There is a deep relationship between these three factors (open innovation, complex adaptive systems, and evolutionary change) and they are arranged in a conceptual order based on the name of the model, not in a temporal order. Conceptually, open innovation at a firm goes through a complex adaptive system and then leads to evolutionary change. However, in reality, a specific complex adaptive system can trigger OI through evolutionary properties at any given firm. The OCE conceptual model in Fig. 1 is based on the conceptual order that is needed to analyze the dynamic changes triggered by OI at an individual firm.
The basic agent of open innovation is the firm. The agent chooses action based on its own independent judgment. Its actions influence other agencies or environments, and also be influenced by agencies and environments. In that sense, social organizations, individuals, and governments can be agents of open innovation as well. Firms make various degrees of open innovation (incremental to radical, inbound or outbound) through diverse channels and corporate open innovation influences NIS, RIS, and SIS.
The complex adaptive system exhibits various levels of emergence from price or product differentiation to change in dominant design or even emergence of new sectors. The complex adaptive system influences evolutionary change in corporate open innovation by way of strange trigger. This might include a certain degree of accordance of fascinating customer groups with the technology regime, technological capabilities of related SIS, RIS, or NIS, and the existence and level of rival firms or suppliers. Of course, unique historical heritage, location, ecosystem, or environment of the innovation system are unique features, which work as an strange trigger with certain effects on the diverse OI activities of a firm.
Corporate OI goes through evolutionary stages in the market, blooming into various types and levels of emergence, or being influenced by strange triggers, under complex adaptive systems. The basic locus of evolution is a market. Corporate OI shows up as dominant design thanks to various evolutionary factors (e.g., economies of scale and scope, economics of networks, or open business models). After all these, corporate OI creates a market lock-in, by initiating path dependence and forming a technology regime. The degree of corporate OI creates a variety of evolutionary effects according to the degree of complexity of the complex adaptive systems.
This OCE conceptual model can be analyzed by case studies. Furthermore, the OCE agent based model (ABM) can be built to simulate the situation (Carcia et al. 2007). It is based on the premise of building the model to analyze the courses of open innovation, the complex adaptive system, and evolutionary change, starting from strategy at a corporate level.
We will now assemble the OCE model in three stages. First, we will build up open innovation factors, processes, and their connections with complex adaptive systems. Second, we will build up diverse complex adaptive system factors and their relationships with open innovation and evolutionary change (Yoon, and Lee 2009;). Third, we will construct evolutionary change that results from a complex adaptive system and its interaction with open innovation (Malerba et al. 1999a, b, 2008).
Open innovation in the OCE model
All innovation based on the inflow and outflow of technologies, knowledge, and ideas crossing the boundary of firms, is considered “open innovation”, which is the intended target of the OCE model.
On one hand, it is a fact that the concept of an agent of open innovation has undergone substantial change over time. Schumpeter thought of an entrepreneur, a person, as the agency of innovation in the initial stage of his research, and a large company as the agency in the latter stage (Schumpeter 1934, 1942). After a discussion of the strategy at a corporate or national level was activated by Porter (1980, 1990), approaches to open innovation strategy in institutes, such as firms, social organizations, or government agencies, have also been discussed, analyzed, and carried forward diversely in direct or indirect methods. In the discussion centered on the firm as the basic agent in OCE model analysis, open innovation inside the firm becomes the target of its strategy. On the other hand, open innovation outside a firm, which can result in a market or system failure, is the target of government policy.
The OCE model, as an open innovation channel, takes into consideration the factors on the technology-pushed side, as well as the factors on market-driven side (e.g., collaboration, acquisition and merging, licensing, customers, suppliers, competing firms, university and national research laboratories, etc.). In reality, an apparent disjunction between changes in technology and productivity can be observed, for instance during the so-called “productivity paradox” of the 1980s and 1990s (Tunzelmann and Wang 2007). The reason is that, first of all, in shaping production function, traditional theories or the Dynamic Capabilities Theory failed to take into consideration the source of new knowledge and technology. It is not only the capabilities of producers that should be taken into consideration as bases of productivity or dynamic change but also the capabilities of consumers (Tunzelmann and Wang 2007). One of the important considerations in the OCE model is the various channels through which knowledge and technology flow in. For example, open innovation considers merging and acquiring (M&A) as an open innovation channel to acquire tacit knowledge. The embeddedness of key capabilities and the knowledge that they embody, often motivate firms to acquire an entire entity to obtain these capabilities, as opposed to simply licensing specific goods or hiring employees (Capron et al. 1998). In fact, the higher the level of corporate scale a firm has, the more it seeks complete, open innovation, including tacit knowledge as well as codified knowledge through M&A.
From the corporate viewpoint, the degree of OI varies from incremental, representing the improvement of existing products, to radical innovation, representing the launching of completely new products on global markets as well as into domestic markets.
The difference in the degree of OI causes the appearance of the emergence at various levels of the complex adaptive system (CAS). Regarding the relation between the level of OI and corporate achievement, there is a reverse U-curve in the quantitative analysis of the relation between the OI and corporate achievements of a great number of firms (Laursen and Salter 2006). However, the relation between the degree of corporate OI activity and corporate achievement will vary according to the corporate environment (Drnevich and Kriauciunas 2011). Namely, it is useless to make a quantitative analysis of the relation between the degree of corporate OI and achievement. Rather, it is reasonable to make an analysis of the dynamic process of the corporate OI strategy, that is, the process for achieving the OI strategy of a specific firm, which is what OCE model analysis is intended for.
Quantitative analysis of the determinants of OI showed a variety of factors that determine the success of OI, such as attitude toward openness, entrepreneurship, internal system for openness, and capability for corporate absorption (Yun and Mohan, 2012a). Of course, the RIS, NIS, or SIS under which a firm functions (as factors external to the firm) were presented by way of a quantitative analysis of factors to determine the level of achievement of OI. According to an OCE model approach, related to the analysis of the processes of dynamic change resulting from a corporate OI strategy, corporate achievement is not determined by the specific factors of open innovation. Rather, the factors that determine corporate achievement during the dynamic processes of OI work differently, and in some cases, the same factors have quite different effects. Namely, only through an analysis of the dynamic effects of open innovation is it possible to find the concrete factors that influence the achievement of OI, at a corporate level, and to analyze this influence.
On one hand, the degree of corporate OI determines how well the firm catches up with the leading firm in the belonging sector. Catch-up strategies basically assumes three patterns, namely, path-following catching-up, stage-skipping catching-up, and path-creating catching-up, which have target sectors of other national innovations (Lee and Lim 2001). In a knowledge-based economy, when a technological life cycle is being shortened, technological catch-up types move from path-following catching-up based on a closed innovation strategy to stage-skipping catching-up, pursuing a medium degree of open innovation and a path-creating catching-up pattern with a high degree of open innovation. The higher the degree of corporate OI, the more rapidly and creatively a related firm can catch up. The model to determine a catch-up pattern includes, as internal determinants, factors representing OI, an example of which is an access to an external knowledge base or other available knowledge and resources.
Complex adaptive systems in the OCE Model
Currently, the topic of complexity is attracting a great deal of interest, but there remains a question of what can be said meaningfully about complexity (Simon 1995). There are several complexity theories like Chaos Theory in mathematics that deal with the complexity of nonlinear dynamic systems whose long-term behavior is unpredictable. Systems theory, which possesses many interacting components, deals with another form of complexity. There is also the Computational Complexity Theory that uses agent-based modeling. This theory is applied to physical and economic issues all together. Complex systems arise naturally in the economy because economic agents, whether they are banks, consumers, firms, or investors, continually adjust their moves, purchasing decisions, prices, and forecasts in response to the situations these moves, decisions, prices, or forecasts create in the market (Arthur 1999, 2009). The complexity in this study includes computational complexity as it uses a computer for analysis. The factors used to configure the system respond to and have an influence on the system itself. For this reason, it is called a complex adaptive system (Fleming and Sorenson 2001). Enterprises, which are representative agents and which make up complex systems, are not just collections of production factors, they are “repositories of competence” that create, coordinate, and deploy knowledge. In this case, it is the knowledge of the “specific connections that seem to work in a particular environment” (Potts 2001).
The degree of complexity of a complex adaptive system is shown concretely by the degree of competition between firms or institutions in that system (Choi et al. 2001; Surana et al. 2005). Complexity arises when the dependencies among the elements become important, and complex adaptive systems are composed of interacting, thoughtful agents such as firms, customers, or banks (Miller and Page 2009).
Complex adaptive systems usually operate for a global optimum, and exhibit many levels of aggregation and interaction (Holland and Miller 1991). Innovation systems are themselves complex adaptive systems composed of complex structures of complex populations (Kastelle et al. 2009). There are several kinds of complex adaptive systems in innovation systems, among them are the national innovation system (NIS), the regional innovation system (RIS), and the sectoral innovation system (SIS). The complexity of NIS as a complex adaptive system varies among countries. In the end, the results of corporate OI can determine the creativeness and complexity of the NIS. The differences in the creativeness and complexity among NISs are determined by corporate open innovation, while the degree or level of corporate OI is influenced by the institutions that form the NIS. As a complex adaptive system, the NIS is a reflection of the firms, major affiliated agencies, complementarities, self-organization, and proper emergence of the complex adaptive systems in it (Manzini 2009).
Meanwhile, the OI of firms can determine the creativeness and complexity of the belonging RIS. At the same time, the openness and creativeness of the belonging cluster or RIS, can determine the degree of OI of the firms (Cooke 2005). For example, in Taiwan, the foundation of the Hsinchu Cluster in which knowledge, technology and capital are free to flow through connections to Silicon Valley, is ascertained through the activation of diverse OI activities of the belonging firms (Saxenian and Hsu 2005).
Corporate OI can increase the creativeness and complexity of a specific sector, while the sectoral innovation system determines the OI of the related firms. Although corporate OI has an effect on the improvement of corporate differentiated competitiveness, it increases the complexity and creativeness of the SIS if it is combined with various positive feedback loops, such as economies of scale, network effects, and open innovation business models. If this occurs, the existing dominant design in the related SIS fades away, and fierce competition occurs to establish a new dominant design. Dominant firms of the belonging SIS change rapidly, new markets are set up, or the initial market size and the SIS scope rapidly expand. In the end, the sectoral specificities in the geography of a corporate location are determined by corporate OI, that is, how the firms in the SIS combine their knowledge, technology, and manufacture of products (Bottazzi et al. 2005).
Complex adaptive systems (NIS, RIS, or SIS) lead open innovation of firms in specific directions. The existence of a fascinating customer group in some sectors plays the role of an strange trigger, and this makes innovation systems accept the OI made in the sector more easily. Triggering effects may also occur in the case of innovation systems with R&D capabilities and technology capabilities focused in the related sector. Certain firms may influence universities by funding, while national university systems crucially affect the competitive advantage of firms in the global market (Francisco et al. 2007). The features and properties of an innovation system act as determinants of acceptance, regarding both the degree and direction of OI of individual firms.
Political intervention by governments is required to enhance the level of openness in complex adaptive systems, or to promote the activation of NIS, RIS, and SIS, through knowledge production, distribution, and consumption. To fix a system failure is to connect technology to markets continuously by making sufficient knowledge and technology available in the innovation system. Government intervention in system failures of complex adaptive systems is aimed at promoting the OI of individual firms. Consequently, the core responsibility of a government is to build an open NIS, open RIS, or open SIS, which in turn produce and distribute new knowledge and technology into the innovation system by enhancing the complexity of the complex adaptive system, that is, the openness of the innovation system.
Evolutionary change in the OCE Model
Evolutionary economics inherited from Schumpeter’s legacy involves coevolution of national industries, technology, and institutions, such as universities, research laboratories, and patents (Nelson 1994). This study does not analyze evolutionary results at the level of the economy, but looks at evolutionary effects reached as a result of differentiation strategies by which OI is established at a specific firm. At the current rates of growth in knowledge, rates of its use and formation of positive feedback loops of new types (e.g., SNS), and the selection of OI strategies at a corporate level, produces rapid evolutionary results in markets. This phenomenon, the evolutionary results from OI at corporate levels in markets, is applicable not only to large, market-dominating companies, but also to SMEs. Before firms could carry forward OI strategies, they should check the evolutionary effects of related OI strategies, which are linked to corporate competitiveness and profits. The time frame from open innovation to evolutionary results is being shortened very sharply. How cutting-edge, new technological products face falling into the commodity trap in such short times is the proof of the current, shortened, technology life cycle.
An evolutionary model of technological change is proposed in which a technological breakthrough, or discontinuity, initiates an era of intense technical variation and selection, culminating in a single dominant design product (Anderson and Tushman 1990). Namely, the pinnacle of evolution in innovation is the very formation of a dominant design. A dominant design goes through a variety of incremental technical progressions according to the differentiation strategies of many firms, by way of open innovation. A dominant design is not fixed. It goes through an evolutionary process created by the OI based on the discontinuous technology of a firm, and then forms another dominant design.
Companies with the best products will not always win, as chance events may cause “lock-in” of inferior technologies (Arthur 1983). The process from dominant design formation to its lock-in is the evolutionary result of open innovation based on new knowledge and technology. Various dynamic evolutionary powers trigger the process from dominant design formation to its lock-in. Similar to biological evolution, the evolution of markets related to technological innovation is never locked-in forever. While switching costs may favor the incumbent during rapid technological change, switching costs can become quickly swamped by switching benefits (Teece et al. 1997). Increasing returns, network effects, economies of scope, and open business models are the forces that enable switching benefits to surpass switching costs.
In economics, positive feedback arises from increasing returns (Arthur 1994). In economies, a positive feedback loop is the driving force making a specific technology win a position of a dominant design on the market and then creating a lock-in. The market mechanisms that make up this positive feedback loop are economies of scale, economies of scope, economies of network, and open business models.
Economies of scale are a positive feedback loop on the supply-side that increases supply so long as profit increases in direct proportion to the increase in supply. In cases where increasing returns are caused by economies of scale, a lock-in for current technology occurs. If one among the competing technologies happens to be adopted by historical events, and increasing returns are created through economies of scale, this technology becomes a dominant design and gradually becomes locked-in (Arthur 1989).
Economies of scope are another positive feedback loop on the supply side. It is more efficient for a single supplier to supply a variety of products than for different suppliers to supply products singly in the same product field. This logic also justifies M&A in microeconomics. When various new types of open innovation occur in a traditional manufacturing industry, they undergo an evolutionary process to a dominant design and a lock-in through economies of scope. However, evolutionary phenomena, based on economies of scale or scope, lose their power the moment the positive feedback stops. If alternative technology appears suddenly through the OI of another firm, and is powered by the positive feedback loop, the existing dominant design can disappear suddenly (Anderson and Tushman 1990).
Economies of network form a positive feedback loop on the demand side. Demand increases geometrically as the bandwagon effect occurs in proportion to the increase in demand. For example, as the number of Microsoft (MS) office users grows, more people are likely to use it. In addition, the exchange and distribution of documents, which were produced by MS Office, becomes more convenient. Accordingly, the number of users of MS office increases more. A variety of social network systems (SNSs, such as Facebook, Twitter, and KakaoTalk) also have positive feedback loops based on economies of network. A creative open innovation based on a new idea evolves into a new dominant design if it is powered by a positive feedback loop thanks to the economies of network based on fortuitous initial users. This positive feedback loop on the demand side has a relatively solid and long-term sustainability. Sales of the QWERTY keyboard have been solidly sustained through keyboards that are more efficient, and it has been developed since it evolved into a dominant design based on the economies of network. Such long sustainability is possible if a new idea reaches the status of dominant design because of the economies of network and even a minimal, steady effect from economies of scale is present at some level.
An open business model has features similar to an evolutionary game. It refers to a phenomenon in which there is a positive feedback loop wherein consumers of products turn into producers of products, then, the said producers turn back into consumers, now concentrating more on related products. For example, the Apple App Store is an open business model where consumers of Apps may turn into producers of Apps then go on to consume more Apps (Chesbrough 2006). Namely, it is a positive feedback loop in which economies of network coincide with economies of scale on both the demand and supply sides. Apple’s iTunes, iBook, and Passbook also possess features of open business models. In the case of firms reaching a dominant design through evolution by use of an open business model, it has a very solid evolutionary quality even if this may not cause rapid growth like that based on economies of scale (based on supply). Its positive feedback loop is made relatively stable by economies of network (demand), and considerable self-supply occurs simultaneously as some consumers become producers as well. In the positive feedback loop of an open business model, consumers turn into suppliers and supply diverse products that are not comparable with those produced due to economies of scope.