FDI or International-Trade-Driven Green Growth of 24 Korean Manufacturing Industries? Evidence from Heterogeneous Panel Based on Non-Causality Test

Round 1

Reviewer 1 Report

The paper entitled as "FDI or International Trade Driven Green Growth of Korean Manufacturing Sector? Evidence from heterogeneous panel based on non-causality test"is quite interesting to the audience of the journal. In the paper, authors investigate FDI, green total factor productivity and international trade within panel econometrics methodologies mainly by focusing on heterogeneity and cross sectional dependence following Dumitrescu-Hurlin type noncausality tests. 

Though paper is interesting, there are serious issues that should be immediately attended. Otherwise, the paper lacks a methodological (not in terms of techniques but textwise mainly) issues that needs attention. 

1) Abstract should be augmented. First three sentences are not required. Also, still it is tried to increase attention of the reader, it does not. Delete and revise. Start with an intest-increasing but short sentence. Than emphasize it is important to evaluate these variables and give reasoning. Then, write a sentence such as The paper aims at evaluation of FDI, Int Trade and green productivity relations within a heterogeneous panel setting. 

Issue 2 for the abstract: It is too fast to talk about DH test. State other methods. No info about dataset. Techciques are panel related but it is said that manufacturing sector in Korea. Which sectors? Which years? Which methods? Which findings? Which policy recommendations? They are either missing or not-satisfactory. 

2) Title could also be changed to include how many sectors are included, what type of sectors? Say, 500 companies in a specific type of sector? 

3) green total factor productivity (GTFP) is not explained well. We have TFP. What is the difference of it? When read in the text, it cannot be understood. 

4) Why would the determination of direction of causality be important? This is not well discussed and emphasized. What policy recommendations are derived? I see that it is stated in the conclusion but it is not understandable. Think of impacts on the society, policy makers, trade policies of Korea. 

5) The referencing style is wrong in the text. Revise to Sustainability's format. 

6) In line 241, it is stated that "First, we start with two stationary variables observed from 24 Korean manufacturing industries from 2011 to 2019." So, T=9 and N=24 and N>T in addition to too low length for time dimension. This is too short for the methods and tests conducted. Time dimension should be increased. Otherwise, the findings are suspiciple. Example: Unit root tests with T=9. Not very appropriate. One could argue that small sample test statistics are available but it is still not appropriate to do analysis with T=9. In unit root tests(CIPS in this setting for example) and in DH causality test which is central in the paper, applying time lags are necessary which also lead to further degrees of freedom loss. I am sorry to say this but the empirical section should be repeated with longer data sets. It is a must. 

7) Heading 3 must be corrected. Also there are problems with subheading to be corrected. Heading 3 could be Methodology and explain methodology here. Don't explain data, your model and method all together. Divide them.

3. Methodology

Explain Measuring GTFP section here because it is not convincing and as I read the paper, I cannot differentiate much from TFP.  Further, there is a literature to measure TFP. Why is the measurement of GTFP, which is by nature very confusing to relate to and differentiate from, so easy? It should be discussed and for readers, it must be very clear what the difference of GTFP is from TFP. 

Before DH test, explain cross-sectional dependence and heterogeneity concepts. Then, explain DH.

Also, I suggest not to keep the paper limited to causality testing. Also estimate models within a panel VAR setting or any method prefered. But concentrate on methods robust to CD and heterogeneity. 

4. Empirical Findings

4.1. Data

Here, delete heading "Descriptives" but use the text. For data first explain the sources, and in detail, the data set. What are the units of data? Which transformations are applied? Logarithmics and then first differences? Explain with examples. Descriptives is a term we use for descriptive statistics AFTER estiamting models. Not for data. This is a misleading heading and might result in misunderstanding. Also, give CD and heterogeneity test results here.  

4.2. Unit Root and Stationarity Tests

4.3. Econometric Findings

and last subheading of 4, I suggest a discussion and policy recommendation section seperate from conclusion.

8) Overall, if I get back to the data section and empirical approach,  "24 Korean manufacturing industries" is not understandable. Are these firms? Manufacturing industry is normally singular. Which firms, are they selected from firms or sectors which engage in green innovations? How is the selection justified? If these are industries selected arbitrarily, this is not very convincing. 

9) What are the implications to sustainability? CO2 emissions are used in the dataset but is it CO2 of the country? In a panel setting, CO2 must be industry - specific emissions.

10) Instead of added value, value-added should be written in text.

11) Table 1 should be extended to include skewness kurtosis JB test and prob. Also, annotations of variables to be added here. 

12) In two places in the text, it says shipments? In the text, value added in total shipments? In table 2, CO2 emissions shipments? 

13) Empirical results will change after increasing time dimension. 

14) Disadvantages should be discussed.

15) Future directions to be added to discussion and policy recommendations section which also should include Comparisons with existent literature.

16) Nonlinear causality literature is omitted. Renewable energy is only in 3 places.  Cointegration is in 1 reference and no discussion for long-run or short-run causality distinctions. Literature section to be extended with this respect. Example papers: 

An Investigation of the Relationship between the Biomass Energy Consumption, Economic Growth and Oil Prices

The Effects of R&D and High Technology Exports on Economic Growth: A Comparative Cointegration Analysis for Turkey and South Korea

Nexus between Industry 4.0 and environmental sustainability: A Fourier panel bootstrap cointegration and causality analysis

The Nonlinear Relationship of Environmental Degradation and Income for the 1870-2011 Period in Selected Developed Countries: The Dynamic Panel-STAR Approach

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper addresses an important issue especially related to green growth and Total factor productivity in green sectors. The emphasis is on Trade versus FDI. The paper harps basically on econometric exercise with sophisticated variety with panel data techniques. The conclusion is not novel but interesting, but obvious like: (i) imports and FDI causes GTFP; (ii) that occurs in capital-intensive competitive manufacturing sector; (iii) import channel works only in fossil energy intensive consumption sectors.  The effect is not that high, meagerly 0.005--0.009. However, imports are not significant. Policy prescriptions seem straight that capital-intensive competitive manufacturing is primary task. This is not new although interesting.

Many theoretical question arises like why imports are not important? Quality of Imports and FDI are overlooked. Also, Why not discussion of environment-friendly technology transfer or R&D discussed? What kind of FDI or import could promote GTFP endogenously? Most importantly, dirty goods import should be prohibited in any country and via supply chain that might affect productive efficiency and that efficiency must be adjusted for quality-environment.  Authors need to disentangle effects of export, import and FDI spillover on TFP in general and GTFP separately. 

As the authors mention about Korea-China FDI-Trade nexus for the recipient China, for Korea the things are different as Korea is much more advanced and have invested a lot in green growth and technology.  Why DH (2012) test is preferred than others? Is there any special reason? The authors should clarify and present alternative results. Equation (5) and (6) are not legible. 

 

What is Dynamic Common Correlated effect mentioned in lines 438--441. That's very vague. Also, what is meant by "serialized and park-based transformational joint industry, thereby promoting the 423 industries’ production efficiency (investKOREA)"? 

I think the author should explain the theoretical causes and reference some literature as Copeland and Taylor, or Nordhaus, Stern report, etc. for background motivation. 

Growth and environment literature is also missing.

 

Thus, subject to major revision it can be accepted. Otherwise, not acceptable in the current format.

Thank you.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors, 

I see that you worked a lot on the final version of the paper. Thank you. My decision is positive for the final version. 

Please attend the minor issue below. 

1) I would like to raise an important note, this is too central for the paper not to be avoided.  Please reattend the interpretation of regression results for FDI in the abstract. Afterwards, in the text.

At line 22: "Furthermore, in the regression analysis, we find that only inward FDI robustly promotes the Korean manufacturing sector's green growth; the positive effect is in the range from 0.005 to 0.009. "

Given that the parameter estimates are too close to 0, writing that FDI has positive effects should be updated with a followup sentence in the following spirit: "though the parameter estimates are poisitve and significant for FDI, they are close to zero, suggesting very limited positive effects which are close to almost zero."

2) Authors noted that they benefited from the references I suggested in the last round. However, they are omitted from this version of the paper.  As I suggest before, a discussion and addition of some of the papers with the econometric contributions should still be added.  

Conluding remarks: The overall improvements in the paper are very satisfactory and I congradulate the authors for their paper and for their contribution to the literature. 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I already reviewed it and gave the report. Thank you

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear authors:

In view of the improvements made, I am of the opinion that the paper is significantly improved and can be published.

Author Response

The authors would like to thank the anonymous reviewer for the useful and constructive comments.