The Heterogeneous Impacts of R&D on Innovation in Services Sector: A Firm-Level Study of Developing ASEAN
Round 1
Reviewer 1 Report
This paper employed a machine learning-based regression analysis to study the relationship between R&D investment and different types of innovation in service organizations in emerging countries. The analysis is based on data from 1500 service companies in ASEAN. It helps the firms and governments promote innovation in emerging countries where service is rapidly increasing, and deploy more advanced analyses using such machine learning in micro- and macro-econometrics surveys.
In particular, it can be said that the result of demonstrating the difference between the R&D investment of service firms and the impacts on each innovation is clearly indicated, and the result may support more detailed companies' investment in innovation and decision-making on policies.
On the other hand, when it comes to academic papers, it is necessary to reinforce the discussion on how to generalize this research. Specifically, how do you discuss the generalization of not only ASEAN but also similar communities, and so on?
It is also advisable to provide an explanation for how the data in this study statistically explain in terms of the population.
In recent years, there has been a movement to provide services in the manufacturing industry. How is the discussion on service-oriented manufacturing in emerging countries as well as SMEs in the traditional service sectors?
It is also expected that they will explicitly discuss how it can contribute to the existing body of knowledge, practical practices, society and so on, as an academic paper.
In addition, it is difficult to read the tables/figures such as Table 5 and Table A4. Please modify these to make it easier to read.
The following are detailed comments with line numbers.
l.19: Is the use of French intentional?
l.44: Check grammar expression.
l.76: Check grammar expression.
l.125: How is the debate on the servitization of manufacturing being positioned?
l.147: Here, ASEAN is included in the hypothesis, but could it be generalized like a model of a community with characteristics like ASEAN?
l.163: What do the x-axis numbers mean?
l.179: Extra and
l.181: Could you explain the population of this sample?
l.204: What do the numbers mean?
l.213: French?
l.214: Why is it appropriate to choose this model?
l.219: What is (xx)?
l.227: Are the variables here different from the variables in equation (2)? Is the problem of multicollinearity of the inverse Mills ratio taken into account?
l.231: normal distribution?
l.262: Please include a description when using a variable/term that first appears
l.284: Please include a description when using a variable/term that first appears
l.303: Where is this column 3?
l.331: Check grammar expression.
l.361: What does this rho mean? Couldn't you give a more detailed explanation around here?
l.397: Please provide a more detailed explanation of the interpretation of complementarity around here.
l.442: Based on these results, please discuss implications for practical, social, academic, etc.
l.448: Mean2 is RD = 1?
l.454: Could you explain the FDI a little more clearly?
Author Response
Thank you so much for your feedback/comments on our manuscript which we find extremely useful. Your insightful suggestions helped improve the quality of our work substantially. We tried our level best to address your comments. While most of the comments are addressed in the revised manuscript at length, here we answer all the queries in brief. Numbers in brackets indicate references (revised manuscript).
The following are detailed comments and answers with line numbers.
l.19: Is the use of French intentional?
Replaced with English words/phrase.
l.44: Check grammar expression.
Addressed
l.76: Check grammar expression.
Addressed
l.125: How is the debate on the servitization of manufacturing being positioned?
We included this debate in the discussion on whether technological innovation is more relevant to manufacturing sector and non-technological innovation to services sector, resulted in three distinct views- the technicist, the demarcation, and the synthesis [16]. The technicist (or assimilation) approach, in particular, rejects any possibility that a technological innovation can be incubated within this sector [18] Contrarily, the demarcation (service-oriented) approach acknowledges the role of non-technological innovation [18]. The synthesis (integrative) approach encompasses both goods and services as well as technological and non-technological modes of innovation [19]. The importance of this framework is also heightened by the fact that the boundaries between goods and services have become blurred [20], which is discussed in servitization of manufacturing literature at length [21].
l.147: Here, ASEAN is included in the hypothesis, but could it be generalized like a model of a community with characteristics like ASEAN?
While the ASEAN is the geographical focus of this study, in the revised paper we discussed the implications (conclusion section) of our results for similar level of developing countries.
l.163: What do the x-axis numbers mean?
The codes reported in horizontal axis represent services subsectors. The description of services subsectors is provided in section 3.
l.179: Extra and
Deleted
l.181: Could you explain the population of this sample?
As N varies based on the types of models (Heckman, Tobit etc) that are applied, we reported population size, strata and primary sampling units in Table 3.
l.204: What do the numbers mean?
In short, a large number of firms do not invest in R&D.
l.213: French?
Replaced with English words/phrase
l.214: Why is it appropriate to choose this model?
While we followed the literature, in the revised paper, we quoted a paper that observed that models akin to CDM are generally estimated applying simultaneous estimators such as likelihood-based estimators and asymptotic least squares, among others, proven to be effective to control for selectivity and endogeneity problems.
l.219: What is (xx)?
The missing reference is added. Ref #39
l.227: Are the variables here different from the variables in equation (2)? Is the problem of multicollinearity of the inverse Mills ratio taken into account?
Yes. We have estimated a Heckman selection model to see if the sample of ASEAN services firms self-select. We are aware that the inclusion of the Inverse Mills Ratio (IMR) often results in multicollinearity having profound consequences for model estimates. As literature suggests, the best solution to this problem is to incorporate exclusion restrictions. With a valid exclusion restriction, the IMR and the X vector in the substantive equation likley to be less correlated, reducing multicollinearity among predictors as well as the correlation between error terms. As such, we included exclusion restrictions following the literature. It is suggested that a condition number below 20 indicates multicollinearity is not a concern. In our case, the number of 12.77, which is below the critical threshold of 20, suggesting that the collinearity is not a problem in our model. (Ref: Bushway, S., Johnson, B.D. and Slocum, L.A., 2007. Is the magic still there? The use of the Heckman two-step correction for selection bias in criminology. Journal of quantitative criminology, 23(2), pp.151-178.)
Nevertheless, as the value of IMR is insignificant, our subsequent estimates are not based on the Heckman selection. We estimated a Type I Tobit model. Thus, did not explain the IMR and multicollinearity issue in the paper at length.
l.231: normal distribution?
Yes
l.262: Please include a description when using a variable/term that first appears
Provided
l.284: Please include a description when using a variable/term that first appears
A brief explanation on lassologit and cvlassologit and lasso2 is provided.
l.303: Where is this column 3?
Mentioned (Table 3).
l.331: Check grammar expression.
Addressed
l.361: What does this rho mean? Couldn't you give a more detailed explanation around here?
Provided. In short, rho represents the correlation of errors between the four equations.
l.397: Please provide a more detailed explanation of the interpretation of complementarity around here.
Provided. Firms’ decision to innovate one type of innovation is not independent of other types of innovation. This reflects that services firms in ASEAN simultaneously innovate both technological and non-technological innovations.
l.442: Based on these results, please discuss implications for practical, social, academic, etc.
In the conclusion section, we discussed the implications of our findings.
l.448: Mean2 is RD = 1?
Yes, corrected.
l.454: Could you explain the FDI a little more clearly?
Provided
In addition, there is a concern on the visual representation of Tables: it is difficult to read the tables/figures such as Table 5 and Table A4. Please modify these to make it easier to read.
The Tables are clearly visible in MS Word version of our manuscript. However, when the word file is converted to PDF by the MDPI system, some graphs/Tables are not readable/ visible.
Reviewer 2 Report
Innovation was mostly studied on manufacturing and empirical research on services firms’ innovative activities is relatively new. Additionally most studies about manufacturing or services focused on the role of product and process innovation and less attention is being paid to the strategies of organizational and marketing innovation (nontechnological innovation). This is important because it could lead to technological (product- and process) innovation and improve overall performance of organizations.
The present work presents the linkages between knowledge input and output concentrating on a group of developing economies of the Association of Southeast Asian Nations (ASEAN). The objectives of the study were to identify the determinants of investment in R&D exploring the heterogeneities in terms of the relative impact of R&D on product, process, organization and marketing innovation as well as if there exist a complementary (substitute) relation concerning firms’ innovation preferences between four types of innovation.
1500 firms of seven developing economies of ASEAN have been contacted, the LASSO machine learning-based regression has been used to identify key predictors likely to influence firms’ R&D propensity and intensity. The knowledge function has been estimated and found, that medium-sized firm, human capital (training) and credit facilities affect firms’ decision to invest in R&D favorably.
Conclusions and English language should be improved
Author Response
Thank you so much for your comments on our manuscript. As suggested, we have revised the conclusion section, which is appended below. Kindly be informed that the changes are highlighted in bold and the numbers in brackets indicate references of our revised manuscript.
Discussion and conclusions:
Given the limited insights on the association between R&D and innovation in services sectors of developing countries, this study, identifying the determinants of firms’ investment in knowledge, first explores the heterogeneous impacts of R&D on product-, process-, organization-, and marketing innovation. Then it examines if there exists a complementary relation concerning firms’ innovation preferences between four types of innovation. As such, we tested two key hypotheses which are interlinked.
To identify firms’ R&D propensity (likelihood) and R&D intensity (actual investment), we first applied a Machine Learning Application and then, estimated conventional models. The techniques of MLA-based regression and the conventional regression produced nearly an identical result. Both the approaches show that medium-sized firms, human capital (training) and access to credit determine firms’ likelihood as well as decision to invest in innovation. Besides, firms’ location, export status and acquisition of quality certificate have positive influence on R&D intensity. Firms’ location and acquisition of quality certificate also marginally increase firms’ likelihood to invest in knowledge. Whereas, if the main or first product generates the largest share of revenue, then firms’ probability of engaging in innovation as well as decision to invest in knowledge is low, which is probably a unique finding of our study, not common in R&D literature. Previous research identified that innovation in services sector, compared to manufacturing and agriculture, is highly heterogeneous. The innovative activities of services firms do not simply mirror that of their manufacturing counterparts [12]. The descriptive statistics show that of the nine sub-sectors of our sample, scientific research, information services, broadcast, financial activities and air transports have either higher propensity or intensity to invest in R&D. Our estimates suggest that of them, the first three subsectors have statistically significant influence over firms’ R&D intensity.
Then we showed the linkages between R&D intensity and innovation, which is methodologically challenging. The common approach that generally applied to show the association between these two variables is based on univariate probit models (Hypothesis 1), masking an important issue concerning firms’ innovation behavior: simultaneity (Hypothesis 2). Taking this into account, we tested our second hypothesis applying an algorithm-based multivariate probit technique.
The univariate probit model shows ASEAN firms’ investment in knowledge translates into innovation irrespective to product-, process-, organization- and marketing innovation. However, the application of the GHK-simulator-based multivariate probit model results suggest that the relationship between different types of innovation is complementary: firms’ strategy to adopt one type of innovation is influenced by other types, which univariate probit model ignore providing potentially biased results. This led to the estimation of R&D’s impact on technological (joint effects of product and process) and non-technological (joint effects on organization and marketing) innovation. The result suggests that while firms in ASEAN innovate both types of innovation, there is a skewed link between non-technological innovation and services sector. The robustness checks based on an alternative approach also back the outcome.
Overcoming the methodological challenges and applying advanced techniques, we contribute to the scarce literature on R&D-innovation linkages in services sector of developing countries. Our findings support the synthesis (the integrative) framework of service sector innovation, which acknowledges the role of both technological and non-technological innovation in the sector[17]. This study reinforces the views that in services, the technological trajectories are not the only or central form of innovation [19]. However, there is a skewed link between non-technological innovation and services sector. The contribution of non-technological innovation has also been recognized by demarcation (service-oriented) approach of innovation[17]. In other words, in service sector, organization- and marketing innovation (or their joint impacts) play greater role compared to product- and process innovation.
Furthermore, the findings of the study have an important policy implication for developing countries. The experience of developing ASEAN, which has undergone steady structural transformation becoming services dominant suggests that the tertiary sector’s investment in knowledge pays off. As the service sector is increasingly playing a key role in developing economies, their catch-up with advanced economies depend to large extent on innovation in this sector. As such, it is suggested that services firms in developing countries allocate more resources in R&D and other knowledge inputs.
Future research, drawing from our study, can analyze further on single product dominant services firms’ (in terms of sales revenue) lower propensity as well as intensity to invest in innovation, preferably using panel data.
Reviewer 3 Report
The paper fills the gap in current research concerning the role of R&D in innovation management in service sector in ASEAN region.
The authors clearly stated the goals of the paper as the determinants of firms´inverstments into knowledge. As a supplementary goal the examination of companys´preferences towards various types of innovation (as per Oslo manual). The goals definition is acceptable.
As for the methodology the authors took advantage of established statistical analysis including probit model as well as LASSO which suit the purpose well. The authors should better explain why these approaches are superior to others regarding the exploration of the topic. The sample of 1500 companies is quite illustrative ans representative.
Theoretical part addresses important resarches that were conducted with respect to to innovation in service sector.
The conclusions appear to be persuasive enough. Medium-sized firm, human capital and credit facilities affect firms’ decision to invest in R&D favorably. A bit more extensive discussion should be devoted to the validation of the model. The authors can take advantage of several expert interviews to prove the validity of the model.
Proof reading by the native English seems to be inevitable.
The authors should highlight the implication of their conclusions for the practice. Apart from some objections the paper is valuable contribution to creating deeper knowledge about the innovation in service sector.
Author Response
Thank you so much for your comments on our manuscript. As suggested, we have addressed your two broad concerns as follows:
1) the validation on the model;
2) implications of the study/conclusion
Query #1
In addition to following the literature, in the revised paper, we provided further evidence to validate our model. One of the main reasons to select the CDM framework is that previous studies show that the model is well-poised to address issues such as selection bias and endogeneity. In this regard, in the revised draft we referred to a paper, which observed that models akin to CDM are generally estimated applying simultaneous estimators such as likelihood-based estimators and asymptotic least squares, among others, proven to be effective to control for selectivity and endogeneity problems.
Query #2: Implications
As advised, we extended our conclusions highlighting the following theoretical and policy implications. Pls note that the numbers in brackets indicate references of our revised manuscript.
Overcoming the methodological challenges and applying advanced techniques, we contribute to the scarce literature on R&D-innovation linkages in services sector of developing countries. Our findings support the synthesis (the integrative) framework of service sector innovation, which acknowledges the role of both technological and non-technological innovation in the sector[17]. This study reinforces the views that in services, the technological trajectories are not the only or central form of innovation [19]. However, there is a skewed link between non-technological innovation and services sector. The contribution of non-technological innovation has also been recognized by demarcation (service-oriented) approach of innovation[17]. In other words, in service sector, organization- and marketing innovation (or their joint impacts) play greater role compared to product- and process innovation.
Furthermore, the findings of the study have an important policy implication for developing countries. The experience of developing ASEAN, which has undergone steady structural transformation becoming services dominant suggests that the tertiary sector’s investment in knowledge pays off. As the service sector is increasingly playing a key role in developing economies, their catch-up with advanced economies depend to large extent on innovation in this sector. As such, it is suggested that services firms in developing countries allocate more resources in R&D and other knowledge inputs.
