Presenation

advertisement
By
Eugenie W. H. Maïga
African Center for Economic Transformation
(ACET)
Background on Burkina Faso
 Land area = 274,000 km2, Population =15.21
million (2008), GNI per capita =$480 (2008).
 In Burkina Faso, the prevalence rate for stunting
(height-for-age) was 39% in 2003.
 In 2009 the gross school enrolment rate for
primary, secondary, and higher education were
78%, 20%, and 3%, respectively.
Gross enrollment rates in primary school by
province in 2004-2005
Source: Ministere de l’Enseignement de Base et de l’Alphabetisation, 2005
Research problem
 Child health and nutrition outcomes in developing
countries are disappointing (prevalence of
stunting in preschool children was 33% in 2000,
De Onis et. al., rates between 30% and 39% are
considered high).
 Link between child health and child nutrition:
malnourished children are more likely to develop
illnesses that can have long lasting effects
throughout their lives.
 Mothers play an important role in child nutrition.
How does her education come into play?
Why are child health outcomes important?
 Costs of related illnesses (physical suffering, time costs
and monetary costs).
 Impacts on ability to learn i.e. on schooling outcomes.
 Impacts on productivity later in life which could
adversely affect economic growth
Objectives
 General objective:
• Examine the relationship between maternal education
and child health in Burkina Faso
 Specific objectives:
• Verify whether a strong causal relationship exists
• Understand the channels through which mother’s
education affects child health
• investigate whether threshold effects exist, that is,
whether specific years or levels of mothers’ education
have unusually large impacts on children’s health.
Change in education policy in Burkina Faso
 Primary education: construction of 3-year primary
schools Ecoles Satellites, ES) in areas of 28
provinces (out of 45) where enrolment rates are
low. As of 2008, there were 304 ES.
 Became effective in the school year 1995-1996.
Instruction is done in two languages, French and
language spoken in the area.
 Trying to favor girls enrollment in those schools
Evolution of enrollment rates in provinces
with and without the program
Provinces with program
Provinces without program
Previous Evidence
 Empirical studies without pathways (did not use
natural experiment): Baya (1998), Appoh and
Krekling(2005).
 Empirical studies without pathways (used natural
experiment): Breierova and Duflo (2004) and
Chou et. al. (2007).
 Empirical studies with pathway(s): Thomas et al.
(1991), Glewwe (1999), Handa (1999), and Webb
and Block (2004).
Contribution
 Many have studied the subject but few studies
have used natural experiments (change in policy)
in identifying the relationship between mother’s
education and child health in developing countries
 Use of distance to school as IV for mother’s
education in a study of the impact of mother’s
education on child health
 Use of both natural experiment and pathways
Conceptual framework
 Draws on Becker’s model of the family, and Rosenzweig
and Schultz (1983) health production function
 Consider the following equations:
U  u (X, H )
H  h(Z , I,  )
M  p1 X  p 2 Z  p 3 I
 Where X is market goods, Z represents goods that affect
child health, H is child health, I represents inputs that
affect utility only through their effect on H and μ is child
health endowment.
Conceptual framework
 Estimating H requires detailed data on health inputs, data
which are not available to this study. A way out of that, is to
use reduced form equations.
 Maximizing U subject to the health production and the
budget constraint yields reduced-form demand functions
for X, Z, H and I.
Data
 Main source: Burkina Faso 2007 Enquete sur les conditions
de vie des menages (survey of living standards).
 Additional data sources include the ministry of primary
education (policy change data: number, location, and year
of construction of the schools, and enrollment
information) and Internet sources (distance calculator).
 Distance between the closest school (village/city name)
and the household village/city of residence computed
using distance calculator at
www.infoplease.com/atlas/calculate-distance.html
Descriptive Statistics: Child Characteristics
Variable
Child’s age in months
Male
Obs
Mean
2,003
27.3
2,007 52.7%
Std.
Dev.
16.8
0.50
Min
0
0
Max
59
1
HAZ
Stunting
Moderate stunting
Severe stunting
2,003
2,003
2,003
2,003
-1.8
47.6%
19.4%
28.3%
2.05
0.50
0.40
0.45
-5.98
0
0
0
5.8
1
1
1
WHZ
Wasting
Moderate wasting
Severe wasting
1,932
1,932
1,932
1,932
-0.3
20.7%
9.5%
11.2%
2.04
0.40
0.29
0.32
-4.96
0
0
0
5
1
1
1
Descriptive Statistics: Regressors
Variable
Zero to 5km of program school
Five to 10km of program school
Number of years ES was available
Mother’s years of education
Mother’s age
Father’s years of education
Father’s age
Monthly expenditures (1,000 FCFA)
Urban residence
Television
Radio
Refrigerator
Bicycle
Moped
Car
Electricity
Cell phone
Mother’s health knowledge
Mother’s bargaining power
Obs
2,007
2,007
2,007
2,007
2,007
2,007
2,004
2,005
2,007
2,007
2,007
2,007
2,007
2,007
2,007
2,007
2,007
2,007
2,007
Mean Std. Dev.
6.9%
0.25
10.1%
0.30
0.4
1.05
1.4
3.12
24.3
3.29
2.1
4.15
35.7
11.44
8.2
12.44
24.5%
0.43
20.1%
0.40
74.9%
0.43
5.4%
0.23
86.8%
0.34
38.5%
0.49
2.4%
0.15
15.8%
0.37
25.0%
0.43
3.7
1.80
65.4%
0.48
Min Max
0
1
0
1
0
5
0
17
14
29
0
17
15
97
0 225
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
8
0
1
.8
.6
0
.2
.4
Fraction
.4
.2
0
Fraction
.6
.8
Descriptive Statistics: Histogram of
parents education
0
5
10
momedu_years
15
20
0
5
10
dadedu_years
15
20
Selected child health outcomes by parents’ education
Stunting
Wasting
50.7%
21.5%
Some
schooling
Father’s
schooling
No schooling
36.1%
17.5%
51.2%
21.2%
Some
schooling
Total
38.4%
19.2%
47.6%
20.7%
Mother
schooling
No schooling
Methods and Procedures
 Parents’ education and child health maybe
affected by many ‘unobservables’. To identify the
effect parents education on child health, I will use
a system of equations:
S i     k  Z  X   i
H i    k   S i  X   i
Methods and procedures
 Use two-stage –least –squares to estimate the
system (education equation and child health
equation) with the primary education program as
an instrument for mother’s education.
 Add income, mother health knowledge, mother’s
bargaining power, etc. to the child health equation
and estimate it.
Table 1: OLS Estimates of Mother’s Health Knowledge, Bargaining Power, and Household
Expenditures on Mother’s Education and other Variables.
VARIABLES
Mother’s education (log)
OLS
Per capita
expenditures
OLS
Health
knowledge
Bargain power
0.300***
(0.031)
-0.001
(0.007)
-0.125***
(0.021)
-0.007***
(0.002)
0.370***
(0.052)
0.541***
(0.064)
0.084*
(0.049)
0.041***
(0.013)
-0.158***
(0.043)
-0.001
(0.004)
0.499***
(0.117)
0.225*
(0.128)
0.416***
(0.094)
-0.023
(0.018)
-0.022
(0.014)
0.018***
(0.004)
-0.028**
(0.013)
-0.000
(0.001)
-0.137***
(0.037)
-0.064
(0.041)
-0.016
(0.026)
1,934
0.164
1,934
0.173
1,934
0.056
Per capita expenditures (log)
Mother’s age
Father’s education (log)
Father’s age
Urban residence
Has a television
Has a radio
Observations
R-squared
OLS
Table 2: Reduced Form Estimates for Child Height-for-Age (HAZ): Pathways Added
VARIABLES
Mother’s education (log)
Child’s age (months)
Father’s education
Base FE
HAZ
Income
HAZ
Health
Knowledge
HAZ
0.133**
(0.064)
-0.204***
(0.035)
-0.008
(0.053)
0.083
(0.061)
-0.203***
(0.035)
-0.005
(0.054)
0.123*
(0.064)
-0.204***
(0.035)
-0.008
(0.053)
Per capita expenditures
(log)
Bargaining
power
HAZ
0.139**
(0.065)
-0.204***
(0.035)
-0.006
(0.053)
0.143
(0.122)
Within 30 minutes of a
health clinic
0.079
(0.062)
-0.203***
(0.035)
-0.004
(0.054)
0.170***
(0.061)
0.012
(0.027)
0.152
(0.119)
0.017
(0.027)
Bargaining power
Number of provinces
0.130*
(0.065)
-0.203***
(0.035)
-0.008
(0.054)
0.169***
(0.060)
Health knowledge
Observations
R-squared
Community
services All pathways
HAZ
HAZ
0.044
(0.107)
0.035
(0.105)
2,000
0.101
1,998
0.104
2,000
0.101
2,000
0.102
2,000
0.101
1,998
0.105
43
43
43
43
43
43
Table 3: Reduced Form Estimates for Child Weight-for-Height (WHZ): Pathways Added
VARIABLES
Mother’s education (log)
Age in months
Father’s education
Base FE
WHZ
Income
WHZ
0.164*
(0.086)
0.011***
(0.003)
-0.140**
(0.058)
0.157*
(0.087)
0.011***
(0.004)
-0.142**
(0.059)
0.008
(0.072)
Per capita expenditures (log)
Health Knowledge
Health
Knowledge
WHZ
Bargaining
power
WHZ
Community
services
WHZ
All
pathways
WHZ
0.165*
(0.087)
0.011***
(0.003)
-0.140**
(0.057)
0.156*
(0.086)
0.012***
(0.003)
-0.143**
(0.059)
0.166*
(0.082)
0.011***
(0.003)
-0.140**
(0.058)
0.154*
(0.087)
0.012***
(0.004)
-0.145**
(0.059)
0.004
(0.071)
-0.003
(0.037)
-0.218**
(0.096)
-0.021
(0.111)
-0.029
(0.117)
-0.002
(0.038)
Bargaining power
-0.221**
(0.097)
Within 30 minutes of a health
clinic
Observations
R-squared
Number of provinces
1,925
0.020
1,923
0.020
1,925
0.020
1,925
0.022
1,925
0.020
1,923
0.022
43
43
43
43
43
43
Table 4: First Stage IV Regression for Child Weight-for-Height (WHZ)
VARIABLES
Zero to 5km of ES (d1)
Five to 10km of ES (d2)
Number of years ES was available (ESyears)
Interaction d1*ESyears
Interaction d2*ESyears
Mother’s
education
-0.041
(0.066)
0.157**
(0.076)
Mother’s
education
-0.008
(0.049)
0.163**
(0.071)
Per capita
expenditures
0.068
(0.092)
0.010
(0.079)
-0.016
(0.020)
0.349***
(0.107)
-0.049
(0.045)
-0.003
(0.019)
0.407***
(0.124)
-0.020
(0.042)
0.072***
(0.010)
-0.023
(0.030)
0.031
(0.035)
0.368***
(0.027)
1923
0.299
0.007
(0.027)
-0.168**
(0.066)
0.048
(0.045)
0.028**
(0.013)
-0.035
(0.044)
-0.007
(0.041)
0.320***
(0.026)
1923
0.161
Health knowledge
Bargaining power
Within 30 minutes of a health clinic
Wealth index
Observations
R-squared
1925
0.117
Table 5: Second Stage IV Regressions for Child Weight-for-Height (WHZ)
VARIABLES
Mother’s education (log)
Age in months
Father’s education
Basic model
WHZ
Full model without income
WHZ
Full model with income
WHZ
1.735***
(0.579)
0.016***
(0.004)
-0.163**
(0.071)
1.736***
(0.551)
0.016***
(0.004)
-0.157**
(0.071)
-0.162***
(0.062)
-0.163
(0.129)
1.287***
(0.326)
0.012***
(0.004)
-0.177**
(0.070)
-0.065
(0.040)
-0.287**
(0.126)
-0.145
(0.149)
-0.033
(0.130)
Health knowledge
Bargaining power
Within 30 minutes of a
health clinic
Per capita expenditures
(log)
Observations
R-squared
Number of provinces
-1.113***
(0.394)
1,925
-0.244
43
1,925
-0.224
43
1,923
-0.227
43
Threshold effects
WHZ threshold regression coefficients
2
0
-2
-2
-1
-1
0
1
Parameter estimate
1
2
3
HAZ threshold regression coefficients
0
_Imomedu_ye_5
_Imomedu_ye_10
Mother's years of education
log_pce
0
_Imomedu_ye_5
_Imomedu_ye_10
Mother's years of education
log_pce
Summary of results
• Strong positive coefficient of mother’s
education on WHZ (IV/FE estimation)
• Positive but insignificant coefficient of
mother’s education for HAZ (OLS
estimation).
• Household per capita expenditures are a
pathway for impact on HAZ.
Counterintuitive sign of father’s education,
bargaining power and household
expenditures in WHZ regression.
Summary of results
 Evidence of existence of threshold effects but need to
consider the cost.
 Keep girls in school as long as possible; design
nutrition programs targeted at girls/women.
 Limitations: could not IV mother’s health knowledge,
quality of bargaining power and health knowledge
variables, no information on religion and ethnicity.
Questions?
Thank you!
Table A1: First Stage IV Regressions for Child Height-for-Age (HAZ)
VARIABLES
Zero to 5km of ES (d1)
Five to 10km of ES (d2)
Number of years ES was available
Interaction d1*ESyears
Interaction d2*ESyears
Mother’s
education
-0.045
(0.063)
0.155**
(0.074)
Mother’s
education
-0.001
(0.059)
0.169**
(0.074)
Mother’s
education
-0.006
(0.047)
0.156**
(0.069)
Per capita
expenditures
0.025
(0.091)
0.023
(0.076)
-0.013
(0.020)
0.326***
(0.106)
-0.051
(0.044)
-0.007
(0.019)
0.329***
(0.112)
-0.042
(0.044)
0.095***
(0.011)
-0.084**
(0.033)
-0.001
(0.019)
0.380***
(0.123)
-0.019
(0.042)
0.070***
(0.010)
-0.026
(0.030)
0.001
(0.027)
-0.165**
(0.067)
0.049
(0.045)
0.030**
(0.013)
-0.029
(0.044)
0.108***
(0.037)
0.030
(0.034)
0.362***
(0.027)
1,998
0.296
32.45
0.000
-0.004
(0.040)
0.314***
(0.026)
1,998
0.157
27.03
0.000
Health knowledge
Bargaining power
Within 30 minutes of a health clinic
Wealth index
Observations
R-squared
F (test of excluded instruments)
P-value of F-test
2,000
0.116
3.11
0.008
2,000
0.177
2.91
0.013
Table A2: Second Stage IV Regressions for Child Height-for-Age (HAZ)
VARIABLES
Mother’s education
Child’s age (months)
Father’s education
Urban residence
Basic model
HAZ
Full model without income
HAZ
Full model with income
HAZ
-0.637
(0.514)
-0.206***
(0.029)
0.028
(0.061)
0.726*
(0.387)
-0.739
(0.521)
-0.207***
(0.029)
0.024
(0.061)
0.713**
(0.331)
0.093
(0.058)
0.130
(0.114)
-0.355
(0.301)
-0.199***
(0.028)
0.034
(0.061)
0.030
(0.204)
0.020
(0.034)
0.217**
(0.106)
0.118
(0.128)
0.030
(0.111)
Health knowledge
Bargaining power
Within 30 minutes of a
health clinic
Per capita expenditures
Observations
R-squared
0.713*
(0.364)
2,000
0.047
2,000
0.038
1,998
0.054
Download