The study of innovation is a relatively young and fast-growing branch of social science. Mainly inspired by the work of Joseph Schumpeter and by other research traditions outside the economics mainstream, it has developed as an interdisciplinary field studying the relationships among economic, technological, organizational, and institutional changes. In his Theory of Economic Development, Schumpeter (1934) pointed out that the main function of entrepreneurs in private firms is to combine existing resources to put forward "new uses and new combinations," or "innovations." These he conceived in a broad sense, so as to encompass new processes and new products as well as new sources of supply of raw materials, new markets, and organizational changes.
A surge of interest in the Schumpeterian theory came at the beginning of the 1980s with the seminal contributions in the evolutionary (Dosi 1982; Nelson and Winter 1982) and neo-Schumpeterian (Freeman et al. 1982) approaches to the study of economic growth, which focused on the central role of innovation for the process of economic development. After such pioneering works, innovation studies have rapidly developed in the last fifteen to twenty years, bringing new insights into the study of economic change and increasingly attracting researchers and policy makers.
As is the case for other young and emerging scientific fields, the rapid development of the discipline has resulted in a great richness and heterogeneity of concepts, theoretical approaches, and empirical results. At present, there still exists no attempt to construct a coherent theoretical framework linking the different findings from this now huge literature. In this respect, there is a need for more systematic efforts to discuss the state of the art of innovation studies and to point out the directions for future research. (1)
This article is intended as a contribution to this purpose. It presents the state of the art of the field of innovation studies, and it discusses the challenges and perspectives for future research. The purpose is not to provide the reader with a complete and detailed overview but rather to focus on recent advances and to point to the challenges ahead.
Our discussion will mainly reflect the evolutionary (or post-Schumpeterian) economic traditions. Other approaches to the study of technology and innovation, such as science and technology studies, business and management literature, and sociological approaches, though potentially relevant, are not considered to the same extent.
The paper has the following structure. First, it separately considers different streams of the literature on innovation by dividing them into different "levels of analysis." The first section will describe the theory of innovation at the microeconomic level. The complexity of the innovative process at the microeconomic level calls for a "systemic" approach to the study of innovation, which will be presented in the second section. Such complexity, moreover, makes the patterns and effects of innovative activity rather heterogeneous across different sectors of the economy. Therefore, the third section will discuss the main challenges in studying "why innovation differs at the meso level." The fourth section will shift to the macroeconomic level, and it will consider the impacts of innovation on the performance of national economies in the long run, in terms of aggregate growth, catching up, and employment trends. Then, the fifth section will point out the main challenges common to the whole field. It will focus on some theoretical and methodological issues that need to be considered by innovation scholars in future research. Finally, the last section will sum up the main conclusions.
Innovation at the Organizational Level
Innovative activities have taken different forms in different historical periods and in different countries (Freeman and Louca 2001; Bruland and Mowery 2005; Lazonick 2005). The British Industrial Revolution in the nineteenth century was based on the on-the-job apprenticeship system, through which craft workers passed on their skills to the next generation. A few decades later, the U.S. innovative revolution was characterized by the rise of the professional manager due to the separation between ownership and control and to the growth of graduate management education. This transition started at the beginning of the twentieth century, and by the end of World War II U.S. industrial corporations had powerful managerial organizations for developing new technologies. During 1970s and 1980s, Japanese companies challenged U.S. industrial corporations in the very industrial sectors in which American corporations had previously gained great competitive advantages. According to William Lazonick (2005), three kinds of institutions constituted the foundation of the Japanese success: cross shareholding, the banking system, and lifelong employment. The importance of lifelong employment in securing Japan's competitive advantage was that, unlike in the USA, there was no separation between salaried and hourly workers, which integrated shop-floor workers into the process of organizational learning.
The Innovative Process and the Innovative Firm
The British, American, and Japanese models of innovative firms are important examples that show how heterogeneous the innovative process tends to be in different historical periods and in different national contexts. Variety over time and space is certainly a fundamental starting point for scholars of innovation. However, the existence of different models of innovation has not discouraged the investigation of the main common factors driving innovation in different historical and geographical situations. In fact, a large number of studies on the innovative process at the organizational level have been carried out since the seminal contributions in the evolutionary economic approach by Giovanni Dosi (1982) and Richard Nelson and Sidney Winter (1977, 1982).
A theory of the innovative firm needs to be based on a few fundamental facts pointed out by evolutionary scholars. (2) First, the technological knowledge of the firm is often tacit and embodied in the minds and in the routines of the individuals within each organization, that is, it is not always codified and easily codifiable. (3) Second, knowledge is not static but dynamic, cumulative, evolving over time. It changes as a result of individuals' creativity, but more commonly new knowledge and innovations are products of intra- and interorganizational interactions. In other words, innovation is a collective endeavor. Third, economic agents operate in an ever-changing and highly uncertain environment, and they have limited information and limited capability to process and interpret such information. In other words, far from being perfect rational maximizers, individuals and firms tend to perform their working and business activities based on a "bounded" kind of rationality (Simon 1959, 1965), which is a limited capability to use information and to make their economic and technological choices. Such characterization of knowledge and technology implies that the heterogeneity in technological capabilities and competencies is important for understanding the process by which new technologies are introduced. Heterogeneity (between firms, technologies, national and historical contexts), therefore, is not an obstacle to the study of innovation but a fundamental precondition for it.
A few decades ago, before evolutionary theories of the innovative firm started to be developed, the common approach to the study of innovation was the so-called "linear model." According to it, the linear sequence originates with an invention that most often comes out of universities or industrial R&D departments. Thereafter, the innovative idea is developed into a working artifact. Finally, the artifact is matched with users' requirements and placed commercially on the market. Several researchers have later criticized the linear model of innovation (more recently, Gittelman and Kogut 2001; Ruttan 2001; Pavitt 2005). They have argued that innovation is a "non-linear cycle of divergent and convergent activities that may repeat over time and at different organizational levels if resources are obtained to renew the cycle" (Van de Ven et al. 1999, 16). The unpredictable nature of innovation is due to the conditions of strong uncertainty in which it occurs, given the unfeasibility of accurately predicting the cost and value of a new product, or the users' demand. Moreover, the innovative process involves the exploration and exploitation of opportunities (March 1991), which can be based either on an advance in technical practice, on a change in market demand, or on a combination of the two (Mowery and Rosenberg 1979).
Recently, the characteristics of the innovative process described above are made more complex by some important trends: (1) the increasing specialization in knowledge production; (2) the increasing complexity in the physical artifacts and in the knowledge base underpinning them; and (3) the continuous matching of technological opportunities with market needs and organizational practices (Pavitt 2005). On the whole, two central features of the innovative process have been stressed in the more recent literature: first, that the innovative process involves the coordination and integration of specialized knowledge and, second, that it requires learning in conditions of uncertainty.
Given the difficulties and complexities associated with the innovative process, great demands are placed on the innovative firm. To manage this inherent uncertainty, the innovative firm must continuously engage in activities of strategizing, financing, and organizing. Furthermore, by definition innovation requires learning about how to transform technologies and how to access markets in order to achieve higher quality and lower production costs. Learning is a...