Solidaridad: a story of co-responsibilities in the Dominican Republic. Ludovic SUBRAN Social Protection, Latin America and the Caribbean

Transcription

1 Solidaridad: a story of co-responsibilities in the Dominican Republic Ludovic SUBRAN Social Protection, Latin America and the Caribbean Poverty and Social Impact Analysis Workshop May 23, 2011

2 Genesis of the PASS Series Strong political will to seize the opportunity in the 3F crisis to reform social assistance in the Dominican Republic Programmatic DPL series to support incremental policy and institutional change on the Performance and Accountability of Social Sectors (PASS) Joint vision to protect and promote the human capital of the poorest citizens, while improving the quality and transparency of public expenditures in social sectors

3 Objectives of the PASS Series I. Enhanced Performance of Social Sectors to Promote Human Capital for the Poor II. Improved Budget Management to Support the Joint Performance of Social Sectors within the CCT Solidaridad III. Gradual Introduction of Performance Informed Budgeting in Social Sectors IV. Enhanced Transparency and Civil Society Oversight of Social Sectors

4 Partnership(s) PSIA or (PSIA) 2? Joint World Bank-Government working team (including the Social Cabinet and Ministry of Economy and Planning) Audience includes: line Ministries, Finance and CCT Intersectoral Committee (including a High-level Seminar in January 2011) Capacity- and team-building on the analytical toolbox (September 2010) Selectivity Question 1: What is the expected human capital impact of the CCT? (Basis: Overhaul of social assistance - Conditioning) Question 2: How progressive is social spending? (Basis: Closing Supply-Gaps Program Programa Cierre de Brechas)

5 PSIA or (PSIA) 2? (2) Iteration(s) and aligned with upstream policy changes linked to the PASS series Phase 1: Initial stakeholder analysis (Sept. 2009) Phase 2: First round of analysis (March - Jan. 2011) Next steps: Evaluation(s), qualitative assessments, second round of analysis (crisis response e.g.) Advice Part of a programmatic Non-Lending Technical Assistance Platform for the analytical underpinnings of policy-making

6

7 What is the expected impact of the CCT Solidaridad on the human capital accumulation of the poor? Ex-ante simulations of the CCT in education (and health)

8 Context and Methodology Context Impact Evaluation in the pipeline (baseline expected early 2011) Difficult justification of the additional requirements of the program Methodology (for education) Bourguignon, Ferreira and Leite (2003) No need for panel (Attanasio, Meghir y Santiago, 2005; Todd and Wolpin, 2005), but standard household data Behavioral model to estimate change in enrollment (and child labor) through a reduced-form demand model for education (with exogenous prices) Confidence estimation of ex-post impacts Robust Comparisons exist for Brazil, Mexico, Nicaragua - Children occupational choices (ex-post and ex-ante ITT ) and poverty indicators Sensitivity analysis (for parametric/design adjustments)

9 Emerging Results Effectiveness Enrollment increase of 6 pp for the poor (71.0% to 76.9%) 9.2% of working children would enroll in schools Cost-efficiency Expected reduction in poverty levels of 8-10 pp Low-cost: 1% of GDP FGT(0) Change (%) FGT (1) Change (%) FGT (2) Change (%) Gini Change (%) Cost As Share of GDP (%) Actual Scenario 1 (Comer+Bonogas +CCT) Target: P1, P2, P3, zona=2, CCT Scenario 2 (Comer+Bonogas +UCT) Target: P1, P2, P3, zona=2, UCT Scenario 3 (ILAE CCT alone) Target: P1, P2, zona=2, CCT Scenario 4 (ILAE UCT alone) Target: P1, P2, zona=2, UCT

10 Emerging Findings in Health Despite more limited information, analysis of lifting financial barriers to health center attendance finds: Under-5 attendance would increase by 9.1 pp (from 38.9% to 48%) and doubles for the rural poor (32.6% to 66%) Primary attention attendance would increase by 16.5 pp (38.9% to 55.4%), and triples for the rural poor (32.3 to 93.3%) Selected pre-natal and immunization determinants (marginal PROBIT regression) show the importance of health service quality beyond lifting financial barriers: Urban residence (for a first child) decreases the probability of being attended by a general practitioner at pre-natal visits by 5 pp Each additional year of mothers education: (i) Decreases the probability primary health care level by 1.1; (ii) Increases immunization rates: BCG by 2.9 pp; Polio 5 pp; DPT 4.4 pp; and Measles by 2.6 pp

11 Selected Recommendations Need for greater collaboration between the ENIGH 2007 team and SIUBEN (Unique Beneficiary Registry Agency) to capture adequate information on the social assistance transformation For results to occur, both hypotheses need to hold: appropriate supply-side interventions and effective operational compliance of co-responsibilities More scenarios and sensitivity analyses can be run (if generosity of benefits change, for instance) and used for the design of impact evaluations and policy recommendations Renewed emphasis on collecting good household data in health

12 How efficient is social spending in the Dominican Republic? Benefit-Incidence Analysis of Education and Health Spending with and without the Closing Supply-Gaps Program

13 Guiding Questions Context How well targeted is social spending? How much will supply-side interventions improve its accuracy? How can it be more progressive? Spending as Share of GDP (%) Real Growth of Spending (%) 40% 2.2% 1.8% 1.4% Educación Salud 35% 30% 25% 20% 15% 10% 5% PIB Educación Salud 1.0% % -5%

14 Cumulative Incidence (%) Emerging Results in Education Spending Education spending is pro-poor for initial and basic education; neutral for middle school education The poorest 40% receive 58% of education spending by level: Higher concentration of children ages 3-17 in the poorest quintiles; Higher rate of enrollment of poor children in public schools Adjusting for demographics, the accuracy of targeting is reduced and spending tends to be neutral Incidence of Education Spending Inicial Básica Media Agregado Cumulative Population (%)

15 Geographic Mapping of Municipal Education Spending Concentrations

16 Emerging Results in Health Health spending is more significantly pro-poor, with a concentration index of Service usage differentiated by quintile is not available High concentration of health spending in the Ozama region 61% of all consult and hospitalization spending, relative to: 32% of the population; 22% of the poor Deviation of Health Spending Incidence by Region El Valle Enriquillo C. Noroeste Yuma Higuamo C. Nordeste C. Sur Valdesia C. Norte Total Hospitalización Consultas Ozama Percentage Points

17 Closing Supply-Gaps Program The Program closes supply-gaps in education, health, and nutrition that limit a more effective impact of the CCT Solidaridad on the human capital of the poor Supply-Gaps in Education, Health, and Nutrition (US$) Education 283,352,771 88% Health 37,701,289 12% Nutrition % TOTAL 322,160,491 This budgeting was included in the medium-term expenditure framework

18 Program Targeting Criteria Used in Education The Closing Supply-Gaps Program used targeting criteria that extended beyond simple poverty mapping to include: Solidaridad beneficiary density; Middle school education enrollment gap; Relative poverty level; Number of poor; Marginal Education Spending Incidence by Targeting Criteria under the Program Without the Cierre Program With the Cierre Program under Different Criteria Solidaridad Middle School Gap Poverty Level Number of Poor Initial Basic Middle

19 Selected Recommendations The supply gaps measurement tools should be applied to other mechanisms for the prioritization of social spending, reapplied and recalculated. Given the current dialogue of expanding education spending to 4% of GDP, the toolkit should be applied to assess the best targeting criteria for spending as well as the incidence of alternative proposals The definition of targeting criteria and investments should be updated Additional targeting criteria should be assessed, including private service supply, demographic factors, and quality indicators An ex-post evaluation should analyze: (i) The assignment and execution of the Program funds by the Steering Commitee (ii)its impact on human capital through Solidaridad, including an operational process evaluation (iii)a Marginal Benefits Incidence Analysis should be conducted to confirm the expected results in education The lessons learned from this process should be shared within the Government to increase awareness of the impact of investing in more and better information for measuring program impacts

20 Thank you!