This introductory paper to the issue puts in perspective the various dimensions that are implied by contemporary innovation processes as a collective phenomenon within and across firms and then stresses the new forms of coordination that arise as responses to the challenges posed by multi-agent, uncertain and dynamic innovation processes.

The “interactive” model proposed by Kline and Rosenberg (1986) shows that innovation proceeds by feedback between R&D, design, manufacturing and marketing services. In other words, the genesis of innovation is the result of systemic linkages between knowledge and the market (Laperche, Uzunidis, Von Tunzelmann, 2008). Global competencies are developed for the adaptation of the firm to its environment and for the innovative process. Productive competencies, marketing, human resources, financing competencies must thus be taken into account (Munier, 2006). They express the possibility of assimilating internal and external information, of creating knowledge and of developing aptitudes to pose new problems and to bring answers, i.e. to innovate. The importance of competencies is also at the level of the relations, which the firm weaves with external partners. Cohen and Levinthal (1990) introduce the concept of “absorptive capacity” to describe this phenomenon. The capacity of the firm to develop an internal knowledge base and to exploit external knowledge is crucial for developing an innovation. Hence, studying the constitution of the firm’s “knowledge capital” (Laperche, 2007), i.e. the set of information and knowledge produced, acquired and combined by the firm in order to innovate is crucial to understand the innovation process.

The current keyword of the innovation process and of the constitution of the firm’s “knowledge capital” is the one of “network”. Complex networks associate large firms, start-ups, universities, public and private research labs, funding institutions, consultancy and specialised business services, etc. dedicated to the innovation process. The importance of networks reveals several major features of the innovation process which have taken place since the beginning of the 21^st century.

First of all, neither innovation, nor networks are Deus ex machina. The complex interactions between institutions are the result of –and result in – a change in the way of considering territories (as generators of resources and not just as blocks of factors) and public policies (giving incentives to the constitution of networks, revealing and accompanying the advantages of territories through short term and structural policies of attractiveness). For any firm, either large and well established or small and young, cooperative arrangements and networking and clustering moves have become inescapable strategic keys for producing/accessing new knowledge and competences, and for being part of the innovative system and new markets in a given industry. Besides the debate initiated by Schumpeter (1912, 1942) which opposes the “entrepreneurial model” vs. the “large firm model”, there are now solid arguments and growing evidence that these two models combine in a new broader model (the “networked model”) based on strong collaboration and networking/clustering among various complementary and interdependent actors.

This model re-shapes innovation as well as competition dynamics on more collective grounds, putting the accessibility to the most strategic partners at the heart of firms’ ability to innovate and to “stay in the game” (Hamdouch, 2007). It also increasingly enlarges the competitive game from mere inter-firm rivalry towards intra- and inter-coalitions, networks and clusters rivalry within and across various spatial scales (cities, regions, countries). Hence, innovation and knowledge creation-diffusion-accumulation processes underlie a complex and dynamic institutional and spatial framework in which industrial strategies, public policies and socially and institutionally embedded behaviours and interacting mechanisms, shape the specific trajectories and performance displayed by firms and territories in a given industry (Hamdouch and Moulaert, 2006).

All these new dynamics are nowadays acknowledged by most national and regional governments (and even local policy-makers) as reflected in the overwhelming “cluster initiatives” one can observe in developed countries as well as in emerging or developing economies. But collaborative and networking-clustering dynamics cannot simply be initiated overnight by the sole virtue of political volunteerism or by the strategic aim of a single firm or institution. These dynamics build on a specific “alchemy” in which various historical, socio-economic and institutional dimensions of the time-space features characterising an industry within and across various territorial scales give opportunities to different actors to engage in innovative activities and to build on local or more distant collaborative relationships. In this perspective, social, scientific and business networks and the way they emerge, structure and evolve appear to be the crucial components of effective and lasting clustering dynamics within certain territories but not in others (Hamdouch, 2008).

Secondly, networks reveal the “necessity” of collaboration, which is due to the complexity, the cost and the risk of the innovation process. The growing burdens on financial investment for the organisation of productive activities implies cooperation between firms and institutions to facilitate continuous, “permanent” development of profitable new goods and services. It thus explains the renewal of the entrepreneur’s function (notably “intrapreneurship,” spin-offs, and outsourcing) as well as the invention of new organizational and inter-organizational modes (alliances, partnerships, networks, clusters) (see Laperche and Uzunidis, 2008). The rise of finance also leads to questions as to the nature of the innovations that result from this collective process. The potential of development of major (and socially useful) innovations stemming from the collective nature of the innovation process seems to be thwarted by the “financialization” of economic activities.

This idea is well explained by Claude Serfati, in the second article of this issue. Multinational corporations, or transnational corporations, are defined by C. Serfati as a category on their own, based upon the centralization of financial assets and a specific organization structure, where the holding company occupies a core position. Financialization induces changes in the governance of the global value chains including the growing rise of dividend pay-outs, the increase in appropriation of value and the decline in production activities, and most importantly, the growing rise of intangible assets that can be considered as a driver for financialization. These new characteristics of large firms’ organization and strategy have strong impacts on innovative activities. The first impact stressed by Claude Serfati is the increase of rent appropriation, induced by the strengthening of intellectual property rights. The second is the reorientation of R&D expenditures toward more short term developments.

Thirdly, networks cannot be and cannot become stationary (referring to the words used by Joseph Aloïs Schumpeter to qualify capitalism: see Schumpeter, 1942). Cooperation and competition characterize their functioning. The study of network dynamics becomes a fruitful field of research. In the third paper of the issue, Brigitte Gay gives us an interesting example of such studies. Borrowing metrics from both sociologists (social capital theories) and statistical physicists (complex network theories), she studies the properties of the biotech sector of the global pharmaceutical industry and the distinct network capital of three firms that dominate the biotech sector for the period studied (2000-2003). Brigitte Gay’s paper puts forward the unstable and dynamic character of networks and studies the strategy that allow some firms at some moment to dominate the networks. She shows that innovation, network dynamics and strategy are entwined and that the positioning of firms within networks is the result of a dynamic management. This issue also includes a category “engineering” where André Slowak studies the elaboration of standards within networks in the automation industry. The author focuses on the knowledge dynamics in standard-setting collaboration. Studying the case of the Profibus community (a leading standard-setting community within industrial automation), he explains the incentives that lead to collaborative standard setting and shows how firms establish standard-setting communities capable of balancing value creation, value distribution and value capture, and hence balancing exploration and exploitation logics (March, 1991).

Coordination is a fundamental part of all innovative firms whatever may be their size, location, or specialization. However, coordination is of particular concern to large companies. Indeed, in spite of the increasing importance of the “start-up” phenomenon with its extension of R&D activities across many actors and territories, today more than ever the large company is the main nerve centre from which innovative goods and services are developed and diffused. This is also true when dealing with innovations developed by (or with) young innovative firms or with other actors (universities, research centres, etc.). Within a historical perspective, particularly, one observes that after the concentration of production means, the definition and the division of production tasks and the constitution of in-house communities of work, the contemporary organization and management of productive resources within large firms typically proceeds through coordinated, decentralized networks, within which big and small firms have particular new characteristics. In the first paper of this issue, Judit Kapás adopts such an historical approach to show and explain the crucial shifts in firm organization. The centralized hierarchy of the factory of the British industrial revolution generated two mutants, the multidividisional form, characterized by a decentralized hierarchy during the second industrial revolution, and the project-based firm in the ICT revolution. These shifts in the organization are explained by the co-evolution of social technology (including the institutional environment, the governance structures and the embedded norms and rules) and physical technology. If new technical systems and new institutional environments generate new types of enterprises, what are the new characteristics of the ICT start-ups born during the last years for the 20th century and the first years of the 21st century?

This is the question raised by Chrystelle Gaujard in the fifth paper of this issue. She uses the Weber’s “idealtype” methodology to build a representation of the organization of work within French ICT start-ups, before and after the burst of the speculative bubble. She shows that start-up idealtypes propose a new organization of work which seems to stimulate innovation and learning and which is based on a mobile structure, where rules are organic, where culture is community, where the leadership style is transformational and where employees are “intrapreneurs” searching for learning and skills.

In the project-based firm, that links the big firm and the young innovative ones within evolving networks, innovations are the result of a dual process of decentralization and appropriation. Controversies pertaining to the organization of networked enterprises, then, focus as much on firm flexibility (in the creation or destruction of production capacities according to the economic context, including the evolution of market context) as on the capacity of the firm to appropriate resources, knowledge and competencies, even without investing in production per se.

The need for coordination within large firms is analyzed in the fourth paper of this issue by Dorra Yahiaoui and Hela Chebbi through the study of the relationship between the headquarters of a multinational company and its subsidiary within the innovation process. It reveals that cooperation, notably between the units of a big multinational company does not happen organically but is the result of a learning process. When an innovation takes shape within a networked multinational corporation, one solution for diffusing it in the different units and markets where the enterprise operates is to transfer it in a top-down manner. However, as explained by Dorra Yahiaoui and Hela Chebbi in the case of a large French telecommunication group, top-down transfers are not an efficient strategy, due to information asymmetry and coordination problems resulting in waste of time and delay. The practice of hybridization, by contrast, is a combination throughout the innovation process of the parent company’s knowledge with those of the subsidiaries. As such, it is a means for technological innovation co-construction and appears to be a new practice for producing transnational innovation, taking into account the characteristics of the subsidiary and of the markets where the new products and services will be implemented.

Firms have thus to facilitate the emergence of creativity at individual or group level. Decentralized organisational structure privileges the implementation of horizontal communication channels involving more intensively middle management levels in strategic decision-making processes. Empirical observations (Nonaka, 1991; Brown and Eisenhardt, 1997) pointed out that these middle managers are usually involved through project teams carrying out new products design. In fact, the implementation of project teams and more interactive groups aims at integrating specialized individuals with complementary skills into looser structures in order to encourage creativity at the individual level. Here the co-ordination structure is based on individual commitment and co-operation among agents within the organization. Individual commitment as opposed to hierarchical control reflects both individual freedom (autonomy) to improvise, and her or his duty to respect organizational priorities and responsibilities.

This perspective is related to the concept of “communities of knowledge”. The knowledge-based economy and the related intense use of new ways of communication and knowledge exchange will contribute to enhance the role of the two specific forms of communities, i.e. epistemic communities and communities of practice, in the processes of organizational learning, of innovation and of knowledge creation within the firm. The main goal of communities of practice is to create, maintain and assimilate knowledge about a shared practice. The communities of practice are oriented toward their members (Lave and Wenger, 1990; Brown and Duguid, 1991). They are considered as supports for enhancing individual competencies through the construction, exchange and sharing of a repertoire of cognitive resources (Wenger, 1998). Interactions between members thus appear as the cornerstone of the model. Epistemic communities, on their part, can be defined as small groups of agents working on a commonly acknowledged subset of knowledge issues and who at the very least accept a commonly understood procedural authority as essential to the success of their knowledge activities (Cowan et al., 1998). Hence, epistemic communities can be defined as groups of agents sharing a common goal of knowledge creation and a common framework allowing the participants to understand this trend (Creplet et al., 2005; Cohendet et al., 2006). The goal of epistemic communities is therefore simultaneously outside and above the community’s members. Innovation implies individuals who belong to multiple communities, and play a role of facilitators between them. This leads to focus on the network connecting communities of practice as the unit of analysis. Communities convert the tacit knowledge of their members into codified knowledge, making it more easily available to support the cross-fertilization between the knowledge domains necessary for innovation. In this way, communities are able to conceptualize these practices by notably belonging to, and participating in, external networks.

This topic is particularly studied in this issue by Thierry Burger Helmchen and Patrick Llerena, who focus on the governance structure of small creative firms. Within networks where several actors and communities are in relation with the firm, and thus where work and knowledge are divided, the issues of coordination and of governance are of major concern. Building on the case of a creative SME in the sector of video games for mobile phones, they show that the governance structure co-evolves with the division of knowledge and of labour. They represent four governance phases observed over a four-year time span and link them with the evolution of the relations between the firms involved and different types of communities (communities of practice, user communities). They show that the governance structure of a creative start-up in a high-tech industry can be influenced by communities peripheral to the firm and conversely that the firm can partly structure these communities.

Finally the last paper of this issue makes a link with the previous issue of the Journal of Innovation Economics dedicated to the economic performances of Russia. Cédric Durand puts forward the comeback of the State in the Russian economy, which is characterized by the implementation since 2004 of new instruments aiming at guiding economic development (such as the extension and the restructuration of public ownership and the mobilization of resources in national projects, the creation of special economic zones, and public investment funds). Cédric Durand proposes the concept of “instumentalized developmentalism” in order to interpret this comeback of the Russian State in which the centralization of resources orchestrated by the State follows the objectives of national independence and the creation of “national champions”, but which induces at the same time the concentration of economic power within a small group of people, an elite closely allied to the “new political leadership”.