Introduction
In the last three to four decades, corporations have faced increasing pressures to conduct business in a socially responsible way. Customers, communities, employees, governments, and even shareholders are all joining in the push for more attention to society’s well-being (Sethi, Reference Sethi2003). To respond to stakeholders’ social responsibility concerns, firms’ strategies are changing to demonstrate social consciousness, including through philanthropy, and enhance legitimacy (Idowu & Papasolomou, Reference Idowu and Papasolomou2007; Chiu & Sharfman, Reference Chiu and Sharfman2011). Corporate philanthropy (CP), as a dimension of corporate social responsibility, refers to gift giving or monetary contributions made by firms to social or charitable causes, such as education, the arts, health care, environmental protection, disaster relief, and the like (Godfrey, Reference Godfrey2005; Wang & Qian, Reference Wang and Qian2011).
Scholars have identified and investigated such antecedents of CP as firm size (e.g., Adams & Hardwick, Reference Adams and Hardwick1998; Brammer & Millington, Reference Brammer and Millington2006), financial performance (e.g., Preston & O’Bannon, Reference Preston and O’Bannon1997; Waddock & Graves, Reference Waddock and Graves1997), slack resources (Sen, Reference Sen2011), leverage (Wang & Qian, Reference Wang and Qian2011), ownership type (e.g., Zhang, Rezaee, & Zhu, Reference Zhang, Rezaee and Zhu2009), advertising intensity (Brammer & Millington, Reference Brammer and Millington2004; Zhang, Zhu, Yue, & Zhu, Reference Zhang, Zhu, Yue and Zhu2010), institutional forces (e.g., Marquis, Glynn, & Davis, Reference Marquis and Qian2007; Gao, Reference Gao2011), industry (Useem, Reference Useem1988; Amato & Amato, Reference Amato and Amato2012), and corporate governance (e.g., Galaskiewicz, Reference Galaskiewicz1997; Bartkus, Morris, & Seifert, Reference Bartkus, Morris and Seifert2002). But they have devoted little attention to the effect of decision-makers’ characteristics (Du, Jian, Du, Feng, & Zeng, Reference Du, Jian, Du, Feng and Zeng2014).
Although CP is an organization-level action, the decision is made by the individuals who own or control the organization (Buchholtz, Amason, & Rutherford, Reference Buchholtz, Amason and Rutherford1999; Hambrick & Quigley, Reference Hambrick and Quigley2014). Like many other organizational decisions or actions, CP is affected by owners’ or top managers’ characteristics, values, and perceptions (Nishii, Gotte, & Raver, Reference Nishii, Gotte and Raver2007; Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009). Further, these determine often the owners/managers’ reputation and standing among their peers and within the community (Fombrun, Reference Fombrun1996). Given the importance of decision-makers’ characteristics, including their altruistic and social interest orientation, in CP, it is surprising that only a handful of studies have examined their effects (e.g., Campbell, Gulas, & Gruca, Reference Campbell, Gulas and Gruca1999; Gomez-Mejia, Cruz, Berrone, & DeCastro, Reference Gomez-Mejia, Cruz, Berrone and DeCastro2011).
Some scholars have introduced the idea of social status and suggested that it could be a significant determinant of socially responsible behavior. Social status is the relative standing or position of an individual in a social system based on prestige, honor, and deference (Berger, Cohen, & Zelditch, Reference Berger, Cohen and Zelditch1972; Thye, Reference Thye2000). It is often associated with social power and to the foundations of social responsibility (Davis, Reference Davis1960; Magee & Galinsky, Reference Magee and Galinsky2008). According to Davis, ‘social responsibilities of businessmen need to be commensurate with their social power’ (Reference Davis1960: 71), and ‘the avoidance of social responsibility leads to gradual erosion of social power’ (Reference Davis1960: 73). Given the association of social power with social status, business owners with high social status are expected by the public to be socially responsible.
Social status is achieved either through ascription (ascribed social status [AsSS]) or through achievements (achieved social status [AcSS]). AsSS is associated to inherited characteristics assigned at birth such as gender, race, and socioeconomic backgrounds; while AcSS refers to the socioeconomic characteristics, acquired during a person’s lifetime (Foladare, Reference Foladare1969; Glorieux & Laurijssen, Reference Glorieux and Laurijssen2011). Since AsSS and AcSS are usually correlated, and the impact of AcSS on CP overshadows that of AsSS, we focus in this study on the effect on CP of business owners’ AcSS.
We focus on business owners’ rather than managers’ AcSS because in most Chinese private enterprises, especially the small- and medium-sized enterprises (SMEs), the owner also acts as the general manager (Huang, Reference Huang2009) and makes almost all important decisions on a daily basis. This is different from many large and publicly traded Western firms in which a hired chief executive officer makes the CP decisions on behalf of shareholders (Galaskiewicz, Reference Galaskiewicz1997; Choi & Wang, Reference Choi and Wang2007). We argue first that business owners’ AcSS increases their visibility to, and thus the expectations and pressures from external stakeholders to support philanthropy.
We also consider the boundary conditions of the effect of business owners’ AcSS on CP. We believe that the pressures a business owner faces to support philanthropy vary with firm visibility. Owners with higher AcSS are scrutinized by the public, and the higher their firms’ visibility the more pressures they would feel to support philanthropy. Therefore, we expect firm visibility to strengthen the positive relationship between business owners’ AcSS and CP. In addition, although business owners with a higher AcSS may contribute more to charity, such decisions may also depend on owners’ power within the firm. For some firms, there is more than one owner and each may have a voice in organizational decisions. In such cases, we expect the owner with a higher ratio of equity in the firm to have more influence. This leads us to assume that the effect of business owners’ AcSS on CP would be strengthened by their equity level.
We conduct this study in the context of China, a large emerging economy. More than three decades of economic reform have given birth to millions of private SMEs. These privately owned SMEs play a very important role in speeding economic growth and creating employment (Anderson, Li, Harrison, & Robson, Reference Anderson, Li, Harrison and Robson2003). In addition, because of the country socialist orientation, an individual’s AcSS is often more valued and respected in China than his/her AsSS. Therefore, China provides us with a good field to test our arguments.
Theories and Hypotheses
Both the resource-based theory (Peteraf & Barney, Reference Peteraf and Barney2003) and the institutional theory (Scott, Reference Scott2014) could help explain the effect of business owners’ AcSS on CP. First, AcSS can be seen as a unique acquired individual resource. Because in SMEs, the owners’ resources can be seen as the firm’s resources, one could argue that owners’ AcSS are also unique firm resources, hard to imitate, and thus at the roots of its competitive advantage. But AcSS is also a social phenomenon, acquired through interaction with other influential social actors over a long period of time. It is fragile and subject to scrutiny by the community. Using Oliver’s (Reference Oliver1997) conceptualization, it is an ‘institutional capital.’ Unlike traditional ‘resource capital,’ which is valued mostly because of its inimitability and value in the firm value chain, an institutional capital influences strategic decisions and in particular could affect decisions about resource capital.
Owners’ AcSS, we said, is related to a firm’s sustainable competitive advantage, and could make its success within a community. But owners’ AcSS is, unlike other resources, institutionally determined. Thus, it comes with limitations on owners’ ability to decide. As argued by Oliver ‘institutional capital is the context surrounding resources and resource strategies that enhances or inhibits the optimal use of valued resource capital’ (Reference Oliver1997: 700). It constrains the ability to make economic choices. All choices are somewhat controlled by AcSS, and we shall argue that this interaction between the resource-based theory and the institutional theory explains that owners of SMEs faced with community expectations would tend to respond so as to protect their institutional capital (i.e., AcSS), even if it is at the expense of existing assets. In family firms, where owners are also powerful, this has also been conceptualized as socioemotional wealth, and has been shown to affect strategic decisions (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, Reference Berrone, Cruz, Gomez-Mejia and Larraza-Kintana2010).
Thus, owners’ AcSS can be conceived as having a dual character. It is an important resource, which can provide distinctiveness and competitive advantage, as stipulated by the resource-based theory (Barney, Reference Barney1991). It is also an institutional resource which affects and constrains the firm’s strategies and economic decisions, as predicted by the institutional theory (Scott, Reference Scott2014). We shall argue that this dual character determines CP giving.
Literature of social status has well documented that high-status individuals often wield more influence (Berger, Fisek, Zorman, & Zelditch, Reference Berger, Fisek, Zorman and Zelditch1977; Piff, Kraus, Côté, Cheng, & Keltner, Reference Piff, Kraus, Côté, Cheng and Keltner2010), receive more opportunities (Cohen, Reference Cohen1994; Thye, Reference Thye2000), are more favorably evaluated (Foschi, Reference Foschi1992), are given more credit (Merton, Reference Merton1968), and are allocated more resources (Berger, Wagner, & Zelditch, Reference Berger, Wagner and Zelditch1985; Godoy et al., Reference Godoy, Reyes-García, Leonard, Huanca, McDade, Vadez and Tanner2007). They also have higher levels of media exposure, obtain greater recognition, receive more orders from sellers (Podolny, Reference Podolny2001), and thus face higher stakeholders’ expectations or pressures, than those who have lower social status, to support philanthropy (Stewart, Reference Stewart2005). This demonstrates the resource capital aspect, but in this paper, we shall argue that this also leads to institutionally determined decisions, notably in philanthropy.
AcSS of business owners in China
The concept of social status has been developed in both sociology and social psychology. In particular, German sociologist, Max Weber (see Gerth & Mills, Reference Gerth and Mills1946; Kelly & Stein, Reference Kelly and Stein2011) proposed that stratification is based on three factors, ‘the three p’s of stratification’: property, prestige, and power. He claimed that social stratification is a result of the interaction of wealth, prestige, and power.
-
∙ Property refers to one’s material possessions and their life chances. If someone has control of property, that person has power over others and can use the property to his or her own benefit.
-
∙ Prestige is also a significant factor in determining one’s place in the stratification system. Property does not always assure power, and frequently people have power and status with prestige and little property.
-
∙ Power is the ability to do what one wants, regardless of the will of others.
We could therefore argue that anything that affects these three dimensions affects AcSS. For example, income would affect property; education, political positions, and personal contributions to community would affect prestige; political ties would affect power. Occupation may affect all of property, prestige, and power.
Specifically, it has been widely recognized in the social science literature that one’s occupation, education, and income are good indices of his or her AcSS (Duncan, Reference Duncan1961; Hollingshead, Reference Hollingshead1975; Evans & Kelley, Reference Evans and Kelley1995; Yamaguchi & Wang, Reference Yamaguchi and Wang2002). China is not an exception. In China, occupation is highly and directly related to one’s AcSS. Government officials, entrepreneurs in large companies, famous scientists, movie stars, and other famous artists tend to enjoy high social status; while farmers, migrant workers, and laid-off workers tend to have low social status. Income is also related to AcSS, especially during the present period of transition. Economic growth and the rise of people’s living standard are seen as the key tasks of the Chinese government (Zhu, Reference Zhu2012). More than three decades of economic reform have opened the door to market opportunities and motivated the general public to pursue private wealth. Striking it rich is now almost everyone’s dream or aspiration (Xiao, Reference Xiao2009). There is a saying in China that ‘money can make the devil turn a mill!’ Thus, income should be an important index for AcSS. Finally, education is also highly valued by the Chinese people. China has a long tradition of valuing teachers and education. Most people believe that ‘knowledge can change fate.’ Scientists and other intellectuals have traditionally been admired, and are widely respected. As a result, there is no doubt that an individual’s education could reflect partially his or her AcSS.
In addition to the above three factors, some Chinese scholars have also pointed out that an individual’s social capital, especially his/her political ties, cultivated by holding a position in the People’s Congress or the Chinese People’s Political Consultative Conference (CPPCC)Footnote 1 , could be seen as an additional factor in the determination of AcSS (e.g., Lin & Bian, Reference Lin and Bian2002; Zhang, Reference Zhang2004; Chou & Xiao, Reference Chou and Xiao2011). For example, deputies of the People’s Congress and the CPPCC at all levels enjoy various privileges according to the constitution and laws. They have the right to discuss state affairs and local important issues, and supervise the public administration system operations. Deputies of the People’s Congress also have the right to participate in the election of governors, to formulate laws and important public policies. They have some level of legal immunity, being exempt from immediate detention even when they commit crimes. Therefore, political ties could be used as another factor to reflect an individual’s AcSS in China.
Also contributing to the social capital dimension is visible action dedicated to the community’s well-being. Volunteer work and personal donations have traditionally been seen as such an expression of concern, and valued by community opinion-makers. The French sociologist Bourdieu (Reference Bourdieu1984) developed theories of social stratification based on esthetic taste in his work Distinction. Bourdieu claims that how one chooses to present one’s social space to the world, one’s esthetic dispositions, depicts one’s status and distances oneself from lower groups. Individual donations, charitable, or philanthropic, are a distinction mechanism, and would be associated with prestige. They thus affect AcSS.
Therefore, in this study, a business owner’s AcSS is reflected by his or her income, education, political ties, and personal donations in money or in nature. We have left out occupation, because we are here focusing on individuals who are all business owners, occupation is not expected to provide any differentiating effect.
AcSS and CP
As argued previously, individuals with higher social status tend to be more influential and enjoy many privileges (Berger et al., Reference Berger, Fisek, Zorman and Zelditch1977; Thye, Reference Thye2000; Ridgeway & Correll, Reference Ridgeway and Correll2006). They also have higher levels of media exposure, achieve greater recognition, and all these generate higher expectations and pressures from the public to contribute to social good (Podolny, Reference Podolny2001; Stewart, Reference Stewart2005; Magee & Galinsky, Reference Magee and Galinsky2008). Following this logic, we expect that firms whose business owners enjoy higher AcSS would contribute more to philanthropic giving.
One important reason to justify such a conclusion is that a firm’s decisions tend to reflect the decision-makers’ characteristics and values (Hambrick & Mason, Reference Hambrick and Mason1984; Nishii, Gotte, & Raver, Reference Nishii, Gotte and Raver2007; Finkelstein, Hambrick, & Cannella, Reference Finkelstein, Hambrick and Cannella2009). Further, we said, organizational decisions in Chinese private firms, especially in SMEs, are generally made by business owners, who usually act as general managers. This owner–manager duality explains that the pressures these individuals receive are also the pressures affecting their firms. We can conflate the owners with their firms, and state that their perceptions of external pressures and attitudes toward philanthropy would affect their firms’ CP decisions.
Since business owners with higher AcSS receive higher media exposure (Podolny, Reference Podolny2001) and thus receive more pressures from the public to support philanthropy, we expect that they will make their firms’ decisions so as to contribute more to CP (Hess & Warren, Reference Hess and Warren2008). This discussion justifies the following hypothesis:
Hypothesis 1: There is a positive relationship between business owners’ AcSS and their firms’ CP.
The moderating effect of firm visibility
Business owners with higher AcSS are likely to contribute more organizational resources to support philanthropy. However, the pressures a business owner receives or perceives are also related to his or her firm’s visibility among external stakeholders. Higher firm visibility would increase public CP expectations and pressures. Thus, we expect firms’ visibility to strengthen the positive relationship between business owners’ AcSS and CP.
A firm’s visibility, although related, is distinct from the owner’s visibility and AcSS. It is generally the result of investments in advertising and promotion and usually part of a firm’s differentiation strategy. The corporate social responsibility literature suggests a direct relationship between firm visibility and CP (e.g., Useem, Reference Useem1988; Zhang et al., Reference Zhang, Zhu, Yue and Zhu2010). Useem (Reference Useem1988) has shown that firms in more visible industries face higher expectations or pressures to support social causes. Zhang et al. (Reference Zhang, Zhu, Yue and Zhu2010) found that a firm’s visibility (measured by advertising intensity) has a positive effect on corporate disaster relief giving in the context of China. The authors believe CP and advertising to complement each other in differentiation strategies.
High visibility attracts more attention from external stakeholders, especially from existing and potential customers and employees (Brammer & Millington, Reference Brammer and Millington2006). Because the general public’s awareness of corporate social responsibility increases rapidly in China (Wang & Qian, Reference Wang and Qian2011; Marquis & Qian, Reference Marquis, Glynn and Davis2014), firms with higher visibility are likely to face more pressures to increase philanthropy (Gao, Reference Gao2011; Wang & Qian, Reference Wang and Qian2011). Thus, if higher AcSS leads business owners to face higher pressures and expectations to support CP, high firm visibility would increase these pressures to do so. Thus, we expect a positive moderation effect of firm visibility on the relationship between business owners’ AcSS and CP.
Hypothesis 2: Firm visibility strengthens the relationship between business owner’ AcSS and their firms’ CP.
The moderating effect of equity level
Although business owners’ AcSS is expected to have a positive effect on CP, the effect would vary with the owner’s influence or power over his or her firm’s decisions. Since business owners’ power in organizational decisions is generally determined by their holdings of the firm’s equity, we expect the owners’ equity level to affect their CP decisions and thus moderate the relationship between owners’ AcSS and CP.
For some private SMEs there is only one owner and this owner makes all major decisions for his or her firm (He & Sommer, Reference He and Sommer2006). The organizational CP decisions fully reflect the owner’s perception of external pressures and his/her own attitude toward CP. However, it is also likely that firms have more than one owner and in such cases, the CP decisions are affected by other owners as well. The voice of a particular owner in such decisions is determined mainly by his/her relative equity level. The CP decision would therefore be a composition of these owners’ influence and AcSS.
Even for firms where ownership and control are separated (Jensen & Meckling, Reference Jensen and Meckling1976), the same logic applies, even though the owners usually shape or direct the organization’s decisions through a board of directors (Fama & Jensen, Reference Fama and Jensen1983; Hafsi & Turgut, Reference Hafsi and Turgut2013). In these situations, only the owners who have a high level of equity have the chance to be represented on the board (Jensen, Reference Jensen1993). The higher the equity level, the higher the chance an owner has to be represented or nominated as a director. In addition, in the boardroom, the owners or directors with higher levels of equity have more power in affecting the firm’s operations and managers’ decisions.
In our sample, some firms have a single owner, but many have more than one owner. Therefore, we expect a business owner’s equity level to strengthen the effect of his or her AcSS on CP. Thus, we propose the following hypothesis:
Hypothesis 3: A business owner’s equity level strengthens the relationship between his/her AcSS and his/her firm’s CP.
Method
Data collection
The data used in this research come from a nationwide survey of private firms in China conducted by the United Front Work Department of the Central Committee of the Chinese Communist Party, the All-China Federation of Industry and Commerce, the State Administration for Industry and Commerce of China, and the China Private Economy Association, in March 2008. This survey is one of a series of surveys to collect information about private firms in China. This survey was conducted across all 31 provinces and all industry sectors in China. The respondents of this survey were main owners of private enterprisesFootnote 2 . The survey received responses from more than 4,000 firms. However, some questionnaires were not fully completed. By excluding questionnaires with important information missing, we obtained a final sample consisting of 2,019 firms.
Most of our sampled firms are SMEs. SMEs are defined differently in different industries, according to the Chinese criteriaFootnote 3 . For example, within industries such as mining, energy, manufacture, and construction, SMEs are defined as those firms with total assets valued at no more than RMB 400 million, or with a number of employees no higher than 2,000, or with annual revenue no higher than RMB 300 million. While SMEs in wholesale and retailing should have no more than 500 employees, or no more than RMB 150 million in revenues for retailers, or no more than RMB 300 million in revenues for wholesalers. In our sample, only six firms do not strictly conform to these criteriaFootnote 4 . It is worthy to note that this definition of SMEs in China is different from most Western developed countries’ definitions (e.g., European Commission, 2003; U.S. International Trade Commission, 2010). This may be due to differences in resource endowments. Most Western developed countries ensure access to a large pool of capital and technological resources, but have a relatively limited pool of workers. In contrast, the workforce in China is abundant but access to financial capital and technological resources is constrained. The SME definition differences make international comparisons of findings more difficult. We will discuss this later.
All empirical results in this study either come directly from the survey or are derived from it. There could be concern about common method bias, but the risk is small, because the survey was anonymous, extensive, including and mixing dozens of variables in an ambiguous order. Also, rather than simply testing the main effect of business owners’ AcSS on CP, we also look at the interaction effects between AcSS and both firm visibility and equity level. It has been shown that such complicated multiple regression analyses are less likely to be affected by common method bias (Chang, Witteloostuijn, & Eden, Reference Chang, Witteloostuijn and Eden2010; Siemsen, Roth, & Oliveira, Reference Siemsen, Roth and Oliveira2010).
Concern about overreporting of CP should also be put to rest. The original aim of this survey was to collect information about private firms’ operations rather than to focus on their philanthropic donations. Checking further, we compared our philanthropy data with those of the 2005 survey. In 2005, the private firms reported giving probability was 70.4%, and the average giving amount was about RMB 66,000 (Liang et al., 2010). The giving amount in our sample is larger but remains within a range considered consistent with the rapid development of the Chinese economy and the decreased number of donating firms during the 2-year gap.
Measurement of variables
CP
CP is measured by the total value of a firm’s donations in RMB in 2007Footnote 5 . Because the data are highly skewed (Adams & Hardwick, Reference Adams and Hardwick1998; Wang & Qian, Reference Wang and Qian2011), we computed its natural logarithm.
AcSS
A business owner’s AcSS is reflected by his or her income, education, political ties, and individual donations over time. Some scholars prefer a composite index (AcSS index) of social status by integrating all the four aspects (see Hauser & Warren, Reference Hauser and Warren1997 for a review). We thus computed the ASS index by averaging the four factors equally. The simple additive protocol is made legitimate for at least two reasons. First, theoretically each of the components is likely to add to social status as argued earlier. One could, however, argue that they might have different weights. Therefore, second to check whether they have different effects, we conducted an exploratory factor analysis. The factor analysis is appropriate (Kaiser-Meyer-Olkin (KMO)=0.640, Barlett χ2=1,167.406, p<.001) and only one factor had an eigenvalue larger than one (Kaiser, Reference Kaiser1960). The smallest factor loading is near 0.5 (0.467 for education). Also, according to Hauser and Warren (Reference Hauser and Warren1997) these variables have similar effects on status. The measurement of variables, as seen below, has also been designed to ensure that these variables’ influences remain reasonably comparable. There are other intersample distance calculation possibilities as shown by Hafsi and Turgut (Reference Hafsi and Turgut2013), but such a protocol is tricky and useful only when measuring small and hard to identify differences.
Hauser and Warren (Reference Hauser and Warren1997) have also reported that some scholars argued that the indices are not always reliable, because variables could work at cross-purpose, and they have suggested that one way to ensure that the results are meaningful would be to confirm by looking at the effects of each component variable (Magnuson & Duncan, Reference Magnuson and Duncan2002). In this study, we decided to look also at the effects of individual variables, to check for inconsistent results.
Income
Income is an ordinal category variable. We first identified the quartiles of business owners’ income according to their total annual income in 2007 in the sampled firms. The three quartile references are RMB 50,000, RMB 100,000, and RMB 200,000, respectively. We correspondingly coded income equal to or less than RMB 50,000 as 1; RMB 50,001–100,000 as 2; RMB 100,001–200,000 as 3; and above RMB 200,000 as 4.
Education
Education is also a categorical variable. Business owners who had received their education in senior middle school or below are coded as 1; junior college as 2; university or college as 3; and postgraduate as 4.
Political ties
Political ties are measured according to the ranks of business owners who act or not as representatives at the People’s Congress or the CPPCC (Faccio, Masulis, & McConnell, Reference Faccio, Masulis and McConnell2006; Gao, Reference Gao2011). Business owners who hold ranks below the county level were coded as 1; those at the county level were coded as 2; at the prefecture level as 3; and at the province level and above as 4.
Individual donations
Individual donations in this study refer to the total donations made by business owners individually and over time. Similar to the measure of income, we divided individual donations into quartiles and coded 1 for business owners who had donated RMB 7,000 or below; 2 for RMB 7,001–50,000; 3 for RMB 50,001–200,000; and 4 above RMB 200,000.
Firm visibility
Firm visibility is measured by a firm’s advertising expenditure scaled by total assets in 2007 (Zhang et al., Reference Zhang, Zhu, Yue and Zhu2010). Since this ratio is highly skewed, we winsorized this variable at the upper and lower 1% tails of the distribution.
Equity level
A business owner’s equity level is measured as the percentage of equity owned by the owner in relation to the total equity of the firm by the end of 2007.
Control variables
Previous studies have found or argued that firm size, firm age, financial performance, leverage, and industry type also affect CP (Useem, Reference Useem1988; Preston & O’Bannon, Reference Preston and O’Bannon1997; Waddock & Graves, Reference Waddock and Graves1997; Adams & Hardwick, Reference Adams and Hardwick1998; Brammer & Millington, Reference Brammer and Millington2006; Zhang, Rezaee, & Zhu, Reference Zhang, Rezaee and Zhu2009; Wang & Qian, Reference Wang and Qian2011). Also, individual characteristics such as age (Edmundson, Reference Edmundson1986; Halfpenny, Reference Halfpenny1999) and gender (Jones & Posnett, Reference Jones and Posnett1991) have been found to affect one’s giving. In addition, a business owner’s gender and age could also be related to his/her AsSS, and thus may affect CP. All these variables are controlled for in this study and measured as follows. Firm size, following Lenway and Rehbein (Reference Lenway and Rehbein1991), is measured by the total value of firm assets in RMB at the end of 2007. A natural logarithm transformation is used to reduce the bias in the coefficient estimation. Firm age is the age of a firm from founding to the end of 2007. Financial performance is measured by the ‘return on assets’ in 2007. Since this ratio is highly skewed, we winsorized this variable at the upper and lower 1% tails of the distribution. Leverage is measured by the ratio of total debt to the total value of assets at the end of 2007. Age is a business owner’s age as of 2007. Sex is the gender of business owners, with male business owners being coded as 1, and female business owners as 0. In this study, the sample firms are located mainly in industries like manufacturing; wholesale and retailing; agriculture; construction and real estate; information service; mining; hotel and catering; and other social services. Thus, we included seven industry dummies to control for industry-related variations.
Analysis and Results
Characteristics of the sample
Characteristics of the sampled firms and business owners are summarized in Table 1. In this sample, 1,302 firms conducted philanthropic giving in 2007, which accounts for 64.5% of the total sample. Giving amount ranges from 0 to RMB 50 million, with a mean of RMB 108,000.9. Business owners’ AcSS, when measured by a composite index, ranges from 1 to 4, with a mean of 2.1.
Characteristics of the sample
Note. N=2,019.
a Lot of firms may have the same level of income or individual giving at each quartile, which explains that numbers of the four levels of firms are not 25% of the total sample.
AcSS=achieved social status; ROA, return on assets.
In total, 670 (or 33.2%) respondents reported an annual income of RMB 50,000 or below; 541 (or 26.8%) reported above RMB 50,000 but no more than RMB 100,000; 420 (or 20.8%) reported above RMB 100,000 but no more than RMB 200,000; and 388 (or 19.2%) reported over RMB 200,000. In total, 808 (or 40%) respondents had received education up to senior middle school; 553 (or 27.4%) had received a junior college degree; 422 (or 20.9%) had received a bachelor degree; and 236 (or 11.7%) had received a postgraduate degree. In total, 1,118 respondents had below county-level political ties or no political ties; 519 (or 25.7%) had county-level ties; 317 (or 15.7%) had prefecture level ties; and 65 (or 3.2%) had province level or higher political ties. In total, 506 (or 25.1%) respondents had donated RMB 7,000 or below in the past; 627 (or 31.1%) had donated more than RMB 7,000 but no more than RMB 50,000; 407 (or 20.2%) had donated more than RMB 50,000 but no more than RMB 200,000; and 479 (or 23.7%) had donated more than RMB 200,000.
Firm visibility ranges from 0 to 26.55%. Business owners’ equity level ranges from 1 to 100%, with a mean of 66.93%. Firm size, measured by the value of total assets, ranges from RMB 20,000 to RMB 2.05 billion. Firm age ranges from 1 to 27. The highest return on assets was 277.78%, while the lowest was −20.25%, with a mean of 18.75%. Leverage of firms ranged from 0 to 99.24%, with a mean of 20.72%. Firms were located mainly in manufacturing, followed by wholesale and retailing, construction and real estate, agriculture, and other industries. The oldest business owner was 83 years old, while the youngest was 20. In total, 1,727 business owners were male, which accounts for 85.5% of all the respondents.
Descriptive statistics
Table 2 summarizes the means, standard deviations, and Pearson correlations among the variables. From the Pearson correlations we can see that all the independent variables including all the control variables except equity level are significantly associated with the dependent variable, CP. We also find significant correlations between most of the independent variables. To mitigate the potential multicollinearity, we mean-centered all the independent variables that are included in the interaction terms and then created interaction terms by multiplying the relevant mean-centered variables (Aiken & West, Reference Aiken and West1991; Li, Poppo, & Zhou, Reference Li, Poppo and Zhou2008). In addition, we also followed the ‘coldiag’ procedure recommended by Belsley, Kuh, and Welsch (Reference Belsley, Kuh and Welsch1980) to check the multicollinearity problem. Results show that the largest condition number is 28.13, below the threshold of 30 (Belsley, Kuh, & Welsch, Reference Belsley, Kuh and Welsch1980), suggesting that multicollinearity is not serious in this study.
Descriptive statistics and Pearson correlation matrix
Note. N=2,019.
AcSS=achieved social status; ROA, return on assets.
Correlations ≥|0.048| are significant at the p<.05 level (two-tailed).
Regression analyses and results
The dependent variable, CP, is a continuous random variable with nonnegative values. There is a corner solution because some firms did not donate to social causes. For such a situation, a censored Tobit regression model is a good choice (Tobin, Reference Tobin1958). As a result, we conducted Tobit analyses to test our hypotheses. Table 3 reports the results when the business owners’ AcSS is measured by a composite index (AcSS index); while Table 4 reports the results when AcSS is measured by separate variables including income, education, political ties, and individual donations.
Tobit regression results for the achieved social status (AcSS) index
Note. Figures in parentheses are standard errors. Industry dummies are included in each model.
ROA, return on assets.
*p<.05, **p<.01 (two-tailed).
Tobit regression results for the achieved social status (AcSS) components
Note. Figures in parentheses are standard errors. Industry dummies are included in each model.
ROA, return on assets.
*p<.05, **p<.01 (two-tailed).
In Table 3, keeping with usual practices, Model 1 includes only control variables and moderators. In Model 2 we add the explanatory variable, AcSS index. In Models 3 and 4 we further test the moderating effects of firm visibility and equity level. Model 5 is the full model including all the independent variables and interaction terms. In Table 4, Models 1–4 are used to examine the main effect of income, education, political ties, and individual donations, respectively. Models 5 and 6 further test the moderating effect of firm visibility and equity level, respectively. We test the moderating effect of firm visibility and equity level separately because the potential high correlations between interaction terms may overinflate the standard error of the coefficient estimates and lead them to be insignificant (Sheng, Zhou, & Li, Reference Sheng, Zhou and Li2011).
Hypothesis 1 predicts a positive relationship between a business owner’s AcSS and CP. In Model 2 of Table 3, the coefficient of AcSS index is positive and significant (β=3.424, p<.01). The results are in line with our expectations. Thus, Hypothesis 1 is supported. When we consider AcSS elements separately, the results in Table 4 show that income (β=0.962, p<.01, see Model 1), political ties (β=0.862, p<.01, see Model 3), and individual donations (β=3.114, p<.01, see Model 4) are significantly and positively associated with CP, while the coefficient of education is not significant. It suggests that the effect of AcSS index on CP is contributed mainly by those three variables.
Hypothesis 2 predicts that firm visibility strengthens the relationship between a business owner’s AcSS and CP. In Model 3 of Table 3, the coefficient of the interaction between AcSS index and firm visibility is significant but negative (β=−0.455, p<.01). The results are contrary to our expectations. Thus, Hypothesis 2 is not supported. This will be discussed later. When we look into the different variables making AcSS, only the coefficient of the interaction term between firm visibility and individual donations is significant though negative (β=−0.420, p<.01, see Model 5 in Table 4). It suggests that among the four indices of AcSS, the effect of individual donations on CP is the only one affected significantly by firm visibility. This again will be discussed later.
Hypothesis 3 predicts a business owner’s equity level in his or her firm strengthens the relationship between AcSS and CP. In Model 4 of Table 3, the coefficient of the interaction term between AcSS index and equity level is not significant. Thus Hypothesis 3 is not supported. However, when we look into the different elements of AcSS, the coefficient of the interaction term between education and equity level is significant and positive (β=0.308, p<.05), in line with our expectations.
It is argued that the validity of significant interaction terms may not be simply judged by its coefficient (Norton, Wang, & Ai, Reference Norton, Wang and Ai2004). For linear models, the interaction effect is also captured by drawing a graph considering low (mean−1 SD) and high (mean+1 SD) values of the dependent and moderating variables. But in nonlinear models, this cannot be done. Thus, we follow Ai and Norton’s (Reference Ai and Norton2003) suggestion to examine the marginal effect of the significant interaction terms. The ‘margins’ command (Stata 11.0 edition) is conducted. All the marginal effects of significant interactions are still significant and their directions did not change, suggesting our results are robust.
Conclusion and Discussion
Using the data of a national survey sample, covering private enterprises in China, this study examines the effect of business owners’ AcSS on their firms’ philanthropic giving. We find, as we expected, that a business owner’s AcSS has a significant and positive effect on CP. A business owner who has a higher level of income, political ties, and individual donations, put more organizational resources to support social causes. However, the effect of education is not significant. This may be attributed to the rapid development of professional education, such as Master of Business Administration and Executive Master of Business Administration, in China during the past decade. Receiving high education is much easier for many business owners today than in the past, and may not be a real factor in social status. In the general perception, it is also possible that education is conflated with income and ownership.
In addition, firm visibility weakens the effect of AcSS on CP. The specific effect of individual donations is the only element of AcSS which is affected, but again negatively. This came as a surprise. We were expecting AcSS, and its components to be moderated positively. The negative moderation could be explained by two possibilities: (1) high firm visibility means high advertising expenses, which reduces a firm’s resources that could be used for social goals. In a sense, there is competition between visibility, as measured here, and giving. (2) It has been also argued that CP and advertising are substitutes, contributing both to visibility, and thus competing for the same expenditure budget (McWilliams & Siegel, Reference McWilliams and Siegel2000; Hull & Rothenberg, Reference Hull and Rothenberg2008). Burt (Reference Burt1983) and Gardberg and Fombrun (Reference Gardberg and Fombrun2006), among others, have also argued that CP, advertising, and research and development, all contribute to a differentiation strategy. The possible substitution effect leads firms who have invested more in advertising to contribute less to charity.
Finally, a business owner’s equity level shows no significant moderating effect on the relationship between the composite AcSS index and CP. This is also inconsistent with our expectations. It may be attributed to the high average equity level (66.93%) in our sample. There may be little variance among owners. It can also be argued that beyond a certain (high) level of equity, there is a limited effect on decision-making power. Looking into more details, it is interesting to observe that the equity level does strengthen the relationship between education and CP. This means that the positive effect of owners’ education on CP is larger when the equity level is higher. This appears to make sense. If owners’ education enhances organizational CP, wealthier owners would push for more CP than less wealthy ones. This effect appears to cumulate the effect of wealth, already shown in the literature, with the effect of education.
These findings confirm our theoretical contention that AcSS acts as a resource and an institutional context. As a resource, it distinguishes among individuals and their firms, making some more attractive than others. The higher the owners’ AcSS, the more likely the presence and power within the community, and the more likely their firm will benefit from many economic and noneconomic advantages as argued earlier. Also, for our purpose here, the higher the owners’ AcSS, the more sensitive they would be to social and institutional pressures, and the more likely their firms would contribute to philanthropic giving. Owners’ AcSS appears therefore as an important marker of CP, and of its effect on the firm’s distinctive behavior. This is the first and most important contribution of this research.
This study has significant other theoretical implications. First, it contributes to the literature on CP by linking business owners’ characteristics – in particular AcSS – to organization-level philanthropic giving. As discussed at the beginning of this paper, prior studies have not paid much attention to the role of business owners’ characteristics in CP (Campbell, Gulas, & Gruca, Reference Campbell, Gulas and Gruca1999; Gomez-Mejia et al., Reference Gomez-Mejia, Cruz, Berrone and DeCastro2011; Du et al., Reference Du, Jian, Du, Feng and Zeng2014), not to mention the role of AcSS. Our study makes up to some extent for this gap, by examining and providing evidence for the effect of business owners’ status on CP.
Second, this study extends the application of AcSS in explaining organizational behavior. Prior studies focus mainly on the effect of social status on lifestyle choices (Contoyannis & Jones, Reference Contoyannis and Jones2004), health risk behaviors (Goy, Rosenberg, & King, Reference Goy, Rosenberg and King2008), education attainment (Chevalier & Lanot, Reference Chevalier and Lanot2002), and the like. Only a handful of studies have examined the effect of chief executive officers’ social status on corporate business behaviors such as research and development investment (Shemesh, Reference Shemesh2010) and acquisition (Lucey, Plaksina, & Dowling, Reference Lucey, Plaksina and Dowling2013). Our study suggests that social status could not only be used to explain corporate business behavior, but also be used to predict or explain corporate social behavior.
Third, this study documents the constraining side of social status or social power. Although the positive effect of social status or power is widely discussed and recognized in social interactions, its constraining side is to a large extent overlooked. Our study suggests that business owners’ high AcSS leads to high pressures or expectations from the public, which further forces them to contribute more to social causes at the organizational level. The findings not only support an institutional viewpoint of CP (Oliver, Reference Oliver1991), but also provide support for Davis’ (Reference Davis1960) argument that social responsibilities come from social power, since social status and social power are highly related.
There are limitations of this study. At first, all the data used were self-reported by the investigated business owners. Although self-reported data were widely used in studies on CP, it is hard to fully reject the risk of running into a social desirability bias. Second, all the data used in this study comes from the same questionnaire. Although, as discussed previously, a common method bias risk is small, it is hard to exclude it altogether. Third, we have only considered the AcSS and excluded the AsSS. However, it is interesting and perhaps necessary to consider both types of social status simultaneously, as they probably overlap and interact widely. Fourth, our study focuses solely on China. Since China is different from other countries in many aspects, we should be cautions when extending our findings to other countries. On the one hand, the definition of SMEs in China is different from most Western developed country standards. This could also affect the generalizability of the findings. On the other hand, social status tends to be country or culture specific. This suggests that social status should consider country and perhaps situational differences. In this respect, it is important to investigate the effects of business owners’ other characteristics (such as personality, leadership styles, etc.) on CP. At last, our study examined the situations of CP by 2007. In recent years, with the remarkable growth of internet-based social media, it becomes much easier and faster for business owners and firm social behavior to be exposed to the public; we expect an even stronger relationship between social status and CP, as higher visibility breeds higher pressures on business owners. These topics call for further and interesting studies in the corporate social responsibility field.
Acknowledgement
Yongqiang Gao is grateful to the financial support of the National Natural Science Foundation of China (NSFC) (No. 71372131, 71531009), and Xiaobin He is grateful to the financial support of the NSFC (No. 71472050).
