What Shapes Local Innovation Policies? Empirical Evidence from Japanese Cities
Round 1
Reviewer 1 Report
First of all I would like to congratulate authors with such good research filling the gap of knowledge in the field of local innovation policies.
It was really a big plasure to read the article and to know more about Japan case.
However, I have some small comments which can help authors to improve the manuscript:
1) It would be good to get the information about the quantitative approach used in the empirical research in the abstract.
2) Authors gave a background of the relevance of the topic in the introduction. However, it would be great to have more detailed explanation of the novelty of this research (very breif but targeted: what authors and what kind of important aspects were revealed already in previous researches in this field?). Something similar is given in the literature review (section 2), but introduction should be the main overview already.
3) It would be better to have clearly emphasized aim of the manuscript in the introduction.
4) The literature review lacks the mentioning of one of the main theoretical approaches - Triple Helix model, which gives argumentation on the role of government in the regional (or local) innovation system.
5) The literature review ends with the accent on the cities-level analysis. However, it seems too narrow just to mention this without any broader argumentation. And reading this the question of "shrinking cities" emerges too. IS there any threat of this phenomenon on local innovation policy in city mnicipalities?
6) Relying on Triple Helix model, the axis of university (or science and research institutions) are missing in hypotheses (just business and government interaction is involved). Do authors have arguments why the research focus avoided this dimension in the context of R&D? Especially when the sample is based on "target projects of these programs" including "single firm 264 projects, inter-firm collaboration, and university-industry collaboration". More over, it may be "university-government gollaboration" axis. Do authors paid attention on this too?
It seems that theoretical background needs additional reinforcement, including all dimensions of Triple Helix (University-Business-Government, plus Business and Innovation supporting institutions).
7) Authors explained: "We obtained responses on 247 programs (response rate: 85%). From these, we used data relating to 151 programs in 131 cities for the empirical analysis". What was the argumentation to exclude other 96 programs (almost 40 percent) from the sample? Need for the argumentation.
8) Giving argumentation of the sample authors emphasized that they were "targeting 129 cities", but programs selected to the programme were from 131 city. It is still unclear if those cities selected for the examination of the determinants of local R&D subsidy programs among cities match cities, selected for the analysis of city programs. If not, why?
9) References part seems unussual. In my personal perspective it seems that information given in brackets is not needed.
One more, congratulations for authors and I hope that my comments will help you to improve this manuscript and/or give you ideas for the future research conduction.
Author Response
Reply to Reviewer 1
We appreciate your comments and revised our manuscript accordingly as far as possible.
1) It would be good to get the information about the quantitative approach used in the empirical research in the abstract.
* We added in the abstract the following sentence suggesting the quantitative approach: “We employ probit models for basic estimations and provide some robustness checks.”
2) Authors gave a background of the relevance of the topic in the introduction. However, it would be great to have more detailed explanation of the novelty of this research (very brief but targeted: what authors and what kind of important aspects were revealed already in previous researches in this field?). Something similar is given in the literature review (section 2), but introduction should be the main overview already.
* We extended the second paragraph of the introduction part including some previous researches concretely and adding some new sentences ahead of the literature review part.
3) It would be better to have clearly emphasized aim of the manuscript in the introduction.
* We emphasized our research aim more clearly in the introduction. More concretely, we added the following sentences in the second and fifth paragraph of this section: “Hence (Thus), our research aims are to empirically investigate the determinants of both implementation and design of R&D subsidy programs by city governments using unique Japanese data.”
4) The literature review lacks the mentioning of one of the main theoretical approaches - Triple Helix model, which gives argumentation on the role of government in the regional (or local) innovation system.
* Thank you also for this comment. We understand that Triple Helix is also an important issue in national and regional innovation systems. Therefore, we added some
basic literature on Triple Helix by Leydesdorff and Etzkowitz to the reference and inserted a new paragraph after the third one in the literature review section in order to underline the important role of (local) governments, although in the literature review we focus on empirical studies from different perspectives. We enriched the literature review with some more recent papers.
5) The literature review ends with the accent on the cities-level analysis. However, it seems too narrow just to mention this without any broader argumentation. And reading this the question of "shrinking cities" emerges too. Is there any threat of this phenomenon on local innovation policy in city municipalities?
* Because the population in Japan is just beginning to decline, the question of “shrinking cities” with declining population and industries is evident in almost every region and city. In fact, this is an important background of the current “Chiho Sosei” (regional revitalization) policy in Japan, rather than a threat to this policy. Therefore, we added a sentence about the background of this new policy trend in the third paragraph of the introduction part as follows: “An important background of this new policy is the decline of the Japanese population especially in rural areas due to rapid aging and low fertility rate”.
* Moreover, we revised the sentences after the above one as follows (the revised part in italic): “This new policy trend gives local governments a new, more important position in regional revitalization, although policy initiatives may differ significantly according to local needs and conditions. This is the major reason for our research focus on local authorities, especially city governments.” We also added another paragraph that argues the importance to focus on city level R&D support policy.
6) Relying on Triple Helix model, the axis of university (or science and research institutions) are missing in hypotheses (just business and government interaction is involved). Do authors have arguments why the research focus avoided this dimension in the context of R&D? Especially when the sample is based on "target projects of these programs" including "single firm 264 projects, inter-firm collaboration, and university-industry collaboration". Moreover, it may be "university-government collaboration" axis. Do authors paid attention on this too?
It seems that theoretical background needs additional reinforcement, including all dimensions of Triple Helix (University-Business-Government, plus Business and Innovation supporting institutions).
* Please note that we consider also the availability of local universities and public research institutes as the potential R&D partners in Hypothesis 2 (now 2a and 2b), and include the relevant variable into the estimation models. Indeed, we have already explained the reasons for Hypothesis 2 as follows “The national government promotes collaborative R&D with other firms, universities, or public research institutes through the allocation of specific subsidies. Accordingly, most local R&D support programs target collaborative R&D projects, as we show later in the data section. Therefore, the availability of potential R&D partners in the region may be an essential element of the implementation of local R&D subsidy programs.” Thus, we neither ignore nor avoid the issue of university-industry collaboration as an important axis of Triple Helix.
* According to your comments, however, we extended the last sentence above as follows: “Therefore, the availability of potential R&D partners in the region including universities and public research institutes may be an essential element of the implementation of local R&D subsidy programs. Moreover, we inserted the same addendum “(including universities and public research institutes)” into Hypotheses 2a and 2b after the words “with more potential R&D partners”.
* There can be regional or local R&D collaborations between universities and governments (public research institutes), which are, however, not the targets of regional or local innovation policies. Nor do local authorities directly support local universities’ research activities, except within the constant budget from local governments. Public support for prefecture or city universities by regular budget is generally not considered as public subsidies.
7) Authors explained: "We obtained responses on 247 programs (response rate: 85%). From these, we used data relating to 151 programs in 131 cities for the empirical analysis". What was the argumentation to exclude other 96 programs (almost 40 percent) from the sample? Need for the argumentation.
* We have already explained that we excluded 96 prefecture programs from the sample in the last sentence of the second paragraph of Section 4 in page 7 (in the older version). The reasons for excluding prefecture programs from the sample and focusing on city programs were discussed in the paragraph just after it: “as program implementation and design are markedly different between prefectures and cities, and because the variety of local programs appears much larger across cities than prefectures”.
* Finally, we revised this section (Section 5 in the new version) fundamentally to avoid any misunderstanding and confusion. Our original survey targeted both prefecture and city programs, while we used only city data (and no prefecture data) for the empirical analyses in this paper. Because the information about prefecture responses is misleading, we decided to eliminate it from the sample description and mentioned it just in a footnote.
8) Giving argumentation of the sample authors emphasized that they were "targeting 129 cities", but programs selected to the program were from 131 city. It is still unclear if those cities selected for the examination of the determinants of local R&D subsidy programs among cities match cities, selected for the analysis of city programs. If not, why?
* We are sorry again for your confusion and misunderstanding. We could not obtain administrative and regional data of two cities among these 131 cities due to city mergers. Thus, our sample for the analysis of program designs is limited to 129 cities due to data constraints. Thus, the sample of the second estimation about program designs (129 cities) is almost the same as the subsample of 131 cities (that implemented R&D subsidy programs) in the first estimation. Eventually, our sample for the second estimation was reduced to 120 cities due to further data constraints. We revised the sampling description in this section accordingly.
9) References part seems unusual. In my personal perspective it seems that information given in brackets is not needed.
* We have just followed the manuscript template in the submission guideline of this journal. It looks strange and unnecessary even for us, because we have never seen such a reference rule, but it is requested by the journal. Yet, supported by your opinion, we delete the information given in parentheses before the authors’ names from the reference.
Reviewer 2 Report
Dear Author!
The study examines local innovation policies based on the evidences from Japanese cities. This is rather interesting issue because especially in European environment the implementation of local R&D subsidy programs at the municipality level is something rare and unusual.
The authors wanted to solve the research problems thanks to collecting the information about R&D subsidy programs from Japanese local authorities websites and their original surveys. They analysed 151 R&D subsidy programs conducted by 131 cities.
The authors wanted to target the variety of local innovation policies at city level in Japan directly and empirically.
Specific comments:
The paper structure is rather correct. In abstract should be included also some information about paper research goals.
In section: Introduction it should be better explain the whole process of R&D project implementation: from the stage regarding the project preparation, next stages link the project implementation with the local government subsidy, at the and the final stage. when the project is settled. The authors should better explain what the cash flow looks like in this kind of project, what the reporting results to the local government look like? It is also important to explain the objectives of R&D local programs and what results local governments expect. This must be explained because, for example, in Europe the European Union subsidies dominate and such examples are poor unknown. Maybe it should be even the next chapter in the article.
In section: Literature review I suggest also to enrich the bibliography and add some “fresh” papers, monographs or research, if possible.
My another suggestion for authors is to consider to add another part of the text with at least one case study illustrating the whole process of typical R&D project development with the local government financial support. It will be quite good empirical explanation of the research problem.
In section: 3. Theoretical Backgrounds and Hypotheses – I suggest to explain the scientific background for each hypothesis because it is not easy to match the above discussion with the below hypothesis.
In section 4. Sample and Data - my suggestion is to clearly explain the research overall goal as well as specific goals. Authors should consider the better clarification of their approach to this research problem as well as the description of the research stages. Why have they chosen this way to solve the research problem?
In my opinion the authors should add another section: Discussion and in clearly way explain how they evaluate each hypothesis. For example the sentence: “Moreover, they support the first half of our last hypothesis in that supply-side factors are more important than demand-side factors in implementing local R&D policies” or “These findings partially support our first hypothesis that the implementation of local R&D support policies depends on demand-side factors of local economy, but fully support our third hypothesis on the supply-side factors” – are not clear for me. In this section they should summarize the previous discussion regarding the data analysis.
Generally the issue is interesting but needs to be corrected at least according to the above remarks.
Good luck!
Author Response
Reply to Reviewer 2
We appreciate your comments and revised our manuscript accordingly as far as possible.
1) The paper structure is rather correct. In abstract should be included also some information about paper research goals.
* We revised the structure of the abstract (especially the third and the fifth sentence) so that the research goals of this paper may be clarified.
2) In section: Introduction it should be better explain the whole process of R&D project implementation: from the stage regarding the project preparation, next stages link the project implementation with the local government subsidy, and the final stage when the project is settled. The authors should better explain what the cash flow looks like in this kind of project, what the reporting results to the local government look like? It is also important to explain the objectives of R&D local programs and what results local governments expect. This must be explained because, for example, in Europe the European Union subsidies dominate and such examples are poor unknown. Maybe it should be even the next chapter in the article.
* We inserted a paragraph in the introduction (after the third paragraph) which explains how unique recent innovation policy in Japan is: Compared to the multilevel innovation policies in the EU countries, where the EU requests regional governments to develop own support policies but municipality policies may be still negligible, Japanese innovation policy involves and encourages city governments to a higher extent.
* However, to the best of our knowledge, the process of R&D project by recipient firms may not differ much across national, prefecture, and city subsidies (The difference may be much larger across technology fields and industry sectors, between collaboration types and firm size classes etc.). Moreover, regarding this study we do not have enough information about the individual process of R&D projects. Maybe it is more important to explain the process of city government’s R&D subsidy program in relation to subsidized R&D projects. We explained it in a new Section 3 after literature review, based on our own survey results, as suggested in your later comment.
3) In section: Literature review I suggest also to enrich the bibliography and add some “fresh” papers, monographs or research, if possible.
* We tried to enrich the bibliography by adding some more “fresh” papers published in 2019 and 2020 in the literature review section (McDonald 2018 (in press); Belucci et al. 2019, Morrison and Doussineau 2019; Okamuro et al. 2019; Sharif and Xing 2019; Okamuro and Nishimura 2020). We also included the papers on the recent “Smart City” policy (Appio et al. 2019; Caragliu and Del Bo 2019). Moreover, according to a suggestion by another reviewer, we added seminal papers on Triple Helix theory (Leydesdorff and Etzkowitz 1996; Etzkowitz and Leydesdorff 2000).
4) My another suggestion for authors is to consider to add another part of the text with at least one case study illustrating the whole process of typical R&D project development with the local government financial support. It will be quite good empirical explanation of the research problem.
* Because this research project targets local authorities and their program managers, we cannot provide any illustrating cases of private R&D projects that are supported by public subsidies. Instead, we added a new section (Section 3) that describes the whole process of local authority’s R&D support program for recipient firms from the call for applications to the project final report, reimbursement of R&D expenditures and the later follow-up. In this process description, we tried to explain what types of innovation city governments expect to support with small budget and short term.
5) In section: 3. Theoretical Backgrounds and Hypotheses – I suggest to explain the scientific background for each hypothesis because it is not easy to match the above discussion with the below hypothesis.
* In order to make it easier to match the theoretical background with each hypothesis, we first rearranged the order of paragraphs so that each hypothesis matches each background in the revised version. Moreover, we tried to show the scientific background of each hypothesis more clearly with additional citations of previous studies.
6) In section 4. Sample and Data - my suggestion is to clearly explain the research overall goal as well as specific goals. Authors should consider the better clarification of their approach to this research problem as well as the description of the research stages. Why have they chosen this way to solve the research problem?
* Although we have already provided a detailed explanation of estimation methods (approaches) and the reasons for selecting these approaches (models) in Section 5.1 (6.1 in the revised version), we fundamentally revised Section 4 (5 in the revised version) to clearly explain how we collected necessary data and selected the samples for our empirical research according to the research goals. Empirical (econometric) analysis is a powerful method to investigate the determinants and effects, and to test hypotheses.
7) In my opinion the authors should add another section: Discussion and in a clear way explain how they evaluate each hypothesis. For example the sentence: “Moreover, they support the first half of our last hypothesis in that supply-side factors are more important than demand-side factors in implementing local R&D policies” or “These findings partially support our first hypothesis that the implementation of local R&D support policies depends on demand-side factors of local economy, but fully support our third hypothesis on the supply-side factors” – are not clear for me. In this section they should summarize the previous discussion regarding the data analysis.
* We regret that our evaluations of the estimation results with regard to the hypotheses may not be clear enough. We first considered adding a new subsection of discussion before the robustness checks, as you suggested, yet eventually we decided to reformulate the hypotheses and revise the evaluation of each hypothesis in the same subsection. In the revised version, for more clear evaluation of the results, we divide each hypothesis except for the last one into two sub--hypotheses such as 1a and 1b, 2a and 2b, as well as 3a and 3b, where Hypotheses 1a, 2a, and 3a refer to program implementation and Hypotheses 1b, 2b, and 3b refer to program design. We deleted Hypothesis 4 and argue instead whether demand-- or supply--side factors matter more for program implementation or program design. Moreover, we changed the order of hypotheses by starting from supply--side factors considering the logical structure.
Round 2
Reviewer 2 Report
I have no additional comments because the author s improved the paper according to my reccommendations.
