Environmental Accounting of Financial Development and Foreign Investment: Spatial Analyses of East Asia
Round 1
Reviewer 1 Report
The paper is interesting and covers a key issue.
In my perspective, two main limitations are evident:
- it is not clear to me the novelty of the paper;
- the conclusions could be stronger.
The paper could be revised regarding these two aspects.
Author Response
Reviewer 1
The paper is interesting and covers a key issue.
In my perspective, two main limitations are evident:
Comment:
- it is not clear to me the novelty of the paper;
Response:
First of all, we would like to thank the honorable referee to provide us very useful comments which significantly improved the quality of our paper.
Description on the past literature on Asia has been discussed, literature gap has been highlighted and novelty of this present research has been incorporated by adding a last paragraph in the introduction section (lines 102-121) and in the second section in (lines 294-299).
Comment:
- the conclusions could be stronger.
The paper could be revised regarding these two aspects.
Response:
Conclusion section is improved by expanding the discussions on concluded results and the policy implication in (lines 471-479, 484-488, 490-492, 496-506, 510-512).
Reviewer 2 Report
1.Base on the introductory section lacks explicit descriptions about the mechanism that relates CO2 emissions and financial development, please increase the most recent literature.
2.The definition of financial development, although it is correct, seems to be inadequate, once there have been more alternative ratios defining the level of financial development. The author has to consider those alternative measures to provide robustness tests for the paper results.
3.In order to test the long-run relationships and avoid the spurious regression among
CO2 emissions, GDP and so on, the author need to reports panel unit root tests or other method tests considering in the literature.
Author Response
Reviewer 2
Comment:
1. Base on the introductory section lacks explicit descriptions about the mechanism that relates CO2 emissions and financial development, please increase the most recent literature.
Response:
First of all, we would like to thank the honorable referee to provide us very useful comments which significantly improved the quality of our paper.
With the help of 4 researches published in 2018, a description on the mechanism of relationship between CO2 emissions and financial development has been added in the second last paragraph of Introduction Section in (lines 87-101).
Comment:
2. The definition of financial development, although it is correct, seems to be inadequate, once there have been more alternative ratios defining the level of financial development. The author has to consider those alternative measures to provide robustness tests for the paper results.
Response:
This is a very bright idea. But, the data on the alternative proxies is not available for the whole time period of our all sample countries. Even, we try to incorporate a commonly available proxy of M2/M3 of GDP ratio but we could not even found the data of these variables for our sample countries and time. Due to this limitation, most of past literature has utilized the same proxy as used by us if data on other proxies is not available. We also could not test the alternative definition of financial development due to non-availability of data for sample period and countries.
Comment:
3. In order to test the long-run relationships and avoid the spurious regression among CO2 emissions, GDP and so on, the author need to reports panel unit root tests or other method tests considering in the literature.
Response:
Three cointegration test proposed by Pedroni (2004), Kao (1999) and Maddala and Wu (1999) are employed which have confirmed the evidence of Cointegration in our hypothesized model. The results are reported in newly added table 1 in (line 363) and discussed in section 4 in (lines 356-362) and references are added in (lines 617-622).
Reviewer 3 Report
Summary: The paper investigates the Environmental Kuznets Curve (EKC) hypothesis in the East Asian countries using a period 1991-2014 along with the impact of Financial Market Development (FMD), Foreign Direct Investment (FDI), trade openness and energy intensity on the CO2 emissions per capita. For this purpose, author(s) regress the non-spatial models and test the existence of spatial dependency in the model, they find statistically significant results.
The paper addresses an interesting subject and, in my opinion , it has the potential to make a useful contribution to the literature.
I have only one comment for author(s).
It is natural to expect that the changes of CO2 emissions in response to FMD, FDI and trade openness changes will not be instantaneous. A further check can investigate the impact of FMD, FDI, trade openness and energy intensity averaged over previuos period on CO2 emissions averaged over the following period.
Author Response
Reviewer 3
Summary: The paper investigates the Environmental Kuznets Curve (EKC) hypothesis in the East Asian countries using a period 1991-2014 along with the impact of Financial Market Development (FMD), Foreign Direct Investment (FDI), trade openness and energy intensity on the CO2 emissions per capita. For this purpose, author(s) regress the non-spatial models and test the existence of spatial dependency in the model, they find statistically significant results.
The paper addresses an interesting subject and, in my opinion, it has the potential to make a useful contribution to the literature.
I have only one comment for author(s).
Comment:
It is natural to expect that the changes of CO2 emissions in response to FMD, FDI and trade openness changes will not be instantaneous. A further check can investigate the impact of FMD, FDI, trade openness and energy intensity averaged over previous period on CO2 emissions averaged over the following period.
Response:
First of all, we would like to thank the honorable referee to provide us very useful comment which significantly improved the quality of our paper.
Considering one lag and one lead of each variable in the hypothesized model, Dynamic OLS is employed and results are added in the last column of table 2 in (line 384) and interpretation has been extended in this regard in (lines 368-375). Further in spatial estimates, lag effect of variables are already presented by weighted variables as per spatial established methodology.
Reviewer 4 Report
Title
Environmental accounting of financial development and foreign investment: Spatial analyses of East Asia.
Summary of the paper
The paper uses spatial econometric techniques to analyse the environmental effects of financial market development, foreign direct investments and trade openness on per capita CO2 emissions and the Environmental Kuznets Curve hypothesis in East Asian countries over the period 1991-2014. The empirical results suggest that foreign direct investment inflows and trade openness contribute to environmental degradation both in the locality and in neighbouring countries, while financial market development seems to contribute to local CO2 emissions per capita only through externality effects. Overall, the Environmental Kuznets Curve hypothesis finds confirmation in the East Asia region.
General Comment
The paper contributes to a very interesting topic by analysing the Environmental Kuznets Curve hypothesis and the determinants of CO2 in a sample of East Asian countries. This is of particular interest, given the fast rate at which these economies are growing and their "relative position" in terms of global environmental issues. The paper is well written, overall. It is easy to read, even though some sentences and paragraphs could be re-phrased. The theoretical and literature review sections could be improved, and the empirical setting asks for further developments. Overall, I find the paper an interesting contribution, although I have few suggestions which, I hope, could help the Author(s) to improve the work and make it suitable for publication.
Specific Comments
(1) Please, go for another round of editing. Although the paper is well written, overall, some sentences and paragraphs are not very fluent.
(2) Please, introduce acronyms properly the very first time they are used in the manuscript (e.g. "Environmental Kuznets Curve (EKC)" at line 42, page 1).
(3) I would suggest the Author(s) to reframe the "Literature Review" and "Theoretical Framework" sections. Specifically, I would suggest to incorporate the two sections into a unique one ("Theoretical Framework") which, first, presents theoretically the environmental issue and introduces the reader to the EKC hypothesis, and, second, discusses the potential determinants of environmental degradation and the economic development-environment relationship drawing on previous contributions on the topic. Thus, it would be useful to introduce the 'spatial externality' issue at this second stage.
(4) In addition, I would suggest to state a bit clearer the key novelty of the paper, both in the introduction and at the end of the "Theoretical Framework" section, by underlining the gap in the literature.
(5) I would suggest the Author(s) to add some statistics on the environmental and economic conditions (income, GDP, etc.) to compare the six countries analyse, in order to help the reader understanding the particular context analysed - one or two tables and/or plots and/or maps.
(6) Please, use a standard notation for denoting the logarithm of a variable, i.e. "log(.)" rather than "l".
(7) How the Hong Kong series for FDI has been interpolated? Few words.
(8) Please, include country and time fixed effects in Equation (1), in particular because it is stated later in the text that it can be estimated by either the "simpler" Pooled OLS approach or the Fixed Effects (FE) one.
(9) Please, add a few references to justify the choice of the SDM as the (potentially) best spatial specification. In particular, the SDM allows to model global externalities, while the alternative for modelling local externalities is the Spatial Durbin Error Model. Following several contributions by James P. LeSage, in the absence of a theoretical justification for global vs. local externalities a Bayesian approach should be used to identify the correct type of spatial externalities based on comparison of model probabilities. This would lead to test essentially four key models: (i) a-spatial model; (ii) Spatial Durbin Model for global externalities; (iii) Spatial Durbin Error Model for local externalities and global shocks; (iv) Spatial X Model (SLX in LeSage's setting) for local externalities only. In such a setting, the Spatial Autoregressive (SAR) model and the Spatial Error Model (SEM) are "nested" within the previous cited ones. LeSage also provides Matlab codes (Spatial Econometrics Library) for Bayesian analysis.
(10) If the Author(s) decide to go for a Bayesian approach, then there should be only one "best" model to consider.
Author Response
Reviewer 4
Title
Environmental accounting of financial development and foreign investment: Spatial analyses of East Asia.
Summary of the paper
The paper uses spatial econometric techniques to analyze the environmental effects of financial market development, foreign direct investments and trade openness on per capita CO2 emissions and the Environmental Kuznets Curve hypothesis in East Asian countries over the period 1991-2014. The empirical results suggest that foreign direct investment inflows and trade openness contribute to environmental degradation both in the locality and in neighboring countries, while financial market development seems to contribute to local CO2 emissions per capita only through externality effects. Overall, the Environmental Kuznets Curve hypothesis finds confirmation in the East Asia region.
General Comment
The paper contributes to a very interesting topic by analysing the Environmental Kuznets Curve hypothesis and the determinants of CO2 in a sample of East Asian countries. This is of particular interest, given the fast rate at which these economies are growing and their "relative position" in terms of global environmental issues. The paper is well written, overall. It is easy to read, even though some sentences and paragraphs could be re-phrased. The theoretical and literature review sections could be improved, and the empirical setting asks for further developments. Overall, I find the paper an interesting contribution, although I have few suggestions which, I hope, could help the Author(s) to improve the work and make it suitable for publication.
Specific Comments
Comment:
(1) Please, go for another round of editing. Although the paper is well written, overall, some sentences and paragraphs are not very fluent.
Response:
First of all, we would like to thank the honorable referee to provide us very useful comments which significantly improved the quality of our paper.
Language issues are cared throughout in the revised article.
Comment:
(2) Please, introduce acronyms properly the very first time they are used in the manuscript (e.g. "Environmental Kuznets Curve (EKC)" at line 42, page 1)
Response:
Missing definitions of acronyms have been incorporated in (lines15, 70, 95, 139, 154, 214, 307-8, 312, 315, 334, 386-387).
Comment:
(3) I would suggest the Author(s) to reframe the "Literature Review" and "Theoretical Framework" sections. Specifically, I would suggest to incorporate the two sections into a unique one ("Theoretical Framework") which, first, presents theoretically the environmental issue and introduces the reader to the EKC hypothesis, and, second, discusses the potential determinants of environmental degradation and the economic development-environment relationship drawing on previous contributions on the topic. Thus, it would be useful to introduce the 'spatial externality' issue at this second stage.
Response:
Following this comment, we have converted two sections into one (Theoretical and Empirical Framework) in (line 122-299). In this section, we have followed the suggested order of presenting the literature. First two paragraphs are discussed the theory of EKC in (lines 123-147). 3rd, 4th, 5th, 6th, 7th, 8th, 9thand 10th paragraphs are discussed the all literature on the determinants of pollution emissions in (lines 148-239). Lastly, spatial dimension is discussed in the 11th, 12th , 13th, 14th and 15th paragraphs in (lines 240-293).
Comment:
(4) In addition, I would suggest to state a bit clearer the key novelty of the paper, both in the introduction and at the end of the "Theoretical Framework" section, by underlining the gap in the literature.
Response:
Description on the past literature on Asia has been discussed, literature gap has been highlighted and novelty of this present research has been incorporated by adding a last paragraph in the introduction section (lines 102-121). Further, literature gap has also been highlighted in last paragraph of Section 2 (lines 294-299).
Comment:
(5) I would suggest the Author(s) to add some statistics on the environmental and economic conditions (income, GDP, etc.) to compare the six countries analyse, in order to help the reader understanding the particular context analysed - one or two tables and/or plots and/or maps.
Response:
In section 4, Figure 2 and 3 are incorporated to show the trends of GDP per capita and CO2 emissions per capita in section 4 in (line 351-354). Description on figures have been added in (lines 340-350).
Comment:
(6) Please, use a standard notation for denoting the logarithm of a variable, i.e. "log(.)" rather than "l".
Response:
We used the natural log in the estimations and have changed notation “l” to “ln”. The respective changes have been done in the equations (lines 304, 326), in the table 2 (line 384), in the table 3 (line 394) and in the descriptions in (lines 307, 370, 408, 493)
Comment:
(7) How the Hong Kong series for FDI has been interpolated? Few words.
Response:
Interpolation method has been incorporated in (lines 317-318).
Comment:
(8) Please, include country and time fixed effects in Equation (1), in particular because it is stated later in the text that it can be estimated by either the "simpler" Pooled OLS approach or the Fixed Effects (FE) one.
Response:
Country and time fixed effects are incorporated in equation 1 in (line 304). Further, a little description about it has also been added in section 3 in (lines 312-315).
Comment:
(9) Please, add a few references to justify the choice of the SDM as the (potentially) best spatial specification. In particular, the SDM allows to model global externalities, while the alternative for modelling local externalities is the Spatial Durbin Error Model. Following several contributions by James P. LeSage, in the absence of a theoretical justification for global vs. local externalities a Bayesian approach should be used to identify the correct type of spatial externalities based on comparison of model probabilities. This would lead to test essentially four key models: (i) a-spatial model; (ii) Spatial Durbin Model for global externalities; (iii) Spatial Durbin Error Model for local externalities and global shocks; (iv) Spatial X Model (SLX in LeSage's setting) for local externalities only. In such a setting, the Spatial Autoregressive (SAR) model and the Spatial Error Model (SEM) are "nested" within the previous cited ones. LeSage also provides Matlab codes (Spatial Econometrics Library) for Bayesian analysis.
Response:
References on spatial econometrics have been incorporated in (lines 323, 324, 330, 338). Further, concerned studies are incorporated in the reference list in (lines 613-616).
Applying Bayesian approach is very bright and healthy comment for us because we were unaware of its benefits before. That why, this comment is very constructive for us to enhance our estimation ability. But currently, we are not able to incorporate this comment because of a reason that we don’t have Matlab at this point in time. All spatial estimations and most of others have been done in STATA. Further, we will try to arrange Matlab and will try to have new estimates through Bayesian if honorable referee suggests that it is very necessary for this present article.
Comment:
(10) If the Author(s) decide to go for a Bayesian approach, then there should be only one "best" model to consider.
Response:
Currently, we are not able to estimate Bayesian approach due to reasons mentioned above.
Round 2
Reviewer 1 Report
The revised version of the paper answers the concerns expressed in my previous report.
Reviewer 4 Report
You did a good job!
