The Impact of Earthquake on Poverty: Learning from the 12 May 2008 Wenchuan Earthquake
Stephen Platt
Round 1
Reviewer 1 Report
Sustainability
The impact of earthquake on poverty: learning from 2 the 12 May 2008 Wenchuan earthquake
This is an important and topical issue that needs this kind of case study research. The idea of relating policy response to the type or level of poverty and level of disaster impact is very interesting and worthwhile.
In their abstract the authors state "Already ten years on, it is a good 19 time to look back on what we learned from the Great Wenchuan earthquake." This implies that they have data for 2017, and intervening years, that they will compare with the pre-disaster state. Yet most of the data presented is only for 1 year after the disaster in 2008. This creates a false impression and needs correcting either by presenting most recent data or amending the abstract and introduction.
They also say " How to combine disaster prevention and mitigation, post-disaster reconstruction and poverty alleviation is a new hot issue." This implies that they will discuss how in Wenchuan the recovery process succeeded or failed to alleviate poverty. They should consider presenting further evidence of this and discuss it further in a discussion section or in the conclusion.
The main issue with this paper, however, is that the authors seem to have got lost in the data and have consequently produced a confusing paper. It would be most helpful if the authors clarified how they present the data. The main distinction is between poverty-stricken and non-poverty-stricken. But data is presented at county, village and households level. There are also wealthy, ordinary, and poverty-stricken households. The authors hop about from one level of data to another and this causes confusion.
It might also be helpful, since so much data is presented, if the authors limited themselves to 2 rather than 4 significant figures e.g. 18% rather than 17.84%. They do this in some places e.g. L193-194.
In detail:
L87 How are "poverty incidence", "poverty depth" and "poverty severity" defined. Would it be sensible to use just one of these indicators? It would be useful to plot the change in poverty over time from pre-disaster to date.
L91 Poverty severity in 2008 107.6% in relation to what?
L97 How many people/households were displaced and for how long?
L107 "According to statistics, the incidence of disasters in poverty-stricken areas is 5-10 times higher than that in the ordinary areas [19]". Is this correct? It seems very high. What evidence is there? I couldn't find this reference Liu, Y.; Yuan, P.; Jia, H.; et al. Time trend analysis of PTSD post-Wenchuan earthquake. Chin J Public Health 2011, 27, 303–304.
L110-113 It is quite difficult to follow the difference between national and provincial key poverty alleviation counties.
L125 Figure 3. What is this telling us? One would obviously expect per capita losses to be lower in poverty-stricken counties since people have less to lose. But the death rate for non-poverty and provincial level poverty-stricken counties is similar. Is this small difference statistically significant?
L134 Yet this seems to be contradicted by "Whether the number of deaths or the proportion of people killed by the disaster, the non-poverty villages are far larger than the poverty villages, 6.9 times and 4.0 times respectively." Which is correct?
Why as the death rate so much higher in non-poverty-stricken villages, but not in counties. Is the important statistic at the village level and not the county?
L146 It is unclear how the various Area A1-A4 villages are defined. This is important since the authors suggest basing policy on these types of village. Maybe a table with definitions would be helpful.
L176 What is "counties >100 ten thousand" is this >100,000 or >1,000,000? Same for "<20 ten thousand".
L194 "Most middle income and wealthy households have suffered only minor damage to their houses." Yet the causality rate was much higher in non-poverty villages (L134).
L200 We are given census data at county level for poverty-stricken and non-poverty-stricken counties but it would useful to know what type of house construction caused the highest casualties in villages?
L209 The building collapse rate for different income levels is given: wealthy households 53%
L174-219 Would it not be clearer to put this data in a data and then only refer to the salient points in the text?
L263 Would it have been possible to fine-tune the investment in disaster resilience measures for different types of village along the lines the authors suggest in L146? Should these differences be reflected in Figure 8, which currently seems to have a singles global policy for Recovery and Reconstruction.
L293-314 The conclusion merely summarises the impact of the disaster on different economic groups rather offering, "Suggestions for improving the mode of combining 22 disaster prevention, post-disaster reconstruction and poverty alleviation were proposed" as promised in the introduction and abstract.
L312 It is not clear how the authors propose to combine "disaster prevention and mitigation/disaster reconstruction with poverty alleviation".
Although understandable there are many issues of English expression that need correcting to achieve clarity, for example:
L42 Change to: As result of this earthquake a large number of people are classed as “3-no personage”, citizens with no housing, no means of production, and no source of income.
L63 Change to: Most of the severely affected counties are in high plateau or mountainous areas.
L200 Change to: In poverty-stricken counties 18% of house were of brick-concrete construction and the rest were of other construction; in contrast in non-poverty counties 37% of houses were of brick-concrete construction (Figure 6).
L228 Change to: Wealthy households increased their capita income slightly in the year following the disaster, whereas in medium income and poor households per capita income significantly reduced.
Author Response
Reviewer 1:
The impact of earthquake on poverty: learning from 2 the 12 May 2008 Wenchuan earthquake
This is an important and topical issue that needs this kind of case study research. The idea of relating policy response to the type or level of poverty and level of disaster impact is very interesting and worthwhile.
Reply:
Thank you for your encouragement and concern. We are trying hard to improve our manuscript. We appreciate for your warm review work earnestly. The comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches.
In their abstract the authors state "Already ten years on, it is a good 19 time to look back on what we learned from the Great Wenchuan earthquake." This implies that they have data for 2017, and intervening years, that they will compare with the pre-disaster state. Yet most of the data presented is only for 1 year after the disaster in 2008. This creates a false impression and needs correcting either by presenting most recent data or amending the abstract and introduction.
Reply:
Thank you for your encouragement and concern. We are trying hard to improve our manuscript. We agree with this comment. It is really true as reviewer pointed out that most of the data presented is only for 1 year after the disaster in 2008. We have presented most recent data for years of 2016 and 2017 (Figure 3 and Figure 9) and analysis according to the Reviewer’s suggestion (Page 4, line 112-116; Page 11, line 315-316; Page 12, 317-319).
They also say " How to combine disaster prevention and mitigation, post-disaster reconstruction and poverty alleviation is a new hot issue." This implies that they will discuss how in Wenchuan the recovery process succeeded or failed to alleviate poverty. They should consider presenting further evidence of this and discuss it further in a discussion section or in the conclusion.
Reply:
We greatly appreciate the reviewers' comment. As Reviewer suggested that we have added how in Wenchuan the recovery process succeeded to alleviate poverty in the conclusion according to the Reviewer’s suggestion (Page 14, line 370-393; Page 15, line 394-406).
The main issue with this paper, however, is that the authors seem to have got lost in the data and have consequently produced a confusing paper. It would be most helpful if the authors clarified how they present the data. The main distinction is between poverty-stricken and non-poverty-stricken. But data is presented at county, village and households level. There are also wealthy, ordinary, and poverty-stricken households. The authors hop about from one level of data to another and this causes confusion. It might also be helpful, since so much data is presented, if the authors limited themselves to 2 rather than 4 significant figures e.g. 18% rather than 17.84%. They do this in some places e.g. L193-194.
Reply:
We agree with this comment. Thank you for pointing this out. It is really true as reviewer pointed out that It would be helpful that we limited much data to 2 rather than 4 significant figures. We have made all the corresponding changes in the revised paper (Page 1, line 40-42; Page 3, line 85-88; Page 4, line 95-98; Page 5, line 123-127; Page 5, line 146-148; Page 7, line 198-215; Page 8, line 232-235; Page 9, line 242-253; Page 10, line 262-263; Page 10).
In detail:
L87 How are "poverty incidence", "poverty depth" and "poverty severity" defined. Would it be sensible to use just one of these indicators? It would be useful to plot the change in poverty over time from pre-disaster to date.
Reply:
We agree with this comment. The definition and calculation formula are as follows,
Poverty incidence (P1) refers to the proportion of the population with disposable income below the poverty line in the total population.
X is the poverty line, n is the poor population, and N is the total population.
Poverty depth (P2) refers to the accumulated poverty gap based on the income level of the poor relative to the poverty line.
ui represents the income level of the No. i poor people, X is the poverty line, n is the poor population, and N is the total population.
Poverty intensity (P3) is based on the income level of the poor relative to the poverty line.
ui represents the income level of the No. i poor people, X is the poverty line, n is the poor population, and N is the total population.
Using one indictor “poverty incidence’, we have plotted the change in poverty over time from pre-disaster to date according to the Reviewer’s suggestion in the revised paper (Page 4, line 112-116).
L91 Poverty severity in 2008 107.6% in relation to what?
Reply:
Thank you for pointing this out. The definition and calculation formula of Poverty severity are as above. Poverty severity is just an indicator. We quote the data in the reference (Page 4, line 98).
L97 How many people/households were displaced and for how long?
Reply:
Thank you for pointing this out. The earthquake left about 4.8 million people homeless. More than 10 million people displaced. Just three months later, the people in the disaster-stricken areas were resettled. This data is only the result of an online search and there is no clear reference. So we didn’t put it in the manuscript.
L107 "According to statistics, the incidence of disasters in poverty-stricken areas is 5-10 times higher than that in the ordinary areas [19]". Is this correct? It seems very high. What evidence is there? I couldn't find this reference Liu, Y.; Yuan, P.; Jia, H.; et al. Time trend analysis of PTSD post-Wenchuan earthquake. Chin J Public Health 2011, 27, 303–304.
Reply:
We agree with this comment. Thank you for pointing this out. This data may be not clear, in order to be more scientific, we deleted this sentence in the revised paper (Page 5, line 120-121).
L110-113 It is quite difficult to follow the difference between national and provincial key poverty alleviation counties.
Reply:
We greatly appreciate the reviewers' comment. It is really true as reviewer pointed out that it is quite difficult to follow the difference between national and provincial key poverty alleviation counties. We have changed it into poverty-stricken counties and non-poverty counties according to the Reviewer’s suggestion in the revised paper (Page 5, line 128-141).
L125 Figure 3. What is this telling us? One would obviously expect per capita losses to be lower in poverty-stricken counties since people have less to lose. But the death rate for non-poverty and provincial level poverty-stricken counties is similar. Is this small difference statistically significant? L134 Yet this seems to be contradicted by "Whether the number of deaths or the proportion of people killed by the disaster, the non-poverty villages are far larger than the poverty villages, 6.9 times and 4.0 times respectively." Which is correct? Why as the death rate so much higher in non-poverty-stricken villages, but not in counties. Is the important statistic at the village level and not the county?
Reply:
Thank you for pointing this out. It is quite difficult to follow the difference between national and provincial key poverty alleviation counties. We have changed it into poverty-stricken counties and non-poverty counties according to the Reviewer’s suggestion in the revised paper. The reason why as the death rate so much higher in non-poverty-stricken villages is the structure of the house. The poor villages are mostly bamboo and wood structures with good seismic performance. Non-poverty villages are dominated by bricks and bricks, but these houses in rural areas lack design standards for seismic fortifications, and collapses lead to high mortality. These houses in rural areas lack the design of seismic fortification standards, mostly for farmers to build themselves. After the collapse, the damage caused to the personnel is greater than that of the bamboo and wood structure, resulting in high mortality (Page 5, line 128-141; Page 6, line 154-159).
L146 It is unclear how the various Area A1-A4 villages are defined. This is important since the authors suggest basing policy on these types of village. Maybe a table with definitions would be helpful.
Reply:
We greatly appreciate the reviewers' comment. We agree with this comment. This is important since the authors suggest basing policy on these types of village. We have added a figure according to the Reviewer’s suggestion in the revised paper (Page 13, line 338-341; figure 10).
L176 What is "counties >100 ten thousand" is this >100,000 or >1,000,000? Same for "<20 ten thousand".
Reply:
Thank you for pointing this out. "counties >100 ten thousand" is this >1,000,000. We have made corresponding changes in the revised paper (Page 7, line 197-199).
L194 "Most middle income and wealthy households have suffered only minor damage to their houses." Yet the causality rate was much higher in non-poverty villages (L134).
Reply:
Thank you for pointing this out. This data is in general terms. But in order to be more scientific, we deleted this sentence in the revised paper (Page 8, line 222-225).
L200 We are given census data at county level for poverty-stricken and non-poverty-stricken counties but it would useful to know what type of house construction caused the highest casualties in villages?
Reply:
We agree with this comment. It would useful to know old-fashioned brick wood and other types of house construction caused the highest casualties in villages (Page 9, line 238-254).
L209 The building collapse rate for different income levels is given: wealthy households 53%
Reply:
Thank you for pointing this out. The proportion of all completely collapsed, partially collapsed, and all completely dangerous buildings of wealthy households reachised 52.833%, while that of the ordinary households were is 72.26%, and that of the poverty-stricken households reached is 78.28% (Figure 7). We have made corresponding changes in the revised paper (Page 9, line 241-244).
L174-219 Would it not be clearer to put this data in a data and then only refer to the salient points in the text?
Reply:
We agree with this comment. Data is the support of the discussion. Thank you for your suggestion, we made the corresponding changes according to the Reviewer’s suggestion in the revised paper (line 190-254).
L263 Would it have been possible to fine-tune the investment in disaster resilience measures for different types of village along the lines the authors suggest in L146? Should these differences be reflected in Figure 8, which currently seems to have a singles global policy for Recovery and Reconstruction.
Reply:
We greatly appreciate the reviewers' comment. We agree with this comment. We have reflected the different types of village (A1, A2, A3, A4 types) and added it into Figure 10 according to the Reviewer’s suggestion in the revised paper (Page 13, line 337-341).
L293-314 The conclusion merely summarises the impact of the disaster on different economic groups rather offering, "Suggestions for improving the mode of combining 22 disaster prevention, post-disaster reconstruction and poverty alleviation were proposed" as promised in the introduction and abstract.
Reply:
We agree with this comment. Thank you for pointing this out. We have made corresponding changes in the revised paper (Page 14, line 370-393; Page 15, line 394-408).
L312 It is not clear how the authors propose to combine "disaster prevention and mitigation/disaster reconstruction with poverty alleviation".
Reply:
Thank you for pointing this out. The post-disaster reconstruction and poverty alleviation should consider the relationship between the degree of disaster impact and the degree of poverty. We have made corresponding changes in the revised paper (Page 6, line 177-183; Page 15, line 403-408).
Although understandable there are many issues of English expression that need correcting to achieve clarity, for example:
L42 Change to: As result of this earthquake a large number of people are classed as “3-no personage”, citizens with no housing, no means of production, and no source of income.
Reply:
We greatly appreciate the reviewers' comment. We have modified it according to the Reviewer’s suggestion in the revised paper (Page 1, line 43-44).
L63 Change to: Most of the severely affected counties are in high plateau or mountainous areas.
Reply:
We greatly appreciate the reviewers' comment. We have modified it according to the Reviewer’s suggestion in the revised paper (Page 3, line 68).
L200 Change to: In poverty-stricken counties 18% of house were of brick-concrete construction and the rest were of other construction; in contrast in non-poverty counties 37% of houses were of brick-concrete construction (Figure 6).
Reply:
We greatly appreciate the reviewers' comment. We have modified it according to the Reviewer’s suggestion in the revised paper (Page 8, line 231-233).
L228 Change to: Wealthy households increased their capita income slightly in the year following the disaster, whereas in medium income and poor households per capita income significantly reduced.
Reply:
We greatly appreciate the reviewers' comment. We have modified it according to the Reviewer’s suggestion in the revised paper (Page 10, line 263-265).
We appreciate for the reviewer’ warm review work earnestly. The comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Once again, thank you very much for your comments and suggestions.
Reviewer 2 Report
This is a major contribution to the knowledge of the impact of the Wenchuan earthquake on poverty and related demographic factors. The survey data is very valuable and shows at the micro level the differential impacts on various types of households/farmers etc. I would like to see the Conclusions discussed in more detail. Particularly lines 305 to 310 need to be clarified and highlighted more in their importance. Language needs to be improved and edited.
Author Response
Reviewer 2:
This is a major contribution to the knowledge of the impact of the Wenchuan earthquake on poverty and related demographic factors. The survey data is very valuable and shows at the micro level the differential impacts on various types of households/farmers etc. I would like to see the Conclusions discussed in more detail. Particularly lines 305 to 310 need to be clarified and highlighted more in their importance. Language needs to be improved and edited.
Reply:
We greatly appreciate the reviewers' comment. We have made some clarifications and Suggestions in the part of Conclusions (Page 14, line 370-393; Page 15, line 394-408). With the help of Charlesworth Author Services ([email protected]), a native English speaker has rechecked the language carefully throughout. We believe that the presentation of this revised manuscript has been greater improved, especially the English writing. Many syntactical errors in the whole text have been carefully corrected (Page 1-16).
We appreciate for the reviewer’ warm review work earnestly. The comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Once again, thank you very much for your comments and suggestions.
Round 2
Reviewer 1 Report
Well done for addressing my previous comments.
