1. November 6, 2003Social Policy Monitoring Network1 Evaluation of the pilot phase of the Social Safety Net (RPS) * in Nicaragua: Health and Nutrition Impacts John Maluccio 1 and Rafael Flores 2 1 International Food Policy Research Institute (IFPRI) and 2 Emory University * Red de Protección Social

2. November 6, 2003Social Policy Monitoring Network2 Presentation  Project description  Evaluation design and sample evolution  Impacts on expenditures on food  Impacts on service utilization and coverage of preventive health interventions  Impacts on nutrition

3. November 6, 2003Social Policy Monitoring Network3 Red de Protección Social (RPS) Social Safety Net IADB (US$ 9 million) GN (US$ 1 million) Pilot Phase

4. November 6, 2003Social Policy Monitoring Network4 The Social Safety Net pilot phase  Outputs  Increase the demand for preventive health care for children < 3 years of age  Decrease school desertion during the first four years of primary school  Increase expenditures in food at the household level

5. November 6, 2003Social Policy Monitoring Network5 Outcome and impact indicators IndicatorChange % of children under age 3 who participate in the growth and development monitoring program + 10% % of children age 12-23 months of age with complete vaccination + 10% % of households that have increased spending on food, as a fraction of the total household expenditures See trend

6. November 6, 2003Social Policy Monitoring Network6 Demand and Supply-side Interventions: Food security, Health & Nutrition Demand  Bimonthly cash transfer (food transfer) to all eligible households  US$ 224 per household per year  Conditional upon attendance at bimonthly educational workshops and on bringing their children < 5 for healthcare appointments Supply  Child growth monitoring  Monthly: 0-2 year olds  Bimonthly: 2-5 year olds  Vaccination  Provision of anti- parasites, vitamins and iron supplements  Bimonthly health and nutrition education workshops

7. November 6, 2003Social Policy Monitoring Network7 Demand and Supply-side Interventions: Education Demand  Bimonthly cash transfer (school attendance transfer) to all eligible households  US$ 112  School supplies transfer US$ 21 (per child/year)  Conditional upon enrollment and regular school attendance of children who have not completed 4 th grade (ages 7-13) Supply  Teacher transfer US$ 5 per child/year

8. November 6, 2003Social Policy Monitoring Network8 Geographic Targeting  Stage 1: rural areas in all 17 departments were eligible. Madriz and Matagalpa were selected due to their needs and capacity to implement the project.  Stage 2: six out of 20 municipalities were selected because they had a participatory development program. 36-61% of the rural population extremely poor and 78-90 % extremely poor or poor.  Stage 3: 42 census “comarcas” out of the 59 were chosen based on a marginality index (highest scores). 50 % of the population extremely poor and 85% extremely poor or poor.

9. November 6, 2003Social Policy Monitoring Network9 Base Line Aug-Sep 2000 Measurement Oct 2001 Measurement Oct 2002 RPSControlRPSControlRPSControl 21 a : 811 b 21:77121:76621:72421:72221:712 Evaluation design: cluster- randomized trial a Census “comarcas”: administrative areas within municipalities that include between one and five small communities averaging 100 households each b Households within 21 comarcas, planned sample size = 882 (42*21)

10. November 6, 2003Social Policy Monitoring Network10 Survey non-response and attrition Households200020012002 Completed Interview 158114901434 Intervention706 (80%)706 (87%) Control690 (78%)690 (90%) Not Interviewed 18391147 Target17641581

11. November 6, 2003Social Policy Monitoring Network11 RPS average impact on per capita annual food expenditures (C$) MeasurementInterventionControlDifference 200230272246781 ** 200131292141988 ** 200028122598214 Diff 2001-2000316-457 ** 774 ** Diff 2002-2000214-352 ** 566 ** ** P<0.01

12. November 6, 2003Social Policy Monitoring Network12 RPS average impact on the composition of food expenditures Food Category 2000-2002 expenditures C$ 2000-2002 food share % Cereals504-3.2 ** Beans-138-3.1 ** Meat656 ** 1.9 ** Milk1220.4 Fats and oils663 ** 1.9 ** Fruits and vegetables 494 ** 2.3 **

13. November 6, 2003Social Policy Monitoring Network13 Conclusions  Additional expenditures as a result of the transfers were spent predominantly on food  RPS prevented the deterioration of the food security situation in the intervention group offsetting the decline seen in the control group  The expenditure on almost every food group increased with the program  Three food groups associated with a better diet increased in absolute expenditures and as a percentage of total food expenditures

14. November 6, 2003Social Policy Monitoring Network14 RPS average impact on percent of children age 0-3 taken to health control in past six months MeasurementInterventionControlDifference 200292.784.97.9 * 200195.579.016.5 ** 200069.872.9-3.1 Diff 2001-200025.7 ** 6.219.5 ** Diff 2002-200023.0 ** 12.0 * 11.0 ** P<0.01 * P<0.05

15. November 6, 2003Social Policy Monitoring Network15 Impact on coverage of preventive health interventions Indicator DD 2001- 2000 DD 2002- 2000 Children age 0-3 weighed in past six months % (mother report) 28.9 ** 17.5 * Children age 12-23 months with updated vaccination % (health card) 14.26.1 ** p<0.01 * p <0.05

16. November 6, 2003Social Policy Monitoring Network16 Conclusions  RPS had a large impact on the utilization of well baby check-ups between 2000 and 2001, however this impact decreased in 2002 due to a continued improvement in the control group  Growth monitoring was markedly increased in both 2001 and 2002. However, childhood immunization did not significantly improve in these years

17. November 6, 2003Social Policy Monitoring Network17 Impact on percentage of children < 5 who are stunted * ** Estimated from ENDESA 1998 and 2001 * * HAZ < -2.00 DD= -5.3 D= -3.1 **

18. November 6, 2003Social Policy Monitoring Network18 Impact on percentage of children < 5 who are wasted * * WHZ < -2.00

19. November 6, 2003Social Policy Monitoring Network19 Impact on percentage of children < 5 who are underweight 1 ** ** Estimated from ENDESA 1998 and 2001 * 1 WAZ < -2.00 DD= -6.0 * D= -1.7 ** * P < 0.05

20. November 6, 2003Social Policy Monitoring Network20 Conclusions  Improved household diet and improved health care services (bimonthly health and nutrition education workshops) for children have combined to improve nutritional status of beneficiary children < 5.  Net impact was a 5 percent point decline in the percentage of children who were stunted.  This decline is more than one and a half (1.7) times faster than the improvement see at the national level between 1998 and 2001.

21. November 6, 2003Social Policy Monitoring Network21 Impact on percent of children age 0-3 given iron supplement in past 12 months MeasurementInterventionControlDifference 200275.441.434.0 ** 200169.040.928.1 ** 200039.136.32.8 Diff 2001-200029.9 ** 4.625.2 ** Diff 2002-200036.3 ** 5.131.1 ** ** P<0.01

22. November 6, 2003Social Policy Monitoring Network22 Impact on percentage of children 6-59 months of age with anemia * * Hemoglobin < 11.0 g/dL ** ** ENM, 2000

23. November 6, 2003Social Policy Monitoring Network23 Impact on average Hemoglobin (g/dL) for children 6-59 months of age MeasurementInterventionControlDifference 200211.211.4-0.2 200011.411.5-0.1 Diff 2002-2000-0.2-0.1

24. November 6, 2003Social Policy Monitoring Network24 Conclusions  RPS has been successful in distributing iron supplements to mothers of children 6-59 months of age during its two years of operation  One out of every three children shows anemia in the area where RPS is operating  Despite improved household diet, improved health care services and improvements in the distribution of iron supplements, RPS had not succeeded in improving the grave situation on anemia nor improving hemoglobin levels

25. November 6, 2003Social Policy Monitoring Network25 Are conditional cash transfer programs capable of improving the nutritional status of pre- school children?

26. November 6, 2003Social Policy Monitoring Network26 Impacts (double differences 2002-2000) IndicatorRPSPRAF a % children age 0-3 taken to health control in past six months / 0-5 last 30 days 11.021.0 *** % children age 0-3 weighed in past six months / 0-5 last 30 days (mother’s report) 17.5 * 21.7 *** % children age 12-23 months with updated vaccination / DTP1/pentavalent (health card) 6.16.9 * *** p<0.001 ** p<0.01 * p <0.05 a Demand

27. November 6, 2003Social Policy Monitoring Network27 Impacts, changes and interventions IndicatorRPSPRAF a % children age 0-5 who are stunted-5.4-0.3 % children age 0-5 who are stunted country level estimation -3.1-2.0 % of total annual household expenditures 134 Supply side interventionsYesNo a Demand