Transcript El SISBEN: Su diseño y evolución - Stone Center for Latin American

Comments on:
The permanent effects of
recessions on child health: evidence
from Peru
By : Adriana Camacho.
Positive aspects
• Interesting and important for public health
policy
• Contributes to evidence in developing countries,
as the impact is the reverse of developed
countries
▫ A very good country to study given strong
recessions
• Deals with selection of mothers into phases of
the economic cycle.
Channels: Recession can affect Health
• Reduce (private and public) resources
1.
2.
1.
Nutrition
Medical care (prenatal, postnatal)
Breastfeeding
▫ Result on nutrition not counterintuitive in this case.
▫ Expect to see neonatal mortality flatter
▫ Exploit age heterogeneity with this channel
2a. Prenatal care visits
▫ If impact of recession starts during pregnancy, lag GDP to check
results with mortality.
2b. Stress
▫ Birth Weight, term of delivery
2c Postnatal access to medical care
▫ Control for type of health insurance or public health programs in
the area. (regional effects in the cross sections)
Include in the paper to Clarify of
Motivate
• literature on:
▫ early child human capital development.
• Health Production function
▫ clarify income-substitution effect
▫ Genetics (mother fe)
• Brief explanation of health care system for
pregnant woman and children.
• Shade recession years in the mortality figure
Other possible outcomes
•
•
•
•
•
•
Probability of breastfeeding
Birth weight
Term delivery
Probability of complete vaccination by age
Intestinal and Respiratory Health problems
Other Z scores for WfA, WfH (easier to change)
Other things to do
• Evidence of pro-poor anti-poor (unequal)
growth
• Twins !! control for them
▫ Could be used to do a robustness check of the
results.
• Heterogeneous effect by age
▫ child cohort
• Educational attainment of recession cohorts to
validate the claim “permanent mark on
children´s human capital”.
▫ Positive selection into survival.
Check
• Compare IMR constructed from DHS with vital
stats. Yours are really high!
• Explain better the way IMR was constructed
when there is an overlap of years.
• Show data of counter-cyclical public expenditure
• Strongest impact in the last two recessions
(mildest ones), evidence problems of
retrospective info?
Minor changes
•
•
•
•
•
Include urban area control
Descriptive stats of GDP and growth
Fig 2. Mortality not Infant Mortality
Explain “continuous” 2004 DHS
Growth rates in fig do not coincide with number in
paper
• Drop Aguero and Robles citation (credit constraints
not tested).
• Conclusions have long discussion on education??
(Drop fig 5, 6 and 7ª). (only significant for non
educated)
Colombia:
8.00%
5.0%
4.0%
7.90%
3.0%
7.80%
7.70%
1.0%
7.60%
0.0%
-1.0%
7.50%
-2.0%
7.40%
-3.0%
7.30%
-4.0%
7.20%
-5.0%
1998
1999
2000
2001
2002
2003
gdp growth
lbw
2.0%