1

Poverty Reduction Through Conditional Cash Transfers (CCTs)

Jehan ArulpragasamCountry Sector Coordinator for Human Development

World Bank Office ManilaJuly 2008

Women enrolling in the Child Support Program in Pakistan

Children in Secondary school in Mexico

2

How does a CCT work?

Cash Transfers • Targeted to poor families• Co-responsibilities• Paid to women

Co-responsibilities for Beneficiaries:• Regular school attendance• Health care:

Vaccines Pre- and post-natal visits Regular check-ups Participate in Health,Nutrition, Population seminars

Interactions:(a) “Income effects” of the transfers: - Immediate poverty relief, redistribution - All relief has some structural effect (demand for education, health)

(b) “Price effect:” Stimulating demand: - Conditionality (co-responsibility) aspects of transfers seeks to foster

behavioral changes - Synergies between simultaneously promoting health and education

Seeking to break poverty trap by providing immediate relief (transfers)

and incentives for investments in health and education

3

Defining Characteristics of CCTs

CCTs - transfer cash*- to poor households chosen through an objective poverty targeting

mechanism- on condition that their children go to school and use preventive

health care

Twin goals:- Immediate income support through cash transfers- Long run poverty reduction through improvements in poor

children’s human capital (health, nutrition, and education)

Not a “dole out” because the poor beneficiaries have to comply with specific, monitorable actions to remain in the program.

* food may work too, though with much higher administrative costs

4

CCT Experience Around the World

5

6

7

Impact Evaluation shows:CCTs reduce poverty

Very well-targeted

Reduce poverty in proportion to coverage and size of transfer

Reduce child labor, but not adult labor

Evidence in Mexico that families invest about 25% of their transfer, with a return that raises their income by 24% over 6 years

Comparison of SA Instruments: Strong Performance of CCTs

0%

10%

20%

30%

40%

50%

Q1 Q2 Q3 Q4 Q5

CCTs

Other Cash

Scholarships

School Feeding

Other Feeding

Source: Lindert, Skoufias, and Shapiro, 2005

8

Impact Evaluation shows:increased school enrollment

Significant effects in all evaluations done

Larger effects for countries and grades with lower initial enrollments

Larger effects for the poor, ethnic minorities, girls, those in rural areas

Mexico

Mexico

Mexico

Nicaragua

Honduras

Ecuador

Colombia

Colombia Brazil

Cambodia

Bangladesh

010

2030

Impa

ct o

f CC

T on

enr

ollm

ent

20 40 60 80 100Enrollment rate at baseline

Source: Schady 2006

9

Impacts on Education

Impact on Dropout Rate

13%

7%9%9%

5%

2%

0%

5%

10%

15%

Mexico Nicaragua Honduras

Control

Treatment

10

Impacts on Education

Impact on Grade Repetition

0

10

20

30

40

Mexico Honduras

Control

Treatment

11

Impact Evaluation shows: higher use of preventive health services

Evidence from four countries (Mexico, Nicaragua, Honduras, Colombia):

1. CCTs increase coverage of some preventive services for children, but not others:– Significant effects on use of growth monitoring services

• Colombia: 23-33 % points • Honduras: 20 % points

– No effects on immunization rates, because already so high

2. Mexico: Increase in use of preventive health care by adults as well as children– Probability that an adult age 50+ has been taken to a preventive visit is 16 to

18% points higher in PROGRESA-Oportunidades communities – Adults more likely to have had check-ups for diabetes and hypertension

3. Mexico: Increase in use of public health facilities may be offset by reduction in use of private health care – Reductions in out-of-pocket expenditures on health– Unclear what implications this has for quality of care received (may vary by

country)

12

• .Nicaragua RPS: Increase in Health Care for Children

0-3 years (in past six months)

0

20

40

60

80

100

120

Health Visit Weighed

Control

Treatment

Honduras PRAF: Increase in Health Visits

0

10

20

30

40

50

60

70

Control Treatment

>5 Prenatal Visits

Child Health Visit

Impact Evaluation shows: increasing number of health visits

13

Impact Evaluation shows: reduction in child labor

Impact on Child Labor (reduction)

22%

11% 12%7%

17%

38%

17%19%

6%11%

5%10% 12%

4%

0%5%

10%15%20%25%30%35%40%

Control

Treatment

Source: Olinto (2004)

14

Impact Evaluation shows:empowerment

Empowerment of women• through paying the transfer to them• by providing venue/need for participation in community

groups, transactions• Increase in documentation – ID cards, formalization of

unions, registration of births• no increase in family violence

Community • Increased organization not explicitly sought, but

sometimes observed, especially among beneficiaries

15

Good record on accountability

Many programs, especially the ‘classics” in Latin America, have shown:

• Monitoring conditionalities – requires strong Management Information System which

safeguards program management more generally– makes it hard to invent ghost beneficiaries

• Robust targeting systems deliver benefits to poor families• Automaticity of payments via banking sector limit possible diversion of

funds• Quality control mechanisms better developed than for many other

programs (spot checks, data base cross-checks, hotlines, etc.)• Good record on transparency• Systemic and robust impact evaluation

These programs have acted as leaders in modernizing social sector management

16

CCT in the Philippines

17

Background: Philippines• Limited progress in reducing poverty

– Slower reduction in poverty than in other countries in the region; one-third of population below national poverty line and close to one-half below $2/day poverty line.

– Poverty rate seems to have increased between 2003 and 2006.

• Education indicators are falling – Primary net enrollment rate has fallen in recent years (from

90.29% in 2002 to 84.4% in 2006).– Primary drop out rate has risen (from 6.9% in 2003 to 7.3% in

2006).

• Troubling health indicators – One of higher maternal mortality rates in the region, esp. among

middle income countries

18

Background: Philippines

Despite huge issue, more could be done to address poverty…..

• Total budget to targeted transfers and subsidies remains small – approx. 0.2% of GDP relative to approx. 2.0% of GDP in some

Latin American Countries.

• Numerous small transfer programs– Largest ones are rice distribution programs– DSWD programs are relatively small (only 3% of national

budget)

• Targeting is extremely poor– Different targeting regimes for different programs (some central

and some local government targeting).– e.g. Only 38% if food-for-school program estimated to go to the

poor. – Targeting susceptible and has long history of political

intervention

19

Motivation

• CCT pilot program offers the opportunity to address:– Chronic poverty and human capital problems to help

break cycle of poverty;– Cushion poorest from shocks (e.g. current food

prices)– Actions detrimental to human capital development of

poor as a result of shocks.

• CCT pilot could also be strategic entry point to help country shift to:– Better targeting systems and better M&E systems– Move from badly targeted commodity subsidies to

better targeted cash transfers.

20

Cheaper delivery than comparable programs

2900

8900

5500

16600

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

Per Child Family

CCT Food for School

21

CCT Design ElementsPhilippines

22

Co-Responsibilities

• Education:– Regular attendance in school among children (6-14 years)

• Health and Nutrition:– All children (0-5 years) to complete entire Dept of Health

protocol

– Full vaccination protocol• Monthly growth monitoring

– All pregnant women to adhere to protocol • Regular prenatal and postnatal check-ups at health clinic• Delivery with skilled health professionals

23

Targeting• Like many CCT programs, combine geographic and household targeting

to determine eligibility

– First, select priority areas:

• Geographic targeting (micro-area poverty maps)

• Conditional on supply side assessment

0601

0304

08

05

09

02

07

CHIAPAS. ZONAS DE OPERACION

0 40 80 Kilometers

S

N

EW

Zonas de OperaciónI CENTRO COPAINALA (MEZCALAPA)I CENTRO TUXTLAII ALTOS TZELTALII ALTOS TZOTZILIII FRONTERIZA COMALAPAIII FRONTERIZA COMITANIV FRAILESCAIX ISTMO COSTASV NORTE BOCHILV NORTE PICHUCALCOVI SELVA OCOSINGOVI SELVA PALENQUEVI SELVA YAJALONVII SIERRAVIII SOCONUSCO FRONTERIZAVIII SOCONUSCO MAPASTEPECVIII SOCONUSCO TAPACHULA

Zonas de Atención

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

###

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

####

#

##

###

##

#

#

##

#

#

#

#

#

##

#

#

#

#

##

#

##

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#######

##

#

#

#

###

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

# #

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

##

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

# #

#

#

#

#

#

#

#

## #

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

##

#

#

# #

#

###

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

##

##

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

# #

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

## #

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

# #

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

##

### ##

###

#

#

#

#

#

##

##

####

#

##

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

# #

#

#

#

##

#

#

#

#

#

#

#

#

#

#

######

#

â

S

N

EW

0 4 8 Kilometers

ZONA II ALTOS TZOTZIL

SEDE

0 1 2 Kilometers

S

N

EW

Distrito Federal

DelegaciónXochimilco

%

%

$

$

$

$

$

$

#

#

#

###

#

#

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

#

##

###

#

#

#

#

#

#

#

#

##

#

$

# #

#

#

##

#

#

##

#

#

#

#

#

#

# #

#

##

#

#

##

#

#

##

#

#

#

#

##

#

# # #

#

#

#

#

#

##

#

#

#

##

# #

##

# #

##

##

#

#

##

#

#

#

#

##

# #

#

#

#

#

#

#

#

##

#

#

#

##

#

#

#

#

#

#

#

#

#

##

#

#

#

#

#

##

#

#

##

%

%

#

#

#

#

#

#

#

#

#

#

#

# #

#

%

#

#

#

#

#

#

#

# #

#

#

#

#

#

#

#

#

#

##

#

#

##

#

#

#

#

##

##

#

#

#

#

#

#

#

#

#

#

#

# ##

#

#

#

%

#

#

#

#

#

#

##

#

#

##

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

##

#

#

#

#

#

#

#

#

#

##

#

#

##

#

#

#

#

#

#

#

#

#

##

#

#

#

# #

#

#

#

# ##

#

##

#

##

#

#

#

## #

#

##

#

#

#

#

XOCHIMILCO

TLAHUAC

Municipal

Limite

Colonia

Hogares en pobrezade capacidades

0 - 2

3 - 7

8 - 17

18 - 44

45 - 123

Simbología

Rural

Localidades#

$

% Urbana

Semi urbana

24

– Collect data on household characteristics

– Apply objective household eligibility criteria• Proxy means-testing (used in most countries)

– Eligibility based on a weighted index of characteristics (score) that are easily observed but not manipulated and that are associated with poverty

– Broader multi-dimensional notion of poverty

– Move away from subjective or “politically motivated” targeting

– Fairly impressive results globally

– Enrollment: Verify and Issue ID Cards to beneficiary households

• Beneficiary households enrolled for 5 years and then “graduate”

Targeting

25

Delivery System: Philippines

• Program Management– Department of Social Welfare and Development (DSWD)

• Money– Program determines payments based on achievement of co-

responsibilities– Payments to be made [bi-monthly]. Using payment through

ATM/cash points. • People

– Health centers and schools to verify co-responsibilities– Social workers (probably) to liase with beneficiary families at

village level. (DSWD has regional presence). Will explain program, relay information, record complaints.

• Pilot Areas - Agusan Sur, Misamis Occ. Pasay, Caloocan,

- Pilot Experience/lessons to be used to design a wider program