INTRODUCTION
Family-controlled firms are those in which ‘a family owner exercises much influence over the firm's affairs’ (Gomez-Mejia, Cruz, Berrone, & De Castro, Reference Gomez-Mejia, Cruz, Berrone and De Castro2011: 658). This form of business is ubiquitous and numerically dominant throughout the world (Au, Craig, & Ramachandran, Reference Au, Craig and Ramachandran2011; Gomez-Mejia, Larraza-Kintana, & Makri, Reference Gomez-Mejia, Larraza-Kintana and Makri2003; La Porta, Lopez-de-Silanes, & Shleifer, Reference La Porta, Lopez-de-Silanes and Shleifer1999; Lu, Au, Peng, & Xu, Reference Lu, Au, Peng and Xu2013). One of the key differences between family-controlled and nonfamily-controlled firms is that the former are more inclined to sacrifice economic gains to accumulate or maintain the noneconomic utility – or, formally, socioemotional wealth (SEW) – which a family would derive from its association with the firm (Gómez-Mejía, Haynes, Núñez-Nickel, Jacobson, & Moyano-Fuentes, Reference Gómez-Mejía, Haynes, Núñez-Nickel, Jacobson and Moyano-Fuentes2007). Studies have shown that SEW consideration is the primary driver of many behaviors of family-controlled firms, including corporate diversification, environmental performance, initial public offering underpricing, and investment in research and development (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, Reference Berrone, Cruz, Gomez-Mejia and Larraza-Kintana2010; Chrisman & Patel, Reference Chrisman and Patel2012; Gomez-Mejia et al, Reference Gómez-Mejía, Haynes, Núñez-Nickel, Jacobson and Moyano-Fuentes2007; Gomez-Mejia, Makri, & Larraza-Kintana, Reference Gomez-Mejia, Makri and Larraza-Kintana2010; Leitterstorf & Rau, Reference Leitterstorf and Rau2014).
Such studies are valuable for showing the uniqueness of family-controlled firms, but they imply that family-controlled firms ‘are homogeneous in their emphasis on SEW considerations’ (Berrone, Cruz, & Gomez-Mejia, Reference Berrone, Cruz and Gomez-Mejia2012: 270). As not all family-controlled firms emphasize SEW to the same extent, the studies do not give a systematic account of the variation among family-controlled firms (Chrisman & Patel, Reference Chrisman and Patel2012; Chua, Chrisman, & Sharma, Reference Chua, Chrisman and Sharma1999; Naldi, Cennamo, Corbetta, & Gomez-Mejia, Reference Naldi, Cennamo, Corbetta and Gomez-Mejia2013). Even though recent studies have suggested that family-controlled firms vary greatly regarding their behaviors driven by SEW considerations (Chrisman & Patel, Reference Chrisman and Patel2012; Chua, Chrisman, Steier, & Rau, Reference Chua, Chrisman, Steier and Rau2012), few studies have assessed or accounted for this heterogeneity (Chrisman & Patel, Reference Chrisman and Patel2012).
Another deficiency in studies applying the SEW perspective is that they used family ownership as a proxy for the existence of SEW, thus ignoring the possibility that family owners might have different levels of motivation to accumulate or maintain SEW. Failing to account for family owners’ varying motivations in our progress toward a theory of SEW is unfortunate, because according to motivation theories, individuals vary greatly regarding their motivations to behave in a particular way (Latham & Pinder, Reference Latham and Pinder2005). To the extent that family owners are the key decision makers in family-controlled firms, an examination of their motivations would contribute to a better understanding of the behaviors that are driven by SEW considerations in family-controlled firms.
This study is an attempt to fill these research gaps with a focus on the corporate philanthropy of family-controlled firms in China (Cui, Liang, & Lu, Reference Cui, Liang and Lu2015). Uniting the SEW perspective with expectancy theory, a classic motivation theory on individual decision making (Vroom, Reference Vroom1964), we propose that family owners’ motivation to engage in corporate philanthropic activities is determined by the perceived probability that corporate philanthropy will lead to a positive outcome (i.e., expectancy) and by the attractiveness of that positive outcome (i.e., valence). Following this logic, we propose that corporate philanthropy is likely to contribute to a positive reward, namely, SEW, and family owners with intrafamily succession intention tend to place a higher value on such a reward. As a result, these family owners are motivated to make more charitable contributions. Moreover, family owners with high social status will raise their expectation that corporate philanthropy is closely related with family SEW (Yiu, Wan, Ng, Chen, & Su, Reference Yiu, Wan, Ng, Chen and Su2014). Hence, family owners’ social status will strengthen the relationship between intrafamily succession intention and corporate philanthropy. In addition, we posit that family owners with religiosity are likely to be intrinsically motivated and value corporate philanthropy itself rather than the extrinsic reward derived from philanthropic activities (i.e., SEW). As a result, intrafamily succession intention will be a weaker motivating factor influencing corporate philanthropy when family owners are already strongly motived by religiosity.
This study makes three main contributions to the literature. First, building on insight from expectancy theory, we propose that family owners’ motivations are key determinants of heterogeneity among family-controlled firms. In particular, we identify that the behaviors of family-controlled firms can be driven by both extrinsic and intrinsic motivations. These findings help to build a stronger theoretical foundation for understanding the behavior of family-controlled firms.
Second, our study contributes to the research on SEW by taking into account both family owners’ ability and motivations to accumulate and maintain SEW. Prior studies on SEW focused mainly on family involvement as a key determinant of behavior related to SEW considerations, as that factor provides family owners the ability to pursue SEW. The results of this study demonstrate that without understanding family owners’ motivations, one may fail to fully grasp the behaviors of family-controlled firms.
Third, this study takes into account the institutional context in China. The recent institutional developments in China have greatly enhanced the social status of family owners and have resulted in increasing numbers of family owners publicly declaring their religious affiliations. These developments provide a unique context for testing the interplay of ability and motivation for family firms’ particularistic behavior.
INSTITUTIONAL BACKGROUND
In the past few decades, the institutional background in China, particularly that related to the growth of privately owned firms, has changed dramatically. Table 1 presents the main institutional changes during this period. As all family-controlled firms are privately owned firms, a discussion of these institutional changes will help us understand the behaviors of family-controlled firms in China.
The Evolving Status of Private Business Owners in China
In traditional Chinese society, which was greatly influenced by Confucianism, private business owners were of relatively low social status. In Confucian teachings, society is divided into the scholar (shi), farmer (nong), artisan (gong), and merchant (shang) classes. Accordingly, private business owners and merchants have traditionally been placed toward the bottom of society in Chinese culture, viewed as not producing anything and only profiting from others’ creations (Mencius, 1895).
The social status of private business owners did not improve after the Communist Party of China (CPC) won the civil war and founded the People's Republic of China (PRC) in 1949. Between 1952 and 1977, the overall political environment was antagonistic toward the private sector, which was regarded as the ‘tail of capitalism’ and completely banned in China. Although private business was revived after the 11th Central Committee's 3rd Plenum, private business owners were still treated with suspicion and even considered despicable in the 1980s. The private ownership of firms was considered to be an inferior form of business ownership for ideological reasons.
This political and social discrimination against private business owners has gradually decreased since the early 1990s. Ideological opposition to private enterprise has subsided and the private sector has been accorded equal status with other sectors. One of the most significant changes in the political status of private business owners occurred during the CPC's 16th Congress in 2002 when the membership charter was revised and private business owners were formally allowed to apply for CPC membership. This loosening of the ideological and political constraints in China raised the social status of private business owners to such an extent that an increasing number of government employees have quit their jobs to start their own businesses.
State Policies toward Religion
Despite the relaxed ideological and political constraints, the Chinese government still maintains tight control over the growth and scope of religious activities. Before the founding of the PRC in 1949, the state maintained loose control over religious groups. Almost all religions, including foreign ones, were tolerated, as long as they did not threaten the stability of the state. Some emperors even had religious affiliations. For example, Emperor Jiajing in the Ming Dynasty was a devoted follower of Taoism, and Emperor Shunzhi in the Qing Dynasty became a devout follower of Buddhism in 1657.
When the PRC was founded in 1949, religious activities were suppressed by the CPC, which is officially atheistic. During the Cultural Revolution from 1966 to 1976, many religious sites and artifacts in China were damaged, and religious scriptures were burned. Monks and priests were ‘wiped away like dust’ through imprisonment and re-education in labor camps.
In the post-1979 era, the pragmatic CPC began to loosen control over various social activities, including religion. In the 1980s, suppression was replaced with tolerance, and a limited number of places for worship, including Protestant and Catholic churches, Buddhist and Taoist temples, and Islamic mosques, were allowed to reopen for religious services.
THEORETICAL BACKGROUND AND HYPOTHESES
SEW Perspective and the Behaviors of Family-controlled Firms
According to the SEW perspective, family-controlled firms place a higher value on SEW than do nonfamily-controlled firms (Berrone et al., Reference Berrone, Cruz and Gomez-Mejia2012). As a result, they are likely to be very interested in keeping control over their firms, perpetuating the family dynasty, and sustaining the family image and reputation. To accumulate or maintain SEW, family owners are willing to engage in behavior that is not driven by economic logic (Berrone et al., Reference Berrone, Cruz, Gomez-Mejia and Larraza-Kintana2010; Gomez-Mejia et al., Reference Gómez-Mejía, Haynes, Núñez-Nickel, Jacobson and Moyano-Fuentes2007).
Recent studies on family-controlled firms have relied frequently on the SEW perspective to predict the differences in the strategic choices and practices of family-controlled and nonfamily-controlled firms (Chrisman & Patel, Reference Chrisman and Patel2012; Gomez-Mejia et al., Reference Gomez-Mejia, Cruz, Berrone and De Castro2011). Focusing mainly on differences in firm behavior between family-controlled and nonfamily-controlled firms (Patel & Chrisman, Reference Patel and Chrisman2014), they have assumed that family-controlled firms ‘are homogeneous in their emphasis on SEW considerations’ (Berrone et al., Reference Berrone, Cruz and Gomez-Mejia2012: 270). However, the family business literature suggests that in some situations, the variations among family-controlled firms can be as large as, if not larger than, the differences between family-controlled and nonfamily-controlled firms (Chrisman & Patel, Reference Chrisman and Patel2012; Chua et al., Reference Chua, Chrisman, Steier and Rau2012). Big differences of emphasis in SEW may actually exist among family-controlled firms.
In addition, most of these studies have used family ownership as a proxy for the existence of SEW (Berrone et al., Reference Berrone, Cruz and Gomez-Mejia2012; Chrisman & Patel, Reference Chrisman and Patel2012). Family ownership can be a valid approximation of SEW when using archival data (Naldi et al., Reference Naldi, Cennamo, Corbetta and Gomez-Mejia2013). However, studies have recently suggested that not all family owners have the same level of motivation to accumulate and maintain SEW (Chrisman & Patel, Reference Chrisman and Patel2012). Hence, directly studying the motivations of family owners might better capture the decision-making process related to SEW considerations (Kellermanns, Eddleston, & Zellweger, Reference Kellermanns, Eddleston and Zellweger2012; Naldi et al., Reference Naldi, Cennamo, Corbetta and Gomez-Mejia2013).
Drivers of Charitable Donations
Expectancy theory explains the process individuals use to make decisions regarding behavioral alternatives (Vroom, Reference Vroom1964). Organizational psychology literature includes a number of different motivational theories; we selected expectancy theory to derive our hypotheses for several reasons. First, this theory has been used for understanding prosocial behavior and organization citizenship behavior (Grant & Berg, Reference Grant, Berg, Cameron and Spreitzer2011; Grant & Mayer, Reference Grant and Mayer2009); thus, it is useful for theorizing how family owners’ motivations may affect their engagement in corporate philanthropic activities. Second, expectancy theory provides a comprehensive framework for assessing individual behaviors (Fudge & Schlacter, Reference Fudge and Schlacter1999). It has the advantage of identifying and systematically relating the various attributes of family owners to their likelihood of making charitable donations. Third, this theory has been subjected to rigorous empirical testing and is one of the most commonly used theories of motivation in organizational settings (Chiang & Jang, Reference Chiang and Jang2008).
According to expectancy theory, there are two basic prerequisites underlying an individual's motivation to engage in a behavior: expectancy and valence (Chen & Miller, Reference Chen and Miller1994; Fudge & Schlacter, Reference Fudge and Schlacter1999; Vroom, Reference Vroom1964). Expectancy is the perceived probability that effort in an activity will lead to desired outcomes (Vroom, Reference Vroom1964). Valence refers to the subjective reward value or the attractiveness of the behavioral outcome (Chen & Miller, Reference Chen and Miller1994; Chiang & Jang, Reference Chiang and Jang2008). Because expectancy and valence are both prerequisites of a behavior, these two factors combine multiplicatively to determine an individual's motivation to behave. In general, empirical studies on individual motivation in the workplace have provided strong support for this theory (Fudge & Schlacter, Reference Fudge and Schlacter1999).
In this study, we use expectancy theory and the SEW perspective to argue that family owners are heterogeneous in their motivations for engaging in corporate philanthropic activities. In particular, family owners with intrafamily succession might have stronger motivations to make charitable contributions because making charitable contributions is highly likely to contribute to a positive outcome (i.e., family image and reputation), and these family owners place a high value on such outcome. In the language of expectancy theory, since both the effort-outcome expectancy and outcome valence are high, these family owners are motivated to make charitable donations.
Moreover, not all family owners have the same effort-outcome expectancy. Only those who are confident that corporate donations have a high likelihood of leading to SEW will engage in such philanthropic activities (Tsiros, Mittal, & Ross, Reference Tsiros, Mittal and Ross2004; Vroom, Reference Vroom1964). Following this logic, as there would be more social monitoring and scrutiny from company stakeholders, family owners who are of high social status, we propose, would tend to believe that corporate philanthropy is closely related to family image and reputation (Zellweger, Nason, Nordqvist, & Brush, Reference Zellweger, Nason, Nordqvist and Brush2013). For this reason, having a higher social status is likely to strengthen the relationship between intrafamily succession intention and corporate philanthropy.
The foregoing arguments are based on the assumption that family owners are self-interested individuals who desire to accumulate and maintain SEW. In other words, family owners are motivated to engage in corporate philanthropic activities mainly because of the extrinsic value derived from such activities (Cennamo, Berrone, Cruz, & Gomez-Mejia, Reference Cennamo, Berrone, Cruz and Gomez-Mejia2012). However, family owners can also be interested in charity in and of itself, and they may derive intrinsic satisfaction by engaging in corporate philanthropic activities. In particular, we propose that those with religious faith tend to value corporate philanthropy intrinsically; thus, they have a lower valence for SEW that is generated from corporate philanthropy. In this sense, religiosity will weaken the relationship between intrafamily succession intention and corporate philanthropy.
Based on these arguments, this research focuses on three factors that are expected to influence family owners’ motivation to make charitable contributions: their intrafamily succession intention, social status, and religiosity. Figure 1 summarizes the proposed relationships between these concepts.
Hypotheses
In family-controlled firms, family owners with intrafamily succession intention are motivated to engage in corporate philanthropic activities, because (1) corporate philanthropy is a critical way to enhance family image and reputation and (2) family image and reputation, a key element of SEW, is highly valued by these family owners.
First, family-controlled firms that make substantial philanthropic contributions are likely to promote a socially responsible public image (Godfrey, Reference Godfrey2005; Wang & Qian, Reference Wang and Qian2011). Corporate philanthropy, the top of the corporate social responsibility pyramid, plays a significant role in establishing and developing corporate reputations (Brammer & Milington, Reference Brammer and Millington2005; Dutton, Dukerich, & Harquail, Reference Dutton, Dukerich and Harquail1994; Saiia, Carroll, & Buchholtz, Reference Saiia, Carroll and Buchholtz2003). Given the close and often inseparable ties between family and the firm in family-controlled firms, corporate philanthropy will often serve as a crucial way to enhance family image and reputation (Zellweger et al., Reference Zellweger, Nason, Nordqvist and Brush2013).
Second, family owners with intrafamily succession intention place great emphasis on family image and reputation (Zellweger, Kellermanns, Chrisman, & Chua, Reference Zellweger, Kellermanns, Chrisman and Chua2011). According to the SEW perspective, intrafamily succession intention implies that the controlling family is oriented toward a long-term vision that ‘extends beyond the lifespan of the current generation’ (Chrisman, Chua, Pearson, & Barnett, Reference Chrisman, Chua, Pearson and Barnett2012: 274). For this reason, as the importance of intrafamily succession increases, family owners are more likely to take steps to develop and perpetuate a positive image and reputation that is critical for long-term success (Berrone et al., Reference Berrone, Cruz and Gomez-Mejia2012; Chrisman & Patel, Reference Chrisman and Patel2012).
Thus, as corporate philanthropy is likely to lead to a positive outcome (i.e., family image and reputation) and this outcome is highly attractive to family owners with an intrafamily succession intention, we hypothesize:
Hypothesis 1: Family owners’ intrafamily succession intention will relate positively to their corporate philanthropy.
While family owners’ intrafamily succession intention is expected to be associated with more charitable contributions of family-controlled firms, we draw upon the expectancy theory to explain how the two individual factors – social status and religiosity – can act as contingency factors qualifying the association. We consider the impact of family owners’ social status first.
Social status refers to an individual's publicly acknowledged social esteem and social worth relative to other individuals in a social hierarchy (Chen, Peterson, Phillips, Podolny, & Ridgeway, Reference Chen, Peterson, Phillips, Podolny and Ridgeway2012). Private business owners had relatively low social status in traditional Chinese society, because, according to the Confucian scholars who did not appreciate the economic contributions of the merchants’ risk-taking and efficiency in facilitating the flow of goods, they selfishly harmed the common people who produced in the fields. The public image of private business owners was even worse between 1949 and 1977 when private businesses were denigrated as the ‘tail of capitalism’. However, the economic reforms announced by the CPC in 1978 greatly improved the social status of private business owners (Au & Sun, Reference Au and Sun1998). A large number of such owners are now members of the People's Congress (PC), the highest organ of state power, and the Chinese People's Political Consultative Conference (CPPCC), the top political advisory body in China (Li and Liang, in press).
Drawing on expectancy theory, we propose that the social status of family owners increases their expectancy that corporate philanthropy is closely related to SEW, particularly regarding family image and status, and thus strengthens the relationship between intrafamily succession intention and corporate philanthropy. When family owners have high social status, they are well known to individuals in their communities (Ma & Parish, Reference Ma and Parish2006; Yiu et al., Reference Yiu, Wan, Ng, Chen and Su2014; Zellweger et al., Reference Zellweger, Nason, Nordqvist and Brush2013). As a result, when family-controlled firms make charitable contributions, their philanthropic behavior is more likely to be noticed and, thus, to contribute to family image and reputation (Berrone et al., Reference Berrone, Cruz, Gomez-Mejia and Larraza-Kintana2010; Zellweger et al., Reference Zellweger, Nason, Nordqvist and Brush2013). Accordingly, family owners with high social status would expect that corporate philanthropy is closely related to family SEW. Hence, high status family owners with intrafamily succession intention will have greater motivation to make charitable donations to maintain or increase their individual and family status.
By contrast, when family owners are of relatively low social status, the relationship between intrafamily succession intention and corporate philanthropy might become weaker. These family owners experience less social monitoring, and their charitable contributions are less likely to be noticed (Zellweger et al., Reference Zellweger, Nason, Nordqvist and Brush2013). Thus, they might have a lower corporate philanthropy–SEW expectancy. Consequently, even if they have intrafamily succession intention, they might be less motivated to make charitable contributions, as they expect that there is only a weak link between corporate philanthropy and family image and reputation. Hence, we hypothesize:
Hypothesis 2: Family owners’ social status will moderate the positive relationship between family owners’ intrafamily succession intention and corporate philanthropy. Specifically, the positive relationship between family owners’ intrafamily succession intention and corporate philanthropy will be stronger for family owners with high social status than for those of low social status.
While the state continues to maintain tight control over religious activities in China, in recent years, more family owners have been willing to publicly profess their religious affiliations. For example, Mr. Pan Shiyi and Ms. Zhang Xin, who cofounded the largest prime office real estate development company in China, SOHO China, wrote on their blogs that they had converted to the Baha’i faith. Similarly, Mr. Huang Rulun, the founder of Century Golden Resources Group, publicly announced his affiliation with Buddhism. Indeed, when we calculated the percentage of family owners professing attachment to a particular religion based on the nationwide survey of privately owned enterprises that we used to test our hypotheses, we found that this percentage had increased from 16.9% to 21% in just two years, from 2008 to 2010.
Family owners with religiosity, we argue, are likely to value corporate philanthropy intrinsically. Although all branches of religion may have different behavioral codes, one trait they have in common is encouraging charity and caring for the poor (McGuire, Newton, Omer, & Sharp, Reference McGuire, Newton and Sharp2012). For example, Buddhist scripture encourages universal compassion, which refers to an active willingness to bear the pain of others (Norberg-Hodge, Reference Norberg-Hodge1997). Similarly, in Christianity, charity is held to be the ultimate perfection of the human spirit. Through daily exposure to religious norms, individuals who identify with particular religious denominations are likely to follow these norms and see charity as an end in and of itself (Parboteeah, Hoegl, & Cullen, Reference Parboteeah, Hoegl and Cullen2007; Ramasamy, Yeung, & Au, Reference Ramasamy, Yeung and Au2010).
This is particularly true in China, where religious affiliation is more of a personal choice than the result of a parental decision or social calling. Because very few religious believers and practitioners remained when the government began to loosen control over religion in 1978, individuals in China, especially family owners, were unlikely to have acquired their religious views from their parents (Lai, Reference Lai2003). As a result, compared to their counterparts in Western countries, family owners in China who are affiliated with a particular religion are more likely to have internalized their religious teachings and to have a genuine interest in making charitable contributions.
Drawing upon expectancy theory, we propose that because family owners with religious faith value corporate philanthropy for itself rather than for the utilitarian benefits derived from it, there will be a weaker relationship between intrafamily succession intention and corporate philanthropy. Specifically, these family owners are likely to donate for intrinsic reasons and rely much less on extrinsic rewards for their motivation (McGuire et al., Reference McGuire, Newton and Sharp2012). That is, when family owners are religious, it is less likely that extrinsic reward (i.e., SEW) will be the main driver of their corporate philanthropy, and the relationship between intrafamily succession intention and corporate philanthropy will be weaker. As a case in point, Mr. Huang Rulun, a Buddhist and number two on the Hurun list of top philanthropic donors in China, stated that he firmly believed that charity should be spontaneous and not conducted for personal glory or gain. Hence, based on this discussion, we hypothesize:
Hypothesis 3: Family owners’ religiosity will negatively moderate the positive relationship between family owners’ intrafamily succession intention and corporate philanthropy. Specifically, the positive relationship between family owners’ intrafamily succession intention and corporate philanthropy will be weaker when family owners are religious than when they are not.
METHOD
Empirical Setting
Our main data source was a nationwide survey conducted in 2010 of privately owned Chinese enterprises. The survey provided a reliable database on privately owned businesses in China and has been used in studies published in leading journals (e.g., Jia, Reference Jia2014; Zhou, Reference Zhou2013). Starting in 1991, the research team conducted a nationwide survey approximately every two years (1991, 1993, and so on until 2010). Unfortunately, the enterprises selected in the previous surveys were not tracked in the latter ones, so a longitudinal investigation of each firm's corporate social performance was not feasible. Thus, we chose the latest version of the survey, the 2010 National Survey, as the dataset for hypothesis testing. Differing from its previous versions, the 2010 National Survey added items measuring family owners’ intrafamily succession intention; thus, it was appropriate for our hypothesis testing. The response rate for this survey was 95%.
We classify a firm as a family-controlled one when a family owns at least 50% of the shares. Most prior studies on family-controlled firms have been on listed firms and have used 5% as the cutoff point defining family-controlled firms. Our focus was on private enterprises; thus, we required a more stringent criterion. The 50% benchmark ensured that the family owner had the discretion to engage in behaviors enhancing family SEW.
After we removed observations with missing key explanatory variables, the final sample contained 1,923 observations. We also used other ownership percentages as cutoff points in the robustness tests.
Dependent Variable
Corporate philanthropy was assessed by the total amount of charitable donations that a firm made in 2009, scaled by the firm's sales. Such scaling was used to control for the effect of firm size; larger firms are generally able to support higher levels of corporate charity. Corporate philanthropy typically accounts for a small proportion of a firm's sales. This might lead to regression coefficients too small to be interpreted. Hence, we multiplied the scaled measure by 1,000. The amount of charitable giving was measured by a question: How much did your firm donate to charity in 2009? A firm's sales revenue was measured by a question: What was the amount of your company's annual revenue? It should be noted that the value of such a measure of corporate philanthropy can be very large when a firm that made charitable contributions generated extremely low sales revenue. However, the hypotheses testing results were basically the same when we excluded firms from our analyses whose values of corporate philanthropy were extremely high.
Independent and Moderating Variables
Intrafamily succession intention was measured by the question: Do you intend to give the next generation ownership or management control of the business? This is a dummy variable, coded 1 if the respondent intends to pass on management or ownership control to the next generation, and 0 otherwise.
We used family owners’ political representation in the PC and CPPCC to assess the family owners’ social status, using a dummy variable coded as 1 if a family owner was a member of the PC or CPPCC, and 0 otherwise. The PC is the highest organ of state power in China, whereas the CPPCC is the top political advisory body. Hence, membership in the PC or CPPCC reflects a family owner's social status in the community (Li & Liang, in press). Furthermore, this measure was less subject to individual bias, as the family owners were required to simply report in the questionnaire whether they were members of the PC or CPPCC or not.
Religiosity was measured by a dummy variable, set to 1 if the family owner was affiliated with a religion, and 0 otherwise.
Control Variables
Several control variables were also included. According to upper echelons theory, family owners’ demographic characteristics can influence key firm decisions. Hence, we controlled for a set of variables concerning demographic characteristics, including age, gender, and education. Owner's age was measured in years. Gender was measured by a dummy variable set to 1 for male and 0 for female. Education was measured by a six-point scale item, asking respondents to indicate their education level (1 = primary school or below; 2 = junior middle school; 3 = senior high school or vocational school; 4 = junior college; 5 = university; 6 = postgraduate). Because family owners’ financial positions may also have a bearing on how much they are willing to donate, we also controlled for family owners’ salaries.
Further, we controlled for a set of firm characteristics. Firm size was measured as the natural log of total employees. Firm age was measured as the number of years since the firm's inception. We also controlled for firm profitability, which has been shown to be associated with corporate philanthropy (Waddock & Graves, Reference Waddock and Graves1997). Firm profitability was measured as the net profit in the year 2008.
In addition, we controlled for family ownership and family management, because both may increase a controlling family's motivation to accumulate or maintain SEW. Family ownership was measured as the percentage of ownership held by the controlling family. Family management is a dummy variable, coded as 1 if one of the family members served as board chair or the CEO of the firm, and 0 otherwise. Finally, 18 industry dummies and 31 region dummies were included in the analyses to control for industry and regional effects.
Estimation Method
In this study, an ordinary-least-squares regression model was used to test the hypotheses. We first entered all of the control variables and then the variable of family owners’ succession intention; interaction effects were added in subsequent models. Because the data we used came from a stratified sample, it was desirable to make statistical adjustments. Accordingly, we clustered standard errors by industry for each regression model.
RESULTS
Table 2 presents the descriptive statistics and a correlation matrix of the variables. The magnitude of the correlations between the independent variables is in the low-to-medium range, suggesting that multicollinearity could be a problem in the testing of the hypotheses. To address this issue, we inspected variance inflation factors (VIFs) in a parallel set of models. The VIFs were well within the limit of 10, indicating that multicollinearity did not seem to have an undue influence on the estimates.
Notes: aNet profit is measured for year t – 1; other variables, for t, n = 1,923. Correlations greater than or equal to |0.05| are significant at p < 0.05. Correlations greater than or equal to |0.06| are significant at p < 0.01.
Table 3 presents the results based on hierarchical multiple regression analysis. Model 1 is the baseline model, including only control variables. Model 2 provides the results of testing the main effect. As presented in Model 1, intrafamily succession intention has a significant and positive influence on corporate philanthropy (β = 8.41, p < 0.01). Hypothesis 1, which predicts that intrafamily succession intention is positively related to corporate philanthropy, is thus supported. It is worth noting that family ownership and family management do not have significant relationships with corporate philanthropy.
Notes: an = 1,923 observations. Bracketed numbers are standardized errors.
b +p < 0.10, *p < 0.05, **p < 0.01, and ***p < 0.001; two-tailed tests.
cAll three models include 18 industry dummy variables and 31 region dummy variables not reported here.
Model 3 in Table 3 shows the results for Hypothesis 2, which predicts that the positive influence of intrafamily succession intention on corporate philanthropy will be heightened by the family owner's social status. The coefficients for the interaction between intrafamily succession intention and family owners’ social status are positive and significant (β = 13.29, p < 0.01). Figure 2 illustrates the moderating effect. Following the procedure proposed by Aiken and West (Reference Aiken and West1991) and Jaccard and Turrisi (Reference Jaccard and Turrisi2003), we performed simple slope tests to examine the relationship between intrafamily succession intention and corporate philanthropy when family owners had high and low social status. The simple slope tests show that intrafamily succession intention is significantly and positively related with corporate philanthropy (p < 0.01) when family owners have high social status (i.e., a social status that is one standard deviation above the mean social status). In contrast, the relationship between intrafamily succession intention and corporate philanthropy is insignificant when family owners have low social status (i.e., a social status that is one standard deviation below the mean social status). Hypothesis 2 is thus supported.
Model 4 in Table 3 presents the results of testing Hypothesis 3, which predicts that the positive effect of intrafamily succession intention on corporate philanthropy will be weaker when the family owners are affiliated with a religion. As predicted, the interaction term between intrafamily succession intention and a family owner's religiosity is negative and significant (β = –28.93, p < 0.01). Figure 3 illustrates the moderating effect. The slope of those with religiosity is flatter than that of those without religious affiliations. A simple slope test shows that intrafamily succession intention is significantly and positively associated with corporate philanthropy (p < 0.01) when family owners are not affiliated with a religion. In contrast, the relationship between intrafamily succession intention and corporate philanthropy is not significant when the family owners are affiliated with a religion. Thus, Hypothesis 3 is supported.
Model 5 in Table 3 shows the results of simultaneously testing Hypotheses 2 and 3 by adding the two interaction terms associated with social status and religiosity. Again, the results provide strong support for the two moderating effect hypotheses.
Robustness CheckFootnote [1]
We examined whether our results were robust to the alternative estimation method, using Heckman's two-stage treatment regression to test our hypotheses. Two-stage treatment effects models are less susceptible to the proliferation of specification errors and distortions. Specifically, the first stage of the procedure is a probit analysis with intrafamily succession intention as the dependent variable. To meet the exclusion restrictions that are necessary for identification in Heckman's model, we included a variable in the probit model that we did not include in the second-stage regression. This was a dummy variable on the next generation's willingness to succeed, coded as 1 if the next generation was willing to take over the family-controlled firm, and 0 otherwise. The results from the treatment effects estimation were qualitatively the same as those reported in Table 3.Footnote [2]
We further verified the robustness of our results by using an alternative measure of social status. In the questionnaire, the family owners were asked to rate their relative social standing in the community by a ten-point scale (1 = highest; 10 = lowest). For the analyses, we transformed this rating by subtracting the original rating from 11. Such a subjective social status measure is common in sociological research (Haug, Reference Haug1977; Ma & Parish, Reference Ma and Parish2006). For example, in a study of private entrepreneurs in China, Ma and Parish (Reference Ma and Parish2006) used the subjective social status to measure the social standings of the family owners. The results of the regression analyses employing this alternative measure of social status were consistent with those reported in Table 3.
Finally, we also used 20% and 30% of firm ownership as the cutoff points to differentiate family-controlled firms from nonfamily-controlled firms. After deleting firms from our sample in which the family owned less than 20% or 30% of the equity, we reran the analyses. The empirical results from these analyses again were qualitatively the same as those reported in Table 3.
DISCUSSION
Recent interest in the role of SEW has contributed to understanding the behaviors of family-controlled firms (e.g., Leitterstorf & Rau, Reference Leitterstorf and Rau2014). We extend this understanding by investigating the variations among family-controlled firms, specifically those related to corporate philanthropy. We found that intrafamily succession intention was significantly related to corporate philanthropy. This finding is consistent with prior studies indicating that this intention is a potential source of heterogeneity among family-controlled firms. Moreover, our empirical results show that intrafamily succession intention plays a much more important role in driving corporate philanthropy when family owners are of relatively high social status and when they are without a religious affiliation. These findings provide support for the argument in prior studies that the personal features of family owners play an important role in explaining their perceptions of SEW (Naldi et al., Reference Naldi, Cennamo, Corbetta and Gomez-Mejia2013) and that ability alone is insufficient to produce family firm particularistic behavior; willingness or motivation must also be present (De Massis, Kotlar, Chua, & Chrisman, Reference De Massis, Kotlar, Chua and Chrisman2014).
Contributions and Implications
This study makes several important theoretical contributions. First, it contributes to the research on family-controlled firms by integrating insights from expectancy theory. As indicated above, a common assumption in empirical studies that have applied the SEW perspective is that family owners are homogeneous in their emphasis on SEW. Although recent studies on SEW have indicated that family owners’ personal features might play a role in influencing their perceptions of SEW and their firm's behaviors (Kellermanns et al., Reference Kellermanns, Eddleston and Zellweger2012), this study is the first to provide specific theoretical and empirical insight into when and how these personal features may shape the behavior of family-controlled firms. Drawing on expectancy theory, we posit that whether a family owner is willing to make charitable contributions is highly dependent on the possibility that these contributions will lead to a positive outcome and attractiveness of this outcome. Accordingly, family owners’ intrafamily succession intention, social status, and religiosity might interact with each other to determine the family owners’ motivation for engaging in corporate philanthropic activities. There have been calls for an examination of the antecedents of heterogeneity among family-controlled firms (Chrisman & Patel, Reference Chrisman and Patel2012; Chua et al., Reference Chua, Chrisman, Steier and Rau2012), and this study is a response, providing a systematic understanding of the motivational factors underlying the behavior of family-controlled firms.
Future research on SEW could benefit by drawing on insight from other psychological theories as well. Zellweger and Dehlen (Reference Zellweger and Dehlen2012) integrated the affect infusion model from cognitive psychology with the SEW perspective and developed a conceptual model explaining how affect might influence family owners’ SEW perception. Our study represents another attempt to use the lens of psychology to explain variations in behavior between family-controlled firms. To the extent that a family owner's decision making related to SEW considerations is a cognitive process, other psychological theories such as goal-setting theory and social cognitive theory might also add value to a better understanding of family SEW.
Second, our study contributes to the research on SEW by suggesting that family involvement in ownership and management may not be a sufficient condition for the family to accumulate and maintain SEW. Indeed, our empirical results in Table 3 show that intrafamily succession intention is more strongly associated with corporate philanthropy than is family involvement in ownership and management. We argue that this occurs because family involvement only provides family members with the ability to maintain or increase SEW, whereas their intrafamily succession intention determines their willingness to do so. Without intrafamily succession intention, family members may not use their influence to pursue family-centered noneconomic goals. To advance this perspective on SEW, future studies should take both family involvement and intrafamily succession intention into consideration.
Third, our findings also contribute to a better understanding of the behavior of family-controlled firms in China. The institutional environment in China has experienced significant changes during the past three decades. Family owners may respond to these changes by adjusting their goals, which, in turn, might have a great impact on their firm's behavior (Peng, Sun, Pinkham, & Chen, Reference Peng, Sun, Pinkham and Chen2009). Since 1978, the Chinese government has gradually abandoned its discrimination against family-controlled firms and softened its official policies toward religion. As a result, family owners are now allowed to seek higher social status, for example, through winning a seat in the PC or the CPPCC. Increasing numbers are also publicly declaring their religious affiliation. In this study, we have examined whether family owners’ social status and religious affiliation, both of which have been greatly shaped by the recent institutional developments, influence their corporate philanthropy. Our empirical finding that social status and religious affiliation interact with intrafamily succession intention to influence family-controlled firms’ charitable contributions greatly enhances our understanding of the evolving nature of family business in China. While a number of recent studies have examined the social status of family owners (Jia, Reference Jia2014), the role of religion in the growth of family-controlled firms and social development of China may also be a fruitful topic for future research. For example, studies could investigate how the religiosity of business owners and corporate executives might have an impact on corporate social entrepreneurship.
In addition, future research might benefit from investigating the effect of institutional changes on family-controlled firms’ charitable donations. For example, loosening controls on family-controlled firms would lead to more family business owners being willing to make charitable contributions. Thus, investigators could attempt to extend this research by examining how formal institutional changes influence the corporate philanthropy of family-controlled firms.
Furthermore, in addition to examining the effect of institutional differences across time, we would like to see more research on the influence of regional institutional differences. For example, Du, Jian, Du, Feng, and Zeng (Reference Du, Jian, Du, Feng and Zeng2014) noted great variations in local religiosity across provinces in China. Local religiosity refers to the intensity of religious beliefs within the local community. In regions where religion plays a prominent role, religious beliefs can become part of the social norm shaping individual behaviors. Hence, more work can be done to examine how local religiosity might interact with family owners’ goals and values in influencing corporate philanthropy.
Limitations
A few limitations of this study should be acknowledged. First, our sample consisted of privately owned firms in China, a major emerging economy. Thus, the nature of the sample may limit the generalizability of the findings to developed economies. China differs from developed economies substantially in terms of institutional arrangements, which might play a role in shaping corporate philanthropy. Hence, future research in this area would benefit from an investigation into whether intrafamily succession intention influences corporate philanthropy in other institutional settings.
Second, although we have argued that family owners with intrafamily succession intention engage in corporate philanthropy to accumulate or preserve their SEW, data limitations prevented us from directly measuring the dimensions of SEW that play a more critical role. Berrone et al. (Reference Berrone, Cruz and Gomez-Mejia2012) pointed out that SEW has multiple dimensions, each of which might play a different role in influencing family-controlled firms’ behavior. Researchers should investigate the underlying mechanisms and examine which SEW dimensions drive corporate philanthropy.
Third, although all religions encourage charity and caring for society, there are slight differences in group norms across religions. In our sample, among those who had a religious affiliation, 80.5% of the family owners were affiliated with Buddhism. The relatively small number of family owners affiliated with other religions prevented us from detecting significant effects because of low statistical power. In-depth interviews with family owners affiliated with different religions might be a more fruitful way to examine the role of religious norms in driving corporate philanthropy.
Finally, because our data were collected by a stratified sampling method, ideally, we would have made adjustments to estimation models by clustering standard errors by both regions and industry sectors. However, because the data on whether a firm was from a rural or urban area were unavailable, we were unable to cluster standard errors by region. Therefore, the empirical results need to be interpreted with caution.
Practical Implications
The results of our study also have a number of practical implications. Government agencies and nongovernmental organizations (NGOs) around the world have been actively promoting corporate philanthropy among different types of companies. However, they have focused mainly on providing economic incentives or generating institutional pressure. The findings of this study suggest that corporate philanthropy is greatly shaped by the goals, social standing, and personal values of family owners. Hence, government agencies and NGOs might consider complementing their current practices with a more nuanced approach that links corporate philanthropy with family owners’ SEW. For example, they might encourage the succession of family businesses to the next generation and seek ways to raise the social status of family owners.
CONCLUSION
This study uses expectancy theory and the SEW perspective to investigate the sources of variation in the behaviors of family-controlled firms. The results show that family owners’ intrafamily succession intention is the key factor that shapes family-controlled firms’ corporate philanthropy. Their social status and religiosity moderate that relationship. This study contributes to a better understanding of the differences among family-controlled firms and to the evolving nature of family businesses, particularly in China.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/mor.2015.10