Improvement of Agricultural Supply Quality in China: Evidence from Jiangsu Province

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Accept in present form

Minor editing of English language required

Author Response

Dear reviewer,

Thank you very much for your positive comment!

1. Minor editing of English language required

Response: Based on your suggestion, we’ve polished the writing with the help of a native English speaker (Ph.D.)

Reviewer 2 Report (Previous Reviewer 2)

The authors did a commendable work in revising their manuscript. In particular, the proposed 3 step methodology appears very valuable for the empirical analysis. However it is very convolute as well, therefore it would be important to include an explanation in the abstract.

The 3 step methodology proposed by the authors is a combination of DEA and SFA. The DEA part is vert popular, while the SFA part is borrowed from Fried et al. (2002). However, I do not understand whether the proposed combination is original or has been adopted previously: in the first case, this should be better emphasized as an added value of the manuscript, especially in the abstract.

In addition, I have few minor issues. Please, see below.

Line 99: the acronym SFA is undefined.

Line 235: the acronym for the randomized cutting edge method cannot be SFA.

Table 1 is very important for the explaination of the methodology, but it is too long and difficult to read, therefore, in my opinion, it would be preferable to convert it into plain text.

The formula at line 385 is pretty obscure. The test statistic seems to be Max_E, so I do not understand what that formula means. Usually, to describe a statistical test is sufficient to report the null hypothesis, the test statistic, and its distribution under the null hypothesis.

Formulas at lines 495 and 496: there is a missing tilde, indicating the assumed probability distribution of v_it and u_it, before +N() and -N(). It would be useful for the reader to tell in words that the distribution of u_it is normal truncated at 0 from the right (i.e., u_it are non-positive).

In general, the format of formulas is pretty bad, making them difficult to read. These should be formatted properly.

The manuscript would benefit from a little revision of the English from the authors

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report (New Reviewer)

1. Can we apply this study in different provinces. Write one or two sentences utility of the study on different provinces of China.

2.       Keywords are not as per journal format, put small 1-2 words keyword rather than phrases.

3.       All the figures in the manuscript need improvement.
Figure 1: Put x- and Y-axis title, also remove years form after each year from the X-axis, Put title for both Y-axis.

Figure 2: Increase the font size in the figure. Due to smaller size, it seems distorted.

Figure 3 and 4: Put x- and Y-axis title, remove background colour. Make more appealable to the reader.

 

4.       Table 1 has lots of character table format. Kindly reduce the characters under characteristics column and last column. This information can go in running text under Literature Review section.

5.       Reference style is not following the journal style (Ln no. 224, 315, 316). Also seems mismatching of journal list and text (Ln. 224). Match and verify at all the places.

6.       Data and Methods can be written as “Materials and Methods”.

7.       Other comments are provided in the attached manuscript.

Comments for author File: Comments.pdf

It is okay. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report (New Reviewer)

# I suggest that the authors reconsider their interpretation of negative technical efficiency. It is noted that negative technical efficiency is more related to technological progress in the agricultural sector rather than cost. The authors should revise their discussion in light of this clarification to ensure accurate representation.

# The abstract mentions differences across regions, but it is unclear how these differences affect economic growth. The authors should expand on this point and provide further analysis or discussion to establish a clear link between regional differences and economic growth outcomes.

# In page 3, paragraph 1, the authors mention applying random envelope analysis. However, I suggest that the authors likely meant Stochastic Frontier Analysis. The authors should correct this inconsistency.

# I suggest adding a paragraph at the end of the introduction to outline the structure of the paper. Providing a clear overview of the paper's organization will enhance reader understanding and facilitate navigation through the manuscript.

# I recommend expanding the timeline covered in Figure 1 to include the period before 2010 since the study spans from 2000 to 2021.

# I suggest shortening Table 1. It would be beneficial for the authors to review the table and consider excluding any redundant or non-essential information, ensuring that the table remains concise while conveying the necessary details effectively.

# One of the columns in Table 2 is labeled "data and method," but there is no mention of the data used in the presented studies. The authors should rectify this omission by providing a brief description or reference to the data sources utilized in the previous studies.

# The hypotheses need to be revised to align them with the technical features of the applied method. The authors should ensure consistency between efficiency and optimization concepts to avoid confusion or misinterpretation.

# On page 10, it is recommended to present the input and output variables in a single table with three columns: variable name, description, and variable type. This structure will enhance readability and make it easier for readers to comprehend and reference the variables used in the study.

# On page 12, the authors should provide justification for selecting variable returns to scale over constant returns to scale and inputs-oriented DEA over outputs-oriented DEA. Applying the RTS-test to determine the best orientation would strengthen the methodology and enhance the robustness of the analysis.

# On page 16, the sentence stating "Firstly, the assumption of the variable return to scale and obtained the set of efficiency values under the assumption of the constant return to scale (CRS) were removed" requires clarification. The authors should revise this sentence to ensure clarity and avoid confusion regarding the methodology used.

# In Table 6 on page 18, the authors use different concepts for the AEC, referring to technical efficiency and technology efficiency. These are two distinct concepts, and the authors should reconsider their terminological consistency throughout the paper. It is recommended that the authors carefully review their terminology and ensure consistent usage of terms and concepts throughout the manuscript

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report (New Reviewer)

Good work with revision and editing. Particularly changing figures and tables had added value to the manuscript. 

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

- Advantages and benefits of the proposed approach should be given in detail. Also, the research gaps and the novelty of this study is not clear.

- The characteristics of current research should be highlighted in the comparative table of literature review from both aspects of theoretical and application. In other words, a comprehensive literature review as well as research gaps should be summarized in in the comparative table.

- Generally, real data are tainted by uncertainty. The authors should discuss the proposed approach under data uncertainty.

- In the variables list of current study, some variables such as number of total agricultural employees (LAB), is integer. Does the proposed approach have the ability to be used in the presence of integer values? For more details see the following references:

Kuosmanen, T., & Matin, R. K. (2009). Theory of integer-valued data envelopment analysis. European Journal of Operational Research, 192(2), 658-667.

Lozano, S., & Villa, G. (2006). Data envelopment analysis of integer-valued inputs and outputs. Computers & Operations Research, 33(10), 3004-3014.

- The authors should compare their results and proposed approach with popular approaches in literature.

- The writing should be improved. Also, the paper should be re-edited to fix its errors in both punctuation and grammar.

- Regarding the use of abbreviations, the article has many mistakes. For example "variable scale reward (VRS)" is not correct and "variable returns to scale (VRS)" is correct.

Author Response

Dear reviewer,

We thank you for your detailed comments and valuable suggestions on our paper. We have carefully considered and fully addressed all the comments raised by the reviewers in this revision. One of the authors is a native speaker of the English language, and she is responsible for the language revision and polishing of the whole paper. Our point-to-point response to your comments is presented as follows.

1. Advantages and benefits of the proposed approach should be given in detail. Also, the research gaps and the novelty of this study is not clear.

Response 1: It has been modified. Please see lines 71-88, 267-274.

2. The characteristics of current research should be highlighted in the comparative table of literature review from both aspects of theoretical and application. In other words, a comprehensive literature review as well as research gaps should be summarized in in the comparative table.

Response 2: It has been modified，please see line 203-212, 267-274.

3. Generally, real data are tainted by uncertainty.The authors should discuss the proposed approach under data uncertainty.

In the variables list of current study, some variables such as number of total agricultural employees (LAB), is integer. Does the proposed approach have the ability to be used in the presence of integer values? For more details see the following references:

Kuosmanen, T., & Matin, R. K. (2009). Theory of integer-valued data envelopment analysis. European Journal of Operational Research, 192(2), 658-667.

Lozano, S., & Villa, G. (2006). Data envelopment analysis of integer-valued inputs and outputs. Computers & Operations Research, 33(10), 3004-3014.

Response 3: It has been modified. Please see reference 33-37.

4. The authors should compare their results and proposed approach with popular approaches in literature.

The writing should be improved. Also, the paper should be re-edited to fix its errors in both punctuation and grammar.

Response 4: It has been modified. Please see lines 203-212

5. Regarding the use of abbreviations, the article has many mistakes. For example "variable scale reward (VRS)" is not correct and "variable returns to scale (VRS)" is correct.

Response 5: It has been modified. Please see lines 431-432

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript addresses the improvement of the agricultural supply quality in the Jiangsu province in 2001-2022. At this purpose, agricultural TFP change is assessed by mean of DEA-based Malmquist index numbers. The topic appears interesting for the journal Sustainability, and the manuscript is in overall well written and discussed.

Section 4.2: I would add an introduction before the begin of subsections.

Section 4.2.2 (Data sources): this paragraph is too short. Consider moving it within section 4.4 before lines 350-355.

Sections 4.3.1, 4.3.2, 3.3.3 are too short. Consider avoiding explicit sectioning.

Lines 348-349: provide a reference for the KS and T tests. Which are the other methods mentioned? Cite them or delete.

Descriptive statistics for the research variables are missing: please add.

Table 1: since with 10 variables there are 2^10 possible combinations, it is infeasible to perform all the possible tests. Therefore, I suppose that a sequential criterion was used to perform the selection, e.g., stepwise forward/backward. Plese clarify this.

Author Response

Dear reviewer,

We thank you for your detailed comments and valuable suggestions on our paper. We have carefully considered and fully addressed all the comments raised by the reviewers in this revision. One of the authors is a native speaker of the English language, and she is responsible for the language revision and polishing of the whole paper. Our point-to-point response to your comments is presented as follows.

1. Section 4.2: I would add an introduction before the begin of subsections.

Response1: Section 4.2 has been changed to Section 4.1. Please see lines 298-312.

2. Section 4.2.2 (Data sources): this paragraph is too short. Consider moving it within section 4.4 before lines 350-355.

Response 2: It has been modified according to your suggestion. Please see lines 477-481

3. Sections 4.3.1, 4.3.2, 3.3.3 are too short. Consider avoiding explicit sectioning.

Response 3: It has been modified.please see lines 360-366.

4. Lines 348-349: provide a reference for the KS and T tests. Which are the other methods mentioned? Cite them or delete.

Response 4: It has been deleted. Please see line 430

5. Descriptive statistics for the research variables are missing: please add.

Response 5: It has been modified. Please see lines 485-489 and table 2-3.

6.Table 1: since with 10 variables there are 2^10 possible combinations, it is infeasible to perform all the possible tests. Therefore, I suppose that a sequential criterion was used to perform the selection, e.g., stepwise forward/backward. Please clarify this.

Response 6: It has been modified. Please see lines 410-469

Author Response File: Author Response.pdf

Reviewer 3 Report

Thank you for the chance to review this paper on potential agricultural production in Jiangsu. The authors use combinations of data envelopment analysis to establish a theoretical optimium. I used these techniques frequently in my career - and my doctorate was on civil servant productivity. I live and work in China. So I feel qualified to review this paper.

Decision: Major Revision. Before I think this paper is publishable, at a minimum you need to:

1. make your model simplier. Figure 1 is so abstract. There are a bunch of generalities which are not explained by the literature review or described in concrete terms (like did farmers use a new fertilizer?). 

2. make your simplified model more concrete.

3, Actually describe the literature you cite. To take one example, " We should adjust and optimize the planting structure of grain, cash crops and forage [18]."  Did Lele and Goswami really say this? In what context? Were they looking at rice? They talk about markets. You don't mention markets at all in your paper.

I could do this for every citation/reference - but my review would be as long as your paper.

4. Be self-critical. Most economists think that productivity and output responds to PRICES (and profit). Prices/profits don't appear anywhere in your design or explanation. Why not? What is wrong with linear regression? Where is trade? Why are not agricultural policies of the CCP cited? Why are not actual producers cited?

5. Novelty/importance. Why should we care about productivity in Jiangsu? Your methods are not new. Your data isn't new. Because we don't get a proper literature review, we can not assess how your results/methods INNOVATE on the literature you cite (or the real literature on productivity). 

The structure has problems. For example, 5.1.1 would be good to know right at the beginning. What are the productivity problems in Jiangsu? All these tiny sections are maddening, and seem unconnected. Again, by simplifying your "story", you would make the paper easier to follow.

Language and typos. There are still many typos. Like "The supply side of produce faces the structural imbalance of effective supply shortage and high inventory." Should be production (on line 113). Notice how abstract this is. What do you mean by this?

This leads to a broader point. The paper is hard to read. I don't mean technical. I mean, the writing prevents one from welcoming the content.

To fix this, try to break up large paragraphs of abstract words with concrete examples. This will make your writing more approachable and give the reader a chance to understand what you are saying.

For example, "At present, China's grain production, inventory and import are increasing at the same time", and agricultural products are" large in amount but poor in quality", with the majority of low-end products in quantity and few high-quality varieties [9]. The over-application of chemical fertilizers and pesticides is not suitable for the development of green and modern agriculture, which does not meet the needs of consumers for superior, green and safe agricultural products [10]. Since
2020, the COVID-19 pandemic has aggravated the imbalance between agricultural
products’ imports and exports. The data from the Customs department showed  that
China’s imports of agricultural products from the USA in 2020 reached 16274 ten million" (lines 114-122).

Give a description of this poor quality. Or the over-production in a particular location in Jiangsu. What does it actually LOOK like? I dont recommend a poem or descriptive novel. But breaking up these very broad, abstract observations with time for the reader to 'see' the problem gives us time to reflect. It also establishes your credibility as an expert.

The write-up in section 4.2 is generic. Just saying they are vectors doesnt help. Please combine sec. 4 with the data sources and variable descriptions.

Also, we have no idea why you use CRR, then BCC, then Malmquist. You just present the textbook definition of each. We dont know why they are presented separately. So how they fit into your design separately?

Its not clear from the literature review or sec 4:

a) where does efficiency come from?

b) Where does progress come from?

c) How are the two different?

Table 1 is not explained. The variables themselves are not explained. I have no idea where these came from.

You just suddenly say you use a KS test. It looks at how normal your distributions are (how close to Gaussian). You seem to be taking a difference (which isnt described mathematically, even though we get textbook maths about CRR and BCC). We then get a t-test whether the change over the period is statistically significantly different from zero (not explained in the paper - but obviously what you are doing).

And anyway, how to use a t-test to non-norrmal data? Why didnt you use non-parametric methods? Im not asking you to use them. I am asking you to be SELF-CRITICAL, and recognize that non-normal data need non-parametric methods.

The addition of random variables doesnt make sense. If ACUL isnt significant but ACUL+IRR is, then obviously only the addition of 1,000 hectres (IRR) has significantly changed over the period. Anyway, you never explain why you are doing this adding.

What I dislike the most about the paper is that there is not theory of causality. We see that agricultural production statistically significantly increased. Somehow grain production (AF) didn't. What is the proposed CAUSE of this productivity. I know DEA looks at combos of factors. But if we don't have some theory. Some explanation -- then we are just data mining.

Figure 1 is a theory of everything, and nothing. It is so complex. And arrows are not an explanation of causality. What is "product structure upgrading" in Figure 1. Or "system and mechanism innovation"?

I must stop here. I can not even interpret section 5 (your main results). Because I do not know what innovations or improvements were actually made. You are measuring something without describing what it is you are measuring. Did they use robo-farming? New crop variants? Make it SIMPLE.

Finally, tell us what your results mean. You say "the rate of ATFP in Jiangsu Province was 28.4%". Does that mean we get 28% more rice? Or use 28% less petrol? Raw numbers have no real world meaning.

I hope this helps. I didnt reject the paper outright because I guess there is a grain of useful material in there. But it has been hidden by abstraction and making the paper overly complicated. And please compare/contrast your methods/findings to the literature. Do not just use other papers as decoration.

You promise in the abstract to "put forward counter-measures and suggestions to promote agricultural structural reform. Thus, rely on agricultural technological innovation, promote the development of modern agricultural industry, implement
clean production, and promote the coordinated development of the three major rural industries." Yet, this most important part of paper actually does not appear. You promise this discussion but I couldnt find it -- more than a passing sentence or two in the conclusion.

Good luck with your revisions!

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The manuscript is improved, however some issues should be addressed:

- Generally, real data are tainted by uncertainty. The authors should discuss the proposed approach under data uncertainty.

- The authors should compare their results and proposed approach with popular approaches in literature.

- Quality of Figure 1 should be improved.

Author Response

Comments and Suggestions for Authors

The manuscript is improved, however some issues should be addressed:

Response: We thank the reviewer for the commendation. Your kind and valuable comments helped us improve it. The additional issues have been addressed accordingly. Kindly let us know if there is anything to be done.

- Generally, real data are tainted by uncertainty. The authors should discuss the proposed approach under data uncertainty.

Response: We thank the reviewer for this insight. Following this comment, we have discussed in detail the uncertainties regarding the data. Please see for your reference.

- The authors should compare their results and proposed approach with popular approaches in literature. 

Response: Thank you for this suggestion. In the revised manuscript, we have compared the results obtained with the most recent literature. Please see for your reference.

- Quality of Figure 1 should be improved.

Response: We thank the reviewer for this comment. Following your suggestion, we have improved the quality of Figure 1. Please see the revised manuscript for your reference.

Author Response File: Author Response.pdf

Reviewer 3 Report

Thank you for the chance to review the changes made to the paper. We have a slightly better presentation yet it failed to address several critical and relevant questions below:

1. what happened to agro productivity in Jiangsu over the period (and its sub-areas),

2. what happened (changed?)

3. WHY?

4. what does it mean?

Think of your paper like a story. A plot. a thesis. One simple idea. I can not find it because there is still all the stuff going on which I described last time.

We are the 2020s. No need to give variable names 5 letters only. For example, ACUL. Just use the whole phrase. So I dont need to keep looking up whaet ACUL is.

I know DEA inside and out. But your paper makes it hard. I wish you had given more thought to my first rounds of comments. I still have no idea what drives productivity in Jiangsu. Or how to influences it.

The KS and t-stat thing is inappropriate. Actually, the methods in general are not used in figuring out the productivity of a level or unit. DEA is used for comparisons.

The references are not used well because you do not describe what the original authors say. There are no page citations. No evidence or critiques from the original sources at all. We have 55 references but only a few talk about agro productivity in China or Jiangsu.

As an economist, I just do not know what your paper is trying to say. Nor can I say you hit on some novelty. You don't use new methods. Or new ways of looking at the subject. "Originality is the key for any academic paper"

So no novelty is easily reason enough to reject the paper.

I suggest you turn this paper into a report to a government or a magazine article without all the jargon.

Good luck!

Author Response

Comments and Suggestions for Authors

Thank you for the chance to review the changes made to the paper. We have a slightly better presentation yet it failed to address several critical and relevant questions below:

1. what happened to agro productivity in Jiangsu over the period (and its sub-areas),
2. what happened (changed?)
3. WHY?
4. what does it mean?

Think of your paper like a story. A plot. a thesis. One simple idea. I can not find it because there is still all the stuff going on which I described last time.

We are the 2020s. No need to give variable names 5 letters only. For example, ACUL. Just use the whole phrase. So I dont need to keep looking up whaet ACUL is.

I know DEA inside and out. But your paper makes it hard. I wish you had given more thought to my first rounds of comments. I still have no idea what drives productivity in Jiangsu. Or how to influences it.

The KS and t-stat thing is inappropriate. Actually, the methods in general are not used in figuring out the productivity of a level or unit. DEA is used for comparisons.

The references are not used well because you do not describe what the original authors say. There are no page citations. No evidence or critiques from the original sources at all. We have 55 references but only a few talk about agro productivity in China or Jiangsu.

As an economist, I just do not know what your paper is trying to say. Nor can I say you hit on some novelty. You don't use new methods. Or new ways of looking at the subject. "Originality is the key for any academic paper"

So no novelty is easily reason enough to reject the paper.

I suggest you turn this paper into a report to a government or a magazine article without all the jargon.

Good luck!

 

Response:

We appreciate the reviewer for your patience so far. It is quite unfortunate to read some of these comments from you. However, we apologize for the negligence on our part. We have come to believe that research is also an avenue for learning, and we appreciate your suggestions in both rounds. We initially worked on the comments you provided, but not up to your expectation. It may be a misunderstanding on our part and we humbly accept it. Following this round of comments, the team has sat down and discussed your comments again and has revised the manuscript according to your suggestions. We have also invited other researchers in the field to help improve the manuscript to your satisfaction. Indeed, your patience during this review process is commendable. It is our objective to make the manuscript the best it could be. Please check the revised manuscript for your reference and let us know if we need to do something more to your satisfaction. Some of the extensive reviews done in the manuscript include the following.

1. We have explained the “jargon” that reduces readability in Section 4.2. We have also included them as a note below the Tables for easy referencing.
2. Extensive grammar checks have been done to improve readability
3. On agricultural productivity in Jiangsu over the years, we have discussed the results to improve the manuscript.
4. References have been updated and the most recent ones included to back our claims.
5. The DEA model is a broad model used recently by scholars in various fields. We adopted this model because we intend to analyze and compare the variables chosen from the Jiangsu province. We believe our initial discussion wasn’t sufficient which made it seem we deviated from using it. However, we believe the revised manuscript will be able to answer the questions posed. Some of these suggestions made about DEA can be considered in our future studies because we deem them valuable.
6. Lastly, we scrutinized the entire manuscript again and restructured the content to make it appealing and improve readability. The introduction part has been improved extensively with information about agricultural technology and productivity. The less relevant information has been removed, and the objectives and contributions to be made have been highlighted. We believe these changes are great efforts by the team to address your issues to your satisfaction.

We hope our revision and responses will meet your satisfaction this time. If there are additional things to do after this round of revision, please let us know.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Unfortunately, despite two rounds of opportunities for the authors, my comments have not been addressed.

- The characteristics of current research should be highlighted in the comparative table of literature review from both aspects of theoretical and application. In other words, a comprehensive literature review as well as research gaps should be summarized in in the comparative table.

- Generally, real data are tainted by uncertainty. The authors should discuss the proposed approach under data uncertainty. In other words, the authors can be applied uncertain DEA models such as stochastic DEA, Fuzzy DEA and Robust DEA. 

- The authors should compare their results and proposed approach with popular approaches in literature.