Progresa

Question 1

Authors and year

Answer 1

Gertler - Boyce 2001

Question 2

Motivation for Progresa

Answer 2

people grow up in poverty and receive poor education, inadequate nutrition and little medical attention
when enter labour market, no adequate endowment of human capital to pull out of poverty
cycle: low HC, low productivity, low wage, no resources for children
intergenerational tragedy of poverty

Question 3

Health related CCT conditions

Answer 3

every family member gets preventative healthcare
children 0-5 and lactating mother must attend clinics for education and nutrition
pregnant women must receive pre natal care

Question 4

Education related CCT conditions

Answer 4

children attending high school
increases with higher grades
biased in favour of females - academic lit shows they have a higher dropout rate

Question 5

Other aspect to Progresa (not CCTs)

Answer 5

free fortified food supplements for young and low weight children

Question 6

How long did households receive CCT?

Answer 6

3 years, then reassessed

Question 7

take up rate of Progresa in elegible areas?

Answer 7

97%

Question 8

was Progresa transfer significant?

Answer 8

1/3 HH income, 46.5% of Mexican federal anti poverty budget

Question 9

Who were payments given to?

Answer 9

female head of household - in accordance with international literature

Question 10

Methodology: RCTs

Answer 10

• had 506 communities, roughly 10% of progresa program, base line survey conducted to later use as a proxy to check accuracy of responses.
• within these communities, eligible households identified using census data
  randomly assigned to control or treatment
  320 treatment, 185 control (received 2 years later)
  ensured random selection of communities that had the same characteristics
  Treatment and control groups had statistically indistinguishable characteristics (education, age, income)

Question 11

Data Sources

Answer 11

1997 Census for determining eligibility
Surveys - 1998 baseline (lack of health data collected) then 4 follow ups every 6 months
IMSS - Solidaridad for health centre utilization

Question 12

Surveyed questions to measure utilization

Answer 12

• attendance to clinics for preventative care
• in patient hospitalizations
• household visits to healthcare providers

Question 13

Data Handling Approach?

Answer 13

difference - in - difference

• controls for area specific characteristics and secular trends
• to compare impact of treatment before and after intervention in treatment and control villages

Question 14

Notes on importance of methodology

Answer 14

• D-I-D model controls for effects that change over time for both of them
• finding right comparison group is a challenge. if not equivalent comparison is not valid
• statistically unbiased only if compare similar groups of households that participate and do not

Question 15

Results

Answer 15

CHILD HEALTH IMPROVED
- children 1-5 have 12% lower incidence of illness
- stronger nutrition measured
ADULT HEALTH IMPROVED
- measured in terms of days incapacitated/in bed due to illness/ difficulties with daily activity
- not much dif 6-17, but less sickness reported in 18-50, 51+
- people more productive under Progresa
UTILZATION OF PUBLIC HEALTH CLINICS FOR HOSPITALIZATION FELL
NO REDUCTION IN PRIVATE PROVIDER UTILZATION
- no crowding out effect
UTILIZATION OF PUBLIC HEALTH CLINICS INCREASED
- for all age groups except 0-2 due to increase in immunized babies
NUTRITION MONITORING SYSTEMS STRENGTHENED

Question 16

Pros of Progresa Program

Answer 16

• replicated around world with similar results

- much more pressure for gov to act now as higher fertility rates increases importance of program

Question 17

Cons of Program

Answer 17

expensive to carry out (2%) GDP
how do you graduate from progresa?
political issues of under reporting income to qualify

Question 18

What does randomisation ensure?

Answer 18

that selection into treatment is uncorrelated to unobserved characteristics - this is very important when evaluating the benefits of a program because those that self select into it are different in many ways that are unobservable by an econometrician.

Question 19

What does a difference in difference model control for?

Answer 19

area specific characteristics and secular trends (occurs over long timeframe)

Question 20

How did adults get healthier?

Answer 20

12% less days off. whilst focus on children, still expected as improved quantity and quality of food supply in house and had to get preventative care

Question 21

What happened to public health centre and public hospital utilization rates

Answer 21

for treatment group, public health centre visits grew faster and hospital visits declined faster - in line with prediction