Sustainability of Impacts of Cash Transfers, Food Transfers, and Behavior Change Communication in Bangladesh: Transfer Modality Research Initiative

The Transfer Modality Research Initiative (TMRI), was assigned following a cluster-randomized controlled trial design in two zones of Bangladesh (north and south). WFP-Bangladesh implemented the intervention, and the International Food Policy Research Institute (IFPRI) conducted the impact evaluation research. Intervention arms were assigned at the village level, stratified by zone, where arms were as follows: (1) cash transfers [north and south]; (2) cash transfers + nutrition behavior communication change (BCC) [north only]; (3) food transfers [north and south]; (4) food transfers + nutrition BCC [south only]; (5) food-cash split [north and south]; and (6) control [north and south]. Within treatment villages, women living in very poor households were targeted to receive benefits for two years, from May 2012-April 2014.

2012-05-15

2014-04-15

1.Household food security
2.Child nutritional status
3.Intimate partner violence

1.Household food security:
Number of food groups consumed in previous 7 days; Number of unique foods consumed in previous 7 days; Number of food groups consumed by household in previous 7 days, weighted by nutritional value and frequency of consumption; Value of food consumption in previous seven days; Household caloric availability in previous seven days; Household Food Insecurity Access Scale score

2.Child nutritional status:
Height for age z scores; Weight for height z scores; Weight for age z scores; Stunting; Wasting; Underweight

3.Intimate partner violence:
Prevalence of any emotional intimate partner violence in previous 12 months; prevalence of any physical intimate partner violence in previous 12 months; prevalence of any emotional or physical intimate partner violence in previous 12 months; prevalence of individual acts of intimate partner violence in previous 12 months; prevalence of controlling behaviors in previous 12 months

1. Household consumption
2. Household income
3. Household asset accumulation
4. Maternal nutrition knowledge
5. Maternal infant and young child feeding practices
6. Child development
7. Child health
8. Parental interaction with children
9. Gender dynamics
10. Psychological well-being
11. Community norms

1. Household consumption:
Value of food consumption in previous seven days; Value of monthly non-food consumption; Value of monthly non-food consumption by specified sub-groups (clothing, medical expenses, education expenses, transport, consumables, semi-durables); Value of monthly household consumption (food and non-food)

2. Household income:
Value of annual labor income; value of annual income derived from all sources

3. Household asset accumulation:
Index of consumer durables; value of consumer durables; index of productive assets; value of productive assets; index of quality of housing stock; area of productive land ownership; index of total asset ownership; value of total asset ownership

4. Maternal nutrition knowledge:
Maternal knowledge of infant and young child feeding (IYCF) practices; knowledge, attitudes, practice of IYCF scored as yes/no and as a total score on all questions

5. Maternal infant and young child feeding practices:
Maternal adherence to WHO-recommended infant and young child feeding practices (IYCF)

6. Child development:
Raven's Progressive Matrices set A and set B scores (component and total), Early Grade Reading Assessment scores (component and total), Early Grade Mathematics Assessment scores (component and total), Strengths and Difficulties Questionnaire score, school enrollment, school attendance

7. Child health:
Mother's report of child illness in previous 2 weeks

8. Parental interaction with children:
Time spent on child stimulation activities in previous 7 days; discipline of child in previous month

9. Gender dynamics:
Men's and women's decisionmaking; men's and women's individual asset ownership; men's and women's mobility; men's and women's labor force participation; men's and women's earnings; men's and women's respect for spouse; men's and women's gender norms; men's and women's attitudes toward intimate partner violence; men's and women's social capital

10. Psychological well-being:
Men's and women's Perceived Stress Scale score; men's and women's Patient Health Questionnaire-9 Depression Test score; men's and women's aspirations on assets, social status, and children's education

11. Community norms:
Community leaders' gender norms; community leaders' perceptions of attitudes toward intimate partner violence violence; community members' perceived social control of intimate partner violence

This study assesses the sustainability of impacts, 4 years post-program, from a pilot safety net program that was implemented from May 2012-April 2014. The intervention, called the Transfer Modality Research Initiative (TMRI), was assigned following a cluster-randomized controlled trial design in two zones of Bangladesh (north and south). Intervention arms were assigned at the village level, stratified by zone, where arms were as follows: (1) cash transfers [north and south]; (2) cash transfers + nutrition behavior communication change (BCC) [north only]; (3) food transfers [north and south]; (4) food transfers + nutrition BCC [south only]; (4) food-cash split [north and south]; and (5) control [north and south]. Within treatment villages, women living in very poor households were targeted to receive benefits for two years.

The objective of the study is to assess, 4 years after a pilot safety net intervention ended in April 2014, the sustainability of the intervention's impacts on households, children, and women. The intervention, called the Transfer Modality Research Initiative (TMRI), was assigned following a cluster-randomized controlled trial design in two zones of Bangladesh (north and south). Intervention arms were assigned at the village level, stratified by zone, where arms were as follows: (1) cash transfers [north and south]; (2) cash transfers + nutrition behavior communication change (BCC) [north only]; (3) food transfers [north and south]; (4) food transfers + nutrition BCC [south only]; (5) food-cash split [north and south]; and (5) control [north and south]. Within treatment villages, women living in very poor households were targeted to receive benefits for two years, from May 2012-April 2014. For the original impact evaluation, longitudinal data on 5000 households were collected from 2012-2015.

For the sustainability study in 2018, given available funding, the sampling frame will include 4000 households. This will include all households that were originally in the following arms: (1) cash transfers [north and south]; (2) cash transfers + nutrition BCC [north only]; (3) food transfers [north and south]; (4) food transfers + nutrition BCC [south only]; (5) control [north and south]. Subject to attrition being uncorrelated with the intervention arms, the resulting sample will allow unbiased estimates of the impacts of various treatments as of 2018, four years after completion of the program.

Randomization done in office by a computer

Village

Yes

500 villages included in original TMRI impact evaluation. Given available funding, planned resurvey in 400 villages.

5000 households surveyed in baseline round of original TMRI impact evaluation. Given available funding, planned resurvey of 4000 households.

Original TMRI impact evaluation's baseline round included 5000 households drawn from 500 villages (10 households per village), as follows:
- Cash transfers: 1000 households (50 villages in north, 50 villages in south)
- Food transfers: 1000 households (50 villages in north, 50 villages in south)
- Food/cash split transfers: 1000 households (50 villages in north, 50 villages in south)
- Cash + nutrition BCC: 500 households (50 villages in north)
- Food + nutrition BCC: 500 households (50 villages in south)
- Control: 1000 households (50 villages in north, 50 villages in south)

Given available funds, the sampling frame for the sustainability study in 2018 will include 4000 households drawn from 400 villages (10 households per village), as follows:
- Cash transfers: 1000 households (50 villages in north, 50 villages in south)
- Food transfers: 1000 households (50 villages in north, 50 villages in south)
- Cash + nutrition BCC: 500 households (50 villages in north)
- Food + nutrition BCC: 500 households (50 villages in south)
- Control: 1000 households (50 villages in north, 50 villages in south)