How R&D Financial Subsidies, Regional R&D Input, and Intellectual Property Protection Affect the Sustainable Patent Output of SMEs: Evidence from China

Round 1

Reviewer 1 Report

I think that your method is solid and appropriate, and the results are clear enough. However, the English editing required to make a publication chance. My comments and suggestions are as follows.

The authors should describe in the Abstract more clearly which is the purpose of the paper, the methodology used, and how the findings of the paper contribute to the existing literature. Furthermore, I suggest that "The research (or empirical) results" can be used instead of "Data" to reveal your conclusions on line 19. I find there are two apparent mistakes in the References list; for examples, you should delete "p" (page),which is not consistent with the others for reference list format, on line 607; and the author's last name should be "Han" only, not "JongsooHan"; please pay attention to formatting the References list. What is the difference between H2a and H2b? Why do you use these two hypotheses in paper? Is what you concern about the type I error and type II error? How to get 31.2% and 67.8% on line 357 and line 358 respectively? I suggest that you should add a path for H5 in Figure 1. Here are a few examples for typos or errors; a. Is the notation "rdsu_it" correct? I believe that it should be "rdfs_it"; b. "Adusted R2" should be corrected as "Adjusted R²" in Table 4 and Table 5; c. "effects" should be corrected as "affects" on line 409; d. "et al" should be corrected as "et al." on line 428.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

General comments:

The paper deals with the effects of R&D financial subsidies on a sustainable patent output in the sample of small and medium enterprises in Chinese regions. The authors study the direct and indirect impact, using as mediating variables: regional R&D personnel input and R&D fund input. Additionally, they verify the moderating role of intellectual property protection in these relationships. The mediation and moderation effects are investigated on the basis of the models specified in equations (1)-(6). Although the study is interesting, there are many flaws in its methodological part. The main drawback is the omission of spatial dependencies in the analyses. There are many papers that confirm the existence of spatial spillovers of R&D across regions. As such, the application of spatial models seems to be obligatory in this type of analyses. It is hard to agree with the statement that “use of the cluster robust standard error to reduce the heteroscedasticity and autocorrelation problems (including spatial spillover effect)” (line 334-335), since applying the cluster robust standard error is not sufficient to solve the problem of the model misspecification (i.e. not including spatial terms). Another problem arises from the time lag between R&D and the patent output. To mitigate this problem, the authors should consider using the dynamic spatial model.

 

 

Specific comments:

It is not clear whether Hypothesis 1 concerns the direct effect or the total effect of RDFS on Sustainable Patent Output of SMEs. Figure 1 indicates that Hypothesis 1 refers to the direct effect. In this case, the justification of Hypothesis 1 is not convincing, since the arguments provided by the authors pertain mainly to the indirect effect through R&D activities of enterprises. The motivation of H1 also suggests that there are some additional mediation variables (e.g. financial conditions of enterprises). It is redundant to include time in the data presented in Table 1. It is clear that the location of the regions does not change, and the data on other indicators suggest that they are time invariant. There are too many abbreviations in the paper. Most variables have several abbreviations, which may be very confusing for the reader. For example, the variable “Market Capacity” in Table 2 has abbreviation “mc”, and in Table 4 and 5 – “Market”, in Table 3 – “Mc”. The specifications of panel data models (1)-(6) are incorrect. The models are over-specified. Additionally, the equations (4)-(6) do not include the variable “ipp” (without interaction) (Hayes, A. Introduction to mediation, moderation, and conditional process analysis. The Guilford Press: New York, NY, USA, 2013). The Hausman test should be carried out for the complete model (line 336-337). According to Baron and Kenny, ?₁ refers to the total effect, not the direct effect (line 348). In this situation, it cannot be used for the acceptance of H1 (see the first comment). There is the lack of direct, mediation and total effect in Table 5.

 

Conclusion

In my opinion, the paper cannot be accepted for publication in the current form due to its methodological flaws and it should be corrected according to the above comments.

Comments for author File: Comments.pdf

Author Response

Please see the attachment。

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments:

The presentation of results is too detailed.

There is still a problem with abbreviations. The variable "Regional Openness" has abbreviation "Ro", and in table 8 probably "Deopen".

 

Author Response

Thank you for your advice. I have shorten the length of empirical results. and the abbreviation of "Regional Openness" was modified to "Ro"