Hierarchical Inconsistency Among Family-Member Top Leaders and Nonfamily Executives' Compensation Levels: Evidence from Chinese Family Firms

Abstract

In this study, we propose hierarchical inconsistency among family-member top leaders as a novel structural reason for nonfamily executives' high compensation in family firms. Hierarchical inconsistency among family-member top leaders is observed when the head in the formal business hierarchy is not ranked the highest among family-member executives and directors in the informal family hierarchy. We argue that this structure triggers contestations between the heads of the two hierarchies, adding complexity and challenges to nonfamily executives' jobs. Family firms with hierarchical inconsistency among family-member top leaders need to offer higher compensation to make up for the demanding features of these professional managers' work. The positive relationship between hierarchical inconsistency and nonfamily executives' compensation is weaker when the proportion of female family-member executives and directors is high and when the formal head's tenure is long. We use data from publicly listed family firms in China to test our arguments. Our study contributes to family business research by suggesting how the interplay between family relationships and formal organizational structures influences job features and compensation decisions.

摘要

本研究提出，家族企业中的一种结构化的因素——家族高层领导者之间的级序不一致——会导致非家族成员的高管获得更高薪酬。这种不一致主要表现在负责公司正式业务领导的层级，没有被排在负责非正式业务的其他家族成员高管之上时。作者认为，这种级序不一致的结构，会导致家族层级中最高领导者之间的相互竞争，因而增加非家族高管成员的工作复杂性和挑战性，相应的，需要提高他们的薪酬，与此同时，作者预测，当董事和高管成员中女性家族成员的比例较高时，或者负责正式业务的最高领导者任期较长时，级序不一致与非家族高管成员薪酬之间的正向关系会减弱。中国上市家族企业的数据检验并支持了上述假设。

Introduction

Nonfamily executives play crucial roles in the operation of family firms (Tabor, Chrisman, Madison, & Vardaman, 2018). They bring managerial expertise, new market opportunities, diverse knowledge, and legitimacy, helping overcome owning families' resource constraints (Guidice, Mero, & Greene, 2013; Sciascia & Mazzola, 2008; Vandekerkhof, Steijvers, Hendriks, & Voordeckers, 2015). Family firms must design compensation policies to attract and retain these professional talents (Chua, Chrisman, & Bergiel, 2009). Prior studies have suggested how the informal family relationships of the controlling family erode nonfamily executives' compensation (e.g., Chen, Chittoor, & Vissa, 2021). They argue that kinship ties induce altruism and affective partiality among family members, harming nonfamily executives' interests and negatively affecting their compensations (Chua et al., 2009; Karra, Tracey, & Phillips, 2006).

However, this argument does not consider the formal organizational structure these family members are embedded in. When the formal organizational structure is incongruent with the informal kinship structure, family members' cognitive, emotional, and behavioral dynamics might be altered (Guo, Luo, & Li, 2022), which might change the daily job demands of nonfamily executives who work with the family. Specifically, incongruence between structures might lead to conflicts rather than altruism among family members, especially those powerful ones at the top of the firm (Guo et al., 2022). As implied by prior research on matrix organizations, people who work in such competing structures confront great job complexity (e.g., Ford & Randolph, 1992). We hence wonder how nonfamily executives' job demands might increase with the mismatch between the formal and informal structures of the family firm and require high compensation to make up for the extra effort.

In particular, we explore how nonfamily executives' compensation levels are influenced by hierarchical inconsistency among family-member top leaders in family firms, a specific type of misalignment between formal organizational and informal family structures. Hierarchical inconsistency among family-member top leaders in the family firm is defined as the situation in which the formal business head and the informal family head are different family members in the family firm (Guo et al., 2022). In other words, a family member who ranks lower than at least one family-member director or executive in the family hierarchy is the head of the formal business hierarchy. In the Chinese business environment, the board chair has the ultimate authority to set firm strategies and possesses the highest formal power in the firm's decision-making (Chung & Luo, 2013; Krause, Li, Ma, & Bruton, 2019). Hence, hierarchical inconsistency occurs in a Chinese family firm when the family-member chair does not rank the highest in the family hierarchy among all family-member executives and directors engaged in the firm's operations. For instance, a chair and his elder brother are embedded in hierarchical inconsistency if the elder brother, who ranks higher in the family hierarchy, works as a top manager or a board member under the formal leadership of the chair.

We argue that family firms with hierarchical inconsistency among family-member top leaders offer higher compensation to nonfamily executives than those without hierarchical inconsistency. Under hierarchical inconsistency, the head in the formal business hierarchy and the head in the informal family hierarchy engage in competitive behaviors, resulting in demanding jobs that require higher compensation levels for nonfamily executives. The positive relationship between hierarchical inconsistency and nonfamily executives' compensation is weakened in two conditions. First, when other family members close to the two heads are motivated and able to mediate their conflicts, nonfamily executives' job demands will be reduced. Second, when the balance between the two heads in the contestation is broken, with one head gaining dominance over the other, nonfamily executives' jobs will be less demanding with a dominant authority. Consistent with these theoretical arguments, we predict that the positive relationship between hierarchical inconsistency and nonfamily executives' compensation level is weakened when the proportion of female family-member executives and directors is high and when the head in the formal business hierarchy has a long tenure. We use data from publicly listed family firms in China between 2006 and 2013 to test these arguments. We also interview two Chinese family firms under hierarchical inconsistency to understand the phenomenon.

This study first and foremost contributes to research on compensation decisions in family businesses. Compared with prior research on how family firms' executive compensation is affected by structural features, such as formal governance structures or informal family relationships, our study brings two new insights. First, we reconsider an assumption from prior research that family relationships give rise to emotional attachment and affective partiality among family members, which erode nonfamily executives' interests and negatively influence their compensation level (Chua et al., 2009). Instead, we argue that the interplay between family relationships and formal organizational structures can result in family members' conflicts and contestations, boosting nonfamily executives' pay. Second, prior research examines how structural elements shape power dynamics between family and nonfamily members. Nonfamily executives' relative power determines their salaries (Chen et al., 2021; Cheng, Lin, & Wei, 2015; Michiels, Voordeckers, Lybaert, & Steijvers, 2012). In contrast, we suggest that power dynamics among family members may alter nonfamily executives' job features. Higher job complexity stimulates nonfamily executives' demand for higher compensation.

Apart from the family business literature, this study also provides insights into executive compensation research in general. Existing research proposes internal and external strategic drivers of top managers' job complexity and compensation (Chen, 2015; Henderson & Fredrickson, 1996; Pathak, Hoskisson, & Johnson, 2014; Philippon & Reshef, 2012). Firms undertaking significant strategic initiatives or encountering severe turbulence in the external environment tend to compensate their executives with higher salaries for challenging jobs. In complement to these studies focusing on strategic forces, we suggest that internal structural forces such as hierarchical inconsistency may also cause executives' greater job demands and higher compensation. Hence, we propose a novel element that may impact top executives' job complexity and have an implication on their compensation levels.

Theoretical Background and Hypotheses Development

Nonfamily Executives' Compensation Levels in Family Firms

Family firms employ nonfamily executives to tackle the resource limitations of the owning families. The importance of these professional managers increases with the growth of family firms since the number of family members available for employment is limited (Block, 2011; Fang, Randolph, Memili, & Chrisman, 2016). Nonfamily executives bring key business competencies, such as strategic planning skills and financial expertise (Sciascia & Mazzola, 2008). They also expand family firms' social networks, introducing new market opportunities to the firms (Guidice et al., 2013). Scholars have found that nonfamily executives are especially valuable in family firms' pursuit of international and innovative strategies, as outside managerial talents can compensate for family members' lack of a diverse knowledge base (Vandekerkhof et al., 2015). Apart from skills, social capital, and knowledge, nonfamily managers also provide legitimacy to family businesses as professionalized firms, inducing trust from various stakeholders (Tabor et al., 2018).

Reaping the benefits generated by professional executives is not without cost. Nonfamily top managers often demand higher compensation than their family peers. Scholars have provided various reasons for their demand for higher pay. First, different from family members who have strong emotional commitments to the family firm and willingly promote its long-term prosperity, professional managers are more calculative in bargaining for their compensations and need more pecuniary benefits to join the family firm (Gomez-Mejia, Larraza-Kintana, & Makri, 2003; McConaughy, 2000; Michiels et al., 2012). Second, nonfamily executives often have limited promotion opportunities in family firms, so they would require higher salaries to stay (Combs, Penney, Crook, & Short, 2010; Yu, Stanley, Li, Eddleston, & Kellermanns, 2020). In contrast, family executives are rewarded with benefits such as job security due to their family connections and high status associated with family identity (Gomez-Mejia et al., 2003; McConaughy, 2000). Hence, family members are willing to accept lower compensation in exchange for these returns (Gomez-Mejia et al., 2003; McConaughy, 2000). Finally, compared with family executives, nonfamily managers are more likely to compete in the external labor market, have more employment choices, and thus can ask for higher pay (Gomez-Mejia et al., 2003; McConaughy, 2000).

However, nonfamily executives' demands for higher compensation may not be satisfied. Family owners' asymmetric altruism and affective partiality toward family managers may lead to the latter's overpayment (Chua et al., 2009). The controlling family may tunnel corporate resources for the private benefits of family executives by giving them exorbitant compensation, damaging their nonfamily colleagues' legitimate interests (Chen et al., 2021; Cheng et al., 2015; Michiels et al., 2012). Family members can dominate the compensation decision process on the board, leaving family executives' higher pay unchecked (Chen et al., 2021). The lower pay received by nonfamily executives may cause feelings of injustice and resentment, resulting in their demotivation and withdrawal from the firm (Chua et al., 2009). Research indicates that family firms may suffer from financial failure when they lack contributions from nonfamily executives (Chua et al., 2009; Fang et al., 2016; Sciascia & Mazzola, 2008).

Despite these insights, we reconsider an argument from prior studies on the determinants of nonfamily executives' compensation. Scholars suggest that family ties give rise to emotional attachment and affective favoritism among family members, which are detrimental to nonfamily executives' interests and negatively influence their compensation (Chen et al., 2021; Chua et al., 2009; Michiels et al., 2012). We argue that when family ties are intertwined with formal organizational structures, they may not result in family members helping each other but affective conflicts and competitive behaviors. Specifically, we propose that the inconsistent hierarchical orderings indicated by family relationships and business structures among family-member top leaders result in nonfamily executives' demanding jobs and high compensation.

Hierarchical Inconsistency Among Family-Member Top Leaders in Family Firms

Hierarchy indicates how individuals are ranked along a valued social dimension based on their power or status (Magee & Galinsky, 2008). Scholars have long acknowledged that individuals often simultaneously occupy ranks in multiple hierarchies, the saliency of which is determined by specific social conditions (Stryker & Macke, 1978). In family firms that combine family control and business operations (Gersick, Davis, Hampton, & Lansberg, 1997), two hierarchical cues are particularly relevant and comparably influential: the formal business hierarchy is deliberately designed by managers and legitimated by laws and regulations (Morand, 1995); the informal family hierarchy is implied by family roles bolstered by cultural norms conferring respect and power based on generation, birth order, and patriarchy (Ainsworth & Cox, 2003). The two hierarchies may confer misaligned ranks among family members in the family firm.

Specifically, among family-member top leaders, hierarchical inconsistency occurs when the formal business head of the firm is a family member whose rank is not the highest in the family hierarchy among all family directors and executives (Guo et al., 2022). The formal head's inferior rank in the family hierarchy and the existence of a different head from the family in the firm may come into being for various reasons. For example, the formal head might be a junior family member with outstanding competence compared with his/her senior relatives and might have either founded the firm or been selected by the family as the formal head. Alternatively, hierarchical inconsistency may reflect that the family firm is under leadership succession, in which old-generation family leaders pass the business on to a young-generation family leader but still work as board members or executives (Le Breton-Miller, Miller, & Lester, 2011).

Prior research demonstrates that, no matter how and why hierarchical inconsistency emerges, this structure alters family members' cognitive, emotional, and behavioral patterns (Guo et al., 2022). Interviews from prior research indicate that the business head confronted significant challenges from the family head, even when the former founded the firm and invited the latter to join the business or the latter chose the former as the successor (Guo et al., 2022). Quantitative results also show that people engaged in competitive behaviors when embedded in hierarchical inconsistency, after taking self-selection into account (Guo et al., 2022). Below, we will delineate how the two heads experience conflicts and compete with each other.

In hierarchical inconsistency, the two heads exert influence based on authorities conferred by different hierarchical orders. The formal business head relies on the formal chain of command to implement his/her decisions, while the informal family head mobilizes people and resources through entrenched family relationships (Guo et al., 2022). While rank orders in hierarchies shape people's expectations of how to interact with others (Stryker & Macke, 1978), misaligned hierarchies cause ambiguity on superiority, implying conflicting expectations (Jackson, 1962; Jensen & Wang, 2018). Incongruent social expectations make it difficult for the two heads to form an organized, stable sense of self (Goffman, 1957; Sessions, Nahrgang, Baer, & Welsh, 2022). Their social experiences are frustrating and uncertain, resulting in their psychological stress (Jackson, 1962; Stryker & Macke, 1978). To alleviate the stress, both heads strive to become the superior rather than the inferior, because an advantageous position means more deference and resource access (Greer, Caruso, & Jehn, 2011; Stryker & Macke, 1978). Hence, they engage in persistent contests to improve their own social standing and undermine that of the other (Guo et al., 2022). Because raising one's rank means lowering the other's superiority, the competition between the two heads is intense (Bendersky & Hays, 2012). They overtly challenge each other, find fault with each other's behaviors, and express objection to each other's decisions (Guo et al., 2022). Other members of the top management team or the board, such as nonfamily executives, are also part of this conflict, because the social conferral of superiority relies on these bystanders' perceptions and judgments (Bendersky & Hays, 2012).

Nonfamily Executives' Demanding Jobs Under Hierarchical Inconsistency

Nonfamily executives' jobs are complex and challenging. Not only do they provide business competence to the family firm but they also navigate the interpersonal dynamics of the owning family in their daily work (Blumentritt, Keyt, & Astrachan, 2007; Huybrechts, Voordeckers, & Lybaert, 2013). Family members often bring competing opinions to the business (Sciascia & Mazzola, 2008). They sometimes vary greatly in their feelings about how to run the business (Blumentritt et al., 2007). Nonfamily executives should know these incompatible expectations and appreciate these key stakeholders' strong feelings (Sciascia & Mazzola, 2008). Professional managers must invest time and energy to keep open and consistent communication with family members (Blumentritt et al., 2007). They must balance family members' varied opinions, helping them achieve consensus (Blumentritt et al., 2007; Tabor et al., 2018).

Hierarchical inconsistency among family members makes nonfamily executives' jobs even more difficult. As stated earlier, the formal business head and the informal family head can exert influence on nonfamily executives based on authorities conferred by different sources. While the formal head pushes forward agendas through formal reporting lines, the informal head employs entrenched family connections and norms to collect information and exercise control (Guo et al., 2022). In other words, nonfamily executives are subject to the dual influence of both heads. Our interviews show that nonfamily executives were indeed influenced by the informal family head as well as the formal boss in their daily work (see Appendix I for the basic information of our two interviews). We interviewed Firm A's nonfamily CEO (E1). The chair of Firm A founded the firm jointly with his elder brother, who served as vice president. Firm A became hierarchically inconsistent because the chair, though younger than the vice president, had an educational background aligned with the firm's core business. E1 told us, ‘Although I formally report to the chair, the elder brother has strong connections with many people in the firm and can learn all the operational details from old employees. The elder brother can also substantially control me and other professional managers’. As indicated in the interview, nonfamily executives such as E1 encounter two bosses in their work, formal and informal.

Prior studies on matrix structures describe employees' demanding work under two bosses (Knight, 1976). When two leaders compete to control the same resources, subordinates are susceptible to work overload and incongruent orders (Barker, Tjosvold, & Andrews, 1988; Joyce, 1986). Employees struggle to meet the demands of two bosses, spending extra energy coordinating excessive needs, which are often inherently incompatible (Ford & Randolph, 1992). To manage the relationships with their two bosses, subordinates must regulate emotions, suppress anger that damages cooperation, and display positive emotions to foster progress (Sy & Côté, 2004). In such work environments, people report frustration, anxiety, and low job satisfaction (Ford & Randolph, 1992; Sy & Côté, 2004).

Similarly, nonfamily executives are continually faced with challenging encounters in daily interactions with the formal business head and the informal family head. Their jobs are even more demanding, as the two heads are strongly motivated to compete for dominance (Bendersky & Hays, 2012). The two leaders overtly attack each other's decisions (Guo et al., 2022), making reconciling their expectations difficult, if not impossible. As the third parties in this competition for superiority, nonfamily executives should be cautious not to irritate either head, as the two heads are both sensitive to these bystanders' attitudes toward them (Bendersky & Hays, 2012). These professional managers must invest tremendous cognitive and emotional energy to meet the competing demands of both heads. As E1 shared,

My work is difficult, sometimes a real headache. The decision-making cost is high. I often don't know whether to follow the chair or his elder brother. For example, the chair once wanted to buy land and build a new plant. I had gone halfway through this project, and suddenly his elder brother said we shouldn't do this. And I had to stop. I often have to oscillate between their different opinions. This brings us professional managers big problems. We should be very careful to make them both satisfied.

Hierarchical inconsistency emerged in Firm B because its founder passed on the business to his son. The son became the firm's chair, while the father still served as a director on the board. The firm's vice president (E2), a nonfamily member, shared his observations:

After stepping down from the chair's position, the father actually had no specific work, but he was involved in recruiting an R&D executive in the hope of helping the company and the son find the right person for the future. However, the father and the son had very different opinions on the recruitment criteria … It could be really embarrassing witnessing the father and the son quarreling with each other. The father thought that he had lost face in front of us … I wanted to be the peacemaker, but it was difficult … The recruited R&D executive, though meeting the expectations of the father, failed to work smoothly with the son in daily operations. This manager had lots of complaints and quit after one to two months.

The labor economics perspective suggests that job candidates are willing to accept aversive jobs requiring additional effort only when a pay differential exists to compensate for their job pain (Campion & Berger, 1990; Glomb, Kammeyer-Mueller, & Rotundo, 2004). Due to the demanding feature of nonfamily executives' jobs under hierarchical inconsistency, those who work for a hierarchically incongruent family firm need higher pay than their peers working in a family firm with a single authority. Otherwise, people would rather choose to serve in a hierarchically consistent firm. Accordingly, we hypothesize the following:

Hypothesis 1 (H1): Compared with family firms with hierarchical consistency among family-member top leaders, those with hierarchical inconsistency will offer higher compensation to nonfamily executives.

We argue that the positive relationship between hierarchical inconsistency and nonfamily executives' compensation attenuates when the proportion of female family executives and directors is high, as they are motivated and able to mediate the conflicts between the two heads. Female family members have the motivation to enhance peace and cohesion of their families. Research in evolutionary psychology shows that women feel more subjectively close to their family members and value kinship connections more than men (Neyer & Lang, 2003). They prioritize family harmony and cooperation over power and achievement (Chen, Fang, MacKenzie, Carter, Chen & Wu, 2018).

Apart from their motivation, female family members are more capable of reconciling conflicts and buffering nonfamily executives from painful situations. The gender role theory argues that the distribution of men and women into different occupational and family roles gives rise to different normative expectations of their behaviors (Eagly & Karau, 1991). By conforming to these norms, men and women would acquire distinct qualities (Grossman & Wood, 1993). Because of their roles as primary family care givers and nurturers, women are more sensitive to others' emotional needs and better at imagining how others feel than men (Hoffman, 1977). Women listen more patiently to others and know better when to provide help (Neff & Karney, 2005). Female leaders demonstrate stronger skills in releasing tension, promoting interpersonal harmony, and maintaining group morale in complex social interactions (Eagly & Karau, 1991). Combined with their tacit understanding of the family's idiosyncratic habits, rituals, and communication styles (Block, 2011; Ensley & Pearson, 2005; Jimenez, 2009), female family executives and directors have unique advantages to being arbitrators under hierarchical inconsistency. They would help interpret the behaviors of one head to the other, take care of both heads' feelings, and mediate their conflicts (Jimenez, 2009). Due to female family members' efforts, disputes between the two heads under hierarchical inconsistency would more likely be contained within the family, reducing the job complexity of nonfamily executives. Nonfamily executives are more likely able to focus on other essential responsibilities, saving their time and efforts from reconciling the incompatible opinions of the two heads.

In sum, the existence of female family executives and directors would lessen the job demands of nonfamily executives under hierarchical inconsistency, attenuating the necessity for offering them higher compensation. Therefore, we propose the following hypothesis:

Hypothesis 2 (H2): The higher the proportion of female family-member executives and directors, the weaker the positive relationship between hierarchical inconsistency and the nonfamily executive's compensation level.

We also propose that the long tenure of the formal business head will weaken the positive relationship between hierarchical inconsistency and nonfamily executives' compensation level. Specifically, we suggest that the administrative devices, personal connections, and legitimacy the business head builds over time would help reduce the job complexity faced by nonfamily executives in hierarchical inconsistency, which in turn mitigates the latter's need for higher compensation. The baseline hypothesis suggests that the business head engages in fierce competitive behaviors with the family head under hierarchical inconsistency. While the former relies on the power conferred by the formal title, the latter possesses informal authority based on habitual interaction patterns rooted in family dynamics. It would be difficult for nonfamily executives to decide whether to follow the business head or the family senior. These professional managers must spend much time communicating with the two heads and struggling to work out solutions to meet both their demands.

However, the balance in the contestation between the two heads will be broken when the business head holds his/her formal position for a longer time and gains dominance over the family head (Miller, 1991). With formal authority, the business head can launch formal administrative systems to gradually accumulate his/her control over the firm (Miller, 1991). Formal conflict mediation mechanisms to reconcile different opinions between the business head and the family senior can be built over the years to favor the business head. In addition, step by step, the business head can populate the board and executive teams with people loyal to him/her (Hill & Phan, 1991). With the help of these personal supporters, the business head has more chances to win absolute dominance over the family head. Finally, increased tenure will promote the legitimacy of the business head's formal authority, making stakeholders take for granted the business head's leadership over his/her family senior (Ocasio, 1994; Shen, 2003). As the formal head's power advantage becomes clearer, nonfamily executives will have less confusion about whose order to follow. After all, most of them have been professionalized to fulfill business responsibilities under the supervision of the formal business head (Meyer & Rowan, 1977; Ocasio, 1994). Rather than being overwhelmed by excessive, competing demands from the two heads (Knight, 1976), nonfamily executives might develop strategies to substantially carry out the business head's orders while addressing the family head's needs superficially.

Overall, nonfamily executives' jobs will become less demanding under hierarchical inconsistency with a longer-tenured formal head, and the necessity to offer them higher compensation will be weakened accordingly. Therefore, we propose the following:

Hypothesis 3 (H3): The longer the tenure of the head in the formal business hierarchy, the weaker the positive relationship between hierarchical inconsistency and the nonfamily executive's compensation level.

We also expect that hierarchical inconsistency of family members on the board and the executive team is unlikely to influence family executives' compensation levels. Prior literature suggests that family executives have strong emotional commitments to the family firm (Gomez-Mejia et al., 2003; McConaughy, 2000; Michiels et al., 2012). By working for the family firm, they harvest nonpecuniary benefits such as employment security and high family-related status and are unlikely to compete in the external labor market (Gomez-Mejia et al., 2003; McConaughy, 2000; Yu et al., 2020). With the exclusive welfare they can get from the family firm and constrained marketability outside the family firm, family executives are unlikely to ask for higher compensation even if they have to deal with the complex dynamics of hierarchical inconsistency (Brookman & Thistle, 2013; Fulmer, 2009).

Methods

Sample

We use publicly listed family firms in China to test our hypotheses. As suggested earlier, hierarchical inconsistency would be influential only when the two misaligned hierarchies are salient with comparable strengths (Stryker & Macke, 1978). Because hierarchies are rank orderings fostered by shared norms and beliefs, their strengths are determined by the institutional environment in which they are embedded (Magee & Galinsky, 2008). The formal business hierarchy and informal family hierarchy of Chinese listed family firms are bolstered by competing institutions that are comparably influential in Chinese society (Guo et al., 2022). Laws and regulations, which have been especially emphasized since the country's market-based institutional transition, protect the formal business hierarchy designed by business leaders. In China, the Company Law prescribes the chairperson as the legal representative and the formal top decision-maker of a company (Chung & Luo, 2013). The chairs in Chinese firms usually have the highest authority to set up company agendas rather than only undertaking monitoring activities similar to those in the US context (Krause et al., 2019).

The informal family hierarchy, deeply entrenched in the family's close interactions and history, is endorsed by Confucianism in China. This old yet vital culture promotes a patrilineal system highlighting younger family members' respect for their elders and a wife's subservience to her husband (Podmore & Chaney, 1974). For instance, traditional Chinese families strictly conform to the norm of filial piety, the unquestioning obedience of children to their fathers (Morris, Podolny, & Sullivan, 2008). Women are subordinate to men in the rigid family hierarchy dominated by older males (Bauer, Wang, Riley, & Zhao, 1992).

We followed prior research and used two criteria to identify family firms from all publicly listed firms in China (Guo et al., 2022). First, the chair's family owned the largest proportion of firm shares. Second, the chair and at least one executive or director in the firm came from the same family. We used these two criteria to ensure that hierarchical (in)consistency among family-member top leaders could be defined. If the chair was a nonfamily member, the formal head did not have a rank in the family hierarchy, and our definition of hierarchical inconsistency was irrelevant. Alternatively, if there was only one family member among the top leaders in the executive team or on the board, he/she could not be embedded in hierarchical (in)consistency, as he/she was the only person ranked in both business and family hierarchies. Hence, these two criteria led us to a relatively narrow range of family firms to fulfill our specific theoretical purpose (Miller, Le Breton-Miller, Lester, & Cannella, 2007).

Our sample consists of publicly listed family firms meeting the two criteria in China between 2006 and 2013. Our data cover only this period, partially because our independent variable was manually coded based on a highly time-consuming process to identify specific kinship ties involved in each family firm (see Appendix II for the description of the coding process). The manually coded data after 2013 are not available. Nevertheless, we believe this specific time frame is a representative period to test our hypotheses. On the one hand, this period witnessed the ongoing progress of marketization in the Chinese economy, which continuously reinforced the influence of laws and regulations on the business activities of listed firms (Guthrie, 2012). On the other hand, the family culture based on Confucianism remained crucial in family dynamics during these years (Ebrey, 2014; Murphy, Tao, & Lu, 2011). Hence, the formal business and informal family hierarchies were comparably strong, and the inconsistency between the two was likely to have pivotal effects on decision-making in the firm.

Nonfamily executives were identified as top managers without kinship ties with the chair's family. The two criteria and the abovementioned time frame led us to 3,185 nonfamily executives in 475 family firms. In total, our sample had 5,881 executive-year observations.

Dependent Variable

Nonfamily executive's compensation level

We used the nonfamily executive's annual cash salary and bonus to calculate the compensation level. Specifically, we measured a nonfamily executive's compensation level with his/her cash salary and bonus in a year divided by the average cash salary and bonus received by family executives in the same firm that year. Because we expected that hierarchical inconsistency only increased nonfamily executives' compensation and did not influence family executives' compensation, we used this ratio variable to reflect the relative increase in the pay of nonfamily executives, taking into account the variation in family executives' pay across firms. Both family and nonfamily executives' compensation levels were collected from the China Stock Market and Accounting Research (CSMAR) database (Shenzhen CSMAR Data Technology Co., Ltd., 2013).

Independent Variable and Moderators

Hierarchical inconsistency among family-member top leaders

We measured this concept with a dichotomous variable. It was coded one if the nonfamily executive worked in a firm where the formal business head (i.e., the board chair) and the informal family head are different family members and zero otherwise. In other words, this variable was coded one if the nonfamily executive worked in a firm where the chair had at least one family executive or director ranked higher than him/her in the family hierarchy. We followed prior research and considered a family executive or director as ranked higher than the chair in the family if this executive or director was the chair's ‘grandparent (-in-law); parent (-in-law); sibling (-in-law) or cousin (-in-law) of the grandparents (-in-laws) or parents (-in-laws); elder sibling or cousin, or their spouse; spouse's elder sibling or cousin; husband’ (Guo et al., 2022: 2195). We manually collected information on kinship ties among executives and directors from firms' prospectuses and annual reports, triangulating them with media articles (see Appendix II for more details).

Proportion of female family-member top leaders

We measured this variable with the number of female family executives and directors divided by the total number of family executives and directors in the firm. We calculated this variable with our manually collected data.

Tenure of the head in the formal business hierarchy

This variable was measured by the number of years the chair served in this position. We collected this information from the CSMAR database.

Control Variables

We controlled the nonfamily executive's characteristics that may influence his/her compensation level (Ren & Zhu, 2010; Wasserman, 2006). Specifically, we controlled for the nonfamily executive's age and gender (male = 1; female = 0). Moreover, we controlled the number of years the nonfamily executive served as a top manager throughout his/her career history. We also included the nonfamily executive's educational level. We used a categorical variable to measure his/her highest education as technical secondary school and below, junior college, bachelor, master, or PhD. To control for their stock incentives, we included a dichotomous variable to indicate whether the nonfamily executive held the firm's shares. We used a dummy variable to indicate whether he/she was a CEO to control for the formal power held by the nonfamily executive. A nonfamily CEO was more powerful than nonfamily executives with other titles.

Some variables reflect the family firm's governance. First, we considered family influence and collected data on the percentage of shares owned by the family (Gomez-Mejia et al., 2003). Second, the chair's formal power might influence compensation decision-making (Chen et al., 2021). Hence, we included a dummy variable to indicate whether the chair was the CEO. Third, we controlled the proportion of shares held by institutional investors (Gomez-Mejia et al., 2003). Fourth, the number of independent directors on the board was included to control for board independence. Lastly, we included the number of board meetings held annually.

We also considered other firm characteristics. First, an alternative explanation for our argument is that hierarchical inconsistency might imply a transition period during which a senior family member prepares to pass on the business to a junior one. A typical example is when the chair works with his/her parents who might retire soon. Family firms might require capable professional talents to help them through transition periods, and these nonfamily executives might negotiate higher compensation for the higher risks in managing firms in transition (Block, 2011; Yu et al., 2020). Hence, we controlled for the years when a firm experienced a leadership transition to address this alternative explanation. Specifically, we coded the chair succession year and the years before and after this year as one, and zero otherwise. Apart from controlling the transition period, we also included firm performance, measured by return on assets (ROA), as prior research found that financial performance would influence executives' compensation (Michiels et al., 2012). In addition, we controlled for firm size, measured by assets (logged), because scholars suggested that top managers' compensation level increased with the size of the firms they managed (Wright, Kroll, & Elenkov, 2002). We also included firm age as a control variable, coded as the number of years since its founding. Finally, industry dummies and year dummies were added to the data analysis. All the information on the control variables was collected and derived from the CSMAR database. All independent, moderating, and control variables were measured at year t, while the dependent variable was measured at year t + 1.

Analysis

We addressed the potential endogeneity of the independent variable by employing the control function approach. This two-stage method is especially adept at dealing with the endogeneity of a dichotomous independent variable (Wooldridge, 2015). In the first stage, we predicted the likelihood of hierarchical inconsistency by estimating a probit model. We added an instrumental variable apart from putting all second-stage control variables into this model.

The instrument selected was the average age of other firms' chairs in the same industry. Using industry averages as instrumental variables is common in economics and management research (Sutton, Devine, Lamont, & Holmes, 2021). The instrument should satisfy the relevance condition (Wooldridge, 2020). We expected that family firms were more likely to have young chairs in industries where young leaders were prevailing. Hierarchical inconsistency was more likely to occur when a young family member became the formal business head. Apart from relevance, we also evaluated the instrument's exclusion restriction (Miguel, Satyanath, & Sergenti, 2004; Wang, Pahnke, & McDonald, 2022). This condition required that the average age of other firms' chairs in the same industry should affect nonfamily executives' compensation only through hierarchical inconsistency. In our research, exclusion restriction could not be explicitly tested because we only identified one instrumental variable and hence could not use the test of overidentifying restrictions to check exogeneity (Stock & Watson, 2019; Wang et al., 2022; Wooldridge, 2020).

Nevertheless, following prior practices (Miguel et al., 2004), we examined two potential violations of the exclusion restriction. First, it is possible that industries with young chairs attract nonfamily executives of similar ages, and nonfamily executives' ages might influence their compensation levels. To explore this possibility, we estimated the effect of our instrumental variable on nonfamily executives' ages. This relationship was found statistically insignificant (β = 0.01, p = 0.750). Therefore, the instrumental variable was unlikely to impact nonfamily executives' compensation levels through their ages. Second, we considered the possibility that the instrumental variable affected our dependent variable through firm age. Industries in which old leaders are prevailing might be mature industries with old firms. Old firms might have rigid compensation policies for nonfamily executives. We found that our instrumental variable was not significantly associated with firm age, hence ruling out the possibility that the instrument impacted nonfamily executives' compensation levels through firm age (β = 0.01, p = 0.630).

Apart from the investigations on potential violations, we followed prior studies to get an intuitive understanding of the exogeneity of our instrument by examining whether it could be a significant predictor in the second-stage model (Ertug, Kotha, & Hedström, 2020) and whether it was correlated with the residual of the second-stage model (Bentley & Kehoe, 2020). We found that the coefficient of the instrumental variable was insignificant when added to the second-stage model (β = −0.04, p = 0.170), suggesting that the instrument was unlikely to impact the dependent variable directly (Ertug et al., 2020). We also calculated the correlation between the residual of the second-stage model and the instrument. The correlation was 0.00, strengthening our confidence that the instrument was valid (Bentley & Kehoe, 2020).

From the first-stage estimation, we computed a generalized residual. In the second-stage models, we conducted random-effects regressions as some variables did not vary much over time (Marquis & Lee, 2013). We added the generalized residual to regressions in the second stage.

Results

Table 1 presents the descriptive statistics and correlation coefficients for the variables. Approximately 26% of nonfamily executives served in family firms with hierarchical inconsistency. The mean of the dependent variable is 0.87, indicating that, on average, nonfamily executives received lower pay than their family peers. The absolute values of the correlation coefficients between hierarchical inconsistency and the two moderating variables are below 0.15, a relatively low level compared with recently published research and indicating a low risk of multicollinearity (Dutt & Joseph, 2019; Feldberg, 2022; Garrido, Giachetti, & Maicas, 2023; Han & Pollock, 2021; Mawdsley, Paolella, & Durand, 2023; Park & Rogan, 2019).

Table 1. Descriptive statistics and correlations (N = 5,881 observations)

Note: All absolute values larger than 0.05 are significant at p < 0.05.

Table 2 shows the first-stage probit model and the second-stage random-effects regressions. Model 1 is the first-stage model. Consistent with our prediction, the coefficient of the instrumental variable is negative and significant (β = −0.18; p = 0.020). Model 2 is the baseline model only with control variables. Model 3 adds the independent variable and the generalized residual calculated from the first-stage estimation. In Models 4 and 5, two moderators are added, respectively. Model 6 is the full model and includes both sets of interactions. The average variance inflation factor across all models is 2.81, below the threshold of 10 (Ryan, 2008), indicating little multicollinearity. The generalized residual is significant in Models 3–6 (β = −0.80, p = 0.000 in Model 3; β = −0.79, p = 0.000 in Model 4; β = −0.69, p = 0.000 in Model 5; β = −0.67, p = 0.001 in Model 6), suggesting the endogenous feature of hierarchical inconsistency.

Table 2. Control function approach and random-effects regressions predicting nonfamily executives' compensation level (ratio variable), 2006–2013

Notes: Robust standard errors in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

H1 predicts the positive relationship between hierarchical inconsistency and nonfamily executives' compensation. As expected, the coefficient of hierarchical inconsistency is significantly positive in Models 3–6 (β = 1.36, p = 0.000 in Model 3; β = 1.37, p = 0.000 in Model 4; β = 1.30, p = 0.000 in Model 5; β = 1.31, p = 0.000 in Model 6). H1 is supported.

H2 posits that the positive effect of hierarchical inconsistency is weakened when the proportion of female family executives and directors is high. Models 4 and 6 show that the coefficient of the interaction between hierarchical inconsistency and the proportion of female family top leaders is negative and significant (β = −0.18, p = 0.045 in Model 4; β = −0.20, p = 0.033 in Model 6). H2 is supported.

Following recent studies, we conducted a marginal effect analysis to demonstrate the moderating effect (Bonini, Deng, Ferrari, John, & Ross, 2022; Busenbark, Graffin, Campbell, & Lee, 2022). As shown in Figure 1A, the marginal effect of hierarchical inconsistency on nonfamily executives' compensation levels decreases with the increase in the proportion of female family executives and directors. The dotted lines indicate 95% confidence intervals around this effect. The range between the two dotted lines does not cross zero, suggesting that the effect is significant.

Figure 1. Marginal effects of hierarchical inconsistency on nonfamily executives' compensation level with 95% confidence intervals

To further check the significance of the interaction effect, we calculated the estimated difference in nonfamily executives' compensation increase under hierarchical inconsistency between the two scenarios of low and high proportions of female family top leaders (the so-called double difference; Zelner, 2009). Specifically, when the proportion of female family top leaders is low (mean − 1 s.d.), the estimated difference in nonfamily executives' compensation between hierarchical inconsistency = 1 and hierarchical inconsistency = 0 is 1.325 (=1.819 − 0.494). When the proportion of female family top leaders is high (mean + 1 s.d.), the estimated difference in nonfamily executives' compensation between hierarchical inconsistency = 1 and hierarchical inconsistency = 0 is 1.223 (=1.776 − 0.553). The double difference is 0.102 (=(1.819 − 0.494) − (1.776 − 0.553)) with a 95% confidence interval of [0.008, 0.197]. The confidence interval does not cross zero, suggesting a significant moderating effect of the proportion of female family executives and directors.

Moreover, to better interpret the meaning of this moderating effect, we set family executives' compensation level and all the control variables at mean value and calculated the monetary value of change in compensation in Chinese Yuan. With a low proportion of female family top leaders, hierarchical inconsistency increases nonfamily executives' compensation at 559,695.92 (=765,899.68 − 206,203.76) Chinese Yuan. With a high proportion of female family top leaders, hierarchical inconsistency increases nonfamily executives' compensation at 517,613.52 (=749,066.72 − 231,453.20) Chinese Yuan. The difference between the two is 42,082.4 Chinese Yuan. In other words, when working with a large proportion of female family members, nonfamily executives under hierarchical inconsistency will receive a payment increase 42,082.4 Chinese Yuan fewer than those who work with a small proportion of female family members under hierarchical inconsistency.

H3 suggests that the positive effect of hierarchical inconsistency is weakened when the chair's tenure is long. In Models 5 and 6, the coefficient of the interaction term is negative and significant (β = −0.04, p = 0.001 in Model 5; β = −0.04, p = 0.001 in Model 6), supporting H3. Figure 1B shows that the marginal effect of hierarchical inconsistency decreases with the increase in formal head's tenure, with the 95% confidence intervals all above zero. We also calculated the double difference – the estimated difference in nonfamily executives' compensation increase under hierarchical inconsistency between the two scenarios of short (mean − 1 s.d.) and long (mean + 1 s.d.) tenures of the chair. The double difference is 0.154 (=(1.661 − 0.430) − (1.750 − 0.673)), with a confidence interval not crossing zero [0.062, 0.246], indicating the significance of the moderating effect.

We interpreted this result with nonfamily executives' absolute compensation levels. By setting family executives' compensation and all the control variables at mean value, we found that hierarchical inconsistency increases nonfamily executives' compensation at 517,613.52 (=698,567.84 − 180,954.32) Chinese Yuan with a short-tenured chair and 454,489.92 (=736,442 − 281,952.08) with a long-tenured chair. The difference is 63,123.6 Chinese Yuan, enhancing our confidence that the formal business head's years in position attenuates the need to provide higher compensation to nonfamily executives under hierarchical inconsistency.

Although the regressions show that hierarchical inconsistency is an important factor in determining nonfamily executives' compensation, the R ² is relatively low, indicating limited variance explained by the regressions. It is not uncommon to observe low R ² in management studies (e.g., Figueiredo & Rawley, 2011; Washburn & Bromiley, 2014; Wombacher & Felfe, 2017). The limited explanatory power of the regressions partially reflects the variety of factors influencing nonfamily executives' compensations, such as firm policies, market conditions, and individual negotiation skills (Bidwell & Briscoe, 2010; Miller, 1995). Though the effect size of hierarchical inconsistency is generally large, it is far from the only important factor influencing nonfamily executives' compensation.

Alternative Explanations

There are at least two alternative explanations for our theoretical arguments. First, it could be argued that hierarchical inconsistency might imply the meritocracy culture of a family firm, as the firm might be led by a junior but capable family member rather than the family senior. A family firm with a meritocracy culture would be more likely to offer generous compensation to capable nonfamily executives rather than designing its compensation policy based on nepotism. First, to take the relationship between capabilities and compensation levels into account, we controlled nonfamily executives' capabilities in the regressions. We used executives' years of working experience as top managers throughout their career paths to measure their capabilities. Models 3–6 in Table 2 show that nonfamily executives' working experience is not significantly associated with their compensation.

Moreover, if this alternative argument were true, we would expect a positive moderating effect of the formal business head's tenure. The reasoning is as follows. A capable junior chair would advocate a culture of meritocracy. The longer this chair holds the formal top position, the more ingrained the meritocracy culture, which leads to even higher compensation for nonfamily executives. However, our empirical result testing H3 demonstrates the opposite, suggesting that this alternative explanation might be less likely.

The second alternative explanation is that hierarchical inconsistency, as a temporal arrangement during the succession period, might not lead to conflict. The family senior is motivated to offer high compensation to nonfamily executives, hoping these professional managers can continue supporting the new leader. Therefore, hierarchical inconsistency does not indicate conflicts but, on the contrary, implies support in the succession period. We argue that the family senior's motivation to support the young chair can be expected not only under hierarchical inconsistency but also in hierarchically consistent firms, where the transition period starts, and the young successor is still in a subordinate position as a mentee of the old chair. In other words, this alternative argument can be applied to any family firm in leadership transition, whether it is hierarchically consistent or not. To take this mechanism into account, we incorporated the transition period as a control variable in the regressions. (See the ‘Control Variables’ subsection for information on measuring transition periods.) As indicated in Table 2, the transition period has a marginally significant positive relationship with nonfamily executives' compensation in general, suggesting the importance of controlling this transition effect.

Moreover, if this alternative explanation were true, we would expect a positive moderating effect of the proportion of female family top leaders. As suggested by evolutionary psychology, female family members, such as mothers, are more likely to take excessive care of the juniors and are more motivated to help them than male family members (Lubatkin, Schulze, Ling, & Dino, 2005; Neyer & Lang, 2003). Hence, nonfamily executives are more likely to receive pay with an even higher premium under hierarchical inconsistency with the existence of many female family top leaders. However, our empirical result of H2 suggests the opposite, making this alternative argument less likely.

Further Analyses

We conducted two supplementary analyses to understand the effect of hierarchical inconsistency further. First, as suggested in our theory development, we did not expect family executives to be compensated for the work conditions under hierarchical inconsistency. Therefore, instead of using a ratio variable, we examined how hierarchical inconsistency influenced the absolute compensation levels of nonfamily executives and family executives, respectively. We collected 2,059 family executive-year observations between 2006 and 2013. Table 3 shows that, consistent with our main analysis, the effect of hierarchical inconsistency on the nonfamily executive's absolute compensation level is significant and positive (β = 29.10; p = 0.012) in Model 7. In contrast, Model 8 demonstrates that the effect of hierarchical inconsistency on the family executive's absolute compensation level is only marginally significant (β = 27.84; p = 0.094).

Table 3. The effect of hierarchical inconsistency on executives' absolute compensation level: nonfamily executives vs. family executives

Notes: Robust standard errors in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

Second, while using a dichotomous variable to measure hierarchical inconsistency helps us focus exclusively on the parsimonious structural misalignment between business and family hierarchies, we ignore the various types of kinship ties between the chair and those who rank higher than him/her in the family hierarchy. Hence, we created a categorical variable that distinguished hierarchical inconsistency into five types. Type 1 was coded one when the chair led his/her parents in the executive team or on the board, and zero otherwise. Type 2 was assigned a value of one when the chair led her husband, and zero otherwise. Type 3 was coded one when the chair led his/her older-generation relatives other than parents, and zero otherwise. Type 4 was coded one when the chair led his/her elder siblings, and zero otherwise. Type 5 was assigned a value of one when the chair led all other types of superior relatives in the same generation not included in Types 2 and 4, and zero otherwise.

Table 4 presents the empirical findings on how these five types of hierarchical inconsistency influence nonfamily executives' compensation (ratio variable). We employed a control function approach with an ordered probit model at the first stage (Model 9). Model 10 shows the second-stage full model and further indicates that all five types of hierarchical inconsistency have significantly positive relationships with nonfamily executives' compensation (β = 1.33, p = 0.001 for Type 1; β = 0.97, p = 0.002 for Type 2; β = 0.83, p = 0.005 for Type 3; β = 0.76, p = 0.001 for Type 4; and β = 0.51, p = 0.003 for Type 5). We conducted a series of Wald tests to compare the coefficients of these five types of hierarchical inconsistency. The Wald tests indicate that the coefficient of Type 1 is significantly larger than those of the other four types. The coefficient of Type 2 is significantly larger than those of Types 3, 4, and 5. Moreover, the coefficients of Types 3 and 4 have no significant differences, but both of them are significantly larger than Type 5. These results support the idea that hierarchical inconsistency as a featured structure influences nonfamily executives' compensation levels, though the strength of the influence varies with the types of kinship ties involved in the structure.

Table 4. The effect of hierarchical inconsistency on nonfamily executives' compensation level (ratio variable): distinguishing different types of hierarchical inconsistency

Notes: Robust standard errors in parentheses. *p < 0.1, **p < 0.05, ***p < 0.01.

Discussion

This study suggests that hierarchical inconsistency among family-member top leaders in family firms leads to higher compensation levels for nonfamily executives. The contestations between the formal business head and the informal family head embedded in hierarchical inconsistency make nonfamily executives' work demanding, resulting in higher pay to compensate for their extra efforts. This positive relationship between hierarchical inconsistency and nonfamily executives' compensation is weaker when the proportion of female family top leaders is higher and when the tenure of the head in the formal hierarchy is longer.

We are able to test these theoretical arguments, thanks to the salient context provided by family firms in China. As a transitional economy, China is a typical context of institutional complexity (Luo, Wang, & Zhang, 2017). Different institutional norms bolster competing hierarchical orders (Magee & Galinsky, 2008), making the country a rich field to witness misaligned hierarchies. Specifically, Confucianism in China explicitly indicates family hierarchical orders (Podmore & Chaney, 1974). Espoused by traditional values, family hierarchy is as influential as business hierarchy in Chinese family firms. Hence, this context allows us to observe salient contexts among people under hierarchical inconsistency. As stated below, to the extent that family hierarchy exists in other cultures and multiple hierarchies prevail in other social systems, the theoretical insights developed from this context can be transferred to family business research in general and inspire the broader literature on executive compensation.

Theoretical Contributions and Practical Implications

Our study contributes to family business research by enhancing our understanding of the determinants of nonfamily executives' compensation levels. We propose a novel structural reason that shapes nonfamily executives' job features and determines their compensation levels. Prior studies have identified structural factors influencing family firms' compensation policies. For instance, formal governance structures such as a board with a low level of independence would fail to monitor the powerful group of family members, leading these people to misappropriate nonfamily executives' interests by giving themselves disproportionally high compensation (Chen et al., 2021; Cheng et al., 2015; Michiels et al., 2012). However, this argument is built upon the assumption that family relationships inevitably give rise to emotional attachment and affective partiality among family members (Chua et al., 2009; Jaskiewicz, Uhlenbruck, Balkin, & Reay, 2013; Karra et al., 2006; Neckebrouck, Schulze, & Zellweger, 2018). Our study reconsiders this assumption and suggests that family relationships as a major type of informal structure in a family firm may not be congruent with the firm's formal structure, which results in family members' conflict and competition rather than the cohesion and favoritism suggested by prior literature.

Moreover, while existing studies on nonfamily executives' compensation have focused on the power dynamics between family and nonfamily members shaped by structural factors that impact professional managers' compensation levels (e.g., Chen et al., 2021), our research highlights that power dynamics among family members may cause heterogeneity of nonfamily managers' job features. The interplay of the two inconsistent hierarchical structures stimulates the heads of the two hierarchies to attack each other and contest superiority. Like matrix organizations' two-boss structure (Ford & Randolph, 1992; Knight, 1976), this structural feature adds extra complexity to nonfamily executives' jobs. Hence, our research advances the knowledge of nonfamily executives' compensation by suggesting a new determinant of their job heterogeneity.

Our study also offers fresh insights into executive compensation research in general. Prior literature has documented the positive relationship between the level of executives' job demands and their compensation. Specifically, executives are found to be compensated for challenging strategic initiatives their firms undertake, such as innovation projects, refocusing programs, turnaround activities, and multi-business operations (Chen, 2015; Henderson & Fredrickson, 1996; Pathak et al., 2014). In addition, changes in the external environment, such as deregulation, may also increase the job demands of executives and hence increase their compensations (Philippon & Reshef, 2012). Unlike existing studies, our research uncovers an internal structural reason for the increase in executives' job complexity. Specifically, we show that the existence of two top authorities in a firm complicates the executives' work dynamics, resulting in their demand for higher pay. Complementary with extant research focusing on the strategic drivers of executive job demands, we propose structural dynamics as an underexplored cause of executives' job complexity and increase in pay.

Finally, while prior research has shown hierarchical inconsistency's instrumental impact on corporate governance (Guo et al., 2022), our study seems to imply a negative effect of hierarchical inconsistency related to corporate governance. The behavioral agency theory emphasizes that corporate governance should aim to motivate executives to perform to the best of their abilities (Pepper & Gore, 2015). Though inducing effective monitoring (Guo et al., 2022), hierarchical inconsistency leads to executives' great job demands, adding to their cognitive and emotional burdens and depleting their energies to make quality strategic decisions (Hambrick, Finkelstein, & Mooney, 2005). We suggest a more balanced view to perceive hierarchical inconsistency's impact on corporate governance.

Apart from theoretical contributions, our study also has practical implications. It is crucial yet challenging for family firms to attract and retain professional talents (Tabor et al., 2018). Our study has three implications for the management of professional executives. First, we suggest that family firms with hierarchical inconsistency should be particularly generous to outside managerial talents because these managers' work environment is more demanding than their peers in family firms without hierarchical inconsistency. Otherwise, according to labor economics (Campion & Berger, 1990), firms with this structural feature would be unlikely to attract enough talent and performance might be jeopardized (Chua et al., 2009).

Second, our study implies possible ways to mitigate nonfamily executives' job complexity under hierarchical inconsistency, reducing the firm's human resource costs. Specifically, family firms with hierarchical inconsistency may encourage more female family members to participate in the business and play mediator roles. The active involvement of female family members would alleviate the job demand of professional managers under hierarchical inconsistency and attenuate their need for high compensation. Alternatively, a clear, stable authority of the formal head will ease the confusion and pain of nonfamily executives in communication and coordination, lowering the likelihood of compensating them with higher pay. Note that our empirical findings indicate that the formal head's tenure is positively associated with nonfamily executives' compensation. This may imply that with time passing, the formal head builds closer personal relationships with the nonfamily executives, treating them more like family members and voluntarily increasing their pay. Nevertheless, this positive main effect of the formal head's tenure does not influence our key arguments. When hierarchical inconsistency exists, the longer tenure of the formal head reduces the complexity of nonfamily executives' work and attenuates their demand for higher compensation.

Finally, firms might want to avoid the hierarchically inconsistent structure to reduce high compensation costs when possible. For instance, in the succession period, the older-generation family leader might consider training the family successor when still holding the formal top position, and then fully withdrawing from any formal positions and leaving the young head a clear, uniform authority to reduce professional managers' job complexity.

Limitations and Future Research Directions

Our study is subject to limitations that imply further research opportunities. First, following prior research (Guo et al., 2022), we identified hierarchical inconsistency by examining whether the heads of the two hierarchies were different family members and measured this key construct with a dichotomous variable in our main analysis. Although we considered different types of kinship ties between the chair and the superior relatives in the supplementary analysis, we did not consider the hierarchical inconsistency experienced by people other than the two heads. Future research may develop more thorough empirical strategies to measure hierarchical inconsistency.

Second, nonfamily executives may engage in various job responsibilities in family firms. Different types of hierarchical inconsistencies might be associated with different daily job dynamics of nonfamily executives. For instance, hierarchical inconsistency between a parent and a child sometimes implies succession, and nonfamily executives might spend lots of time and energy facilitating the succession by mentoring the young successor and balancing the interests of different stakeholders during the turbulent transition period (Aronoff, McClure, & Ward, 2003; Michel & Kammerlander, 2015). In contrast, managers working in firms with hierarchical inconsistency between siblings might be less likely to focus on such responsibilities. Future research may explore how job complexity varies with specific responsibilities and is associated with different types of hierarchical inconsistency.

Third, although our study suggests that hierarchical inconsistency may result in higher pay for nonfamily executives, we do not know to what extent this higher pay would benefit the firm. Existing literature has argued that the pay dispersion in the top management team is negatively associated with firm performance because it induces feelings of inequity and damages cooperation (Ensley, Pearson, & Sardeshmukh, 2007; Jaskiewicz, Block, Combs, & Miller, 2017). Our empirical data suggest that, on average, nonfamily executives receive lower pay than their family peers in Chinese family firms, and hierarchical inconsistency increases the compensations of nonfamily executives relative to those of family executives. In other words, hierarchical inconsistency seems to foster the salary equity between family and nonfamily top managers in Chinese family businesses. Nevertheless, scholars have proposed competing theories and found mixed results on whether nonfamily executives receive higher pay than family executives in other contexts (e.g., Chua et al., 2009; Gomez-Mejia et al., 2003). Future studies could investigate the contingency factors determining how hierarchical inconsistency influences firm performance through compensation decisions.

Finally, our theoretical arguments on executive compensation may be applied to nonfamily firms. As suggested by extant research, hierarchical inconsistency also exists when the formal organizational hierarchy is misaligned with other salient informal hierarchical orders in the context, such as age, education, race, and gender (e.g., Park & Westphal, 2013). In nonfamily firms, to the extent that an informal hierarchy is as influential as the formal hierarchy among the top leaders, the heads of the two hierarchies may compete for dominance. We expect higher compensation for executives who are the bystanders in this contestation. Future research could test this hypothesis in other contexts.

Conclusion

Our study proposes a novel structural reason for nonfamily executives' high compensation levels in family firms. We argue that family top leaders' hierarchical inconsistency stimulates competitive behaviors between the heads of the two hierarchies, bringing extra complexity and challenges for nonfamily executives' work. Higher compensation is needed to make up for the demanding jobs of these professional managers. Our study provides new insights into family business research by proposing how the interplay between business and family structural elements shapes the job features and compensation of nonfamily executives.