Social Structure and Intermediation: Market-making ... · PDF fileSocial Structure and Intermediation: Market-making Strategies in International Exchange* Paul D. Ellis Hong Kong Polytechnic

Social Structure and Intermediation:Market-making Strategies in InternationalExchange*

Paul D. EllisHong Kong Polytechnic University

Information gaps between markets create opportunities for internationaltrade intermediaries to negotiate cross-border exchanges. Faced with the prospect ofeventually being eliminated from these exchanges, intermediaries must continuallysearch for new opportunities to mediate international exchange. In this paper anoriginal explanation is derived from the core principles of structural hole theory toexplain how these market-making firms operate in the tension found between theinevitable decay of existing exchange relationships and the uncertainty of findingreplacement sources of income.

INTRODUCTION

International trade intermediaries have played an unparalleled role in internationalbusiness ever since Europe’s maritime powers created the first chartered tradingcompanies in the early seventeenth century. In their quest for profitable trade routesthese forerunners to the modern multinational enterprise promoted colonialism,engaged in state-sponsored piracy, trafficked drugs, annexed entire nations, andcarved much of the world into the geopolitical shape it is today. Although corpo-rate strategies have since become more civil, international trade intermediaries(ITIs) remain a potent force in many of the world’s major markets. In Asia alonethe 131 ITIs listed among the region’s top 1000 firms have a combined annualturnover of more than US$1.6 trillion (Anwar and Alex, 1999).

ITIs are distinguished from multinational enterprises primarily by their empha-sis on the organization of trade. ITIs are also unique in their unrivalled ability toreinvent themselves ‘generation after generation’ ( Jones, 1998a). It is somewhatsurprising therefore, that mainstream organizational research has generally dis-

Journal of Management Studies 40:7 November 20030022-2380

© Blackwell Publishing Ltd 2003. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ,UK and 350 Main Street, Malden, MA 02148, USA.

Address for reprints: Paul Ellis, Department of Business Studies, Hong Kong Polytechnic University,Hung Hom, Kowloon, Hong Kong ([email protected]).

missed the ITI as a relic of a bygone age. This is unfortunate for two reasons. First,for better or worse, over the past four centuries ITIs have demonstrated theirability to integrate the poor two-thirds of the world into the global economy (Ellis,2003). Indeed, their very existence depends on their knack for making marketswhere none previously existed and then moving on when those markets mature.Second, as one of the most enduring species within the genus intermediary – trulythe dominant corporate form of the information age – ITIs ‘encapsulate veryneatly the vision of a firm as an information processing system’ (Casson, 1997, p.245). For these reasons chiefly, the study of ITIs has much to offer both manage-ment theorists and society at large.

The aim of this paper is to develop a generic explanation for those organiza-tional routines which ensure the survival of ITIs in mediated cross-borderexchanges. The paper is divided into four parts. In the opening section the eclec-tic literature on the topic is briefly surveyed and this is followed by a more in-depthreview of the dominant theoretical frameworks which have been used to accountfor ITI behaviour. In the third part of the paper the entrepreneurial role of thetrade intermediary is assessed in light of recent insights drawn from structural holetheory and this leads to the derivation of four propositions.

BACKGROUND

International trade intermediaries have been studied in their many guises andnatural habitats and an extensive mosaic of literature on the topic has emerged.To date considerable research has been done on North American ITIs (Bello andWilliamson, 1985; Howard and Maskulka, 1988; Kelly and Lecraw, 1985; Perry,1990), Asian ITIs (Cho, 1984; Ellis, 1998; MacBean, 1996; Sarathy, 1985), andthe many European variations on the theme (Carlos and Nicholas, 1988;Mattsson, 1990; Sugiyama, 1987). While research into ITIs is largely nationalisticin scope (Ellis, 2001), with few truly cross-national studies in existence (cf. Armine,1987), within this chauvinistic body of literature common research themes haveemerged relating to, for example, patterns of diversification (Balabanis and Baker,1993a, 1993b; Cho, 1987) and organizational evolution (Ellis, 2001; Kim, 1986;Perry, 1990). Two especially fertile topics for study have been (i) the unique andhighly diversified Japanese version of ITI known as the sogo shosha (Kojima andOzawa, 1984; Shao and Herbig, 1993; Shin, 1989; Yoshihara, 1982) and (ii) the‘US Export Trading Companies Act’ of 1982 (Armine et al., 1986; Howard andMaskulka, 1988; Kaikati, 1984; Williams and Baliga, 1983). (These otherwise dis-similar topics are intrinsically connected as the rationale for the latter was inspiredin part by the success of the former (Kaikati, 1984).)

Even a cursory glance at the studies just listed reveals that ITI research haswaxed and waned in correspondence with the economic performance of Japan.After Japan entered a sustained period of recession in the early 1990s, the number

1684 P. D. Ellis

© Blackwell Publishing Ltd 2003

of papers extolling the might of the Japanese general trading companies declinedmarkedly. At around the same time, scholars and policy-makers were beginning toacknowledge the failure of the ETC Act (1982) to live up to expectations (Czinkotaand Onto, 1990). The result? A once active area of international business researchentered a period of hiatus.

Now research investigating ITI behaviour is enjoying a renaissance of sorts. Tworecent papers published in a leading international business journal have promotedthe study of ITIs as the missing link in export development research (Peng andIlinitch, 1998) and endeavored to more fully classify the nature of the ITI service-offering (Balabanis, 2000). A recently edited volume (Jones, 1998b) has made asubstantial contribution towards understanding the many varieties of multina-tional trading company. Qualitative research conducive to theory building hasemerged in place of the simple descriptive surveys of the earlier era (Buchan, 1994;Ellis, 1998; Jones, 1996; Munro, 1998). Indeed, case studies of high performingITIs are even finding their way on to the syllabi of prestigious business schools(e.g., Long and Seet, 1995; O’Connell, 1995).

Perhaps most significantly, from an epistemological perspective, ITI researchwhich was once largely descriptive and classificatory in nature (e.g., Brasch, 1978;Cho, 1984; Howard and Maskulka, 1988), has now progressed towards goals ofexplanation (e.g., Balabanis, 1998; Peng, 1998) signalling the maturation of thefield. New theories emphasizing the role of the ITI as an information hub (e.g.,Casson, 1997, chapter 9) seem especially promising in this regard. However, todate theoretical advancement has been slow and piecemeal due in part to a lackof a consensus on matters of definition (Armine, 1987; Bello and Williamson,1985; Brasch, 1978; Cho, 1987). By way of illustration, trade intermediaries arereferred to as export management companies or export trading companies in theUSA, foreign trading companies in China, senmon shosha in Japan, and Chonghap-

mooyeok-sangsa in Korea (Cho, 1987; MacBean, 1996). They may also be catego-rized according to the type and range of services offered, hence the terms general

trading company, specialized trading company, commodity trading company, and fed-erated export management group (Armine, 1987). For these reasons, very few uni-versally applicable definitions are available (cf. Balabanis and Baker, 1993b) andthe more common approach is to define the ITI in terms of the service it pro-vides. In this regard, Casson’s (1998) recent definition is arguably the most usefulyet least restrictive: a trading company is any firm that specializes in market-making intermediation, an activity which may involve brokerage (selling on behalfof another) or reselling (taking title to the goods traded). Building on this defini-tion, an international trade intermediary may be defined as any market-makingintermediary involved in the indirect export of a product between a supplier anda buyer located in separate countries. The high number of permutations on thisfundamental intermediary role reflects either differences in the scale and scope ofoperations (Cho, 1987), the stage of organizational development (Balabanis and

Social Structure and Intermediation 1685


Baker, 1993a; Ellis, 2001; Kim, 1986), or idiosyncrasies in the local institutionalor cultural context (Kojima and Ozawa, 1984; Perry, 1990; Yoshihara, 1982).

ALTERNATIVE EXPLANATIONS FOR INTERNATIONAL TRADEINTERMEDIARIES

Uncertainty in economic activity is correlated with the geographic, cultural, andtemporal distance separating buyers from sellers. Trading companies exploit thisuncertainty and its inherent risks by developing and mediating international tradeflows (Kojima and Ozawa, 1984). The resulting connections between ITIs andtheir trading partners have been analysed from the perspective of agency theory,which is concerned with the optimal contract governing relations between partiespursuing potentially incompatible goals (Eisenhardt, 1989; Jensen and Meckling,1976), and transaction cost analysis, which considers the efficiency tradeoffs inher-ent in different exchange structures (Coase, 1937; Rindfleisch and Heide, 1997;Williamson, 1975, 1985, 1996).

Agency Theory

In a brokerage-type trading relationship, a manufacturer will contract with anexternal ITI who will assume the export marketing responsibilities for the manu-facturer’s product in return for a commission fee. Contractual problems will arisewhen the intermediary (or agent) acts opportunistically in the event that their inter-ests conflict with those of the manufacturer (or principal) (Sharma, 1997). Thisarrangement is particularly troublesome for the principal because each party pos-sesses different information about the task to be performed and this imbalance orasymmetry favours the agent. For example, to win the contract in the first placethe intermediary may be tempted to conceal or misrepresent information regard-ing their abilities and resources. If the principal cannot fully verify these claims inadvance, they may inadvertently suffer from a problem of adverse selection. A secondpotential problem, moral hazard, may arise after the contract has been signed whenthe principal is unable to ensure that the agent delivers the agreed-upon level ofeffort (Eisenhardt, 1989). The central issue of agency theory therefore, concernsthe structuring of exchange relationships with the appropriate mix of incentivesand penalties to ensure intermediaries are motivated to perform the delegated taskin accordance with the expectations of the principal (Lassar and Kerr, 1996).

The chartered trading companies of seventeenth century England provide aclassic example of contracting to reduce agency costs. These early multinationalshad to devise methods for controlling agents in far-flung parts of the world in anage when market information was highly imperfect. Information asymmetriesmeant that agents based in India or Hudson Bay could easily pursue their own self-enrichment at the expense of their principals back in distant London (Carlos and

1686 P. D. Ellis


Nicholas, 1988). To counteract this temptation, the trading companies providedgenerous compensation packages, required managers to take oaths of allegianceand periodically opened and read the private mail of employees. As a case in point,the Royal African Company dealt with the problem of moral hazard by requiringemployees to post a bond ranging from six to ten times their annual salary. Thisbond, combined with thorough searches for contraband items every time a shipreached an English port, evidently induced the required level of effort while atten-uating opportunism (Carlos, 1992). Such measures, however, did little to protect theRoyal African Company against the problem of adverse selection. In this regardthe Hudson Bay Company of the same era was unique in its development of anapprenticeship programme to train managers. By promoting managers though theranks, the company was able to effectively screen out unsuitable agents while devel-oping a class of managers closely allied to the goals of the organization. This system,which replaced an earlier bonding arrangement, may partly explain why theHudson Bay Company survives today (Carlos, 1992; Carlos and Nicholas, 1988).

Transaction Cost Analysis

The hazards associated with relying on intermediaries may render the agency rela-tionship inefficient in a given situation and compel the would-be exporter to pursueother modes of exchange, such as direct exporting. The matching of appropriateexchange characteristics (or governance problems) with exchange modes (or gov-ernance structures) describes the domain of transaction cost analysis (Heide andJohn, 1988, 1992). Transaction cost analysis, or TCA, is based on the axiom thatfirms will choose exchange modes which minimize the costs of transacting. Toovercome the barriers to trade created by uncertainty, potential exporters will incurat least four discrete types of transaction costs:

(1) Search costs associated with gathering information to identify and evaluatepotential trading partners.

(2) Contracting costs associated with negotiating and writing agreements.(3) Monitoring costs associated with ensuring that each party honors the prede-

termined set of obligations.(4) Enforcement costs associated with ex post bargaining and the sanctioning of

trading parties that fail to perform according to the agreement (Dyer, 1997,p. 536).

All things being equal, TCA suggests that instead of performing export-relatedtasks in-house, a potential exporter will be better off using a specialist trade inter-mediary in order to tap into the benefits that the middleman is able to obtain inthe form of scale economies in distribution (Anderson and Coughlan, 1987). Bycarrying product lines for several manufacturers, even small ITIs are able to export



more cheaply than a single manufacturer going it alone (Perry, 1990), and com-petitive pressures will compel intermediaries to pass along these lower costs tomanufacturers (Williamson, 1979). Thus, from a TCA perspective, ITIs existbecause they are able to lower transaction costs for their clients ( Jones, 1998a;Peng and Ilinitch, 1998; Roehl, 1983).

However, the value of employing an intermediary will vary with the nature ofa given exchange. For example, in situations where the manufacturer fears that theintermediary may act opportunistically, the costs of transacting will increase intandem with the specificity of the assets being traded (Erramilli and Rao, 1993;Pilling et al., 1994). This happens because of the additional costs incurred in safe-guarding assets that have little value outside of particular exchange relationships( John and Weitz, 1988). In such situations the cost-minimizing properties ofvarious governance structures (organizations, markets, and hybrid arrangements),will be brought to bear to determine the most efficient means for structuring theexchange (Heide and John, 1992). For example, in a study of channel preferencesby US manufacturers, it was shown that for export settings involving complexproducts with a high service component, respondents tended to prefer integratedchannels to relying on external intermediaries (Anderson and Coughlan, 1987).In such cases, investments in distributor training represent a sunk cost that can notbe easily recovered in the event that the distributor shirks his or her responsibili-ties and needs to be replaced. Consequently, the additional bureaucratic and coor-dination costs of conducting the exchange within the firm’s boundaries may beless than the combined ex ante costs of negotiating stringent contracts to demar-cate the expected level of service and the ex post costs associated with monitoringand enforcing contractual obligations.

Provided the opportunistic tendencies of middlemen are kept in check by com-petitive product markets (Dwyer and Oh, 1988; Klein et al., 1990), the value-addedcontribution of the intermediary will be correlated with the uncertainty or riskinherent in the particular exchange setting. Generally export intermediaries willbe preferred in the early stages of opening new markets (Shin, 1989), for exchangeinvolving more distant markets (Klein and Roth, 1990), or when the manufacturerhas limited international experience (Anderson and Gatignon, 1986).

If TCA constitutes a valid explanation for the existence of ITIs, it follows thatthe performance of trade intermediaries will be directly related to their ability tominimize transaction costs for others. However, this question has rarely been askedin TCA research (Aulakh and Kotabe, 1997). Existing studies have generallysought to verify whether governance choices conform to the core TCA predictionsrelating to various exchange conditions with superior outcomes assumed to resultfrom more efficient matches (Shelanski and Klein, 1995). Peng’s (1998) survey of195 US ITIs is a recent exception to this practice. In this study the author hypoth-esized that trade intermediaries’ performance would reflect the presence of skillsdirectly relating to the four transaction costs identified above. The results showed

1688 P. D. Ellis


that intermediaries who were more knowledgeable about foreign markets (lower-ing their search costs), had superior negotiating abilities (lowering their negotia-tion costs), and were able to take title to the goods they traded (lowering theirclients’ monitoring and enforcement costs), generally recorded higher per capitasales and trading margins.

Summary

Agency theory and TCA offer powerful insights into ITI behaviour. Yet, both areinherently partial in their explanation for the existence of trade intermediaries.With its emphasis on the potential conflict residing within principal-agent rela-tions, agency theory makes no attempt to accommodate those trading activitieswhich involve reselling. In merchant trading situations where the middleman takestitle to the traded goods, there is no delegation of decision-making responsibilitiesto the intermediary. In such cases the ITI is effectively no different from a domes-tic customer and no agency relationship exists. TCA is less restrictive in scope buthas nevertheless been criticized for emphasizing efficiency considerations overvarious strategic objectives, such as those relating to competitive positioning, whichmay also affect governance decisions (Aulakh and Kotabe, 1997). Strategic vari-ables, along with transaction-specific characteristics, will influence the choice ofexchange mode (Hill et al., 1990), and in noncompetitive industries firms with slackresources may choose to retain control even if such choices create transactionalinefficiencies (Anderson and Gatignon, 1986).

Intermediaries exist because they promote and add value to mediated exchangerelationships. At the level of the individual exchange intermediaries create valueby reducing the transaction and agency costs within a channel (Casson, 1997).Improvements in transactional efficiencies can have a significant wealth-creatingeffect. Historical studies have shown that cumulative efficiency gains in distribu-tion can lead to an overall improvement in the allocative efficiency of the economywith a corresponding rise in per capita living standards (Porter and Livesay, 1971).In addition to improving the productivity of the channel however, intermediariesmay engage in innovation leading to a second source of value creation being real-ized. For ITIs innovation may be defined in terms of opening up new markets anddiscovering new sources of supply (Casson, 1997). Whenever a trade intermedi-ary develops a new value chain the result is a reorganization of available resourceswithin two or more local economies. This has implications for the future deploy-ment of resources in each economy affected (Ellis, 2003). Insofar as economicvalue is derived from the use or combination of resources (Penrose, 1959; Schum-peter, 1934), then value creating possibilities will be redefined every time exchangealters the set of resources available for successive rounds of combination (Moran,1996; Moran and Ghoshal, 1999). Consequently, the immediate effect of creatingnew value chains will be a more efficient reorganization of the local production



matrix. This in turn will stimulate new forms of exchange as traders become moti-vated to find markets for the additional outputs now being produced. In short,‘every change in the use of resources – every reorganization of productive activ-ities – creates the opportunity for a further change which would not have existedotherwise’ (Kaldor, 1972, p. 1245). This never-ending process of reorganization isparticularly evident in those sectors of the economy characterized by competition,elastic demand, and increasing returns to scale. When these three conditions arepresent, economic growth, triggered by the pursuit of new markets for industrialoutput, may become a self-reinforcing phenomenon (Young, 1928).

In summary, the value added contribution of ITIs can be defined two ways: (i)in terms of the finite efficiency gains brought to the channel; and (ii) in terms ofthe open-ended Schumpetarian innovation that results from creating something(trade) out of nothing (missing markets). TCA and agency theory are helpful inunderstanding the first kind of value creation but are irrelevant to the study of thesecond kind. From time to time trade intermediaries may play a catalytic role increating exchange relations where none previously existed, but neither agencytheory nor TCA make any attempt to account for this type of market-makingbehaviour. Rather, these existing theories seem more concerned with explainingwhy middlemen are inclined to act opportunistically some of the time than withexplaining how they manage to be entrepreneurial most of the time. Indeed, a widelyshared view in channel research is that trade intermediaries passively wait to be approached by manufacturers looking to reduce their exporting costs (e.g.,Anderson and Coughlan, 1987; Klein and Roth, 1990; Peng and Ilinitch, 1998).Yet there is extensive evidence to show that ITIs usually take the initiative inapproaching potential clients (e.g., Brasch, 1978; Haigh, 1994; Mattsson, 1990).Proactive searching is consistent with the definition adopted here that ITIs are spe-cialist market-making firms. Finally, TCA defines the criteria for survival, namelythe minimization of transaction costs for others, but offers no insight into how thiscan be achieved beyond making loosely-specified references to capturing scaleeconomies in distribution. Despite the validated claims of TCA, much remainsunknown about the actual ways and means by which ITIs manage to minimizesearch, negotiation and policing costs.

TOWARDS A NEW EXPLANATION

It is readily evident that differences in the quality of information among poten-tially transacting parties create opportunities for intermediaries. These infor-mation asymmetries may provide the basis for both negative behaviours (e.g.,opportunism) and positive behaviours (e.g., coordination of new exchange). Analternative approach to the study of ITIs therefore, is one which de-emphasizesthe costs associated with controlling the former, and highlights the value-addedcontribution of the latter (Ghoshal and Moran, 1996; Madhok, 1997; Nayyar,

1690 P. D. Ellis


1990). From this perspective the ITI is viewed as an information broker whosemain role is the coordination of information relating to the organization of trade(Casson, 1998; Etgar and Zusman, 1982).

The information handled by the intermediary may be of a routine nature, suchas that relating to the processing of clients’ orders, or it may be strategic in natureand relate to the identification of new opportunities to mediate exchange (Casson,1998). The handling of routine information is subject to economies of scale andthus provides the intermediary with an opportunity to improve the efficiency ofthe exchange. Organizational survival, however, ultimately depends on the man-agement of strategic information and this is where the intermediary must be mostentrepreneurial. Consequently, the remainder of this discussion will be concernedwith the issue of how ITIs collect and manage information which is germane totheir continued existence.

The Facts of Intermediary Life

Traders operate at the boundaries between groups. Initially their chief source ofcompetitive advantage stems from the knowledge gap separating potential buyersfrom sellers (Nayyar, 1990). Traders exploit these gaps by mediating exchangeflows between locations distinguished chiefly by their price differentials. As this isan inherently risky undertaking, traders will endeavour to develop market-relatedexpertise as a means of risk-minimization. This superior information gives themtheir ‘edge’ in the market and is the basis for their survival. However, the infor-mation asymmetries that exist between a particular buyer and seller and which theITI is able to exploit, generally diminish over time compelling the trader to con-tinually seek out new opportunities to broker relationships. Consequently, survivalof the intermediary will be a function of (a) prolonging the inevitable decline ofexisting exchange relationships and (b) replacing lost business with new relation-ships. When the addition of income from new business no longer matches the lossof income from exiting clients, the intermediary will have started down the pathto organizational extinction.

Any theory purporting to explain the endurance of the ITI must account forthe operational realities just identified. This implies an accommodation of the fol-lowing questions:

(1) How do ITIs identify and evaluate new exchange partners?(2) How do ITIs negotiate the terms of trade?(3) How do ITIs preserve exchange relationships in the face of decline?

If intermediaries can be considered to be entrepreneurial responses to marketimperfections separating buyers from sellers, then useful insights into these ques-



tions may be provided by social network theory which recognizes that knowledgeof entrepreneurial opportunities diffuses unevenly through society and that uniqueinformation benefits accrue to those individuals connected by non-redundantbridge ties to distant social clusters (Aldrich and Zimmer, 1986; Rogers andKincaid, 1981; Weiman, 1989). The relationship between social structure andentrepreneurial outcomes is neatly captured in Burt’s (1992, 1997, 2000) structuralhole theory which describes how social capital is a function of the brokerageopportunities inherent within social networks. The gist of Burt’s argument may beparaphrased as follows: Trading behaviour is explained in terms of access to ‘holes’in the social structure of the trading arena. This idea is distilled below.

Structural Hole Theory

In any market, imperfections separating buyers from sellers will lead to the emer-gence of entrepreneurial middlemen who match supply with demand. Wherepotential exchange parties are located in different countries, the most significantobstacles impeding direct exchange are initially those barriers which inhibit thefree flow of information across national boundaries. Although information willeventually spread to all people in a market, it will circulate within groups beforeit circulates between groups. Consequently, not everyone can be simultaneouslyaware of the opportunities in all groups and some individuals will be quicker toperceive market opportunities than others resulting in a short-term competitiveadvantage (Burt, 2000).

In pre-equilibrium markets opportunities may present themselves in the formof gaps or ‘holes’ between individuals possessing complementary resources orinformation (such as a rubber plantation in a developing country and a sneakerfactory in another). These holes represent unrealized market opportunities whichmay be exploited by a third individual who recognizes that the needs of one partymay be served by the skills or resources of another. The entrepreneur who per-ceives the opportunity to be the tertius gaudens, or the ‘third who profits from thedisunion of others’, is able to do so by virtue of his or her central position withina network of nonredundant contacts (Burt, 1992). It is the network – itself adynamic social construction existing between the player and the environment –which induces both the motive and the entrepreneurial opportunity to secure thebenefits of productive relationships in imperfect markets (Burt, 1992; Nahapietand Ghoshal, 1998).

Although Burt’s (1992) structural hole thesis has been advanced as a generaltheory ‘about the competition for the benefits of relationships’, the frameworkwould appear to have special relevance for understanding the behaviour of thosefirms whose sole raison d’être is the mediation of exchange relationships acrossnational boundaries. Two reasons support this. First, international trade interme-

1692 P. D. Ellis


diaries, by definition, operate on a social frontier between two or more countriesand as Burt (1992, p. 163) explains:

Hole effects are most evident for managers operating on a social frontier. Asocial frontier is any place where two worlds meet, where people of one kindmeet people of another kind. Individuals who live on a social frontier are morelikely to live by their entrepreneurial wits than are individuals in a sociallyhomogenous environment.

Second, Burt’s (1992, p. 9) claim that ‘social capital is the final arbiter of com-petitive success’ seems especially pertinent in a business characterized by very lowentry barriers and where performance is less a function of individual ability thanin creating value through contacts with other people. In other words, success fortrade intermediaries is less a question of financial and human capital than socialcapital which, in turn, is defined by the brokerage opportunities afforded by eachtrader’s network. Viewed from a structural hole perspective, ITIs exist, not byvirtue of their privy market knowledge per se, but by exploiting their unique tertius

position at the intersection of player relations.Trade intermediaries innovate by matching buyers and suppliers who otherwise

might have no contact with each other. The execution of this marketing functionis, however, structurally constrained by the information and control benefits inher-ent within each trader’s idiosyncratic network. Information benefits describe whogets to learn of new trading opportunities whereas control benefits determine who gets to mediate new relationships and for how long. From these underlyingpremises, four propositional statements can be derived. These relate to the identification and evaluation of new exchange partners, the negotiation of theterms of trade, and the maintenance or preservation of relationships already inexistence.

PROPOSITIONS

Identifying Exchange Partners

As markets mature, channels shorten (Sharma and Dominguez, 1992). Risingcompetition and improved information flows may lead to the middleman beingsqueezed out of the mediated exchange. Consequently, traders must continuallyidentify new opportunities to broker exchange relationships. Without the replen-ishment of lost income traders run the risk of being eliminated altogether.

Opportunities to broker international exchange may be identified via a numberof ways including advertisements in trade-related publications, attendance at tradefairs, market research, client referrals, and so on. In the search for new trading



partners it can be assumed that ITIs will tend to rely on those search behaviourswhich minimize the costs of information acquisition. Moreover, these costs will beunique to each trader reflecting the personal stock of social capital. If there aresignificant latent opportunities inherent in an individual’s social network thensearch patterns will tend to be based on personal sources of information. In suchcircumstances social interaction will be the main basis for the diffusion ofinformation. Conversely, if the individual does not possess substantial social capital,that is, network opportunities are comparatively small, then individual searchbehaviour will tend to be based on more impersonal sources of information.

Network constraint. The ability of an individual to participate in the diffusion ofinformation across national boundaries will be contingent on the value of thesocial capital inherent within that person’s network. Interpersonal networks serveas conduits for market information. Several studies have shown how personal tieswith extended family (Kotkin, 1992; Rauch, 1996), distribution channels (Hsing,1999), local government (Barrett, 1997), and even the military (Peng, 1998; casestudy #3) can provide traders with a valuable source of information regardingmarket opportunities. In their survey of Malagasy agricultural traders, Fafchampsand Minten (1999) found that respondents rated their relationships with othertraders, suppliers and clients as their most important sources of information onmarket conditions. More than 90 per cent of their information needs were metvia their trading networks suggesting that relationships ‘reduce search costs’ (p. 18).Social relations with others in the network may provide the intermediary withtimely access to resources. However, the extent to which the trader is able to par-ticipate in and control this process will be constrained by his or her network char-acteristics. In short, in the ongoing search for new trading opportunities sparse,diverse networks will be more efficient and less constraining than small dense net-works (Burt, 2000). Network size or reach determines the quantity of informationone is exposed to, whereas network diversity affects the quality of information(Aldrich and Zimmer, 1986; Birley et al., 1991).

Experience. While there are advantages to being connected to large, diverse net-works, such networks take time to develop suggesting a correlation with personalexperience in the market. Social capital accumulates over a career (Burt, 2000).Although this link has not yet been tested empirically, the available evidence tends to support the claim (Nijssen et al., 1999; Warms, 1990). For example, inBjörkman and Köck’s (1995) study of Western companies with investments in the Peoples’ Republic of China, interviewees with ‘long China experience’ ratedgood personal relations most highly.

Exchange uncertainty. Hole effects will be most noticeable where there are obstaclesto the flow of information between two groups. Such conditions arise when the

1694 P. D. Ellis


cultural distance separating buyer and seller is high and when market imper-fections are many. Cultural distance impedes information flow by adding semioticcomplexity to the encoding and decoding of messages relayed across marketboundaries. Market imperfections, such as a lack of contract-enforcing institutions,further obscure signals by rendering price an unreliable indicator of value. Wheninternational exchange is couched in uncertainty, the potential for opportunism –at any point in the channel – is magnified. In such an environment contract writing becomes impractical and actors will tend to rely on those potential part-ners about whom they have the greatest knowledge (Ganesan, 1994; Larson, 1992;Podolny, 1994). Fluctuations in price will compound these variables by introduc-ing a dynamic element of risk and creating opportunities for speculation (Casson,1998).

In combination, cultural distance, market imperfections and volatility interactto create substantial barriers to trade. Significant costs will be incurred in over-coming these barriers and therefore any intermediary with the social resources toreduce these costs will have the means to profitably broker exchange relationships.A classic example of this is provided by the Maghribi traders of North Africa. TheMaghribi traders were Jewish middlemen who operated in the Muslim Mediter-ranean during the eleventh century (Greif, 1989, 1992). The Maghribi traderscoped with the uncertainty and complexity of long-distance trade by relying onbusiness associates within a nonanonymous institution referred to as the coalition.Coalition members were simultaneously merchants with their own trading inter-ests and agents acting on behalf of other members. Relations within the coalitionwere governed by the understanding that no merchant would ever employ an agentwho had cheated another coalition member. Thus, membership in the coalitionlowered the costs of trade by reducing the premium needed to keep an agenthonest. Agency relations were preserved, not by the inadequate legal system of theday, but by the premium entailed in membership within the coalition (Greif, 1989).Evidence of the value of social capital within the coalition is found in the obser-vation that links between Maghribi and non-Maghribi traders, whether Jewish orMuslim, were rare and that the volume of trade was primarily limited by the coali-tion’s size (Greif, 1992).

Summary. The identification of new opportunities to mediate internationalexchange is affected by two variables: (1) the uncertainty separating would-beexchange parties; and (2) the social capital of the broker who would bring the twoparties together. Exchange uncertainty is a cumbersome label for what is really aninformation gap. The gap is an obstacle to the current identification of suitablepartners (who do I know in foreign markets?) with a shadow extending into thefuture (who can I trust in volatile markets?). The information gap or hole providesthe setting for a tertius-broker to mediate new exchange relationships. The abilityto do this, however, is defined not in terms of the broker’s personal attributes which



are merely correlates of entrepreneurial behaviour, but in terms of the informa-tion benefits inherent in the broker’s relationships with others (Burt, 1992).

This social capital explanation may be contrasted with the normative viewadopted in the export marketing literature which advocates a more formalapproach to partner identification based on the collection of objective informa-tion gathered systematically via market research (Douglas and Craig, 1983; Root,1994). Although this textbook model has been implicitly accepted as the receivedview in mainstream marketing for many years, several recent studies have revealedthat international exchange partners are frequently identified on the basis of priorsocial ties (Axelsson and Johanson, 1992; Bonaccorsi, 1992; Ellis, 2000; Wong andEllis, 2002). This emerging body of evidence complements the intuitive appeal ofthe structural hole theory and suggests the following proposition:

Proposition 1: The use of social ties as a means for identifying new opportuni-ties to mediate international exchange will be positively related to:

(a) the diversity and reach of the trader’s network;(b) the experience of the trader in the market;(c) the degree of uncertainty surrounding the exchange, which in turn will

reflect (i) the cultural distance and (ii) the extent of market imperfections separating the buyer and seller.

Evaluating Potential Exchange Partners

In identifying and evaluating market opportunities, traders may acquire informa-tion via ties embedded in networks of social relationships or via arm’s-length tieswith strangers. There are clear advantages to identifying exchange partners on thebasis of network ties (Egan and Mody, 1992; Larson, 1992). For example, if a newclient is introduced on the basis of a referral from an existing supplier, it does notserve the purposes of the existing supplier to pass along the name of someoneknown to be a risky proposition. No such qualitative screening is possible for poten-tial partners identified solely through impersonal means (e.g., in response to anadvertisement).

Trust. Embedded ties not only facilitate the screening of potential exchange part-ners, but they also promote the development of trust in new relationships. Theycan do this in several ways. First, for the reasons just discussed, social ties are morelikely than arms-length ties to lead to the a priori identification of trustworthy part-ners (Gulati, 1995; Lyon, 2000). Moreover, exchange partners referred by knownothers will be more inclined to share information without being asked to do so(Rauch, 1996). That is, they are more likely to engage in trust-creating behaviours.Further, new exchanges which are embedded within existing social relationshipswill come ‘primed’ with social resources such as uncertainty-reducing norms

1696 P. D. Ellis


regarding behavioural expectations and ‘an initial stock of trust appropriated froma preexisting social relationship’ (Uzzi, 1996, p. 680). In short, exchanges em-bedded in existing personal relationships are more likely to be characterized bytrust (a relationship variable) between partners who are themselves more likely to be trustworthy (an individual trait).

Commitment. Exchange partners identified on the basis of third party referrals andprevious personal relations are more likely to be credited as initially trustworthythan partners identified via anonymous sources (Uzzi, 1996). Trust has beendescribed in terms of confidence in, and a willingness to rely on, the exchangepartner (Moorman et al., 1993). Consequently, trust promotes commitment to arelationship (Morgan and Hunt, 1994; Moorman et al., 1992; Nielson, 1998). Itfollows then that traders will display greater commitment to new partners identi-fied on the basis of social ties than those identified under more fortuitous (e.g., achance meeting at a trade fair) or impersonal encounters.

Start-up speed. The contracting costs associated with arms-length transactions maybe reduced for exchanges based on embedded ties reflecting the added value oftacit information gleaned from known contacts. One implication is that em-bedded exchanges promote economies of time resulting in faster start-ups andrapid access to new opportunities emerging within the network. As Uzzi (1997)has observed, where partners are either known or vouched for by known others,transactional details can be negotiated ‘on the fly’ or even after the deal is com-pleted when time is of the essence.

Summary. In markets characterized by risk or rapidly closing windows of oppor-tunity, social ties provide unique advantages for evaluating potential exchange part-ners. Exchange partners identified on the basis of the recommendations of othersare likely to be more trustworthy than strangers and the resulting ventures willmore likely be characterized by higher levels of trust and commitment. In addi-tion, embedded ties may enable potential exchange parties to move quickly to capitalize on market opportunities without being waylaid by the need to install allthe requisite safeguards. In contrast, arms-length exchanges will require additionaltime to develop trust and prepare contracts which protect against opportunism.The preceding arguments can be expressed in proposition-form as follows:

Proposition 2: In contrast with arms-length exchanges, the use of social ties as ameans for identifying new opportunities to mediate international exchange willbe positively related to:

(a) the identification of exchange partners who are perceived to be more trustworthy;

(b) the trader’s resulting degree of commitment to the new venture;(c) quicker start-up times.



Negotiating Exchange

Information is one thing, control is another. While the identification of potentialexchange partners will be influenced by the information benefits of the trader’snetwork, the terms of trade agreed upon will reflect the control benefits or thedegree of autonomy enjoyed by the intermediary. Autonomy is concerned withthe replaceability of the intermediary; a trader who can be easily replaced has lowautonomy in the mediated exchange. A trader’s degree of autonomy will first beevident at the negotiation stage of the exchange process.

Structural autonomy. The greatest value in brokering connections across a structuralhole goes to the first few people to do it, when uncertainty is highest (Burt, 2000).This is the point when the two parties linked by the middleman are the least famil-iar with one another. In the absence of a common reference point, the mutually-linked intermediary has the freedom to exploit the ‘swirling mix of preferences’and to play the demands of one party against the other. The intermediary mayeven contribute to this uncertainty by strategically moving accurate, ambiguous,or distorted information between the two parties (Burt, 2000). When the uncer-tainty of exchange is at its highest, the tertius enjoys the greatest freedom to nego-tiate for favourable terms of trade. However, this opportunity diminishes in directproportion to the number of rivals or peers who could replace the intermediary.In the current context the term ‘peers’ would include other ITIs (industry rivalry),the exchange partners themselves (the threat of integration), as well as those banks,government agencies, and freight forwarders offering substitute intermediation ser-vices. Peers undermine the trader’s autonomy in a given exchange by providing acompetitive frame of reference against which claims for compensation may bejudged (Burt, 1997). In short, peers erode the value of the intermediary’s socialcapital by inhibiting the freedom of the trader to settle for anything other thanthe market rate.

Summary. In a perfectly competitive market players lack the freedom to negotiatethe terms of trade. But in an imperfect market multiple rates of return are pos-sible because disconnections between players leave some people unaware of the benefits they could offer others (Burt, 1999). The value of a trader’s social capitalin an imperfect market will be correlated with the degree of uncertainty separat-ing potential exchange partners. However, the trader’s latitude in exploiting thisuncertainty will be constrained by the presence of others against whom the trader’sterms can be assessed. This can be expressed as follows:

Proposition 3: The negotiated rate of return (or trading margin) will be positivelyrelated to the trader’s degree of structural autonomy in the mediated exchangewhich, in turn, will be inversely related to the number of peers (reflecting the

1698 P. D. Ellis


replaceability of the trader) and positively related to the exchange uncertaintyseparating buyers from sellers.

Maintaining Exchange

Ultimately mediated exchanges decay. As markets mature and competition inten-sifies information and resources will begin to flow more freely between producersand consumers undermining the role of the intermediary. For example, economicdevelopment led to a decline in the number of ITIs based in the San Franciscoarea from 1400 in the late 1940s to less than 100 by 1988 (Perry, 1992). At thedomestic and international levels, pressures to eliminate the middleman will invari-ably rise as market imperfections diminish.

Within specific exchange relationships intermediaries may find themselvescaught on the horns of a dilemma. If they perform their marketing function well,generating substantial sales or orders for their clients, they may inadvertentlyprompt either the buyer or seller to internalize those activities performed by thetrader. For example, an ITI that is successful in developing a new export marketruns the risk of being eliminated from the channel by a manufacturer keen to reapthe benefits of more direct entry modes. Once the ‘liability of newness’ has beenovercome, the manufacturer may stand to gain more by eliminating the middle-man. On the other hand, if intermediaries perform poorly they run the risk ofbeing replaced, either by another trader or by one of the exchange parties.

Given time, competitive threats to the existence of trade intermediaries willeventually emerge irrespective of, and possibly because of, their level of perfor-mance in the indirect export channel. These threats originate from two sources:other middlemen (or organizations offering substitute services, e.g., banks) and theexchange parties themselves. For these reasons the traders interviewed by Perry(1992) estimated that few in their business survive as long as ten years.

While scholars (e.g., Hennart and Kryda, 1998; Peng, 1998) have long recog-nized the ‘built-in mortality’ of mediated exchange relationships, no satisfactoryexplanation of the process of decline has yet been developed. Rather, within theinternational business literature the main emphasis has been on those relationalvariables, such as commitment, which are presumed to forestall the inevitable (see,for example, Barovick and Anderson, 1992; Castaldi et al., 1992). However, com-mitment to one’s clients, while commendable, will not of itself sustain the ITIsrole in the mediated exchange. Rather, the duration of any given trading rela-tionship will be dependent upon the control benefits inherent in the particulartertius strategy.

Switching incentives. Middlemen are eliminated when they are replaced either byother middlemen or by exchange parties assuming responsibility for the middle-man function. It follows then that an intermediary’s position in a cross-border



exchange will persist for just as long as the incentives for preserving or main-taining the arrangement counteract the pressures for elimination. Pressures forelimination, or ‘switching incentives’ if viewed from the perspective of the manufacturer-supplier, have been described in terms of (i) dissatisfaction with the intermediary’s current level of performance, (ii) the accumulation of marketknowledge, (iii) export market growth, and (iv) growth of the company (Benito et al., 1999). The first switching incentive will result in the ITI being replaced by another intermediary, while the other incentives will stimulate integration of channel functions by the supplier.

Switching costs. Fully aware of the pressures for elimination posed by both rivalsand other channel members, intermediaries can be expected to engage in depen-dence-balancing actions designed to preserve their role in the mediated exchange.One strategy is to make relationship-specific investments and thus create exit barriers for clients. Investments may be supplier-specific or customer-specific.Supplier-specific investments would include the effort expended by the interme-diary to gain a unique understanding of how the supplier operates in order to beable to promote the supplier’s products more effectively (Weiss and Kurland, 1997).Such investments make the intermediary more valuable to the supplier as any deci-sion to terminate the agency relationship must now be offset against the additionalcosts incurred in replacing this enhanced level of service (Anderson and Schmittlein, 1984). Conversely, customer-specific investments may take the formof building close personal relationships with buyers (Ganesan, 1994; Weiss andKurland, 1997). Suppliers will hesitate to terminate agency agreements if there isa prospect of losing key customers with whom the intermediary has formed a close bond (Heide and John, 1988). The anticipation of high switching costs asso-ciated with such barriers may result in clients’ desire to maintain the relationship(Lewin and Johnston, 1997).

Estimating duration. Mediated exchange relationships terminate altogether when the benefits of integration outweigh the costs of doing so. Assuming the use ofITIs in international exchange as the default hypothesis (a lá TCA), implies thatany change in the cost/benefit ratio of the current channel arrangement must beassessed in light of the conditions present at the initiation of the mediatedexchange. This can be done by imagining an X–Y graph where movements awayfrom the Y-axis reflect changes in the underlying costs of the exchange and move-ments away from the X-axis reflect changes in the underlying benefits. Again,adopting the perspective of the supplier, the exchange conditions at time t = 0 are described by the point (0,0) on the graph. The costs of altering the existingarrangement can be expected to fall (move to the left of the Y-axis) as the supplieracquires either direct knowledge of the foreign market or slack resources withwhich it may purchase such information. Counteracting this are transaction-

1700 P. D. Ellis


specific investments made by the intermediary which increase the costs of alteringthe existing exchange structure (reflected in movements to the right of the Y-axis).On the benefit side of the equation, growth in the export market will create anincentive for the supplier to integrate the intermediary functions (as reflected in amovement upwards from the X-axis). Conversely, the emergence of additionaluncertainty in the market (captured in a movement downwards from the X-axis)will reinforce the intermediary’s position in the exchange.

Taking these factors together and assuming that, all things being equal, thestatus quo will prevail, the duration of any mediated exchange can be expressedin the following equation:

This simple ‘duration equation’ is useful for discriminating between those variableswhich are salient in determining relationship duration and those which have indirect effects. For example, it could be tempting to infer a negative correlationbetween relationship duration and the competitive intensity generated by the pres-ence of industry rivalry and substitute intermediary services. However, the numberof peers who are able to replace the intermediary in a given exchange is merelycorrelated (inversely) with the degree of ambiguity regarding the ITI’s level ofperformance. Peers provide a competitive frame of reference against which thedelivered effort may be gauged (Burt, 1992). If the ITI is executing its marketingtasks at a competitive price, the existing arrangement will remain unchanged.

What if the ITI is performing well above the supplier’s expectations? A highlevel of intermediary performance, resulting perhaps from growth in the exportmarket, may indeed stimulate the supplier to investigate the possible benefits of integration. However, again the status quo will not be challenged unless thesubsequent accumulation of market knowledge or slack resources causes Calt to fall below the t = 0 baseline.

Finally, the duration equation can be used to reconcile contradictory hypothe-ses logically derived from the axioms of TCA. For example, high uncertainty in amarket can and has been correlated with both market-based ( Jones et al., 1997) and hierarchical modes of governance (Klein et al., 1990). Both arguments arepersuasive. On the one hand, in situations of high demand uncertainty, decou-pling of activities via subcontracting or outsourcing increases organizational flex-ibility improving the firm’s ability to respond to a wide range of unforeseencontingencies ( Jones et al., 1997). On the other hand, uncertainty in foreignmarkets impedes forecasting making it difficult to write comprehensive contracts.Therefore, high external uncertainty will lead to high transaction costs encourag-ing high levels of integration to reduce such costs and to improve adaptive sequen-tial decision-making in the market ( John and Weitz, 1988; Klein et al., 1990).

IfC

CB

Bthen maintain independent channels, otherwise integrate.

alt alt[1]£



Reconciling these divergent claims is not difficult when one realizes that uncer-tainty, by itself, can say little about the suitability of particular exchange modes.Uncertainty simply refers to the cost of acquiring information which must be offsetagainst the value assigned to that information by the purchaser. In the case ofinternational exchange a supplier may assign a high value to information relatingto a fast-growing market and opt for a direct entry mode for all the reasons iden-tified by Klein et al. (1990). Conversely, a less desirable (but equally uncertain)market may only warrant indirect exports, as per Jones et al. (1997).

Summary. Whether traders perform the marketing function poorly or well, they runthe risk of being eliminated from the indirect export channel. The duration of anyparticular mediated exchange will be circumscribed by this Catch-22 situation. Thespecific point of relationship termination however, will be found in the com-plex interplay of a variety of switching incentives and costs as perceived by theexchange parties. In essence, ITIs will not be replaced until the pressures for elimination outweigh the pressures for preservation. This can be expressed inpropositional form as follows:

Proposition 4: Mediated exchange relationships will end when the pressures foreliminating the intermediary outweigh the pressures for preserving the relationship:

(a) Pressures for elimination will be positively related to (i) the number ofpeers who could replace the trader when clients perceive the trader to be performing poorly and (ii) the perceived gains from integration when clients perceive the trader to be performing well.

(b) Pressures for preserving the relationship will be positively related to (i) the presence of relationship-specific investments made by the trader and (ii) the perceived degree of exchange uncertainty.

CONCLUSIONS

Despite their historical and contemporary significance in cross-border trade, ITIsremain outside the scope of mainstream organizational research. Even within theinterdisciplinary domain of international business these highly adaptable servicefirms have been largely overlooked in deference to the multinational enterprise.One possible explanation for this state of affairs is that although ITIs are conceptually fascinating at many levels, very few generic models of ITI behaviourexist. Much of the available research is highly particularistic in focus with little rel-evance beyond the country in which the study was conducted. Recently this situ-ation has begun to change. TCA scholars are now beginning to realize that becauseITIs operate largely by minimizing transaction costs for others, they constitute an ideal setting for investigating the hitherto ignored link between governance

1702 P. D. Ellis


structures and performance (Peng, 1998). In addition, respected economists andbusiness historians are turning their powerful analytical lenses towards ITIs (seefor example, Casson (1998) and Jones (1996) respectively). While such trends augur well for the future of ITI research, fundamental questions regarding theentrepreneurial role of traders persist: How do traders create new value chains?How do they make markets? What measures do they enact to sustain their place inmediated exchanges?

Market-making intermediaries such as ITIs are primarily vehicles for exchange.The limitations of agency theory and TCA in explaining these organizationalforms stem from the fact that exchange is a source of at least two kinds ofvalue creation. First, value is created whenever exchange improves the allocativeefficiency of the economy. This exchange process benefits from the writing ofcontracts that ensure goal congruency between principals and agents and from the appropriate matching of exchange modes with exchange characteristics.Thus, both agency theory and TCA are concerned with this first type of valuecreation. But ITIs also engage in a second, more innovative type of value creationin the form of stimulating new trade flows. In contrast with the finite gains from trade achievable by improving the efficiency of trade (doing the same thingbetter), this type of value creation, which is captured in the persistent search for new markets, is potentially limitless in scope (Young, 1928). It is in the organization and management of this second type of trade that ITIs must be mostentrepreneurial.

ITIs as entrepreneurs are a diverse breed and trading entities from one part ofthe world may have little in common with those elsewhere. Structural hole theoryoffers a neat way of sidestepping this issue by explaining entrepreneurship, not interms of individual or organizational traits, but in terms of the competition forthe benefits of productive relationships (Burt, 1992). Every trade intermediary has social capital, but the landscape of opportunity – in effect, the size of the structural hole being spanned – determines the value of that social capital. Theintermediary as entrepreneur creates value by participating in, and controlling,the diffusion of market information across national boundaries. The ability to dothis will be influenced by the information and control benefits inherent within hisor her social network. Information benefits come into play at the initial stages ofexchange creation whereas control benefits will influence the rate of return andthe duration of the mediated exchange.

NOTES

*Earlier versions of this paper were read by Gabriel Benito, Ron Burt, and Mark Casson. The authoris grateful for the constructive feedback provided. However, the usual caveat applies. This projectwas supported by the Research Grants Council of the HKSAR (Project no. PolyU 5234/99H).[1] C and B denote the costs and benefits of using an intermediary, whereas Calt and Balt denote the

costs and benefits of performing intermediary functions in-house.



REFERENCES

Aldrich, H. and Zimmer, C. (1986). ‘Entrepreneurship through social networks.’ In Sexton, D. L.and Smilor, R. W. (Eds), The Art and Science of Entrepreneurship. Cambridge, MA: Ballinger, 2–23.

Anderson, E. and Coughlan, A. T. (1987). ‘International market entry and expansion via inde-pendent or integrated channels of distribution’. Journal of Marketing, 51, January, 71–82.

Anderson, E. and Gatignon, H. (1986). ‘Modes of foreign entry: a transaction cost analysis andpropositions’. Journal of International Business Studies, Fall, 1–26.

Anderson, E. and Schmittlein, D. (1984). ‘Integration of the sales force: an empirical examination’.Rand Journal of Economics, 15, Autumn, 385–95.

Anwar, S. T. and Alex, R. (1999). ‘Market Behavior and Performance Patterns of Asia’s General andSpecialized Trading Companies: A Comparative Study’. Paper presented at the Academy ofInternational Business annual meeting, Charleston, SC.

Armine, L. S. (1987). ‘Toward a conceptualization of export trading companies in world markets’.Advances in International Marketing, 2, 1, 199–238.

Armine, L. S., Cavusgil, S. T. and Weinstein, R. I. (1986). ‘Japanese sogo shosha and the US exporttrading companies’. Journal of the Academy of Marketing Science, 14, 3, 21–32.

Aulakh, P. S. and Kotabe, M. (1997). ‘Antecedents and performance implications of channel integration in foreign markets’. Journal of International Business Studies, First Quarter, 145–75.

Axelsson, B. and Johanson, J. (1992). ‘Foreign market entry – the textbook vs. the network view’. InAxelsson, B. and Easton, G. (Eds), Industrial Networks: A New View of Reality. London: Routledge,218–34.

Balabanis, G. (1998). ‘Antecedents of cooperation, conflict and relationship longevity in an interna-tional trade intermediary’s supply chain’. Journal of Global Marketing, 12, 2, 25–46.

Balabanis, G. (2000). ‘Factors affecting export intermediaries’ service offerings: the British example’.Journal of International Business Studies, 31, 1, 83–99.

Balabanis, G. and Baker, M. (1993a). ‘Barriers to the development of a European general tradingcompany’. Journal of Euromarketing, 2, 3, 45–72.

Balabanis, G. I. and Baker, M. J. (1993b). ‘Determinants and direction of trading companies’ devel-opment in Europe’. European Journal of Management, 27, 9, 58–76.

Barovick, R. and Anderson, P. (1992). ‘EMCs/ETCs: what they are, how they work’. Business America,13 July, 2–5.

Barrett, C. B. (1997). ‘Food marketing liberalization and trader entry: evidence from Madagascar’.World Development, 25, 5, 763–77.

Bello, D. C. and Williamson, N. C. (1985). ‘Contractual arrangement and marketing practices in theindirect export channel’. Journal of International Business Studies, 16, 2, 65–82.

Benito, G., Pedersen, T. and Petersen, B. (1999). ‘Export Channel Dynamics: An Empirical Analy-sis of Changes in the Organization of Foreign Distribution’. Paper presented at the annualmeeting of the European International Business Academy, Manchester.

Birley, S., Cromie, S. and Myers, A. (1991). ‘Entrepreneurial networks: their emergence in Irelandand overseas’. International Small Business Journal, 9, 4, 56–74.

Björkman, I. and Kock, S. (1995). ‘Social relationships and business networks: the case of Westerncompanies in China’. International Business Review, 4, 4, 519–35.

Bonaccorsi, A. (1992). ‘On the relationship between firm size and export intensity’. Journal of Inter-national Business Studies, 23, 4, 605–35.

Brasch, J. J. (1978). ‘Export management companies’. Journal of International Business Studies, 9, 2,59–71.

Buchan, P. B. (1994). ‘The East India Company 1749–1800: the evolution of a territorial strategyand the changing role of directors’. Business and Economic History, 23, 1, 52–61.

Burt, R. S. (1992). Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard Uni-versity Press.

Burt, R. S. (1997). ‘The contingent value of social capital’. Administrative Science Quarterly, 42,339–65.

Burt, R. S. (1999). ‘Entrepreneurs, distrust, and third parties: a strategic look at the dark side ofdense networks’. In Thompson, L., Levine, J. and Messick, D. (Eds), Shared Cognition in Organi-zations. Hillsdale, NJ: Lawrence Erlbaum, 213–43.

Burt, R. S. (2000). ‘The network structure of social capital’. In Sutton, R. I. and Staw, B. M. (Eds),Research in Organizational Behavior. Greenwich, CT: JAI Press.

1704 P. D. Ellis


Carlos, A. M. (1992). ‘Principal-agent problems in early trading companies: a tale of two firms’. AEAPapers and Proceedings, 82, 2, 140–5.

Carlos, A. M. and Nicholas, S. (1988). ‘Giants of an earlier capitalism: the chartered trading com-panies as modern multinationals’. Business History Review, 62, Autumn, 398–419.

Casson, M. (1997). Information and Organization: A New Perspective on the Theory of the Firm. Oxford:Clarendon Press.

Casson, M. (1998). ‘The economic analysis of multinational trading companies’. In Jones, G. (Ed.),The Multinational Traders. London: Routledge, 22–47.

Castaldi, R. M., De Noble, A. and Kantor, J. (1992). ‘The intermediary service requirements ofCanadian and American exporters’. International Marketing Review, 9, 2, 21–40.

Cho, D. S. (1984). ‘The anatomy of the Korean general trading company’. Journal of Business Research,12, 241–55.

Cho, D. S. (1987). The General Trading Company. Lexington, MA: D.C. Heath.Coase, R. H. (1937). ‘The nature of the firm’. Economica, 4, November, 386–405.Czinkota, M. R. and Onto, J. G. (1990). ‘Export trading companies: a promise yet unfulfilled’. Devel-

opments in Marketing Science, 13, 135–9.Douglas, S. P. and Craig, C. S. (1983). International Marketing Research. Englewood Cliffs, NJ: Prentice

Hall.Dwyer, F. and Oh, S. (1988). ‘A transaction cost perspective on vertical contractual structure and

interchannel competitive strategies’. Journal of Marketing, 52, April, 21–34.Dyer, J. H. (1997). ‘Effective interfirm collaboration: how firms minimize their transaction costs and

maximize transaction value’. Strategic Management Journal, 18, 7, 535–56.Egan, M. L. and Mody, A. (1992). ‘Buyer-seller links in export development’. World Development, 20,

3, 321–34.Eisenhardt, K. M. (1989). ‘Agency theory: an assessment and review’. Academy of Management Review,

14, 1, 57–74.Ellis, P. (1998). ‘Evolution of a Multinational Trading Corporation: The Li & Fung Story’.

Paper presented at the Europe International Business Academy annual meeting, Jerusalem,Israel.

Ellis, P. (2000). ‘Social ties and foreign market entry’. Journal of International Business Studies, 31, 3,1–27.

Ellis, P. (2001). ‘Adaptive strategies of trading companies’. International Business Review, 10, 235–59.Ellis, P. (2003). ‘Are international trade intermediaries catalysts in economic development? A new

research agenda’. Journal of International Marketing, 11, 1, 73–96.Erramilli, M. K. and Rao, C. P. (1993). ‘Service firms’ international entry-mode choice: a modified

transaction-cost analysis approach’. Journal of Marketing, 57, July, 19–38.Etgar, M. and Zusman, P. (1982). ‘The marketing intermediary as an information seller: a new

approach’. Journal of Business, 55, 4, 505–15.Fafchamps, M. and Minten, B. (1999). ‘Relationships and traders in Madagascar’. Journal of Devel-

opment Studies, 35, 6, 1–35.Ganesan, S. (1994). ‘Determinants of long-term orientation in buyer-seller relationships’. Journal of

Marketing, 58, April, 1–19.Ghoshal, S. and Moran, P. (1996). ‘Bad for practice: a critique of the transaction cost theory’. Academy

of Management Review, 21, 1, 13–47.Greif, A. (1989). ‘Reputation and coalitions in medieval trade: evidence on the Maghribi traders’.

Journal of Economic History, 49, December, 857–82.Greif, A. (1992). ‘Institutions and international trade: lessons from the commercial revolution’. AEA

Papers and Proceedings, 82, 2, 128–33.Gulati, R. (1995). ‘Social structure and alliance formation patterns: a longitudinal analysis’. Admin-

istrative Science Quarterly, 40, December, 619–52.Haigh, R. W. (1994). ‘Thinking of exporting? Export management companies could be the answer’.

Columbia Journal of World Business, Winter, 66–81.Heide, J. B. (1994). ‘Interorganizational governance in marketing channels’. Journal of Marketing, 58,

January, 71–85.Heide, J. B. and John, G. (1988). ‘The role of dependence balancing in safeguarding transaction-

specific assets in conventional channels’. Journal of Marketing, 52, January, 20–35.Heide, J. B. and John, G. (1992). ‘Do norms matter in marketing relationships?’. Journal of Market-

ing, 56, April, 32–44.



Hennart, J. F. and Kryda, G. M. (1998). ‘Why do traders invest in manufacturing?’. In Jones, G.(Ed.), The Multinational Traders. London: Routledge, 213–27.

Hill, C., Hwang, P. and Kim, W. C. (1990). ‘An eclectic theory of the choice of international entrymode’. Strategic Management Journal, 11, 117–128.

Howard, D. G. and Maskulka, J. M. (1988). ‘Will American export trading companies replace tra-ditional export management companies?’. International Marketing Review, 5, 4, 41–50.

Hsing, Y. T. (1999). ‘Trading companies in Taiwan’s fashion shoe networks’. Journal of InternationalEconomics, 48, 101–20.

Jensen, M. C. and Meckling, W. H. (1976). ‘Theory of the firm: managerial behavior, agency costsand ownership structure’. Journal of Financial Economics, 3, 305–60.

John, G. and Weitz, B. A. (1988). ‘Forward integration into distribution: an empirical test of trans-action cost analysis’. Journal of Law, Economics, and Organization, 4, 2, 337–55.

Jones, G. (1996). ‘Diversification strategies and corporate governance in trading companies: Anglo-Japanese comparisons since the late nineteenth century’. Business and Economic History, 25, 2,103–18.

Jones, G. (1998a). ‘Multinational trading companies in history and theory’. In Jones, G. (Ed.), TheMultinational Traders. London: Routledge, 1–21.

Jones, G. (Ed.) (1998b). The Multinational Traders. London: Routledge.Jones, C., Hesterly, W. S. and Borgatti, S. P. (1997). ‘A general theory of network governance:

exchange conditions and social mechanisms’. Academy of Management Review, 22, 4, 911–45.Kaikati, J. (1984). ‘The Export Trading Company Act: a viable international marketing tool’.

California Management Review, 27, 1, 59–70.Kaldor, N. (1972). ‘The irrelevance of equilibrium economics’. Economic Journal, 82, December,

1237–55.Kelly, E. R. and Lecraw, D. (1985). ‘Trading houses to spur Canadian exports’. Business Quarterly, 50,

1, 36–43.Kim, W. C. (1986). ‘Global diffusion of the general trading company concept’. Sloan Management

Review, Summer, 35–43.Klein, S. and Roth, V. (1990). ‘Determinants of export channel structure: the effects of experience

and psychic distance reconsidered’. International Marketing Review, 7, 5, 27–38.Klein, S., Frazier, G. and Roth, V. (1990). ‘A transaction cost analysis model of channel integration

in international markets’. Journal of Marketing Research, 26, May, 196–208.Kojima, K. and Ozawa, T. (1984). Japan’s General Trading Companies: Merchants of Economic Development.

Paris: OECD.Kotkin, J. (1992). Tribes: How Race, Religion, and Identity Determine Success in the New Global Economy. New

York: Random House.Larson, A. (1992). ‘Network dyads in entrepreneurial settings: a study of the governance of exchange

relationships’. Administrative Science Quarterly, 37, March, 76–104.Lassar, W. M. and Kerr, J. L. (1996). ‘Strategy and control in supplier-distributor relationships: an

agency perspective’. Strategic Management Journal, 17, 613–32.Lewin, J. E. and Johnston, W. J. (1997). ‘Relationship marketing theory in practice: a case study’.

Journal of Business Research, 39, 23–31.Long, D. and Seet, R. (1995). Li and Fung (A). HBS case study #9-396-107, Boston, MA: HBS

Publishing.Lyon, F. (2000). ‘Trust, networks and norms: the creation of social capital in agricultural economies

in Ghana’. World Development, 28, 4, 663–81.MacBean, A. (1996). ‘China’s foreign trade corporations: their role in economic reform and export

success’. In Child, J. and Lu, Y. (Eds), Management Issues in China: Volume II. London: Routledge,183–200.

Madhok, A. (1997). ‘Cost, value and foreign market entry mode: the transaction and the firm’.Strategic Management Journal, 18, 39–61.

Mattsson, J. (1990). ‘Trading companies and small and medium-sized firms: functional roles in inter-national commercial relations’. European Journal of Marketing, 24, 3, 42–56.

Moran, P. (1996). ‘Value creation by firms’. Academy of Management Proceedings, 41–5.Moran, P. and Ghoshal, S. (1999). ‘Markets, firms, and the process of economic development’.

Academy of Management Review, 24, 3, 390–412.Morgan, R. M. and Hunt, S. D. (1994). ‘The commitment-trust theory of relationship marketing’.

Journal of Marketing, 58, July, 20–38.

1706 P. D. Ellis


Moorman, C., Zaltman, G. and Deshpande, R. (1992). ‘Relationships between providers and usersof market research: the dynamics of trust within and between organizations’. Journal of Mar-keting Research, 29, August, 314–28.

Moorman, C., Deshpande, R. and Zaltman, G. (1993). ‘Factors affecting trust in market researchrelationships’. Journal of Marketing, 57, January, 81–101.

Munro, J. F. (1998). ‘From regional trade to global shipping: Mackinnon Mackenzie & Co withinthe Mackinnon Enterprise Network’. In Jones, G. (Ed.), The Multinational Traders. London:Routledge, 48–65.

Nahapiet, J. and Ghoshal, S. (1998). ‘Social capital, intellectual capital, and the organizational advantage’. Academy of Management Review, 23, 2, 242–66.

Nayyar, P. R. (1990). ‘Information asymmetries: a source of competitive advantage for diversifiedservice firms’. Strategic Management Journal, 11, 513–19.

Nielson, C. C. (1998). ‘An empirical examination of the role of “closeness” in industrial buyer-sellerrelationships’. European Journal of Marketing, 32, 5/6, 441–63.

Nijssen, E., Douglas, S. and Calis, G. (1999). ‘Gathering and using information for the selection oftrading partners’. European Journal of Marketing, 33, 1/2, 143–62.

O’Connell, J. (1995). Li and Fung (Trading) Ltd, HBS case study #9-396-075, Boston, MA: HBSPublishing.

Peng, M. W. (1998). Behind the Success and Failure of US Export Intermediaries. Westport, CT:Quorum.

Peng, M. W. and Ilinitch, A. Y. (1998). ‘Export intermediary firms: a note on export developmentresearch’. Journal of International Business Studies, 29, 3, 609–20.

Penrose, E. T. (1959). The Theory of the Growth of the Firm. New York: Wiley.Perry, A. C. (1990). ‘The evolution of the US international trade intermediary in the 1980s: a

dynamic model’. Journal of International Business Studies, First Quarter, 133–53.Perry, A. C. (1992). ‘US international trade intermediaries: a field study investigation’. International

Marketing Review, 9, 2, 7–20.Pilling, B. K., Crosby, L. A. and Jackson, D. W. (1994). ‘Relational bonds in industrial exchange:

an experimental test of the transaction cost economic framework’. Journal of Business Research,30, 237–51.

Podolny, J. (1994). ‘Market uncertainty and the social character of economic exchange’. Administra-tive Science Quarterly, 39, September, 458–83.

Porter, G. and Livesay, H. C. (1971). Merchants and Manufacturers: Studies in the Changing Structure of Nine-teenth Century Marketing. Baltimore, MD: Johns Hopkins Press.

Rauch, J. E. (1996). ‘Trade and Search: Social Capital, Sogo Shosha, and Spillovers’. NBER WorkingPaper 5618, Cambridge, MA: National Bureau of Economic Research.

Rindfleisch, A. and Heide, J. B. (1997). ‘Transaction cost analysis: past, present, and future applica-tions’. Journal of Marketing, 61, October, 30–54.

Roehl, T. (1983). ‘A transactions cost approach to international trading structures: the case of theJapanese general trading companies’. Hitosubashi Journal of Economics, 24, 119–35.

Rogers, E. M. and Kincaid, D. L. (1981). Communication Networks: Toward a New Paradigm for Research.New York: Free Press.

Root, F. R. (1994). Entry Strategies for International Markets. New York: Lexington Books.Sarathy, R. (1985). ‘Japanese trading companies: can they be copied?’. Journal of International Business

Studies, Summer, 101–19.Schumpeter, J. A. (1934). The Theory of Economic Development. Cambridge, MA: Harvard University

Press.Shao, A. T. and Herbig, P. (1993). ‘The future of sogo shosha in a global economy’. International

Marketing Review, 10, 5, 37–55.Sharma, A. (1997). ‘Professional as agent: knowledge asymmetry in agency exchange’. Academy of

Management Review, 22, 3, 758–98.Sharma, A. and Dominguez, L. V. (1992). ‘Channel evolution: a framework for analysis’. Journal of

the Academy of Marketing Science, 20, 1, 1–15.Shelanski, H. and Klein, P. G. (1995). ‘Empirical research in transaction cost economics: a review

and assessment’. Journal of Law, Economics, and Organization, 11, 2, 335–61.Shin, K. S. (1989). ‘Information, transaction costs, and the organization of distribution: the case

of Japan’s general trading companies’. Journal of the Japanese and International Economies, 3,292–307.



Sugiyama, S. (1987). ‘A British trading firm in the Far East: John Swire and Sons, 1867–1914’. InYonekawa, S. and Yoshihara, H. (Eds), Business History of General Trading Companies. Japan: Uni-versity of Tokyo Press, 171–202.

Uzzi, B. (1996). ‘The sources and consequences of embeddedness for the economic performance oforganizations: the network effect’. American Sociological Review, 61, August, 674–98.

Uzzi, B. (1997). ‘Social structure and competition in interfirm networks: the paradox of embed-dedness’. Administrative Science Quarterly, 42, March, 35–67.

Warms, R. L. (1990). ‘Who are the merchants? Ethnic identity and trade in southwestern Mali’.Ethnic Groups, 8, 1, 57–72.

Weiman, G. (1989). ‘Social networks and communication’. In Asante, M. K. and Gudykunst, W. B.(Eds), Handbook of International and Intercultural Communication. Newbury Park, CA: Sage, 186–203.

Weiss, A. M. and Kurland, N. (1997). ‘Holding distribution channel relationships together: the roleof transaction-specific assets’. Organization Science, 8, 6, 612–23.

Williams, H. R. and Baliga, G. M. (1983). ‘The US Export Trading Company Act of 1982’. Journalof World Trade Law, 17, 224–35.

Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. New York: Free Press.Williamson, O. E. (1979). ‘Transaction-cost economics: the governance of contractual relations’. The

Journal of Law and Economics, 22, 233–61.Williamson, O. E. (1985). The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting.

New York: Free Press.Williamson, O. E. (1996). The Mechanisms of Governance. New York: Oxford University Press.Wong, P. and Ellis, P. (2002). ‘Social ties and partner identification in Sino-Hong Kong International

Joint ventures’. Journal of International Business Studies, 33, 2, 267–89.Yoshihara, K. (1982). Sogo Shosha: The Vanguard of the Japanese Economy. Oxford: Oxford University

Press.Young, A. A. (1928). ‘Increasing returns and economic progress’. Economic Journal, 38, 152, 527–42.

1708 P. D. Ellis


Documents

Social Structure and Intermediation: Market-making ... · PDF fileSocial Structure and Intermediation: Market-making Strategies in International Exchange* Paul D. Ellis Hong Kong Polytechnic