INTRODUCTION Managing Multinationals in a Knowledge Economy: Economics, Culture,

Transcription

1 SEARCHING FOR THEORIES OF DYNAMIC RELATIONSHIPS IN BUSINESS STRATEGY: COMMENT ON JOHN DUNNING S RELATIONAL ASSETS, NETWORKS, AND INTERNATIONAL BUSINESS ACTIVITY PAPER Will Mitchell INTRODUCTION Professor Dunning offers a thoughtful paper (Dunning, 0b) concerning relational assets and international business activity. The work is sweeping in its scope and concepts. Given the breadth, my commentary inevitably will need to focus on some aspects at the expense of others. Rather than to critique the paper exhaustively, then, my goal is to identify intriguing aspects of the argument and to offer some tentative suggestions for extending and revising the argument. Managing Multinationals in a Knowledge Economy: Economics, Culture, and Human Resources Advances in International Management, Volume, Copyright 0 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0-/doi:./S0-(0)0-

2 WILL MITCHELL OVERVIEW Professor Dunning first argues that social capital plays a critically important role in business strategy and performance, where social capital consists of resources that are embedded in networks of relationships. He then focuses on the concept of relational assets, which he views as a dimension of social capital. Although the working definition of relational assets is somewhat complex, Professor Dunning s core idea is that relational assets comprise an actor s ability to form and govern beneficial relationships with other actors, including other firms and individuals within a firm. Firms and individuals use their relational assets to gain access to other actors assets and to coordinate the use of their partners assets with the focal actor s own resources. Relational assets, in Professor Dunning s approach, have a wide range of characteristics. At the most general level, relational assets emphasise attitudinal attributes such as values, honesty, trust, cultural sensitivity, and reciprocity. The paper offers several useful ways of identifying these ambiguous assets, ranging from firm-level measures such as alliance experience and reputations to country-level measures such as corruption and civic engagement. The basic idea is that the ability to leverage resources that other actors control arises from the ability to engender trust in one s own judgement and intentions. The paper further argues that relational assets are becoming increasingly important in business practice. The drivers of importance include the growing roles of knowledge intensive economic activities, growth in cross-border business, technological convergences, and general societal changes. Professor Dunning then places relational assets in the context of his Ownership Location Internalisation (OLI) eclectic paradigm of MNE activity. Relational assets provide ownership advantages through superiority in coordinating the use of functional assets. He proposes that MNEs tend to have more relational assets than domestic firms, owing to a greater number and intensity of linkages. Relational assets provide locational advantages through superiority in business infrastructures. He proposes that MNEs are increasingly seeking locations that offer superior country-based relational assets, particularly in knowledge and service-intensive industries. Relational assets provide internalisation advantages in that they provide linkages to many other assets. He argues that the growing importance of relational assets means that the locus of internalisation is shifting from internalisation of functional assets to internalisation of governance skills, such that firms increasingly gain value through their ability to govern their access to other actors assets. He proposes that dyadic and network forms of inter-firm organisation will increase, which in turn will increase the value of firms relational assets (i.e. their ability to govern inter-actor organisation). Thus, firms may need

3 Searching for Theories of Dynamic Relationships in Business Strategy to emphasise internalisation of the ability to govern (relational assets), as much or more as the ability to produce (functional assets). At the most general level, then, Professor Dunning proposes that firms ability to create and use relational assets will have an increasingly important influence on MNE activity. Moreover, if MNEs do in fact tend to have more relational assets than domestic firms do, then the growing importance of relational assets would drive an increase in MNE-related business activity. Professor Dunning suggests several implications of his discussion. International business theory needs to give greater attention to relational assets as key sources of firm advantage. Business managers need to develop greater skill in creating relational assets. Policy-makers need to improve social capital and relational assets within their environments. Supra-national agencies need to foster international respect for the underpinnings of relational assets, such as trust and reciprocity. STRENGTH The strength of the paper lies in the thoughtful development of the idea that the ability to govern the use of assets across actors is a critical component of business skill, whether domestically or internationally. The discussion helps reinforce the shift that international management theory, and strategy theory more generally, has made in recent years. Rather than focus on stocks of physical assets as the primary sources of competitive advantage, many thoughtful analysts now emphasise the more nebulous but critically important notion of business capabilities as sources of business advantage. Indeed, at a somewhat more subtle level, the discussion creates a shift of emphasis within knowledge intensive views of business activity. Knowledge theorists have long argued that what a firm knows how to do is as or more important in driving business performance as the stocks of physical assets that the firm owns. The notion of relational assets refines this sense of knowledge-intensity, by emphasising that individual employees and firms need the ability to work with a wider range of assets than the boundaries of any one function or organisation can contain. Moreover, the notion of relational assets provides a multi-level assessment of this governance ability, ranging from individual to firm to social to cross-border abilities to create and manage linkages among resources. A related strength of the paper is that it has sufficient scope to offer opportunities for development. The opportunities arise both in refining the existing argument and in undertaking more fundamental changes that might lead to substantial revision of the core argument.

4 0 WILL MITCHELL REFINEMENTS Four refinements offer immediate opportunities for fruitful thought. First, it would be useful to offer a more focused definition of the core notion of relational assets. The current working definition is somewhat complex. Professor Dunning states that We shall define a firm s R-assets as: The stock of a firm s willingness and capability to access, create, and shape economically and to coordinate the resources and capabilities, necessary to beneficial relationships; sustain and upgrade the quality of these relationships. Such relationships which may take various forms though always conducted by and between individuals, may take place both within the confines of a particular firm, or between that firm and other organisations or individuals (Dunning, 0b, p.). This is a helpful start at a definition, because it establishes a broad scope for the assessment but leaves some ambiguity in terms of identifying the core concept. Moreover, the working definition runs the risk of confusing the existence of the assets (capability to create and coordinate relationships) with the outcome of using the assets (relationship quality). I will return to this point later, but it is important to recognise that relationships frequently have negative consequences as well as beneficial aspects. Let me offer a somewhat more succinct attempt at a definition, which emphasises the assets rather than their outcome: Relational assets are the mechanisms with which an actor forms and governs relationships with other actors. Second, once we have an accepted definition, it would be useful to identify the underlying mechanisms that comprise relational assets. Professor Dunning offers several indicators of the presence of relational assets, ranging from alliance experience at the firm level to common R&D capacity at the dyadic level to degree of corruption at the national level. These are viable initial indicators but for the most part are not the underlying mechanisms themselves. Recent work in alliance strategy has begun to identify specific mechanisms for governing inter-actor relationships. Research by Miguel Rivera (0), for instance, has been categorising and evaluating the mechanisms that firms use to protect and coordinate the resources they use in alliances. Let me suggest some options for identifying relational asset mechanisms at the various levels of analysis, recognising that these are far from exhaustive lists. Individual-level and intra-firm relational asset mechanisms include participation in cross-functional work groups and matrix organisations. Firm-level relational asset mechanisms include alliance management staff groups, relationship-training programs, membership on other firms boards of directors, trade association memberships, and lobbying initiatives. Country-level relational asset mechanisms include proprietary right laws and norms that facilitate exchanges, as well as laws and norms that enforce

5 Searching for Theories of Dynamic Relationships in Business Strategy corporate and public fiscal reliability. The common feature of these mechanisms, across levels, is that they are identifiable and at least partially adjustable processes that facilitate the ability and willingness of one actor to commit resources in joint activity with another actor. Third, it would be useful to speculate how firms create relational assets. This issue is one that arises in several similar literatures, such as the resource-based view of the firm and evolutionary theories, which tend to emphasise strong path dependence and difficult replicability as key features of valuable assets. Clearly, many relational assets, like other firm capabilities, will emerge over time in largely unplanned and unanticipated ways. It would be conceptually limiting, though, to posit that relational assets primarily are the result of lucky or unlucky accidents of past history. Instead, let us consider a few ways in which firms and other actors can actively manage the process of creating relational assets. At the individual level, some individuals spend substantial time and effort in creating linkages with other people, whether through common schooling or through work-related activity. These linkages, in turn, provide means of facilitating common projects. At the firm level, some companies invest substantial time and effort in creating linkage abilities. One example is the recent trend at companies such as General Electric and Coke to creating knowledge/learning officers and staffs, who are responsible for facilitating knowledge flow within a firm. Similarly, companies such as Eli Lilly and others have created alliance management staff organisations, with the explicit charge of managing partnership relationships with other firms. Typically, such corporate-wide initiatives arise at relatively high levels of the firm. In addition, though, some initiatives that lead to the formation of relationship assets arise through operating practice. For instance, recent work by Zhao, Anand and Mitchell (0) has found that overseas training programs in international joint ventures sometimes lead to the formation of bridge networks. During intensive on-the-job overseas teaching programs, trainees often gain detailed knowledge of who knows what at another organisation. The trainees can then use their networks to obtain needed knowledge and facilitate operating interactions long after their training session ends. The common point in these examples is that firms can actively manage the process of creating relational assets, as well as using relational assets that happen to result from emergent processes. Fourth, it would be useful to provide greater comparative specificity in the propositions. Professor Dunning (0b) offers a thoughtful set of conjectures in each of the key sections of the paper, concerning the existence and effect of relational assets. Some of the propositions are quite direct, e.g. the hypothesis that MNEs are likely to possess a greater stock of R-assets relative to non MNEs (p...). Such direct hypotheses have several strengths: they allow us to focus on

6 WILL MITCHELL specific aspects of the argument, and they provide potentially testable implications of the argument. Other propositions, though, are somewhat more ambiguous, particularly in the frequent argument that relational assets are becoming more common and/or more important than in the past, e.g. R-assets are becoming a more important component of the resources and capabilities of firms engaging in cross border activity (p...) and Location specific R-assets are becoming a more important influence on the location choices of MNEs, both between and within countries (p...). The first issue with such temporally oriented proposition lies in specifying the comparison, such as the same firm or individual at an earlier point in time, a pool of all firms and individuals at some earlier time, or a set of country-specific or industry-specific actors at some earlier time. The second issue arises in setting a time point for comparison, such as any time in the industrialised era, since World War II, during the post-vietnam era in the U.S., or since the growth of non-u.s. multinationals during the past quarter century. Providing such comparison group and time-point specifications would help us find the bounds to the argument, as well as give greater focus to some aspects of the argument. FUNDAMENTAL QUESTIONS Beyond the refinements, five issues suggest more fundamental questions that we might want to address in developing the notion of relational assets. First, it would be useful to drill down into the argument that relational assets are more important now than they have been in the past and, moreover, that MNEs have more relational assets than domestic firms. An alternative argument would be that relational assets have always been important, but what has changed is the types and scope of relationships that matter. In parallel, the difference between MNEs and domestic firms may lie more in the types of relational assets than in the number of relational assets. Certainly, since the beginning of the industrial era at least, firms have needed to manage relationships within their boundaries and to draw on resources outside their own organisational boundaries in order to survive. Steel plants required partnerships with mines and end-product producers, as well as with unions, local and national political parties, and other external agencies. Similarly, early rail, telecommunications, oil, chemical, and pharmaceutical firms needed commercial and political linkages to become established and to grow, both domestically and across borders. In parallel, there is a growing literature that stresses that domestic firms in emerging economies depend strikingly on networks of relationships for their survival and success, where the networks include personal linkages among family members, commercial linkages with suppliers and distributors, and linkages with political authorities. Notions of governance skills

7 Searching for Theories of Dynamic Relationships in Business Strategy and relationship power have longstanding centrality in institutional economic and sociological theories, whether early theorists such as Marx, Coase, and Commons or more recent discussions in transaction-cost theory, resource dependence and network theory, agency theory, evolutionary economics, and economic sociology. Thus, it is not at all clear either that relationships are more important now than they were in the past or that multinational firms have more relational assets than domestic firms do. Rather than a change in importance of relational assets, then, what may have changed is the nature and scope of relational assets. As business has increasingly become national and international in scope, relational assets have needed greater formalisation. When the scope of business activities lies within a small region, many commercial and political ties arise through informal networks. An engineer at a chemical company, for instance, may talk at lunch with an engineer at an electronics company across the street, to jointly identify an opportunity to tie their development activities together. An engineer at software or hardware firm in a high-tech ghetto may consult informally with an academic scientist in a local university. Similarly, a senior manager of a local firm can jawbone a local regulator during a meeting of a local social club. Equivalently, a family-based business can rely on informal personal ties among the family members who run its various activities to govern the linkages among those activities. When business expands its geographic and managerial scope, such informal ties continue to be important, but often become insufficient. Instead, the business often will need to create trade organisations, explicit alliance agreements, specialised lobbying efforts, detailed training programs, and other formal relationship asset mechanisms. We need to take care, then, in distinguishing between the incidence of relationship assets and the type of relationship assets that arise in practice at different points in commercial history. Second, it would be useful to extend the mechanisms that underlie relational assets into areas of power, as well as those of trust. Professor Dunning s approach to relationships emphasises the importance of inter-actor trust and affinity. These nice aspects of relationships clearly facilitate inter-actor activity. At the same time, though, brute-force social and economic powers also play central roles in relationships. The ability of one actor to gain access to another actor s resources, which is a central part of relational assets, often depends on the first actor s power over the second. This power may arise because the actor controls supply sources or distribution channels that are important for the second. The power may arise because one actor has ties with political actors who can influence the other actor s fate. The power may arise from greater social status of one actor s leaders. The power may arise because an MNE has the credible threat of moving its operations to another country if a government or union or supplier does not grant it the terms

8 WILL MITCHELL that it demands. The power may arise from other sources. But, no matter what the source of the power, it undoubtedly plays a role in managing relationships. Hence, a theory of relational assets needs to take power into account. Third, it would be useful to assess which aspects of the argument reflect underlying aspects of relational assets and which assertions stem more from personal values and goals. This issue is particularly salient at the country level. Several of the arguments in the paper address the importance of creating desirable social environments for business activity and list factors such as drugs, terrorism, crime, divorce, and pornography as socially dysfunctional problems. At a high level, there is a strong argument that the social environment affects investment desirability. Certainly, many aspects of the social environment contribute to measures of country risk and other investment guides that researchers and managers use regularly. Similarly, the current wave of corporate crime and mispractice in the U.S. is having a direct negative effect on investor confidence and corporate investment. Nonetheless, substantial issues arise when attempting to identify which social factors create problems and which have little impact on business activity. For instance, the U.S. has substantial crime levels and drug use, as well as an incarceration rate that is among the highest in the world, but also enjoys substantial international investment. Are crime and drug use irrelevant in their impact on relational assets, then, or do only some types of crime and drug use matter, or perhaps crime and drug use in concert with some other factors? Hence, paring away the some of the more sweeping assertions, which appear to rely on laudable but perhaps conceptually peripheral values, would help identify and focus on the core aspects of the thesis. In parallel, the paper argues that one of the drivers of relational asset growth is that societies are increasingly valuing quality of life, rather than concentrating on purely efficiency related issues. While one might want to believe that a trend to improved life quality exists, there appears to be an element of hope within the argument. At least two issues arise. First, it is not clear that pure efficiency has been a strong social goal in the past, whether in developed or developing economies. Second, recent trends in market economies may actually speak to greater emphasis on economic efficiency, through corporate reorganisation and growing incidence of cross-border acquisitions throughout the world, perhaps at the cost of reducing corporate ability to address the needs of multiple stakeholders. Fourth, it would be helpful to address the degree to which within-firm and inter-firm relational assets differ conceptually. It is intriguing and insightful to encompass both levels of analysis with the relational assets concept. At the same time, though, it is sometimes difficult to tease out when personal assets are the

9 Searching for Theories of Dynamic Relationships in Business Strategy drivers of commercial activity and when firm-level assets are more important. As a starting point, it would be useful to distinguish between relational assets that depend only on individual level action and those that involve routine-based activity. This distinction would help us isolate differential ways in which relational assets arise, as well as differential ways in which firms can manage and use their relational assets. One key difference is that routine-based assets often emerge through unanticipated processes and require greater concerted effort to manage and use. In parallel, though, the routine-based assets may be more firm-specific than individually based assets, to the extent that individual employees are able to move from firm to firm. Fifth, it would be helpful to assess the role that relational assets play in creating relationship lock-in, which sometimes counters their role in creating business value. For the most part, the paper takes a positive view of relational assets, explicitly viewing them as means of achieving beneficial access to resources. At the same time, though, relational assets may well tend to create semi-permanent ties to particular actors. These ties are beneficial as long as the actors resources have value in a given environment. When an environment changes, whether due to change in technology, customer tastes, political regimes, or the death or exit of a partner, such lock-in can constrain firms along failing paths. Recent work by Singh and Mitchell (), for instance, shows that having partnerships with other firms helps a company grow and survive but that failure rates rise above industry averages when firms lose access to their partners, especially if they lack the ability to form new partnerships. Similarly, many Japanese firms benefited from relationship ties that helped them develop new technology and expand internationally during the post-war era. During the past decade of economic stagnation in Japan, though, those same ties have inhibited corporate adjustment and contributed to the country s inability to break out of recession. Hence, we may want to distinguish conceptually between relational assets that allow an actor to form ties with other actors and relational assets that allow an actor to break existing ties when the other actors resources lose value or when the other actor disappears. In turn, we can discriminate empirically among mechanisms for forming relationships, mechanisms for managing relationships, and mechanisms for ending relationships. CONCLUSION Overall, then, the paper offers fascinating arguments and insightful suggestions. Professor Dunning draws from his experience to identify an important avenue within international business theory and practice. The breadth of the implications

10 WILL MITCHELL of relational assets which apply to individuals, firms, countries, and international organisations makes this line of argument particularly important. ACKNOWLEDGMENT This paper (Dunning, 0b) was presented in the AIM-IMD Distinguished Scholar Forum at the 0 Academy of Management Meeting in Denver, which is an expanded version of Dunning (0a). REFERENCES Dunning, J. H. (0a). Relational assets, networks, and international business activity. In: F. J. Contractor & P. Lorange (Eds), Cooperative Strategies and Alliances (pp. ). Oxford: Elsevier. Dunning, J. H. (0b). Relational assets, networks, and international business activity: Extended thoughts on Dunning 0a. Presented in the AIM-IMD Distinguished Scholar Forum, Academy of Management Meeting, Denver. Rivera, M. (0). How do firms learn from alliances? HEC Graduate School of Management, Paris, unpublished dissertation. Singh, K., & Mitchell, W. (). Precarious collaboration: Business survival after partners shut down or form new partnerships. Strategic Management Journal,,. (Special Issue on Evolutionary Perspectives on Strategy.) Zhao, J., Anand, J., & Mitchell, W. (0). Transferring collective knowledge: Teaching and learning in the Chinese auto industry. Duke University Working Paper Series.