a few comments

Post Reply
Luca Tiberti
Posts: 23
Joined: Thu Jun 07, 2018 6:31 am

a few comments

Post by Luca Tiberti » Mon Jun 11, 2018 1:49 pm

Thanks Faith for your presentation. I have some comments I would like to share with you and your team:

- In the Introduction, you should soon say what you do, your main contribution and then provide some previews of the results.
- Background of the study (section 2) is too long. It is more appropriate for a report, but much less for a scientific paper. For example, we do not need of all those info and figures.
- Theoretical framework:
o I would encourage to use another model (see, e.g., Chang et al (2012) “Labor supply, income, and welfare of the farm household”, Labour Economics, 2012, vol. 19, issue 3, 427-437 - enclosed)
o There seems to be a few mistakes in your model (e.g., I do not see the production function, the FOCs do not seem fully correct and it is not clear how you derived them)
o Also, unclear how you moved from the FOCs to the empirical model (see between equations 6 and 7)
- Econometric specification:
o This section should be introduced better, as it starts too abruptly.
o Also, the econometric models should be chosen with respect to well identified econometric issues (such as sample selection). The current presentation does not follow such logic and it appears a bit confusing (you propose 5 econometric approaches, but they are not well documented with respect to why they are proposed to be used)
o The main outcome and explanatory variables should be justified by the theoretical framework, and defined here. For example, how off-farm employment is defined exactly? Why, in the regression analyses, you used an aggregate variable for off-farm employment? The various off-farm options are too heterogeneous to be combined together. I would personally privilege the off-farm alternative
- Empirical results:
o The discussion over the validity of the instrumental variables is too short and not yet fully convincing; we do not even know (unless I missed them) which are the IVs. And additional usual tests should be carried out.
o I found that the discussion on the determinants is too long. Your paper should not be a determinants paper, but a causal analysis.
o You found that off-farm employment reduces agricultural production and welfare, but the arguments provided to defend such (apparently counterintuitive) results are not convincing and are sometimes contradictory (see for example in the conclusions). The theoretical framework should help to find possible cases and dynamics why this would happen. As it is, the theoretical model does not help.
o You should define exactly the sample of analysis. Did you include only farmers with or without off-farm labour?
- Other points:
o All monetary variables should be expressed in real terms.
o I do not understand the definition of off-farm1: does it identify household with on-farm labour only?
o Which are the variables in the y-axis of figure 3?
o Check the number of observations per each off-farm variables (much lower than all other variables)
o Table A: should be improved. Are IVs for the off-farm combined variable? I am not convinced by the IV.
o Very different results (in terms of magnitude, but sometimes also for their sign) across the different models.
Chang et al..pdf
(572.21 KiB) Downloaded 58 times

Post Reply

Return to “Non-farm participation, agriculture productivity and farmers welfare in East-Africa – PMMA 20055”

Who is online

Users browsing this forum: No registered users and 1 guest