MODELLING THE IMPACT OF ECONOMIC INTEGRATION
ON MULTINATIONALS’ STRATEGIES

 

 

Ana Teresa Tavares *
University of Reading
University of Porto (Faculty of Economics)

Address:

University of Reading
Department of Economics
PO Box 218 – Whiteknights
Reading, RG6 6AA, United Kingdom

Phone: ++ 44 (0) 118 9875123
Fax: ++ 44 (0) 118 9750236

Email: A.T.C.D.P.Tavares@reading.ac.uk

 

*I would like to thank Simon Burke, Mark Casson, Bob Pearce and Nigel Wadeson for invaluable comments and suggestions made during the writing of this paper. Partial financial support from the Subprograma Ciência e Tecnologia do 2o Quadro Comunitário de Apoio is acknowledged.

 

MODELLING THE IMPACT OF ECONOMIC INTEGRATION
ON MULTINATIONALS’ STRATEGIES

 

 

 

Abstract

 

Economic integration between countries and corporate integration effected by multinational enterprises (MNEs) represent undoubtedly two of the most determinant shaping factors in the contemporary economy.

These two phenomena seem to coevolve, and develop (at least partially) a mutual influence.

The paper focuses on the impact of economic integration on multinationals’ strategies. It models the decision-making processes intra-MNE in response to the imminent formation of a trading bloc between two hypothetical countries (a ‘peripheral’ and a ‘core’ country, with distinct comparative advantages and macroeconomic conditions). The previous scenario of trade fragmentation is compared and contrasted with the post-integration situation in terms of the implications for the multinational’s activities and strategic choices.

There are considerable formalisation opportunities in this field. A discrete choice model is proposed, in which the strategy set available to a MNE when economic integration occurs is clearly identified and a system of equations solved through subsequent mathematical programming. Boundaries of strategic optimality are drawn, based on a trade-off between linkage and production costs, providing results that show which is the best strategy in distinct scenarios.

Underlying the approach used is the concept of the MNE as a differentiated network and the idea of asymmetry in the strategies/roles that subsidiaries may develop. It is argued that economic integration allows more rationality in the MNE’s decision-making process, materialised for instance in specialisation according to comparative advantages (among other factors). The need to adopt a systems perspective is defended, and the explicit inclusion of information costs is an essential feature of the model.

The version here presented is marked by generality and flexibility to accommodate further refinements and changes in assumptions. It purports to act as a feasibility study, setting out its own research agenda, that may be developed to incorporate phenomena such as vertical integration, multi-product manufacturing and specialised R & D operations. It also allows the inclusion of at least one more country (preferably belonging to another trading scheme), transcending the boundaries of one bloc and enabling the consideration of global issues.

Key words: economic integration, corporate integration, systems, coevolution, discrete choice model.

 

 

Introduction

A general process of integration is observable in several aspects of the contemporary global economy. Two instances in which this process assumes particular relevance are economic integration through the formation and interaction of trade groupings and corporate integration effected by multinational enterprises (MNEs). The paper explores certain aspects of the coevolution process between firms and trading systems.

Dunning and Robson (1988) emphasised the need to explore the interface between economic and corporate integration. Previous analyses (e.g. Scaperlanda and Mauer [ 1969] , Goldberg [ 1972] , Lunn [ 1980] , Scaperlanda and Balough [ 1983] ) studied the impact of the formation of a trading bloc on the magnitude of FDI inflows. This macro-level perspective, which neglected the role of the firm, is unsatisfactory, and a micro-focused, institutional analysis, shifting the emphasis to the MNE (the institution that is ultimately responsible for these - often fragmentary – FDI decisions), seems more appropriate.

The motivation for the model here proposed stems from the considerable neglect to which this economic-corporate integration dialectic has been subject, and from the formalisation opportunities it provides. The paper conceptualises the MNE as a decision-making system that reacts to environmental shocks (e.g. economic integration), and models the decision-making processes intra-MNE in response to them. Hence the focus is on the internal organisation of the firm and on how it may evolve when two countries previously divided by a host of trade barriers form an economically integrated area. One of the keys to understand this interface consists of relating changes in MNEs’ organisational structure and subsidiaries’ roles to the dynamic processes induced by economic integration.

The paper’s emphasis is on generality. The undeniable virtues and intellectual appeal of generality are consistently defeated by its inherent complexity. Hence, a research strategy to tackle more specific aspects is identified, and in this vein a model (always a partial and selective representation of reality) is proposed. The model purports to act as a (contextualised) feasibility study, setting out a research agenda that may be worth pursuing.

The paper will be organised as follows. The first major section will discuss the main background ideas underlying the approach used, including the need to use a systemic perspective and the relevance of information costs and their impact on organisational decisions and structure. The second main part will deal with the specifics of the model proposed. The strategy set available to the archetypal MNE when economic integration occurs is specified and the model is solved through the identification of dominant strategies under certain conditions and subsequent mathematical programming. The following section will analyse and discuss the results obtained. These results will show which are the optimal loci of production and R & D for different combinations of the parameter values of the cost structure equations. Finally, some concluding remarks and suggestions for future research will be outlined.

Systems and Information: Two fundamental Building Blocks

A Systems Approach as a Logical Imperative

The concept of system as a "set of elements standing in interrelation" (Bertalanffy, 1968) is extremely pertinent in a study of the impact of economic integration on MNEs’ strategies. A possible reason why this view is not yet popular in this field is its intrinsic complexity. However, it can be shown that the main principles and analytical tools of systems theory are applicable to both economic and corporate integration in a straightforward manner.

First and foremost, a dynamic system is characterised by having coherent elements, with patterned relationships, and forming a purposive whole (Beer, 1966). Goal directedness, self-awareness, and convergence to a compromise purpose should a priori be features of any successful economic integration effort or multinational firm.

A fundamental notion is systemic embeddedness. Systems are composed of subsystems and are embedded in other (larger) systems. Subsidiaries have distinct departments and are part of the MNE group. Economic agreements (macro-regions) reunite different countries, which in turn are composed by micro-regions and localities, and are embedded in the wider global economy. These can be seen as articulated ‘levels of recursion’ (Beer, 1985), successively and logically nested. Recursivity is, therefore, inseparable from the notion of embeddedness.

Given this complexity and interaction of various levels, a ‘system-in-focus’ (Beer, 1985) must be adopted. The paper will concentrate on the macro-region (economic grouping) and will model a situation intra-economic bloc. The case of the macro-region, by linking national and global systems, has particular contemporary relevance.

The division of labour (as well as the division of knowledge) in dynamic systems creates the need for coordination, that must account for the concomitant forces of differentiation and integration (Lawrence and Lorsch, 1967) in complex organisational systems. Coordination is a central problem at all recursion levels. However, it would be reductionist to see coordination merely as a problem. Economic integration processes, more than posing a challenge to firms’ adaptation, provide also enhanced opportunities (e.g. capitalising on the interdependence and complementarity of both countries’ and firms’ characteristics).

Another feature of any system can be synthesised in the idea of ‘wholeness’, or gestalt. Each part makes sense in terms of the others: the whole defines itself (Beer, 1985). Synergy (the whole is more than the sum of the parts) is a fundamental aspect. Hence, there is a need to study a system as an entity rather than as a conglomeration of parts (Ackoff, 1959), as well as the patterns of linkages that underwrite their reciprocal dependence.

Homeostasis refers to the stability of a system’s internal environment, despite the system’s having to cope with an unpredictable external environment (Beer, 1985). Homeostatic regulation ensures the maintenance of internal balance, in particular when conflict with pressures from other systems occurs. As viable systems cannot be entirely regulated from the outside (Beer, 1966), the MNE needs to implement self-regulation mechanisms (Ashby, 1956; Beer, 1979; Gupta et al. 1999), or some type of organised self-organisation. Formal and informal anti-oscillatory devices, such as a solid corporate ethos, must be developed.

Equilibrium is crucial. As a ‘tendency toward nonchange’ (Seiler, 1967), it is continuously challenged by strong drives for change and growth (like the ones posed by imminent integration). Thus, a system must reach a compromise between its dynamic pressures and its inertia and frictions. Organisational equilibrium, though dynamic, is often perhaps not optimal, due to this mix of cooperation, conflict, and incentive asymmetry between distinct subsystems. For subsidiaries and countries to cooperate they must have a motivation, either of a positive (some sort of recognition/payment) or negative nature (fear of chaotic disruption). Ultimately trust among members and loyalty to the organisation (‘systemic belonging’) constitute the best cement and basis for cooperation (Casson, 1990; Simon, 1997).

To give a practical example, the very notion of globalisation in intrinsically systemic. Globalisation provides a special momentum to develop a systems view, to understand the various challenges and opportunities that are so frequently mentioned (although rarely explained). In very simple terms, the global/regional interface is linked to the issue of how (sub)systems are embedded in higher order systems. Hence, one of the possible ways to conceptualise globalisation relates to this complex juxtaposition and interdependence of subsystems, and to the considerable heterogeneity both intra- and inter-systems. The proliferation and fluidity of linkages and blurring of boundaries are also inherent features of the process. Thus, the global economy can be seen as a ‘systemic envelope’ of several regionally integrated interacting systems. As hinted before, a truly systemic view was rarely attempted in terms of economic and corporate integration. The theory of the firm and in particular of the MNE provide a few exceptions worth mentioning. More specifically Coase (1937) and Buckley and Casson (1976) adopted a systems perspective. Buckley and Casson (1976) stressed the interdependence and connectivity between different activities in the MNE. Knowledge flows were explicitly considered, which represented a further advance relatively to the then prevalent materialistic view of the economy. Casson (1990) also emphasised the need to adopt a systems view of international business. The sheer amount of linkages in modern MNEs and cross-effects created when each linkage is embedded in a system of other linkages make this approach a logical necessity rather than a choice.

The Economic Role of Information, Information Costs, and their Impact on Organisational Structure

Neoclassical theory envisaged the firm as a ‘black box’, with no ‘insides’ (Penrose, 1995). Information was costless and complete, and no concern with the internal organisation of the firm existed. The relative neglect informational issues were voted until recently (as more materialistic views were privileged) was reversed with the pioneering work of Hayek (1945), Richardson (1972) and Marshak and Radner (1972). The perspective of the firm as a specialised decision-making unit, whose function is to improve coordination by structuring information flow (Casson, 1995) is a main building block of the present analysis.

Information has a non-negligible economic role, or rather a multiplicity of roles. Hayek (1945) emphasised the role of information in the coordination of trade and economic activities in general. Another of its fundamental roles is as a ‘belief-changer’, or a reducer of uncertainty. Information always increases knowledge about something. By leading to revise expectations, it may contribute decisively to improve decisions.(1). Marshak (1974) stressed the value of information as an asset. Its value derives from the impact on practical decisions taken by the firm (Burke and Casson, 1998), as it allows both better judgement and better coordination, thus being conducive to superior allocative efficiency and increased welfare.

In fact, information affects decision-making not only directly as illustrated, but also indirectly (e.g. through its effect on the goals of the different groups in the system [Wadeson, 1997]). Furthermore, as information is difficult to sell, it is possible to gain rents from it (Casson, 1995b). All these perspectives are extremely applicable in the context of information flows intra-MNE/inter-subsidiaries. They will be explicitly included in the subsequent analysis.

Informational issues are also linked with the systems approach defended above. Information has always been at the forefront of systems theory (Shannon and Weaver, 1949), given the primacy of its role in regulatory systems. By counteracting entropy, it underwrites viability.

Information costs (2). include the costs of collecting, processing, storing and communicating information, and also the costs of forming the judgements required for a decision (Casson, 1997). There are also costs of consolidating information intra-firm (inter-subsidiaries) that can be considerable if strategic withholding of information occurs. The magnitude of information costs has a multiplicity of determinants. Two key variables are search and communication. Communication, like decision-making, is always costly and imperfect (Sah and Stiglitz, 1986). Tacitness increases communication costs. Sharing and combining different knowledge fragments, in particular "of the particular circumstances of time and place" (Hayek, 1937; 1945) is never free. Nelson and Winter (1982) stressed that rules, procedures and routines possessed by each firm (subsidiary) may be difficult to communicate due to their tacit nature.

Internal communication of information tends to be rather defective at the firm level (i.e. subsidiaries may not share information in order to enhance their bargaining power intra-group). This is particularly valid for the most creative subsidiaries. In this vein, the subsidiary may be envisaged as a monopolist. This brings the issues of the non-public good nature of information intra-MNE and how important is secrecy, both as a rent-appropriating strategy, and as a motivation for subsidiaries to excel in their innovative capabilities and contribute to the organisations that they are a part of. Asset specificity and this type of monopoly power determines that information may be withheld by subsidiaries, to sustain their power in the MNE network, both vis-à-vis HQ and their ‘sister’ subsidiaries. Thus, the HQ-subsidiaries interface may be conceptualised as a principal-agent problem (Jensen and Meckling, 1976).

Economic integration will be probably conducive to a process of structural change in the MNE involving rationalisation, largely consisting of more specialised activities developed by each subsidiary. A trade-off between the benefits of specialisation and the costs of communication is likely to exist (Wadeson, 1997). When economic integration occurs, information flows from HQ to subsidiaries cannot be restricted to ‘mainstream managerial directives’. There will be more than the previous ‘costs of order-giving’, and also an intensified need for two-way information flows and feedback that will not be costless as well.

Two contrasting forces have been pervasive recently. On the one hand, information costs are falling (mainly due to technological change), and on the other more information is generated, needing to faster processing. The model will account for these factors.

The structure of institutions can be seen as a rational response to the need to economise on information costs. For Radner (1992), Carter (1995), and Casson and Wadeson (1998), information costs provide a key to the firm’s internal organisation. Decentralised information in the firm raises an organisational problem (Carter, 1995). As relevant information can be communicated in varying ways, there are multiple organisational structures (‘architectures’), with distinct effectiveness and costs (Sah and Stiglitz, 1986; Buckley and Carter, 1996).

To take the economic role of information and the costs it implies properly into account has great relevance to understand the firm’s internal organisation and decision-making. Organisations exist not only to process information about their members but also about their environment, e.g. scanning the environment and using the information to formulate a business strategy (Casson and Wadeson, 1996). This approach is potentially more realistic than a perspective based only on deterring shirking and opportunism (Buckley and Carter, 1996). There are wide-ranging implications of considering information costs, notably when the MNE needs to make an optimal strategic choice that would maximise the value of its objective function. Since information is crucial for finding the best decision, it should be included in any model involving a choice between distinct strategies, as will be shown subsequently.

The Model

Preliminary considerations and methodology

The model identifies the strategy set available to a MNE when economic integration occurs. The emergence of a trading bloc is conceptualised as an influential exogenous shock in the MNE’s environment, to which it must react in order to remain competitive given the new threats and opportunities implied by the new trade scenario. The type of model deemed adequate is a discrete choice model in which all available strategies are identified and evaluated in order to find the best possible course of action, given specific working assumptions and the firm’s goal. Examples of the economic application of this methodology can be found in recent literature (Buckley and Casson, 1998; Casson and Wadeson, 1996).

The MNE acts as a profit maximiser. Given its products’ prices, demand conditions and market competition, the firm aims at minimising its system-wide cost (Cheung, 1983), i.e., the total cost of developing, producing and distributing its output in all countries in which it operates. Cost minimisation is fundamental for the firm’s survival and performance in a competitive industrial system (and reasonably realistic according to what firms de facto do).

Like in all attempts at formalisation, this model relies on a certain set of assumptions that, though intrinsically (and intentionally) restrictive, offer a workable theoretical framework permitting to analyse the phenomena under consideration. The deliberate simplicity of these assumptions is legitimated by both the imperative of tractability of the model and by the flexibility that the proposed framework offers to further extension and refinement.(3).

The status quo before economic integration

Consider two countries, X and Y. Until now trade between them has been fragmented by a host of trade barriers, notably a combination of tariffs, non-tariff trade barriers (NTBs) and quotas. X and Y are considering the scenario of founding an economically integrated scheme.

The two prospective members have different macroeconomic characteristics (e.g. GDP per capita), factor endowments, and development stages, allowing to consider X as a ‘peripheral’ and Y as a ‘core’ country (no spatial connotation implied). The ‘core’ country has a better established scientific community and superior infrastructural conditions. Labour costs are inferior in X. Y has a static comparative advantage in technologically more sophisticated goods/productive stages, whereas X is more competitive in labour-intensive products/stages.

A typical multinational enterprise (M), by assumption originating in Y, has been operating in both countries for a considerable time. It basically develops three value-adding activities, viz. production (P), distribution (D) and research and development (R & D).

In the prevailing trade fragmented environment, M has only one R & D lab located in its home country Y. It has production and distribution activities in both X and Y, given that trade barriers render exports relatively too expensive vis-à-vis local production. M acts as a cost minimiser, and this paper’s focus will be on the cost minimisation in the economic bloc, though more realistically any MNE will tend to pursue efficiency at the global level.

Production undertaken by M’s subsidiaries in X and Y is fundamentally of an autarkic nature. Although there is hierarchical control from the main HQ in Y, subsidiaries operate in a rather independent manner (focusing exclusively on their local markets, as Fig. 1 illustrates).

It is assumed that demand for the firm’s products is fairly price-elastic (so trade barriers are effective), and few productive factors need to be traded across borders. It is straightforward to conclude that this strategy is a priori inefficient as it does not involve specialisation in accordance to local comparative advantage or scope for economies of scale in production.

After economic integration

Freedom of trade between X and Y will eventually induce appropriate adjustments in the firm’s internal organisation. Confronted with the new fluid trade context M will probably face two dichotomous pressures. On the one hand, due to the decrease in export costs resulting from the erosion of trade barriers, there may be an incentive to re-siting all operations back to the home country Y and export to X (centralisation). On the other hand, an enhanced scope for coordination means that the firm will be better positioned to exploit the advantages of corporate internalisation and could decentralise (to part or all countries of the bloc, according possibly to local comparative advantage and to the overall goal of cost minimisation).

Distinct locations attract distinct types of subsidiaries. This is consistent with the recent literature on subsidiary’s roles/strategy (White and Poynter, 1984; D’Cruz, 1986; Pearce, 1992; Birkinshaw and Hood, 1995; Taggart, 1996) and with the idea of distinct motivations underlying international production (Dunning, 1993). It also reflects the conception of the MNE as a differentiated network (Forsgren and Johanson, 1991), and the emphasis on the MNE’s intra-systemic heterogeneity. In this paper a parsimonious, tripartite typology of subsidiaries’ roles/strategies will be instrumental, based on the ‘scope’ typology (4). originally proposed by White and Poynter (1984), and from which many variations have been derived. An autarkic subsidiary supplies exclusively the host country. It is typical from constrained trade environments (as before integration). It is a market-seeking, import-substituting subsidiary type, producing the full range of the firm’s goods (wide product scope) and having a narrow market scope. Functional/value-added scope is usually quite limited. A rationalised manufacturer produces a part or one of the firm’s goods or specialises in a productive stage (narrow product scope), for export markets (wide market scope) and has a quite limited value-added scope (usually less than its autarkic counterpart). Basically they obey HQ directives. Finally, a creative/high value-added subsidiary is characterised by narrow product scope, considerable market scope (for regional/global markets), having the resources and the mandate to develop, produce and potentially market the product(s) for which it is responsible.

After economic integration, most probably, the profit-maximising firm will abandon its inefficient approach of producing ‘everything everywhere’ and will specialise in order to benefit from local conditions and coordination opportunities. The higher the complexity of the MNE’s network, the greater the scope for coordination. Then the firm will give up having autarkic subsidiaries and will mainly undertake rationalised operations.(5). Potentially a creative subsidiary will be set up to develop products for the whole economic area. It is expected that such process may bring welfare-enhancing effects due to matching operations to local comparative advantage. There is a crucial change from negative to positive location advantages (Dunning, 1977), i.e., from trade restraints to selected sources of competitiveness.

Basic structure of the model

It is possible to develop a family of discrete choice models to analyse the problem of strategic decisions of MNEs when economic integration occurs. The version here presented is the simplest, having considerable flexibility to be developed in a multiplicity of ways and to accommodate further complications and relaxation of assumptions as determined by imperatives and aims of future research along these lines.

The present is a two-country model. The MNE operates in countries X and Y, that have respectively a comparative advantage in labour-intensive and knowledge-intensive activities.

The MNE produces only one consumer good, the same version being sold in markets X and Y. No adaptation and local R & D are needed. By assumption the good is labour-intensive.

Three value-adding activities are developed by the archetypal firm, namely production (P), distribution (D) and R & D (for more details see the next section).

The MNE is a system whose parts are linked by a complex web of distinct flows (6). (see Fig. 2).

 

* Material (products - from P to D, one-way)

The cost of material flows is proportional to output. It is assumed that, for this good, X and Y markets have the same dimension (=1; unitary). Total production capacity is set =2, so if there is only one plant it will produce 2 and if there are two each will produce 1. When half of the production is transferred from a factory to distribution (if only one producer exists), linkages concerning material flows from P to D are multiplied by 0.5 in the respective equations. Cross-hauling of production, incompatible with the profit-maximising assumption, does not occur.

* Intangible (knowledge - R Û P and R Û D, two-way). This type of flow is very important and has been vastly neglected. Knowledge flows from research to production (e.g. new productive techniques) and vice-versa (when a situation occurs and improvements can occur by doing further research). Knowledge also flows from distribution to research (information about factors like market conditions, consumers’ tastes and fashions) and from research to distribution (how to change the product/distribution strategy to reflect demand volatility).

There is not a direct knowledge flow from production to distribution and vice versa. As there are two-way flows linking R Û P and R Û D, it is plausible to assume that the link P-D is intermediated by R & D (a common denominator, avoiding unnecessary duplication).

The cost of intangible flows is assumed unrelated to output, depending just on the number of linkages established. The cost of intangible flows is greater than the cost of material flows, ceteris paribus (due to tacitness and difficulties in communicating/transmitting information and knowledge). When flows are two-way, they will be multiplied by 2 in the cost equations.

i (iota) = premium of intangible flows versus material flows (per flow)

Elements of the cost structure

In the transition before-after economic integration, some components of the cost structure will increase or decrease in absolute terms (e.g. production costs, trade barriers) and some will increase/decrease in relative terms (e.g. cost of international versus domestic linkages).

Examining more carefully these components, the following categories appear relevant:

I) Costs related to the three value-adding activities:

I.1) Production costs - three possible production strategies exist: production in X only; production in Y only; and production in both X and Y (as before economic integration). It is assumed that price is high enough so that it is always attractive to serve both markets. Comparing production costs, in rational terms they should not be higher after integration. In the limit they would remain equal. When markets were fragmented, P occurred in both X and Y. Production costs were probably greater, since after plant location can be rationally chosen and then production concentrated in one site, the other market being served by exports. There is a potential opportunity cost of loss of economies of scale, if they are relevant in the industry, and of goodwill in case divestment will occur in one of the countries.

Fixed cost of production is ignored. As the MNE had already factories in X and Y, if it centralises production, fixed capital equipment and machinery are assumed easily transferable to the other location. If the firm still decides to carry out production in both countries, the plants will be the same, thus it is not a major loss of explanatory power to ignore fixed costs of production and set up costs. The good is assumed unchanged after integration (there may be a need to adapt technology in case of rationalisation but it can be ignored without much risk of loss of rigour or generality). Variable production costs are proxied by labour costs. Production is labour-intensive, and raw materials represent on the one hand a relatively small part of the total variable cost, and on the other hand their costs do not differ considerably between X and Y, so their exclusion is not too problematical. One unit of labour generates one unit of output.

p (pi) = production costs (unit) in X.

q = output (number of units).

P = p * q = total production costs

a (alfa) = premium to produce the good in Y rather than in X

I.2) R & D costs

As Demsetz (1991) emphasised, knowledge is costly to produce, maintain and use, hence R & D costs, as ‘costs of generating knowledge’, must be given due consideration. Knowledge is a public good intra-MNE (though this assumption is open to challenge) (7). , thus only one R & D lab exists (in either X or Y) and knowledge there generated is transferred throughout the firm without friction. Subsidiaries are aware of how crucial is knowledge sharing for the MNE’s success and for their own viability, so incentives to behave opportunistically are reduced.

Hypothetically, R & D before occurred in the main HQ in Y. Since Y is comparatively advantaged in scientific activities, R & D will have a natural propensity to occur there. Given the postulated no need for adaptation these costs tend to be equal before and after integration.

Variable cost refers to units of scientific knowledge. In the present framework, knowledge generation uses only scientific labour, one unit of output using one unit of this input. Hence, the overall R & D cost coincides with its variable cost. Fixed cost (set up of the R & D lab and respective equipment) can be ignored as in principle the R & D lab will be the same after integration (given Y’s comparative advantage in knowledge-intensive activities and the constancy in the products’ characteristics) or, if not, the equipment can be easily transferable.

r (ro) = research costs (per unit produced) to research in Y.

R = r * q = total research costs

b (beta) = premium to research in X

I.3) Distribution costs

A distribution facility (due to goodwill and to enhance the penetration and capture of market share) exists in all countries (X and Y) of the economically integrated space. Therefore, distribution costs do not constitute a decision criterion. The firm distributes its products in X and Y through its own channels. Hence, a fixed cost (set up of this network) has already been incurred. A certain amount per unit distributed constitutes the variable cost of distribution.

d (delta) = distribution costs (unit)

D = d * q (total distribution costs)

II) Linkage costs - This category encompasses three major types of costs, notably:

II.1) Transport costs associated with moving the good or resources. Concerning these costs (a factor weighing in the decision exports vs. local production), some industries will have problems in decentralising (e.g. ‘heavy’ industries). For intangibles, the relevant costs (called transfer costs) include training costs, postal costs, and telecommunications (Casson, 2000).

II.2.) Transaction costs involved by the three types of trade barriers, notably tariffs, costs of surmounting NTBs, and the implicit cost of the existence of quotas. In what concerns costs of surmounting trade barriers (pre-integration), the distinct types of restraints imply transaction costs, and integration results in setting this component to zero, in the limit.

II.3) Information costs. This category includes information costs of both a transactional and of a non-transactional nature, the latter referring to information costs not associated with the prevention of attitudes such as shirking and opportunism. A vast part of information costs are implied in devising an appropriate business strategy rather than preventing this kind of entropic (from a system’s perspective) behaviour (Buckley and Carter, 1996; Casson,1997b).

The concept of information costs employed here encompasses the costs of obtaining, transmitting/communicating, and also of consolidating information, as information is dispersed in distinct parts of the organisation (Hayek, 1945; Aoki, 1986; Radner, 1992).

With the new free trade scenario, the concomitant scope for rationalisation and intensified competition, there will probably be an increasing need for communication and consolidation. Communication cost refers mainly to intra-firm/inter-subsidiary flows. After integration, the information flow inter-subsidiaries is not limited to the previous central HQ order-giving.

In sum, fluidity brings the potential for greater volatility, increasing need for fast decisions, prompt reaction to new competitive pressures, and more considerable need for permanent and timely information exchange and feedback. For these reasons, it may be hypothesised that ceteris paribus information costs (especially of a non-transactional nature) will tend to increase with integration, and this is likely to drive the general pattern of linkage costs.

At this stage, linkage costs will not differentiate among these three distinct categories. The second type of linkage costs can fairly be ignored post-integration, whereas transport costs and information costs will be essential to decide on the optimal strategy to adopt under certain conditions to be specified. Moreover, as pointed out before, it is hypothesised that information costs will play a key role in the overall magnitude and evolution of linkage costs.

Only one variable (L) will be used to denote linkage costs. As already explained, linkage costs are assumed independent of output and only contingent on the number of linkages. The cost of international linkages is greater than that of domestic linkages, ceteris paribus.

l (lambda) = cost per linkage

n = number of linkages

L = linkage costs = lambda * n

g (gamma) = premium to establish an international linkage versus a domestic one

Until now, the main distinction between costs pre- vs. post-integration resides in the costs of production (if pervasive economies of scale or benefits of specialisation due to comparative advantage exist), and in the linkage costs involved. Hence, the problem consists of comparing the decrease in production costs with the potential increment in other (linkage) costs. This kind of trade-off has already received support in existing literature (Teece, 1983) (8).

Strategy set available

Six possible strategies (S = 1, 2,…, 6) exist (see Table 1 and Figure 2). As distribution (D) will always occur in both countries, distribution costs will be the same for all possible courses of action, being irrelevant as a discriminating variable. Hence distinct strategies can be characterised just by the locus of production and by the locus of research.

Table 1. Strategy set available to the MNE

Production (P)

R & D

Distribution (D)

S1

X

X

X + Y

S2

X

Y

X + Y

S3

Y

X

X + Y

S4

Y

Y

X + Y

S5

X + Y

X

X + Y

S6

X + Y

Y

X + Y

An important element to solve such a problem is to identify clearly dominated strategies and eliminate them as early as possible (Casson, 2000). Economic principles are fundamental in the early identification and elimination of dominated strategies, as well as in deciding the appropriate order of search among non-dominated counterparts. In this case, S5 is dominated by S6. In turn, S6 is dominated by S2. Furthermore, S3 is dominated by S2. This is justified by former assumptions concerning the distribution of comparative advantages. It will be proved that in fact strategies 1, 2 and 4 dominate the other strategic alternatives.

Formalisation of the equations corresponding to the six distinct available strategies

Each equation Ki formalises the cost structure associated with each strategy Si (see Table 2 for full specification) and has four additive components: production costs [ KP] , R & D costs [ KR] , distribution costs [ KD] and linkage costs [ KL] . All assumptions expounded hold.

Table 2. Cost structure associated with the six possible strategies

Strategies

Production
costs

R & D costs

Distribution
costs

Linkage costs

S1

 

P

R*(1+b )

D

0.5*L+0.5*L*(1+g )+2*L*(1+i )+2*L*(1+i )+2*L*(1+g )*(1+i )

S2

 

P

R

D

0.5*L+0.5*L*(1+g )+2*L*(1+g )*(1+i )+2*L*(1+g )*(1+i )+2*L*(1+i )

S3

 

P*(1+a )

R*(1+b )

D

0.5*L+0.5*L*(1+g )+2*L*(1+g )*(1+i )*2*L*(1+i )+2*L*(1+g )*(1+i )

S4

 

P*(1+a )

R

D

0.5*L+0.5*L*(1+g )+2*L*(1+i )+2*L*(1+i )*(1+g )+2*L*(1+i )

S5

 

0.5*P+0.5*P*(1+a )

R*(1+b )

D

0.5*L+0.5*L+2*L*(1+i )+2*L*(1+g )*(1+i )+2*L*(1*i )+2*L*(1+g )*(1+i )

S6

 

0.5*P+0.5*P*(1+a )

R

D

0.5*L+0.5*L+2*L*(1+i )+2*L*(1+g )*(1+i )+2*L*(1+g )*(1+i )+2*L*(1+i )

As an illustration, take equation K1:

K1 = [ KP] + [ KR] + [ KD] + [ KL]

= [ P] +[ R*(1+b )] +[ D] +[ 0.5*L+0.5*L *(1+g )+2*L*(1+i )+2*L*(1+i )+2*L*(1+g )*(1+i )]

= [ P] +[ R*(1+b )] +[ D] +[ 0.5*L*(2+g )+2*L(1+i )*(3+g )]

Since it corresponds to S1, production and R & D occur exclusively in X, and distribution in both X and Y. Production costs (represented by P) are not penalised by any premium as production is undertaken in the most adequate location, X. R & D occurs in a country comparatively disadvantaged in scientific labour, so premium b is applicable. The first two terms of linkage costs relate to the one-way flow P Þ D. Half of the production is distributed in X and half is exported to Y (incurring the premium g ). The next component of the equation relates to the two-way flow P Û R. Since it is bilateral, the respective cost is multiplied by 2. Then as it is intangible, premium i applies. The last two terms correspond to the two-way flow R Û D. As these linkages involve both countries, linkage costs will arise internally (in X, the only premium applicable is i ) and externally (the flow R Û D in Y has also to incur a premium g as it is international). All other equations follow the same logic, with the peculiarity that, as in S5 and S6 half is produced in X and half in Y, there is a factor of 0.5 applicable to certain cost components. As indicated before, strategies 3, 5 and 6 can be ignored as they are dominated by the others. The basic decision rule is cost minimisation, that is, the theory underlying this model predicts that the strategy with the lowest cost will be chosen.

Analysis and Interpretation of Results

Allowing p and l to vary and solving the system of equations described, it is proved that only strategies 1, 2 and 4 are sensible and that each is valid for a certain range of (p ,l ) combinations, as the following figure shows.

In order to plot this two-dimensional diagram, all the parameters but two need to be fixed. Supported by the previous discussion, Fig. 3 illustrates the production/linkage costs trade-off.

Since the advantage of this model lies in its flexibility and generality, the important aspect does not concern so much the specific values each parameter assumes, but the possibility of putting into the cost equations any value estimated by the potential user of the model, or MNE in question. All manufacturing firms of a reasonable dimension have their cost accounting department which would produce such estimates, and that is the merit of developing a generally applicable model so that each firm can include the specific costs they face. However, the values here used should be sensible, both in absolute and in relative (rank between costs) terms. Of course, what is appropriate depends on the industry concerned.

When such type of estimates is needed, the possibility of calibration should be explored if reliable estimates had been found in previous credible work. However, it was not viable to use calibrated values exactly due to the main advantage of the model, i.e., its generality. Once it is applicable to all industries, distances between countries, types of flows, and so forth, it becomes impossible to suggest a value for each variable. What may be a logical point of discussion is the relative magnitude of the components of the cost structure.

In Figure 3, the following parameter values were assumed:

q = 1000; n = 10; r = 0.4; d = 1; a = 0.5; b = 0.3; g = 0.6; i = 0.2

In this example, the output level and number of linkages were set to respectively 1000 and 10. This does not offer great controversy, any number being acceptable (though the number of linkages will be probably smaller that the number of units produced).

The different premia existent in the model were ranked as follows. The premium to establish an international vs. a domestic linkage was assumed to be higher than the difference between producing the good in the country with the comparative advantage country and in its non-advantaged counterpart. In terms of R & D, the comparative disadvantage onus is slightly smaller than the one concerning production. A relatively small premium penalised intangible flows vs. those of a material nature, mainly due to tacitness and communication difficulties. The greater the importance of tacitness, the higher the premium. It was assumed that it is more expensive to distribute one unit of the product than the amount of unit research costs.

Fig. 3 above shows that for relatively high linkage costs and relatively low production costs, S4 will be optimal. As production costs are relatively low, to produce in X is not so crucial, but due to high linkage costs it is rational to have production and R & D in the same country to avoid further inter-country linkages. Since Y is comparatively advantaged in knowledge-intensive activities, S4 then becomes the best given all assumptions.

When relatively high production costs and high linkage costs occur simultaneously S1 will be optimal. Given high linkage costs it will be important to produce and research in the same country, and due to the important magnitude of production costs production in X is advisable.

Lastly, if linkage costs are relatively low and production costs relatively high, S2 will be the most sensible. S2 reflects complete accordance with the principle of comparative advantage for both production and R & D and thus is the best option in a significant proportion of cases.

Questioning the economic rationality behind the boundaries that can be seen in the diagram, they are determined by the points of tangency of isocosts corresponding to distinct strategies.

Conclusion and Suggestions for Future Research

The paper modelled the strategy set available to a multinational firm when economic integration occurs between two countries in which the MNE was hitherto developing an autarkic approach. The model identifies six possible strategies, which are evaluated given specified assumptions combined with the general principle of cost minimisation. Three of the original strategies are dominated and it is proven that the remaining strategic options define optimal courses of action according to distinct combinations of production and linkage costs.

As could be expected, reconfiguration of the internal organisation of the firm tends to occur as a consequence or a rational reaction to the key environmental shock under analysis (economic integration and the concomitant fluidity of trade and opportunities for organisational redesign that it provides). A systemic view and explicit consideration of the ‘environmental envelope’ of the firm is deemed fundamental to understand this phenomenon.

In terms of future research along these lines, a whole family of models can be derived. Important variations such as vertical integration (considering two productive stages), a multi-product firm (two goods), and specialised R & D operations (basic/adaptation) are potential avenues worth pursuing, at the cost of increasing the number of dimensions and therefore the computational complexity of the model. As presented here, the model is tractable and flexible, and a judicious choice of assumptions will enable the inclusion of these possible variants.

On the other hand, a promising development may be a three-country model that will permit to evaluate the impact on a third country of integration between two other countries. Since an economically integrated scheme is not a hermetic reality, this would represent a step forward by enabling global issues to be discussed, as well as regional and national dimensions. It is expected to address an eventual inter-bloc dialectic in a future paper.

 

References

Ackoff, Russell. 1959. Games, Decision and Organization. General Systems, 4: 145-50.

Aoki, Masahiko. 1986. Horizontal vs. Vertical Information Structure of the Firm. American Economic Review, 76: 971-83.

Ashby, William Ross. 1956. An Introduction to Cybernetics. London: Chapman and Hall.

Beer, Stafford. 1966. Decision and Control: The Meaning of Operational Research and Management Cybernetics. Chichester: John Wiley and Sons.

Beer, Stafford. 1967. Management Science. New York: Doubleday & Company.

Beer, Stafford. 1985. Diagnosing the System for Organizations. Chichester: John Wiley and Sons.

Bertalanffy, Ludwig von. 1968. General System Theory: Foundations-Development-Applications. Harmondsworth: Penguin Books.

Birkinshaw, Julian and Hood, Neil. 1997. An Empirical Study of Development Processes in Foreign-Owned Subsidiaries in Canada and Scotland. Management International Review, 37(4): 339-64.

Buckley, Peter J. and Carter, Martin J. 1996. The Economics of Business Process Design: Motivation, Information and Coordination within the Firm. International Journal of the Economics of Business, 3(1): 5-24.

Buckley, Peter J. and Casson, Mark C. 1976. The Future of the Multinational Enterprise. London: Macmillan.

Buckley, Peter J. and Casson, Mark C. 1998. Analyzing Foreign Market Entry Strategies: Extending the Internalisation Approach. Journal of International Business Studies, 29(3): 539-561.

Burke, Simon P. and Casson, Mark C. 1998. Information Strategies in Foreign Market Entry: Investing in Sales and Distribution Facilities. Paper presented at the Academy of International Business Conference, Vienna, October 1998.

Carter, Martin J. 1995. Information and the Division of Labour: Implications for the Firm’s Choice of Organisation. The Economic Journal, 105(March): 385-97.

Casson, Mark C. 1990. Enterprise and Competitiveness: A Systems View of International Business. Oxford: Clarendon Press.

Casson, Mark C. 1995. The Nature of the Firm Reconsidered: Information Synthesis and Entrepreneurial Organisation. University of Reading Discussion Papers in Economics, Series A, Vol. VIII, No. 332.

Casson, Mark C. 1995b. Information and Economic Organisation. University of Reading Discussion Papers in Economics, Series A, Vol. VIII, No. 317.

Casson, Mark C. 1997. Information and Organization. Oxford: Clarendon Press.

Casson, Mark C. 1997b. Institutional Economics and Business History: A Way Forward?. University of Reading Discussion Papers in Economics and Management, Series A, Vol. X, No. 362.

Casson, Mark C. 2000. ‘The Boundaries of Firms: A Global Systems Perspective’. Chapter 3 in Economics of International Business: A New Research Agenda, Cheltenham: Edward Elgar, forthcoming.

Casson, Mark C. and Wadeson, Nigel. 1996. Information Strategies and the Theory of the Firm. International Journal of the Economics of Business, 3(3): 307-30.

Casson, Mark C. and Wadeson, Nigel. 1998. Communication Costs and the Boundaries of the Firm. International Journal of the Economics of Business, 5(1): 5-27.

Cheung, Steven N. S. 1983. The Contractual Nature of the Firm, Journal of Law and Economics, 26 (April): 386-405.

Coase, Ronald H. 1937. The Nature of the Firm. Economica, New Series, 4: 386-405.

D’Cruz, Joseph. 1986. Strategic Management of Subsidiaries. In Étemad, Hamid and Séguin-Dulude, Louise., editors, Managing the Multinational Subsidiary. London: Croom Helm, 75-89.

Demsetz, Harold. 1991. The Theory of the Firm Revisited. In Williamson, Oliver E. and Winter, Sidney G., editors, The Nature of the Firm: Origins, Evolution, and Development. Oxford: Oxford University Press, 159-78.

Dunning, John H. 1977. Trade, Location of Economic Activity and the Multinational Enterprise: The Search for an Eclectic Approach. In Ohlin, Bertil, Hesselbon, P.O. and Wijkman, P.M., editors, The International Allocation of Economic Activity. London: Macmillan, 395-418.

Dunning, John H. 1993. Multinational Enterprises and the World Economy. Wokingham: Addison-Wesley.

Dunning, John H. and Robson, Peter, editors. 1988. Multinationals and the European Community. Oxford: Basil Blackwell.

Forsgren, Mats and Johanson, J. 1991. Managing Networks in International Business. Philadelphia: Gordon and Breach.

Goldberg, Murray A. 1972. The Determinants of US Direct Investment in the EC: Comment. The American Economic Review, 693-99.

Gupta, Anil K., Govindarajan, Vijay, and Malhotra, Ayesha. 1999. Feedback-Seeking Behavior within Multinational Corporations. Strategic Management Journal, 20: 205-22.

Hayek, Friedrich A. von. 1937. Economics and Knowledge. Economica, New Series, 4: 33-56, reprinted in Individualism and Economic Order (1949). London: Routledge and Kegan Paul.

Hayek, Friedrich A. von. 1945. The Use of Knowledge in Society. American Economic Review, 35(4): 519-30.

Hood, Neil and Young, Stephen. 1988. Inward Investment and the EC: UK Evidence on Corporate Integration Strategies. In Dunning, John H. and Robson, Peter, editors, op. cit.

Jensen, Michael C. and Meckling, William H. 1976. The Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure. Journal of Financial Economics, 305-60.

Lawrence, Paul and Lorsch, Jay. 1967. Organizations and Environments: Managing Differentiation and Integration. Boston (MA): The Belknap Press of Harvard Business Press.

Lunn, John. 1980. Determinants of US Direct Investment in the EEC: Further Evidence. European Economic Review, 13: 93-101.

Marschak, Jacob. 1974. Economic Information, Decision and Prediction, Dordrecht: D. Reidel.

Marschak, Jacob and Radner, Roy. 1972. Economic Theory of Teams. New Haven: Yale University Press.

Nelson, Richard R. and Winter, Sidney. 1982. An Evolutionary Theory of Economic Change. Cambridge (MA): The Belknap Press of Harvard University Press.

Pearce, Robert D. 1992. World Product Mandates and MNE Specialisation. Scandinavian International Business Review, 1(2): 38-58.

Pearce, Robert D. and Papanastassiou, Marina. 1997. European Markets and the Strategic Roles of Multinational Enterprise Subsidiaries in the UK. Journal of Common Market Studies, 35(2): 241-66.

Penrose, Edith. 1995. Foreword to The Theory of the Growth of the Firm (Third Edition). Oxford: Basil Blackwell.

Polanyi, Michael. 1967. The Tacit Dimension. London: Routledge.

Radner, Roy. 1992. Hierarchy: The Economics of Managing. Journal of Economic Literature, 30 (September): 1382-1415.

Richardson, George. 1972. The Organization of Industry. The Economic Journal, 82(September): 883-96.

Sah, R.K. and Stiglitz, Joseph E. 1986. The Architecture of Economic Systems: Hierarchies and Polyarchies. American Economic Review, 76(4): 716-27.

Scaperlanda, Anthony and Balough, R. 1983. Determinants of US Direct Investment in the EEC: Revisited. European Economic Review, 21: 381-90.

Scaperlanda, Anthony and Mauer, L.J. 1969. The Determinants of US Direct Investment in the EEC. American Economic Review, 59(4): 558-68.

Seiler, J. A. 1967. Systems Analysis in Organizational Behaviour. The Irwin-Dorsey Series in Behavioural Science, Homewood (Illinois): Richard D. Irwin, Inc. and The Dorsey Press.

Shannon, Claude E. and Weaver, Warren. 1949. The Mathematical Theory of Communication. Urbana (Il): University of Illinois Press).

Simon, Herbert A. 1997. Administrative Behaviour (Fourth Edition). New York: The Free Press.

Taggart, James. 1996. Multinational Manufacturing Subsidiaries in Scotland: Strategic Roles and Economic Impact. International Business Review, 5: 447-68.

Tavares, Ana T. and Pearce, Robert D. 1998. Regional Economic Integration and the Strategic (Re) Positioning of MNEs’ Subsidiaries: A Conceptual Investigation. University of Reading Discussion Papers in International Investment and Management, No. 254, Series B, Vol. XI.

Teece, David. 1983. Technological and Organizational Factors in the Theory of the Multinational Enterprise. In Casson, Mark C., editor, The Growth of International Business. London: George Allen and Unwyn.

Wadeson, Nigel. 1997. Two-way Communication Costs and the Boundaries of the Firm. University of Reading Discussion Papers in Economics and Management, Series A, Vol. X, No. 370.

White, Roderick and Poynter, Thomas. 1984. Strategies for Foreign-Owned Subsidiaries in Canada. Business Quarterly, 49(2): 59-69.

 

NOTES

(1). This reflects the idea immanent in the 'principle of materiality' of accounting (an information is 'material' when the decision taken after the reception of information is distinct than the one taken in its absence).
(2). See Casson (1997; 1997b) for an explanation of the distinction between information costs and transaction costs. Although the issue cannot be discussed here at length, it is incorporated in the model provided.
(3). The efficiency of discrete choice models is inversely proportional to the number of dimensions considered, and if the number of these dimensions increases, a problem known as 'combinatorial explosion' (Casson and Wadeson, 1996) is likely to occur.
(4). Which classifies distinct subsidiaries along the dimensions of market, product, and value-added scope.
(5). Previous empirical studies (Hood and Young, 1988; Taggart, 1996; Pearce and Papanastassiou, 1997; Tavares and Pearce, 1998) concluded that this is indeed the most common evolutionary path in subsidiary development.
(6). The numbers in this Figure denote the distinct strategies explained in Table 1.
(7). Both on the grounds of zero marginal cost of transfer/use and non-excludability, the two defining characteristics of a public good. First of all, knowledge transfer is frequently costly, and the more tacit (Polanyi, 1967) it is, the more onerous it tends to be to transfer it. Non-excludability might not hold, as it is often possible to prevent other agents/institutions from accessing the knowledge (e.g. MNE/subsidiaries).
(8). Teece (1983) analysed explicitly the trade-off between production costs and governance/transaction costs. The present category of linkage costs is more general than Teece's transaction costs, encompassing also information costs of a non-transactional kind, that are imperative for the efficient carrying out of the firm's activities. Distribution and R & D costs are also contemplated in the present analysis, even though due to the specified assumptions they do not impact on the choice of the best strategy in the way linkage and production costs do.