Is user involvement in product development beneficial for firms’ innovation performance? In this paper, we theoretically derive and empirically test conditions under which sourcing external knowledge from users is expected to be most beneficial for firms. We claim that the more knowledgeable users (lead users) tend to further develop promising designs, prototypes or ideas in collaboration with the industry’s innovative firms that possess the absorptive capacity to enable a successful commercialization. Thus, our model suggests that the relationship between innovation performance and user involvement is positively moderated by a firm’s absorptive capacity. To test our hypotheses, we exploit firm-level data from the Swiss Innovation Survey (SIS), the Swiss equivalent of the Community Innovation Survey (CIS) in the European Union. Preliminary results of our estimations confirm that the larger a firm’s absorptive capacity (i.e. more internal R&D) the more it benefits from user involvement.
Many studies find collaborative innovation with customers or user innovation to be crucial for firms’ development of new products and services (see the reviews of Greer & Lei, 2012; and Bogers et al., 2010). So far, beyond theoretical conjectures and exemplars, there is no systematic examination of the effect of user involvement on manufacturers’ innovation performance across industries and sectors. Thus, the economic significance remains limited to firms and select industry cases (see Chatterji & Fabrizio, 2014; Chatterji & Fabrizio, 2012; and Foss et al., 2011).
Our theory is rooted in the discussion on the locus of innovation. The repeated use of a product (i.e. tools, instruments, equipment, materials, etc.) is likely to increase the product specific knowledge of the user through experience and training effects. Consistent with the sticky information hypothesis (von Hippel, 1994), the acquired knowledge can have a sizeable tacit dimension which producers can hardly access, but would potentially reveal the users’ needs that are critical for innovation (Bogers et al., 2010). Consequently, the locus of innovation shifts more towards the user. Through the process of product specific knowledge acquisition the user recognizes whether his or her needs are met by the usability of the product. We argue that many “non-satiated” users are likely to find a higher quality substitute in the market satisfying their unmet needs. However, if “non-satiated” users are knowledgeable and already used high-end products, they are left with no other option than engaging in the process of innovation themselves. The lack of knowledge on commercialization and mass production incentivizes users to collaborate with the high-end producers that possess the absorptive capacity (Cohen and Levinthal, 1990). Consequently, innovative firms that produce goods and services at the high-end of the market are expected to collaborate with lead users (von Hippel, 1986) that developed more valuable designs, prototypes or ideas while less innovative firms show a higher risk of engaging in less successful user collaborations.
Our empirical estimations make use of a merged data set comprising a) comprehensive survey data stemming from the Swiss Innovation Survey (SIS), which covers Swiss firms’ innovation activities and b) a data set based on a detailed internet search for firms that used user designs in their innovation process between 2010 and 2012. The internet search was necessary in order to distinguish between two different types of users, these are, private users (households) and professional users (other enterprises). The SIS survey is conducted by the Swiss Economic Institute (KOF) at the ETH Zuerich (www.kof.ethz.ch) and it is based on a stratified random sample from the Swiss business census on firms with more than five employees, covering all relevant industries (29) in the manufacturing, construction, and service sector. The SIS is the Swiss equivalent of the Community Innovation Survey in the European Union. These surveys served as the empirical background for many innovation related studies and proofed to be of high quality (see, e.g., Laursen and Salter (2006) for CIS, Trantopoulos, von Krogh, Wallin, Woerter (2016) for SIS). For the purpose of this study we use data from two waves of this survey, i.e. the survey in the year 2013 and 2011. Although we mainly rely on specific user questions in the year 2013, we use the information for specific questions from the 2011 survey dealing with the endogeneity of the use of user designs in the innovation process of a firm. Based on the survey results, about 12.5% of Swiss firms introduced new or modified products that were partially or entirely developed by customers.
We estimate the relationship between customer designs and innovation productivity.
Inno_Prod=β0+β1User_D+β2Controls+γInno_Prod-1+u (1)
where User_D represents our variables of main interest and Controls refers to a comprehensive control vector. u is orthogonal to User_D. We proxy unobserved productivity factors that are correlated with User_D with Inno_Prod-1, the innovation productivity of a firm in the previous period. To test our hypothesis we introduce let User_D interact with the internal R&D variable (R&D_Exp-1).
Innovation productivity (Inno_Prod) is measured by the natural logarithm of innovative sales per employee (in full-time equivalent). D is a binary variable and represents customer designs in general or different types of customer designs. R&D_Exp-1 is measured by R&D expenditures per employee in the previous period. Controls contains variables referring to a firms export activities, firm size, technological potential, appropriability, and the competitive situation of a firm approximated by a variable about the intensity of price competition, non-price competition, and number of principal competitors in the main sales market of a firm worldwide. Moreover we control for important knowledge sources for the innovation activities of a firm in order to secure that our measure for customer designs does not measure the general openness of the innovation process of the focal firm.
Preliminary results are reported in Table 1. While user involvement does not show a significant direct effect on innovation performance, the interaction with internal R&D (User_D X R&D_Exp-1) shows a significant positive effect. Moreover, the results show that if firms with low absorptive capacities (i.e. low internal R&D) might even be harmful for innovation performance.
The_Performance_Effects_of_User_Innovation.pdf 