Economic Reforms, Labour Markets and Formal Sector Employment: Evidence from India
Round 1
Reviewer 1 Report
The paper is well written. The only trouble I had was inconsistent citations throughout the paper. Please ensure this is consistent.
Another issue is the English, even the first word of the first paragraph begins with "T" space then "In". I do not know what the authors wanted to write.
Author Response
This is my response to Reviewer 1's comments.
Reviewer 2 Report
The paper studies an interesting theme, namely how structural reforms affected formal employment in a sample of Indian states. Though the topic of the paper is interesting, there are significant problems and hence I recommend major revisions before proceeding to a possible resubmission.
Below is a list of my principal observations:
The paper seems disorganized. It presents a great deal of data but the author (s) is constantly referring to figures and tables in other sections of the document. I found it difficult to follow as I repeatedly had to go different pages to observe the information the author wanted me to see.
Figure 1 is unclear. A better description of what it's trying to convey is necessary
The indicator developed by Besley and Burgess - indicator used by the author to construct his (her) Flex 2 index - seems arbitrary and ambiguous, which may generate a number of technical problems. The author needs to justify the utilization of the indicator and of the Flex 2 index to show that the results reported are not biased
Are lines 403-412 and 413-422 the same?
All the regressions included in the paper lack control variables (i.e., gdp per capita, pop, coastal region or not, etc) and so omitted variable bias, among other problems, is a real concern
As argued by Brambor, et al (2005), it is difficult to assign a meaning to interaction coefficients, but the author seems to do just that by stating that the overall impact of reforms on employment is the sum of coefficients (lines 464-473).
Endogeneity checks should go to the appendix
Author Response
Response to Reviewer 2 Comments
Point 1: The paper seems disorganized. It presents a great deal of data but the author (s) is constantly referring to figures and tables in other sections of the document. I found it difficult to follow as I repeatedly had to go different pages to observe the information the author wanted me to see.
Response 1: I provided a number of references to figures and tables in the document as I wanted to be clear and specific (the journal template does not appear to allow for cross-referencing). However, if it helps with readability, I am happy to remove some of these references in a published version of the paper.
Point 2: Figure 1 is unclear. A better description of what it's trying to convey is necessary.
Response 2: I have attempted to provide a better description of Figure 1 in lines 278-292 (Page 7, changes tracked).
Point 3: The indicator developed by Besley and Burgess - indicator used by the author to construct his (her) Flex 2 index - seems arbitrary and ambiguous, which may generate a number of technical problems. The author needs to justify the utilization of the indicator and of the Flex 2 index to show that the results reported are not biased.
Response 3: Section 3.3 of the manuscript outlines the construction of the FLEX 2 index in some detail, and explains why it is likely to be the best indicator of labour market flexibility (from among the alternatives). Nonetheless, in Table 14 (Page 31), I show that the main results do not change, in terms of magnitude and direction, when we use the FLEX 1 index or the FLEX 3 index instead of the FLEX 2 index. (In the text, justification is provided in lines 877-881 on Page 30 and Page 31).
Point 4: Are lines 403-412 and 413-422 the same?
Response 4: Yes, that is correct, thank you. I have deleted lines 413-422 (now lines 422-431 on Page 13, changes tracked) to address that.
Point 5: All the regressions included in the paper lack control variables (i.e., gdp per capita, pop, coastal region or not, etc) and so omitted variable bias, among other problems, is a real concern.
Response 5: All the baseline regressions include state fixed effects, which account for time-invariant state-specific features (such as whether a state has a coastline or not). As regards variables which vary across states and over time (including GDP per capita and population), I include state-year interaction fixed effects in Column 3 of Table 13 (Page 29 and Page 30), and this does not affect my main findings. I have reiterated this in lines 844-850 (Page 29).
Point 6: As argued by Brambor, et al (2005), it is difficult to assign a meaning to interaction coefficients, but the author seems to do just that by stating that the overall impact of reforms on employment is the sum of coefficients (lines 464-473).
Response 6: Technically, the above is correct. I have made the necessary amendment to Section 4 (now lines 476-485, Page 14, changes tracked). As LMk is an indicator variable, however, this does not affect the interpretation of any of the results discussed in Section 5.
Point 7: Endogeneity checks should go to the appendix.
Response 7: I have addressed this by moving the endogeneity checks (originally Table 13 and Table 14) to the Appendix (now Table A1 and Table A2, on Page 37 and Page 38), and adjusting the text referencing where relevant (on Page 28 and Page 29).
Author Response File: 
Author Response.docx
Reviewer 3 Report
I consider that the article presents an interesting topic for India but it can be improved in some aspects:
The summary can be clearer, for example, the data used are cross-section, time series or panel? What are the most important implications of the results?
There are some errors in the document. For example, on line 28 start with a T.
The article is very extensive and not very clear in the ideas presented. I believe that the most important results and implications of the work could be simplified and presented clearly.
In the econometric part, it could be clearer what type of data is used and add the diagnostic tests of each model, for example, the normality and specification tests of the model.
In the conclusions, I recommend emphasizing the most important results and implications, weaknesses of the work and future lines of research.
Author Response
Response to Reviewer 3 Comments
Point 1: The summary can be clearer, for example, the data used are cross-section, time series or panel? What are the most important implications of the results?
Response 1: I have attempted to do this in the abstract or summary on Page 1 (changes tracked).
Point 2: There are some errors in the document. For example, on line 28 start with a T.
Response 2: The “T” in the first paragraph was a typo that arose when the paper was imported into the journal template – this has now been addressed and I do not think that there are further linguistic errors in the paper.
Point 3: The article is very extensive and not very clear in the ideas presented. I believe that the most important results and implications of the work could be simplified and presented clearly.
Response 3: The article is extensive as there is a lot of background and contextual work. I have moved some of the robustness checks (Section 6.2) to the Appendix (line 1148 onwards), and I believe that the introduction and conclusion (discussion) sections present the key findings in a clear and concise manner. I am also willing to move some of the contextual discussion to an appendix if that might help.
Point 4: In the econometric part, it could be clearer what type of data is used and add the diagnostic tests of each model, for example, the normality and specification tests of the model.
Response 4: I have attempted to describe the survey data more clearly in Section 3.1 (Page 7, changes tracked). As regards diagnostic tests, I do not believe that they pose a concern for this study, given the (very) large sample size and the implications for normality (through the Central Limit Theorem and similar).
Point 5: In the conclusions, I recommend emphasizing the most important results and implications, weaknesses of the work and future lines of research.
Response 5: Thank you. I have made a few changes to the conclusions or discussion section on Page 33 and Page 34 (changes tracked), emphasising the above.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Interesting final product
Author Response
-
Reviewer 3 Report
I believe that the article has improved and can be published in the journal.
Author Response
-
