Comments made by Discussants on 11 June 2018

Dikshya Singh
Posts: 1
Joined: Fri Jun 08, 2018 3:21 am

Comments made by Discussants on 11 June 2018

Post by Dikshya Singh » Tue Jun 12, 2018 4:02 am

Congratulations for such an insightful presentation. Here are some of our comments.

• You have rightly pointed out in the proposal that panel data is being used due to account for unobserved heterogeneity. It is our understanding that in your panel data same households are tracked over time. In your specification (1), fixed effects for time are there but there are not any household fixed effects. Just wanted to point out if it is a typo or are you thinking of some other way to account for the heterogeneity? If household fixed effects are not to be used or cannot be used with the method you propose to employ, then the advantage of using panel data would be the increase in sample size, not the ability to sweep out time-invariant unobserved heterogeneity. Please clarify this.

• Highlight the difference between internal and external remittance/migration.

• It would be great if you can point out the composition of the migrant household, whether both father and mother migrate? What is the trend? Decision making and/or expenditure characteristics may vary depending on who is in-charge in the migrant household.

• A little more insight into educational system in Cambodia could be helpful. If the migrant or remittance-receiving households were sending their children to public school or private schools. I skimmed through some news stories published in Cambodian media that said enrolment rates in the private schools are growing. But, then again that could be only in urban areas. If there is some similar distinction in rural areas as well then it could provide clearer picture. If the enrolment is more in the public schools, which are free, then increased income due to remittance may have less direct impact, unlike in the cases of private schools that charge hefty fees. For public schooling, increased income due to remittance may reduce the urgency for households to make their kids skip school in favour of household chores, presuming they could afford hiring help. So if possible, you could consider having three categories in your outcome variable: not enrolled, enrolled in private school and enrolled in public school. Of course, doing this would call for exploring an alternative specification -- for example, something along the lines of multinomial logit. This point could help recommending some action-oriented policy measures related to education policies, if there is some differentiated impact based on private versus public schooling.

• In the policy implications front also, findings of this study could help targeting specific policies and programs, for example, maybe providing better guidance to the young people that might have been forced to leave school to take care of family members.

Posts: 5
Joined: Mon Jun 04, 2018 1:05 am

Re: Comments made by Discussants on 11 June 2018

Post by chanhang » Tue Jun 12, 2018 9:55 am

Hello Dykshya Singh,

Thank you very much for your comments.



Return to “Migration, Remittances and Child Schooling in Rural Cambodia – PMMA 20209”

Who is online

Users browsing this forum: No registered users and 2 guests