GLOBAL FIRMS IN INTERNATIONALIZED NETWORKS
Ulf Andersson, Mats Forsgren, Cecilia Pahlberg and Peter Thilenius
Abstract
This paper is part of a research pmject with its main focus on Swedish multinational ccapanies. The purpose of the pmject is to study the tension between parent cmpanies in Seden and a n&er of subsidiaries. chicfly located in Eumpe. Our sims an? ta find what constitutes the strwgth of the subsidiaries. and what possibilities do they have to influence the corporate strategy. The presented model outlines the backgro& for an eopirical study which Will be conducted during 1991. The "K&l rewgnizes feur groups of variables, all of wbich reflect different aspects of a subsidiary‘s relative base of pcwx.
The pm_iect is partly financed by Tbc Swedish Gwncil for Hanagenent and Work Life Issues
Background
Most literature on international business presents an hierar-chic perspective, where the corporate executive committee controls the subsidiaries and makes the main strategic decisions. The internationalization of firms is based on the gradual exploitation of a firm-specific asset, controlled by the corporate executive committee. Investments are in these theories thought to emanate from, and be dependent on, the corporate executive committee, both technologically and in questions about knowledge.
Several studies seek to find the optimal way to structure an internationalized firm. According to these studies, organizational and strategic issues in highly internationalized firms always have a pronounced corporate executive committee perspective. Such is-sues cancern questions about the subsidiary's allowed degree of autonomy and how the total global structure of the firm should be developed.
The underlying assumption in these studies is the corporate ex-ecutive committee's ability to control and integrate a number of subsidiaries in other countries. The subsidiaries in this perspec-tive are acperspec-tive mostly at a lotal level. The main influence over the development of the firm is, according to this picture, in the hands of the corporate executive committee at a strategic
ment level. Empirital findings during the 70's and 80's indicate however that a substantial proportion of foreign investments were initiated and accomplished by subsidiaries. The investments were based on activities in the subsidiaries rather than in the corpo-rate executive committee.
There is, thus, some evidente suggesting that the hierarchical top management picture outlined above should be replaced by a more complex one, with a number of geographically dispersed, more or less mutually dependent power centres, which beside the corporate executive committee exert considerable influence over the develop-ment of the total firm. The former periphery of subsidiaries has developed into significant centres for investments, activities and influence, motivated by the economic power in the industrial structures to which the subsidiaries belong. This is most impor-tant for firms which are highly internationalized and have a large share of their business and staff outside the parent Company. (Forsgren, et al. 1990)
Our basic assumption is, thus, that the fundamental conflict between the subsidiaries' need for freedom of action and possibi-lities of lotal adaptation, and at the same time, the need for a collected direction and thereby central control from the corporate executive committee's point of view, leads to difficulties in the strategic management and control of the firm.
Managing international enterprises
Studies about management and control in international companies are dominated by two questions, (1) the centralization and decen-tralization continuum and (2) the design of the management infor-mation systems, including the evaluation of the subsidiaries' ac-tivities in economic terms. These questions are usually associated with the information processing of the corporate executive commit-tee. The economic control systems are assumed to be designed in accordance with their need for, and access to information neces-sary for the control and integration of the firm and the subsidia-ries.
One example is the degree of decentralization of some decisions to the subsidiaries. In this perspective there is a question about the balante between on the one hand, the corporate executive committee's need to co-ordinate the firm due to operational linka-ges between different subsidiaries and, on the other, the possibi-lities they have to obtain and process the necessary information. The top management's task is to choose a leve1 of decentralization which is appropriate according to the above-mentioned need for, and access to information and the ability to process the same. In a similar way, the economic and other report systems are assumed to be designed considering e.g. the subsidiaries' connections to the rest of the Company and the corporate executive committee's requirements for information as a basis for decision-making.
Characteristic for these studies is the perspective where the corporate executive committee designs these control systems, usu-ally within the frame of the established organization structure. In our project, on the other hand, we consider control in an orga-nization from a perspective which is connected rather to power and position in a network of business relationships. Control has in this perspective a wider meaning, and is more a question of social relations within the formal organization (Astley and Sachdeva, 1984). In this perspective, control can emanate from three totally different sources, i.e. hierarchical level, resource-control and network position.
The hierarchical aspect of control deals mainly with questions about the formal ways of supervising a firm. Control connected to the hierarchical levels incorporates the above-mentioned centrali-zation and decentralicentrali-zation continuum, but also the degree of formal evaluation systems used by the corporate executive commit-tee. Control through hierarchical means is ultimately based on the legitimate right to control, i.e. the traditional organizationally sanctioned right, not on the control over actual resources.
The setond source emanates from the control over actual resour-tes. Control in this sense consists of the influence which an organizational unit can exercise on other units as a result of its control over resources on which they are dependent. This control
of resources is usually connected to the ability to obtain criti-ca1 resources from the environment.
The third source from which control can originate is connected to the differentiation and integration in an organizational sys-tem, like an internationalized firm. The integration creates a system of activity-based connections between the units, an inter-nal network. If we consider the formal structure within the hie-rarthy as a "prescribed" network, we see the activity-based con-nections as an emergent network (Tiehy et al., 1979). Usually, there is a certain correspondence between the activity-based network and the formal structure, i.e. "the organization provides the pegs upon which the emergent network hangs" (Tiehy et al., op.cit.). A unit can have more or less of a central position in an internal network. The more central the unit, the more important its role for other units' possibilities of obtaining the resources which, due to the resource-control, are at different places in the network. Centrality can, in this perspective, be considered a so-urce for control above the control that derives from the resoso-urce control.
Fig. 1:
Sources
of ton tro 1Hierarchical control is one-sided and based on the, through the organizational structure, legitimated right to control. The other two sources for control are reciprocal, originating in resource-control and network position. The traditional resource-control through the hierarchy can thus be either intensified or abolished by the other two sources of control. The corporate executive committee's possi-bilities to manage the subsidiary depend on the corporate
execu-tive committee's control over resources, important for the subsi-diary, and its centrality in the internal network. Mind, this in-ternal network includes the subsidiary. If, however, the resource control is in the hands of the subsidiary and the subsidiary has a central position in the internal network, the corporate executive committee's possibilities to control are limited. The control can then even be more apparent in the opposite direction.
The discussion has so far been concentrated to the internal re-lationships. The concept "environment" has only been used as a summary concept for everything outside the Company. Implicitly, there is also the assumption of correspondence between the formal structure and the activity-based structure, although this corres-pondence is not complete (Astley, Sachdeva op.cit.). The fundamen-tal components in this analysis are illustrated in Fig. 2 below. Fig. 2: Intra-organizational control perspective
In this first analysis, the relation between hierarchy, resource-control and position is fairly clear. The control over resources can be dispersed to several units in the internationali-zed firm inasmuch as these have connections with the environment. A correspondence between the hierarchically based control and the control arising from a unit's possibility to import resources from the environment is thereby not necessary. However, it can be
sumed that there is a positive correlation between leve1 of cen-trality and hierarchical level. The higher in the hierarchy the unit is, the higher the degree of centrality is assumed to be, due to the fatt that higher levels have more co-ordinating and integrating functions than lower levels. B should consequently have a higher degree of centrality than A. In most cases, two out of three control sources should favour the corporate executive committee over the subsidiary.
Nevertheless this conclusion is doubtful if the assumption about a correlation between the formal structure and an internal network is dropped. Another defect of the analysis is that the environment is still "anonymous". The contact with the environment is totally connected to the resource-control. But if we broaden the network concept to include also specific relationships to other units outside the Company, the two concepts centrality and resource-control acquire another meaning. The struttural power of an actor can be assumed to arise from the centrality in the network (Bartlett and Gh.oshal, 1990). A unit can be more or less central in the industrial network of which it is a part, no matter how much of the network is inside or outside the Company. Fig. 2 above should therefore be complemented as in Fig. 3 below.
In this perspective, A may have a higher degree of centrality than 6 if consideration is taken to the position which A occupies in the industrial network that also includes B. Thus, the position in the industrial network must be the starting point, not the po-sition in the network we ca11 the internal network. Centrality as a source for control Will not necessarily coincide with the hie-rarchical level, as is the case if we only perform an analysis in an intra-organizational perspective. If a subsidiary occupies a central position in an industrial network which is important for other units in the internationalized firm, the subsidiary has a base for influence irrespective of its position in the hierarchy. Consequently, the presence in an industrial network, as a base for the control of a unit, is decided both from the importante of the network to other parts of the internationalized firm and the sub-sidiary's position in the network in tomparison with other units of the internationalized firm. (Larsson 1985, Forsgren 1989)
The concept of industrial network
The concept of industrial network is based on the existence of lasting relationships between actors. The industrial network is furthermore a product of its history. The actors have memories of their interactions and have also developed knowledge about their network. At each point in time the firm occupies a certain posi-tion. The individual firm is dependent on resources controlled by others, and gains access to these through the network position. The development of positions takes time and effort and the posi-tion defines future opportunities and constraints for the strate-gic development. (Johanson 1989)
Industrial networks include several actors, linked to each other through specific relationships. The industrial network model is based on assumptions about industrial activities. Each activity is dependent on the performance of other activities which must precede or are expected to follow. The activities are like links, forming chains, which are more or less extensive and closely lin-ked to each other.
Industrial networks are unbounded and can extend without li-mits. Boundaries suitable for analytical purposes can be drawn, as a result of perspectives, intentions and interpretations. The ac-tors in the industrial networks can be firms, divisions, business units or individuals. No actor can perform his activities isolated from the others. An actor is also usually involved in a number of activity-chains, which makes the firm dependent, directly or indi-rectly, on activities in several directions. These dependences can be general or specific. (Johanson 1989)
An important basis for the network approach is found in social exchange theory. The focus was on dyadic exchange relations, until Emerson conceptualized dyadic relations as embedded in larger ex-change networks (Cook, (ed) 1987). He showed that exex-change proces-ses in dyads are systematically affected by the network in which they are embedded, and that the structure of the network is the major determinant of the exchange processes.
A fundamental assumption in exchange networks is that the indi-vidual actor's behaviour reflects the structure. If two actors have the same location in a network, they are under the same struttural constraints and should behave in a similar way. They are considered to occupy strutturally equivalent positions. Varia-tions in their behavior do not reflect the struttural property and should be attributed to individual actors. The logital unit of analysis is thus position rather than individual actors. (Cook (ed), 1987)
The development of an industrial network results from exchanges between actors and is not designed by an individual actor as a consequence of a plan or strategic decision-making. The design and character of connections between the exchange relationships dic-tate the structure of the network. However, each actor can initi-ate exchange, which can result in struttural changes. Usually pro-duct exchange is the primary aspect of exchange in business, but information and social exchange should also be considered.
An important factor influencing the exchange processes is dis-tance, both geographic and psychic. These two are correlated, but exceptions can easily be found. Psychic distance is defined as factors preventing and disturbing the information flows, such as
differentes in culture, language, leve1 of industrial develcrmcn( etc. (Engwall (ed), 1975)
The actors do not have the same amount of power. Their power is based on direct control over their own activities and indi,~;. control over other activities through relationships with othri ,lc., tors (Pfeffer & Salancik 1978). This indirect control derives from the network position, the strength of the relationship and -the other actors' relative importante for each other (Yamagishi et al. 1988). In every industrial network, there is a power-structure in the respect that the actors in the network have different power to att and influence the actions of others. The power is more or less concentrated to one or several actors, and this structure can have an influence on other actors as well as on the future development of the network.
The power-dependence relationships are based on the exchan:lc ol resources. An actor can lessen his dependence and thereby increase his power by cultivating alternative sources of desired resources. Perrucci and Pilisuk (1970) suggest that an agency, in an effort to obtain scarce resources, can mobilize more power the more it expands its interorganizational connections. According to Benson (1975) increased co-operation between organizations enhantes these organizations' capacity to dominate their environment. Thus organizations with extensive linkages to the environment are most
likely to be powerful. (Provan et al. 1980)
A subsidiary has, in principle, the same possibilities as other actors in the network to initiate exchange and thereby establish new relationships and change the network. Some of the relations-hips are to units within the same firm, others to units outside. With a resource-dependence perspective, the internal network con-sists of actors who are more or less dependent upon each other. The resources existing within the subsidiary and the internal re-lationships to other firm-units determine the position within the internationalized firm. The subsidiary's position in relation to the external actors is a question of the critical resources oxis-ting within the organization hut also the density and character of the relationships to these other actors.
The subsidiary’s relative power base - a model
This part of the paper presents the outlines of a model which is intended to assess a subsidiary' relative power base. In i.11 11. is the relationship between a divisional management, situated within the parent Company in Sweden, and a subsidiary abroad.'
(See Fig. 4)
The divisional management has a global responsibility and its primary function cancerns strategic issues. The subsidiary has strategic as well as operative responsibilities.
Fig. 4: Fota 7 leve75
nternationolized firm
paate executke committee Divisional manogement Subsidiary
In order to determine the relative power base, it is necessary to survey the network. As it is impossible to cover all rela-tionships, we shall concentrate on the most important actors, wit-hin as well as outside the firm.
The model recognizes four independent variables, where three reflect different aspects of the actual structures while the 1. In a nunber of subsidiaries, there is a divisional structure. In those cases, the rela-tionship between the subsidiary management and the lotal divisional management is also of interest. and Will be studied.
fourth is related to a certain unit's degree of knowledge about these struttural aspects.
Model 1: Subsidiary's power base
Subsidiary's relative power base
1
A A A
The first variable in the model connects to the hierarchy as a source for control. The formal hierarchical positions of the sub-sidiary and the divisional management constitute legitimate bases for power. To estimate the relative power in the formal structure it is important to see the relative hierarchical distance between the actors.
The setond variable is connected to the resource-dependence perspective. Our approach is to determine the relative economic importante of the fotal industrial network of which the subsidiary is a part. The main questions here cancern the extent to which the subsidiary directly contributes to the firm's total profit but also the indirect contributions generated through the internal network.
The third variable emerges from the positions which the subsi-diary and the divisional management occupy in the fotal industrial network. These positions play important roles as determinants of the subsidiary's relative power base.
The fourth variable differs from the others inasmuch as it is related to knowledge about the structures. Knowledge cannot be
absolute, i.e. no single actor can have a total perception of and knowledge about the structures. The main question is therefore to estimate an actor's relative degree of knowledge. One of the crucial questions here is the divisional management's knowledge of
important actors in the fotal industrial network.
The presented model is useful in both a dynamit and a static perspective. In a static perspective the model primarily gives a struttural outline of the conditions in the fotal industrial network, and thereby the subsidiary's momentary power base. It might be more appropriate to adapt a dynamit perspective, as it is more likely that the model contains a time-lag effect, i.e. the four independent variables are interlinked and connected over a period. The variables Will be further developed below.
(1). Formal positions
The usual assumption about hierarchical power is that the hig-her a unit's position, and thig-hereby formal authority, the greater its base for power, all other variables excluded. This derives from the assumption that a higher formal position in the hierarchy gives a better possibility to survey the environment and upcoming situations. The higher leve1 units have, in this perspective, a function as a beach-head between units involved in different in-dustrial branthes. The beach-head aspect is significant in the re-spett that it identifies an important position in the formal structure. A unit occupying a beach-head position has in an intra-organizational perspective an important power base as it handles a large amount of uncertainty. (Hickson, et al., 1971)
Normally, in this aspect the divisional management has a relative advantage (compared to the subsidiary) through the contacts they have with other divisional managements and the top management in the firm. This leads to the assumption that a unit's exclusive right to connections with other units within the firm gives a greater power base.
It is necessary as operative implications of the model to sug-gest some possible indicators of a subsidiary's relative formal position in an intra-organizational perspective. One possible
indicator is the number of formal levels between the subsidiary and the divisional management, which hints at how close they are in the hierarchical system. Another indicator can be the number of formal levels between the divisional management and the corporate executive committee. This indicator reflects the divisional management's possibility to survey the part of the total firm structure that includes the subsidiary. Closely related to the hierarchical leveis is the span of control. This indicator re-flects the corporate executive committee's number of subordinated units, including the subsidiary. In a stritt formal perspective, a wider span of control may indicate a somewhat larger power base for the subsidiary as the corporate executive committee to a grea-ter extent has to decentralize.
Still another indicator of a subsidiary's power base is the ex-istence of a matrix structure within the firm. The actual question in this case is whether the subsidiary has formal channels to other subsidiaries, divisional managements and/or corporate execu-tive committee management. The number and character of these for-mal channels are important for the subsjdiary's forfor-mal position relative to other firm units.
A plausible assumption is that the formal position not only in-fluences the subsidiary's power base directly, but also has an ef-fett on network position and knowledge about the network.
(2). Economic importante of the network
In a resource-dependence perspective an organization has a gre-ater power base if it controls resources valuable to other organi-zations. Applied to our model, the subsidiary's and division's significance can be estimated by their economic importante. In the first place, the subsidiary itself can be of extensive economic importante for the division, which in turn can be important to the firm. Setondly, the indirect contributions from the subsidiary's internal and external counterparts have to be considered, as their contributions can be partly dependent on the subsidiary's perfor-mance.
The economic importante of the subsidiary's network shows the
degree of value the subsidiary has for the division and the whole firm. The purpose of this variable is to highlight the subsidi-ary's relative economic strength or importante in the division and in the firm. Naturally, the subsidiary's contribution to the divi-sion's total profit and turnover are possible indicators of econo-mic importante, but we also see others such as e.g. future possi-bilities for the subsidiary's products as judged by the divisional management.
It is in most cases adequate to consider the economic impor-tance of the subsidiary as an indispensable prerequisite for its ability to influence. Even if the subsidiary has a strong position in the industrial environment, this position can only be used as a power base when the subsidiary and its internal and external coun-terparts
per se
have a sufficient economic importante.(3). Network positions
While the economic importante of a subsidiary reflects the va-lue added to the division, the network position demonstrates the subsidiary's ability to control this value. The network position is closely related to the concept of centrality, as this is used in the intra- and inter-organizational perspectives.
Above all three aspects of centrality are of interest; (1) to what extent a unit serves as a gatekeeper or beach-head between two networks or different parts of the same network (Krackhardt 1990, p. 35), (2) the number and the intensity of the connections the unit has, and (3) what would be the consequences for the network if the unit was removed.
These three aspects lead to some possible indicators of the subsidiary's network position. The first is the extent to which the fotal subsidiary is the divisions' only or principal contact with the external network. Another indicator is how many connec-tions the subsidiary has and how intense they are economically, technologically and socially. The question is who is dependent upon whom in the most important connections, either external or internal. The third indicator is based on the proportional reduction in total flows. This is a somewhat different approach
outlined by Yamagishi (Cook (ed), 1987). It incorporates the remo-val of a unit in an interlinked system. In an intra-organizational perspective the aim is to estimate the consequences of an imagi-nary removal of the subsidiary. This intellectual experiment Will show the hypothetical reduction in internal flows within the firm, as if the subsidiary had ceased to exist. With an industrial network perspective, the main questions are whether the removed subsidiary is replaceable, and what the medium-range effects, up to five years, would be in the network.
Although the divisional management is seen mainly as an administrative rather than an operative unit, with the prime purpose to manage the total activities, we cannot ignore the fatt that the divisional management sometimes has an operative role which is comparable with the role of the subsidiary in the network. It is therefore possible, and necessary, to determine the network position for the divisional management by using similar indicators as for the subsidiary.
Again the beach-head aspect becomes important, but in the oppo-site perspective, i.e. to what extent is the divisional management the subsidiary's only contact with the network, both the internal and/or the external. The number and intensity of the divisional management's contacts with other actors in the network is also im-portant. Moreover it is interesting to consider the proportional reduction in flows for the subsidiary, if the divisional management were removed, i.e. what the medium-range consequences would be for the network if the divisional management ceased to exist.
(4). Knowledge about the network
The fourth variable differs from the other three, being con-nected to knowledge rather than to the actual structures. Know-ledge, in our model is not general but specific in the respect that it refers only to the different management leveis' knowledge about the network in which they att. Knowledge is in our perspec-tive a rather tangible subjett, because it considers actual rela-tionships between actors in a business system. The variable,
knowledge about the network, can be viewed as a source of power. According to Krackhardt (1990) power in a network does not only accrue to those who occupy central positions but also to those who have an accurate perception of the network. In an industrial network perspective, the greater the knowledge an actor has about the actual structures, actors and relationships, the easier it is to exercise control. Knowledge has thus an indirect influence on the relative power base. In a sense this variable counteracts with the other three in determining the subsidiary's relative power base. For example, the subsidiary's relative power base may be ex-pected to diminish if the divisional management has a clear know-ledge about, and contacts with, the subsidiary's most important customers, suppliers etc.
Some possible indicators of knowledge about the network are to what extent the divisional management knows to which central sup-pliers, customers and other important actors the subsidiary has relationships. A setond indicator is to what extent the divisional management has contacts with or relationships to actors in the network. A third indicator is the psychic distance between the di-visional management and the subsidiary.
The subsidiary's relative power base
There are numerous definitions of power. From a general point of view, power can be defined as a capacity to bring about conse-quences or to affect the outcome of an issue. According to Larsson (1985) power is both relative and relational and can be exercised by both parties in a relationship.
In most definitions, power need not be exercised to exist. Wrong (1968) emphasized that although power is defined as the pscity to control others, there is a differente between the ca-pscity to control and the actual performance of control. The for-mer is talled potential power, the latter enacted power. This con-ceptualization is adopted by Provan et al. (1980). Figure 5 below illustrates the interrelation between the two power concepts and different measurement approaches.
Fig. 5: The interrelation of conceptual and measorement approaches to power.
Paten&!P<lWcr Concept"<rlizOtion EnocttA
I Pora
! SubjoctNs Pa-C.i"kd PCWCI
Type of uaosLaement
Source: Provan et al. (1980) p. 208.
The figure shows that there is a differente between objective and subjective measurement approaches. Objective measurement met-hods aim to estimate the two basic concepts of power. Subjective methods do not give the same clear-tut distinction between the concepts, but rather an integrated picture of power which Provan talls perceived power. Perceived power thus gives possible biases in the measures. To obtain a complete picture, our aim is to use both measurement approaches in studying power relationships. Objective measurements give us the unbiased power conditions, whilst the subjective methods add the experienced power to form the power bases in the system.
Possible indicators are for example the divisional management's and the subsidiary's opinion of each unit's power, i.e. which unit is considered to have most influence in the firm. Another possible indicator is third partyls opinion of where the greatest influence on the network development is.
Other relevant indicators are the influence on critical events, e.g. (1) the subsidiary's part of total investments in the divi-sion during the last five years, (2) the subsidiary's influence on the division's most important investments during the last one, two
and five years, and (3) the divisional managementis influence on the subsidiary's most important investments during the last one, two and five years.
In our opinion, the subsidiary's participation in the divisional management, together with the divisional management representation in the corporate executive committee is also an indicator of the power base. The time which the divisional management devotes to developing relationships to actors in the subsidiary's network should also be investigated as well as the time the divisional management uses to visit the subsidiary. The main problem with these indicators is to assign them to the different concepts of power. As the measurement is based on both objective and subjective methods we anticipate some difficulties in the analysis of power bases. However, in a theoretical perspective the distinction is useful.
Empirital notions and concluding remarks
The model presented above constitutes the theoretical framework of an empirital study which Will be conducted mainly during 1991. The study Will include six of the largest Swedish multinational firms. In each firm a number of subsidiaries Will be studied. The focus will be on subsidiaries located mainly in Europe. To cover the influence from both geographic and psychic distance, some ad-ditional studies Will be made outside Europe.
As mentioned before, the firms often have structures which dif-fer from the one outlined in our model. A subsidiary may, for in-stance, tonsist of several divisions.
In
those cases, our aim isto study as many divisions as possible in order to obtain as plete a picture as possible of the tension between the parent com-pany and the subsidiary.
References
Astley, G and Sachdeva, P, "Struttural Sources of Intraorganizational Power: A Theoretical Synthesis", Academy
of
Management Review, 1984, Vol 9, No 1, pp 104-113Bartlett, C and Ghoshal, S, ' The Multinational Corporation as an Interorganizational Network", Academy of Management Review (forthcoming) Benson, K, "The Interorganizational Network as a
Politital Economy", Administrative Science Quarterly, Val 20, 1975, pp 229-249
Cook, K, ed, "An Exchange Theoretical Approach to Network Positions" by Yamagishi, T, Social fxchange
Theory, SAGE Publications, 1987
Engwall, L, (ed), "The Internationalization of the Firm Four Swedish Cases" by Johanson, J and Wiedersheim -Paul, F, Uppsala Contribotions to Business Research, Uppsala, 1984
forsgren, M, Managing the InternationaIization Process, Routledge, London and New York, 1989
Forsgren, M, Holm, U and Johanson, J, "The
Internationalization of the Setond Degree" in Young, S (ed), Eut-ope and the MultinationaIs: Issues and Responses for the 199Os, Edwar Elgar (forthcoming) Hickson, D J, Hinings, C R, Lee, C A, Schneck, R E, and
Pennings, J M, "A Strategic Contingencies'
Theory of Intraorganizational Power", Administrative Science Quarterly, Vol 16, 1971, pp 216-227
Johanson, J, "Business relationships and Industrial Networks", Reprint Series 1989/7, Department of Business Studies, Uppsala University
Krackhardt, D, "Assessing the Politital Landscape: Structure, Cognition, and Power in Organizations", Administrative Science Quarterly, Vol 35, 1990, pp 342-360
Larsson, A, Structure and Change. Power in the
Transnational Enterprise, Acta Universitatis Upsaliensis, 1985, No 23
Pfeffer, J and Salancik, G, The External Control of Organizations: A Resource Dependence Perspective,
Harper Row, New York, 1978
Perrucci, R and Pilisuk, M, ' Leaders and ruling elites: The interorganizational bases of connnunity power", American Sociologital Review, 35, 1970, pp 1040-1057
Provan, K, Beyer, J and Kruytbosch, C, "Environmental Linkages And Power in Resource-Dependence Relations between Organizations", Administrative Science Quarterly, Vol 25, 1980, pp 200-225
Tiehy, N. M, Tushman, M. L and Fornbrun, C, "Social Network Analysis for Organizations", Academy of Management Review, Vol 4, 1979, pp 507-519 Wrong, D, "Some Problems in Defining social Power",
American Journal of Sociology, Vol 73, 1968, pp 673-681
Yamagishi, P, Gillmore, M and Cook, K. "Network Connections and the Distribution of Power in Exchange Networks", American Journal of Sociology, Val 93, No 4, 1988, pp 633-651
