Internationalization often involves the transfer of a firm’s knowledge from one country to another and can be conceived of as a process of learning and accumulation of experiential knowledge (Eriksson, Johanson, Majkgård, & Sharma, 2000). Those that learn efficiently from their internationalization experience are able to expand faster and with fewer mistakes (Tsang, 1999). In fact, the well-known Uppsala model of internationalization is based on the argument that firms learn from their overseas expansion experience (Johanson & Vahlne, 1977), and organizational learning is a major theoretical lens adopted in the study of foreign direct investment (FDI) over the past two decades (e.g., Barkema & Vermeulen, 1998; Li, Li, & Shapiro, 2012; Martin & Salomon, 2003; Tsang, 2002a; Tsang & Yamanoi, 2016; Zahra, Ireland, & Hitt, 2000). Internationalization often poses special learning challenges for family firms because of the unique characteristics of these firms (Tsang, 2002b), making it an intellectually stimulating topic of study. Generally speaking, applying a learning lens could bring theoretical insights to the study of family business, as forcefully maintained by Hamilton’s (2011) recent qualitative study based on a small sample of family firms located in England.

These three streams of research—family firms, internationalization and organizational learning—have been growing steadily over the past two to three decades. While there are studies addressing two of the three streams, studies that combine all three are rare. Pukall and Calabrò’s (2014) review of 72 journal articles published between 1980 and 2012 on the internationalization of family firms indicates that only two—presumably Basly (2007) and Tsang (2002b)—adopt a theoretical framework of organizational learning or the knowledge-based view. Yet, such studies could have some unique and significant theoretical and managerial implications. For example, Tsao and Lien (2013: 193) argued that “because family owners tend to have a longer investment horizon, they are able to travel farther along the firm’s learning curve. In other words, they acquire skills and capabilities necessary for greater information-processing capacity, which allow them to solve complex problems of internationalization.” Their argument is surely interesting, but they did not follow up with a proper empirical study. Their analysis of a sample of publicly listed Taiwanese firms during the period from 2000 to 2009 simply fails to provide any direct empirical evidence either for or against their argument.

To summarize, organizational learning is a key perspective used by researchers to study the internationalization of firms because internationalization is fundamentally a learning process for both newcomers and experienced players. However, in the specific case of family firm internationalization, learning is rarely adopted as a theoretical lens. This is unfortunate because a more comprehensive understanding of the phenomenon is hampered as a result. To address this deficiency in the literature, my paper discusses the internationalization of family firms, based on Huber’s (1991) typology of organizational learning processes. It shows the theoretical richness of a learning perspective and contributes to the family business and internationalization literatures by comparing the learning processes between traditional family firms, professional family firms and non-family firms with respect to internationalization. It enhances our understanding of the learning-related issues when family firms internationalize and it contributes to a more comprehensive view of the phenomenon. In particular, theoretically-significant propositions can be derived from the differences among the three types of firms to guide empirical research.

Nature of family firms

By family firm, I mean firms that are effectively controlled through ownership and managed by a single family. Effective control refers to “a controlling owner’s ability to add, direct, or dispose of a firm’s assets without recourse to a third party, such as an institutional investor, a bank, or a business partner” (Carney, 2005: 254). Depending on the institutional context and stock distribution, an absolute majority of stock ownership may not be required to achieve effective control. Such family firms should be distinguished from those in which the family influences the operation without having effective control (Chua, Chrisman, & Sharma, 1999).

Family firms constitute a heterogeneous group with variations along multiple dimensions (Chrisman, Chua, & Sharma, 2005). For example, based on a cross-sectional survey of family firms in the UK, Westhead and Howorth (2007) identified seven empirical types of family firms. For the sake of this discussion, I divide family firms into two main types—traditional and professional—along the dimension of professionalization, which refers to “hiring full-time, non-family employees, particularly with the delegation of managerial authority” (Stewart & Hitt, 2012: 59). Professionalization entails other aspects of management, such as the extent of formalization, merit-based appraisal system and promotion, organized training, and boards with the presence of independent directors (Chua, Chrisman, & Bergiel, 2009; Tsui-Auch, 2004).

Traditional family firms refer to those that have a low degree of professionalization. A typical case would be a small or medium-sized firm whose top management team (TMT) consists mostly of family members. Strategic decision-making authority is highly centralized within the owning family. Promotion is based on not only performance, but also the employee’s relationship with the family. Operational rules and regulations, if any, may not be uniformly applied to all the employees.

In contrast, professional family firms are usually, but not always, larger in size, with a substantial number of TMT members being non-family members. Aside from the fact that the firm is effectively controlled and actively run by a single family, it is similar to a professionally-managed non-family firm in virtually all aspects. An excellent example is the business conglomerate of Li Ka-shing, a noted Hong Kong entrepreneur. Although he and his two sons make the critical strategic decisions, such as the recent decision to gradually reduce investment commitment in mainland China (Ma, 2015), they have to rely on a huge team of professional managers to run the conglomerate’s operations in dozens of countries.

Note that the classification of traditional versus professional family firms is not binary; rather, the “traditional” and “professional” labels form the two ends of a continuum. For instance, Tsui-Auch (2003) conducted a comparative case study of two small and medium-sized Chinese family firms in Singapore. The smaller one was a typical traditional family firm. Although the other one adopted a professional management system and was considerably closer to the “professional” end of the continuum, it was by no means comparable to the extent of professionalization found in, say, Li Ka-shing’s conglomerate.

It is likely that a traditional family firm will evolve to become a professional one as it grows in size. A large organization often has a more formalized structure and more rules and regulations governing its operation. The size limitation of the owning family necessitates the hiring of more outsiders into the TMT as the business expands. In particular, when a family firm operates in a considerable number of foreign countries, it is not geographically feasible for a few family members to be in charge of all such operations, and outsiders have to be heavily involved in running the operations. In this respect, the traditional-professional dichotomy is particularly relevant to the study of family firms’ internationalization.

Organizational learning processes

Huber’s (1991) seminal work provides a rather comprehensive typology categorizing the major processes of organizational learning. Although the framework is over a quarter of a century old, to date it is still heavily cited by researchers. To my knowledge, there has not been a better framework put forth for classifying learning processes. In this section, I adopt his typology as a guiding framework and discuss the characteristics of family firms with respect to these processes in the context of internationalization. The discussion is based on prior studies of the various aspects of family firms’ internationalization—including my own studies of Chinese family firms (Tsang, 2001, 2002b)—and other studies of family firms or internationalization in general. Admittedly, there is conjecture that may not be empirically well-grounded and would need to be confirmed by future empirical tests.

Although the percentage of a firm’s sales that are generated from export revenues is often used to measure the degree of internationalization in empirical studies (e.g., Arregle, Naldi, Nordqvist, & Hitt, 2012; Fernhaber, Gilbert, & McDougall, 2008; Zahra, 2003), the following discussion will focus on FDI activities, which involve higher levels of commitment, risk and complexity than exporting (Anderson & Gatignon, 1986). Such activities are more practically challenging for the focal firm and also more intellectually interesting. For example, the need to monitor a foreign operation can be a taxing demand on the oft-constrained managerial capacity of a family firm (Arregle, Duran, Hitt, & van Essen, 2017).

Huber’s (1991) typology consists of five subprocesses of knowledge acquisition, four subprocesses of information interpretation and two subprocesses of organizational memory. In this paper I only include the five subprocesses of knowledge acquisition—experiential learning, vicarious learning, congenital learning, searching and grafting—because they have so far attracted more research attention than the subprocesses of information interpretation or organizational memory. For example, in reviewing the international entrepreneurship literature with a focus on the issues of learning and knowledge management, De Clercq, Sapienza, Yavuz, and Zhou (2012) only used the five knowledge acquisition subprocesses to organize their review. Table 1 summarizes the discussion of the learning processes.

Table 1 Learning characteristics of traditional family, professional family, and non-family firms in the context of internationalization
Full size table

Knowledge acquisition

Huber (1991: 88) used the word “voluminous” to describe the literature on knowledge acquisition, suggesting that this is the learning process that has drawn the highest level of research attention. This is somewhat expected because a normal process for an organization to learn starts with obtaining knowledge from whatever sources that are available to it.

Experiential learning

Among the five subprocesses, experiential learning is probably the most researched one. After their birth, organizations acquire some of their knowledge through their own direct experience, mostly in an unintentional or unsystematic manner (Huber, 1991). Using the concept of experiential learning, Johanson and Vahlne (1977) stressed that market-specific knowledge can only be gradually obtained through experience in foreign markets, and the process of internationalization entails incremental commitments of resources. The Uppsala model has prompted a large number of empirical studies testing its experiential learning argument, with mixed support for the model (Petersen & Pedersen, 1997).

The experience of setting up and running an overseas venture can be roughly divided into two types, namely, operational and strategic. The former experience is gained from day-to-day operation, while the latter from general management and making strategic decisions. Experience gained at the individual level largely determines the extent of experiential learning at the organizational level. My case study of six family and four non-family firms in Singapore indicates that there is no significant difference between the two kinds of firms in terms of operational experiential learning (Tsang, 2002b). This result is not surprising because, for family firms, the routine duty of managing the various functional aspects of an overseas venture often falls on the shoulders of non-family members. Therefore, whether the firm is family owned and managed is not a major factor affecting the outcome of learning.

In contrast, since members of the owning family of a traditional family firm usually occupy top or senior managerial positions, strategic experience is mostly concentrated in the family. Moreover, in the case of Chinese family firms, where authority and control are highly centralized, making critical strategic decisions is normally a family affair that takes place behind closed doors with little participation from senior managers who do not belong to the family. In short, strategic experience is usually not well shared among TMT members of a traditional family firm.

The case of professional family firms is somewhat different. More non-family members are likely to be present in the TMT, and their participation in the decision-making process is a normal routine. As such, strategic experience is also shared among non-family members. A caveat is that, for decisions of paramount importance, such as the above-mentioned decision by Li Ka-shing to reduce his investment commitment in mainland China, the owning family may not be willing to openly discuss the rationale underlying such decisions with non-family members. There may be private, socioemotional-related family motives that these decisions are intended to fulfill. Regardless of traditional or professional family firms, once a key strategic decision—such as whether to set up a new or to close down an existing overseas venture—has been made, non-family members have to be involved in implementing the decision, and through the process, gain operational experience.

Vicarious learning

No organizations can rely solely on the knowledge generated from their own experience. Learning from the experience of others is essential, especially when venturing into a new domain such as investing in a foreign country. Using a sample of 150 US publicly-held new ventures, Fernhaber and Li (2010) found that international market entries by these new ventures depended in part on the degree of internationalization of other firms in the venture’s home country industry. They explained this finding by arguing that these ventures imitated their more experienced peers. Since imitation is considered one form of vicarious learning (Cyert & March, 1963), their study provides indirect evidence of vicarious learning. The depth of vicarious learning depends a great deal on the extent to which the focal firm can access the operation of the learning target. Most of the new ventures in Fernhaber and Li’s (2010) study were not likely to learn much from their industry peers, most of whom were also their competitors.

The extended family connections of a family firm may facilitate vicarious learning. One of the traditional family firms in Singapore (referred to as Firm X hereafter) that I studied leveraged such connections (Tsang, 2001).Footnote 1 The founder of the company was born into a large Taiwanese family. Its core business was electronics. However, the company ventured into a new industry in a foreign country by setting up a plant in China manufacturing construction materials. The knowledge to operate the plant was acquired through the founder’s extended family connections: His father had a similar business in Taiwan. Although it is possible that normal social network ties may also provide similar access to such knowledge, a family tie is often superior to a social network tie. As an antecedent of knowledge transfer, trust between two parties evolves as the transfer takes place between them (Inkpen & Currall, 2004). It can take a considerable amount of time for trust to reach the level at which one party is willing to share critical and sensitive knowledge with the other. On the other hand, the trust entailed by a family tie—such as the one between the founder and his father in the aforementioned company—is often strong and preexists. Moreover, family ties are exclusive: One can have many friends but only one father, thus alleviating the worry that one’s competitors may be able to get access to the same knowledge source.

The above example indicates family firms’ ability to use strong personal ties to fill “institutional voids”—the relative lack of market intermediaries, regulatory systems and contract enforcing mechanisms—of emerging economies (Khanna & Palepu, 1997). Although family firms operate as a single entity, they are unique in that two forms of social capital coexist: the family’s and the whole organization’s (Arregle, Hitt, Sirmon, & Very, 2007), with the objective of building a cohesive internal “community.” Extending this community spirit, family firms tend to establish deeper, more personalized relationships with outside stakeholders than do non-family firms, as indicated by the findings of Miller, Lee, Chang, and Le Breton-Miller’s (2009) survey of Korean high-technology businesses. These relationships enhance family firms’ vicarious learning opportunities. However, as the extent of professionalization of a family firm increases, managers who are non-family members are more motivated by short-term economic rationales and less concerned with long-term, and often quite personal, partnerships when dealing with such outside stakeholders as suppliers and customers (Miller et al., 2009). Thus, while extended family members are a reliable source of vicarious learning for both traditional and professional family firms, the latter are in a less favorable position to tap the source of outside stakeholders. Lastly, without the help of any family or extended family ties, non-family firms learn mostly from stakeholders that they interact with, such as suppliers, competitors and strategic alliance partners. The extent of learning depends a great deal on the closeness of their relationships with these parties.

Congenital learning

Every organization is endowed with the knowledge and experience of the founder and other members of the founding team at birth. Such knowledge and experience form the foundation upon which a new organization makes sense of its environment, perceives opportunities and plans its future actions. Oviatt and McDougall (2005) argued that firms in which members of the founding team have lived abroad or have prior overseas work experience exhibit speedier foreign market entries and greater commitment to internationalization. Similarly, in building a dynamic capability model of born global firms, Weerawardena, Mort, Liesch, and Knight (2007) proposed that owner-managers’ prior international experience indirectly contributes to the speed of first international entry and the extent and scope of internationalization. In another model-building attempt—this time, a model of the influence of knowledge on internationalization—Casillas, Moreno, Acedo, Gallego, and Ramos (2009) proposed that the education and prior experience of the founder and TMT members have an impact on the recognition of internationalization as an opportunity for improving firm performance. Among the few empirical studies in this domain is Bruneel, Yli-Renko, and Clarysse’s (2010) analysis of 114 young technology-based firms in Belgium. Their results indicate that the lower a firm’s experiential learning from international operations, the more significant the effects of the founding team’s prior international knowledge base (i.e., congenital learning) on the extent of internationalization, although they do not find a direct effect of congenital learning on internationalization.

Compared with non-family firms, family firms are in a disadvantageous position as far as congenital learning is concerned. A typical family firm is established by a few members of one family, who are usually couples and/or siblings, with the assistance of non-family members. Since the founding family dominates the strategic decision-making that determines the firm’s path of growth, non-family members’ prior experience contributes little in this respect. Congenital learning is therefore constrained by the somewhat homogeneous backgrounds of the members of the founding family. Moreover, the younger members of the family often grow up in the business run by their parents, contributing to another form of congenital learning. In contrast, a non-family firm is usually established by several entrepreneurs who do not have blood relationships. More often than not, none of these founding members dominates the management of the company. Their backgrounds are in general more heterogeneous than those of the members of a founding family. As such, compared to family firms, there is a higher chance that at least some founding members of a non-family firm have backgrounds, such as overseas education and work experience, that would enable them to spearhead internationalization initiatives.

A similar argument is also applicable to the distinction between traditional and professional family firms. When a traditional family firm sets up a new operation in a foreign country, it tends to send one or more family members to head the operation, at least during the initial stage. The owning family is thus the main source of congenital learning for the operation. In contrast, a professional family firm is likely to dispatch non-family members, who may be existing managers or new recruits, to run the operation, resulting in a wider scope of congenital learning. In brief, professional family firms are better positioned for congenital learning than traditional ones. By the same token, since the founding team members of non-family firms normally do not have any family relationships and are likely to have a variety of expertise in order to deal with the various challenges faced by start-ups, these companies generally enjoy higher-quality congenital learning than either traditional or professional family firms.

Searching

Searching can refer to wide-ranging scanning of the organization’s external environment or to more focused examination of a particular segment of the environment in response to certain threats or opportunities (Huber, 1991). When a firm considers expanding overseas, a natural first step is to scan the environment, especially foreign countries, and look for suitable investment deals. In spite of this direct link between searching and internationalization, few empirical studies directly measure the construct and investigate its effects on internationalization. A more common approach is to treat it as a component of a broader construct. For instance, Zhou (2007) included survey items related to searching when measuring proactiveness, which is a dimension of international entrepreneurial proclivity, and Armario, Ruiz, and Armario (2008) did the same when measuring intelligence generation, which is a dimension of market orientation.

Empirical studies of the searching behavior of family firms during the process of internationalization are even rarer. My study sheds some light on this matter (Tsang, 2002b). The founder of Firm X hired a personal assistant, who was a non-family member, to gather information about potential FDI projects. He also personally collected information from his friends and relatives, and, through this informal social network, he kept abreast of current market dynamics, policy changes and potential investment opportunities. A key distinction between family and non-family firms in my study is the degree of formalization, which refers to the extent to which the information collected is recorded and presented in a standardized and organized way. As expected, traditional family firms have a much lower degree of formalization. The haphazard method of information collection is well illustrated by the following quote from the founder of another Chinese family firm (referred to as Firm Y hereafter) in Singapore:

I have been in this business for twenty years. I know all the guys in the industry—who can do a good job and who cannot deliver. I play golf with executives from multinational corporations and our local friends. It’s easy for me to get market information, say, which firm is going bankrupt, which is axing the top guy, etc. They trust me and give me hints. I can’t expect any employees or even a manager to be able to tap such information. (Tsui-Auch, 2003: 211)

Firm Y was the more professionally managed firm in Tsui-Auch’s (2003) study mentioned earlier. Firms X and Y—one traditional and the other more professional—are highly consistent in that both founders were heavily involved in the searching process and that both did it in an unsystematic manner.

The reliance of family firms on family members’ social networks restricts the scope of search. As mentioned, members of the same family tend to have somewhat homogeneous backgrounds and thus overlapping social networks. Another issue facing traditional family firms is that, even if non-family members obtain some relevant or useful information through their personal connections, they may not bother to pass the information to the owning family because, as someone outside the family, they may doubt whether their information will be taken seriously. Professional family firms fare better in this respect. Since non-family members actively participate in day-to-day management and their views are well respected by the owning family, they tend to be more willing to share their search results. Moreover, professional family firms adopt a more formalized style of searching, in line with their more systematic approach to management. Lastly, keeping other things constant, non-family firms tend to have more systematic and formalized information collection than either traditional or professional family firms. Moreover, their scope of search is not constrained by a manager’s family background and is tailored to the purpose at hand.

Grafting

A quick method for an organization to enhance its knowledge base is to acquire new members who possess knowledge not currently available within the organization. For example, grafting is often the major motive underlying acquisitions in high-tech industries. In these industries, “the pace and magnitude of technological change, as well as the breadth and depth of knowledge-based resources required to compete, may not allow firms to internally develop all the technologies and capabilities they need to stay competitive” (Ranft & Lord, 2002: 420). When a firm considers expanding its operations overseas for the first time or to a new foreign country, it is not likely to possess the necessary knowledge. Thus, grafting can be an efficient solution. Unfortunately, De Clercq et al.’s (2012) review of 48 journal articles that study internationalization in the field of international entrepreneurship does not find any studies of grafting.

Compared with non-family firms, a key consideration for family firms when hiring managers from outside is concerned with the issue of trust, especially if the manager will be assigned to work in a foreign operation beyond the immediate supervision of the owning family at the headquarters. Family firms are usually regarded as “high trust” organizations (Jones, 1983). This view is not surprising since the level of trust existing among members of a family is normally higher than that among managers who do not share any blood relationships. Trust seldom extends beyond the family. A family may therefore hesitate to hire a senior manager who is not a family member to make up for its deficient FDI knowledge and would instead prefer to hire someone it can trust. An illustrative example is Firm X, which operated a hotel in China, whereas its core business was electronics. The two industries are distantly related. As a remedy, the company hired the former general manager of a five-star hotel in Taipei that was owned by the founder’s extended family in Taiwan and placed him in charge of the hotel (Tsang, 2001). Since that manager was previously hired by the founder’s extended family, he was perceived to be trustworthy. In short, trust is a barrier that family firms have to overcome when acquiring knowledge through grafting as far as senior managerial positions are concerned.

Trust in strangers varies across cultures. Both Tsui-Auch’s (2003, 2004) studies and my own were based on Chinese family firms. When commenting on the issue of trust in Chinese society, Redding (1990: 66) noted:

The key feature would appear to be that you trust your family absolutely, your friends and acquaintances to the degree that mutual dependence has been established and face invested in them. With everybody else you make no assumptions about their goodwill.

Within the Asia Pacific region, it seems that Japanese society has a higher level of trust in strangers, as indicated by Chang and Shim’s (2015) study of the transition from family to professional management among Japanese family firms. There are high-profile anecdotes of such transitions, such as that of Panasonic (formerly known as Matsushita Electric, founded by Konosuke Matsushita). Similar cases of professional managers taking the helm of large Chinese family firms are relatively rare, as evidenced by Tsui-Auch’s (2004) study of such firms in Singapore.

Regardless of culture, professional family firms have more non-family members occupying senior positions, exhibiting a higher level of trust than traditional family firms. As such, professional family firms are better positioned to use grafting as a means of dealing with knowledge deficiency. For instance, when planning to enter a foreign country for the first time, they may not hesitate to hire someone who has had managerial experience in that country to lead the project or even hire a local manager for that purpose. In contrast, traditional family firms adopt a more conservative approach and may treat grafting as a last resort. It goes without saying that even for professional family firms, preference will be given to the job candidate who has connections with the owning family when two candidates are similarly qualified. Finally, non-family firms practice merit-based hiring, considering only a job candidate’s contribution to the firm’s pool of knowledge and experience.

Information distribution

How information is distributed in an organization is crucial to its learning, as “organizations often do not know what they know” (Huber, 1991: 100). Sometimes organizations find themselves reinventing the wheel because they fail to retrieve the knowledge that they already possess. One edge of family firms over their non-family counterparts is that knowledge is generally better shared among family members than among TMT members of non-family firms:

Over the many years of shared experiences between relatives special words, phrases, expressions, and body movements evolve that have agreed upon meanings. Private languages, “family languages,” allow family members to communicate more efficiently than is generally possible among nonrelatives, even among close friends. This can permit relatives to exchange more information with greater privacy and arrive at decisions more rapidly than can two nonrelatives. (Tagiuri & Davis, 1996: 204–205)

The common socialization within a family, the family’s identity and the commitment to the long-term success of the family firm together facilitate the development of trusting relationships among family members for sharing knowledge, especially tacit knowledge that is crucial for building competitive advantages (Jaskiewicz, Uhlenbruck, Balkin, & Reay, 2013).

That said, relationship conflicts do occur occasionally and may result in family members trying to fight each other, putting personal interests above family interests, and being unwilling to share business information (Chirico & Salvato, 2008), as illustrated by the following quote from a third-generation owner-manager in a multiple-case study of the Italian winery industry:

Sometimes conflicts are very dangerous for our firm. The young generation has a different vision and they would like to create a new firm. But this is a great mistake. (Kammerlander, Dessi, Bird, Floris, & Murru, 2015: 339)

Family firms’ lack of trust in managers who do not belong to the family hinders knowledge sharing. The following quote is a continuation of the above quote from the founder of Firm Y:

I can’t tell the managers all the information I got and all the plans I have. Business is a secret. As for my brother, I can tell him. He knows the guys too. He has helped me all through the years. (Tsui-Auch, 2003: 211)

Such an attitude is not conducive to getting members of the management team on the same page and to leveraging the knowledge of each. Non-family members may be left in the dark about the rationale underlying a new strategic initiative.

In the context of internationalization, as mentioned, my study indicates that there is no significant difference between family and non-family firms in terms of operational experiential learning (Tsang, 2002b), implying that the related knowledge sharing is more or less the same for both types of firms. In contrast, since important strategic decision-making in traditional family firms often excludes non-family members, strategic management experience is mostly held within the family. In addition, the traditional family firms in my study tended to have a small team—consisting mostly of family members—when negotiating an FDI project, whereas non-family firms tended to have a larger team consisting of managers from different functional departments. As such, the related experience is also not as well shared among non-family members of traditional family firms. This hoarding of strategic information within the family may slow down the implementation of a complicated FDI project.

Professional family firms, on the other hand, have a more even distribution of strategic information among both family and non-family members because of the active participation of the latter in decision-making. That said, critical information related to confidential family motives is still held within the family only. As to non-family firms, the information managers possess depends a great deal on the functions for which they are responsible. Explicit mechanisms may be needed to encourage information sharing.

Information interpretation

After information is acquired by an organization, it must first be interpreted before actions can be proposed and implemented. The outcome of interpretation depends a great deal on who is involved in the process. As discussed, strategic information is concentrated in the family of a traditional family firm. The high level of trust, cohesive bonding and relatively similar backgrounds among family members may lead to groupthink when such information is interpreted. Worse still, when a family firm, whether traditional or professional, has an owner with a dominant personality—who is usually the founder or key successor and has superior power over the management of the firm—constructive questioning and debate of ideas may be curtailed (Ensley & Pearson, 2005). Efferin and Hartono’s (2015) case study of a family-owned Indonesian construction firm illustrates this effect. One family member spoke of the founder, Mr. TD, as follows:

Yes, Mr. TD was dominant during the meeting. He is our leader. We shouldn’t challenge him openly. We will surely evaluate and give him input during the execution later, but we can’t say “no, it’s impossible” during the meeting. It would be impolite. We must show respect to leader. (Efferin & Hartono, 2015: 146)

Gersick, Davis, Hampton, and Lansberg (1997) warned that in more extreme cases, dominant owners may be reluctant to seek input, advice or ideas from other family members, and thus reduce or even eliminate most idea or task conflict. Such conflict can often improve the quality of information interpretation and analysis.

For Firm X, after information about a potential FDI project had been collected by the founder and his personal assistant, it was interpreted and analyzed by the founder, his wife and the managing director (his old classmate). As the owner of the firm, the founder had the power to make the final decision. His personal assistant described his style as one that used a “sixth sense,” ignoring modern investment analysis methods such as discounted cash flows and payback periods (Tsang, 2002b).

In contrast, the non-family firms in my study used a more structured approach. The management of these firms organized the collected information, systematically analyzed it, produced a formal investment proposal (enclosing a feasibility study) and submitted the proposal to their board of directors for discussion and evaluation. In the process, a fairly large number of managers at different levels and of various functional areas were involved. Therefore, unlike family firms, the outcome was not subject to the personal, and sometimes capricious, style of a single individual—the dominant owner. With a higher degree of formalization and systematization, professional family firms’ approach is closer to that of non-family firms and the risk of groupthink is lower than that of traditional family firms. That said, the views of core family members would carry more weight than those of non-family TMT members who may be even more experienced and knowledgeable, adversely affecting the quality of interpretation.

A related factor is concerned with succession in family business. In a family firm, the successors to top management positions, such as the CEO and the chairman of the board of directors, are selected based on kinship from a small pool of family members. Kinship in terms of closeness to the founder often overrides competence in the promotion of family heirs (Pérez-González, 2006). In Tsui-Auch’s (2004) study of large Chinese family firms in Singapore, kinship remains the most important criterion for selecting top leadership positions even during a firm’s transition from traditional to more professional status. A consequence is that these successors are likely to be less competent than their counterparts in non-family firms, and the quality of information interpretation thus suffers.

Organizational memory

As an analogue of individual memory, organizational memory refers to the stored information generated from an organization’s history of operation that may affect its future decisions (Walsh & Ungson, 1991). The information is stored in two types of bins: human and non-human. Congenital learning in fact refers to the information in the human storage bin when an organization is created. Non-human storage bins exist in many different forms, including operations manuals, regulations, rules, organizational charts, reports, memos, computer software and so on. For simplicity of discussion, anything that is in an individual’s mind or memory is regarded as being kept in the human storage bin.

A key characteristic of the organizational memory of family firms is succinctly manifested by the aforementioned assertion of the founder of Firm Y that he could not share all the information he had collected with his subordinates because business is a secret (Tsui-Auch, 2003). Important strategic knowledge is kept in the human storage bin among family members. This is in fact a strength of family firms relative to their non-family counterparts. Personnel turnover among senior management often results in a considerable loss of the human component of an organization’s memory. The loss is even more damaging if the departing manager subsequently joins a competitor. Given the bonds among family members, the TMT of a family firm is generally more stable than that of a non-family firm. The odds that a family member will join a competitor are low, unless the family is torn apart by serious conflict. Therefore strategic knowledge and its associated competitive advantage are better protected in family firms, especially in the case of traditional family firms.

Moreover, if a family firm manages to implement a succession plan within the family, the transfer of knowledge from one generation to the next will be smooth and will often take the form of stories shared among family members, as exemplified by this quote from a fifth-generation owner-manager in the above-mentioned study of Italian wineries.

Our firm is a very old family firm. In our final product, there is the story of our family. That story transmitted across generations represents the core of our firm. (Kammerlander et al., 2015: 338)

In contrast, leadership succession in non-family firms is more complicated. The two CEOs concerned may not even know one another if the successor is hired from outside. The new CEO may learn little from the departing CEO, resulting in a loss of organizational memory. Similar problems often arise in the succession of other TMT members. On the other hand, when one family member takes over the position of another, transfer of knowledge is less of a problem and knowledge is likely to be conserved. The following quote from a family member of an Italian family firm drives this point home nicely:

Every generation brings something more which creates value in the business. Our overall level of knowledge has increased through generations. We have grown up in the family firm as persons, managers and now owners. The second generation was able to teach us directly and indirectly all the tricks of the trade in production, administration and distribution. Our parents also taught us how to communicate and cooperate with each other and solve problems. (Chirico, 2008: 442)

Despite this strength, a key weakness of traditional family firms is that its organizational memory relies too much on the owner. It is not difficult to imagine the devastating blow to Firm X if the founder suddenly died. His death would immediately and irrevocably wipe out the most important chunk of the firm’s strategic knowledge.

Gersick, Lansberg, Desjardins, and Dunn (1999) proposed that family firms may move through three ownership stages—controlling owner, sibling partnership and cousin consortium—as they develop. Each stage has different implications for the firm’s organizational memory. For instance, the above discussion of the impact of the death of Firm X’s founder refers to the controlling owner stage. For the other two stages, strategic knowledge is better shared among family members, while leakage of such knowledge to other firms is also more likely because some of these members may not be very committed to the firm’s success and may even join its competitors. A related issue is that, as the ownership of a family firm is passed from one generation to another, the number of family members is likely to grow and some members of later generations may not be actively involved in the firm. However, they can get access to important strategic knowledge through interacting with other family members who manage the firm. As such, knowledge protection becomes more difficult.

A critical internationalization challenge facing family firms is how to staff their foreign operations. Since these operations are not in physical proximity to the headquarters, the managers running the operations have to be trustworthy in the eyes of the owning family. It is thus natural that family firms tend to assign family members to be in charge of key expatriate positions (Tsang, 2002b). However, this practice becomes increasingly infeasible as a family firm’s overseas expansion continues because of the constraint of family size. Family firms often lack sufficient managerial capacity to monitor geographically-dispersed operating units (Carney, 1998), especially in the case of traditional family firms. When the extent of internationalization reaches a certain point, non-family members have to be hired to manage some of the overseas operations. Integrating the experience of these managers with that of the family can be a challenging problem.

Furthermore, as previously mentioned, traditional family firms in general have a lower degree of formalization and thus store less knowledge in the non-human component of its memory. The non-human storage bin would be almost entirely incapable of retaining non-family members’ experience and their departure could lead to the loss of virtually all of their experience. Given their practice of actively involving non-family members in management and their higher levels of formalization, professional family firms are in a better position to tackle these problems than their traditional counterparts. For instance, they may be able to capture some of the non-family members’ FDI experience by requiring them to submit regular written reports (non-human storage bin) and to share the experience with their TMT (human storage bin).

Discussion

The above section discusses the differences and similarities between family and non-family firms as well as between traditional and professional family firms with respect to Huber’s (1991) major learning processes in the context of internationalization. Each type of firm has some unique characteristics of learning. Excluding non-family TMT members from making crucial strategic decisions, traditional family firms have a high concentration of strategic knowledge within the family. Leakage of the knowledge is prevented by family members’ commitment to the success of the business. The process of internationalization is constrained by the personal connections of key family members in identifying FDI opportunities and by the reluctance to hire outsiders to run overseas operations. The willingness to trust non-family members and let them participate in strategic decision-making enables professional family firms to overcome these constraints.

There are many examples of professional family firms that are truly global in their operation. One excellent example is Li Ka-shing’s conglomerate mentioned above. Although these firms have a more even distribution of information among TMT members, critical information related to confidential family motives is still kept within the family only. Non-family firms are free from factors related to family ties among their managers. Members of their founding teams are likely to have a greater variety of experience and expertise than those of family firms, resulting in better congenital learning. The knowledge possessed by managers depends a great deal on the functions for which they are responsible. Succession of key TMT members often results in losing knowledge in the human storage bin. Moreover, the risk of leaking critical knowledge is considerable because of managers’ opportunistic behavior.

The organizational learning perspective not only deepens our understanding of family firms’ internationalization but also enhances researchers’ ability to investigate the phenomenon. By way of illustration, a sample proposition is included in Table 1 for each learning process or subprocess with respect to the internationalization of traditional family firms. For example, Arregle et al.’s (2017) meta-analysis of 76 studies indicates that family firms have a narrower geographic scope of internationalization for FDI than other firms. This result is not surprising, given that family firms’ search for FDI opportunities is constrained by the personal ties of their key family members. In fact, detailed information about the ties of the owner of a traditional family firm may enable a fairly accurate prediction of which countries the firm will invest in (P4 of Table 1). However, the relationship between the owner’s background and FDI is weaker in the case of professional family firms. As another example, the discussion of congenital learning suggests that it takes a longer time for a traditional family firm’s new foreign operation to become profitable (P3 of Table 1). The main reason is that the firm tends to select expatriate managers from a restricted pool of family members, and it is likely that those chosen are not well qualified for the job. Without a solid knowledge foundation to start with, the foreign operation needs more time to establish its foothold. This problem of congenital learning is less serious for professional family firms, not to mention non-family firms. In short, other than the sample propositions listed in Table 1, more can be derived from the differences in learning behaviors among the three types of firms to guide empirical research.

Although this paper focuses on FDI, differences in learning affect other aspects of firm behavior. For instance, when exploring opportunities for exporting, professional family firms are likely to engage in detailed and systematic market research of foreign countries. Conversely, traditional family firms tend to rely more on haphazard foreign market information collected by their family members. Moreover, being constrained by the availability of family members or trusted non-family members, the latter are less likely than the former to set up overseas sales offices.

As another example, consider R&D investments. Since such investments can be substantial and results are often uncertain, traditional family firms are reluctant to hire outsiders who possess relevant background knowledge to lead their R&D units. Instead, they may appoint family members or trusted non-family members and expect these individuals to learn on the job. However, R&D often involves specialized knowledge that is hard to pick up through learning by doing. Lacking sufficient knowledge, such individuals are not qualified to be in charge of the units. Thus, the R&D investments of traditional family firms tend to have lower returns than those of their professional counterparts.

Most of the examples used in the discussion are Chinese family firms, drawn primarily from my two studies (Tsang, 2001, 2002b) and Tsui-Auch’s (2003, 2004). Although this type of family firm constitutes a dominant economic force in the Asia Pacific region, some of its unique characteristics may not be shared by other types of family firms, as illustrated by the aforementioned issue of trusting strangers. An implication is that the pace of localization—the hiring of local managers in the host country to run an operation—is likely to be slower for Chinese family firms than family firms of other cultural origins.

That said, family firms worldwide do seem to share some common characteristics. For instance, based on a sample of 360 firms, 160 of them being family-controlled and the rest non-family-controlled, extracted from the Compustat database, Gomez-Mejia, Makri, and Kintana (2010) showed that family firms prefer to invest in countries that are culturally similar to their home country.Footnote 2 This finding is consistent with my own study of Chinese family firms and is likely to be somewhat generalizable to family firms of various cultural origins, at least during the early stage of internationalization. The reason is that the owning family, in the case of a traditional or professional family firm, has the final say in choosing a foreign country to invest in. It is likely that most family members would prefer a country with which they feel culturally comfortable, that is, a country that is culturally close to the home country. That said, the generational stage of a family is also a key factor here. For example, it is rather common that founders of Chinese family firms send their children to receive education in Western countries (Tsui-Auch, 2004). As such, the younger generation may be less reluctant to invest in countries with cultures that are dissimilar to the Chinese culture. Again, this point is likely to be generalizable, in that later generations of family firms tend to invest in a wider range of countries than the first generation.

Conclusion

Various theoretical lenses have been used to study family firms in general (Gedajlovic, Carney, Chrisman, & Kellermanns, 2012) and the internationalization of these firms in particular (Pukall & Calabrò, 2014). My discussion of family firms’ learning processes supports Hamilton’s (2011) argument that a learning lens can bring theoretical insights to the study of family business. Organizational learning could in fact be integrated with other theoretical lenses to achieve a deeper understanding of a phenomenon. For example, the concept of socioemotional wealth offers a relatively new theoretical lens (Berrone, Cruz, & Gomez-Mejia, 2012). Pukall and Calabrò (2014) provided an interesting conceptual discussion of how considerations of family firms to preserve socioemotional wealth can affect the type and speed of learning related to internationalization. However, without further empirical tests, their argument will remain conjecture.

It should be noted that the stream of research on organizational learning is itself evolving. As the number of studies grows, the perspective is further developed and strengthened, offering some new conceptual angles, one example of which is the reconceptualization of organizational unlearning (Tsang, 2017; Tsang & Zahra, 2008). For instance, it would be interesting to study how family firms unlearn obsolete practices in their foreign ventures or how their unlearning style affects the process of internationalization.

To conclude, this paper shows that organizational learning is an intellectually vibrant—yet not adequately explored—option for family business researchers. I look forward to seeing future research in this direction.