INTRODUCTION
Corporate financial fraud, defined as the ‘intentional misrepresentation of amounts or disclosures in the financial statements' (Apostolou, Hassell, & Webber, Reference Apostolou, Hassell and Webber2000: 181), can result in negative and sometimes ruinous economic outcomes (Baucus & Baucus, Reference Baucus and Baucus1997; Davidson, Worrell, & Lee, Reference Davidson, Worrell and Lee1994; Karpoff & Lott, Reference Karpoff and Lott1993). To address the issue of financial fraud, governments around the world have established agencies, such as the Securities and Exchange Commission (SEC) in the United States, to investigate corporate financial fraud and punish convicted firms. Fraud punishment levied by such government agencies would lead to stigma that may result in the loss of an organization's social approval and legitimacy and even threaten organizational survival (Carberry & King, Reference Carberry and King2012; Tracey & Phillips, Reference Tracey and Phillips2016). Considering the negative consequences of the threats resulting from public fraud punishment, organizational leaders will attempt to eliminate such threats (Carberry & King, Reference Carberry and King2012; Wang, Reference Wang2010). For example, a firm punished by the government (convicted firm hereafter) can signal the efforts it is making to refrain from further fraud occurrence, or it can decouple the organization from stigma by blaming a particular employee, team, or unit (Elsbach & Sutton, Reference Elsbach and Sutton1992; Suchman, Reference Suchman1995).
However, given the generally nontransparent political and regulatory environment in emerging markets (Marquis & Raynard, Reference Marquis and Raynard2015), merely reacting to public criticism and stigma caused by fraud punishment may not be enough. In order to remove potential threats in the future, convicted firms need to use institutional strategies to shape their institutional environment purposefully and strategically after fraud punishment. Nevertheless, few studies have explored the ways in which organizations strategically shape their institutional environment in emerging markets (Marquis & Raynard, Reference Marquis and Raynard2015). This study aims to fill this gap by focusing on China, which is the largest emerging market, and by examining firms’ institutional strategy after being punished by the Chinese government for fraudulent activities.
Although existing studies have examined firms’ reactions to their corporate wrongdoings being made public (e.g., Arthaud-Day, Certo, Dalton, & Dalton, Reference Arthaud-Day, Certo, Dalton and Dalton2006; Zavyalova, Pfarrer, Reger, & Shapiro, Reference Zavyalova, Pfarrer, Reger and Shapiro2012), little attention has been paid to reaction to punishment from key stakeholders such as the government. Given the extensive government involvement in firms’ operations in emerging markets such as China (Du & Luo, Reference Du and Luo2016; Wright, Filatotchev, Hoskisson, & Peng, Reference Wright, Filatotchev, Hoskisson and Peng2005), punishment by the government can result in serious and far-reaching consequences for firms, including legitimacy loss and even threat to survival. In such a context, firms’ responses become critical. As such, this study focuses on the relationship between fraud punishment by the government and subsequent corporate philanthropy, the latter of which is often used as a strategy to please the government and the public in China (Li, Song, & Wu, Reference Li, Song and Wu2015).
We adopt sociological institutional theory (institutional theory hereafter) (DiMaggio & Powell, Reference DiMaggio and Powell1983; Meyer & Rowan, Reference Meyer and Rowan1977; Scott, Reference Scott1995) as our overarching framework/theory to explain the philanthropic behavior of firms after they have been punished for financial fraud by the government. Institutional theory suggests that every organization tries to obtain or maintain their legitimacy (Scott, Reference Scott1995). As such, institutional constituents are able to influence organizations’ actions by conferring or withdrawing legitimacy (DiMaggio & Powell, Reference DiMaggio and Powell1983; Meyer & Rowan, Reference Meyer and Rowan1977; Parsons, Reference Parsons1951; Scott, Reference Scott1995). Drawing on insights from institutional theory (DiMaggio & Powell, Reference DiMaggio and Powell1983; Scott, Reference Scott1995), we propose that as the largest institutional constituent in China, the government keeps listed firms in check by punishing those committing fraud. Since being punished by the government often incurs a loss of legitimacy, in order to prevent further punishment from various constituents as well as to regain legitimacy, the convicted firms may use a legitimacy regaining strategy, which is one of the key institutional strategies, to shape their institutional environment purposefully (Suchman, Reference Suchman1995). In this study, we focus on philanthropic contributions as an important legitimacy regaining strategy in emerging markets. Specifically, we propose that convicted firms would use corporate philanthropy to regain legitimacy from the public and the government after being punished for fraud. As punishment severity gives some indication of how much legitimacy the convicted firms are likely to lose (Wang, Reference Wang2010), we hypothesize punishment severity is positively related to the increase of corporate philanthropy among these firms.
Furthermore, we propose that the convicted firms’ media visibility and political embeddedness, which increase their vulnerability to the legitimacy loss from the public and the government respectively, can strengthen their motivation to regain legitimacy, thereby influencing the relationship between punishment severity and corporate philanthropy increase. Considering that high media visibility can increase the attention and scrutiny the convicted firms receive from the public, leading to a greater loss of legitimacy, legitimacy regaining strategies are more needed for convicted firms with high media visibility. Moreover, political embeddedness enables convicted firms to gain more attention from the government and to better understand the implicit roles of governmental punishment, leading to more anxiety and thereby strengthening their motivation to regain legitimacy from the government through contributing more. As such, we propose that variables representing firms’ political embeddedness – dominant state ownership and national political appointment – can strengthen the influence of punishment severity on corporate philanthropy increase.
We tested our hypotheses using a sample of Chinese-listed firms that were punished for corporate financial fraud in the ten years from 2004 to 2013 and employing the two-stage Heckman selection model (Heckman, Reference Heckman1979) as the method. Our study has important contributions and implications. First, it contributes to the institutional theory literature by enriching our knowledge about the institutional strategies that firms use in emerging markets. While extensive research has focused on how institutions shape organizations’ behavior, less attention has been paid to examining the ways in which organizations purposefully and strategically shape their institutional environment, especially in emerging markets (Marquis & Raynard, Reference Marquis and Raynard2015). We extend the research on institutional strategies by focusing on corporate philanthropy as a legitimacy regaining strategy in emerging markets. Specifically, we provide evidence that unlike their counterparts in developed markets that may engage in lobbying, campaign contributions, and public relations campaigns (Mellahi, Frynas, Sun, & Siegel, Reference Mellahi, Frynas, Sun and Siegel2016), convicted firms in China may simply use corporate philanthropy to regain legitimacy from the public and the government in the wake of fraud punishment. Moreover, we also contribute to the institutional theory literature by exploring the underlying mechanisms leading to convicted firms’ different responses to the institutional stimulus by the government.
Second, we also contribute to the literature on political embeddedness by demonstrating that political embeddedness would make convicted firms’ more sensitive to governmental punishment in the Chinese context. Unlike prior studies that predominantly focus on the benefits of political embeddedness (e.g., Haveman, Jia, Shi, & Wang, Reference Haveman, Jia, Shi and Wang2017; Sheng, Zhou, & Li, Reference Sheng, Zhou and Li2011), we provide evidence that in emerging markets with immature formal institutions and powerful government, such as China, political embeddedness is also associated with significant costs, thus enriching our understanding of political embeddedness in emerging markets.
Finally, our study also has important practical implications. Our results show that although corporate philanthropy plays a strategic role in shaping the institutional environment in emerging markets such as China, not all convicted firms use such a strategic tool to regain legitimacy in the wake of fraud punishment. This implies that firms in these markets can employ their knowledge about the institutions to strategically shape their institutional environment. Another implication is that regulators should be alert to the donations by convicted firms, especially highly visible ones, and differentiate symbolic compliance from substantial compliance.
THEORETICAL BACKGROUND AND HYPOTHESES
Institutional Theory and Institutional Strategies in Emerging Markets
Institutional theory focuses on how social pressure toward conformity shapes organizations’ actions (DiMaggio & Powell, Reference DiMaggio and Powell1983; Meyer & Rowan, Reference Meyer and Rowan1977; Scott, Reference Scott1995). It suggests that organizations seek approval from institutional constituents, and thus they are susceptible to social influence. One vital component of institutional theory is legitimacy, which refers to ‘a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions' (Suchman, Reference Suchman1995: 574). Institutional theory assumes that firms seek legitimacy from institutional constituents to secure resources for survival (Deephouse, Reference Deephouse1999; Scott, Reference Scott1995). One of the main theses of institutionalists is that every organization tries to enhance or protect its legitimacy (Scott, Reference Scott1995). Legitimacy concerns motivate firms to adopt practices that are socially desirable (Deephouse, Reference Deephouse1999).
Although extant institutional literature mainly focuses on how institutions shape social conformity through legitimacy, there has been growing recognition that organizations can use institutional strategies to purposefully and strategically shape their institutional environment (Marquis & Raynard, Reference Marquis and Raynard2015). The term institutional strategy was first considered by Lawrence (Reference Lawrence1999). Based on the work of Lawrence (Reference Lawrence1999), Marquis and Raynard (Reference Marquis and Raynard2015) developed this concept and defined it as ‘the comprehensive set of plans and actions directed at leveraging and shaping socio-political and cultural institutions to maintain or improve an organization's competitive position' (Marquis & Raynard, Reference Marquis and Raynard2015: 284). Different from earlier perspectives that treated institutions as top-down pressures constraining social and organizational behavior, institutional strategy research highlights the strategic interactions between organizations and their institutional environments (Lawrence, Reference Lawrence1999).
Although the implementation of economic liberalization and privatization policies has promoted economic transition, the government in emerging economies still has a high degree of control over key resources (Peng & Luo, Reference Peng and Luo2000). In this situation, organizations need to adopt a strategic perspective to manage their relationship with the government. Thus, institutional strategies are critical to an organization's performance and long-term viability in emerging markets (Marquis & Raynard, Reference Marquis and Raynard2015).
Institutional Development and the Role of Government in China
Despite being one of the largest economies in the world, China is characterized by the underdevelopment of formal market institutions, particularly legal systems and regulatory policies (Allen, Qian, & Qian, Reference Allen, Qian and Qian2005; Hoskisson, Eden, Lau, & Wright, Reference Hoskisson, Eden, Lau and Wright2000). The Chinese government has made great efforts to improve the institutional environment (Chen, Sun, & Wu, Reference Chen, Sun and Wu2010), but the latter remains far from mature in terms of investor protection and information transparency. As a result, opportunistic behaviors such as financial fraud events are prevalent (Boisot & Child, Reference Boisot and Child1988; Yiu, Xu, & Wan, Reference Yiu, Xu and Wan2014). Moreover, although the economic transition is in full swing with economic liberalization and privatization policies newly implemented, it creates varied and sometimes incompatible regulations, increasing uncertainty and enhancing opportunities for state predation (Bian & Zhang, Reference Bian, Zhang, Brass, Labianca, Mehra, Halgin and Borgatti2014; Manion, Reference Manion2004; Naughton, Reference Naughton2007). The government controls key resources and many business activities and therefore has the ability to shape the views of other institutional constituents and of the firms about what kinds of corporate behaviors are acceptable (Dobbin & Sutton, Reference Dobbin and Sutton1998).
A good example of governmental influence is its efforts to monitor and curb corporate financial fraud. The Chinese central government has established the China Securities Regulatory Commission (CSRC), which is a government agency similar to the SEC in the United States that regulates the nationwide securities market to ensure its lawful operation (Friedman, Reference Friedman2002). CSRC carries out checks, investigations, and prosecutions of corporate financial fraud. CSRC punishes the convicted firms and discloses the details to the public. The fraud punishments levied by CSRC have far-reaching consequences for convicted firms as they can greatly influence reactions of other institutional constituents (Chen, Firth, Gao, & Rui, Reference Chen, Firth, Gao and Rui2005).
In particular, when a firm is convicted of financial fraud, punishment by the CSRC indicates to the other government agencies that the firm has acted inappropriately. In this situation, the other government agencies may withdraw legitimacy provided for the firm. For instance, one of the largest financial fraud scandals in China involved the Green-Land Biological Technology Company. In 2013, this company was punished by the CSRC for several financially fraudulent acts, including inflating its profits and exaggerating its assets. As a result, the company suffered a huge loss. The government subsidy it received declined from RMB 19 million in 2012 to RMB 0.61 million in 2013 and to 0 in 2014, far less than the average subsidy provided to its industry peers (at RMB 17.8 million in 2013 and RMB 25.4 million in 2014). Moreover, fraud punishment levied by CSRC may further lead to the loss of the firm's legitimacy from various institutional constituents. Since the mass media are controlled by the government, other institutional constituents often turn to government regulators for information when assessing firms’ legitimacy (Bansal & Clelland, Reference Bansal and Clelland2004; Yu, Sengul, & Lester, Reference Yu, Sengul and Lester2008). In this situation, in addition to directly reacting to public criticism and pressure caused by fraud punishment, convicted firms also need to use legitimacy regaining strategies to manage legitimacy purposefully in China.
The literature to date has explicitly focused on a developed market economy context and identified some important strategies for managing legitimacy, such as lobbying and organizing political committees (Dorobantu, Kaul, & Zelner, Reference Dorobantu, Kaul and Zelner2017). However, such institutional strategies may be ill-suited for the context of China. Given the important role corporate philanthropy plays in China (Wang & Qian, Reference Wang and Qian2011), we focus on corporate philanthropy and propose that it can be a valid legitimacy regaining strategy to help convicted firms regain legitimacy both from the public and the government. We summarized our review of past relevant conceptual and empirical studies in Table 1 and Table 2.
Table 1. Summary of conceptual papers in the literature
Table 2. Summary of empirical findings in the literature
Notes: * p < .05, ** p < .01, *** p < .001, two-tailed tests.
Corporate Philanthropy as a Legitimacy Regaining Strategy After Fraud Punishment in China
When a listed firm is punished for financial fraud by CSRC, the punishment information would be released to the public. The severity of punishment is usually commensurate with the severity of the fraudulent behavior. Given that fraud punishment would bring stigma to convicted firms (Carberry & King, Reference Carberry and King2012; Tracey & Phillips, Reference Tracey and Phillips2016), the public and the government, which are the two important institutional constituents in China, may withdraw legitimacy from the convicted firms (Romani, Grappi, & Bagozzi, Reference Romani, Grappi and Bagozzi2013; Sullivan, Haunschild, & Page, Reference Sullivan, Haunschild and Page2007; Wang, Reference Wang2010). In particular, punishments that are more severe are more visible to the public and the government (Skowronski & Carlston, Reference Skowronski and Carlston1987, Reference Skowronski and Carlston1989), and thus they are more likely to elicit unfavorable responses. Hence, such punishments are associated with greater legitimacy loss, which in turn can lead to greater uncertainty over the future survival and success of the convicted firms (Ruef & Scott, Reference Ruef and Scott1998).
According to institutional theory, convicted firms would adopt certain highly visible and relevant strategies to regain legitimacy both from the public and the government (Meyer & Rowan, Reference Meyer and Rowan1977). In China, corporate philanthropy can be such a legitimacy regaining strategy that helps convicted firms manage legitimacy after being punished by the government. There are three reasons for the success of this strategy.
First, corporate philanthropy can help convicted firms divert the attention of various institutional constituents away from the wrongdoing, thus alleviating the loss of legitimacy. Because socially responsible activities of firms are often viewed favorably by institutional constituents (Aguinis & Glavas, Reference Aguinis and Glavas2012; Campbell, Reference Campbell2007) such as the public, the media that need to cater to the constituents’ preferences may cover more about the firms’ responsible activities (Cahan, Chen, Chen, & Nguyen, Reference Cahan, Chen, Chen and Nguyen2015). Such media coverage in turn can attract the attention of various institutional constituents. As such, convicted firms may use corporate philanthropy, which is one type of socially responsible act, to divert institutional constituents’ attention away from the financial fraud, so as to reduce the loss of legitimacy.
Second, corporate philanthropy can help convicted firms regain legitimacy from the public (Pava & Krausz, Reference Pava and Krausz1997). A wealth of research suggests that corporate philanthropy can help firms develop favorable relationships with the public (Brammer & Millington, Reference Brammer and Millington2005; Sen, Bhattacharya, & Korschun, Reference Sen, Bhattacharya and Korschun2006). In particular, Chinese people are deeply influenced by traditional culture that attaches great importance to showing compassion and kindness toward others. Buddhism, Daoism, and Confucianism are all part of the traditional Chinese culture. Buddhists believe that compassion is a virtue and people should be kindhearted and benevolent. Similarly, Daoism emphasizes the importance of listening to others’ needs. Confucianism regards generous people as sages (Wang & Qian, Reference Wang and Qian2011). As corporate philanthropy conforms to traditional Chinese values, it is often highly praised by Chinese people and thus able to help firms regain legitimacy from the public.
Third, corporate philanthropy can also help convicted firms regain legitimacy from the government. One important responsibility of a government is to promote social welfare (Henderson, Reference Henderson1968). However, China is such a vast country that there are generally not enough resources to engage in community and social welfare projects (Dickson, Reference Dickson2003). As such, corporate philanthropy that can reduce the Chinese government's financial burden is often highly appreciated by the government (Dickson, Reference Dickson2003). Moreover, the Chinese government is heavily involved in charity and retains tight control over the registration of charitable organizations. The majority of registered charities are closely monitored by various government departments (Li et al., Reference Li, Song and Wu2015). In this sense, firms that commit to philanthropic activities can be viewed as supporting the government. In particular, the government would consider public opinion when regulating firms (Burstein, Reference Burstein1998). Corporate philanthropy that is favorable to the public can further help convicted firms regain legitimacy from the government.
Fraud punishments that are more severe are also more visible (Skowronski & Carlston, Reference Skowronski and Carlston1987, Reference Skowronski and Carlston1989) and may result in more legitimacy loss. Therefore, convicted firms facing severe punishment are likely to make more philanthropic contributions to regain legitimacy both from the public and the government. We posit the following:
Hypothesis 1:
The severity of punishment imposed by the government on a firm convicted of fraud is positively related to the increase in the firm's corporate philanthropic contributions.
Moderating Effects of Media Visibility and Political Embeddedness
As the government is powerful and controls most of the resources in emerging markets such as China (Peng & Luo, Reference Peng and Luo2000), the public and government agencies, which are two important institutional constituents of firms, often look for signals from governmental actions before making decisions. In the context of fraud punishment by the government, the public and government agencies are very likely to withdraw the legitimacy granted to the convicted firm, leading to a substantive loss for the firm. As such, factors that represent convicted firms’ vulnerability to these two institutional constituencies’ withdrawing of legitimacy may drive the firms to make a response. Media visibility and political embeddedness are the two factors that can affect the firm's vulnerability to the legitimacy loss from the public and the government, respectively. Specifically, we propose that media visibility can strengthen convicted firms’ motivation to regain legitimacy from the public by eliciting more attention and pressure from the public, while political embeddedness can strengthen convicted firms’ motivation to regain legitimacy from the government by eliciting more attention from government agencies and increasing the firms’ anxiety about government punishment. We provide detailed arguments below.
Media visibility
Although different firms might be subject to the same institutional influence, they may respond differently to institutional influence depending on their current conditions (Berrone, Fosfuri, Gelabert, & Gomez-Mejia, Reference Berrone, Fosfuri, Gelabert and Gomez-Mejia2013). Firms’ media visibility is such a condition that may moderate the effect of fraud punishment on corporate philanthropy by increasing the convicted firms’ necessity and motivation to use a legitimacy regaining strategy.
Convicted firms with high media visibility are more vulnerable to the public's withdrawal of legitimacy. High media visibility means a high level of public attention on the firm (Pfarrer, Pollock, & Rindova, Reference Pfarrer, Pollock and Rindova2010). By attracting more public attention, firms’ media visibility helps reduce information asymmetry between the public and the firm (Brammer & Millington, Reference Brammer and Millington2006). A decrease in information asymmetry gives the public a clearer picture of the firms’ actions, including the actions of financial fraud. Since the public often monitor more closely the activities of firms with high media visibility (Brooks, Highhouse, Russell, & Mohr, Reference Brooks, Highhouse, Russell and Mohr2003), the consequences of fraud punishment are amplified for such firms. As such, given a certain level of fraud punishment, convicted firms with high media visibility would face greater loss of legitimacy from the public. Considering the greater threat to their legitimacy and survival, it is more necessary for the convicted firms with high media visibility to use corporate philanthropy as a legitimacy regaining strategy.
To sum up, high media visibility means greater attention and scrutiny from the public, which may lead to more legitimacy loss for the convicted firms. Accordingly, media visibility increases the necessity of convicted firms to use corporate philanthropy to regain legitimacy from the public. Thus, we propose that the effects of fraud punishment on corporate philanthropy would be stronger for firms with high media visibility. Stated formally, we hypothesize the following:
Hypothesis 2:
Media visibility positively moderates the relationship between punishment severity and the increase of corporate philanthropy such that the effect of punishment severity is stronger for convicted firms with high media visibility.
Political embeddedness
Political embeddedness refers to ‘bureaucratic, instrumental, or affective ties to the state and its actors' (Michelson, Reference Michelson2007: 352). Given that convicted firms with more political embeddedness often enjoy more preferential treatment by the government (Faccio, Reference Faccio2006; Wang, Wong, & Xia, Reference Wang, Wong and Xia2008), they would suffer from more loss than other firms do once such preferential treatment is withdrawn. Moreover, in the context of fraud punishment, convicted firms with political embeddedness can make a more informed interpretation of fraud punishment and thus be more aware of the negative consequences. Both the threat of the withdrawal of preferential treatment and the knowledge of it would produce more anxiety in convicted firms, which in turn would drive them to resort to corporate philanthropy as a legitimacy regaining strategy to relieve the anxiety. In this study, we focus on two variables that represent political embeddedness: dominant state ownership and national political appointment.
One important factor representing political embeddedness is dominant state ownership (Marquis & Qian, Reference Marquis and Qian2014; Okhmatovskiy, Reference Okhmatovskiy2010). We argue that convicted firms controlled by the central Chinese government are more vulnerable to legitimacy loss. It has been widely observed that state-owned firms may enjoy preferential treatment in terms of inputs and access to product and capital markets (Wang et al., Reference Wang, Wong and Xia2008). The fraud punishment levied by the government means the firms are about to lose such preferential treatment. Thus, fraud punishment can produce more anxiety in executives of firms that are controlled by the central government, making them more likely to take actions to regain legitimacy after the punishment.
Moreover, executives of the firms in which the central government has controlling ownership are usually deeply influenced by the Chinese government system. They are usually exposed to various governmental policies and are busy dealing with such policies (Fang, Reference Fang2000), and thus they are more able to understand the implicit roles of governmental punishment. Specifically, close connection to the central government allows executives to make a more informed interpretation of the rationale behind the imposed punishment, producing more anxiety in them. Executives of firms controlled by the central government are more likely to regard the punishments imposed as exemplary punishments used by the government and realize that the fraud punishment can lead to the loss of their legitimacy from the government.
Given the vulnerability and anxiety, executives of convicted firms that are controlled by the central government are more likely to resort to corporate philanthropy as a legitimacy regaining strategy to relieve their anxiety. It suggests that dominant state ownership can strengthen the effects of fraud punishment on corporate philanthropy. Stated formally:
Hypothesis 3:
Dominant state ownership of firms positively moderates the relationship between punishment severity and the increase of corporate philanthropy, such that the effect of punishment severity is stronger for convicted firms with dominant state ownership.
National political appointment is another important factor representing political embeddedness (Marquis & Qian, Reference Marquis and Qian2014). We argue that convicted firms with national political appointment are also more vulnerable to legitimacy loss. In China, successful business leaders would be elected as members of prestigious state organs such as the National People's Congress (NPC) and the National Chinese People's Political Consultative Conference (CPPCC) (Zhang, Marquis, & Qiao, Reference Zhang, Marquis and Qiao2016). Similar to state-owned firms, firms whose executives have such political appointments also enjoy various preferential treatments (Li et al., Reference Li, Song and Wu2015). When they are punished for fraud by the government, they are likely to lose legitimacy from the government as well as the preferential treatment they previously enjoyed. In addition, since political status is often vulnerable to competition (Zhang et al., Reference Zhang, Marquis and Qiao2016), fraud punishment would produce more anxiety in executives with national political appointment, in that they would worry more about losing political appointment and the political resources brought by political appointment. In fear of this, executives of convicted firms with national political appointment are more likely to take actions to regain legitimacy after the punishment.
Moreover, executives who are elected to national political councils would regularly participate in policy discussions led by the central government (Li, Meng, & Zhang, Reference Li, Meng and Zhang2006). Through formal or informal meetings, such executives would be more familiar with and learn more about the priorities in Chinese government systems (Jia & Zhang, Reference Jia and Zhang2013), enabling them to make a more informed interpretation of fraud punishment. They may interpret fraud punishments imposed on their firms as exemplary punishments used by the government, and thus they would view the punishment more negatively and anxiously than others do. In order to relieve their anxiety, they are more likely to use corporate philanthropy as a legitimacy regaining strategy after punishment.
The arguments above are consistent with the study of Zhang et al. (Reference Zhang, Marquis and Qiao2016), which theorized that political appointment is likely to bind firms to government pressure. Stated formally:
Hypothesis 4:
National political appointment positively moderates the relationship between punishment severity and the increase of corporate philanthropy such that the effect of punishment severity is stronger for convicted firms with executives who have national political appointment.
METHODS
Sample and Data
Our goal is to examine corporate philanthropy after fraud punishment. However, simply conducting regression analysis with a sample of fraudulent firms would not be appropriate. This is because firms that are punished for financial fraud may differ systematically from those that are not. Specifically, it is possible that the factors affecting whether a firm engages in financial fraud may be correlated with our dependent variable—corporate philanthropy increase. We therefore used the two-stage Heckman selection model (Heckman, Reference Heckman1979) to correct for any sample selection biases. In such analyses, parameters estimated from the first-stage probit model based on information that represents all the firms in a population are incorporated into the second stage.
The sample of our first-stage probit model consists of all the public firms from 2004 to 2013 that are listed on the Shenzhen or Shanghai Stock Exchanges. Excluding missing information, our sample of the first-stage model consists of 16,220 firm-year observations across ten years, pertaining to 2,490 unique firms. The sample of our second-stage model consists of all the Chinese-listed firms that were punished for financial fraud during the ten years from 2004 to 2013. We began with all the 1,239 cases of fraud punishment imposed on listed firms from 2004 to 2013 that are recorded in the China Stock Market Accounting Research database (CSMAR). Then, we eliminated 26 fraud punishment cases that involved listed firms in China's B-share stock market because that market is aimed at foreign investors and is governed by much stricter regulations than is the A-share market.
We employed multiple sources to construct our data set. We collected data on punishment severity, corporate philanthropy, dominant state ownership, national political appointment, and other control variables from the CSMAR. As one of the largest databases of Chinese-listed firms, the CSMAR serves as the primary source of information on the Chinese stock market and the financial statements of listed firms (Wang & Qian, Reference Wang and Qian2011). Data on media visibility were collected from a Chinese newspaper database[1] that includes almost all important Chinese newspapers and publications. In addition, we also cross-checked all the observations with missing data and outliers using the data from the companies’ annual reports. After removing observations with missing values for key variables, our final sample of the second-stage model comprised 965 fraud punishment cases pertaining to 722 convicted firms.
Measures
Corporate philanthropy increase
Because we focus on firms’ post-fraud philanthropic activities, corporate philanthropy increase was calculated as the difference between the amount of corporate donations (in millions of yuan) in the year immediately after the fraud punishment year and that in the punishment year. Corporate donations consist of cash and gifts-in-kind.
Punishment severity
As the penalty amount represents the severity of punishment (Yiu et al., Reference Yiu, Xu and Wan2014), punishment severity was measured as the total penalty amount (in millions of yuan) imposed on a given firm for financial fraud in the observation year. We aggregated the records of punishment cases within the same year for every convicted firm, based on which we then coded the data of punishment severity. More specifically, if a firm was punished more than once for financial fraud in a specific year, we aggregated the punishment data to determine punishment severity (i.e., total penalty) in that year.
Media visibility
Following prior literature (Bednar, Reference Bednar2012; Jia & Zhang, Reference Jia and Zhang2015), we collected data on media visibility from newspaper database. Specifically, we searched for coverage in the newspaper database, which includes almost all important Chinese newspapers and publications. We include articles that mention the firms in the title or lead paragraph and that were published during the year that fraud firms were punished (Jia & Zhang, Reference Jia and Zhang2015). Similar to Jia and Zhang (Reference Jia and Zhang2015), we used the number of news media reports as a measure of media visibility.
Dominant state ownership
Following prior study (Luo, Wang, & Zhang, Reference Luo, Wang and Zhang2017), we measured dominant state ownership as a dummy variable, which was coded as 1 if the dominant (i.e., largest) shareholder of a firm was the central government or its agencies and 0 otherwise. The CSMAR database provided detailed information on the dominant shareholder (as well as the other nine largest shareholders).
National political appointment
Consistent with Luo et al. (Reference Luo, Wang and Zhang2017), we coded this variable as 1 if a firm's board chairman or CEO served as a delegate to the NPC or the CPPCC, which are the two most important national political councils (Zheng, Singh, & Mitchell, Reference Zheng, Singh and Mitchell2015).
Control variables
We control for factors that can potentially affect corporate philanthropy increase. As previous research has shown that larger firms engage in more corporate philanthropy than smaller firms (Amato & Amato, Reference Amato and Amato2007), we controlled for firm size, measured as the natural log of the firm's total number of employees. We also controlled for firm age, measured as the number of years since the firm was established. In addition, because ownership has been associated with corporate philanthropy (Gao, Reference Gao2011; Oh, Chang, & Martynov, Reference Oh, Chang and Martynov2011), we controlled for state ownership, management ownership, and foreign ownership, measured as the percentage of shares held by the Chinese government or its agencies (Xu, Zhang, & Chen, Reference Xu, Zhang and Chen2018), the percentage of shares held by top managers, and the percentage of shares held by foreign investors, respectively. Furthermore, prior studies have shown board independence to influence corporate philanthropy (Jia & Coffey, Reference Jia and Coffey1992), and thus we controlled for this variable, too, calculated as the number of independent directors divided by board size.
Researchers have reported that a firm's financial performance affects its philanthropic decisions (Waddock & Graves, Reference Waddock and Graves1997), and thus we controlled for return on assets (ROA) and Tobin's Q. Also, given that slack resources influence corporate philanthropy (Seifert, Morris, & Bartkus, Reference Seifert, Morris and Bartkus2004), we controlled for debt asset ratio measured as the ratio of long-term debts to total assets (Barnett & Salomon, Reference Barnett and Salomon2006; Waddock & Graves, Reference Waddock and Graves1997). We also controlled for fraud punishment history coded as 1 if a firm was punished for financial fraud in the prior year and 0 otherwise. In addition, history of philanthropy increase was controlled. This variable was coded as 1 if a firm increased the amount of corporate philanthropy in the prior year and 0 otherwise. Firms’ prior use of philanthropy as a response to fraud punishment may influence their next response after fraud punishment, thus we controlled for history of philanthropic response. This variable was coded as 1 if a firm increased the amount of corporate philanthropy after fraud punishment in the prior three years and 0 otherwise.
Furthermore, to control for the influence of cross-industry differences in philanthropic contributions, we included five industry dummies representing six industry categories identified by the CSMAR (Wang & Qian, Reference Wang and Qian2011). Similarly, we also included region dummies to exclude the influence of regional variations in China. Furthermore, year dummies were included to tease out time-related differences. Finally, in addition to an administrative penalty, other types of punishment may be handed out to convicted firms, such as a warning or the confiscation of illegal income. We thus included five dummies of punishment type to control for their effects on corporate philanthropy.
Following prior literature (Chen, Firth, Gao, & Rui, Reference Chen, Firth, Gao and Rui2006; Jia & Coffey, Reference Jia and Coffey1992; Walls, Berrone, & Phan, Reference Walls, Berrone and Phan2012; Yiu et al., Reference Yiu, Xu and Wan2014), we controlled for factors that can potentially affect corporate financial fraud engagement. Specifically, we controlled for firm size, firm age, state ownership, management ownership, foreign ownership, board independence, ROA, Tobin's Q, debt asset ratio, fraud punishment history, industry dummies, region dummies, and year dummies in the first-stage probit model.
All independent, moderator, and control variables that vary by year were lagged by one year to predict financial fraud engagement and corporate philanthropy increase.
Estimation Method
The first stage of the Heckman process involved estimating the degree to which a firm's likelihood of being punished for financial fraud can be predicted by various factors. The likelihood of a firm's being punished for financial fraud was estimated by applying a probit model to the entire sample of firms. A dummy variable was the dependent variable of the first-stage probit model. This variable was coded as 1 if a firm was punished for financial fraud in the observation year and 0 otherwise. We calculated an adjustment term, the inverse Mills ratio, from the first-stage probit regression. The ratio was then included as a control variable in the main second-stage equation (Heckman, Reference Heckman1979), which examined the relationship between punishment severity and corporate philanthropy increase using the sample of firms that had been punished for financial fraud. Because some of the firms in our sample may have been punished several times over the years, some of the observations may belong to the same firm, and thus, may be interdependent. In order to mitigate the potential threat of heteroscedasticity and to address potential correlations among the residuals of events that may be interdependent, we used the ‘cluster' (by firm ID) function in STATA to adjust the standard errors (Rotnitzky & Jewell, Reference Rotnitzky and Jewell1990; White, Reference White1980).
RESULTS
Descriptive statistics and the correlation matrix are presented in Table 3 and Table 4. Table 3 includes the variables used in the first-stage probit model of the two-stage Heckman analysis. The descriptive statistics and correlation matrix for the key variables used in the second stage of the Heckman analysis are presented in Table 4. To rule out multicollinearity, we calculated the variance inflation factor (VIF) and the condition index statistic (CIS), which are two important indicators of collinearity among predictors. All VIF values are below 1.4, which is well below the generally accepted limit of 10.0 (Neter, Kutner, Nachtsheim, & Wasserman, Reference Neter, Kutner, Nachtsheim and Wasserman1996) and also below the more conservative estimate of 2. The mean and maximum of the CISs are both below 5.0, well below the recommended threshold of 30 (Belsley, Kuh, & Welsch, Reference Belsley, Kuh and Welsch1981).
Table 3. Descriptive statistics and bivariate correlations of the first-stage variables
Notes: N = 16220. Correlations exceeding 0.01 are significant at p < 0.05 (two-tailed)
Table 4. Descriptive statistics and bivariate correlations of the second-stage variables
Notes: N = 965. Correlations exceeding 0.06 are significant at p < 0.05 (two-tailed).
Table 5 presents the results of the first-stage Heckman selection model, which is a probit regression of the choice of financial fraud punishment against the factors thought to predict whether a firm will be punished for financial fraud. The dependent variable is the dummy variable fraud punishment dummy, indicating whether a firm was punished for financial fraud. Model 1 includes variables to predict fraud punishment dummy. We report odds ratios to better interpret the magnitude of effects. An odds ratio greater than 1 indicates that when the value of the independent variable increases, the likelihood of the occurrence of the fraud punishment will also increase. In contrast, an odds ratio less than 1 indicates that the likelihood of the occurrence of the fraud punishment will decrease with the increase in the independent variable. The result shows that larger firms are less likely to be punished for financial fraud (β = −0.06, odds ratio = 0.94, p < 0.001). Firm age is positively related to financial fraud punishment (β = 0.01, odds ratio = 1.01, p < 0.05). The coefficient of state ownership is negative and significant (β = −0.32, odds ratio = 0.72, p < 0.01). Financial performance measured as ROA (β = −0.57, odds ratio = 0.56, p < 0.001) and Tobin's Q (β = −0.02, odds ratio = 0.98, p < 0.01) is negatively related to financial fraud punishment. Debt asset ratio is also negatively associated with financial fraud punishment (β = −0.46, odds ratio = 0.63, p < 0.001). Moreover, firms with fraud punishment history are more likely to engage in financial fraud and be punished again.
Table 5. Probit estimates for the Heckman first-stage probit model: Fraud punishment dummy as the dependent variable
Notes: N = 16220. †p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed tests. Unstandardized coefficients are presented.
Table 6 presents the results of Heckman's second-stage estimation by controlling for the inverse Mills ratio obtained from the first-stage probit model in Table 5, accounting for selection bias. Model 1 is the baseline model with only control variables. As Model 1 shows, the history of philanthropy increase is negatively associated to corporate philanthropy increase, while the influence of fraud punishment history on corporate philanthropy increase is not significant. A possible explanation is that the influence of fraud punishment on corporate philanthropy would diminish over time. However, contrary to our expectation, the effects of most of the traditional variables, such as firms’ financial status (e.g., ROA, Tobin's Q, and debt ratio) and ownership (e.g., state ownership), are not significant. The possible cause of the null effects of most of the control variables is that our dataset in the second stage of the Heckman test contains only convicted firms, which is different from a traditional dataset that contains all firms (e.g., Cubbage & Brooks, Reference Cubbage and Brooks2012). When making decisions about corporate philanthropy, convicted firms that are punished for financial fraud may focus mainly on the short-term, urgent objective of removing the negative effect of fraud punishment while neglecting other traditional concerns such as financial resource and philanthropic pressure from stockholders.
Table 6. Estimates for the Heckman second-stage model: Philanthropy increase as the dependent variable
Notes: N = 965. †p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001, two-tailed tests. Unstandardized coefficients are presented. Standard errors and p values in parentheses.
In Model 2, we added punishment severity. The positive and statistically significant coefficient (β = 0.19, t = 4.73, p < 0.001) suggests that punishment severity exerts a positive effect on corporate philanthropy increase, which provides support for Hypothesis 1. The magnitude of the effect suggests that an increase in punishment severity of 1 million yuan leads to an increase in corporate philanthropy by approximately 0.19 million yuan. This variable accounts for 11 percent (ΔR 2 = 0.11) of the explained variance in corporate philanthropy increase.
Then, we added interaction terms. In Model 3, the coefficient of the interaction between punishment severity and media visibility is statistically significant and positive (β = 0.87, t = 6.77, p < 0.001). This interaction term accounts for 1 percent (ΔR 2 = 0.01) of the explained variance in corporate philanthropy increase. To gain more insights into this interaction effect, we conducted simple slope tests (Aiken & West, Reference Aiken and West1991). The test shows that the effect of punishment severity on corporate philanthropy increase is significantly positive (simple slope b = 0.97, p <0.001) for firms with high media visibility but becomes negative (simple slope b = −0.77, p <0.001) for firms with low media visibility. These results strongly support Hypothesis 2.
Model 4 tests Hypothesis 3, which predicts that the relationship between punishment severity and corporate philanthropy increase is stronger for firms with dominant state ownership. The result shows that the moderating effect of dominant state ownership on the relationship between punishment severity and corporate philanthropy increase is statistically significant (β = 1.87, t = 1.98, p < 0.05). Simple slope analysis (Aiken & West, Reference Aiken and West1991) indicates that the effect of punishment severity on corporate philanthropy increase is significantly positive (simple slope b = 2.02, p < 0.05) for firms with the central government as the dominant shareholder while it is weaker (simple slope b = 0.10, p < 0.05) for firms with others as the dominant shareholders. Therefore, Hypothesis 3 is also strongly supported.
Hypothesis 4 argues that national political appointment can strengthen the influence of punishment severity on corporate philanthropy increase. The coefficient of the interaction between punishment severity and national political appointment is statistically significant and positive (β = 1.83, t = 2.00, p < 0.05). The effect of punishment severity on corporate philanthropy increase is significantly positive for firms with executives who have national political appointment (simple slope b = 1.93, p < 0.05) while it is weaker for firms without executives who have national political appointment (simple slope b = 0.10, p < 0.05). These results strongly support Hypothesis 4.
It is worth noting that the R-squares of our models indicate there is still a large proportion of unexplained variance in the outcome variable. We speculate that other types of variables such as those of institutional pressures might be the reasons for unexplained variance. Scholars have suggested that, besides regulatory pressures, normative and cognitive pressures are important drivers of corporate social responsibility (Campbell, Reference Campbell2007). While we have included in our models the variables related to industry and region as the control variables, we were unable to fully control for the influence of institutional pressures in this study. Notwithstanding, our results did support the significance of the incremental explanatory power contributed by the hypothesized main and moderating effects.
Robustness Checks
To check the robustness of our findings, we conducted a number of further tests. First, in our main analysis, we used absolute values to measure corporate philanthropy increase and punishment severity. But given that larger firms may donate more money than smaller firms and may also be subject to larger penalties, we also used relative measures of corporate philanthropy increase and punishment severity. More specifically, we divided corporate philanthropy increase and punishment severity by firms’ total assets (natural logarithm). The results are similar to those using absolute values for the two variables.
Second, because the corporate philanthropy increase variable may be skewed, we conducted a robustness test using an alternative measure. Following Wang and Qian (Reference Wang and Qian2011), we transformed corporate philanthropy increase into natural logarithmic form. As not all values of this variable were positive, and the minimum value was -8,874,000, we first added 8,874,000 to all values of philanthropy increase prior to log transformation. The results are similar to those of our main analyses.
Third, as prior fraud punishment may also influence the firm's current response, in order to control for its influence, in our main tests we included fraud punishment history as a control variable, which was measured as whether a firm was punished for fraud in the prior year in our main test. In the robustness test, we used alternative measures of fraud punishment history. Specifically, we measured fraud punishment history as a dummy variable coded as 1 if a firm was punished for fraud in the prior three years and 0 otherwise. The results are robust to that of the main tests, and the influence of fraud punishment history on corporate philanthropy increase is still not significant.
Fourth, we conducted another robustness test by examining the aggregated influence of the severity of fraud punishments in the current and the prior years on corporate philanthropy increase. Specifically, we aggregated the data on the penalties levied on the firm for fraud in the observation year and in the previous two years. Because the influence of fraud punishment may diminish over time, following previous studies (Yiu et al., Reference Yiu, Xu and Wan2014) we use a three-year time-diminishing measure of punishment severity. More specifically, we took the weighted average of the penalties levied on a firm for fraud in the observation year and in the previous two years:
where t refers to the observation year in our main analysis. The results are consistent with our main analysis.
Fifth, as some existing empirical evidence used organizational size as a measure of firm visibility (Brammer & Millington, Reference Brammer and Millington2006), we conducted another robustness test by using firm size to measure a firm's visibility. Specifically, we used the natural log of the firm's total number of employees. The results remain similar to those using the number of news media reports as a measure of visibility.
Finally, the key assumption of this paper is that convicted firms would make more philanthropic contributions to regain legitimacy since fraud punishment levied by government agencies would lead to the loss of the firms’ legitimacy. In the main tests, we only compared the punishment effects among firms being punished for fraud. In order to verify the assumption, we examined whether firms being punished for fraud would make more philanthropic contributions than firms without fraud punishment. Specifically, using panel data of all the public firms from 2004 to 2013, we examined the effect of fraud punishment dummy on corporate philanthropy increase (Wooldridge, Reference Wooldridge2003). The effect of fraud punishment dummy on corporate philanthropy increase is statistically significant (β = 0.12, t = 1.94, p < 0.10). The result provides additional support for our hypotheses. All the above analyses demonstrate the robustness of our findings.
Post Hoc Test
Corporate philanthropy history and fraud punishment history
To gain more insights into the relationship between punishment severity and corporate philanthropy increase, we conducted some post hoc tests. First, previous research has shown the importance of a company's history of corporate social behaviors for its current behaviors (Godfrey, Reference Godfrey2005; Tilcsik & Marquis, Reference Tilcsik and Marquis2013). A firm's history of philanthropy may also affect its response to fraud punishment. There may be reasons to expect that a history of generosity will weaken the positive effect of punishment severity on corporate philanthropy increase. Companies with a solid record of charitable behaviors might have built up a reservoir of public goodwill (Peloza, Reference Peloza2006), which could reduce pressures for philanthropy increase after fraud punishment. As such, we examined the moderating role of firms’ prior history of philanthropy. We measured a company's history of philanthropy as its annual average of charitable contributions (logged) in the prior three years before the firm was punished. Contrary to expectation, we find that the coefficient of the interaction between punishment severity and history of philanthropy appointment is positive but not statistically significant (β = 0.05, t = 0.61, n.s.). The reason may be that firms with a good record of philanthropy tend to face increased public expectation and scrutiny. The heightened expectations and scrutiny could compel such firms to be more generous in response to fraud punishment. We also examined the moderating role of firms’ fraud punishment history. The results showed that the moderating effect of fraud punishment history is not significant, suggesting that repeated punishment cannot strengthen convicted firms’ motivation to use corporate philanthropy as an institutional strategy.
Ascribed vs. achieved political connection
Zhang et al. (Reference Zhang, Marquis and Qiao2016) identified two types of political connections – ascribed and achieved. Based on their study, ascribed political connections refer to political ties that executives built while they had been government officials, and achieved political connections are defined as political appointments to prestigious state organs acquired after executives became successful business leaders. Zhang et al. (Reference Zhang, Marquis and Qiao2016) theorized that achieved political connections are likely to bind firms to government pressure, while ascribed political connections can buffer firms from the government pressure. Consistent with their argument, we find that national political appointment (achieved political connections) can make firms more sensitive to fraud punishment by strengthening the influence of punishment severity on corporate philanthropy. Given that ascribed political connections may serve a buffer function, will ascribed political connections weaken the influence of fraud punishment severity on corporate philanthropy increase? As such, we also examined the moderating role of firms’ national ascribed political connections. National ascribed political connections were coded as 1 if a firm's Chairman or CEO had been a government official and 0 otherwise. The results show that the coefficient of the interaction between punishment severity and national ascribed political connections is negative but not statistically significant (β = 0.94, t = 1.56, n.s.). The results suggest that ascribed political connections cannot buffer the influence of fraud punishment on corporate philanthropy.
Political embeddedness with central government vs. local government
In the current study, we propose that convicted firms would regard corporate philanthropy as an institutional strategy to regain their legitimacy and such firms’ political embeddedness with the central government would strengthen the effect of punishment severity on corporate philanthropy increase. Then, will convicted firms’ political embeddedness with the local government moderate the effect of punishment severity on corporate philanthropy increase? As such, we examined the moderating role of firms’ political embeddedness with the local government. Specifically, we examined the moderating effects of dominant local government ownership and local government political appointment on the relationship between punishment severity and corporate philanthropy increase. Dominant local government ownership was coded as 1 if the dominant shareholder of a firm was the local government (government at provincial, city, or county level) or its agencies and 0 otherwise. Local government political appointment was coded as 1 if a firm's Chairman or CEO served as a delegate to the NPC or CPPCC of the local government and 0 otherwise. The results show that the moderating effects of these two variables were not significant. The results provide additional evidence that convicted firms would regard corporate philanthropy as an institutional strategy to regain their legitimacy from the central rather than the local government.
Recipient of philanthropy
Although government departments and state-controlled charitable organizations are the major recipients of philanthropic giving in China (Li et al., Reference Li, Song and Wu2015), we tried to verify our findings by searching for information on the recipients of corporate philanthropy. Specifically, we searched detailed information of corporate philanthropy from firms’ annual reports, CSR reports, official company websites, micro-blogs, and the Internet. We found that among the 965 cases in our sample, most of them have no information on recipients and only 58 have the information on recipients disclosed. As such, the recipients can be categorized into several types, including government agencies (15 cases), disaster relief projects (21 cases), schools (16 cases), intra-organizational charity projects (11 cases), poverty alleviation projects (3 cases), and unknown recipients (907 cases). We then examined the influence of punishment severity on different types of donations after fraud punishment. The results showed that punishment severity was positively associated with both the amount of donations contributed to disaster relief projects and the amount of donations with unknown recipients, and such effects are strengthened by media visibility. The result may be caused by the fact that most of the disaster relief projects are run by Red Cross government-related funds and the fact that donations with unknown recipients also go to government-related projects or funds (Li et al., Reference Li, Song and Wu2015).
DISCUSSION
In this study, we use institutional theory as our overarching perspective to investigate how fraud punishment by the government influences corporate philanthropy. Our supposition is that firms that were punished for financial fraud would use corporate philanthropy as a legitimacy regaining strategy to purposefully and strategically shape their institutional environment. Specifically, we argue that convicted firms would make a higher level of corporate philanthropic contributions in an attempt to regain legitimacy after being punished for fraud by the government.
Our empirical tests on a sample of Chinese-listed firms show that punishment severity is positively related to corporate philanthropy increase. The finding indicates that convicted firms would regard corporate philanthropy as an institutional strategy to regain legitimacy. It also suggests that a more severe punishment can generate a greater loss of legitimacy, thereby leading to more corporate philanthropy. The result is consistent with past findings that firms can cultivate good relations with stakeholders through philanthropic activity (Brammer & Millington, Reference Brammer and Millington2005; Sen et al., Reference Sen, Bhattacharya and Korschun2006). It also echoes the perspective of institutional theory that firms adopt relevant practices to enhance or protect their legitimacy (Deephouse, Reference Deephouse1999; Scott, Reference Scott1995).
Furthermore, our results also reveal that media visibility can strengthen the effect of punishment severity on corporate philanthropy increase. The finding indicates that media visibility can enhance the convicted firm's necessity of using a legitimacy regaining strategy by increasing public attention and legitimacy pressure on the firm. Previous studies (Chiu & Sharfman, Reference Chiu and Sharfman2011; Jia & Zhang, Reference Jia and Zhang2015) found that media visibility would affect top managers’ decisions about corporate social responsibility. Our finding goes beyond this to demonstrate that media visibility can strengthen convicted firms’ tendency to use corporate philanthropy as the legitimacy regaining strategy after being punished.
We also found that political embeddedness represented by dominant state ownership and national political appointment can strengthen the effect of punishment severity on corporate philanthropy increase. The results indicate that political embeddedness can elicit more attention from the government and enhance convicted firms’ tacit knowledge of government punishment in the institutional context, thereby increasing their anxiety and driving them to more likely use corporate philanthropy as a response to the punishment levied by the government. Previous studies (e.g., Luo et al., Reference Luo, Wang and Zhang2017) found that firms with political embeddedness would respond to governmental guidelines more swiftly. Our study not only is consistent with their findings but also demonstrates that political embeddedness strengthens convicted firms’ motivation to respond to government actions by corporate philanthropy.
Theoretical Contributions and Implications
This study has several theoretical implications. First, in response to recent calls to explore the ways in which organizations purposefully and strategically shape their institutional environment (Marquis & Raynard, Reference Marquis and Raynard2015), we contribute to the literature on institutional strategies by regarding corporate philanthropy as a legitimacy regaining strategy. Although the extant literature that focuses on developed market contexts has identified some institutional strategies, such as lobbying and organizing political committees, little attention has been paid to specific institutional strategies in emerging markets. More importantly, institutional strategies that are effective in developed economy contexts may be ill-suited for, or misaligned with, the idiosyncratic conditions of emerging markets. Our study finds that corporate philanthropy is one institutional strategy that convicted firms use to regain legitimacy after fraud punishment, which provides evidence that institutional strategies can be different in China from those in developed markets. By doing so, our study broadens the theoretical scope of the literature on institutional strategies.
Second, we also contribute to institutional theory by exploring the mechanisms that drive convicted firms to respond differently to the institutional stimulus by the government. Although theoretical studies have suggested that organizational responses to institutional pressures can vary (Oliver, Reference Oliver1991, Reference Oliver1997), the issue has not gained much attention from researchers until recently (e.g., Berrone et al., Reference Berrone, Fosfuri, Gelabert and Gomez-Mejia2013; Chatterji & Toffel, Reference Chatterji and Toffel2010). Our findings indicate that convicted firms’ vulnerability to the loss of legitimacy from the public and the government can strengthen the effect of fraud punishment on corporate philanthropy increase, thus enriching the literature by uncovering the mechanisms linking institutional stimuli and firm response in the context of China.
Third, we contribute to the literature on political embeddedness by showing that political embeddedness would make convicted firms more sensitive to governmental punishment and lead to a higher cost. Previous studies have predominantly focused on the benefits of political embeddedness and argue that firms with political embeddedness may enjoy preferential treatment by the government (Faccio, Reference Faccio2006; Wang et al., Reference Wang, Wong and Xia2008), such as lighter taxation. Only a few scholars have realized that political embeddedness can have some costs by binding firms to the government (e.g., Okhmatovskiy, Reference Okhmatovskiy2010; Zhang et al., Reference Zhang, Marquis and Qiao2016). Our findings illustrate that a high level of political embeddedness makes firms more vulnerable to government influence, thereby leading to more philanthropic contributions after punishment, which provides evidence for the less-studied view of the costs of political embeddedness.
In addition to theoretical contributions, the findings of our study have practical implications for regulators. Our study shows that in emerging markets such as China, corporate philanthropy can be regarded as an institutional strategy to shape firms’ institutional environment after fraud punishment, and convicted firms vulnerable to legitimacy loss are more likely to use such a strategy to shape their institutional environment. Our study suggests that regulators should be alert to the donations of convicted firms and differentiate symbolic compliance from substantial compliance. Given that corporate philanthropy can be used as a legitimacy regaining strategy for convicted firms, those firms that would continue their wrongdoing may take advantage of this. If such motivations are not discerned by the government, there is little chance of controlling corporate financial fraud. Therefore, regulators should watch more closely the convicted firms that make more philanthropic contributions after fraud punishment.
Limitations and Future Research Directions
Although this study makes a number of notable contributions, it is not without its limitations. First, we focus on fraud punishment by the government in emerging markets, which is the largest and most powerful institution in the context. However, the effects of punishment by other institutional constituents (e.g., investors and customers) may be different from that by the government. Comparing their effects can also make a contribution to the literature.
Second, we only examined corporate philanthropy as the outcome of fraud punishment and we do not know whether firms also engage in other activities after being punished for fraud. Simultaneous investigation of other outcomes such as public apology, corrective behaviors, or improvement of corporate governance may give us a more complete picture and help us differentiate substantial compliances from symbolic ones.
Third, we focused on financial fraud as a typical example of corporate wrongdoing; examining the government's punishment for other types of wrongdoing and the firms’ various reactions would certainly yield additional insights. For example, when firms are punished for wrongdoing that has far-reaching negative consequences for society, such as environmental pollution or fake vaccines, simply reacting by making more philanthropic contributions may not be enough. Studies comparing the impacts of different types of wrongdoing are especially encouraged.
Fourth, in this study we argue that convicted firms would use corporate philanthropy to regain legitimacy both from the public and the government after being punished for fraud, and our results provide support for this argument. However, data limitation prevented us from directly testing these mechanisms. It also prevented us from differentiating the types of corporate philanthropy. Future studies could test them directly by using survey data or case analysis.
Another fruitful direction for future research would be to investigate whether our findings can be replicated in other countries and whether firm reactions to government punishment differ across countries. This type of analysis of the proposed effects may enrich our understanding of the boundaries of governmental influence in different institutional contexts. It is particularly interesting to investigate whether and how multinational firms react to punishments by governments in different host countries.
Finally, future research can work toward including other variables that may have an impact on convicted firms’ philanthropic decisions. The results show that the traditional variables controlled in our study only explain a small portion of the variance of convicted firms’ philanthropy increase, suggesting that some important factors may be ignored. As prior studies have demonstrated the effects of a CEO's psychological characteristics on corporate decisions (Hambrick & Mason, Reference Hambrick and Mason1984), such characteristics may be important factors that can explain convicted firms’ philanthropic response after fraud punishment. For instance, as hubristic CEOs tend to overestimate the resources they possess and underestimate the legitimacy required (Tang, Qian, Chen, & Shen, Reference Tang, Qian, Chen and Shen2015), they may attach less importance to a legitimacy regaining strategy after fraud punishment. The inclusion of pertinent variables regarding a CEOs’ psychological characteristics may help to further explain the convicted firms’ philanthropic decisions after fraud punishment.
CONCLUSION
This study adopts the institutional perspective to explore how fraud punishment by the government influences corporate philanthropy and the mechanisms underlying such influence. Using data of Chinese-listed companies, our analysis shows that fraud punishment severity is positively associated with the increase in corporate philanthropy of the convicted firm. Furthermore, we also find that media visibility, as well as political embeddedness represented by dominant state ownership and national political appointment, can strengthen such an effect. This study thus enhances our understanding of the role that corporate philanthropy plays as an institutional strategy for convicted firms to regain legitimacy from the public and the government, as well as the underlying mechanisms.
