Professional CEO and Internationalization: An empirical study for Chinese family firms

(1)

Professional CEO and Internationalization:

An empirical study for Chinese family firms

Jie Zhang

Supervisor: Prof. Dr. C. L. M. Hermes

Co-assessor: Mr. H. Vrolijk

January, 2015

University of Groningen - International Financial Management University of Uppsala – Economics & Business

Esdoornlaan 694 +31 6 17 62 89 10

(2)

1

ABSTRACT

This study examines the effect of having professional CEOs on the degree of internationalization based on a sample of Chinese family firms that have CEO transitions during 2008 to 2012. Confronting the succession of family business and an increasing number of influential Chinese firms at the global stage, the traditional management in family firms is on the way of transferring into professional management. In this paper, both positive and negative aspects of having professional CEOs are addressed considering Chinese specific situations. Further discussions on the one hand focus on how professional CEOs help release the risk bearing burden from the families who have concentrated ownership according to the theory of the separation of ownership and management, validated by a significant empirical result; on the other hand suggest controlling agency costs through a better monitoring board with more Independent Non-executive Directors (INDs), which is yet unproven by this study. Additionally, in a short period after the CEO change to professionals, family control in decision making is still important to improve internationalization.

JEL classification: M12; F23.

(3)

2

The term of “Professional CEO” is derived early in 1841 from the U.S. and became a unified mature notion in the Western developed countries. However in China, this concept is still new and undeveloped. Chinese firms, compared with U.S. and U.K. firms, are often managed by family members rather than non-family professional managers (Fan, Wei and Xu, 2011). As reported by Forbes China1 in 2014, among 1.2 million firms in the mainland China, 1.025 million are private-owned firms, surpassing 80% proportion, among which more than 80% are founder-managed firms, 80% of which will face the leadership succession to the second generation or the professional CEO in the next decade. While China Association for Professional Managers2_covered

that 82% offspring are not very willing to take over the business from their parents, then the possibility of welcoming professional managers into family business is promoted.

Meanwhile, China is also described as an active and successful exporting country in 1990s since the implementation of open-door policy. But it is only recently that a number of Chinese firms are recognized as global leaders and players, e.g. Huawei, Alibaba Group, Jingdong and Lenovo. The examples of successful internationalization have pushed going abroad to be one of the strategic decisions that firms may consider. In other words, firms tend to emulate others’ successful actions to develop themselves. Nevertheless family firms, which do not possess necessary knowledge and talents to enter the foreign markets, confront resource risks. The defective institutional system in China may further hinder the strategic behavior of expanding abroad due to the lack of financial support (Liang et al., 2014) and the undeveloped investor protection system demotivates family business owners to disperse their shares, thus bearing a high level of risk aversion and reluctant to make risky strategic decisions.

1_{Three innovative thoughts about successful succession for hundred-years-old firms. Available at}

http://www.forbeschina.com/review/201409/0037154.shtml.

2_{CAPM was founded in Beijing on 30 Jun 2012 by National Committee for Professional managers’}

(5)

4 Therefore, confronting the opportunity of adopting professional CEOs, whether Chinese family firms can benefit from this transition to enhance the internationalization level is of concern in this paper. To answer this question, arguments pro and against professional CEOs are presented. On the one side, the advantages of professionals over family members as CEOs are discussed regarding managerial resources. Also based on the theory of separation of ownership and management, risk bearing burden of family firms is reduced when having professional CEOs, leading to more internationalization. On the other, agency issues produced from having an outside member in family firms are argued considering the immature market of professional managers in China. To compensate for the ineffective institutional system, family firms can improve their internal governance with independent non-executive directors (hereafter INDs). IND system is also regarded as a sign of professional management (Fiegener et al., 2000;

Stewart and Hitt, 2012). Based on this theoretical scenario, and by referring to previous studies about the impact of family control on firm performance or firm value (Qilin and Wen, 2003; Lin and Hu, 2007; Peng and Jiang, 2010) and on firm internationalization (Zahra, 2003; Liang et al. 2014), this study extends to examine whether having professional CEOs can facilitate international operations of Chinese family firms.

(6)

5

2. Theory and Hypothesis

This chapter is organized by four parts. The first part is the application of OLI theory to emphasize the importance of relational assets in internationalization and Chinese kinship that limit the international development for Chinese family firms. The second part presents the relations between family ownership and internationalization. The third part outlines both positive and negative impacts of professional CEOs. Based on the double-sided arguments, the fourth part extends the discussion through the theory of the separation of ownership and management as well as the monitoring mechanism of INDs.

2.1 Internationalization of Chinese family firms

China, as one of the largest emerging economies, overtook Japan as the world’s third largest exporter after Germany and the U.S. since 2004 (Child and Rodrigues, 2005). The underlying motives behind firms going abroad may be explained by Dunning’s (1988) ownership-location-internalization (OLI) theory or eclectic paradigm. More specifically, the internationalization activities performed by firms intend to apply their ownership or firm-specific strengths in overseas markets, where they exploit locational advantages, thus to develop business more efficiently and achieve possible economies of scale.

(7)

6 different parts of a rapidly ramifying enterprise”. Therefore, professional outsiders are becoming essential to enhance the competence of family firms.

2.2 Family ownership and Internationalization

Family ownership is highlighted by La Porta et al. (1999), who found that the ownership of 20 largest listed firms in the richest 27 countries are mostly highly concentrated. But researches have different opinions about the impacts of family ownership on a firm’s tendency to internationalize. Zahra (2003) proves a positive relationship for U.S. family firms, declaring that owner-manager may peruse long-term growth to create wealth for their families and next generations. George et al. (2005) work out the opposite result for Swedish small medium-sized family business, and Fernández and Nieto (2005) also show a negative connection for Spanish firms. They argue that the risk aversion is induced with a higher ownership, leading to less strategic movements.

In this paper, based on above arguments, the institution-based theory of corporate governance proposed in the study of Peng and Jiang (2010), is applied to explain this relation in Chinese family firms. La Porta et al. (1999) point out that in countries with a lower level of minority shareholder protection, people are reluctant to involuntarily disperse their shares. Usually those who hold more shares have more say about how to spend the money from all the shareholders, thus small shareholders’ benefits are often expropriated by large shareholders under undeveloped minority shareholder protection

(Claessens et al. 2000). In this case, Chinese family firms are trying to keep ownership within the family as much as possible to guarantee their control powers. But due to the concentrated ownership, the risk-averse attitude towards strategic decisions is produced (Ward, 1997; Naldi et al., 2007). This is because with a lot of wealth invested in the firm, the desire to take risks is undermined due to the possibilities of financial loss.

(8)

7 Forbes China3_{, the majority of family firms are founder-managed. As the initial equity}

investors of the firms, the founders are more concerned with the survival of the firm (Zahra, 2003). To protect the legacy for future generations, they are inclined to maintain control powers and are more alerted to venturing activities, such as expanding to foreign markets, which may place the firm in risks. Consequently, Chinese family firms are supposed to be more conservative and risk averse, indicating a less tendency of internationalization.

Hypothesis 1: The share of family ownership is negatively related to the degree of internationalization in Chinese family firms.

2.3 Professional CEO and Chinese Family business

Compared with family ownership, family involvement in management is another variable that affects a firm’s tendency to internationalize (Zahra, 2003; Peng and Jiang, 2010; Liang et al., 2014). Previous scholars sent firms surveys and questionnaires to define the degree of family involvement when setting company goals and making strategic decisions, like international expansion (Zahra 2003; Fernández and Nieto, 2005; Liang et al., 2014). In this paper, this methodology is not applied because questionnaires mostly contain subjective questions that may be diverted from the reality and responders can have false answers affected by personal reasons. But a family CEO or an outsider CEO is considered when calculating the degree of family involvement in these literatures, thus in this paper a family member or a professional as CEO is regarded as the measure of low or high family involvement4.

2.3.1 Professional CEO and the Transition in China

As reported by Forbes China, founding family business in China nowadays face the succession of the second generation or hiring professional CEOs. Zhang and Ma

3_{Since 2009, Forbes China starts to investigate Chinese family firms and publish the survey reports}

each year. Available at http://www.forbeschina.com/billionaires/family.

(9)

8 (2009) argue that more overseas education and the recent rapid economic progress bring modern management into Chinese family business, indicating that the traditional management is on the way being substituted by professional management. This process, according to Daily and Dalton (1992), is inevitable once the existing expertise inside the firm no longer supports its growth and competence in the industry. At this turning point, the founders choose to delegate the control to more professional managers (Whisler, 1988). Further, this process is mentioned by Zhang and Ma (2009) as market imperative, indicating that the potential development of entering global markets has pushed the adoption of professional management into Chinese family business.5

This transition taking place in China is due to the difference between family members and professionals as CEOs. On the one hand, from resource-based view, internationalization pursues to apply firm-specific resources in overseas markets to make profits. Hereinto, the knowledge of foreign markets and the perceptions of the diversity between home and host country can help overcome psychic distance, which is disturbing the entry into the foreign markets (Vahlne and Nordström, 1992). Especially in emerging countries, like China, lack the necessary experience in foreign markets. Nevertheless family firms are short of such competent resources and managerial capabilities (Dyer, 1989; Hitt et al., 1997). Besides, internationalization is an uncertain and risky strategic decision (Liang et al., 2014) with a short-term financial loss, and requires a long-term resource investment and talent supply to generate returns. Thus family firms that opt to open global markets demand professionals, aligned with Dunning (1988), who declares that intangible competitive resources, such as knowledge, skills and experiences, significantly influence the firm’s tendency to internationalize.

On the other hand, family business lags behind regarding strategic growth due to the inflexibility of the leadership (Fernández and Nieto, 2005), implying that the owners

5_{The causality issue is about the internationalization forcing the increase of professional CEOs in}

(10)

9 are conservative and resistant to change strategies in a dynamic market. This prevents the family firms from responding sensibly to distinct and unfamiliar foreign markets. But professional CEOs would not stick to past practices, because they tend to make decisions based on logic, not intuition (Dyer, 1989), thus focusing on the objectivity (Schein, 1967). They also favor decentralization of decision-making (Fernández and Nieto, 2005), which is also beneficial in responding to local markets quickly. Therefore professional CEOs will perform better in international markets, thus contribute to the process of internationalization.

Hypothesis 2a: Having professional CEOs is positively related to the degree of internationalization in Chinese family firms.

2.3.2 Professional CEO and Agency Issus in Chinese context

Arguments against the positive effects of professional CEOs are related to agency costs, which are stated in the work of Jensen and Meckling (1979). Based on their illustrations, agency costs exist as long as two or more people are working together. The definition include the monitoring costs by the principals and costs that the managers would pay, such as the reduction of remuneration, if they betrayed their commitment to align their actions with the interests of the principals as well as the residual loss of principals’ welfare incurred when divergent behaviors are taken by those two parties. As this paper concentrates on the perspective from the family firms’ owners who invest the most among various principals, monitoring costs and the residual loss will be discussed in this section.

(11)

10 governance in China. Adoption of a professional CEO into family business and strategic decision about internationalization both bring uncertainties into the firms. According to North (1990), an appropriate institutional protection can help reduce such uncertainties. In his argument, uncertainties are derived from other individuals’ behaviors and human mind has limited capacities to process all the information of others. Under such circumstance, institutional framework is enforced to make sure that people behave in a certain standard. Managerial self-serving behaviors are thus minimized. However, with a defective institution, Chinese family firms are not able to lessen these uncertainties. As clarified by Peng and Jiang (2010), an effective investor protection environment in the U.S. and U.K. encourages founding families to delegate daily management to professional CEOs. By contrast, in China, with a much lower level of investor protection, family business owners employing professional CEOs are taking a risk of having high monitoring costs.

Meanwhile, information asymmetry caused by Chinese cultural traditions of familism and distrust of outsiders (Zhang and Ma, 2009) can also negatively influence the performance of internationalization. McCollom (1988) states that family members share information very often and the communication inside the firm is intensive, but in a less formalized way as described by (Zhang and Ma, 2009). To be specific, the family has developed its own specific communication system, which constitute the culture in a family firm. Further Chinese family business owners used to hide confidential information within the family (Qilin and Wen, 2003). The lack of a formal information flow channel makes difficult for outside CEOs to play their roles, leading to divergent actions of the family and the CEO and causing residual loss. Hall and Nordqvist (2008) also indicate that the unfamiliarity with family business would invalidate the professionals’ expertise and limit their advantages.

(12)

11 owners do not actually understand how the managers conduct the international activities. In this case, monitoring cost is increased to curb mangers’ self-serving behaviors.

According to Zahra (2005), agency costs may inhibit the venturing activities of family firms, including investing in foreign markets. Agency costs could weaken the competitiveness of the firm in international markets regarding the lack of information knowledge and supports from the family who are engaged in supervision rather than strategic activities. Besides, effected by agency costs, family firms could also loss supports from shareholders. Without necessary financial resources, family firms in emerging countries, like China, find difficulties to implement international operations (Liang et al., 2014). Since professional CEOs in Chinese family firms generate substantial agency costs as discussed above, the internationalization is hindered.

Hypothesis 2b: Having professional CEOs is negatively related to the degree of internationalization in Chinese family firms.

2.4 Further discussions about Professional CEO

At this point, positive and negative arguments regarding professional CEO come to a tie and it is hard to predict if there is a significant tendency to either aspect. Thus the remaining part of this chapter aim to magnify the positive effects of professional CEO by discussing the interaction with family ownership while moderate the negative effects of professional CEO by introducing INDs.

2.4.1 Interaction of Family Ownership and Professional CEO

(13)

12 But when key decision managers do not bear an extensive return, their decisions would be left unaffected by wealth. To put it more precisely, family business owners would be less suppressed by the risks of financial loss if they assign the management to professionals. Compared with family members, professional CEOs, who are employed from outside the family, hold much lower shares (Claessens et al., 2002). According to Dyer (1989), professional CEOs gain their power from the key positions, but not from the ownership, thus are able to analyze risks more rationally. Thus the negative relations between family ownership and the level of internationalization will be reduced.

Later, Claver et al. (2008) argue that the level of risk aversion may decrease when the goal congruence is achieved between owners and managers, thus promoting active behaviors of strategic growth, such as internationalization. The means to achieve the goal congruence include providing share incentives (Jensen and Meckling, 1979). Also in China, CEOs in most cases, are seated on the board of directors, participating in the discussions about company strategies and development policy. Therefore, the reach of interest alignment stands a fair chance.

Hypothesis 3: The negative connection between family ownership and the level of internationalization is weaker for firms with having professional CEOs.

2.4.2 Interaction of Professional CEO and IND

The corporate governance can be improved by including INDs to curb the potential agency costs in family firms with professional CEOs. Generally speaking, the function of the board has two dimensions as suggested by Jones et al. (2008): One is to monitor management and the other is to provide resources6. Aiming to reduce agency costs and maximize firm value, the board composition is improved with independent non-executive directors (INDs), also known as outside directors, which is derived from the U.S. in the 1930s. This Anglo-American corporate governance is developed during the

6_{The second dimension is not discussed here as this section is aiming to discuss the moderating effect}

(14)

13 1970s. Following the trend, China, as reported by Ji and Wang (2005), applied the independent director system very late until the 21st century.

Even though family owners, after delegating decision management to professional CEOs, can play the role of monitoring, the adoption of an outside board improves the mechanism. IND, distinct from executive directors or affiliate directors, is completely independent from the management level. They are part-time directors who process incredible intellect and abilities of judgments since they avoid any transactions or relationships with the firm. Hence they are the best candidates for the roles of “ratification and monitoring” (Fama and Jensen, 1983). Zahra and Pearce (1989) also agree that the role of INDs in monitoring is the ultimate solution for a valid corporate control. Further, recalling to section 2.4.1, Fama and Jensen (1983) contend that the mechanism underlying for the success of the separation of ownership and management is an effective monitoring system to control the agency problems. Adoption of INDs is also represented by Fiegener et al. (2000) as one essential part of organizational “life-cycle process” to implement more “professional management”. In this context, the role of INDs is particularly vital in Chinese family firms that are going to hire professional CEOs and pulling through the transition.

Further, Peng and Jiang (2010) argue that external governance mechanisms can compensate for invalid internal corporate monitoring. But oppositely, the undeveloped institutional environments in China require the firms to set up and rely on their own valid internal control systems. Thus it’s important for Chinese family firms, in which professional CEOs are at a high possibility of acting inconsistently with the interests of the family, to recruit more INDs to better control the managerial opportunism (Sherman et al., 1998; Chen and Jaggi, 2001). Accordingly, the positive impact of INDs on better monitoring and minimizing agency costs is expected.

(15)

14

3. Methodology

3.1 Sample and Model

The definition of family firm in this paper is generally aligned with Qilin and Wen (2003), who suggest that the ultimate owner is an individual or a family and this owner directly or indirectly is the largest shareholder of the firm. Based on this principle, the selection can be divided into four steps. Firstly, the sample is roughly collected from CSMAR database7, in which 551 Chinese publicly listed firms are selected if there are any CEO changes during 2008 to 2012. With different CEOs in those years, the distinct effect of a professional CEO on the degree of internationalization in a specific family firm is easier to be tested. Next, all these firms are mapped with their foreign sales data, which can be retrieved from Wind database8. 253 firms are left after cutting off those without available foreign sales in any of those five years. Thirdly, collectively-owned firms that are set up by multiple individuals without kinship and those firms that changed the ownership structure from state-owned to family-owned should be taken out. This step requires a deeper look at more detailed information about the ultimate owners and developing history of each firm with either the annual reports or the disclosures in CSMAR database. 218 firms survived this filtering. Lastly, after checking board information and family membership9, 143 family firms remain in the end.

Following previous papers (Zahra, 2003; Qilin and Wen, 2003; Fernández and Nieto, 2005), Panel Least Squares is performed to test the hypothesis in this study. The Hausman Test shows that the estimation results of the fixed effects model is better than the random effects model. Thus Panel Least Squares with cross-section fixed effects is the method applied in this paper. The sample is a balanced panel of 143 firms in a five year time series, composing 735 observations. Hypothesis 1 and 2a, 2b will be tested

7_{CSMAR database is developed by GTA Information Technology Co., Ltd., based on the standards of}

international database such as CRSP and COMPUSTAT, focusing on Chinese financial market and economic research. Available at http://csmar.gtadata.com.

8_{Wind Information Co., Ltd (Wind Info), headquartered in the Lujiazui Financial Center in Shanghai,}

is a leading integrated service provider of financial data, information, and software. Available at http://www.wind.com.cn/En/Default.aspx.

(16)

15 first. If the test results support 2a significantly, Hypothesis 3 is then tested; otherwise, Hypothesis 4 will be tested; or if there show no significant results, both Hypothesis 3 and 4 will be examined. Moreover, based on the pattern of CEO changes in the collected sample, further analysis is conducted and empirical test is performed for the particular firms that changed their CEOs from family members to professionals. This additional step helps to better answer the question that whether professional CEOs improve the internationalization or not. Details will be shown in section 5.

3.2 Dependent Variable

Exporting is the first step of internationalization (Johanson and Vahlne, 1977) and especially an easy mode of entering the global markets due to a lower risk (Cerrato and Piva, 2012). Thus in this paper the degree of internationalization is defined as export intensity. According to Zahra (2003), Fernández and Nieto (2005), George et al. (2005), Liang et al. (2014), this variable can be measured as the ratio of foreign sales to total sales. Data were downloaded from Wind database for period 2008 to 2012.

3.3 Independent Variables

Family ownership: Zahra (2003) calculates family ownership by combining the percentage of total shares owned by the owner and the percent owned by the close family members as the family ownership. Liang et al. (2014) uses the same technique. Close family members include parents, children, sisters, brothers and also those who are married to this family. Here the calculation of shares should be clarified. Mainly two kinds of family ownership measures exist in China, cash flow rights and voting rights, as studied by Qilin and Wen (2003). Cash flow rights focus on how much wealth the family invest, while voting rights, which can be larger than cash flow rights through pyramids (Figure 1) and cross-shareholding (Figure 2)10. Referred to La Porta et al. (1999), who pointed out that in countries with poor minority shareholders protection,

10 _{Tekušová and Kohlhammer (2008)}_{states “Pyramid structures are built by a tree structure of company}

(17)

16 voting rights, rather than cash flow rights, serve as the proxy of control power that the family is trying to protect. Thus in this paper, the voting rights percentage of the ultimate owners and the family members are applied as family ownership to measure family control. Data are available from CSMAR database.

Figure 1: Pyramid Structure

The ultimate owner controls Firm N with cash flow right of 50%*50%+40%*10%=29%, while the voting right is 60% through pyramid holding.

Source: Adapted from Figure 1 in the work of Tekušová and Kohlhammer (2008).

Figure 2: Cross Shareholding Structure

The ultimate owner controls Firm N with cash flow right of 50%*25%=12.5%, while the voting right is 50% through cross shareholding. This is a simple example from the paper of Claessens et al. (2000), but cross shareholding is a much more complicated ownership structure in many other cases.

Tekušová and Kohlhammer (2008), in their paper, constructed a visual analysis system to show the shareholder networks.

Source: Adapted from the example provided in the paper of Claessens et al. (2000).

(18)

17 Family involvement: In this study, family involvement is defined by two measures. One is the measure of professional CEO. Referred to Lin and Hu (2007), if CEO is not a family member of the largest shareholder or the ultimate owner, then the CEO is defined as professional CEO. Dummy variable is set equal to 1 if it’s professional CEO. For both measurements, checking of family membership is the critical and tricky part, which is first identified by sharing the same family name. But many family names are common in China, such as Liu, Li, Chen, Chang and Wang, thus first names are also inspected in such situations. When the first names contain two characters, one of them should be shared within family members. CSMAR discloses the family relationship among top ten shareholders. Related firm information on the website, like Sina Finance and Yahoo Finance, present the former working experience of the CEO, which could serve as a reference. Annual reports and other information online are carefully looked through to make the assessment as accurate as possible. Forums like Sina, Baidu and Tianya11 can also be a help. If the family membership is not confirmed, the case will be deleted from the sample. The other is the measure of IND. The board of directors in Chinese firms are the top level for decision making. The adoption of IND is conducted as the ratio of INDs on the board. But the rest of the board is not entirely occupied by family members, since corporate employees can also take part in the board, thus the ratio of family members on the board is measured as well.

3.4 Control Variables

Firm characteristics should be controlled for this study.

Firm size: Firm size can influence the access to the necessary resources in international operations (Zahra, 2003). Moreover, larger firms tend to consider internationalization as “natural steps of growth” (Liang et al., 2014). The current study uses the natural log of full time employees as the firm size12, and data is obtained from

11_{Tianya is a famous social interaction platform in China, with 200 million users each month.}

Available at http://bbs.tianya.cn/list-stocks-1.shtml.

12_{Previous literatures measure firm size differently as employees (Zahra, 2003; Fernández and Nieto,}

(19)

18 Orbis13_{, validated through CSMAR database.}

Firm age: The longer the firms are running, the higher possibility for them to consider further international growth. Firm age is measured, following previous studies (Zahra, 2003; Liang et al., 2014) by number of years since the firm is founded.

Past performance: Since previous studies found a positive relation between past firm performance and the degree of internationalization (Zahra, 2003), the return on asset (ROA), profit before tax/total assets, of last year is obtained through annual reports and Orbis.

R&D intensity: As implied by Dunning (1988), intangible assets are important firm specific advantages to internationalize. A high R&D intensity indicates the competitive intangible assets in a firm, leading to a high tendency to go abroad, as the analysis done by Zahra (2003). Especially when Chinese firms are experiencing the progress of technologies in recent years, this firm character should be controlled14. In this study, R&D intensity is measured by the number of patent and trademarks. Data is available in Orbis database. Unfortunately, a strong dispersion within the data cause problems for a feasible regression result and natural log is not efficient since many firms have zero patents or trademarks. Therefore, after adding one to each original number to avoid zeroes, natural log is conducted to keep the data valid in a possible maximum extent.

Second shareholder: Qilin and Wen (2003) state that “one dominating stock monopolizes”15 is common in Chinese family firms. The second largest shareholders are more focused on protecting their own interests from being plundered by the largest shareholder, thus prefer to get a quick return. The extra supervision performed by the

13_{Orbis is a database of extensive financial data and information about worldwide firms.}

14_{A reverse relationship between R&D and internationalization also exist as studied by Erdener and}

Shapiro (2005), who suggest that some Chinese firms are going abroad to access high technology industries.

15_{“One dominating stock state-owned monopolizes” is the word used by OECD to describe the}

(20)

19 second largest shareholder may lead to underinvestment of the largest shareholder (Zhang, 2012), hence hindering long-term investments, including internationalization, which needs a continuous resource commitment (Liang et al., 2014). Following Laeven and Levine (2004), 10% shares is the sign of a second largest shareholder, otherwise the firm has only one largest shareholder. Thus a firm with a second shareholder who has more than 10% shares is coded 1, and 0 elsewise in this study. Information can be found in annual reports.

Indirect shareholding: Indirect shareholding refers to pyramids (Figure 1) and cross-shareholding (Figure 2) in this paper. Because family ownership as earlier stated in section 3.3 does not show whether the family holds the voting rights directly or not, this variable is included. Through indirect shareholding, the family can gain its control power with less cash flow invested. The level of risk aversion associated with internationalization is therefore lowered. A firm with indirect shareholding is coded 1, indicating and 0 otherwise.

4. Results and Discussions

4.1 Descriptive statistics

(21)

20

Table 1：Descriptive statistics

This table shows five descriptive statistics, namely mean, median, maximum, minimum and standard deviation of all the variables. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark, Second Shareholder and Indirect Shareholding are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares. Indirect Shareholding is another dummy set to 1 if ownership holding is by pyramids or crossholding.

Variables N Mean Median Maximum Minimum Std. Dev.

Foreign sales percentage 715 0.234 0.132 0.990 0.000 0.257

Professional CEO 715 0.690 1.000 1.000 0.000 0.463 Family Ownership 715 0.357 0.334 0.796 0.075 0.151 FR 715 0.171 0.143 0.857 0.000 0.114 IR 715 0.366 0.333 0.600 0.143 0.050 Firm Size* 715 3.288 3.286 4.901 1.342 0.444 Firm Age 715 13.175 13.000 27.000 2.000 4.403 Past Performance 715 0.055 0.053 0.821 -1.016 0.110 Patent* 715 0.825 0.778 2.879 0.000 0.660 Trademark* 715 0.117 0.000 1.255 0.000 0.221 Second Shareholder 715 0.302 0.000 1.000 0.000 0.459 Indirect Shareholding 715 0.730 1.000 1.000 0.000 0.444 *This variable is logged.

Table 2 shows family ownership that is calculated averagely in five years. If taking 20% as the benchmark16 for family firms stated by Claessens et al. (2000), about 88% firms in this study hold more than 20% shares, implying that the sample is reasonable.

Faccio and Lang (2002) list the control rights for public firms in western European countries with the mean of 38.48% and Claessens et al. (2000) make a similar list for firms in East Asian countries, whereby the mean is 24.41%. Both datasets are collected from public traded firms where the largest shareholder has more than 5% voting rights.

16_{Define holding more than 20% shares as the benchmark of whether a firm is family controlled, because}

(22)

21 If cutting off non-family firms according to the benchmark, the average for family firms in both studies should be higher. In comparison, Chinese family ownership in China in this sample, 35.7% on average, is not as much as concentrated. The underlying reason can be the step-by-step enterprise revolution strategy taken by China. To be more specific, the predecessors of Chinese family firms in this study are mostly state-owned and collectively-owned firms. The step-by-step strategy is identified by the government selling the shares gradually and slowly to the family or other legal persons (Qilin and Wen, 2003), leading to a low concentration. The other reason could be related to the transition period mentioned in section 2.3.1. As Child and Rodrigues (2005) reveals, the government would hold part of the shares to take control of the business in a transitional stage.

Table 2: Family ownership distribution of Chinese family firms

This table showsthe distribution of family ownership for of 143 observing firms. The ownership of the firm is calculated on a five-year average.

Family Ownership N Percent Cumulative Percent

<10% 1 0.70% 0.70% 10%-19.99% 16 11.19% 11.89% 20%-29.99% 39 27.27% 39.16% 30%-39.99% 40 27.97% 67.13% 40%-49.99% 25 17.48% 84.62% 50%-69.99% 18 12.59% 97.20% >69.99% 4 2.80% 100.00% Total 143 100%

Source: Made by author.

(23)

22

Table 3：Correlation matrix

This table shows correlation matrix of all the variables. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark, Second Shareholder and Indirect Shareholding are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares. Indirect Shareholding is another dummy set to 1 if ownership holding is by pyramids or crossholding.

01 02 03 04 05 06 07 08 09 10 11 12

01.Foreign sales percentage 1.000

02.Professional CEO 0.008 1.000 03.Family Ownership -0.034 -0.191 1.000 04.FR -0.001 -0.375 0.362 1.000 05.IR -0.090 -0.057 -0.064 0.021 1.000 06.Firm Size* 0.183 0.094 -0.026 0.024 -0.122 1.000 07.Firm Age -0.108 0.063 -0.079 -0.052 0.060 -0.039 1.000 08.Past Performance -0.001 -0.099 0.110 0.116 -0.042 0.008 -0.036 1.000 09.Patent* 0.107 -0.146 -0.018 0.103 0.103 0.166 -0.199 0.126 1.000 10.Trademark* 0.244 -0.044 0.083 0.050 -0.029 0.183 -0.039 0.043 0.271 1.000 11.Second shareholder 0.105 0.119 -0.095 -0.007 0.000 0.021 -0.058 0.053 -0.003 0.001 1.000 12.Indirect Shareholding 0.026 0.246 -0.017 -0.100 -0.139 0.118 0.267 -0.129 -0.233 -0.139 -0.005 1.000

(24)

23

4.2 Regression results

Table 4：Foreign sales percentage for Chinese family firms

This table shows the results of Panel Least Squares with cross-section fixed effects model about foreign sales percentage. Model 1 is to test hypothesis 1 and 2, Model 2 and model 3 are designed for hypothesis 3 and 4 respectively. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark, Second Shareholder and Indirect Shareholding are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares. Indirect Shareholding is another dummy set to 1 if ownership shareholding is by pyramids or crossholding.

Model 1 Model 2 Model 3

Hypothesis 1 &2 Hypothesis 3 Hypothesis 4

Constant 0.191 0.246 0.177 Independent Variables Family Ownership -0.203*** -0.305*** -0.203*** Professional CEO 0.008 -0.059* 0.027 FR -0.008 -0.000 -0.006 IR 0.061 0.078 0.099 Interaction

Professional CEO*Family Ownership 0.171**

IR*Professional CEO -0.052 Control Variables Firm Size 0.074*** 0.068*** 0.074*** Firm Age -0.013*** -0.014*** -0.013*** Past Performance 0.031 0.028 0.031 Patent -0.011 -0.006 -0.011 Trademark -0.118** -0.121** -0.118** Second shareholder -0.055*** -0.057*** -0.055*** Indirect Shareholding 0.078 0.079 0.078 R-squared 0.902 0.903 0.902 Adjusted R-squared 0.875 0.876 0.875 Observations 715 715 715 * significant at 10% level ** significant at 5% level *** significant at 1% level

(25)

24 addressed. By adding professional CEO lag variable in the equation, the relation between foreign sales percentage and professional CEO lag is examined. The insignificant p value suggests that foreign sales change cannot force the increase of professional CEOs in family firms. Thus causality issue is of less concern in this study.

As represented in Table 4, the regression results of Model 1 strongly support Hypothesis 1 that family ownership is significantly negative related to the degree of internationalization at 1% level. Most recently, Liang et al. (2014) have done the empirical analysis for private Chinese family firms that have 75% ownership averagely and found a significant negative relation based on the same argument about risk awareness. Compared with Liang et al. (2014), the sample in this study are all listed firms with a much lower family ownership (averagely 35.7%) but yet the empirical result is in line with Liang et al. (2014)17. Nevertheless, the result for professional CEO is not crystal clear since the positive sign is very low and insignificant. Thus Hypothesis 3 and 4 are further conducted under Model 2 and 3.

In Model 2, the interaction of family ownership and professional CEO is added into the equation, corresponding to Hypothesis 3. The result shows a positive sign and a statistically significance at 5% level. It obviously verifies Hypothesis 3 that having professional CEOs is critical to reduce the negative relation between family ownership and internationalization. Both Resource-based theory and Dunning’s eclectic paradigm have emphasized the importance of intangible assets, including competent managerial capabilities and knowledge in family firms, in the event of internationalization. Also as implied by the arguments from Fama and Jensen (1983), ownership should be separated from the management so that family business owners are released from risk bearing burden.

17_{They also argue that with higher family ownership (above 76.13%), the family would try to peruse}

(26)

25 The change of the sign for professional CEO in Model 2 should also be noticed18_.

The significant negative coefficient may suggest a potential negative effect of professional CEO, indicating the need of a stronger monitory mechanism. Moreover, the proportion of INDs in the board shows a positive sign, suggesting the necessity of performing Model 3. However, the interaction of IR and professional CEO gives a negative coefficient, while both IR and professional CEO have positive signs. But without statistical significance, the interaction suggests that the roles played by INDs are not as much efficient as expected. The possible explanation is concerned with whether the assumption of high independence of INDs is true. In China, reported by Ji and Wang (2005), 63% INDs are selected by the board itself and more than 36% INDs are directly recommended by the largest shareholder, the family business owner. Gabrielsson and Huse (2005) also state that INDs are often connected to the dominant family. Therefore, the governance system is not as effective as expected. Then the reason for family firms adopting more INDs is due to the second role of the board19, offering resources. Especially the extensive personal networks managed by INDs (Voordeckers et al., 2007) enable family firms to gain access to crucial resources and information. As reported by People’s Daily Online20, many Chinese political officials are employed by family firms as INDs because of their abundant social networks. Thus currently INDs lack independence to perform monitoring role but are attracted to family firms because of the resource providing role.

Furthermore, the empirical results of control variables, firm age and R&D intensity turn out to be against the predictions. Carr et al. (2010) demonstrates that younger firms generate more sales growth, since well-established firms lack the flexibility to adapt to new environments. This argument becomes more valid in Chinese context, since founders share the same feature of inflexibility. As to trademarks, Child and Rodrigues

18 _{According to Brunninge et al. (2007), “in the presence of interactions, the coefficients for independent} terms making up the interactions convey no meaningful, but possibly misleading information”, thus the coefficients of professional CEO and family ownership should just be a reference without too much emphasis.

19_{Recalling to Section 2.4.2.}

(27)

26

(2005) state that Chinese family firms are seeking recognized brand assets in foreign markets to create a competitive position. Thus those without trademarks are more aggressive in entering international markets.

5. Additional studies about the adoption of Professional CEO

Table 5 is constructed according to the pattern of CEO succession during 2008 to 2012. The sample of 143 family firms are divided into four types. Among them, 102 family firms already have professional CEOs in 2008. 69 of these firms keep transferring the position to professionals (P-P), while 33 samples replaced the professionals with family members (P-F). In comparison, only 41 firms have family CEOs at the beginning. Among them, 14 firms keep passing CEO positions to family members (F-F), and 27 firms choose to hire professionals instead (F-P). In the end, 96 family firms have professional CEOs. Splitting up those four types of family firms, the changes of foreign sales percentage during these years are presented. Firms that stick to the family CEOs have the lowest level of internationalization.

Table 5: Average foreign sales percentage for four types of Chinese family firms This table showsthe changes of foreign sales percentage respectively for four types of changing CEOs regarding 143 samples during 2008 to 2012: professional CEO to professional CEO, professional CEO to family CEO, family CEO to family CEO and family CEO to professional CEO.

CEO change type N Foreign sales

percentage 2008 2009 2010 2011 2012 Average

P-P 69 48.25% 25.71% 22.31% 22.90% 21.73% 22.97% 23.13%

P-F 33 23.08% 31.75% 25.02% 24.57% 22.60% 23.16% 25.42%

F-F 14 9.79% 18.18% 14.21% 16.22% 15.89% 17.01% 16.43%

F-P 27 18.88% 29.21% 25.54% 24.16% 24.08% 23.61% 25.32%

(28)

27 do the comparison (Chart 1). For P-F group, five firms show a leap of foreign sales after returning the CEO position to family members, three of them doubled in the percentage and the other two showing a tenfold increase. For F-P group, the count for firms of doubled foreign sales growth is the same and one of them increases by nearly ten times. However, foreign sales in P-F group dropped in a larger scale than F-P group. In P-F group, 10 firms dropped the foreign sales by greater than half, two of which have no foreign sales afterwards and the reduction of 19 firms are more than 20%; while 7 firms in F-P group have decreased their foreign sales more than half and only 9 firms show a more than 20% reduction.

Overall, even though the whole sample have shown a general decline of foreign sales during 2008 to 2010 according to Table 5, Chart 1 illustrates that changing CEOs from family members to professionals indeed helps to moderate the decline based on the comparison with family firms that change CEOs in the other way around, supporting the expectation of a positive effect of professional CEOs on internationalization.

Chart 1: The comparison between F-P and P-F regarding foreign sales change before and after CEO succession

Two growth rates of 1,437% and 3,356% in P-F group, and one in F-P group, 869% are not included in the chart, since these three rates are too large that could obscure other figures.

(29)

28

Table 6：Foreign sales percentage for F-P Chinese family firms

This table shows the results of Panel Least Squares with cross-section fixed effects model about foreign sales percentage for F-P subsample. Model 1 is to test hypothesis 1 and 2, Model 2 and model 3 are designed for hypothesis 3 and 4 respectively. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark and Second Shareholder are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares.

Model 1 Model 2 Model 3

Hypothesis 1 &2 Hypothesis 3 Hypothesis 4

Constant 0.269 0.265 0.325 Independent Variables Family Ownership -0.415* -0.413* -0.412* Professional CEO -0.013 -0.011 -0.109 FR 0.350 0.349 0.358* IR 0.082 0.082 -0.072 Interaction

Professional CEO*Family Ownership -0.004

IR*Professional CEO 0.266 Control Variables Firm Size 0.078 0.079 0.075 Firm Age -0.019* -0.019* -0.019* Past Performance 0.082 0.082 0.074 Patent -0.164 -0.165 -0.160 Trademark 0.772 0.776 0.770 Second shareholder -0.162*** -0.162*** -0.163*** R-squared 0.924 0.924 0.925 Adjusted R-squared 0.897 0.895 0.896 Observations 135 135 135 * significant at 10% level ** significant at 5% level *** significant at 1% level

(30)

29 is also checked with adding professional CEO lag variable. The result shows no significant p value. Descriptive statistics and correlation matrix can be found in Appendix 1 and 2. Multicollinearity is as well not a problem among these variables.

Table 6 represents a different result for F-P group compared with the whole sample. In Model 2, the interaction between family ownership and professional CEO gives no significant coefficient. The only thing worth attention is the ratio of family members on the board, which is positively related to the internationalization, distinct from the results for the whole sample. Especially in Model 3, this variable shows significance at 10% level, and the sign for the ratio of INDs changes from positive to negative. These results indicate that at the early stage (less than five years21) after family firms changing their CEOs from family members to professionals, INDs and professionals are not helping to increase the degree of internationalization, but rather family members still play significant roles in the decision making process. Since professionals may not be familiar with family business at the start, the strategic decisions can be delayed, as implied by Hall and Nordqvist (2008) who contend that a less understanding of family business culture leads to the invalidation of professionals’ expertise. Family control help professionals to understand family business situations and make strategic decisions timely, giving rise to more internationalization. Also, agency costs may be produced from this unfamiliarity. The more involvement of family members, the more alignment may be achieved between owners and professional CEOs. Moreover, this result can be explained by the family business owners focusing on monitoring and control after delegating decision management to professionals, which is suggested by Fama and Jensen (1983) as the division of decision control from decision management.

6. Conclusion

The impact of family control on firm value has been analyzed widely. But not many empirical studies about firm internationalization have been done in terms of

21_{Almost half of the family firms in this study have adopted professional CEOs before 2008. Among}

(31)

30 family control and the different situations in emerging markets are much less discussed previously. The link between professional CEOs and firm internationalization is further limited in Chinese context due to the late awareness and the slow development of local market for professional CEOs. Thus this paper responds to the call for understanding the current situations of Chinese family firms, combining family business succession issue and internationalization.

This study argues that the separation of ownership and management is essential for the future international growth of Chinese family firms. Thus in order to improve the level of internationalization, family firms should hire professional CEOs who have superior managerial capabilities and who can make strategic decisions without the influence of a substantial wealth. Even though the positive effect is not obvious and family control on the board is still important at the very beginning, according to the empirical results, having professional CEOs is believed to improve foreign sales in the long run. In addition, IND is brought in to deal with agency costs in this paper. But the results suggest that IND mechanism in China is not constructed in perfection, showing a less effective monitoring role, while with a wide pool of social connections, INDs are still demanded in family firms that are committed to global expansion.

(32)

31

References

Brunninge, O., Nordqvist, M., & Wiklund, J. (2007). Corporate governance and strategic change in SMEs: The effects of ownership, board composition and top management teams. Small Business Economics, 29(3), 295-308.

Carr, J. C., Haggard, K. S., Hmieleski, K. M., & Zahra, S. A. (2010). A study of the moderating effects of firm age at internationalization on firm survival and short‐term growth. Strategic Entrepreneurship Journal, 4(2), 183-192.

Cerrato, D., & Piva, M. (2012). The internationalization of small and medium-sized enterprises: the effect of family management, human capital and foreign ownership. Journal of Management & Governance, 16(4), 617-644.

Chen, C. J., & Jaggi, B. (2001). Association between independent non-executive directors, family control and financial disclosures in Hong Kong. Journal of Accounting

and Public Policy, 19(4), 285-310.

Child, J., & Rodrigues, S. B. (2005). The Internationalization of Chinese Firms: A Case for Theoretical Extension?[1]. Management and organization review,1(3), 381-410.

Claessens, S., Djankov, S., & Lang, L. H. (2000). The separation of ownership and control in East Asian corporations. Journal of financial Economics, 58(1), 81-112.

Claessens, S., Djankov, S., Fan, J. P., & Lang, L. H. (2002). Disentangling the incentive and entrenchment effects of large shareholdings. The Journal of Finance, 57(6), 2741-2771.

Claver, E., Rienda, L., & Quer, D. (2008). Family firms' risk perception: empirical evidence on the internationalization process. Journal of Small Business and Enterprise

(33)

32

Daily, C. M., & Dalton, D. R. (1992). Financial performance of founder-managed versus professionally managed small corporations. Journal of small business

management, 30(2), 25-34.

Dunning, J. H. (1988). The eclectic paradigm of international production: a restatement and some possible extensions. Journal of international business studies, 1-31.

Dunning, J. H. (2002). Relational assets, networks and international business activity. Cooperative strategies and alliances, 569-593.

Dyer, W. G. (1989). Integrating professional management into a family owned business.

Family Business Review, 2(3), 221-235.

Erdener, C., & Shapiro, D. M. (2005). The internationalization of Chinese family enterprises and Dunning's eclectic MNE paradigm. Management and Organization

Review, 1(3), 411-436.

Faccio, M., & Lang, L. H. (2002). The ultimate ownership of Western European corporations. Journal of financial economics, 65(3), 365-395.

Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. Journal of

law and economics, 301-325.

Fan, J. P., Wei, K. C., & Xu, X. (2011). Corporate finance and governance in emerging markets: A selective review and an agenda for future research. Journal of Corporate Finance, 17(2), 207-214.

Fernández, Z., & Nieto, M. J. (2005). Internationalization Strategy of Small and Medium‐Sized Family Businesses: Some Influential Factors. Family Business Review,

(34)

33

Fiegener, M. K., Brown, B. M., Dreux, D. R., & Dennis Jr, W. J. (2000). The adoption of outside boards by small private US firms. Entrepreneurship & Regional

Development, 12(4), 291-309.

Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity (pp. 61-7). New York: Free Press.

Gabrielsson, J., & Huse, M. (2005). Outside directors in SME boards: a call for theoretical reflections. Corporate Board: role, duties & composition, 1(1), 28-37.

George, G., Wiklund, J., & Zahra, S. A. (2005). Ownership and the internationalization of small firms. Journal of Management, 31(2), 210-233.

Hall, A., & Nordqvist, M. (2008). Professional management in family businesses: Toward an extended understanding. Family Business Review,21(1), 51-69.

Hitt, M. A., Hoskisson, R. E., & Kim, H. (1997). International diversification: Effects on innovation and firm performance in product-diversified firms.Academy of

Management journal, 40(4), 767-798.

Jensen, M. C., & Meckling, W. H. (1979). Theory of the firm: Managerial behavior,

agency costs, and ownership structure (pp. 163-231). Springer Netherlands.

Ji, C. H., Wang, W. (2005). Research on Independent Director Institution in China.

Journal of Central University of Finance & Economics, 7, 013.

(35)

34

Jones, C. D., Makri, M., & Gomez‐Mejia, L. R. (2008). Affiliate directors and perceived risk bearing in publicly traded, family‐controlled firms: The case of diversification. Entrepreneurship Theory and Practice, 32(6), 1007-1026.

Laeven, L., & Levine, R. (2004). Beyond the biggest: Do other large shareholders influence corporate valuations. Unpublished working paper. University of Minnesota.

Liang, X., Wang, L., & Cui, Z. (2014). Chinese Private Firms and Internationalization Effects of Family Involvement in Management and Family Ownership. Family

Business Review, 27(2), 126-141.

Lin, S. H., & Hu, S. Y. (2007). A family member or professional management? The choice of a CEO and its impact on performance. Corporate Governance: An

International Review, 15(6), 1348-1362.

Lin, W. T. (2012). Family ownership and internationalization processes: Internationalization pace, internationalization scope, and internationalization rhythm.

European Management Journal, 30(1), 47-56.

Liu, Y., Ahlstrom, D., & Yeh, K. S. (2006). The separation of ownership and management in Taiwan's public companies: An empirical study. International Business

Review, 15(4), 415-435.

McCollom, M. E. (1988). Integration in the family firm: When the family system replaces controls and culture. Family Business Review, 1(4), 399-417.

(36)

35

North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge university press.

Peng, M. W., & Jiang, Y. (2010). Institutions behind family ownership and control in large firms. Journal of Management Studies, 47(2), 253-273.

Porta, R., Lopez‐de‐Silanes, F., & Shleifer, A. (1999). Corporate ownership around the world. The journal of finance, 54(2), 471-517.

Qilin, S., & Wen, Z. (2003). Family Control and Firm Value: Evidence from China Listed Companies [J]. Economic Research Journal, 8, 004.

Sharma, A. (1997). Professional as agent: Knowledge asymmetry in agency exchange.

Academy of Management Review, 22(3), 758-798.

Sherman, H. D., Kashlak, R. J., & Joshi, M. P. (1998). The effect of the board and executive committee characteristics on the degree of internationalization.Journal of

International Management, 4(4), 311-335.

Stewart, A., & Hitt, M. A. (2012). Why can’t a family business be more like a nonfamily business? Modes of professionalization in family firms. Family Business Review, 25(1), 58-86.

Tekušová, T., & Kohlhammer, J. (2008, January). Visual analysis and exploration of complex corporate shareholder networks. In Electronic Imaging 2008 (pp. 68090F-68090F). International Society for Optics and Photonics.

(37)

36

Voordeckers, W., Van Gils, A., & Van den Heuvel, J. (2007). Board Composition in Small and Medium‐Sized Family Firms*. Journal of Small Business

Management, 45(1), 137-156.

Ward, J. L. (1997). Growing the family business: Special challenges and best practices. Family Business Review, 10(4), 323-337.

Whisler, T. L. (1988). The role of the board in the threshold firm. Family Business

Review, 1(3), 309-321.

Yeung, H. W. (2006). Change and continuity in Southeast Asian ethnic Chinese business. Asia Pacific Journal of Management, 23(3), 229-254.

Zahra, S. A., & Pearce, J. A. (1989). Boards of directors and corporate financial performance: A review and integrative model. Journal of management, 15(2), 291-334.

Zahra, S. A. (2003). International expansion of US manufacturing family businesses: The effect of ownership and involvement. Journal of Business Venturing, 18(4), 495-512.

Zahra, S. A. (2005). Entrepreneurial risk taking in family firms. Family Business

Review, 18(1), 23-40.

Zhang, J., & Ma, H. (2009). Adoption of professional management in Chinese family business: A multilevel analysis of impetuses and impediments. Asia Pacific Journal of

Management, 26(1), 119-139.

(38)

37

Appendix

Appendix 1: Descriptive statistics

This table shows five descriptive statistics, namely mean, median, maximum, minimum and standard deviation of all the variables for F-P subsample. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark and Second Shareholder are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares.

Variables N Mean Median Maximum Minimum Std. Dev.

Foreign sales percentage 715 0.253 0.121 0.983 0.000 0.277

(39)

38

Appendix 2: Correlation matrix

This table shows correlation matrix of all the variables for F-P subsample. Professional CEO, Family Ownership, FR (the percentage of family members on the board) and IR (the percentage of INDs on the board) are main variables. Professional CEO is a dummy variable which equals 1 if the CEO is from outside the family. Firm Size, Firm Age, Past Performance, Patent, Trademark and Second Shareholder are control variables. Second Shareholder is a dummy variable if the firm has a shareholder with more than 10% shares.

01 02 03 04 05 06 07 08 09 10 11

01.Foreign sales percentage 1.000

Professional CEO and Internationalization: An empirical study for Chinese family firms

Professional CEO and Internationalization:

An empirical study for Chinese family firms

Contents