The Effects of Family Firms’ Heterogeneity on Innovation Performance

Master Thesis

The Effects of Family Firms’ Heterogeneity on Innovation Performance

Jiaying Wang S3717712

Supervisor: Dr. M. C. Sestu Co-Assessor: Dr. O. Lindahl

January 20, 2020

Abstract

Recently, family business scholars have started to consider family firms as heterogeneous groups when studying family firms’ entrepreneurial performance. This paper mainly explores the effects of heterogeneity of family firms on innovation performance by discussing three sources of heterogeneity in family firm: family involvement in ownership, family involvement in management, and the number of family generations involved. A dataset of Chinese listed family firms in manufacturing industry is used for empirical analysis. The empirical results show that family ownership is positively related to innovation performance while the number family generations involved and the presence of a family CEO have negative influence on innovation performance, respectively. Family involvement in management is both positively and negatively related to innovation performance, and the type of relationship is decided by the measurement indicator.

Keywords: heterogeneity, family business, family ownership, family management,

generation, innovation performance, patents, R&D intensity

Table of Contents

1. INTRODUCTION ... 5

2. LITERATURE REVIEW AND HYPOTHESIS ... 8

2.1 I

F

F

... 8

2.2 T

H

F

F

... 10

2.3 F

I

I

P

... 12

2.3.1 Family Ownership and Innovation Performance ... 12

2.3.2 Family Management and Innovation Performance ... 16

2.3.3 Multiple Generations and Innovation Performance ... 23

3. METHODOLOGY ... 26

3.1 S

... 26

3.2 V

D

... 28

3.2.1 Dependent Variables ... 28

3.2.2 Independent Variables ... 29

3.2.3 Control Variables ... 29

3.3 M

... 31

4. RESULTS ... 31

5. CONCLUSION AND DISCUSSION ... 37

6. LIMITATIONS AND FUTURE RESEARCH ... 42

7. REFERENCE ... 44

List of Figures and Tables

Figure 1: Conception Model ……… 24

Table1: Descriptive Statistics ……….. 32

Table 2: Matrix of Correlations ………... 34

Table 3: Poisson Regression Results ………... 35

Table 4: OLS Regression Results ……… 36

1. Introduction

Nowadays, family firms are the most prevalent business organization form globally, and the enduring influence of family involvement on business and society is an essential part for all types of world economies (De Massis et al., 2013). The rapid economic development and fierce business competitions around the world push family firms to recognize that they have to consistently explore more competitive advantages for long term survival (Aghion et al., 2005). Innovativeness, as one of the most important entrepreneurial capacities that firms can adopt to achieve competitive advantage, thereby has naturally become a hot topic and a promising research field with increasing attentions in family business research in recent years (Erdogan et al., 2019).

Previous studies in regard to innovativeness of family business have devoted

substantial attentions to the typical distinctions between family firms and non-family

firms, since there are strong theoretical reasons to believe that both antecedents and

effects of innovation activities are quite different among them (Decker & Gunther,

2017). Family involvement, a key attribute distinguishing family firms from non-family

firms, is always considered to be the most dominant factor that causes the diverse

innovation behaviors between family firms and non-family firms. However, the results

are quite mixed when it comes to whether or not family firms are more innovative than

their non-family counterparts. For example, some studies argue that family firms can

effectively alleviate various agency problems (Fama & Jensen, 1983), making them

become more successful innovators than firms of other ownership structures. Others

assert that nepotism and the reduced access to external labor markets can significantly

limit innovativeness more in family firms than in other firms (Carney, 2005). Many of

such comparisons, although received empirical support in literatures, implicitly regard

family firms as homogeneous entities, which is a reason to explain the inconsistent

results of whether or not family firms are more innovative when compared with non-

family firms.

Even though making comparisons is important, there are evidences show that in fact, the variations of behavior and performance among family firms are as large as the variations between family and non-family organizations (Bennedsen et al., 2010;

Chrisman & Patel, 2012). Figuring out variations among family firms will not merely better explain the heterogeneous relationship between family involvement and behaviors (Chrisman et al, 2012; Lichtenhaler & Muethel, 2012), but also provide a deeper comprehending for family business theory research. As a result, more recent studies have started to use moderators, mediators or continuous measures of family involvement in recognition of the heterogeneity in family firms (Chua et al., 2012), and are trying to better explain the sources of heterogeneity among family firms as well.

Yet, in terms of the innovation performance of family firms, still we do not know enough, especially when exploring from the perspective of diverse sources of family firm heterogeneity. Consequently, in order to understand from a different point of view for the study of family business innovativeness, distinguishing and figuring out characteristics in family firms are quite necessary. Hence, this paper addresses the following research question: whether and to what extent differences in the heterogeneity of family firms affect their performance in innovation?

Current empirical evidences on this question are quite inconsistent. Some studies

find that the higher levels of family involvement lead to more positive influence on firm

entrepreneurial activities (Zahra, 2005), while others suggest this relationship in an

opposite way (Cerrato & Piva, 2012; Fernandez & Nieto, 2005). One possible

explanation for the contradictory empirical results is that most existing studies have

failed to distinguish between two major aspects of heterogeneity in family firm: family

involvement in ownership and family involvement in management, and separating them

to understand how they differentially affect innovation performance will offer new

evidence for family business innovativeness research. Thereby, in order to reconcile

this empirical puzzle and have a better understanding of the influence of family firm

heterogeneity on a firm’s tendency to innovation, this paper mainly focuses on the

contextual factors of heterogeneity in family firms. To be specific, in order to answer

the research question better, this paper mainly studies three characteristics of family

firms: family involvement in ownership, family involvement in management, and the number of family generations involved in a firm.

I propose that family firms are heterogeneous as they have different degree of family involvement in ownership, management, as well as generations, and the different involvement degrees then lead to diverse innovation performance within family firms.

Four relevant hypotheses are proposed in this paper. Specifically, a high level of family involvement in ownership is negatively related to firm innovation performance; a high level of family involvement in management is positively/negatively related to firm innovation performance; the presence of a family CEO is positively/negatively related to firm innovation performance; the more generation dispersions in a family firm, the lower innovation performance the family firm has. These hypotheses are tested through quantitative methods based on information of 579 Chinese listed family firms in manufacturing industry. Data for variables such as patents, family ownership proportion, the presence of a family CEO, ratio of family member in senior management role, the number of generations involved in a family firm, together with other seven control variables are collected.

Providing empirical evidence from China, this paper aims to narrow the research gap of current studies by focusing on a different perspective, a shift from viewing family firms as homogeneous entities to deeply studying on contextual factors of heterogeneity in family firms. This paper mainly offers three main contributions to the current literatures on innovativeness of family business. First, providing new evidence for heterogeneity of family firms by studying three sources of it: family involvement in ownership, family involvement in management, and family generations involved in a firm. Second, clarifying whether and to what degree these three sources respectively affect family firms’ innovation behaviors. Third, investigating family involvement in managerial positions by distinguishing between the ratio of family managers involved and the presence of a family CEO.

The paper is structured as follows. In the following section, I will review in detail

the literature on family firms’ heterogeneity and family firms’ innovativeness. The

hypotheses will then be developed based on this theoretical framework. Next,

methodology and data will be presented, and then analyze the data to test hypotheses.

Conclusion and discussion will then be presented, followed by the insights for limitations and future research.

2. Literature Review and Hypothesis

2.1 Innovation in Family Firms

The ability to innovate is generally recognized as a critical factor for firms to grow, success and survive in today’s competitive economic environment. Previous studies show that technology innovation is an essential avenue by which firms can pursue new opportunities as well as a pivotal dimension of firms’ entrepreneurial orientation (Rauch et al., 2009). Firms that stress technological innovation tend to support novelty and engage in creative processes (Lumpkin & Dess, 1996). Recently, family business scholars have begun to realize the necessity for family firms to be innovative (Naldi et al., 2007). The study of innovative behavior in family firms is essential since family firms are governed by unique sets of norms, cultures, and processes. These resources, which will not be found in other forms of organization, can be transferred into valuable tools for family firms to improve performance in innovation (Kellermanns et al., 2012).

Because innovation facilitates firms to “renew firms, enhance competitive advantage, spur growth, create new employment opportunities, and generate wealth” (Hayton &

Kelley, 2006), for family firms desired to gain more competitive advantages through improving innovation capacity, exploring how family involvement influences their innovativeness seems to be quite significant.

When studying innovation in the firm level, scholars often strengthen two key

concepts: innovation input and innovation output (Adams et al., 2006), which are two

important indicators for firms’ innovation performance. Innovation input mainly

measures resources that are put into innovation activities while innovation output is the

outcome that generated from innovation activities (Carayannis & Provance, 2008).

Family businesses scholars have paid substantial attentions to study family firms’

innovativeness as for both innovation input and innovation output. Studies on innovation input consistently indicate a lower level of innovation investments in family firms as compared with their non-family counterparts (De Massis et al., 2013). A prevalent explanation is that family firms usually serve two goals: to improve economic efficiency and to enlarge family social interests, which sometimes compete and even conflict with each other (Astrachan & Jaskiewicz, 2008). The particular combination of economic and non-economic goals in family businesses influences their firm strategic behavior (Chrisman et al., 2012). For family owners, they are loss averse for non-financial implementation that derived from business, making them risk averse to business opportunities that might reduce that implementation (Chrisman & Patel, 2012).

Given that R&D expenditures are sunk costs without certain benefits in a short time, family firms hence are more willing to pursue conservative innovation strategies associate with less intensive R&D investments (Miller et al., 2011). Substantial literatures can be found to support this perspective. For example, via using a sample of firms listed in the Standard & Poor’s 500, Block (2012) points out a negative effect of family ownership on R&D intensity. Lower levels of R&D investment in family firms are also observed by Chen & Hsu (2009) in listed Taiwanese corporations and Munari et al. (2010) by using a sample of publicly traded firms in Europe.

The relationship between family involvement and innovation has typically been studied in terms of outcomes (Chua et al., 2012). However, previous studies provide quite mixed results, which could be explained by using different innovation output indicators (Classen et al., 2012). Studies looking at patent data generally show a negative relationship between family involvement and innovation outcomes. Block et al. (2013) assert that family ownership has a negative effect on the number of patent citations, an indicator for the economic and technological importance of innovations.

They suggest that the pursuit of socioemotional wealth for the family may conflict with

the realization of ambitious innovation projects. Naldi et al. (2007) mention a lower

level of entrepreneurial orientation in family business, which is in line Block’s idea,

and as a result, family firms prefer to choose modest innovation strategies that are less

likely to challenge family financial and managerial control (Classen et al., 2012). On the other hand, when highlighting the capacity to introduce new products or services, scholars suggest a more positive connection between family involvement and innovation outcomes. Family firms possess unique resources and characteristics that can facilitate to transfer innovation inputs into innovation outputs, and at the same time benefit the implementation of innovation (Classen et al., 2012). Exemplary characteristics of family firms that stimulate innovation behavior are their long-term orientation (Zellweger, 2007), stewardship behavior (Eddleston & Kellermanns, 2007) and informal knowledge sharing (Zahra, 2012). Through analyzing firms across 47 developing markets, Ayyagari et al. (2011) find that family firms introduce more new products compared with non-family firms. The study of Craig & Dibrell (2006) also offers quite similar empirical implications of superior innovation capacity in family businesses.

2.2 The Heterogeneity of Family Firms

Family firms are businesses in which a family’s involvement significantly influences a firm’s behavior, strategy and performance, so that to achieve a vision of how the firm will benefit the family (Chua et al., 1999), and therefore it is generally accepted that family involvement in the business makes the family business be unique.

Today it is widely recognized that family firms are the most prevalent form of organization in the world (La Porta et al. 1999). However, although there still are studies largely discussing and attempting to show how family firms are different from their non-family competitors in the aspects of entrepreneurial behavior, strategy and performance, scholars have long realized that family firms are not homogeneous (Sharma et al., 1997), and have started to focus on the characteristics that shared by different family firms.

The heterogeneity of family firms implies the degree of typical family firm

characteristics may be contingent on what constitutes a family business (Schmid et al.,

2015). One convincing reason of causing the heterogeneity in family firm is its diverse sources (Chua et al., 2012). For instance, the heterogeneity in family firm can be deeply affected by family particular culture, which roots from family history, family social relations and the beliefs as well as values that embedded in the family (Dyer, 1986).

Some family firms are prone to develop cultures that make their organizations inflexible, resistant to change and inclined to stick to traditions, hence becoming less favorable to new proactive entrepreneurial strategies (Hall et al., 2001). Besides, family governance, especially family involvement in firm ownership and management team, is a primary source that makes family firm be heterogeneous. Indeed, family firms are heterogeneous as they differ in the extent of family involvement in ownership and family involvement in management, which emphasize two different important sides of firm governance. Family involvement in ownership mainly focuses on the family’s role as business owners while family involvement in management highlights the family’s role in business management and operation (Liang et al., 2013). The varying levels of family ownership and family management then exerts impacts on the firm’s entrepreneurial strategies, decision-making, capacities, and performance in business (Sharma & Nordqvist, 2013).

An integrated family business theory should not only be able to distinguish between family and non-family firms, but also be able to understand the variations among family firms (Schmid et al., 2015). Family business researchers have long sought to understand the deep-rooted impacts of family involvement in ownership and management on firm goals, resources, performance and strategies (Sharma et al., 1997), for instance in regard to performance in innovation (Chua et al., 2012), so that to provide better patterns of incentives, authority, norms for resource allocation, and conflict resolution for family business development (Carney, 2005; Daily et al., 2003).

Due to the importance of family firm heterogeneity and the necessity to figure out both

antecedents and outcomes of it, in the following sections I will discuss different sources

of heterogeneity in family firm as well as their impacts on firm innovation performance

concretely.

2.3 Family Involvement and Innovation Performance

The heterogeneity in family firms may influence their options on innovation behaviors (Nieto et al., 2015). Family involvement in ownership and family involvement in management emphasize different aspects of family governance and control. Because of different roles in firms performed by the family, the ways of how they affect a firm’s tendency and strategies to innovation might differ as well.

2.3.1 Family Ownership and Innovation Performance

Based on Daspit et al. (2018), family involvement in ownership can vary along with six dimensions, which are: the number of families or family members involved in ownership, family ownership proportion, the dispersion of ownership among family members, the relationships in owners, the demographic characteristics of owners, and the nature of owners’ involvement in governance. Each dimension, separately or in combination, can have influential behavioral and performance implications at both the firm and family level. Usually, the number, dispersion, and relational variables are combined to identify different ownership structures, which are also agreed with the description of three basic forms of family ownership of business that mentioned by Gersick et al. (1997): controlling owner, sibling partnership, and cousin consortium.

Lansberg (1999) further argue that family firms may vary in regard to whether over time they choose to recycle through the same ownership form, move to a new form, or revert back to the previous form. The options and consequences of each ownership form have distinct features that are impacted by and influence the incentive structures, norms of authority and prevailing legitimacy in a family firm (De Massis et al., 2012).

Accordingly, these features create valuable contexts in which certain governance

mechanisms are likely to be more effective than others to achieve prioritized goals

(Sharma & Nordqvist, 2013). Researches about the consequences that different

ownership dimensions or forms brought are noted as well. A study of Swiss family

firms figures out an inverted U-shape relationship between ownership dispersion and

internal family ownership transfers, which suggests that life stage and multi- generational involvement are significant predictors of ownership transitions (Wiklund et al., 2013). Another study with the U.S. start-up venture teams asserts that when teams are consisted of couples, they will achieve first sales earlier than non-family teams or teams with biologically linked members (Brannon et al., 2013). Furthermore, research believes that owners’ personal characteristics such as age, gender, ethnicity, and nationality may influence a firm’s behaviors, decision-making process, and performance outcomes significantly, thus generating questions of the applicability of research findings from one region to another (Daspit & Long, 2014; Khayesi et al., 2014).

In addition, family firms are heterogeneous and differ in the degree of ownership concentration (Sciascia & Mazzola, 2008). Scholars indicate that the diverse levels of family ownership concentration may result in various entrepreneurial strategies and performance in family firms. Usually, the larger shares a family owns in the firm, the higher the firm-specific investments in human and financial capital and the higher the dependence on the company performance (Andres 2008; Munari et al. 2010). More specifically, when family owners have large ownership shares, a substantial portion of their personal wealth will be tied with the firm, so they may become increasingly risk averse, fearing that failed strategic and decisions will damage their personal wealth (Bianco et al., 2012; Liu et al., 2011). On the contrary, when family members own a relatively small ownership share, the embeddedness of the firm in a family will be quite low and only a limited amount of family capital is probably at risk, therefore, family owners are more willing to make investments in potentially risky projects such as increasing investments in developing firm innovation (Miller et al., 2009).

Research suggests that ownership significantly influences a firm’s strategic

choices, especially in family firms where owners hold a significant equity stake

(George et al., 2005; Zahra, 1996). First of all, family ownership might impose capital

constrains that may limit a family firm from investing in innovation (Carney, 2005). In

the process of technological innovation, R&D investment plays a crucial role. The

investment in R&D facilitates the advancement of corporate innovation and is widely

considered to be a key factor to gain and sustain a competitive advantage (Ettlie, 1998).

However, agency theory tends to indicate that family firms as vehicles by which the owners can pursue private family interests at the expense of the firm, the minority, and particularly the non-family shareholders (Le Breton-Miller & Miller, 2009). As a result, scholars argue there is a negative relationship between family firm ownership and R&D investment. For instance, Block (2012) suggests that family-owned firms often suffer from unique internal conflicts which create new agency costs associated with R&D expenditure and ultimately enable lower levels of R&D investment. Similarly, Munari et al. (2010) find that family ownership is negatively associated with R&D investment due to a more limited propensity for risk-taking among their controlling shareholders.

High levels of R&D investment allow firms to accumulate stronger technological and market capabilities for both process and product innovation (Grant, 2002). Accordingly, family firms without enough R&D investment may have rare opportunities to improve their innovation capacity due to the lack of relevant resources. Moreover, family members have strong emotional attachments to original entrepreneurial strategies, thus they are always reluctant to change and adopt a conservative stance (Habbershon et al., 2003; Vago, 2004). Indeed, family firms are often unwilling to recruit outside managers, and the scarcity of more talented employees in managing R&D projects might harm the improvement of innovation capacity. Family firms also have the tendency to avoid external financing, as capital sources from outside are more likely to threaten the family’s ownership and control to the firm (Carney, 2005). However, by often limiting access to external financial resources (Gallo et al., 2004) and avoiding pursuing new opportunities that may have potential to damage assets, family firms may ultimately reduce the possibility to select valuable innovation projects.

Additionally, studies point out that an organization’s absorptive capacity is

positively related to its innovation performance (Kostopoulos et al., 2011; Levinthal,

1990). The knowledge-based view argues that in the contemporary knowledge-

intensive business environment, firms increasingly rely on external information sources

to promote innovation and improve performance (Morgan & Berthon, 2008). A firm’s

ability to leverage knowledge and technology available out of boundaries highly

depends on its absorptive capacity (Cohen & Levinthal, 1990), which is defined as a firm level capability to recognize the value of external technology, assimilate it, and apply it to commercial ends (Zahra & George, 2002). However, scholars illustrate that family ownership, an important antecedent of absorptive capacity (Kotlar et al., 2019), may exert negative effects on a family firm’s absorptive capacity and thereby harm a firm’s performance in innovation. Kotlar et al. (2013) suggest that as one of the important components of absorptive capacity, the ability of technology acquisition is lower in family firms than in non-family firms, which lead to a lower level of potential absorptive capacity in family firms. Chrisman & Patel (2012) state that R&D investments, as stock of prior knowledge, is lower and more heterogeneous in family firms, and as a result may cause a lower overall absorptive capacity in family firms.

Indeed, family ownership may produce a type of emotional attachment that regarding the firm as “our business” (Demsetz & Lehn, 1985). Such emotion may impact the hierarchical structure and informal social relations within the firm (Cannella et al., 2015), setting the constraints of what kind of sources should be acquired, assimilated, transformed, and exploited (Lane et al., 2006). For family members who master high ownership degree of the firm, since firm is the lifeblood of the family, the costs of failure always outweigh the benefits of success. Due to the fear of being failed and the need to keep the business developing for multiple future generations (Habbershon &

Williams, 1999), the owners of family firms always take family interest as their first priority. High ownership degree often suggests high control rights. As a result, when it comes to acquire, allocate, transfer and exploite external sources, family firms’ owners will consider the type that can benefit the family first instead of that can benefit the firms’ comprehensive development. Under such a situation, knowledge about innovation activity may not be picked up from outside and will not be allocated or transferred within family firms, because the development of innovation is a slow process with high failure risks that cannot bring profits within a short period (Martin &

Scott, 2000). Instead, owners are more willing to invest in less uncertain assets which

are more predictable and stability such as buildings and production machineries

(Anderson, 2012). Without getting enough resources from outside and absorbing

current knowledge timely, firms will be failed to carry on innovativeness research and lose opportunities to enhance innovation capacity. Consequently, based on the reasonings above, I propose the following hypothesis:

H1: A high level of family involvement in ownership is negatively related to the innovation performance of family firms.

2.3.2 Family Management and Innovation Performance

Family involvement in managerial positions is another significant source of heterogeneity in family businesses which should be considered, which primarily entails family participation in strategic decision making (Sciascia & Mazzola, 2008), including strategies to enhance innovation performance (Anderson & Reeb, 2003; Kowalewski et al., 2010). The dimensions of family involvement in management are quite similar to those associated with family involvement in ownership, since the number, proportion, and relations all come into play when family members do the positions in the managerial hierarchy (Daspit et al., 2018). Yet, there are studies put forward that in fact, ownership does not impact family firm profitability, and what really matters is the degree to which the family is involved in firm management team (Sacristan-Navarro et al., 2011).

There are different forms of family involvement in firm management team, and each of them may impact firm entrepreneurial performance and allow the implementation of family goals (Daspit et al., 2018). For example, the ratio of family members in the management team, such as family members in the position of board of director, board of supervisor, or senior management is thought to affect how the firm engages in innovation. Studies illustrate that high ratio of family member involved in management team may have negative impacts on family firm performance (Sciascia &

Mazzola, 2008; Matzler et al., 2014), and the degree to which these negative effects

exist is altered by the level of dynamism in the business environment (Chirico & Bau,

2014). A similar effect of family involvement in management is also noted for other

firm strategies and behaviors, like those associated with firm internationalization (Liang et al., 2013). Besides, the family firm generations involved might change how family management affects the firm performance. Sciascia et al. (2014) point out that family involvement in management is positively related to firm performance when the firm is at the later-generational stage of development, which is the time when most of family firms begin to professionalize and prefer financial over socioemotional wealth.

The presence of a family CEO seems to benefit firm performance in broad measures as well. However, whether or not family CEOs can bring advantages to their family firm heavily relies on the generation in the position (Minichilli et al., 2010). When comparing with non-family firms, family firms led by the second-generation CEOs underperform whereas family firms led by the first-generation CEOs outperform typically (Villalonga & Amit, 2006). CEO duality also matters to family firm performance. Braun and Sharma (2007) believe that under the situations where the roles of CEO and chair are separated, the board can keep control of decision-making process as well as to monitor the behaviors of family agents more effectively, thereby encouraging a greater extent of investment in R&D. On the contrary, the presence CEO duality cannot serve as an effective governance control to prevent the risks of increasing family entrenchment (Chen & Hsu, 2009), which will harm the profits of firms.

Moreover, the involvement of founder is another type of family involvement in firm management. McConaughy and Phillips (1999) talk about the “founder effect” in their study with respect to differences among family generations. They clarify that the performance of family firms is particularly strong when the founder is still active as CEO. In consistent with McConaughy and Phillips, Sacristan-Navarro et al. (2011) argue that founder CEOs and descendant CEOs have varying effects on firm performance owing to their different behavior, and the firm performance will be superior when the firm has a founder-CEO in charge.

A relatively large body of evidence asserts that the degree of family involvement

in management carries on serious implications for firm performance and innovation

(Erdogan et al., 2019). Concerning managerial involvement, studies suggest that family

firm behavior improves when firm owners involve other family members in business

management (Eddleston & Kellermanns, 2007). According to the stewardship perspective, high levels of family involvement in management may benefit innovative behavior, since family members are better able to identify both challenges and opportunities that a firm is facing, understand where the organization is headed, and make decisions that can maximize firm value (Zahra, 2005). Similarly, because family involvement extends beyond decision-making to the implementation of decisions, it is easier for family members to make adjustments when having outcomes that beyond expectations, which potentially assists to alleviate losses (Eddleston & Kellermanns, 2007). Family firms that encourage family members to participate in firm management would benefit from the development of psychological ownership and share destiny among family members, thus increasing family members’ sense of responsibility and commitment to the firm (Kellermanns et al., 2012). Consequently, family members will devote more attentions to monitoring innovation process and capturing the gains of innovative behavior. In comparison, when family firms control internal environment with few family managers, it may delay the firm’s ability to exploit opportunities associated with their innovations. For instance, Zahra et al. (2004) illustrate that when a family firm has to concern itself with the opportunism behaviors of non-family managers, the resulting internal controls may decrease the firm’s capacity to respond effectively to environmental changes and to pursue more market opportunities.

Organizational structures that destroy environmental responsiveness, decentralization, and entrepreneurial alertness are less likely to pursue new innovation opportunities (Hayton & Kelley, 2006). Besides, the information asymmetry associated with innovation activities makes it difficult for outside managers to know the value of innovation investment, which may pressure managers to overstate the importance of current bottom-line incomes and ignore innovation investment (Froot et al., 1992).

Thereby, I expect family firms with greater family member involvement in management will benefit more to innovative behavior than those with less family involvement in management and then propose the following hypothesis:

H2a: A high level of family involvement in management is positively related to the

innovation performance of family firms.

Family involvement in management may influence a firm’s innovation performance in a negative way as well. Knowledge-based innovation is strategically efficient only when widely dispersed knowledge can be shared, synthesized, and utilized in unique approaches (Zahra et al., 2007). However, family business scholars propose that the “familiness” phenomenon can discourage non-family members from sharing knowledge (Chirico, 2008). To be specific, high degree of family management only improves formal knowledge sharing among family members and limits knowledge sharing among family and non-family members (Zahra et al., 2007). The short of knowledge resources and the lack of informal knowledge sharing may reduce the knowledge creation efficiency and hence weaken technological innovation development. Even if family firms have their own diversified knowledge resources, the knowledge integration and utilization of non-family members may be inefficient because of the negative influence caused by nepotism (Tsao et al., 2009). Too much family involvement in firm management undermines the recombination of dispersed knowledge that possessed by non-family members and reduces the implementation efficiency of innovation strategies as well (Liang et al., 2013).

Moreover, family involvement in management limits the ability to solve organizational conflicts in time (Carney, 2005). A high level of family involvement in management usually means a high level of power concentration in firms (Cai, 2012).

In organizations with high power concentration such as organizations with pyramid structures, conflicts cannot be resolved instantly and directly (Goel et al., 2011).

Because not only general employees have few avenues to express their own ideas, but

also superior leaders who control the family business would seldom like to

communicate with employees positively, disagreements and conflicts between

employees and leaders always appear. When these disagreements or conflicts cannot

be expressed and solved timely, employees may become low motivated to work

diligently for family firm, and responsibility avoidance and free-riding behaviors may

also arise because of it (Bernheim & Stark, 1988). Furthermore, as disagreements or

conflicts are either avoided or not confronted directly, there is a potential for them to

gradually become deeper and more severe, which has the possibility to spill over into business area and then hurt the family business (Lee & Rogoff, 2006). Because these negative outcomes relate to the performance and even the survival of family firms, managers have to shift their attentions from how to assist their family firm to gain competitive advantages to how to solve conflicts and create a harmony atmosphere for the organization. As such, effectiveness in the innovation process cannot be achieved.

Based on the discussions above, I propose the following hypothesis:

H2b: A high level of family involvement in management is negatively related to the innovation performance of family firms.

A typical form of family involvement in management is to have family members act as firm CEOs (Liu et al., 2012), and the notion that the family CEOs matter is also at the core of the family business literatures (Jiang & Peng, 2011). The behavior of family firms with a family member as CEO has been indicated to differ from that of family firms with an outside CEO (Miller et al., 2013).

Prior research on innovation has increasingly highlighted the important role of firm leadership, particularly CEOs in the innovation process (Jung et al., 2003). In fact, firm CEOs, not merely have substantial rights in resource allocation decisions (Hambrick & Mason, 1984), but also monitor and direct the usage of those resources.

CEOs encourage, select, and nurture innovation activities that emerge within the firm

(West et al., 2003), and create conditions for the subsequent implementation of

innovation (Crossan & Apaydin, 2010). Thus, scholars insist that family CEOs exert

positive influences on firms’ innovativeness. For instance, by study the moderating

effect of family CEO on the relationship between family ownership and firm’s

performance, Al-Dubai et al. (2014) point out that family firms managed by family

CEOs outperform firms not managed by family CEOs in innovation capacity. Duran et

al. (2014) illustrate that family firms with a family CEO have higher innovation outputs

than those without a family CEO through a meta-analysis. Usually, firm leaders such

as CEOs and general managers hold high proportion of stakes and their interests are

highly connected with both family and non-family shareholders (Jiang & Peng, 2011).

Therefore, they are more incentive to effectively monitor innovation activities in a powerful way and ensure efficient conversion of innovation inputs into innovation outputs (Duran et al., 2014). When the CEO is a family member, his or her willingness to efficiently take charge of innovation process will be even higher (Anderson & Reeb, 2003). Previous research in the context of technological innovation in the pharmaceutical industry (Cardinal, 2001) shows that such increased control on firms and projects results in a more efficient and less wasteful innovation process.

The resource-based view also suggests that appointing family members as CEOs is a beneficial way to improve family firms’ innovation performance (Barney, 1991).

Family firms benefit from the accumulation of human capital within the firm and the creation of efficient routines when converting innovation input into innovation output (Carnes & Ireland, 2013). When firm CEO is acted as a family member, his or her individual human capital, especially knowledge about the internal affairs of the firm, might be beneficial for harvesting such advantageous firm-level human capital in the innovation process. Specifically, family CEOs are endowed with superior and tacit knowledge about their firm’s stakeholders and routines, which outside CEOs might not possess or possess in a lower degree, and such knowledge will allow for superior resources orchestration (Sirmon et al., 2011). Given that resource management is a challenging task that is at the core of the innovation process (Subramaniam & Youndt, 2005), such individual knowledge will lead to superior innovation outputs (Carnes &

Ireland, 2013). Besides, family CEOs build trustworthy relationships with customers,

suppliers and partners in other firms, which may provide family firms with valuable

complementary innovative assets and knowledge to improve innovation (Duran et al.,

2014). Family CEOs are likely to follow with their firms no matter how bad situations

they are facing. Such characteristics may enhance partners’ trust and render partners to

actively offer valuable feedbacks to the firm in the process of developing innovation

capacity. Because trust is an important antecedent for harnessing a firm’s network in

the innovation process (Phelps, 2010), it seems that family firms with trustworthy

feature have better innovation performance. Based on the discussions above, I propose

the following hypothesis:

H3a: The presence of a family CEO is positively related to the innovation performance of family firms.

However, having a family CEO can sometimes be detrimental to innovation development in the firm level. First of all, studies assert that family firms with a family CEO have lower innovation inputs than those without a family CEO (Duran et al., 2014).

As discussed in H1, wealth concentration renders family firm owners be sensitive to the uncertainty inherent to their investments (Bianco et al., 2013), and consequently, incentivizes them to avoid investments in uncertain projects (Miller & Folta, 2002).

Family CEOs typically have their personal wealth concentrated in their own firm and possess control through ownership rights (Chrisman et al., 2005). As such, family CEOs are likely to suspend investment in innovation due to the risky, unpredictable, and long- term-oriented characteristics of innovation activities. Because of the lack of innovation inputs, family firms can hardly explore for innovativeness. Secondly, not all family CEOs can perform well and contribute to innovation process. A family may only have few qualified family members, so family owners who motivated by the desires of family succession at the expense of non-family minority shareholders, may appoint incompetent family members as CEOs (Arregle et al., 2007). Once in the position, those family CEOs may feel they are not threatened anymore since the high level of parents’

altruism cannot be attacked easily, and thus, those family CEOs who are appointed by their family do not have to maximize efforts to keep their positions, not to mention to contribute for family firms’ innovation performance.

Moreover, family CEOs are often failed to solve organizational conflicts in an

efficient way as well, not just because the altruism makes family CEOs hard to making

fair judgements, but because family relations are often too complicated to manage

effectively (Arregle et al., 2007). And those organizational conflicts, if not solved in

time, will impede or even damage firm entrepreneurial performance, including

innovation. It is well known that CEOs make decisions for the most important business

and are considered to be the most powerful organizational leaders in firms (Minichilli

et al., 2010). Due to the family connections as well as altruism between a firm’s CEO

and owners, with a family CEO in charge, it is easier for family owners to undertake some unfair transactions that can privately benefit themselves but harm non-family minority shareholders (Jiang & Peng, 2011; Villalonga & Amit, 2006). Yet, even if some family CEOs can realize the unfair operations in firms, they seldom point them out or stop them. As family CEOs usually have common interests with family owners, once they stop those unfair operations and harm the owners’ interests, theirs will be harmed as well. Because CEOs always neglect of those unfair transactions among firms, gradually employees will feel disappointed to their leaders as well as to the firm.

Accordingly, employees may not work wholeheartedly, not to mention contribute creative ideas to help firm develop innovation activities. Besides, family CEOs themselves may have a hard time dealing with other family members. Family internal conflicts such as sibling rivalry and power competition often appear among family managers (Gomez-Mejia et al., 2001). Family CEOs also unwilling to allow other family members to participate in the decision-making process of family firms (Eddelston & Kellermanns, 2007), which could cause family animosities (Jiang & Peng, 2011). Although CEOs are leaders of firms, finding comprehensive approaches to figure out internal organizational conflicts is still a difficult task for them. Dealing with firm internal conflicts and animosities can exhaust CEOs time and energy, which may not only distract CEOs’ attentions from making innovative strategies, but also impede CEOs to catch advanced knowledge associated with innovation. Therefore, I propose the following hypothesis:

H3b: The presence of a family CEO is negatively related to the innovation performance of family firms.

2.3.3 Multiple Generations and Innovation Performance

Family control can be diverse from unified control by the founding generation to

co-control held by multiple generations (Gersick et al., 1997). As regards the

generational ownership, scholars argue that low levels of generational ownership

dispersion are more likely to result in performance advantages when firms engage in innovative behavior (Kellermanns et al., 2012), which is mainly owning to the “founder effect” of the first-generation family firms. In fact, research points that founder-led family firms perform better than non-family firms or family firms led by later generations (Miller et al., 2007). First generation family members are so unique that they are headed by entrepreneurial founders who have the ability to recognize business opportunities and exploit through the creation of a new business (Aldrich & Cliff, 2003).

Founders are seen to be critical because they make the initial decisions about the organization’s mission, goals, and strategy. They also make daily business operating decisions that may influence the organization as it grows and evolves (Davis &

Harveston, 1999). Family members in the first generation often regards the family business as their whole life achievement and is required to possess superior entrepreneurial and technological skills to explore firms’ innovativeness in order to maintain and extend family firm’s success (Fahlenbrach, 2009). Similarly, Le Breton- Miller and Miller (2013) describe family members of earlier generational stages as being committed to the family firm and having a strong emotional attachment with each other. Moreover, due to historical ties and founders’ anticipation of handing over their family business to descendent generations, family owners have longer investment horizons and commitment. This extended horizon encourages family members to focus on the firm’s long-term value, leading to more investments in innovation and less shortsighted behaviors (Tsao & Lien, 2013).

Unlike founders, family successors display a reduced identification with the firm, and also family members are less attached to each other (Gersick et al., 1997). They are more prone to be conservative and interested in preserving the family’s current wealth.

Because there is no guarantee of financial success, subsequent generations may be less willing to support innovation, instead, focusing more on how the expenses of pursuing innovations may threaten family wealth (Kellermanns et al., 2012). Besides, inertial forces, which lead firms to follow previously successful business strategies (Hannan &

Freeman, 1984), may restrict succeeding generations to explore innovativeness. That is

to say, after the successful operating management of the first generation, family firms

have already established so long time that have developed a solid customer base and reputation, hence subsequent generations can easily maintain a good entrepreneurial performance simply by following previous well-established practices and strategies such as in innovation, rather than looking for more competitive advantages by themselves. Gradually, prior practices will become obsolete products and without new innovative strategies advanced with the times, family firms that managed by successors cannot have any advantages and will not perform well in the fierce business competitions.

Additionally, as family firms become more mature, they are passed on to subsequent generations and thus family ownership will be dispersed in a larger number of family members. Potential negative consequences such as family members conflicts (Davis & Englis, 1999), parental altruism (Schulze et al., 2003), and selection of top managers from a limited pool of human talent, which is the family itself (Schulze et al., 2001), will arise gradually. All these problems increase agency costs that will prevent family firms from creating new innovative projects (Block, 2012). In brief, when compared with multiple ownership generation dispersions in family firms, innovative behavior is much more integral and outstanding to the ownership dispersion in single generation, especially in the first generation, as they strive to establish themselves in their industry. Thus, I propose the following hypothesis:

H4: The more family generations involved in a family firm, the lower innovation performance the family firm will have.

Figure 1 illustrates four hypotheses developed in this paper.

Figure 1: Conception Model

3. Methodology

3.1 Sample selection

The hypotheses are tested through quantitative methods based on information of 579 Chinese listed family firms in manufacturing industry. The samples include all family firms listed on the Shanghai and Shenzhen Chinese Stock Exchanges in 2016.

Information about innovation performance, family firm ownership, family members in

the management team, the presence of family CEO, numbers of generation involved in

firms, together with seven control variables are collected. All data are obtained from

the China Stock Market & Accounting Research Database (CSMAR), designed and

developed by Shenzhen GTA Information Technology Company, a major provider of

Chinese data (Cai, 2012). The CSMAR consists professional and comprehensive

information of listed family businesses in China since 2008. Family business in

CSMAR is defined into three categories: single natural person entrepreneurs, multiple

natural persons entrepreneur firm, and multiple persons family firm. To be specific,

single natural person entrepreneur firm indicates the actual controller is individual and

has no relatives holding shares or serving as directors, supervisors and senior management in the listed firm or controlling shareholding company; multiple natural persons entrepreneur firm refers to actual controller is multiple natural persons and none of them has relatives, meanwhile none of their relatives holds shares or serves as directors, supervisors and senior management in the listed firm or controlling shareholding company; and multiple persons family firm is that besides the actual controller, at least one of the family members holds shares, manages or takes charge of the listed firm or controlling shareholding company.

This paper focuses on family firms from twenty different manufacturing industries, which has two reasons. Firstly, manufacturing has traditionally played a key role in the economic development of developing countries such as China (Haraguchi et al., 2017).

Even though in the twenty-first century, manufacturing development still remains relevant and necessary for those countries trying to catch up with more advanced economies and to provide increasing standards of living for their citizens (Naude &

Szirmai, 2012). Secondly, since joined the World Trade Organization in 2001, China has become the world factory with huge amounts of trade surplus. Therefore, it is quite necessary to study what consequences Chinese manufacturing has been brought in these years and figure out superior strategies for future manufacturing development. Given the cultural and institutional diversity across regions in China (Redfern & Crawford, 2008), and since previous studies using Chinese companies find a significant effect of the geographical location on firm performance (Husted et al., 2016; Ioannou &

Serafeim, 2012), I further divide the firms into seven regions, which are South China,

East China, Central China, North China, Northwest China, Southwest China and

Northeast China. Firm size may affect the rate that family firms develop their

technological capabilities as well as the abilities to keep competitive as well (Zahra et

al., 2007), which may influence firms’ innovativeness (Katila, 2002). Thus, I exclude

micro-enterprises with less than 10 employees so as to focus on firms that are large

enough to experience and demonstrate some managerial decision making as well as

family involvement and influence (De Massis et al. 2013). The original firms are 1931,

and after the filtration based on industry and firm size, overall 579 firms are left. And

it is noted that after filtering, all these 579 firms are the type of multiple persons family firm.

3.2 Variable Description

3.2.1 Dependent Variables

The dependent variable of this paper is firms’ innovation performance. Several concepts have been discussed in literatures for capturing innovation performance (Sofka & Grimpe, 2009). While some concepts focus on innovation input such as R&D expenditure, R&D investment, and R&D intensity, others concentrate on the outcomes of innovation activities like patents, new process and products. In this paper, the degree of firms’ innovation performance is measured via the number of patents, denoted by Patents, which refers to the number of patents that a Chinese listed family firm cumulative obtained in China’s domestic until the end of 2017. Patents has been generally considered to be the key indicator of innovation capacity, both in national level and in firm level. The use of patents as a measure of innovation performance is also widely supported in literatures. Tsao & Lien (2013) use the number of patents granted to a firm to measure Taiwan’s publicly listed firms to exam the impact of family management on firm performance and innovation implication of internationalization.

Similarly, Matzler et al. (2014) use patents as an indicator to test how family governance influence innovation output.

To the purpose of conducting robustness tests, I use R&D Intensity as an

alternative indicator for firms’ innovation performance. R&D intensity is a kind of

input indicator that mainly measures R&D resources that put into the innovation

process, as well as a mainstream approach for the measurement of innovation

performance. Consistent with prior studies (Chen & Hsu, 2009; Cohen & Levinthal,

1990), in this paper R&D intensity is defined as the ratio between a firm’s R&D

expenditure and total sales in each fiscal year (De Massis et al., 2012, Boly et al., 2013).

3.2.2 Independent Variables

As the independent variable of this study, the heterogeneity of family firms is measured from four dimensions: family involvement in ownership, family involvement in management, the presence of family CEO, and family generations dispersion.

Followed studies of Liang et al. (2012) and De Massis et al. (2013), family involvement in ownership is measured by family ownership proportion, denoted by Family Ownership. In CSMAR database, the calculation of family ownership proportion is to multiple the proportion held by the actual controller in each layer of shareholding relation chain or to sum after multiplying the proportion held by the actual controller in each layer of each shareholding relation chain. Family involvement in management is measured by the ratio of family members in the role of senior management (Matzler et al., 2014; Liang et al., 2012), denoted by Family Management. A dummy variable is used to capture the presence of a family CEO, denoted by Family CEO, which takes value of 1 if a controlling family member acted as the CEO, and 0 otherwise (Chu, 2011;

Peng & Jiang, 2010). Family Generation is used to measure the number of family generations involved in the business (Miller et al., 2007; Kellermanns et al., 2012).

3.2.3 Control Variables

A series of control variables are included in the regression model in order to control for firms and their effects on innovation performance.

Industry In order to catch the impact of different types of manufacturing industry on family firms’ innovation performance, twenty different types of manufacturing industry are measured as factor variables.

Region The region is divided into seven different categories and is caught as factor variables: South China, East China, Central China, North China, Northwest China, Northeast China and Southwest China, so that to check if region-specific constraints exert influences on family firms’ performance in innovation.

Firm Size Firm size is measured by the number of full-time employees in family firms (Lu & Beamish, 2001; Calabro et al., 2013). Since employee numbers are highly skewed, a logarithmic transformation of the number of employees is used in the regression analysis.

Firm Age Craig & Moores (2006) insist that innovation performance is related to the life stages of firms. Older firms may have better information disclosure and corporate governance mechanisms (Claessens et al., 2002), and hence perform better.

Therefore, firm age is included in control variables and is measured by the natural logarithmic transformation of the number of years the firm has been in existence (Davis

& Harveston, 2000), which is also used to control for the possibility of entrenchment in family firms (Chrisman & Patel, 2012)

Familization Way A dummy variable is set to catch the way of family familization in order to control the diverse familization approaches, which takes value of 1 if the firm came to market as a family firm (direct establishment), and takes value of 0 if the firm came to market as a state-owned or non-natural person/family holding firm and then turns to family firm due to equity transfer, restructuring (indirect establishment).

Firm Performance Firm performance is measured from the accounting-based

indicator by Return on Assets (ROA), which is calculated as the net income divided by

the book value of total assets (Anderson & Reeb, 2003; Perez-Gonzalez, 2006). Firm

performance will control the influence of a firm’s prior performance on strategic

decisions regarding innovation investment (Barker & Mueller, 2002).

Firm Leverage Usually, family firms show lower debt levels when compared to non- family firms (Naldi et al., 2007). High leverage would increase a firm’s interest payment costs and hence may harm performance (Chang & Wong, 2004). Thus, firm leverage, defined as a firm’s total debt divided by total assets, is included as a control variable (Munoz-Bullon & Sanchez-Bueno, 2011).

3.3 Measures

To reduce the impact of outliers, I first winsorize all variables at the 1 and 99 percent levels. Poisson regression analysis is used to test hypotheses. Poisson regression models are attractive for innovation performance measurement since not only these models are appropriate for integer data (counts of events), but they also account counts that are aggregated over time periods. In this paper, the dependent variable is non-negative count of patents, and observations are cumulative obtained in China’s domestic until the end of year 2017. Poisson regression is thus an appropriate method to use. Because the dependent variable of robustness test (R&D intensity) is a continuous variable, OLS regression is used when doing robustness test.

4. Results

Table 1 reports descriptive statistics for main variables used in the regression

models. The average number of patents is 79.169, average family ownership is 41.94%,

and average family members in the senior management role is 0.186%. In addition, the

family CEO of 0.972 indicates that among these 579 firms, more than 97% of them

have family member as CEO, indicating that Chinese listed family firms commonly

appoint family members as firm CEO. Familization way of 0.912 refers that within

these 579 family firms, over 91% are established directly whereas about 9% are

established indirectly. The matrix of correlations among the main variables are

presented in Table 2. Table 2 also displays the variance inflation factors (VIFs), which shows that the multi-collinearity is not a matter of concern in this study, since the maximum value of the observed variance inflation factors (VIF) is far below 10.

Table1: Descriptive Statistics

Variables N M SD Min Max

Family Ownership 579 41.94 16.632 9.776 83.33

Family Generation 579 1.259 0.439 1 2

Family CEO (dummy) 579 0.972 0.164 0 1

Family Management 579 0.186 0.158 0 0.75

Firm Age (in log) 579 2.712 0.316 1.946 3.332

Firm Size (in log) 579 7.41 0.985 5.485 10.212

Familization Way (dummy) 579 0.912 0.284 0 1

Leverage 579 0.331 0.17 0.051 0.769

Firm Performance 579 0.066 0.053 -0.1 0.216

Patents 579 79.169 206.608 0 1480

R&D Intensity 579 0.008 0.024 0 0.145

Table 3 reports the results of the Poisson regression. Model 1 only contains control variables. Firm age is found to have a significant and negative impact to patents (p <

0.01), therefore the longer a family firm has been established, the lower numbers of patents it obtains. Firm size, however, has a significant and positive effect on patents (p < 0.01), which indicates that large firms usually generate a large number of patents.

Familization way is referred to have a negative effect on patents, but this relationship

is insignificant (p > 0.1). Firm performance has a negative and significant impact on

patent (p < 0.01), which means that family firms with high level of return on assets are

likely to have lower numbers of patents. Firm leverage has influence on patents

significantly and negatively (p < 0.01). Model 1 also shows that geography is indeed a

matter of number of patents for Chinese family firms. Based on the results in Model 1,

when taken central China as baseline, family firms in north China and southwest China

have obtained more patents while family firms in east China, northeast China,

northwest China and south China have fewer patents. Besides, among these seven

regions, in regard to the obtained patents number, family firms in northwest China have

the least whereas family firms in southeast China have the most. This finding is thus

consistent with the argument that in China, the economy development in coastal regions (southeast China) far outweighs that in inland regions (northwest China), because the coastal regions are open economic zones where firms can easily integrate into the international economy, collaborate with foreign companies in investment, production- manufacturing as well as distribution, and access to the most advanced resources in the world (Demurger et al., 2002). As for the industry control, although all of them belong to manufacturing relevant industries, it is obvious that different types of manufacturing have different levels of influence on patents.

Model 2 adds family ownership. A positive and significant relationship (p < 0.01) is found between family ownership and patents, thus Hypothesis 1 is not supported.

Model 3 then adds family management and family CEO. A highly significant and positive effect is reported for family influence through participation in senior management roles (p < 0.01), thereby providing support for Hypotheses 2a. The presence of a family CEO has a significant and negative effect on patents (p < 0.01), which supports Hypothesis 3b. Model 4 includes all independent variables. Family generation is found to exert a negative influence on patents, which is also significant (p

< 0.01). This finding supports Hypothesis 4 and suggests that the more family

generations involved in a firm, the fewer number of patents this firm has been

accumulated obtained.

Table 2: Matrix of Correlations

Variables VIF (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)

(1) Patents 1.000

(2) R&D Intensity 0.017 1.000

(3) Family Ownership 1.20 0.025 -0.049 1.000

(4) Family Generation 1.05 0.006 -0.033 0.000 1.000

(5) Family CEO (dummy) 1.03 -0.058 0.008 0.060 -0.093** 1.000

(6) Family Management 1.21 -0.017 -0.070* 0.242*** 0.104** 0.070* 1.000

(7) Firm Age (in log) 1.07 -0.037 -0.009 -0.057 0.092** -0.023 -0.025 1.000

(8) Firm Size (in log) 1.33 0.191*** -0.036 -0.096** 0.044 -0.074* -0.148*** 0.010 1.000

(9) Familization Way (dummy) 1.15 0.011 -0.020 0.230*** -0.052 0.133*** 0.145*** -0.235*** -0.068 1.000

(10) Firm Performance 1.21 -0.047 -0.104** 0.285*** -0.050 0.012 0.139*** -0.005 0.128*** 0.034 1.000

(11) Leverage 1.33 0.118*** -0.060 -0.144*** -0.057 -0.016 -0.134*** -0.039 0.419*** -0.115*** -0.179*** 1.000 Notes: N=579

Significant levels: *** p < 0.01; ** p < 0.05; * p < 0.10.