The Effect of Rainfall on Economic Growth in Thailand: A Blessing for Poor Provinces
Round 1
Reviewer 1 Report
The paper has been reviewed as recommended. I appreciate that the authors followed the guidelines suggested. Please give some last readings in order to fix some minor spelling mistakes.
Author Response
Response to Reviewer 1 Comments
Point 1: The paper has been reviewed as recommended. I appreciate that the authors followed the guidelines suggested. Please give some last readings in order to fix some minor spelling mistakes.
Response 1: Thank you very much for your helpful comments. We already corrected all minor spelling mistakes throughout the manuscript. Please see “track changes” function in Microsoft Word.
Reviewer 2 Report
I suggest that authors should replace "the researchers" with "we", where they use "the researchers" to mean themselves, throughout the manuscript. Reported speech is discouraged in a journal article especially if you mean yourselves.
Author Response
Point 1: I suggest that authors should replace "the researchers" with "we", where they use "the researchers" to mean themselves, throughout the manuscript. Reported speech is discouraged in a journal article especially if you mean yourselves.
Response 1: Thank you very much for your useful comments. We replaced “the researchers” with “we” in case to mean ourselves throughout the manuscript. Please see “track changes” function in Microsoft Word.
Round 2
Reviewer 1 Report
The tasks have been fully accomplished. I appreciate the fine work done by the authors in fulfilling all the requests.
Author Response
Dear Reviewer 1, We are very grateful to your comments that greatly improve our manuscript then we still have been conducting them to develop the further study.
Reviewer 2 Report
Authors did not replace "the researchers" from lines 57, 70, 79, 206, 208, and 391 with "we". Yes we know they are researchers but I advised that reported speech is discouraged in academic writing especially if one is referring to oneself. Please work on this!
Author Response
Dear Reviewer 2, Point 1: Authors did not replace "the researchers" from lines 57, 70, 79, 206, 208, and 391 with "we". Yes we know they are researchers but I advised that reported speech is discouraged in academic writing especially if one is referring to oneself. Please work on this! Response 1: Thank you very much for your kind suggestion. We already corrected "the researchers" with "we" using the “track changes” function in Microsoft Word. Please see line 57, 68, 77, 204, 206, and 389. Thank you very much again that you taking the time to give very helpful comments, we still have been conducting them to improve our future study.
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
The article deals with the effect of rainfall on economic growth in Thailand by economic sectors and subsectors using feasible generalised least square estimator in the regression model.
The article is well written, comprehensive and easy to read.
No special comments.
Reviewer 2 Report
The conceptual framework is missing, the empirics are not well presented. More comments are included.
Comments for author File: 
Comments.pdf
Reviewer 3 Report
Paper title: "The effect of rainfall effect on economic growth."
I have no comment whatsoever on the introduction and literature review. However, discussion on the results in introduction section would need revision if the authors revise the paper's empirical approach based on the comments below.
The overall results seem fine, as shown in Table 2. There was negative significant effect of rainfall on GDP growth in overall 76 provinces, but its effect become positive towards poor provinces.
Major comments:
However I have a reservation regarding the conflicting results according to sectors. My reservation is established based on the argument below:
The paper's objectives, as stated in page 2 (line 64-66):
"The purpose of this study was to investigate how rainfall affects Thai economic growth. The researchers’ first objective involved estimating the
impact of rainfall on GDP growth at the provincial level, via a detailed examination of individual economic sectors as agriculture, industry, and
services."
The above statement about the paper's objective is quite confusing and conflicting each other.
First objective is "estimating the impact of rainfall on GDP growth at the provincial level" -->this is pretty straightforward and understandable. The potential model is to estimate the rainfall effect on GDP based on provincial data of both variables, in addition of other control variables.
Then suddenly the author continues with this sentence "via a detailed examination of individual economic sectors as agriculture, industry, and services." --> Does the author mean that the same model above is applied to agricultural/industrial/service sector? Does this sentence mean the data on sectoral GDP is now regressed against sectoral rainfall?
Then, it has been stated in the empirical model, the data come from each province (denoted i), in data sources there are 76 provinces.
Thus, the data would come from 76 provincial units in Thailand, and it is about each province's precipitation, temperature, population growth. This is fine if the author would wanted to achieve the first objective above.
As for the "detailed examination of individual economic sectors as agriculture, industry, and services," questions that come to my mind:
Out of 76 provinces, how many provinces are classified as fully agricultural/industrial/service province? (in other words, only one type of
economic activity present in the present) Another possible case, does each province contain all three types of economic sectors?
Therefore, my main concern is in the provincial data used in the study:
Let consider the first case (No 1 above) ie there is only one type of economic sector in each province, then the sample size for Table 3 results (for each Panel A, B, C) would not be 76. It will be 76 when we sum all panels (Panel A + Panel B + Panel C sample sizes).
If the second case is true (No 2 above), ie three types of economic sectors are available in any particular province, then the whole results from Table 3 and Table 4 are misleading. How could the data from each province be used in estimating the sectoral rainfall effects on sectoral GDP?
So this is the one obvious flaw that I can detect from the paper's empirical analysis.
My suggestion:
Please provide a clearer explanation on data sources and the segregation used in the study.
Please clearly show the economic sector of each of the 76 provinces. Assuming each province has only 1 sector, then you may proceed with the estimation that have been done to get the results in Table 3 and 4, only now the sample size is no longer 76 (76 is the overall sample size).
If the province has all three types of economic sectors, the Table 3 and 4 results are no longer valid. Sectoral estimations are required, and this must be done on sectoral data, and not on provincial data.
Other minor comments:
Panel unit root test is not needed when cross sectional units (76 provinces) is greater than t (21 years of data). The better model would be fixed effects or random effects. Please do the Hausman test whether it is fixed effects or random effects, or test via BP-LM test if it is simply pooled OLS or random effects. Heteroskedasticity and serial correlation in the error term can be tested and corrected quite easily if the author uses FE/RE model. Do not use "Industrial sector..." as main heading for section 4.2 and Table 3 and 4 since industry is one type of sectors, as mentioned in Table 1, apart from Agriculture and Service.
Reviewer 4 Report
Although the research is introduced with the best purposes (the introduction provide a good literature and the research design looks quite appropriate), starting from the methodology you can observe the diversion of the research through a profoundly wrong path.
Good about the panel application, but I detect a very wrong use of the variables, especially if we consider that the results are counterintuitive, as I absolutely cannot think that the general impact of rainfall is negative. I see a R^2 of 0.23, which is rather a bad result, which means that the estimation needs to be better conducted.
Please, please, make sure to improve the English of this paper, which is absolutely awful. I would suggest to revise and resubmit the work, beacuse as it looks now, it needs a huge need to be re-thought.
