Case Studies 2

Case Studies
ECO2117
Myra Yazbeck
Department of Economics
University of Ottawa
Myra Yazbeck
c (University of Ottawa) 1 / 30
FREE DISTRIBUTION OR COST-SHARING ? EVIDENCE FROM A
RANDOMIZED MALARIA PREVENTION EXPERIMENT.
Quarterly Journal of Economics (2010)
(Case Study 1)
Myra Yazbeck
Malaria and insecticide treated bednets
We agreed last time that we ideally we would like to be able to compare the
purchase of bednets at difference prices (where prices were random...).
The reality is that some clinics in some villages may give bednets for free other
villages may not have this same system.
What is the right counterfactual in this case ?
It might be the case that these bednets are free where malaria is a huge
problem and in this case people would have paid for them anyway because they
need them.
So how can we think of this situation in this case ?
Myra Yazbeck
Purchase
The situation looks this when bednets
way when are
bednets are expensive
expensive :
Purchases
High Low
malaria malaria
Myra Yazbeck
What if the bednets were given for free ? How can we think of this
situation in this case ?
Myra Yazbeck
The situation Purchase
looks this way when bednets
when bednets are
are free : free
Purchases
High Low
malaria malaria
Myra Yazbeck
What is the true effect ?
Myra Yazbeck
The true effectTrue
of priceeffect of price
on purchase on way
looks this purchase
:
Free
Free
Expensive
Purchases
Expensive
High Low
malaria malaria
Myra Yazbeck
What if we compared free region with expensive region as they are stated
initially ?
Myra Yazbeck
Estimate of the effect
Our estimate of price
of effect if weif compare
we were low
to compare
and highlow malaria
malaria region vs. high
regions
malaria region :
Estimate
d effect
Purchases
High Low
malaria malaria
Myra Yazbeck
How should we think of the bias ?
Myra Yazbeck
The bias
Bias in the estimation :
Effect Bias
Effect
Purchases
Bias
High Low
malaria malaria
Myra Yazbeck
Why does it matter ? remember that we are trying to estimate the

elasticity of demand.
Myra Yazbeck
Observed demand at various prices
Demand observed at different prices :
Purchase
0 10 20 30
Myra Yazbeck
Demand we would observe in region with free bed net, if bednets were not free
Demand observed in free bednets areas had bed nets not been free :
Purchase
0 10 20 30
Myra Yazbeck
Bias in elasticity
Bias in the elasticity of demand :
Purchase
Myra Yazbeck
Problem :
If we rely on what we observed : two places with two different prices

(with no further information) we can end up picking a selection bias
effect.
Selection bias : here when faced with the same price, people
behave differently when they are in different places (high malaria
low malaria region).
Selection bias will mask the true effect.
In the setting we have at hand we cannot separate the selection bias
effect from the true effect : we do not know how people would have
behaved with a low price in the high price region (and vice versa).
What is the solution in this case ?
Myra Yazbeck
A solution :
Randomly assign prices in the same region.

People who face high prices and people who face low prices have
no systematic differences.
Note that there will be some random noise → the sample must be
large enough and there will be some uncertainty around the mean
effects.
Myra Yazbeck
Pascaline Dupas and Jessica Cohen conduct a field experiment in which

they offer bednets at different prices in different clinics (these clinics were
randomized).
Track purchase and usage in those clinics.
Compare purchase and usage at each price.
Myra Yazbeck
Myra Yazbeck
Myra Yazbeck
Number of sales falls quickly with price.

Usage conditional on getting the net does not fall.
Effective take up falls quickly with price.
We will revisit this question in the health chapter again.
Myra Yazbeck
CASH, FOOD OR VOUCHERS ?EVIDENCE FROM RANDOMIZED
EXPERIMENT IN NORTHERN ECUADOR.
Journal of Development Economics (2014)
(Case Study 2)
Myra Yazbeck
Cash, food, or vouchers ? Evidence from a rand. exp. in northern Ecuador
Responding to a request from the government of Ecuador in April 2011, the

World Food Program (WFP) expanded its assistance to address the food
security and nutrition needs of Colombian refugees and to support their
integration into Ecuadorian communities.
The new program was designed as a prospective randomized control trial and
consisted of six monthly transfers of cash, food vouchers, or food to Colombian
refugees and poor Ecuadorian households.
The objectives of the program were three-fold :
To improve food consumption by facilitating access to more nutritious
foods.
To increase the role of women in household decision-making related to
food consumption.
To reduce tensions between Colombian refugees and host Ecuadorian
populations.
Myra Yazbeck
Cash, food, or vouchers ? Evidence from a rand. exp. in northern Ecuador
The program was implemented in seven urban centers in the provinces of Carchi
and Sucumbos. Both Carcgi and Sucumbos are northern borders provinces that
receive high influxes of Colombian refugees and cross-border traffic.
Carchi is located in the northern highlands and Sucumbos is located in the
Amazonian lowlands → each has distinct cultural, socioeconomic and
geographic features.
Neighbourhoods (Barrios) are chosen within these urban centers for intervention
with the World Food Program in consultation with the United Nation High
commissioner for Refugees.
Each household in the selected barrios was visited, mapped, and administered
a one-page questionnaire that consisted of basic demographic and
socioeconomic questions. These questions were used to develop a proxy
means test to define program eligibility.
Myra Yazbeck
Case Study : Inclusion/Exclusion
Based on point scores by nationality, the decision was made to automatically

enroll all Colombian and mixed-nationality households
All households who reported receiving the government’s social safety net
transfer program, the Bono de Desarrollo Humano (BDH) were automatically
excluded from eligibility.
→ Households residing in the selected barrios with low socioeconomic status as
measured by the proxy means test that met the criteria described above were
eligible to participate in the program.
During enrollment and sensitization, the program was described as a poverty
and food security transfer targeted toward women, and therefore, the majority of
the entitlement cardholders were expected to be women (if not woman in the
household men were allowed to participate).
Overall, approximately 79% of cardholders in Carchi and 73% of cardholders in
Sucumbos were women.
Myra Yazbeck
Treatments : Transfers in kind and cash
Participating households received benefits from May 2011 to October 2011.
The value of the monthly transfer was standardized across all treatment arms at
US $40 per month per household. The $40 was transferred monthly onto
pre-programmed ATM cards (they had to be taken out in bundles of $10).
The food vouchers were also valued at $40 and given in denominations of $20,
redeemable for a list of nutritionally-approved foods at central supermarkets in
each urban center. The list of approved foods consisted of cereals, tubers, fruits,
vegetables, legumes, meats, fish, milk products, and eggs (note : they are
non-transferable).
The food transfer was valued according to regional market prices at $40 and
included rice (24 kg), vegetable oil (4 l), lentils (8 kg), and canned sardines (8
cans of 0.425 kg).
Myra Yazbeck
Treatments : Education
To ensure a consistent approach to knowledge transfer, a curriculum was

developed by WFP to be covered during each monthly training session.
Program sensitization and information.

Family nutrition.
Food and nutrition for pregnant and lactating women.
Nutrition for children ages 0 to 12 months.
Nutrition for children ages 12 to 24 months.
All participants regardless of transfer modality participated in this training,
and transfers were conditional on attendance.
There were constant flyers distributed to remind participants to be mindful
of diets and nutrition facts etc...
Myra Yazbeck
Randomization
The program evaluation was based on random assignment. Randomization was

conducted in two stages :
First stage : barrios were randomized to either the treatment group or the
control group ;
Second stage : all treatment clusters (geographical units within barrios)
were randomized to cash, food voucher, or food transfer.
Because the geographic area in each urban center was relatively small, this
measure was taken to avoid having a cluster assigned to the control group
within the same barrio as a cluster assigned to the treatment group and
consequently causing discontent among potential beneficiaries.
In total 80 barrios and 145 clusters were randomized into the four intervention
arms control, cash, vouchers and food.
Myra Yazbeck
Randomization
Estimation strategy relies on the randomized design of the transfer program.
Random assignment of clusters assures that, on average, households will have

similar baseline characteristics across treatment and control arms.
Such a design eliminates systematic differences between beneficiaries and

non-beneficiaries and minimizes the risk of bias in the impact estimates due to
“selection effects".
Myra Yazbeck

Case Studies 2

Încărcat de

Informații document

Descriere originală:

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Case Studies 2

Încărcat de

Drepturi de autor:

Formate disponibile

Case Studies

What is the right counterfactual in this case ?

So how can we think of this situation in this case ?

What is the true effect ?

How should we think of the bias ?

Why does it matter ? remember that we are trying to estimate the

If we rely on what we observed : two places with two different prices

Randomly assign prices in the same region.

Pascaline Dupas and Jessica Cohen conduct a field experiment in which

Number of sales falls quickly with price.

Responding to a request from the government of Ecuador in April 2011, the

Based on point scores by nationality, the decision was made to automatically

Participating households received benefits from May 2011 to October 2011.

To ensure a consistent approach to knowledge transfer, a curriculum was

Program sensitization and information.

The program evaluation was based on random assignment. Randomization was

Estimation strategy relies on the randomized design of the transfer program.

Random assignment of clusters assures that, on average, households will have

Such a design eliminates systematic differences between beneficiaries and

S-ar putea să vă placă și