The Features of Rural Labor Transfer and Cultural Differences: Evidence from China’s Southwest Mountainous Areas
Round 1
Reviewer 1 Report
The paper studies the rural labor transfer using micro survey data of Xichang province in China. The authors find that the the employment in the agriculture sector has declined overtime during the period 2006-2016. The main findings of the study suggest that rural labor transfer have regionality, multi-pattern and gradient features.
One of the important issues with this paper is that the research question, literature review and findings are poorly explained.
The paper talks about spatial properties of rural labor transfer but ignores the seasonality aspect, time trends of rural labor transfer.
The figures are not clearly explained. In the figures 3, 4 and 5 there is sudden drop/increase in the indicators in the years 2008 and 2009 and the reason why we see that change is not clearly explained.
In table 3, number of rural laborers is negative. The reason behind this is not clear. The explanation for negative coefficient of urbanization is not clear. Also the authors mention that " urbanization rate is inferior to Number of rural transfer and crop acreage. I am not sure how the authors came to this conclusion.
Author Response
Point 1 One of the important issues with this paper is that the research question, literature review and findings are poorly explained.
Response 1: Thanks for your comments. We had clearly summarized the research question of our study in line 84-88 following your comments, and we compared the case study results with existing literature in discussion (references 3, 22, 23, 24, 25, etc.). By comparison, we answered the research questions raised in the introduction, and further detailedly explained the main conclusions of the manuscript. And more, we had further revised our discussions and conclusions in order to present more clearly our findings, contributions to academic development and its scientific significant in this field.
Point 2 The paper talks about spatial properties of rural labor transfer but ignores the seasonality aspect, time trends of rural labor transfer.
Response 2: Thanks. Your suggestions are good guidance and encouragement for our further research work, it will be an interesting topic to discuss the correlation effect of seasons on agricultural production and rural labor transfer. In the 3.3 part of the manuscript, although we analyzed the spatial properties of rural labor transfer, we did ignore the seasonal change of labor transfer. However, we mentioned the seasonality of part-time farm laborers transfer in 3.5 (lines 357-359), and discussed short-term (seasonality) and round-trip transfers in 4.2 (lines 428-429) in this revision. The main point of this paper is to compare the change of rural labor transfer over the past 10 years with the trend of inter-annual change. So, regular seasonal changes do not affect the results of inter-annual comparisons.
Point 3 The figures are not clearly explained. In the figures 2, 4 and 5 there is sudden drop/increase in the indicators in the years 2008 and 2009 and the reason why we see that change is not clearly explained.
Response 3: Thanks for your comments, we had modified detail in figures and added clear explanation in this revision per your comments. According to the verified source data in figure 2, the permanent population in 2007 decreased by 6,400 compared with 2006, while the registered population increased by 8,100 compared with 2006. As a result, the floating population in 2007 decreased by 14,500 compared with 2006. However, the permanent population in 2008 was higher than that in 2006 and began to grow gradually. As shown in figure 2, registered population and permanent population are represented on the left axis, while floating population is represented on the right axis, different units and scale cause seemingly significant drop.
Our population data comes from the statistical yearbook, so the statistical caliber error will bring some data error. Although we tried our best, we still can't find the permanent population data of Xichang before 2006, so we can't judge the exact reason of population change from 2007 to 2009. However, it is certain that influenced by the return of rural migrant workers from eastern cities, tourism development in Xichang and immigrants from hydropower project since 2009, the permanent population of Xichang began to grow rapidly, which led to a significant increase in the floating population in 2009 compared with 2008.
Figure 2
Thanks for your comments about figure 4, and we had made a clear and further revision and explanation in lines 221-227 per your suggestion. The industrial added value in 2008 was only 2.6 million yuan lower than that in 2007, while the transportation industry was 1.9 billion yuan higher than that in 2007. In May 2008, the Wenchuan earthquake had a devastating impact on the social economy and population in western Sichuan. After the earthquake, the central government invested heavily in rebuilding and upgrading infrastructure in the western mountainous areas of Sichuan to repair the damage caused by the earthquake, leading to a rapid increase in investment in the transportation industry. The industrial sector in Xichang suffered minor damage from the quake and recovered quickly in 2009. So that leads to the rise of transportation and the opposite decline of the industrial sector in 2008. However, the special changes in 2008 had limited influence on the change trend of Xichang over 10 years.
Point 4 In table 3, number of rural laborers is negative. The reason behind this is not clear. The explanation for negative coefficient of urbanization is not clear. Also the authors mention that " urbanization rate is inferior to Number of rural transfer and crop acreage. I am not sure how the authors came to this conclusion.
Response 3: Thanks for your comment very much and we had revised the manuscript following your comments. Point-by-point reply is as follows:
（1） We have made a clear and further explanation in this manuscript of lines 314-318. Number of rural laborers is negative because the townships with more rural laborers are located in the river valley (Figure 7 in manuscript), its agriculture is more developed, and more rural laborers are engaged in agriculture sector (the figure below). Thereby the towns themselves can provide sufficient agricultural employment opportunities and expectant incomes for rural labor, so it reduced ratio of rural labor transfer.
Figure. The relationship between the number of rural laborers
and the number of rural laborers engaged in agriculture
（2） We have made modifications in the manuscript following your comments, and the explanation of this part can be found in lines 321-323. Urbanization rate in township is just inner townships, and Ratio of rural labor transfer means that the proportion of the rural labor transferred to outside of the townships. The higher urbanization rate in townships, the more non-agriculture employment opportunities and less rural surplus laborers. So, local urbanization in townships can attract the surplus rural labor and reduce the willingness of transfer. Therefore, China now encourages small and medium-sized cities and towns to divert population from the rural labor who have transferred to large cities, which shows that the development of urbanization rate in township slow down the local rural labor transfer. Thereby Urbanization rate in township is negative coefficient.
（3） We had added explanation in lines 325-327 following your comments. The absolute value coefficient of Urbanization rate in township and crop acreage is less than that of Number of rural laborers, and the significance of Number of rural laborers and crop acreage is stronger than that of Urbanization rate in township (Table 3). Therefore, we believe that the negative impact of the Urbanization rate in township on the Ratio of rural labor transfer is inferior to Number of rural laborers and crop acreage. We had revised it to that “the hindering significance of Urbanization rate in township is inferior to Number of rural laborers and crop acreage”.
Author Response File: 
Author Response.docx
Reviewer 2 Report
In figure 2 is not clear if “Registered population” and “Permanent population” should be identified in the left or right axis.
Figure 3: is the change of population measured “per/km2”?
In figure 4 how would you explain the rise of transportation industry and the opposite decline of the industrial sector in 2008?
Table 3 shows that “Registered population emigration” is not statistically significant the proportion of rural labor transfer. Could you explain this (unexpected) result?
In the same table the variable “Number of rural laborers” shows a negative coefficient. Authors should better explain it.
According to the sensitivity of the “Urbanization rate in township”, it seems that the urbanization process should slow down over the time. This is against the empirical observation.
There are some typos (e.g., row 37: “his process includes a transfer” should be “this process includes a transfer”)
All figures and tables should be better described
Comments for author File: 
Comments.pdf
Author Response
Point 1 One of the issues of the paper is that Authors should explain more clearly the research question and how their contribution is original compared to the existing literature, since the paper states that “Over all the world, rural labor transfer is in line with classical theory”.
Response 2: Thanks for your comments. In this revision, we had clearly summarized the research question of our study in lines 84-88 following your comments, and we compared the case study results with existing literature in discussion (references 3, 22, 23, 24, 25, etc.). By comparison, we answered the research questions raised in the introduction, and made further modifications to the discussion and conclusions to clarify the scientific significance of our findings.
As for “Over all the world, rural labor transfer is in line with classical theory”, this expression is not very strict, we had modified it as "On global or national scale, rural labor transfer is in line with classical theory." in lines 393 following your comments. And what we're trying to say is that even though the study on a large scale is consistent with classical theory, on a small scale, the regional features and cultural differences of rural labor transfer are more significant.
Point 2 In figure 2 is not clear if “Registered population” and “Permanent population” should be identified in the left or right axis.
Response 2: Thank you. We had revised this error to identified“Registered population” and “Permanent population” per your suggestion. Now, “Registered Population” and “Permanent Population” is symbolized in the left axis and “Floating Population” is symbolized in the right axis.
Point 3 Figure 3: is the change of population measured “per/km2”?
Response 3: Thanks for your suggestions. In this version, we had modified the units to “per/km2” in Figure 3c per your suggestion.
Point 4 In figure 4 how would you explain the rise of transportation industry and the opposite decline of the industrial sector in 2008?
Response 4: Thanks for your comments, and we had made a clear and further revision and explanation in lines 221-227 following your comments. When we checked the source data again, we found that the industrial added value in 2008 was 2.6 million yuan lower than that in 2007, while the transportation industry was 1.9 billion yuan higher than that in 2007. In May 2008, the Wenchuan earthquake had a devastating impact on the social economy and population in western Sichuan. After the earthquake, the central government invested heavily in rebuilding and upgrading infrastructure in the western mountainous areas of Sichuan to repair the damage caused by the earthquake, leading to a rapid increase in investment in the transportation industry. The industrial sector in Xichang suffered minor damage from the quake and recovered quickly in 2009. So that leads to the rise of transportation industry and the opposite decline of the industrial sector in 2008. However, the special changes in 2008 had limited influence on the change trend of Xichang over 10 years.
Point 5 Table 3 shows that “Registered population emigration” is not statistically significant the proportion of rural labor transfer. Could you explain this (unexpected) result?
Response 5: Thanks for your comments. We had explained this unexpected result in lines 345-350 following your comments, and revised it for stating the reasons more clearly. In the social context of China, Transfer hukou is not easy for rural labor without housing property or systematic employment in urban, usually only young people with higher education can obtain urban hukou due to official employment. So, the promote role of “Registered population emigration” on the rural labor transfer is extremely limited.
Point 6 In the same table the variable “Number of rural laborers” shows a negative coefficient. Authors should better explain it.
Response 6: Thanks for your comments, and we had made a clear and further explanation in this revision of lines 314-318. Number of rural laborers is negative because the townships with more rural laborers are located in the river valley (Figure 7 in manuscript), its agriculture is more developed, and more rural laborers are engaged in agriculture sector (the figure below). Thereby the towns themselves can provide sufficient agricultural employment opportunities and expectant incomes for rural labor, so it reduced ratio of rural labor transfer.
Figure. The relationship between the number of rural laborers
and the number of rural laborers engaged in agriculture
Point 7 According to the sensitivity of the “Urbanization rate in township”, it seems that the urbanization process should slow down over the time. This is against the empirical observation.
Response 7: Thank you for your comments, and we had made a clear and further explanation in this revision of lines 321-323. This question is exactly the scientific issue that this manuscript wants to discuss. On a regional or large-scale scale, urbanization will certainly promote the transfer of rural labor over time, especially in large cities, which will attract more and more rural population. However, in small scale regions, local urbanization will slow down the excessive rural population to gather in central cities, especially township urbanization. In this study, the urbanization of townships provides non-agricultural employment and public services for the nearby rural labor, thus reducing the large-scale labor transfer to outside and increasing internal transfer in township. In fact, with the balanced development of regions, the urbanization of small townships is developing rapidly, while the central cities will not over expand and overcrowd. Therefore, China now encourages small and medium-sized cities and towns to divert population from the rural labor who have transferred to large cities, which shows that the development of urbanization rate in township slow down the local rural labor transfer.
Point 8 There are some typos (e.g., row 37: “his process includes a transfer” should be “this process includes a transfer”)
Response 8: Thank you very much. This is a simple mistake caused by our carelessness. We have checked full text to avoid similar mistake, and the manuscript was also professionally polished by AJE.
Pont 9 All figures and tables should be better described
Response 9: Thanks for your comments. We updated and revised the citation position of the full text of the figures and tables following your comments, and made a clearer explanation of the research results expressed by the figures and tables, especially Figure 9 and Table 3. In addition, we also modified some details in the figures and tables to present the research content more accurately.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The answers to the questions looks good to me. Thank you for making the changes as requested.
Author Response
we thank you very much for giving us an opportunity to improve our manuscript and appreciate you for the positive and professional comments and suggestions.
Reviewer 2 Report
Thank you for your response.
I only suggest to remove section 2.4.1 Multiple linear regression. It is really scholastic and I suppose all the readers of a scientifica paper already know it.
Author Response
Thank you for your suggestions. We have removed the section 2.4.1 about multiple linear regression per reviewer’s suggestions, and updated the references order and the sections order in the revision.
