Division of Labor between Innovation Intermediaries for SMEs: Productivity Effects of Interfirm Organizations in Japan¹

1. The author would like to thank two anonymous referees for their comments and suggestions. The usual caveats apply. The author would like to thank members of the research project “The Role of Public Research Institutions in the Japan's National Innovation System” at the Research Institute of Economy, Trade and Industry (RIETI), with a special mention to Patarapong Intarakumnerd. This study was funded by the Japan Society for the Promotion of Science (15K03411), the Murata Science Foundation, and the Nomura Foundation.

Abstract

Innovation intermediaries are individuals or organizations that help others improve productivity. This study examines how small‐ and medium‐sized enterprises (SMEs) chose intermediaries and how intermediaries affected total factor productivity (TFP) growth of participants through different channels. Estimated switching regression models reveal that cooperative associations improved TFP of participants through cost sharing, such as joint logistics, while voluntary groups improved TFP of participants through knowledge sharing, such as joint R&D. Innovative SMEs appear to have exploited different intermediaries so that the benefit from each intermediary would be complementary to TFP growth. The results suggest the division of labor between intermediaries.

1. The author would like to thank two anonymous referees for their comments and suggestions. The usual caveats apply. The author would like to thank members of the research project “The Role of Public Research Institutions in the Japan's National Innovation System” at the Research Institute of Economy, Trade and Industry (RIETI), with a special mention to Patarapong Intarakumnerd. This study was funded by the Japan Society for the Promotion of Science (15K03411), the Murata Science Foundation, and the Nomura Foundation.

1. The author would like to thank two anonymous referees for their comments and suggestions. The usual caveats apply. The author would like to thank members of the research project “The Role of Public Research Institutions in the Japan's National Innovation System” at the Research Institute of Economy, Trade and Industry (RIETI), with a special mention to Patarapong Intarakumnerd. This study was funded by the Japan Society for the Promotion of Science (15K03411), the Murata Science Foundation, and the Nomura Foundation.

Notes

1. The author would like to thank two anonymous referees for their comments and suggestions. The usual caveats apply. The author would like to thank members of the research project “The Role of Public Research Institutions in the Japan's National Innovation System” at the Research Institute of Economy, Trade and Industry (RIETI), with a special mention to Patarapong Intarakumnerd. This study was funded by the Japan Society for the Promotion of Science (15K03411), the Murata Science Foundation, and the Nomura Foundation.

1 The root of SME cooperative associations in Japan can be traced back to informal cartels among manufacturers and exporters in industries like ceramics and textiles in the late 19th century. They controlled price, prevented low‐quality goods from being exported, and enforced sanctions against a participant that broke the cartel. They were gradually institutionalized as cooperative associations by the government before and during WWII. In the occupation era (1945–1951), they were dismantled by the enactment of the Anti‐Monopoly Act in 1947 (Aoyama 1999). However, they were restored under the increasingly intense Cold War, reflected in domestic political conflict between the conservative party and the party in opposition. The establishment and promotion of cooperative associations of SMEs was initially put forward by the conservative government, which was seriously concerned about winning the general election of 1947. The conservative government aimed to attract the self‐employed and small business owners who used to be a base of the opposition political party by offering them a financial advantage of policy loans and debt guarantee through government‐affiliated financial institutes. This symbolizes SME policy as “compensation scheme” for a political crisis that has been established since then (Calder 1988). Although the party in opposition won the general election, this SME policy proposed by the conservative party was inherited by the new government, resulting in the enactment of the SME Cooperative Association Law in 1949.

2 The number of exempted cartels based on Article 7 was 652 in 1966 at its peak, and then decreased to zero in 1995.

3 This enactment had been presaged by the establishment of Clearinghouse for Technology and Market Information by the Small and Medium‐sized Enterprise Agency in 1981, which aimed to help small firm managers exchange information about innovations and business opportunities.

4 This process of joint innovation has been known among participants as “discovering each other,” “exploiting each other,” and “exploring together.”

5 Estimation is based on the full‐information maximum likelihood method using the “movestay” command in Stata (Lokshin and Sajaia 2004).

6 Coefficients of joint activities are expected to precisely capture the impact of a specific joint activity under a specific interfirm organization. However, it is possible that the impact of the same joint activity that the firm participated in another interfirm organization can be mixed. This is because the dataset employed in this study was compiled not at the project level but at the firm level. It is not possible from the dataset to rigorously discern the former impact from the latter. Therefore, care should be taken in interpreting the coefficients of joint activities.

7 Hall, Mairesse, and Mohnen (2010) argues that it is preferable to regress the TFP growth on the change of X in order to control for unobserved time‐invariant heterogeneity among firms. As I will describe in this section, the survey data used in this study had collected information on interfirm organizations only in 1992 and the 1995 survey did not collect the information, which made it impossible for this study to adopt this approach.

8 The ninety‐ninth percentile of the number of employees ranges from 719 to 4794. There are 33 firms juristically defined as SMEs that have more than 1,000 employees.

9 Belderbos et al. (2013) shows that positive impacts of subcontracting demonstrated by this dataset (1992–1995) would have disappeared in 2000s because R&D‐intensive core firms had replaced domestic subcontracting networks with global supply chain networks and foreign direct investments, resulting in a significant decrease in geographical spillover to SMEs in major industrial clusters.

10 The knowledge spillover theory of entrepreneurship suggests that localized spillover of information on new opportunities could increase competitive pressures due to excessive entry, resulting in suppressed entrepreneurship in the region (Acs et al. 2009). Although this may have affected the impact of voluntary groups, it is not possible for this study to make a further inference without relevant and reliable data.

Additional information

Funding

Japan Society for the Promotion of Science

Funding

The Murata Science Foundation

Funding

The Nomura Foundation

Notes on contributors

Nobuya Fukugawa

Nobuya Fukugawa is associate professor in the Graduate School of Engineering, Tohoku University.