on political view of privatization advocate that the new private shareholders de-emphasize political and government objectives that prevailed under high concentrated government ownership (Boycko et al. 1996; Boubakri et al. 2013). In particular, leading research on privatization point out that after government divestiture, the privatized firms’ financial performance improves significantly (D’Souza and Megginson 1999; Gupta 2005; Boubakri et al. 2005a, b; Bai et al. 2009), while it is detrimental from macro-level social perspective, such as employment, income distribution, and poverty (Chong and López-de-Silanes 2005; Birdsall and Nellis 2003; Bayliss 2002).

Taken together, these evidences and arguments suggest that government ownership relinquishment via privatization may cause distortions in important decisions. We thus predict that with privatization, we will observe a negative impact on firms’ CSR performance. Given this negative view, we draw the following hypothesis:

Hypothesis (H1)

Privatization is negatively associated with a firm’s CSR performance.

2.2 The role of firms’ political connection

A firm’s political connection may affect the link between privatization and CSR performance for the following reasons. First, the supporters of institutional theory postulate that external factors (institutional pressure) are the key drivers of strategic decisions including environmental protection and social investment (Colwell and Joshi 2013; Albertini 2017; Marquis and Qian 2014). Institutional pressure seems to be more pronounced for politically connected firms, because they maneuver in different and stringent regulatory environment (Wang et al. 2008). In similar vein, Reimsbach et al. (2018) argue that politically connected counterparts perceive comparatively strict monitoring from regulatory institutions that sensitize their strategic decisions. Given this increased attention and monitoring, we expect that politically connected board members of privatized firms will conform socially-constrained norms that are extensively obligated upon them by regulatory institutions and government.

Alternatively, prior literature suggest that political dependency reflects strategic choices that can be more likely to observe in private firms (Xin and Pearce 1996). Similarly, Li et al. (2008) and Puffer et al. (2010) argue that private owners depend heavily on political connections to cultivate advantageous link with the government in emerging markets (China and Russia)—for receiving favor in critical business transactions. Such evidence echo those of Kozhikode and Li (2012), who demonstrated that Indian private banks utilize political opportunities to protect its ownership and property rights. In general, institutional connections impact the firm’s future by mitigating uncertainties, receiving government contracts, and relaxation of regulatory constraints (Hillman 2005; Goldman et al. 2013; Chizema et al. 2015). These preferential treatments may come at a cost—as such the strategic actions (including social policies) of privatized firms will be more responsive to the benefits associated with political connections.

Finally, based on previous studies, government control is often not subject to relinquish even though government ownership is, as they often retain control on privatized firms’ important decisions by political networks (Bortolotti and Faccio 2006; Boubakri et al. 2008, 2011, 2013). Taken all together, this implies that political connection will facilitate government in privatized firms by taking favorable decisions, including those of CSR (Huang and Zhao 2016). Therefore, we may expect that board’ political connections in privatization’s context to be beneficial, would help in mitigating negative impacts of privatization on CSR performance. Based on the above arguments, we predict the following hypothesis:

Hypothesis (H2)

Board’s political connections weaken the negative relationship between privatization and CSR performance.

3 Methodology

3.1 Data sample

Our dataset is composed of Chinese A-share firms listed on the Shanghai and Shenzhen stock exchanges in the years from 2010 to 2015. We chose 2010 as a starting point in data sample because most listed companies received CSR ratings starting in 2010. We used a variety of data sources to collect data on CSR ratings, including companies’ annual reports, the China Stock Market and Accounting Research (CSMAR), and Rankins (RKS), a standardized agency established in 2007 that provides independent ratings to Chinese listed firms (Wu et al. 2014). We excluded firms observations for which data were missing on explanatory variables, ending with a sample of 10,804 firm-year observations.

3.2 Measures

3.2.1 Dependent variable

Our study used CSR performance as a dependent variable, measured as CSR ratings given by the RKS to a firm for the corresponding year. RKS provides ratings to listed firms based on their CSR indicators following the Global Reporting Initiative (GRI) approach. RKS usually uses seventy CSR indicators to evaluate a company’s CSR performance. These indicators are divided into three major sub-categories: CSR strategy and stakeholder involvement (14 items), technical sufficiency (11 items); and disclosure content (45 items). A composite social performance score is formulated through the weighted average of ratings scores of these three sub-categories (30%, 20%, and 50%, respectively). RKS’s CSR ratings have been shown to be reliable in studies of Luo et al. (2013), Marquis and Qian (2014), Zheng et al. (2014), Lau et al. (2016). The RKS ratings range from 0 to 100, with the higher ratings indicating better CSR performance.

3.2.2 Independent variable

A privatized firm refers to a firm in which state ownership has been reduced via privatization (Chen et al. 2018). We follow the prior privatization studies of Claessens and Djankov (2002); Bai et al. (2009); Chen et al. (2018); Boubakri et al. (2005a, b), and use a dummy variable (PRIV) for Chinese firms that equals 1 if a firm has been privatized its portion of state ownership through private sale following secondary privatization reform, and 0 otherwise.

3.2.3 Moderating variables

We considered the board’s political connections as a moderating variable because boards of directors play a distinct role in making strategic decisions. Directors with such backgrounds are more likely to lead the rest of the directors in regard to the challenges and related opportunities of political connections. Political connectedness refers to owners, directors or any senior management members who have recently or even formerly been officials of the military or the local or central government (Zhang 2017; Reimsbach et al. 2018; Höllerer 2013). This study uses the proxies for political connectedness as the total number of directors with political connections (Num_PC) and the proportion of directors with political connections (Pro_PC).

3.2.4 Control variables

Our study also includes several firm characteristics as control variables to check their effect on CSR performance. For instance, studies have shown that board characteristics have a strong influence on the flow of information and companies’ policies that may either improve or hinder companies’ CSR (Lau et al. 2016; Fernandez-Gago et al. 2018; Ali et al. 2019). Therefore, we add the board size (B_Size), measured as the number of directors who serve on the board and the number of independent directors (Ind_D) as used by Tan and Ma (2016). High levels of firm ownership by board members and institutional investors may also persuade the firm to engage in social objectives to build the firm’s reputation, so we control for board ownership (B_Own), which equal the proportion of shares owned by board members, and institutional holdings (Inst_Own), measured as the proportion of shares owned by institutional investors.

A firm’s economic characteristics can also impact its social investments (Khan et al. 2019). Specifically, highly profitable firms and large firms receive more stakeholder attention than other firms do. Therefore, we use return on assets (ROA) as a measure of profitability and firm size (Size), measured as the log of total assets. In addition, younger firms are more likely to invest in social objectives than their older counterparts (Marquis and Qian 2014), so we consider firm age (F_Age), measured as the number of years since a company was first listed on a stock exchange. Finally, we account for financial leverage (Lev), measured as total debt to total assets. Our study also includes year and industry dummies in its regression analyses to control for their fixed effects. The detail descriptions are given in (Table 1).

Table 1 Variables definition and details
Full size table

3.3 Descriptive statistics and correlations

Table 2 presents summary statistics for all the variables used in analysis. For the 10804 observations, the mean value of CSR score is 28.49, reflecting the sample companies’ average CSR performance. Table 2 also shows that on average, 10% companies among the study sample are identified where government have reduced their ownership following secondary privatization reform. Moreover, it indicates that each firm has two directors on average which are politically connected and their proportion is 25% to total directors serving on board.

Table 2 Descriptive statistics and correlation
Full size table

Furthermore, Table 2 highlights results for correlation among estimated variables. The coefficient values of all the independent variables are lower than 0.60, suggesting that there is no issue of multicollinearity and estimated variables are sufficiently independent. In addition, the Table 2 shows negative correlation between independent variable (PRIV) and CSR which provides preliminary evidence that privatization is negatively associated with firms’ CSR performance. In addition, political connection show positive correlation with CSR.

3.4 Statistical model

Given the change in firms’ ownership pattern after secondary privatization, a panel framework is more suitable to our main hypotheses. In this regard, we follow the previous studies (Faccio 2010; Boubakri et al. 2011; Chen et al. 2018) and use ordinary least square (OLS) regression as a baseline methodology to estimate the Eqs. (1) and (2).

$$CSR_{it} = \alpha + \beta_{1} PRIV_{it} + \beta_{2} PC_{it} + \mathop \sum \limits_{i = 1}^{n} \beta_{n} F\_Control_{it} + \varepsilon_{it}$$
(1)
$$CSR_{it} = \alpha + \beta_{1} PRIV_{it} + \beta_{2} PC_{it} + \beta_{3} PRIV_{it} \times PC_{it} + \mathop \sum \limits_{i = 1}^{n} \beta_{n} F\_Control_{it} + \varepsilon_{it}$$
(2)

where CSR indicates firm’s social performance scaled by RKS ratings; PRIV is representing privatization of state ownership in a firm; PC refers board’s political connection measured by number and proportion of politically connected directors; PRIV × PC shows interaction between privatization and proxies of politically connected board of directors; while F_Control refers firm level control variables.

4 Results and discussion

4.1 Privatization and CSR

Table 3 reports the findings from the estimation models. The Model 1 presents findings about our first hypothesis (H1), in which we assume that privatization is negatively related to CSR. Giving support to H1, we find that privatization (PRIV) tends to negatively influence companies’ CSR performance (β = − 1.69, p < 0.01). This result suggests that privatization relinquishes government control and direct interference in firms’ important decisions, which significantly and negatively affects the intensity of firms’ CSR activities. Previous studies suggest, that privatization reforms made firms more independent and irresponsible in terms of promoting public policies (Chong and López-de-Silanes 2005; Naughton 2006; Liao et al. 2014), since our findings imply that privatization undermine corporate social involvement that is detrimental for society. In addition, the private investors may overweight their personal motives (i.e., economic objectives) after privatization because of their sole responsibility (generating profit as much possible) as suggested by shareholder theory of CSR (Friedman 1970), which may deter privatized firms’ social motives. However, privatized firms’ less likelihood toward CSR may build negative reputation in the eyes of stakeholders (i.e., government, public/customers and media), in which case our findings instruct the investors and business owners.

Table 3 Influence of board’s political connections on the relationship between reduced government ownership and CSR performance
Full size table

4.2 The role of board’s political connection

Table 3 also present results regarding our second proposition, in which we predict that board’s political connections may significantly moderate the relationship between privatization and CSR. To test this proposition, we first examine the direct effect of board’s political connection on CSR. Doing so, we find positive and significant results in Model 2 and 3 for the board’s political connections using the number of politically connected directors (Num_PC) and their proportion on the board (Pro_PC) as its proxies. The results of Model 2 (β = 0.21, p < 0.05) and Model 3 (β = 1.75, p < 0.05) affirm that a higher number and proportion of politically connected directors enhances companies’ CSR performance. Our findings validate the previous evidences that political connections can improve a firm’s social performance because connected firms face stricter monitoring (Jiang et al. 2015; Reimsbach et al. 2018).

Next, we extend our analysis a step farther to examine whether boards’ political connection affects the link between privatization and CSR. Models 4 and 5 test the moderating effects of board members’ political connections (PRIV × Num_PC and PRIV × Pro_PC). In these models, the coefficients of both interactions are positive and highly significant (β = 0.67, p < 0.05, β = 7.04, p < 0.05), indicating that the board’s political connections weaken the negative link between PRIV and CSR, lending support to our second hypothesis (H2). These findings suggest that the board’s political connections provide indirect ways to government to be involved in setting corporate policies including those of social policies. Instead, board’s political connection can direct firm’s social behavior and objectives because they can easily challenge the mindset of other managerial staff for implementing social policies through political power. Furthermore, prior literature on privatization document that privatized firms’ political connection brings some preferential treatment for their firms (Boubakri et al. 2013; Borisova et al. 2012; Guedhami et al. 2014) that might come at a cost in terms of prioritizing political and social objectives. In that case, privatized firms should be aware of consequences regarding board’s political connection not only in sense of getting economic benefits but also to comply with government and political expectations.

The results of all control variables are reported in Table 3. The relationship of ROA to CSR is insignificant, but all remaining control variables have influential effects on CSR. For instance, board size (B_Size) generates significant negative effects on CSR, suggesting that larger boards always have conflicting goals in setting CSR policies. However, the positive coefficients for independent directors (Ind_D) on the board show that they are more cautious about using CSR activities to build trust and the firm’s image (Fernandez-Gago et al. 2018). The results of the board ownership (B_Own) and institutional ownership (Inst_Own) also show positive effects on the adoption of CSR activities, as these owners are rely on government-owned institutions (banks) for financial support. We also found that firm size (Size) positively influence CSR practices because large firms have the resources to invest in CSR projects. On the other side, the analysis shows negative influences of firm age (F_Age) and financial leverage (Lev) on CSR.

5 Additional tests

Although state-influenced markets (such as France, Brazil, South Korea, Russia, China, etc.) have reduced government ownership in firms through privatization reforms, but its hold remains on SOEs (Jiang and Kim 2015; Kang and Moon 2011; Stan et al. 2014; Mariotti and Marzano 2019). Motivated by large proportion of Chinese SOEs with heterogeneous institutional environment, we employed a sub-sample with 5051 firm-year observations to determine empirically whether government divestiture (PRIV) do matters for SOEs’ CSR performance.

The results of models 1–5 reported in Table 4 for state-owned firms. Model 1 highlights that PRIV negatively affects the social expenditures of SOEs (β = − 1.12, p < 0.05). This findings suggest that, reductions in governmental ownership changes firms’ priorities because of an inconstant managerial behavior of firms in state-influenced economies, so the extent of relinquishment of government ownership can be negatively related to the extent of SOEs’ social performance.

Table 4 Additional analysis
Full size table

Political connections are commonly found among board members in state-owned firms and exert strong influence on these firms. We therefore repeat out hypothesis 2 for the sub-sample and apply separate models 2–5, where we find positive and significant coefficients for political connections. Specifically, the positive results of interaction variables in models 4 and 5 (PRIV × Num_PC; β = 0.58, p < 0.05 and PRIV × Pro_PC; β = 5.61, p < 0.01, respectively) show that politically connected boards tend to weaken the negative association between privatization of government ownership and CSR performance. However, our findings conclude that political connections facilitate the government’s effort to align firms’ social objectives with those of the government and to improve the companies’ likelihood of engaging in CSR activities.

6 Robustness check

One of major concerns with privatization and government withdrawal is endogeneity of government decisions whether to privatize state ownership or not. Leading scholars of privatization, Megginson and Netter (2001) point out that government may keep inefficient firms and privatize the healthiest first making privatization “look good”, that can raise issue of self-selection bias and may mislead the main findings.

To address this issue, we use we use a Heckman (1979) 2-stage selection model. To predict the probability of firms to have reduction in state ownership, we use a probit model in the first stage, where we regress privatization (PRIV) on firm-level characteristics, including year and industry dummies. This method allow us to estimate the inverse Mills ratio known as “Lambda”. Turning to the second stage of Heckman, we add the estimated Lambda to the main regression with an independent and all control variables. However, Panel A of Table 5 only report the results of the second stage of Heckman for the sake of space. Using this model, we consistently find that privatization (PRIV) loads negative and significant impacts on firms’ CSR performance, but this negative relationship is weaker for firms with board’s political connection. In addition, Panel A shows that the result for Lambda is negative and insignificant, suggesting that there is no self-selection issue.

Table 5 Robustness
Full size table

We also apply other robustness tests to check the validity of our findings. We considered ordinary least squares (OLS) estimations for the main regression findings, but the OLS requirement that observations be independent cannot be satisfied in our study, as we used more than a single observation from each of sample firms, which may increase the probability to estimate biased results. Therefore, we employed OLS regressions by clustering the standard error by firm (Cluster-OLS) to tackle this concern. Furthermore, we expect that privatization of government ownership may not change the firm’s social behavior abruptly. Such firms might take some time to change their behavior that will negatively influence firms’ CSR performance. To predict this probability, we use the one-year lagged method for privatization (Lagged-PRIV) by replacing PRIV in the main regression model.

In Table 5, Panel B reports the Cluster-OLS results, while Panel C reports results for one-year lagged-PRIV results. These results confirm that privatization’s (PRIV) impact on firms’ CSR performance remains significant and negative. Similarly, the moderating variables (PRIV × Num_PC and PRIV × Pro_PC) play negatively moderating roles between PRIV and CSR, which is consistent with the main OLS regression results shown in Table 3. In sum, we find that our main evidence on firms’ CSR performance in context of privatization is consistent and robust to potential endogeneity concerns.

7 Conclusion

Based on earlier studies that have largely investigated the impacts of privatization in relation to specific firms’ stakeholder such as creditors and shareholders, this study goes a step further by demonstrating the impact of privatization on CSR and the moderating role of a politically connected board of directors. The results, obtained from a sample of listed firms of the second-largest economy—China over the time period 2010–2015, confirm that firms that experience a reduction in government ownership through privatization become less likely to engage in CSR. We also find that the political connections of board of directors weaken the negative relationship between privatization and CSR, suggesting that firms’ social expenditures result primarily from governmental and political involvement.

This paper contributes to the research studies on privatization by demonstrating that firms being privatized tend to show less likelihood toward CSR. Our main evidence implies that with privatization of government ownership, firms’ private objectives start to prevail, that resulting in negative effects on their CSR performance. Instead, perceptions that privatization of government ownership are expected to be costly for society are thus supported by our findings. Secondly, we extend the CSR literature by examining the moderating effect of boards’ political connection in context of privatization. Our evidence on its moderating effect suggest that, even after relinquishment of government ownership, politically connected directors will facilitate government in directing firms’ social behavior, thus will mitigate the negative impacts of privatization on CSR. Taken together, our findings imply that privatization may undermine firms’ CSR intensity, but privatized firms’ political connection can drive their social objectives due to higher imposed pressure from regulatory institutions.

Our study findings also provide insights for other developing countries, especially those suffering from unbalanced government intervention and treatment. Cognizant of inconstant behavior and uncertainties in transitional economies, like China, privatized firms may improve their CSR performance by imitating the positive role models. Instead, political connection is an important CSR driver in China that provides guidance to China’s policy makers. For instance, for improving the Chinese firms’ CSR performance, it would be useful to install government bureaucrats on board level, since our findings indicate that this is one of efficient approach to boost CSR performance. However, our findings should be carefully interpreted and utilized when making policy, as prior studies consider one of limitation of heavy political connections that it may reduce the efficiency of the privatized firms with respect of their financial performance (Fan et al. 2007; Tu et al. 2013).

Beyond this, there still remains that can be planned in future research. First, firms’ adherence to social responsibilities scaled by CSR ratings is considered a rigorous method of measuring CSR performance, but disclosing quality CSR reports can also be an important way to measure firms’ likelihood of pursuing CSR activities that can be examined in future studies. Future research can also replace CSR with earnings management, dividend policies and governance policies using the same explanatory variable. Second, future studies may develop more complex models that explore not only the role of firm-level determinants but also the market level factors, such as market structure (Kim et al. 2019), which may also affect the social outcomes of privatized and public firms. Finally, we focused on China’s institutional context, but the objective of this research can be studied in other institutional settings, which would help to generalize this research’s findings.