University-industry R & D collaboration has been attracting much attention as an effective means to promote innovation. Also in Japan, public support has been provided to such local R & D collaboration through cluster policies. However, no empirical evaluation has been made using micro data regarding MEXT （Ministry of Education, Culture, Sport, Science and Technology）’s knowledge cluster policies, which are comparable to the cluster policies in Europe. Thus, using micro data from public statistics and employing fixed effect panel estimation, this paper empirically examines the direct effects of these cluster policies on R & D expenditures of cluster universities and firms, on the business performance （including sales growth and productivity） of participating firms as well as its spillover effects on all manufacturing plants in cluster regions. Estimation results show that after cluster participation, ① universities and public research institutes significantly increased internal and external research expenditures and research fund from private firms, ② participating firms’ business performance did not significantly improve, and ③ manufacturing plants in cluster areas rather significantly decreased labor productivity, compared to their counterparts that were not located in cluster areas. These results suggest that MEXT’s cluster policies promoted university-industry R & D collaboration, which, however, did not induce overall growth of participating firms’ and local manufacturers’ productivity.

　産業ク

In Japan, business start-up activity has been stagnating since the early 1990s, so the government began providing start-up support in the second half of the 1990s. Start-up support was explicitly introduced in the amended “Basic Law of Small and Medium Enterprises （SMEs）” in 1999 as a new and major goal of the SME policy. Additionally, the new “Basic Law” prescribes an appropriate sharing of SME policy resources between the central and prefecture or municipality governments considering regional and local circumstances. The national policy program for approving the “Start-up Support Business Plans” proposed by municipalities that began in early 2014 based on the “Law for the Strengthening of Industrial Competitiveness” integrated these two directions of the new “Basic Law” into a single policy for the first time. This policy also formed the basis for the “Regional Revitalization” policy that began in September 2014.

This policy has a unique scheme characterized by multilevel collaboration between national and local governments, and local public-private partnerships. The national government encourages municipalities to draw up and submit their own business plans to support start-ups in collaboration with local supporters, including chambers of commerce and industry, and local banks. Upon approval, the national government subsidizes municipality governments, local supporters, and individual founders. By January 2016, 1,000 proposals were approved from among 1,740 municipalities in Japan.

I empirically investigated the effects of this policy on local start-up ratios using municipality panel datasets. The estimation results for panel fixed-effects DID （difference-in-differences） models suggest that this new policy significantly increased the start-up ratios in municipalities with approved proposals compared to those without approved proposals, particularly the ratio of new corporations and subsidiaries of existing firms, as well as in municipalities that appeared unfavorable for start-ups.