개방형 혁신이 혁신성과에 기여하는가? : 한국 중소기업으로부터의 실증적 증거

Small and Medium-sized Enterprises (SMEs) play an important role in national economy and are also important objects of economic and industrial policies in industrialized economies. On the other hand, innovation is a critical factor for sustainable competitive advantage of firm as well as for long-run economic growth. Therefore, most innovative firms invest much resource in innovation activities including R&D, one of important business activities in firms. However, innovation can be more difficult challenge for SMEs because SMEs, in general, have lacks of innovation resources and complementary assets such as marketing and production capabilities, etc. for commercializing innovative outputs. In addition, environmental changes around innovation characterized by decreasing product life cycles, increasing uncertainty and risk of R&D, etc. are also other barriers for innovation of firms including SMEs. Accordingly, many firms tend to search new knowledge from various external sources and to pursue to achieve innovations through cooperation with broadly distributed external innovators for gaining access to or sharing innovation resources or knowledge of external innovators. This innovation strategy can be incorporated within a new conceptual framework called ‘Open Innovation', which nowadays is recognized as a new innovation paradigm by many academic scholars and business managers and is spreading rapidly to academics and business practices even though there is a controversy surrounding the open innovation. As a result, theoretical discussions and case studies related to the open innovation are fast growing. Nonetheless, there are lacks of empirical literatures dealing with the open innovation activities of firms, particularly SMEs, and thus we have little empirical evidences about how opening for innovation can shape innovation performance of firms (including SMEs). Under this context, this research empirically investigates the relationship between open innovation and the innovation performance of firms by focusing on SMEs in South Korea in basis on a large scale survey, named ‘Survey on Technology of SMEs', which has been conducted by a public organization, Korea Foundation of Small and Medium Enterprises, in South Korea from 1995. Objective of this empirical research is to complement previous literatures on open innovation and to give some implications to business managers and policy-makers interested in development and implementation of innovation strategy or policy. To do so, this research explored the effectiveness of the open innovation on innovation performance of SMEs especially by focusing on ‘Inbound' open innovation (outside-in process) even though there are many important open innovation mechanisms categorized into ‘Outbound' type (inside-out process) and the coupled process of Inbound and Outbound. More concretely speaking, the paper examined the effectiveness on SMEs' innovation performance of opening R&D processes in two perspectives of the openness, investments in open R&D activities and networking for open R&D. Open R&D activities were divided by two different categories such as ‘collaboration R&D' and ‘bought-in R&D', whereas networking for open R&D was focused on formal relationship with different types of external innovators and examined in a viewpoint of network size. Furthermore, this empirical paper also investigated how the internal R&D effort influences innovation performance of SMEs because most firms invest much resource in internal innovation activities simultaneously in spite of pursuing to open innovation strategies. In addition, this study tried to obtain empirical evidences about how interactions between the internal R&D effort and the open R&D efforts influence the innovation performance of SMEs. The data used in this study were those that had been collected for total 3,400 sample SMEs that had conducted R&D during the past two years (from 2005 to 2006), which is a largest survey of innovation activities of SMEs in South Korea. However, because the innovation performance of firms as the dependent variable was measured by sales from new or advanced products (including services) from innovation activities, the sample size used in model estimations was 2,911 by eliminating the data for firms that pursued only to process innovations during the sample period. The reason why the measure of the innovation performance of firms was used is to reflect recent trends of studies on innovation based on CIS (Community Innovation Survey) even though this approach has a weakness not of considering process innovation performance. From analysis of the sample characteristics during the past two years, 54.3% (1,580 firms) of the sample conducted open R&D for innovation, and portion of firms conduced the collaboration R&D was higher than that of the bought-in R&D during the period. Share of firms with employees less than 100 was 74.1%, and average age of sample SMEs was 13.3 years since the year of foundation. Startups that were defined by firms with age no more than 3 years shared 7.1% of total sample firms, and average R&D intensity of sample firms was 12.1%. On the other hand, results of correlation analysis showed there is negative correlation between the in-house R&D investments and the investments in the open R&D including both the collaboration R&D and the bought-in R&D, which implies potential substitution between both. In addition, size of firm showed positive correlations with the internal R&D investments as well as the open R&D investments including two sub-categories, which imply a possibility that larger firms might play more important role especially in the outside-in process of open innovation. In estimating the models which are relating the independent variables to the innovation performance as the dependent variable, several variables were introduced to control potential effects on the innovation performance of other factors different from independent variables. Those control variables include R&D intensity, size of firm, age of firm, a dummy representing startups, 14 dummies of industries. The models were estimated by Tobit regression because the dependent variable had censored distribution inherently and the sample had a selection bias to firms with R&D activities. Empirical findings can be summarized as follows : first, the internal R&D investments affected positively the innovation performance of SMEs. This is consistent with arguments that the in-house R&D efforts play an important role in increasing the innovation performance of firms. Second, SMEs conducted open R&D achieved higher innovation performance than others. In addition, estimation results showed that increasing the investments in open R&D activities was effective for enhancing the innovation performance of SMEs regardless of open R&D categories, i.e. the collaboration R&D and the bought-in R&D. Third, empirical findings also indicated that expanding formal innovation networks for open R&D enhances the innovation performance of SMEs. Forth, although both the in-house R&D effort and the open R&D effort has positive relations to the innovation performance of SMEs respectively, it was showed that both have negative interactions to the enhancement of innovation performance of SMEs. In other words, the in-house R&D investments negatively moderated the effectiveness on the innovation performance of both of open R&D investments and the size of open R&D network. Accordingly, we found the evidence that strengthening both the in-house R&D effort and the open R&D effort (investments and formal networking) simultaneously affects negatively the innovation performance of SMEs. In addition to complementing previous literatures on the open innovation mainly focusing on some advanced countries, these empirical findings show that the open innovation strategy is valid for enhancing the innovation performance of SMEs but business managers or policy-makers need to be careful in implementing or supporting the open innovation activities so as not to give rise to negative interactions with in-house innovation activities of firms.