THE IMPACT OF EMBEDDEDNESS ON THE ECONOMIC PERFORMANCE OF FIRMS

This paper presents the main existing notions of embeddedness and emphasizes its impact on the economic performance of a firm. Many scientists have been trying to elaborate this confusingly polyvalent concept, but still there is now a plethora of meanings and definitions of what embeddedness might be or consist of. Different distinctions between the dimensions and levels of embeddedness exist, the most popular among them being the micro and macro level distinction of embeddedness. In this paper, attention is focused on the micro level and the effects that embeddedness exerts on the performance of firms. The main positive and negative features of embeddedness in the economic performance of a firm are presented.

The concept of embeddedness is attaining attention of many researchers, economists and social scientists in particular. Among them, economists have been eager to embrace the notion of embedded ness, because it assumes that firms are closely linked to their local production environments in a world of increasing globalisation. Embeddedness not only accounts for the importance of trust-based networks for regional development, but also incorporates the idea that socio-cultural and institutional factors may be essential for a good economic performance of firms. Therefore, trying to explain the reasons why under the same market conditions some of the economic actors are able to act in the market and others are not, the concept of embeddedness becomes one of the main notions that are able to clarify those non-market factors. By this time, most of scientific studies have been focused on the societal benefits that arise from the process of embeddedness. However, there is an urgent need to unravel the notion of embeddedness and its impact on the economic performance of the firm.
Many well known scientists starting with Polanyi (1944), Granovetter (1985), then Uzzi (1996), Woolcock (1998), Zukin and DiMaggio (1990), Oinas (1997), Boschma (1999), Hess (2004) and others try to elaborate this confusingly polyvalent concept, but still there is now a plethora of meanings and definitions of what embeddedness might be or consist of. Some of them emphasize distinction between political, cultural, structural and cognitive mechanisms. Others make a distinction between micro and macro embeddedness, focus on temporal embeddedness, technological embeddedness, or recognize embeddedness as having four basic forms -cognitive, cultural, political, and structural. In this paper, attention will be focused on the micro level and the effects that embeddedness exerts on the performance of economic actors (firms). All these different distinctions show that the notion of embeddednes is very complex and ill-defined. In Lithuania, the notion of embeddedness has not yet been widely analysed; accordingly, the impact of this phenomenon has not been explored so far. Therefore, the demand to present the main existing ideas about this concept, to highlight its impact on economic performance of firms is relevant.
The purpose of this paper is to present the main existing notions of embeddedness and to clarify its impact on the economic performance of a firm. Therefore the main tasks are: • to overview the main theoretical perspectives of embeddedness • to present the dimensions and levels of embeddedness • to clarify the impact the process of embeddedness may have on the economic performance of a firm.
The paper is based on analysis of the scientific literature.

Theoretical perspectives of embeddedness
According to Oinas (1997), the notion of embeddedness seems to capture all possible aspects in a firm's environment. This is why it 66 is problematic: it encompasses too many things with the result of being ambiguous. Boschma et a1. (1999) state that the concept of embeddedness is imprecise and ill defined.
However, there is a number of conceptual problems with regard to embeddedness. This becomes clear after having a look at the social science literature, including economic geography and business studies, where a variety of meanings linked with embeddedness appear. Many studies in the new regionalism tradition pay attention almost exclusively to local and regional systems of economic and social relations, arguing that the 'local' embeddedness of actors leads to an institutional thickness, which is thought to be a crucial factor of success for regions in the continuously globalizing economy. While every single publication about embeddedness unfailingly pays tribute to Granovetter's (1985) seminal paper and also mentions Polanyi's (1944) contribution, the question remains of the extent to which the theorizations of embeddedness used in the more recent literature have or have not moved away from the original concepts elaborated by Polanyi (1992). This means that the analytical scales of embeddedness need to be scrutinized, both in the original concepts and in their recent adoptions, in order to get a clearer and more consistent understanding of who or what the socially embedded actors (firms or individuals) are, and in what these actors are actually embedded.
Such conceptual problems as the lack of clarity have recently been highlighted by Markusen (1999) in her paper on 'fuzzy concepts' in regional analysis, which has started a lively debate about recent developments in critical regional studies (Hudson, 2002). Interestingly, A. Markusen does not explicitly refer to embeddedness as a fuzzy notion, although it clearly has all the attributes of one. While using the concept of 'networking and co-operative competition in industrial districts' (Markusen, 1999) as an example of fuzziness, the related and underlying theory of embeddedness does not attract her attention.
Polanyi can without doubt be considered the father of the embeddedness concept (Swedberg and Granovetter, 1992;Barber, 1995). Arising from a strong dissatisfaction with the absolutization of the market and its underlying rationale of self-regulation and economizing behaviour, he demonstrated that the economy is enmeshed in institutions, both economic and noneconomic (Polanyi, 1992). He called this view a substantive definition of economics, as opposed to the formal definition supported by economists and market ideologists.
As Gertler (2001) stated, if accepting the central insight of Polanyi (1944) that the market is a socially constructed and governed and not a "natural", given, inevitable form, then it makes perfect sense that firms in market economies should also be modified to some extent by their social-institutional environment. This view is shared by the business systems literature, which follows a similar argument (Whitley, 1999;Kristensen, 1996). According to their exponents, business systems can be defined as different kinds of economic coordination and control systems shaped by an institutional environment specific to different societies and nation states. Although the notion of embeddedness is not specifically elaborated in this strand of research, it is quite obvious that there is an inherent understanding of the societal embeddedness of firms in their national and macro-regulatory environment. As Whitley (1999) argues, even under globalization there is still a tendency for national business systems to retain their specific character-istics. Thus emerges a variety of capitalisms that resists tendencies of a global homogenization of organizational models and a corresponding convergence of business systems. While both Polanyi and Whitley emphasize the role of society in shaping the economy, the business systems literature clearly focuses on particular economic actors, namely on the firm. By doing so, it moves away from Polanyi's more structural conceptualization towards another approach which has had the biggest influence on embeddedness research until today.
Following Granovetter (1985), the embeddedness literature has blamed neo-classical economics for an 'under-socialised' view of economic relations, which emphasises a rational, self-interested behaviour hardly affected by social relations. Neo-classical economics regards actors as individuals who act independently and maximise their utility, and exchange goods and services in one-off deals, based solely on price and quality signals. Their actions are devoid of social context, divorced from social norms, social networks and trust (Coleman 1990). This view gives a normative basis to the free market, in which social relations are regarded as obstacles to the proper functioning of competitive markets.
As Uzzi (1996) noted, the concept of embeddedness does not explain concretely how social ties affect economic outcomes. Embeddedness as such does not clarify the processes of adjustment of entrepreneurs. It only points to the fact that economic actors use more variables than prices and profits alone on which to base their decisions. Moreover, it is far from clear whether embeddedness may have the positive economic effects suggested by Granovetter's notion of the strength of weak ties. In fact, it may have adverse impacts because of lock-in effects. Woolcock (1998) presents a simpler typology of social capital which builds on the notions of 'autonomy' and 'embeddedness', which can be related back to Beeker and Granovetter. Embeddedness is derived from the strength and solidarity of the group -its level of integration and social cohesiveness within itself. In Granovetter's (1985) and Woolcock's (1998) opinion, all forms of exchange, be they cultural, economic, political, or social, are inherently embedded in social relationships (Granovetter, 1985;Woolcock 1998). The level of embeddedness will determine the durability of the groups and how they respond to stress. Likewise, however, these groups with high embeddedness can become more vulnerable to certain types of external shocks due to their lack of flexibility and responsiveness; and they can have quite oppressive or restrictive effects on non-members. It is not only vulnerability that is a cost of embeddedness, but it also constrains choices, flexibility, and sources of infonnation for individual members of the group.
The classifications and typologies have become even more complex and often confusing as the subsequent literature has added more and more forms of embeddedness to the ones already existing. While the issue of agency in the organization business literature has pretty much homed in on firms and their networks, the question of what the firms are embedded in has generated multiple meanings and a number of research perspectives on network embeddedness (Halinen et al., 1998). In Halinen and 'Ibrnroos's paper on the evolution of business networks, the reader is provided with three perspectives (actor-network, dyad-network and micronet-macronet), two dimensions (horizontal and vertical) and six types of embeddedness: social, political, market, technological, temporal and spatial.
Halinen and Tornroos argue that business exchange is embedded in various technological systems and in the development of these systems at the corporate and societallevels. By this they mean that firms are dependent on particular technologies, a view that departs from the common ground of embeddedness concepts discussed thus far. The same is true for their discussion of market embeddedness, pointing out the fact that each business actor is embedded in a specific market defined in terms of products and services offered, -the clientele served, the functions perfonned and the time and territory encompassed by the company's operations. Apart from the fact that in this definition the actors are no longer embedded in a set of social relations but rather in markets defined as 'products' and 'services', these markets are concomitantly seen as a temporal and spatial framework in which the business actors are embedded. This actually makes it unnecessary to conceptualize the distinct types of temporal and spatial embeddedness.
Embeddedness plays an important role in social relations among the economic actors. This is one of the reasons why the new economic geography has adopted the concept of embeddedness since the early 1990s, starting with the work of Dicken and Thrift (1992), firms are produced through a historical process of embedding which involves an interaction between the specific cognitive, cultural, social, political and economic characteristics of a firm's 'home territory', geographically dispersed operations and the competitive and technological pressures which impinge upon it. It is worth noting that this understanding of embeddedness actually resembles more of Polanyi's original idea of societal embeddedness, i.e. the history of economic actors and the cultural imprint of the 'home territory' rather than emphasizing only locally 'bounded' economic activities as in much of the subsequent geographical literature. It therefore is closer to the original embeddedness conceptualizations than most economic-geographical interpretations since.
However, the concept of embeddedness still is imprecise and ill-defined. In the model of local growth it has spawned in geography, sociology and some areas of management science, the concept is undeniably vague, but it has directed attention to the nature of relationships between finns and their socio-spatial environments that are neither well understood nor particularly well conceptualized (Oinas 1997). Therefore there is a need to overview the concept of embeddedness in more detail. emphasizing the dimensions and levels of this concept.

Dimensions and levels of embeddedness
As noted before, embeddedness is an increasingly popular but confusingly polyvalent concept. Indeed, there is now a plethora of meanings and definitions of what embeddedness might consist of, the most prominent classification probably being Zukin and DiMaggio's (1990) distinction between political, cultural, structural and cognitive mechanisms (Baum et al. 1996;Tzeng et al., 2000). Other authors make a distinction between micro and macro embeddedness (Halinen et al., 1998;Fletcher et al., 2001), different forms of social embeddedness (Jessop, 2001), or focus on temporal embeddedness, technological embeddedness, etc. These terrninologies need to be unravelled if we want to get a clearer picture of the common ground and substantive meanings of different concepts.
Following Maskell's (1998) opinion, structural embeddedness identifies the manner in which firms are incorporated into local, placebased networks that facilitate and promote information exchange and learning. Moreover, Grabber (1993) steted that structural embeddedness has been recognized as having four essential characteristics: reciprocity, interdependence, loose couplings and asymmetric power relations: • Reciprocity refers to recurrent transactions between networked firms that are more than simply repetitive and involve relationships that do not have immediate equivalence in each transaction but achieve some approximate balance over the life of an exchange relationship (Polanyi, 1944). • Interdependence reflects the elements of trust and mutual orientation in longterm exchange relationships that enable firms to exchange resources and information crucial for high pedorrnance but difficult to value and transfer via market ties (Uzzi, 1996). It is central to network learning and local innovation capacities. • Loose couplings, or integrated separateness, recognizes the ability of firms networked in a place individually to shift their partners while maintaining an essentially stable district framework of interaction. On the other hand, Hess (2004) introduces much wider perspectives. He points out that embeddedness basically signifies the social relationships between economic and non-economic actors (individuals as well as aggregate groups of individuals, firms), and economic action is grounded in societal structures.
Societal embeddedness. According to Hess (2004), it signifies the importance of where an actor comes from, considering the societal background, influencing and shaping the action of individuals and collective actors within their respective societies and outside them. M. Hess considers this type as the most closely linked with the original idea of embeddedness. Societal embeddedness also reflects the business systems idea of an institutional and regulatory framework that affects and in part determines an actor's behaviour on the individual level via the cognitive mechanisms detailed by Zukin and DiMaggio (1990), or on the aggregate level of a firm as pointed out by Whitley (1992) and his colleagues.
Network embeddedness. Following Hess' (2004) opinion, it describes the structure of relationships among a set of individuals and forms regardless of their country of origin or local anchoring in particular places. It can de defined as the architecture, durability and stability of these relations, both formal and informal, which determines the actors' individual network embeddedness as well as the structure and evolution of the network. While the former refers to an individual's or firm's relationships with other actors, the latter consists not only of business agents involved in the production of a particular good or service, but also takes the broader institutional networks including non-business agents (e.g., government and non-government organizations) into account. As Hess (2004) stated, 70 network embeddedness can be regarded as the product of a process of trust building between network agents, which is important for successful and stable relationships. Even within intrafirm networks where the relationships are structured by ownership integration and control, trust between the different firm units and the different stakeholders involved might be a crucial factor, such as in the case of joint ventures (Hess, 2004).
Territorial embeddedness. In Hess' (2004) opinion, this type of embeddedness considers the extent to which an actor is 'anchored' in particular territories or places. Economic actors become embedded there in the sense that they absorb, and in some cases become constrained by the economic activities and social dynamics that already exist in those places. An example here is the way in which the networks of particular firms may take advantage of clusters of small and medium enterprises that predate the establishment of subsidiary operations by such firms. According to Hess (2004), the location or anchoring down of external firms in particular places might generate a new local or regional network of economic and social relations, involving existing firms as well as attracting new ones. As Harrison (1992) and Amin (1994) noticed, under mentioned circumstances embeddedness may become a key element in regional economic growth and in capturing global opportunities. From the development point of view, then, the mode of territorial embeddedness or the degree of an actor's commitment to a particular location is an important factor for value creation, enhancement and capture.
These three dimensions of embeddedness are of course closely knitted to one another and, in combination, form the space-time context of socio-economic activity. In Hess' (2004) opinion, trying to create a spatial-temporal concept and to avoid a static view of agency and social structure, the three proposed categories of embeddedness have to consider developments over time and changes in the spatial configuration of networks on different scales. 10 paraphrase AmiD's (2002) notion, with a historicized sense of scale, embeddedness (like globalization) can be interpreted as a spatial process elevating the tension between territorial relationships and transterritorial developments.
Besides the mentioned above, the process of embeddedness can be defined in several other ways. According to Boschma et al.
(1999), two major levels at which social relations may affect economic behaviour and performance can be distinguished. At the first level, there are the social relations with known actors that involve dyadic relations or a larger network of relations (mostly based on bonds of friendship or kinship) that associate with the social embedding of inter-firm relationships. This interpretation refers to the social dimension of economic relationships between firms at the micro-level The micro-level dimension of embeddedness. A more limited meaning of embeddedness is associated with the social embedding of economic relations. Boschma et al.
(1999) elaborates on insights of transaction costs economics and network analysis in order to build a picture of the main differences between market-based and embedded economic relationships. This implies that embeddedness may be of different degrees, and that the neoclassical form of economic transaction should not be ruled out just because it ignores the social dimension. Following Uzzi's (1997) opinion, it is very likely that reality involves a mix of both types of transaction, creating a continuum of embeddedness.
Thble 1 gives a short overview of the key contrasts between market-based relationships and embedded relationships at the microlevel. First, the embedded form of economic exchange involves a sequence of reciprocated transactions. According to Uzzi (1996), network theory argues that embeddedness shifts the actors' motivations away from the narrow pursuit of immediate economic gains towards the enrichment of relationships through trust and reciprocity. Second, profitmaximising motives guide the market-based transaction that is limited to the exchange of information on price and quality. This stands in contrast to embedded relationships. Interactive learning presupposes an orientation to "communicative rationality", that is, an orientation to an understanding that transcends the  (Grabher, 1993). Third, the stability of network-based interactions leads to interdependence between actors, rather than the discrete exchange relations between independent actors in markets, and coordination processes within hierarchies. Fourth, conflicts in embedded ties are resolved through voice-based strategies based on joint-problem solving rather than walking out strategies in market exchanges. Finally, embedded relationships are characterised by a loose coupling, in which the exchange partners retain some autonomy. Grabher (1993) maintains that networks consist of actors possessing power, which he regards as a functional element of networks. Power can be considered an element that obscures rather than clarifies the lines drawn between market-based and embedded relationships. In market-based relationships, power may, or may not, be involved, depending on the positions of trading actors. The same applies for embedded relationships, which are reflected by the different positions in the literature with respect to the impact of power in networks on innovative performance. It can be summarised that the industrial district literature puts emphasis on the collaboration between equal network partners, which is regarded as a stimulus for interactive learning. In contrast, Grabher (1993) points out that a radical change requires power in networks. Moreover, Thylor (1999) has stressed that network partners often follow strategies of exclusion, which are to the loss of newcomers and new initiatives.
The macro-level dimension of embeddedness. Considering the socio-cultural environment the firms operate in, the macro level of embeddedness can be distinguished.

72
There can be made a distinction between the 'institutional environment' at the macro-level (such as norms and values of conduct) and institutional arrangements at the micro-level, in which these norms and values are embodied in specific exchange relations.
According to Zukin and Di Maggio (1990), the contextual dimension of embeddedness consists of many aspects. Political embeddedness identifies economic actors embedded in the institutional rules of the game. Cultural embeddedness refers to sets of shared values, like trust, that may be specific to a group of entrepreneurs. Finally, cognitive embeddedness deals with the knowledge aspects of embeddedness. It is expedient to restrict attention to the notion of social capital, which comes close to the broad notion of cultural embeddedness. Considering social capital, it refers to features of social organisation, such as networks, norms and trust, that facilitate coordination and co-operation for mutual benefit. The notion of social capital is considered relevant here, because it puts emphasis on the nature of socio-cultural relations that may bring and hold together economic actors. In this respect, it functions as a sort of 'glue' for collective action. Therefore, this concept may bridge the gap between the contextual dimension of embeddedness at the macro-level and the social embedding of inter-firm relationships at the micro-level. More precisely, social capital may constitute a productive resource that enables cooperation and increases performances of firms due to lower transaction costs, brings about the collective learning process and flexibility. Boschma et al. (1999) also emphasise the geographical proximity. According to them, firms located close to each other have more face-to-face contacts and can easily build up trust, which leads to more personal and thus embedded relationships between firms. But that does not mean that embedded relations are necessarily of a local nature. Although often suggested otherwise, embeddedness may well have a non-local dimension (Oinas 1997). In this respect, Hausmann's (1996) distinction between organisational, social and spatial proximity is relevant. He claims that for inter-organisationallearning, social or organisational proximity may be more important than spatial proximity, but that spatial proximity strongly facilitates these effects.

The impacts of embeddedness on the economic performance of firms
Firstly, it is worth mentioning that the notion of embeddedness has changed the characterisation of the firm itself. According to Boschma et al. (1999), firms are regarded as open systems. This creates a certain fuzziness in the boundaries of production units and firms. As Harrison (1992) stated, firms relate to one another by interpenetrating one another's formal organizational boundaries rather than solely through the price-mediated exchange of commodities. This implies that the acquisition of some inputs, like knowledge, is not a clear matter either. Knowledge as a resource is difficult to contain within one production unit or firm for a long time according to Lambooy (1997), because it is related to human experience and interaction and thus to a collective process sensitive to social values.
In economic geography, Granovetter's notion of embeddedness has been widely adopted, because it revives the idea that firms are firmly linked to their local production environment in a world of increasing globalisation. Harrison (1992) has applied this notion to the phenomenon of industrial districts. According to him, the industrial district is all about interdependence of firms, flexible firm boundaries, co-operative competition and the importance of trust in reproducing sustained collaboration among economic actors. Storper (1997) argues that not only market relations are important, but that more encompassing systems of interdependencies need to evolve, especially within regions. The conclusion of Storper's argument is that enterprises cannot really survive without non-market interdependencies, which are a condition for and a result of a certain embeddedness of economic actors in regional structures.
As noted before, the concept of embeddedness itself does not provide an explanation for economic performance. There can be many aspects to this question, but here the focus is put on the micro-level of inter-firm relationships. In this section the main idea is to link more directly the notion of embeddedness to the adaptive capability of firms, which is essential for their competitiveness. We outline the positions adopted in the economics literature, which reduce to the positive aspects (strength of weak ties) and negative sides (lock-in) of embedded inter-firm relationships, thefore an inverted-U relationship between the degree of embeddedness of a firm and its innovative performance can be proposed in context.
Granovetter's principle of strength of weak ties underpins the positive effects of embeddedness on the micro-level economic performance. According to Boschma et al. (1999), four advantages of embeddedness can be defined. Closely related to these factors are the characteristics of embedded inter-firm relationships shown in Thble 2.
In Thble 2 there are summarised the positive effects of embeddedness on firms. The negative Table 2. Advanlllges of embeddedness and its positive effect on firms Advantage of Characteristics of embeddedfirms embeddedness

Lower tnns-action
• Trust-based supplier relationships reduce the risk of opportunistic behacosts viour by exchange partners and reduces transaction costs on the specification and monitoring of contracts • When trust is high, there is less need to specify all the details of a transaction in fonnal written contracts • Norms that are shared effectively constrain opportunistic behaviour, so the need to control and monitor transactions is also reduced 2. New cooperative • Cooperation is vital for the competitiveness of small firms institutions • As Harrison (1992) states, firms are said to co-operate on getting new work into the district, in forming consortia to obtain cheap credit, in jointly purchasing raw materials, in bidding on large projects and in conducting joint research 3. Greater RexibiHty • Embedded relationships greatly enhance flexibility, because the partners retain some autonomy, and autonomy prevents lock-in • In strongly embedded networks, independent and autonomous firms both fiercely compete and closely cooperate • With good communications, independent partners may be able to shift goals and strategies more easily 4. Organisational learning • Embedded relationships favour the transmission and exchange ofknowledge and information and, thus, learning and innovation (Boschma 1999) • Because of joint problem solving, trust and dense information exchange, embedded relationships stimulate intel'active learning and innovations • They lower the costs of search, trust-based relationships also facilitate the exchange of tacit knowledge which is much more difficult to communicate and to trade through markets • Embedded relationships reflect a social and open attitude of 'communicative rationality', which is conducive to interactive learning effects of embeddedness on the micro-level economic perfonnance can be associated with the weakness of strong ties, or 'lock-in'. In essence, this comes down to a poor ability to interpret new infonnation or an incapability to adjust accordingly. There may be several reasons for this, such as loyalty, long-tenn commitment and cognitive lock-in (in the sense that routines in inter-firm relationships obscure the view on new technologies or new market possibilities). These arguments are briefly explained below.
tions are based on emotional bonds of friendship and kinship. According to Nooteboom (1999), strong ties may have the disadvantage of generating too much personal interaction and loyalty, to the detriment of productive work, criticism and flexibility. In this respect, embedded relationships, based on trust and positive values, provide firms with a false sense of certainty. Accordingly, too much of 'social behaviour' may have negative consequences in a world with calculating actors, in markets where technologies and policies continually change in conditions of uncertainty, and where opportunism is a common attitude.
In Boschma's et al. (1999) opinion, embedded ties hold the possibility of underestimating opportunism, especially when rela-Another negative effect can become a longterm or too much commitment, because they may lock buyers and suppliers into established ways of doing things or into specific technological trajectories at the expense of their own innovative and learning capacity. In this way, they not only become scaled off from new market developments, but also they are incapable of adapting. Thinking evolutionarily, Boschma (1999) and Hudson (1999) explained it in terms of 'routines', 'path-dependent' behaviour, and cognitive 'lock-in'.
Eventually, there is a considerable controversy in the economics literature about the effects of embeddedness on the micro-level economic performance. On the one hand, neoclassical economics would claim that embedded relationships are inefficient and will have a negative impact on the performance of firms. On the other hand, the embeddedness literature would say the opposite: the more embedded the relationships of a firm are, the more superior its performance, because its social relationships facilitate the communication and exchange of knowledge and thus learning and innovation.
Therefore, following Uzzi's (1996Uzzi's ( , 1997 proposal, it is expedient to go beyond this debate by integrating these two contrasting views into an inverted-U shaped relationship between ernbeddedness and innovative performance at the firm level. By doing so it might be proposed that the social dimension of economic relationships has a positive influence on the performance of a firm up to a certain threshold, after which adverse impacts come into being as a result of lock-in (contrary to the embeddedness model). According to Uzzi (1997), in the beginning embedded inter-firm exchange promotes the eco-nomic performance of firms. Collective agreements, shared investments, concentrated exchange of information and knowledge between partners, characterised by interdependency, reciprocity and trust, lower transaction costs, reduce risks and uncertainty and increase the access to tacit knowledge. In other words, social relationships stimulate learning and innovation. It is worth noting that these positive effects can turn negative when the embedded relationships become too closely tied. Then the economic efficiency becomes vulnerable to unforeseen, exogenous shocks that may ruin previously safe exchange relationships, especially when these relationships are rather isolated from vital external information on markets or technology, what results in the loss of competitiveness.
This expected relationship has been to some extent observed and verified in various empirical studies by Uzzi (1996Uzzi ( , 1997Uzzi ( , 1999 and he came to the conclusion that firms organized in networks have higher survival chances than do firms which maintain arm'slength market relationships. The positive effect of embeddedness reaches a threshold, however, after which point the positive effect reverses itself (Uzzi 1996), so he suggests that an optimum in terms of adaptive capacity can be reached when the network of a firm consists of a mixture of arm's-length ties and embedded relationships.

Conclusion
The aim of this paper was to clarify the meaning of the concept of embeddedness and to highlight its impact on the economic performance of firms. To do so is a challenging task, because the notion of embeddedness has something promising to offer. Summarising it can be said that the behaviour of economic actors may be anchored in trust-based inter-firm networks in which personal relations, long-term interdependence, knowledge exchange and reciprocity are key the elements. In this respect, embeddedness may be of great importance for the economic performance and competitiveness of firms and, in the end, for the economic performance of regions. Reassessing the ideas of Polanyi and Granovetter, it becomes clear that embeddedness plays a crucial role in economic activities, not only in premodern societies but also in modern market economies. Besides, it is not only the price mechanism that shapes the nature of economic exchange, but also the social interaction of individual and collective actors.
In the literature on embeddedness there is a big variety of different interpretations of the dimensions of this concept. Despite the other typology, the distinction between a micro-level dimension of embeddedness and a macro-level dimension of embeddedness allows to link the contextual dimension of embeddedness (here defined as social capital) to the nature and the quality of exchange relationships between firms.
Summarising the main ideas about the impact of embeddedness on the economic performance of firms, four main conclusions can be distinguished: 1. Embeddedness is hard to measure empirically, because it may be manifested at various levels. Generally, the higher the level of analysis, the harder it is to define or measure embeddedness. According to Oinas (1997), the embeddedness of a firm should be interpreted somehow as the sum of individual embeddedness of key decision-makers (owners, managers, key employees) in a firm, and the embeddedness of the firm as a collective actor in its external environment. 2. It is difficult to find a relation between the social context at the macro-level and embedded relationships at the microlevel. This is because the firms' strategies may actively shape and change their social relations and their surrounding socio-cultural environment. 3. In the micro-level analysis, explanation of the embedded relations is based mainly on the concept of social capital and too little attention is paid to other factors that may affect the performance of a firm. Moreover, it seems important to control the size of firms and the durability of relationships, although it is still uncertain what impact these parameters might have on the social embedding of economic performance. 4. There is considerable controversy in the literature about the relationship between embeddedness and innovative performance. In this paper, it was visualised as the anticipated combination of positive and negative effects of embedded inter-firm relationships on innovative performance as an inverted Ushaped relationship by extending Uzzi's work on networks. However, there is a need to identify what factors shape this inverted U relationship, and the concept of embeddedness needs to be explored more deeply, both theoretically and empirically.