Determinants of Economic Growth across the European Union: A Panel Data Analysis on Small and Medium Enterprises

Round 1

Reviewer 1 Report

Authors have revised the manucsript according to my comments; however, I don't think the revision is substantive.

1. There is still no sufficient theoretical analysis supporting the hypothese. Literature review cannot be considered as theoretical analysis.
2. Table 3 shows that some explanatory variable are highly correlated.
3. Othere factors that affect the economic growth are not controlled in the regression.
4. I don't think that the GMM and 2SLS method can rule out the endogeneity caused by the two-way causality.

Author Response

REPLY TO REVIEWER 1

Dear Colleague,

 

I am extremely grateful for your input and helpful comments regarding the article, which assisted me in improving it. I have addressed them in a point-by-point manner. Please find the answers below. 

Comment 1: There is still no sufficient theoretical analysis supporting the hypotheses. Literature review cannot be considered as theoretical analysis.

Reply 1: Thank you for your comment. The theoretical analysis has been expanded in order to follow your valuable suggestion. Please find the changes in the revised version of the manuscript, in the introduction section.   

Comment 2: Table 3 shows that some explanatory variable are highly correlated.

Reply 2: Thank you for your comment. Not all the highly correlated explanatory variables were included together into the same model. Moreover, multicollinearity was investigated with the variance inflation factor (VIF) for all estimated models in the manuscript. The VIF values were far below the threshold of 10, which would show multicollinearity, according to the mainstream literature (see Neter, Kutner, Nachtsheim and Wasserman, 1996). Hence, the results yielded by the econometric models are not biased by the multicollinearity risk.  

Comment 3: Other factors that affect the economic growth are not controlled in the regression.

Reply 3: Thank you for your comment. In the literature there are studies that signal other factors impacting on economic growth, such as FDI-inflow, FDI-outflow, etc. Since this manuscript deals with the contribution of SMEs to economic growth, no other database was identified (besides the ones mentioned in the manuscript) that could provide data on other factors regarding the impact of SMEs activity on economic growth. Of course, upcoming studies focused on expanding the country sample could consider also expanding the set of variables by introducing such additional control factors.       

Comment 4: I don't think that the GMM and 2SLS method can rule out the endogeneity caused by the two-way causality.

Reply 4: Thank you for your comment. The GMM method, described in the literature by Arellano and Bover (1995), Blundell and Bond (1998), Roodman (2009), is used in econometric models for controlling various sources of endogeneity (e.g., reverse causality, omitted variables bias, simultaneity, measurement error). For instance, recent studies such as Ullah, Akhtar and Zaefarian (2018) support the use of GMM for dynamic panel data, since it can eliminate endogeneity “by internally transforming data”. Konstantakopoulou (2022) uses the GMM estimator of Arellano-Bover/Blundell-Bond to examine the impact of health quality on tourism since it overcomes various issues, including endogeneity and correlation of independent variables. Zhang (2021), Shabani and Shahnazi (2019), Eslamloueyan and Jafari (2021) also note that GMM has been used extensively in order to control endogeneity. Other authors such as Bascle (2008) and Ma and Yao (2022) used either 2SLS or a combination of 2SLS and GMM to control for endogeneity.

Below you can find the papers that were mentioned above:

1. Roodman, D. (2009). How to do xtabond2: An introduction to difference and system GMM in Stata. The Stata Journal, 9(1), 86–136;
2. Ullah, S., Akhtar, P., & Zaefarian, G. (2018). Dealing with endogeneity bias: The generalized method of moments (GMM) for panel data. Industrial Marketing Management, 71, 69-78;
3. Konstantakopoulou, I. (2022). Does health quality affect tourism? Evidence from system GMM estimates. Economic Analysis and Policy, 73, 425-440;
4. Zhang, H.-F. (2021). Iterative GMM for partially linear single-index models with partly endogenous regressors. Computational Statistics & Data Analysis, 156;
5. Shabani, Z.D., & Shahnazi, R. (2019). Energy consumption, carbon dioxide emissions, information and communications technology, and gross domestic product in Iranian economic sectors: A panel causality analysis. Energy, 169, 1064-1078;
6. Eslamloueyan, K., & Jafari, M. (2021). Do high human capital and strong institutions make oil-rich developing countries immune to the oil curse? Energy Policy, 158;
7. Bascle, G. (2008). Controlling for endogeneity with instrumental variables in strategic management research. Strategic Organization, 6(3), 285–327;
8. Ma, Y., & Yao, C. (2022). Openness, financial structure, and bank risk: International evidence. International Review of Financial Analysis, 81.

Hence, I have addressed the need to use both methods in the study. Please find these changes in the revised version of the manuscript.

 

 

Reviewer 2 Report

This paper investigates the relationship between economic growth, imports and exports regarding the activity of small and medium enterprises within the EU-28 using a panel generalized method of moments (GMM) analysis and a panel two-stage least squares (2SLS) analysis. However, I have some major remarks as follows:

1) What does “Table 1 [16: 11]…” mean in line 57? It could be a typo error, kindly check and confirm.

2) The novelty of the article is that it provides analyses for an extensive time frame covering the period before, during and after the 2008 global financial crisis and the onset of the pandemic crisis. What is the significance difference between this paper with the past studies? The authors need to highlight the significant differences in this paper.

3) At the literature review section, the authors should highlight the research gap at the end of this section. Is there any past study focusing on a panel two-stage least squares (2SLS) analysis, and the motivation of using this analysis.

4) In this study, the analysis actually covers one period of study, which is 2005-2020, but not different periods, such as during and after the 2008 global financial crisis and the onset of the pandemic crisis. The statement in line 139-141 is confusing.  

5) Citation for the references [35-37] is not appropriate and addressed properly. What is the relevancy of these references [35-37] in this paper? Need to remove these citations if they are not relevant.

Author Response

REPLY TO REVIEWER 2

Dear Colleague,

 

I am extremely grateful for your input and helpful comments regarding the article, which assisted me in improving it. I have addressed them in a point-by-point manner. Please find the answers below. 

Comment 1: What does “Table 1 [16: 11]…” mean in line 57? It could be a typo error, kindly check and confirm.

Reply 1: Thank you for your comment. The sentence does not include any error or typo. The table comprising the criteria for categorizing SMEs within the European Union was taken from the source numbered 16 in the previous version of the manuscript, from page 11.

Comment 2: The novelty of the article is that it provides analyses for an extensive time frame covering the period before, during and after the 2008 global financial crisis and the onset of the pandemic crisis. What is the significance difference between this paper with the past studies? The authors need to highlight the significant differences in this paper.

Reply 2: Thank you for your comment. This sentence with the novelty has been removed from the manuscript since the study did not tackle in depth differences between certain periods within the time span 2005-2020. The question of the 2008 financial crisis has been addressed only slightly.    

Comment 3: At the literature review section, the authors should highlight the research gap at the end of this section. Is there any past study focusing on a panel two-stage least squares (2SLS) analysis, and the motivation of using this analysis.

Reply 3: Thank you for your comment. Following your valuable advice, I have tackled the research gap aspect and included it at the end of the literature review section. I have also motivated the choice of using both 2SLS and GMM in this study. The GMM method, described in the literature by Arellano and Bover (1995), Blundell and Bond (1998), Roodman (2009), is used in econometric models for controlling various sources of endogeneity (e.g., reverse causality, omitted variables bias, simultaneity, measurement error). For instance, recent studies such as Ullah, Akhtar and Zaefarian (2018) support the use of GMM for dynamic panel data, since it can eliminate endogeneity “by internally transforming data”. Konstantakopoulou (2022) uses the GMM estimator of Arellano-Bover/Blundell-Bond to examine the impact of health quality on tourism since it overcomes various issues, including endogeneity and correlation of independent variables. Zhang (2021), Shabani and Shahnazi (2019), Eslamloueyan and Jafari (2021) also note that GMM has been used extensively in order to control endogeneity. Other authors such as Bascle (2008) and Ma and Yao (2022) used either 2SLS or a combination of 2SLS and GMM to control for endogeneity.

Below you can find the papers that were mentioned above:

1. Roodman, D. (2009). How to do xtabond2: An introduction to difference and system GMM in Stata. The Stata Journal, 9(1), 86–136;
2. Ullah, S., Akhtar, P., & Zaefarian, G. (2018). Dealing with endogeneity bias: The generalized method of moments (GMM) for panel data. Industrial Marketing Management, 71, 69-78;
3. Konstantakopoulou, I. (2022). Does health quality affect tourism? Evidence from system GMM estimates. Economic Analysis and Policy, 73, 425-440;
4. Zhang, H.-F. (2021). Iterative GMM for partially linear single-index models with partly endogenous regressors. Computational Statistics & Data Analysis, 156;
5. Shabani, Z.D., & Shahnazi, R. (2019). Energy consumption, carbon dioxide emissions, information and communications technology, and gross domestic product in Iranian economic sectors: A panel causality analysis. Energy, 169, 1064-1078;
6. Eslamloueyan, K., & Jafari, M. (2021). Do high human capital and strong institutions make oil-rich developing countries immune to the oil curse? Energy Policy, 158;
7. Bascle, G. (2008). Controlling for endogeneity with instrumental variables in strategic management research. Strategic Organization, 6(3), 285–327;
8. Ma, Y., & Yao, C. (2022). Openness, financial structure, and bank risk: International evidence. International Review of Financial Analysis, 81.

Please find these changes in the revised version of the manuscript.  

Comment 4: In this study, the analysis actually covers one period of study, which is 2005-2020, but not different periods, such as during and after the 2008 global financial crisis and the onset of the pandemic crisis. The statement in line 139-141 is confusing.

Reply 4: Thank you very much for your suggestion. The wording was reconsidered and the statement removed in order to avoid potential misunderstandings. Thank you once again. Please find the change in the revised version of the manuscript.  

Comment 5: Citation for the references [35-37] is not appropriate and addressed properly. What is the relevancy of these references [35-37] in this paper? Need to remove these citations if they are not relevant.

Reply 5: Thank you for your comment. I have removed two of the citations from the manuscript. The remaining one is relevant since it deals with the question of taxes levied by public authorities from companies, which are ultimately used to finance public goods.   

Reviewer 3 Report

Accept in present form.

Author Response

REPLY TO REVIEWER 3

Dear Colleague,

 

I am extremely grateful for your comments regarding the article.

Comment 1: Accept in present form.

Reply 1: Thank you very much for your support, I am extremely grateful. Inputs like yours make us researchers continue on this engaging path in order to contribute to the ongoing research conversation. Thank you once again.     

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Congratulation！I think the paper can be accepted for publication in present form.

Reviewer 2 Report

The authors have addressed all the comments and revised the manuscript accordingly. Therefore, I have no further comments. 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

This paper investigates the relationship between GDP growth and intra- and extra-EU imports and exports of SMEs using the panel data of EU countries in 2008–2017. Empirical results indicate that online imports and exports together with extra-EU imports and exports have significant effects on  economic growth.

Comments

1. The effects of importsand exports on economic growth is not a novel research topic. What is the marginal contribution of this paper? Authors should state clearly the contribution in the introduction.
2.  This paper lacks detailed and thorough theoretical analysis of why and how the imports and exports of SMEs affect the economic growth in EU countries. The proposal of hypothese should be supported by adequate theoretical analysis.
3.  There are many factors that can affect the economic growth and these factors should be controlled in the regression model. Control variables are missing in the panel data models of this paper.
4.  All explanatory variables are related to imports and exports. Therefore, there may exists collinearity between these variables, which may cause the bias of estimation results.
5.  Economic growth may also affect the the imports and exports of SMEs. The two-way causality may lead to the endogeneity.

Reviewer 2 Report

The current manuscript is written and presented with details in the research steps and results. Some major points are required to improve or clarify:

1. At line 181 you write that “t represents the period of analysis (i.e., 2008-2017)”. That means t has value equal with 15 and is not the case. More rigorous I think that is the year in which variables are registered/presented and has values that vary from 2008 to 2017.
2. For the descriptive statistics is better to mention significance the results obtained for asymmetry, kurtosis, non-normality of distributions. What are implications in econometric analysis of non-normality of distributions?
3. Why in the table 2, descriptive statistics, you have different numbers of values for the variables in the study (336, 234, 231, 135, 110 and 229)? Why in the Table 4, the presentations of estimated econometric models, there are 83 observations? Please mention the cause of this differences.
4. Why the period taking in analysis stop in the 2017?
5. The Pearson correlation coefficient can be used for variables that follow the normal distributions or for time series that are stationary. But you mention the non-normality of distribution. Is it better to use a nonparametric indicator for correlation?
6. The interpretation of results obtained from the estimated econometric models are theoretically. Please mention the unit of measurement of variables in the interpretations from page 8 (not “one unit”).

Reviewer 3 Report

The article is confusing and difficult for readers to understand scientific results. Even though I have read the article three times, there are still points that I do not understand. Especially in some critical points, the author(s) did not explain enough or were not mentioned in the paper. It is unclear what is in Tab. 1. These criteria have long been known. The authors do not offer an increase in scientific knowledge in this direction.

The article is completely lacking in scientific results.

There are typo and linguistic mistakes.

The abstract should contain Purpose/Objectives, Methods/Analysis, Findings, and Novelty /Improvement.

Please explain your research content in the Introduction more definitely.

Please add a flowchart to show the research methodology visually.

The methods of empirical economic research used by the authors are not explained.

Please compare the results of the present study with previous
studies and analyze the research results completely.

The Conclusion section needs to be described scientifically. Kindly frame it along the following lines:
i. Main findings of the present study (200-300 words)
ii. Comparison with other studies (200-500 words)
iii. Implication and explanation of findings (150-300 words)
iv. Strengths and limitations (150-300 words)
v. Conclusion, recommendation, and future direction (200-500 words).

Overall, the article is not well written and organized. In my idea, the manuscript does not contain novelty, scientific, or practically significant results.

Reviewer 4 Report

This paper investigates the relationship 17 between economic growth, imports and exports regarding the activity of small and medium 18 enterprises within the European Union (EU-28). However, I have some major remarks as follows:

1. The research gap and motivation of study are not highlighted in the paper.

2. Lack of literature review to discuss the past related studies in this paper.

3. Lack of novelty and significance of study in this paper.

4. The citations [30-32] in the conclusion section are not appropriate without further elaboration.

5. The authors should cover the latest data beyond 2017 instead of future research. 

Reviewer 5 Report

To support the sake of the trade at world to investigate the enterprises with a category would be excellent contribution to the literature. I read all the article and saw that the authors' applied to the panel investigation to support the global initiatives to pull out the importance of the global trade.