Transcript Trends in calorie deficiency in India

Agriculture-Nutrition Linkages In India:
Emerging findings from the Tackling Agriculture Nutrition
Disconnect in India (TANDI) initiative
Suneetha Kadiyala,
Research Fellow
Poverty Health and Nutrition Division,
International Food Policy Research Institute, IFPRI
LCIRAH Lunch Seminar, 20th July 2011
Acknowledgements

IFPRI: D. Headey, A. Chiu, P. Menon, J. Harris, S. Gillespie & others

TANDI Core Group:





Mahendra Dev, Indira Gandhi Institute of Development Research (IGIDR)
Harshi Sachdev, Sitaram Bhartia Institute of Science and Research
Sukhadeo Thorat, Indian Institute of Dalit Studies
Rajani Ved, National Health Systems Resource Center
S. Parasuraman, Tata Institute of Social Sciences (TISS)

Consultants: Priya Bhagowalia

Funding: BMGF
Outline

Some key trends in development indicators in India

Trends in undernutrition: Some global and regional comparisons

Conceptual frameworks and pathways


Determinants of child nutrition
Agriculture -nutrition pathways

6 sets of emerging findings

Lingering questions and some points for discussion
Trends in Indian development
5 facts
Trends in Indian development: 5 facts
Sustained & rapid economic
growth trajectory
Steady declines in poverty
rates, but 42% < $1.25.day
Agriculture sector
performance is below the
target growth rates of 4%
Discouraging progress in
human development
1.
2.
3.
4.

5.
HDI (134/182); GGI (114/134)
Glacial progress in nutrition
Growth Rate in agriculture and
overall GDP (1984/85-2007/08)
8
7.3
7
%
g
r
o
w
t
h
6.3
6.1
6
4.9
5
3.8
4
3
3.0
2.9
2.3
2
1
0
1984/85 to
1991/92
1992/93 to
1999/00
GDPA
2000/01 to
2007/08
1984/85 to
2007/08
GDP
Source: Gulati &Shreedhar
Considerable variations in progress by state & social groups
Undernutrition in India
Some global and regional comparisons
2009 Global Hunger Index
2008 India State Hunger Index
India State Hunger Index scores
-Serious in 4 states
-Alarming in 12 states
-Extremely alarming in 1 State
October 14, 2008
India State Hunger Index 2008
Indian states rank poorly even globally
India
GHI
RANK
65
Burkina Faso
66
India (23.7)
67
Zimbabwe
Gujarat
GHI RANK
State (ISHI score)
69
Haiti
Gujarat (24.69)
70
Madhya Pradesh
Punjab
GHI
RANK
State (ISHI score)
33
Nicaragua
Punjab (13.64)
34
Ghana
Global Hunger Index and India State Hunger Index 2008
Bangladesh
GHI RANK
State (ISHI score)
81
Chad
Madhya Pradesh (30.9)
82
Ethiopia
Child Nutrition among under-5s in South Asia
(2005 – 2007)
60
43
Percentage of under-fives
49
48
50
43
41
42
39
40
31
Underweight
30
20
Wasting
Stunting
21
20
17
13
15
17
14
10
0
India
Source: Menon, Bamezai et al
Bangladesh
Nepal
Sri Lanka
Pakistan
Source: UNICEF http://www.childinfo.org/undernutrition_nutritional_status.php
1. National Family Health Survey, India, 2005-06
2. Bangladesh Demographic Health Survey, 2007
3. Nepal Demographic Health Survey, 2006
4. Sri Lanka Demographic Health Survey, 2006-07
5. Pakistan National Nutrition Survey 2001-02, Re-analyzed by WHO Nov 2007
Maternal BMI (weighted) in select South Asian countries
Mean maternal BMI
Percentage of mothers with mean BMI less
than 18.5
21.2
70
20.99
21
60
20.8
50
20.36
20.4
India
Bangladesh
20.2
20
Nepal
Percentage
Mean BMI
20.6
40
37.98
India
Bangladesh
31.88
Nepal
30
25.20
19.99
20
19.8
19.6
19.4
Source: Menon, Bamezai et al
10
0
Source: Author’s estimates based on DHS datasets
1. National Family Health Survey, India, 2005-06
2. Bangladesh Demographic Health Survey, 2007
3. Nepal Demographic Health Survey, 2006
Tackling Undernutrition in India
The causes and consequences of undernutrition
cut across sectors and so do the solutions
7/20/2015
Long term consequences
Adult size, intellectual ability, economic
productivity, reproductive performance,
metabolic and cardiovascular diseases
Short term consequences
Morbidity, disability , mortality
Direct interventions
Infant feeding
Vitamin A, Zinc
Hygiene
Biofortification
Indirect Interventions
Agriculture (pattern,
pace)
Social protection
Education
Health Systems
Women’s empowerment
Economic growth;
poverty reduction;
environment ;
institutions and
governance
Adapted from Black et al, 2008; Ruel 2008
Maternal and child undernutrition
Inadequate
dietary intake
Household
food
security
Underlying
causes
Diseases
maternal &
child care
Immediate
causes
Hygiene;
Access to
health services
Income poverty: employment , dwelling, assets,
remittances, transfers
Lack of capital: Financial, human physical ,
social and natural
Basic causes
Social , economic and political context
Status of 12 essential direct interventions in India
100
The GOAL : 100%
90
80
Early Inititation of
Breastfeeding
Exclusive BF (0-6 Months)
Gap
70
60
Introduction of CF at 6-9
Months
3 Expected IYCF Practices
50
Iron-rich Foods
40
All basic Immunisations
30
Stools Safely disposed
Vitamin A Supplementation
(<3s)
Adolescent Girls (15-19
Years) Non-Anemic*
HH - Adequately Iodised Salt
20
10
0
India
7/20/2015
Interventions to reduce undernutrition

Effective health and nutrition interventions are essential, but not
enough:
 Direct interventions, if scaled up and implemented effectively, will
address only 1/3rd of the burden

Indirect interventions essential to address 2/3rds of the
undernutrition burden
 There is not enough understanding of how to make indirect
interventions “pro-nutrition”
Tackling the Agriculture-Nutrition Disconnect in
India (TANDI)
Some emerging findings
7/20/2015
TANDI’s Approach

Facilitate transdisciplinary dialogue and research on key
knowledge gaps on linkages between agriculture and nutrition in
India
 What are the pathways between agriculture and nutrition?
 Is the potential being realized?
 What can be done to increase the realization of the potential?

Progress to date:




Consensus reached on pathways and analytical frameworks
Systematic literature review
Data set audit undertaken
Empirical analyses underway
Conceptualizing the pathways between agriculture and
nutrition
Agriculture as a source of food
2. Agriculture as a source of income (employs
2/3rd of rural and half of India’s labor force)
3. Agricultural policy and food prices
4. Expenditure patterns: how income derived
from agriculture is actually spent
Gender dimensions (83% of female rural labor
force is in agriculture)
5. Women’s status and intrahousehold decisions
and resource allocation
6. Women’s ability to manage young child care
7. Women’s own nutritional status
1.
Agriculture is a
key driver of
poverty
reduction
But
Pathways to
nutrition are
diverse and
interconnected
Important household level "leakages” along the
agriculture-child nutrition pathways

In B’desh, a 1% increase in income improved the height-for-age scores by
just 0.03%.
Source: Ahmed 1993
Emerging findings 1
Cross-country evidence
7/20/2015
Is the agriculture-nutrition potential being realized? CrossCountry evidence (Headey 2011)

Does agriculture growth (vs. non-ag growth) influence nutrition?

A GDP per cap growth of 5% reduces stunting prevalence around
0.9 percentage points


Ag growth has a significant and negative effect on stunting
Effects much stronger when Indian states are excluded from the sample –
i.e. disconnect between agriculture & nutrition in India?
Agricultural growth is certainly a huge driver of changes in
energy supply
It is hard to disentangle channels – is it poverty reduction, increased
energy availability, or increased dietary diversity?
1.00
0.80
% Change in
0.60
calorie
availability from
1% growth in 0.40
agricultural
0.20
production
0.00
-0.20
Source: Headey 2011
Initial level of calorie consumption (kcal)
The impacts of 10% increase in income on stunting
Agric. effects when Agric. effects when Food effects when
ag GDP
ag pop.
initially <$150 per
share=30%*
share=60%*
capita
Change in stunting prevalence from
10% increase in average income or DES
0.0
-1.0
-2.0
-3.0
-4.0
Source : Headey 2011
Energy supply
effects
Emerging findings 2
The data disconnect
7/20/2015
India specific studies are sparse

Systematic search of 15 databases





Only 71 articles of varying scale, scope, methodology and rigor attempted
to address the issue of ag-nutrition
None measured nutrition status
The main reason for this paucity pertains to data
Early on in TANDI we saw some value added in exploring the
different datasets in the Indian context, seeing where the gaps were
and whether there might be “quick wins” from merging data
Starting point was to think of the characteristics of an ideal dataset
linking agriculture and nutrition
Type 1 datasets: “Nutrition/health” datasets with little
economic information
Survey name
Years
Panel
NFHS I
NFHS II
NFHS III
NNMB
1992-1993 1998-1999 2005-2006 Various yrs
No
No
No
Yes
Sample
National Coverage
Representative at…
Yes
Yes
Yes
No
Yes (<4 yrs)
Yes[ii]
state
district
Anthropometrics
Women
Men
Access to health services
Access to water & sanitation
Feeding & health practices
Gender, caste, ethnicity
Expend/consumption, including food
Agriculture: production, inputs, etc.
Income : farm, nonfarm
Limited[iii]
Yes
Yes
Yes
Very little
No[v]
No[v]
Yes
Yes
Yes
No
Yes (<3 yrs)
Yes
Yes
Yes
Yes
No
Yes (<5 yrs)
Yes
Yes
Yes[vii]
Yes[xi]
Yes
Yes
Yes
Yes
Yes
Yes
Very little Very little
No[ix]
No[xiii]
7/20/2015
No[ix]
No[xiii]
9-10 states
Mostly rural
Yes
Yes[xlvi]
Yes[xlvii]
Yes[xlviii]
Yes
No
No
No
Limited[xlix]
Yes[l]
No
Yes
Type 2 datasets: Agricultural/economic datasets with little
nutrition or health information
Survey name
Years
Panel
Sample
National Coverage
Represent-ative at…
Anthropometrics
NSS
state
dist
<5 yrs
Women
Men
Access to health services
Access to water & sanitation
Feeding & health practices
Gender, caste, ethnicity
Expenditure, consumption, including food
Agriculture: production, inputs, etc.
Income : farm, nonfarm
ICRISAT (VLS) Agric Census
Various yrs
1975-85;
since 1950 1989; 2001-06
No
Yes
All India
Villages
Yes
No
Yes
No
Yes
No
No
No
No
No
No
No
Yes
No
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Limited
Yes
No
Yes
7/20/2015
1995-96;
2000-01
No
All India
Yes
Yes
Yes
No
No
No
No
No
No
Limited
No
Yes
No
Type 3 datasets: “Hopeful”
(e.g. IHDS, ICRSAT VDS, Young Lives Panel)
Survey name
Years
Panel info*
REDS
1998-1999
Yes
DLHS I
1998-1999
No
IHDS-II
2004-05
Yes
Sample
National Coverage
Representative at…
Rural only
Yes
Yes
No
Yes
No
No
No
Yes
No
Yes
Yes
Yes
Yes
National
Yes
Yes
Yes
Yes
No
No
Yes
Yes
Yes
Yes
No
No
No
National
yes
yes
No
yes
yes
yes
yes
Some
yes
yes
Some
Some
yes
Anthropometrics
state
district
Children
Women
Men
Access to health services
Access to water & sanitation
Feeding & health practices
Gender, caste, ethnicity
Expend/consumption, including food
Agriculture: production, inputs, etc.
Income : farm, nonfarm
Emerging findings 3
Trends in agriculture performance and nutrition
outcomes
7/20/2015
Trends in prevalence of underweight children, per capita
food production and expenditures (1970-2008)

22% point decline in
underweight when food
production increased
As food production per
capita slowed from the
mid 1990s, so too did
nutritional improvements
71%
Prevalence of underweight children (%)

80
60
60
49%
44%
40
Green Revolution:
steady growth in
food prod.
Reform period:
slower growth in
food prod.
20

At the national level, there
does not appear to be a
disconnect, but large
interstate variations
0
40
20
% Children with low weight for height
Net food production per capita
urban expenditure per capita
rural expenditure per capita
% Children with low weight for height
0
Food production and mean incomes (2005 PPP)*
80
Agriculture performance and anthropometric outcomes
Weak association between ag. growth and stunting
1992-2005
Improvements in Stunting
Ag. GDP/worker
High /moderate
Low
High
Bihar, Tamil Nadu, Himachal Pradesh
M.P , Gujarat, A.P
Low
U.P, West Bengal, Karnataka, Orissa, Punjab
Rajasthan, Haryana
Strong association between grain production and stunting
1992-2005
Improvements in Stunting
Grain production
High
Low
High
Karnataka, Orissa , Punjab, West Bengal
Gujarat, A.P,
Low
Himachal, Maharashtra, Tamilnadu
Rajasthan, M.P
Source: Headey, Chiu and Kadiyala, 2011
Weak relationship between agriculture performance and
nutrition status outcomes (income linkage)
Estimated elasticities between malnutrition and welfare indicators
Independent variable
Weight
Dependent Variable
Stunting
Low BMI, (Women)
Asset Index
None
0.09
-0.45#
GDP per cap growth
None
0.08
-0.63
Agr. GDP/worker
Ag. initial
share in total
employment
-0.20#
-0.44*
-0.14#
-0.29*
Non-Agr. GDP/worker
*significant differences from zero at the 10% level,
# indicates marginal insignificance at the 10% level
Source: Headey, Chiu and Kadiyala 2011
Emerging findings 4
Trends in dietary patterns
7/20/2015
Trends in average household calorie availability and noncereal calorie shares
Source: Headey, Chiu and Kadiyala 2011
Dietary Diversity : Sources of Calorie Intake in the Rural
Areas (Food Consumption linkage)
Economic status, by deciles
1993-94
Milk and
Vegs & meat,
Deciles Cereals Pulses fruits
etc.
0-10
81.8
3.7
4.1
10.5
10-20
78.7
3.9
4.3
13.1
20-30
76.8
3.9
4.5
14.7
30-40
74.3
4.2
4.6
16.8
40-50
72.5
4.2
4.8
18.5
50-60
70.8
4.2
4.8
20.2
60-70
68.3
4.4
5.0
22.3
70-80
65.3
4.4
5.2
25.0
80-90
62.0
4.7
5.4
27.9
90-100
55.7
5.2
5.8
33.3
All
India
71.2
4.3
4.8
19.7
Source: Dubey and Thorat 2011
2004-05
Cereals
78.9
74.7
72.3
69.8
68.1
64.2
63.9
61.1
57.2
51.5
Pulses
3.5
3.7
3.8
3.9
3.9
3.9
4.1
4.3
4.4
4.7
Vegs &
fruits
4.6
4.9
5.1
5.2
5.2
5.2
5.5
5.6
5.8
6.6
67.7
4.0
5.3
Milk and
meat, etc.
13.1
16.7
18.8
21.1
22.8
26.7
26.5
29.0
32.6
37.2
23.1
Changes in prices (1983-2000) explain changes in
consumption (food price linkage )
Lowest Income group
Highest Income group
200
200
y = -0.55x + 157
R² = 0.53
% change in consumption
Meat, fish,
eggs
100
Vegetables
Edible oils
50
Milk
RIce
0
Sugar
Wheat
Pulses
-50
Coarse cereals
50
100
150
200
250
300
350
400
Percentage change in consumer prices
But what explains changes in prices?
Source: Calculation from NSSO
y = -0.87x + 231
R² = 0.83
Meat, fish,
eggs
100
Edible oils
Vegetables
50
Milk
0
Sugar
Rice Wheat
-50
Pulses
Coarse cereals
-100
-100
0
Fruits
150
150
% change in consumption
Fruits
0
50
100
150
200
250
300
Percentage change in consumer prices
350
400
Rising pulse prices linked to long term neglect of pulse
sector by government policies
Index of per capita consumption
(1970=1)
Rising coarse grain prices linked dietary diversification >>
rising use of coarse grains in meat and diary production
10
9
8
7
6
5
4
3
2
1
0
Source: Calculations from USDA 2011 data
Meat
consumption
Coarse grain as
feed
Dairy
consumption
Coarse grain for
consumption
Emerging findings 5
Implications of income growth for child nutrition
7/20/2015
Does income influence child’s dietary diversity?
Dependent variable
Number of observations
Age bracket
Maternal characteristics
Mother’s education level (1,2,3)a
Mother’s age
Mother’s in professional work
Mother’s in clerical work
Mother’s in sales work
Mother’s in services work
Mother’s in manual work
Mother’s in agricultural work
Wealth effects (base = 1st quintile)
2nd Quintile
3rd Quintile
4th Quintile
Richest quintile
Location effects (rural=base)
Capital city
Small city
Rural town
R squared
Consumed at least 4 food groups (=1)
4419
3995
3428
6-11 months
12-17 months
18-23 months
0.012
0.001
0.059
0.004
0.017
-0.015
0.010
0.005
0.042**
0.000
0.16
0.362**
-0.055
-0.051
0.031
0.025
0.058***
0.003**
0.256**
0.239**
-0.040
0.107
0.014
-0.001
0.017
0.019*
0.037***
0.076***
0.032*
0.061***
0.110***
0.160***
0.052**
0.078***
0.112***
0.168***
0.009
0.007
0.016
0.085
0.01
0.054
0.0267/20/2015
0.126
-0.006
-0.006
0.033
0.123
Does child’ dietary diversity affect nutrition status
outcomes?
0
-0.5
-0.834
-1
-1.157
-1.5
-1.354
Did not meet minimum diet
diversity (FG<4)
-1.496
-1.785
-1.783
-2
Met minimum diet diversity
(FG>=4)
-2.5
HAZ
WAZ
WHZ
P<0.001 for HAZ & WAZ; p=0.10 for WHZ
Source: Aguayo et al, 2011
Emerging findings 6
Maternal employment and time use effects
7/20/2015
Maternal employment in agriculture and child care

Some evidence of mother’s employment status on child survival
(Kishor and Sulabha 1998)

Long hours and physical work of agriculture may lead to
neglect of children
 Also provides possibility for other adults or older children to take care
of infants
Impact of employment in agriculture on maternal and
child nutrition
Stage 1: Occupation choice on mother's BMI
Mother's BMI
-0.08
-0.31
0.33*
-0.35***
-0.54***
-0.62***
Not working
Clerical
Sales
Services
Manual
Agriculture
Professional (reference category)
Stage 2: Mother's BMI on children's nutrition
Mother's BMI
Child's HAZ
Child's WAZ
0.030***
0.060***
*,**,*** indicate significance at the 10%, 5% and 1% levels respectively
Who takes care of the children?
Percentage of each caretaking category
100%
6%
13%
24%
26%
36%
80%
Mother
33%
41%
67%
60%
88%
46%
94%
40%
43%
31%
47%
70%
Other Adult
64%
Other Children
21%
20%
24%
8%
0%
4%
0%
18%
12%
4%
Mother's occupation
29%
24%
27%
Impact of care giving status on stunting
Prevalence of stunting (<2 std.
Deviations)
Children
<36
months
Height-for-age z-score
Childre
Children
n <18
Children
<36
Children Children <24
months <24 months months <18 months
months
Care provided by
mother
-0.06***
-0.04
-0.05*
0.18**
0.11
0.16
Care provided by
other adults
-0.04**
-0.07**
-0.06**
0.13
0.20*
0.17
Care provided by older child (reference category)
R2
0.069
0.049
0.054
0.064
0.039
0.044
Observations
4897
2566
3357
4897
2566
3357
Discussion points
7/20/2015
Some discussion points
Weak agricultural growth, but there is indeed an agriculturenutrition disconnect at state level
Diets have diversified, but not for the poor
Through improved food production and prices, agriculture
influences diet quality
At the household, it is unclear if engagement in agriculture
influences consumption patterns and nutrition status in India







Agriculture may adversely affects mother’s BMI, and childcare, but
observable effects are quite small
Is this an agricultural problem or a rural problem?
Strong wealth effects on dietary diversity of children>>child
anthropometry
Knowledge gaps—some discussion points
Several!!!!
 Why are some states doing better than others (positive
outliers)?
 Who is being left out of the growth process? Why?
 What are the gender implications of the current agriculture
interventions/policies? How do they mediate nutrition
outcomes?
 Implications of child labor
 Convergence of various services
 Political economy to strengthen rural economy (farm and nonfarm)

Comments suggestions appreciated!!!