Document of The World Bank Report No: ICR00006287 IMPLEMENTATION COMPLETION AND RESULTS REPORT ON A LOAN IN THE AMOUNT OF US$1 BILLION TO THE FEDERATIVE REPUBLIC OF BRAZIL FOR THE BRAZIL: INCOME SUPPORT FOR THE POOR AFFECTED BY COVID-19 PROJECT June 2023 Social Protection and Jobs Global Practice Latin America and Caribbean Region CURRENCY EQUIVALENTS (Exchange Rate Effective May 18, 2023) Currency Unit = Brazilian Real (R$) R$ 4.97= US$1 FISCAL YEAR July 1 - June 30 ABBREVIATIONS AND ACRONYMS AB Auxilio Brasil AE Emergency Aid (Auxilio Emergencial) APL Adaptable Program Lending ASA Advisory Services and Analytics BE Extraordinary Benefit (Benefício Extraordinário) BEm Emergency Employment and Income Maintenance Program (Benefício Emergencial de Manutenção do Emprego e da Renda) BF Family Allowance Program (Bolsa Familia) CCT Conditional Cash Transfer CPF Country Partnership Framework EFO Externally Financed Outputs ESF Environmental and Social Framework GDP Gross Domestic Product GoB Government of Brazil GRM Grievance Redress Mechanism ICR Implementation Completion and Results Report IFI International Financial Institution IFR Interim Financial Report ISR Implementation Status and Results Report M&E Monitoring and Evaluation MIS Management Information System MoC Ministry of Citizenship MoE Ministry of Economy PAD Project Appraisal Document PDI Project Development Objective Indicator PDO Project Development Objective PIU Project Implementation Unit PLR Performance and Learning Review PNAD-C Continuous National Household Sample Survey (Pesquisa Nacional por Amostra de Domicílios Contínua) POF Household Budget Survey (Pesquisa de Orçamentos Familiares) POM Project Operations Manual SD Unemployment Insurance Program (Seguro Desemprego) SENARC National Secretariat for Citizen Income (Secretaria Nacional de Renda de Cidadania) TA Technical Assistance Regional Vice President: Carlos Felipe Jaramillo Country Director: Johannes Zutt Regional Director: Luis Benveniste Practice Manager: Pablo Gottret Task Team Leader(s): Matteo Morgandi, Gabriel Lara Ibarra, Liliana Sousa ICR Main Contributor: Theresa Jones TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 4 A. CONTEXT AT APPRAISAL .........................................................................................................4 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION ................................................................7 II. OUTCOME ...................................................................................................................... 8 A. RELEVANCE OF PDOs ..............................................................................................................8 B. ACHIEVEMENT OF PDOs (EFFICACY) ........................................................................................9 C. EFFICIENCY ........................................................................................................................... 12 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 12 E. OTHER OUTCOMES AND IMPACTS ......................................................................................... 13 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 14 A. KEY FACTORS DURING PREPARATION ................................................................................... 14 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 15 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 16 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 16 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 18 C. BANK PERFORMANCE ........................................................................................................... 19 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 20 V. LESSONS AND RECOMMENDATIONS ............................................................................. 21 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 23 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 35 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 38 ANNEX 4. ECONOMIC ANALYSIS ........................................................................................... 39 ANNEX 5. BORROWER COMPLETION REPORT....................................................................... 42 ANNEX 6. SUPPLEMENTARY INFORMATION ON SELECTED SOCIAL PROTECTION PROGRAMS 68 ANNEX 7. SELECT ADVISORY SERVICES AND ANALYTICS ....................................................... 72 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P174197 Brazil: Income Support for the Poor affected by COVID-19 Country Financing Instrument Brazil Investment Project Financing Original EA Category Revised EA Category Organizations Borrower Implementing Agency Federative Republic of Brazil Ministry of Social Development Project Development Objective (PDO) Original PDO The objective of the project is to mitigate the negative economic effects associated with COVID-19 on income and risks of damaging human capital of poor families. Page 1 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 1,000,000,000 1,000,000,000 1,000,000,000 IBRD-91720 Total 1,000,000,000 1,000,000,000 1,000,000,000 Non-World Bank Financing 0 0 0 Total 0 0 0 Total Project Cost 1,000,000,000 1,000,000,000 1,000,000,000 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 29-Oct-2020 02-Jul-2021 24-Aug-2022 31-Dec-2022 31-Dec-2022 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Satisfactory Substantial RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 04-Feb-2021 Moderately Satisfactory Satisfactory 0 02 07-Aug-2021 Moderately Satisfactory Satisfactory 0 03 25-Feb-2022 Moderately Satisfactory Satisfactory 200.00 04 06-Oct-2022 Satisfactory Highly Satisfactory 997.50 Page 2 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) SECTORS AND THEMES Sectors Major Sector/Sector (%) Social Protection 100 Social Protection 100 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Social Development and Protection 100 Social Protection 100 Social Safety Nets 100 Human Development and Gender 100 Disease Control 100 Pandemic Response 100 Environment and Natural Resource Management 25 Climate change 25 Adaptation 25 ADM STAFF Role At Approval At ICR Regional Vice President: Carlos Felipe Jaramillo Carlos Felipe Jaramillo Country Director: Paloma Anos Casero Johannes C.M. Zutt Director: Luis Benveniste Luis Benveniste Practice Manager: Pablo Gottret Pablo Gottret Matteo Morgandi, Liliana Do Matteo Morgandi, Gabriel Lara Task Team Leader(s): Couto Sousa Ibarra ICR Contributing Author: Theresa Jones Page 3 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL 1. Brazil entered the COVID-19 crisis still recovering from the 2014–2016 recession, with slow poverty reduction and economic growth. Since the peak of the recession in 2015–2016, economic recovery had remained weak, with only 1 percent growth of gross domestic product (GDP) over 2017– 2019. The income of the poorest 40 percent remained below the pre-2014 crisis level. In 2018, about 20 percent of the population lived on less than US$5.50 per day, including 9.3 million people on less than US$1.90 per day—the World Bank global reference for extreme poverty. Thus, as Brazil entered COVID- 19, large segments of the population were highly vulnerable. As a result of the dramatic decrease in domestic and international demand caused by the pandemic, Brazil was expected to fall into a deep recession with a drop in GDP of 8 percent projected for 2020 with acute implications for poverty and employment. 2. The coverage of Brazil’s social protection system was notably skewed toward the elderly and families of formal workers, even if reaching about 75 percent of the population before the COVID-19 crisis. Before COVID-19, expenditures on social protection in Brazil were large by international benchmarks (estimated at 15 percent of GDP in 2015), dominated by old age and disability pensions. Unless they qualified for Bolsa Familia (BF), informal workers and their households were largely unprotected. 3. Social protection measures were at the center of the fiscal package to respond to COVID-19 in Brazil. The social protection measures adopted by the Government of Brazil (GoB) had two objectives: (a) to make social distancing possible for the economically affected and (b) to mitigate the negative impacts of the crisis on welfare and human capital. In April 2020, the Government introduced Emergency Aid (Auxílio Emergencial, AE), a generous temporary program for the self-employed and families working in the informal economy who became temporarily poor; for formal workers at risk of losing their income if dismissed, the Government introduced a wage subsidy, Emergency Employment and Income Maintenance Program (Benefício Emergencial de Manutenção do Emprego e da Renda, Bem).1 Among existing programs, the Unemployment Insurance Program (Seguro Desemprego, SD) was financed to meet increasing claims by formal dismissed workers, and BF was used to reach in scale the pre-pandemic poor, which is of interest to this project. The expansion was both ‘vertical’ and ‘horizontal’: BF increased its coverage by incorporating in March 2020 an additional 1.2 million eligible families, who had been on a waitlist before the COVID-19 due to budget constraints of the program.2 The value of the transfer of BF was also temporarily raised to match at least the amount received by families covered by the AE benefit, which was more than three times the size of pre-COVID BF. The benefits evolved during the life of the project are further described in Annex 6. 4. The Ministry of Economy (MoE) of the GoB requested the World Bank and five other international financial institutions (IFIs) to finance the social protection measures that would be 1 The AE provided cash transfers to people outside formal wage employment but not eligible for social assistance. The BEm aimed to maintain employment relationships by introducing a system of short-time work. 2 During 2017 and 2018, extreme poverty continued to increase in spite of the timid recovery. As a result of the exhaustion of efficiency savings, the BF started to build up a ‘waiting list’ of uncovered poor families in 2019. Page 4 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) deployed to respond to the COVID-19 crisis. Six IFIs, including the World Bank,3 coordinated their financing over BF, SD, AE, and Bem, based on institutional experience and prior engagements with the GoB, for a total of more than US$4 billion. The World Bank took the lead in several aspects during the process: it prepared a common interim financial report (IFR) framework for all donors (to avoid double counting of financed expenditures), and two bilateral financers also adopted part of the World Bank Results Framework and Project Operations Manual (POM). 5. The rapid expansion of social protection was possible due to the investments made by the GoB, including with World Bank support, in the two previous decades. This included a physical network of social assistance offices, a sophisticated and well-updated social registry of applicants to social assistance (Cadastro Único) and digital delivery and payment systems.4 Since 2003, the World Bank had been engaged in supporting the expansion of the social protection system for the poorest, including two operations to develop BF since its inception and the Cadastro Único (BR Bolsa Familia 1st APL, P087713, and the Second Bolsa Familia Project, P101504), as well as providing for years Advisory Services and Analytics (ASA) across the social protection spectrum. Hence, in 2020, the World Bank chose to finance the expansion of BF, considering its past financial and analytical support and the strategic value of the benefits for the World Bank Group COVID-19 Crisis Response Approach. Other donors focused on AE, BEm, and SD though three IFIs also co-financed the budget line for BF families. Other IFIs financed the stock of beneficiaries already in the BF program, while the World Bank financing was provided conditional on being used for beneficiaries added since the pandemic. Designed as such, the World Bank loan ensured the countercyclical policy choice to grow BF, after years of zero real growth and mounting waiting lists. 6. The World Bank devoted a new ASA associated with this operation. Its objectives included assisting the GoB in adapting the design of the BF program to meet the additional challenges posed by the COVID-19 pandemic,5 rethink how social protection should evolve in the post-pandemic era, and document the lessons learned from Brazil’s response for other countries (Lara de Arruda, Pedro; Marina Lazarotto De Andrade; Tiago Falcao Silva; Diana Teixeira Barbosa; and Matteo Morgandi. 2022a. Sharing Lessons Learned from Brazil's Social Protection Response to COVID-19: The Auxílio Emergencial Program. Washington, DC: World Bank Group). The World Bank Group Board approved the project with a near- unanimous statement of support, emphasizing the importance of donor-coordinated response and of the ASA associated with the loan. Annex 7 describes a sample of the World Bank ASAs on social protection developed before and during the pandemic. Theory of Change (Results Chain) 7. The Theory of Change was that the BF expansion would lead to two outcomes explicitly derived from the Project Development Objective (PDO). One, the transfer would protect beneficiaries of the program against shortfalls in income. This would help mitigate welfare losses due to income shocks with long-term implications, including malnourishment and school dropouts. Two, these families would be incentivized to keep children in school and attend regular health checks because the program monitors compliance with human capital conditionalities. The expected impacts were for beneficiary families to 3 The other five institutions were the Interamerican Development Bank (BID), the New Development Bank (NDB), Agence Francaise de Developpement (AFD), Kreditanstalt für Wiederaufbau (KfW) Development Bank, and Cooperacion Andina de Fomento (CAF). 4 See annex 6 for more detailed information. 5 See annex 7 for detailed information on the ASA. Page 5 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) reduce the risk of adopting negative coping strategies, including children dropping out of school, not receiving recommended vaccines, or not attending regular health checks (once services became available). Figure 1. Results Chain in Project Appraisal Document Problem Statement: Economic losses induced by COVID-19 adversely affect poor and vulnerable families’ welfare including malnourishment and increased youth school dropouts. Activities Provide financing of BF expansion Outputs Registration of applicants Cash transfers are paid to 1.2 Outcomes Beneficiary eligibility verification million BF beneficiary families system with cross-check of other databases Beneficiary elegibility confirmed Poor families'income and human Impacts capital are protected from Implement the Comunication Applicants' complaints and negative economic effects Strategy questions are addressed associated with COVID-19: CCT beneficiary children reduce risks of dropping out of school, not Citizens have clarity on elegibility Poor and extreme poor families receiving recommended vaccines, Implement the grievance redress are protected against further mechanism (GRM) and criteria for inclusion or not having health check, when shortfall in income after COVID-19 services are available through CCT Beneficiary families reduce risk of CCT families are incentivized to adopting negative coping maintain children in education and strategies due to livelihood losses health checks through monitoring associated with COVID-19 Critical Assumptions: (1) The BF program will continue to be adequately funded in 2021 and (2) COVID-19 pandemic infection rates slow down and by January 2021, education will have resumed and health centers will remain operational. Project Development Objectives (PDOs) 8. The PDO was to mitigate the negative economic effects associated with COVID-19 on income and risks of damaging human capital of poor families. Key Expected Outcomes and Outcome Indicators 9. The project had four outcome indicators to measure progress toward achieving the development objective. • Families who are maintained above extreme poverty program threshold with project financing (Number; disaggregated by sex and by vulnerable group category of direct recipient) • Mean income support received with project financing by families as a share of the income at the time of application (Percentage) • Children 6–17 benefiting from CCT6 with project financing whose school attendance is being monitored by the program (Percentage; disaggregated by vulnerable group category) 6 CCT = Conditional cash transfer. Page 6 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) • Children under 7 years old benefiting from CCT with project financing, who are being monitored by the program for compliance with health check-ups (Percentage; disaggregated by vulnerable group category). Components 10. The project had a single component: Conditional Cash Transfers for poor and extreme poor families. Project financing would cover the cost of BF transfers corresponding to its countercyclical expansion as a response to the COVID-19 pandemic. The BF expansion was defined as the increase in coverage of the BF compared to a baseline value of 13.05 million families, the number in the program up to March 2020. At that point, 1.2 million families on a waitlist were authorized to enter the BF. These families had been found eligible for the BF before the COVID-19 crisis, but they could not be added to the beneficiary rosters because of budget restrictions. The project would reimburse the average value of BF benefits paid to the number of families in excess of the baseline coverage level pre-pandemic.7 The number of covered families could exceed 1.2 million depending on the 2021 budget, revisions in the benefit level, and percentage and duration of World Bank financing. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION Revised PDOs and Outcome Targets 11. The PDO was not revised. Revised PDO Indicators 12. The PDO indicators (PDIs) were not revised. Revised Components 13. The single component was not revised. Other Changes 14. During project implementation, the GoB replaced BF with AB. On August 9, 2021, the GoB replaced BF with AB through Provisional Measure 1061, later converted to Law 14.284 (December 29, 2021). According to a provision in the Project Loan Agreement (paragraph 5 of the appendix), the definition of BF included “any successor acceptable to the Bank.” The Ministry of Citizenship (MoC) and the World Bank team confirmed the equivalence of AB and BF during the February 2022 supervision mission and the POM was revised accordingly.8 BF families migrated automatically to AB. Operational and institutional arrangements did not change. 7 In addition to the ‘BF’ benefit, most BF families received additional benefits, which varied over time, during project implementation. These benefits were not financed by the project (see annex 6 for detailed information). 8 Since AB benefits maintained the BF objectives— to support the basic income of poor and extremely poor families through cash transfers and to stimulate access to public policies to improve their welfare through monitoring of conditions related to health and education—the World Bank mission concluded that AB qualified as a “successor acceptable to the Bank .” Page 7 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Rationale for Changes and Their Implication on the Original Theory of Change 15. The replacement of BF by AB did not modify the original Theory of Change. The substitution of the BF program was necessitated by the decision of the GoB to replace it with the AB program. Based on the analysis that BF and AB were ‘equivalent’ programs in their objectives, the replacement was judged acceptable to the World Bank and thus eligible for the same project financing as BF, according to the provisions of the Loan Agreement.9 II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating Rating: High 16. The relevance of the PDO is rated as High, as it supported the objectives of the GoB’s National Contingency Plan for Human Infection by COVID-19, as well as the World Bank Country Partnership Framework (CPF) objectives. The project supported the GoB’s efforts to address the economic consequences of the COVID-19 pandemic by financing social protection responses that support households’ welfare and mitigate risks of loss of human capital. The project contributed to measures protecting the poor and vulnerable, particularly women, from income shocks related to COVID-19. 17. The original CPF for Brazil for FY18–FY23 (Report No. 113259-BR) was in place from the time the project was appraised to when it closed, but its focus was adjusted in the interim to respond to COVID- 19. A Performance and Learning Review (PLR) of the CPF was carried out in 2022 (Report No. 143636-BR) with the aim to assess implementation of the CPF until January 2022 and propose adjustments up to June 2023 (CPF Realignment). The latter included the countercyclical emergency support to protect the poor and vulnerable to address the impacts of the pandemic. The support provided by the project was captured under CPF Focus Area 1: Fiscal Consolidation and Government Effectiveness for Sustainable, Inclusive and Efficient Service Delivery. Under specific Objective 1.2, the PLR added a supplementary indicator to capture the shift in World Bank Group emphasis (‘new families who are maintained above extreme poverty with conditional cash transfer program supported by the Bank’), which mirrored a PDI. The PLR notes that the long-standing dialogue in social protection and just-in-time technical assistance (TA) provided by the World Bank to Brazil allowed for rapid and effective lending when the GoB requested it. The ASA associated with the project was also central to the CPF realignment as it would support the GoB in facilitating economic recovery, by informing reforms to the social protection system. The PLR also highlighted the World Bank’s role in coordinating the joint response of six IFIs, which together accounted for two-thirds of the GoB’s response to the pandemic through new spending in 2020. 9While AB’s poverty-focused benefits were eligible for project financing, other AB categorical benefits were not (bonuses for athletes, winners of academic competitions and individuals entering formal employment, and daycare transfers) nor was the Extraordinary Benefit Program’ (Benefício Extraordinário, BE) designed to bring all AB recipients to a minimum benefit amount of R$400 or the temporary Benefício Complementar program to bring recipients to a minimum benefit of R$600. Page 8 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) B. ACHIEVEMENT OF PDOs (EFFICACY) Rating: Substantial 18. According to the PDO, the project had two objectives: 1. Mitigate the negative economic effects associated with COVID-19 on the income of poor families and 2. Mitigate the risks of damaging human capital of poor families. The achievement of each objective is evaluated below. Assessment of Achievement of Each Objective/Outcome 19. Objective 1: Mitigate the negative economic effects associated with COVID-19 on the income of poor families was fully achieved. There were two PDIs to measure the achievement of this objective: the number of families and the income support received. The first indicator was surpassed by a large margin, while the second indicator was largely achieved with the actual result only marginally below the target. 20. The expansion of BF/AB significantly exceeded even the higher estimates of program beneficiaries stated in the Project Appraisal Document (PAD) while the income gains for families were slightly lower than expected. At the end of 2021, BF had expanded to include nearly 1.9 million families, already above the planned 1.2 million. The successor AB, however, raised both eligibility thresholds and the generosity of benefits which had declined in real terms. Those reforms, combined with a removal of budget constraints, led to a substantial growth in Program coverage by roughly an additional 6 million families by June 2022, by which date the project was completely disbursed. By the project closing date (December 2022), nearly 10 million families had been added. Given the larger-than-expected expansion in BF/AB, the disaggregated targets for PDI1 (by gender and vulnerable group) were also exceeded by a large margin (see paragraph 44). The income support indicator for BF/AB ended up slightly below the target of 55 percent (53.33 percent) at the time of project closure (Table 1) and at 54.5 percent at the time of the last disbursement in June 2022. Importantly, all beneficiary families for the project in the same period received a significantly higher amount than what this indicator accounts for, due to complementary transfers via Auxilio Emergencial in 2021 and Beneficio Extraordinario in 2022,10 which were mechanically providing top-ups to any recipient of AB/BF during the project period (see annex 6 for details). Table 1. Results for PDIs 1 and 2 PDI1 Families who are maintained above extreme Baseline: 0 poverty program threshold with project financing11 Target: 1,200,000 Actual: 9,959,838 (December 2022) PDI2 Mean income support received with project Baseline: 0 financing by families as a share of the income at the Target: 55.0 time of application (Percentage)12 Actual: 53.32 (December 2022) 10 This indicator depends both on the declared income level of applicants and on the generosity of the benefit. The average income of applicants likely increased after 2020 as the program raised its eligibility thresholds by about 10 percent in November 2021 (to account for inflation). The program also experienced a moderate increase in the average benefit, and the cost-of-living increase was largely addressed through the complementary benefit to all AB/BF recipients—Beneficio Extraordinario. However, the positive effect on mitigating poverty of this top-up is not visible in PDI 2. The Project Implementation Unit (PIU) computed that, when accounting for both AB and BE, PDI2 would surpass 70 percent in 2022 (See Government Completion Report). 11 For the calculation of the indicators related to the first PDO objective, “with project financing” was defined as the differe nce between the March 2020 baseline of 13.05 families and the updated coverage of BF/AB. 12 The indicator was computed only for the AB benefit, and it does not include BE (see annex 6), which families also received at the time. The PIU computed that, if BE was considered, this PDI would rise to above 70 percent. Page 9 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 21. Objective 2: Mitigate the risks of damaging human capital of poor families was partially achieved. As shown in the Theory of Change and the selection of PDIs, the PDO is highly specific and thus intentionally conservative monitoring of compliance with education and health conditions provided incentives to CCT families to maintain school attendance and health checks for children. At preparation, it was expected that production of data by line ministries regarding compliance with health and education conditionalities would resume during the implementation period of the project. The targets set for both PDIs were equivalent to the pre-pandemic shares of children being monitored by the BF program. The monitoring of compliance with conditions did resume in 2022 and achieved a substantial coverage of children in the Program. However, it did not fully reach pre-pandemic monitoring levels for either education or health due to several reasons. In the case of education, public schools remained closed on average for about 9.5 months, with those in the northeast region closing even longer (nearly a year). The Ministry of Education began to monitor school attendance only in February/March 2022, introducing a new system to record school attendance. The system suffered some operational glitches and gaps in availability, problems that were compounded by the learning curve for municipal staff to use the new system. As a result, the MoC was not able to report complete results until the third quarter of 2022. In the case of health, from the beginning of the pandemic until mid-2021, the Ministry of Health suspended the obligation of health authorities to record information on health visits of children and pregnant women, as well as discouraged health visits by these groups to reduce the chance of contagion and permit health staff to concentrate efforts on addressing the pandemic. Targets for the related disaggregated indicators were only met for compliance monitoring of health conditions for the subgroup of Quilombola (afro-descendants living in traditional areas). Compliance monitoring of education conditions for indigenous children had the largest relative gap compared to the target (63.32 percent versus 85 percent) (see annex 1B). 22. While targets for the resumption of monitoring of compliance with health and education conditions were not reached, the achievement of Objective 1 of the project likely contributed to mitigating the risks of damage to the human capital of poor families. The two PDIs ‘mapped’ to the second PDO objective related to monitoring of health and education conditions by the program (Table 2). Nevertheless, the project’s Theory of Change correctly recognized the role of the cash transfer on its own to help achieve two outcomes (attendance at schools and health check-ups) to ultimately have a positive impact on protecting families’ human capital accumulation. This second order, yet key, outcome results from the fact that the additional cash can reduce household liquidity constraints, thus covering the direct, indirect, and opportunity costs associated with school enrolment, attendance, and retention as well as use of health services.13 Table 2. Results for PDIs 3 and 4 PDI3 Children 6–17 benefiting from CCT with project financing,14 Baseline: 0 whose school attendance is being monitored by the Target: 85.0 program (Percentage) Actual: 70.5 (December 2022) PDI4 Children under 7 years old benefiting from CCT with Baseline: 0 project financing, who are being monitored by the Target: 60.0 13 Bastagli, Francesca; Jessica Hagen-Zanker; Luke Harman; Valentina Barca; Georgina Sturge and Tanja Schmidt, with Luca Pellerano. 2016. “Cash Transfers: What Does the Evidence Say? A rigorous review of programme impact and of the role of design and implementation features” Overseas Development Institute. 14 In the case of the second PDO, the indicators are calculated for the entire BF/AB program, not just the expansion. Page 10 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) program for compliance with health checkups (Percentage) Actual: 45.32 (December 2022) 23. Moreover, data for late 2021 from a nationally representative phone survey provide suggestive evidence that children of BF families did not disproportionately reduce the utilization of health and education services during the pandemic. (a) Preliminary results from the Continuous National Household Sample Survey (PNAD-C) 2022 also suggest that children in BF families in extreme poverty (those with an income per capita lower than R$ 210) showed attendance rates of 96.5 percent, compared to 93.1 percent among those not receiving BF with similar income levels. Also, 96.6 percent of children in BF families attended school, compared to 95.7 percent among non-BF recipients in Cadastro Único (typically with higher income). (b) A survey collected by the World Bank team at the end of 2022 on a sample of 1,300 AB recipients, who enrolled in the program after December 2021, revealed that utilization of health and education services were sustained at high levels. In particular, • 95 percent of poor and extreme poor15 children in AB visited a primary health unit within the previous 12 months. • 88 percent of extreme poor children and 92 percent of the poor children in AB received a vaccine in the previous 12 months. • 80 percent of pregnant women in extreme poor families and 90 percent of the women in poor families and in AB received prenatal checkups (though sample sizes were small for this indicator). • 89 percent of children in poor and in extreme poor families attended school. • 89 percent of enrolled children in extreme poor and 93 percent in poor families attended school in the previous week. Justification of Overall Efficacy Rating 24. Overall efficacy is rated as Substantial. The first objective is considered fully achieved, with one indicator surpassing by a large margin what it was originally intended to achieve and the other indicator achieving 97 percent. Although the PDIs for the second objective fell short of the targets, the data demonstrate that the program was monitoring health and education conditions for a substantial share of children in the program by the end of the project, even if not at the pre-pandemic level. Moreover, survey data demonstrate that a high proportion of BF children were attending school and going for health check- ups, two actual intended outcomes of the project’s Theory of Change. Such behavior is consistent with a 15Poor and extreme poor families are defined according to the declared income before enrollment in the program in Cadastro Único, at the time the sampling for the impact study occurred. Poor families defined as with income below R$ 210 per capita and extreme poor below R$ 105 per capita. Page 11 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) positive incentive effect of the program’s monitoring of conditionalities.16 The overall rating of Substantial also factors in the likely contribution of the cash transfer on its own (Objective 1) to mitigate the risk of damage to human capital. C. EFFICIENCY Rating: High Assessment of Efficiency and Rating 25. Efficiency is rated as High. At appraisal, the estimated economic efficiency for the project was measured by BF targeting performance and the impact of BF expansion on poverty reduction. For the Implementation Completion and Results Report (ICR), these estimates were updated using PNAD-C 2022 survey data (see annex 4) and show that Bolsa Familia’s successor, Auxilio Brasil, maintained strong targeting performance despite the rapid expansion (double the pre-pandemic coverage): over three- fourth of beneficiaries (76 percent) were in the bottom 30 percent of the income distribution, including 37 percent in the bottom decile and 61 percent in the bottom quintile.17 Moreover, simulations using the BraSIM tool (the same as presented in the PAD) suggest that the benefits from the BF/AB program led to a decrease in extreme poverty of 3.6 percentage points in 2022. This implies that, in the absence of the program, extreme poverty would have been 10.3 percent (instead of a simulated 6.7 percent). Moreover, despite strong support from the BF/AB program, there is no evidence of negative impacts on beneficiaries’ attachment to the labor market, as was also expected in the original economic analysis. 26. Project design and implementation aspects contributed to efficiency. The GoB made use of the provision for retroactive financing and the project fully disbursed six months before the closing date. The closing date was not extended, and staff turnover was not an issue. Operating costs were low, the PIU roles were fulfilled by existing staff in the ministry, trained by the World Bank fiduciary and Environmental and Social Framework (ESF) teams at the project onset. The common arrangements agreed among donors for financial reporting and audits also fostered efficiency and economies of scale for the project (see paragraph 30 for more details). The actual financing provided by the six donors was broadly in line with PAD estimates, with only minor differences in the distribution of support by program and amount. D. JUSTIFICATION OF OVERALL OUTCOME RATING 27. The overall outcome rating is Satisfactory. Based on the high relevance of PDO objectives, a rating of Substantial for efficacy and a rating of High for efficiency, the overall outcome rating is assessed as Satisfactory. 16 While it is not possible to pinpoint the explanation for this result, there is international evidence that perceptions of conditions and messaging of prescribed behavior, alone or together with the monitoring of compliance, can promote desired behaviors (Bastagli, Francesca; Jessica Hagen-Zanker; Luke Harman; Valentina Barca; Georgina Sturge and Tanja Schmidt, with Luca Pellerano. 2016. “Cash Transfers: What does the Evidence Say? ” Overseas Development Institute, page 267). This would be expected given the long history and reputation of the BF program as a conditional transfer in Brazil. 17 The 2022 distribution of beneficiaries is slightly less concentrated in the bottom 20 percent compared to 2019 but still very good targeting performance relative to other countries in Latin America and the Caribbean (LAC) with significantly less ambitious levels of coverage. Page 12 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) E. OTHER OUTCOMES AND IMPACTS Gender 28. The project overachieved its gender-related targets mainly because of the much larger-than- expected expansion of BF/AB, but the share of women enrolled in the expansion was less than PAD estimates. The project supported a reduction in gender gaps in earnings among the population eligible for BF, by supporting the expansion of a program that prioritized women as beneficiaries, as tracked by the gender disaggregation of enrollments for women in specific groups (indigenous and Quilombolas), as well as ‘other’ (the remainder of beneficiaries).18 Given that the expansion of BF/AB significantly exceeded what was expected in the PAD baseline, the three numerical targets set for women (indigenous, Quilombolas, and other) were exceeded by large margins (nearly five times higher in the case of the ‘other’ group). However, the share of women enrolled in BF/AB became lower by the end of the project than at the baseline (68 percent versus 82 percent for indigenous, 61 percent versus 73 percent for Quilombolas, and 76 percent versus 80 percent for other groups). A likely reason is the rise in single person families in AB, which are more dominated by men: this unexpected outcome was incentivized by the introduction of Beneficio Extraordinario, the top-up introduced in 2021 (and extinguished in 2023), which gave an additional generous benefit per family regardless of household size, which led to incentives for men to register as separate entities than their (unmarried) partners.19 During the life of the project, the team also conducted two analytical studies that informed the ministry on constraints on the participation of women to work and determinants of their insertion in the labor market (see annex 7). Institutional Strengthening 29. The project solely financed BF/AB cash transfers to beneficiaries, but parallel TA is likely to have contributed to institutional strengthening. Implementation built on the sound experience of the Government in implementing cash transfer operations through established processes and systems that allowed for orderly implementation and eligibility controls. Prior World Bank operations had supported the strengthening of institutional capacity for the delivery of the BF program from its inception in 2003, first through the BR Bolsa Familia 1st Adaptable Program Lending (APL) (P087713) and between 2012 and 2018 through the Second Bolsa Familia Project (P101504). Both projects included a large TA component to help build capacity of the BF delivery institution, the management model of Cadastro Único, and monitoring and evaluation (M&E). In the case of this project, parallel TA provided during implementation generated technical notes on the lessons learned from AE (Lara de Arruda et al. 2022) and supported innovations to the AB benefit, in part incorporated in the new AB Law 2021.20 18 The design of BF/AB incorporated a strong push to enroll women—overrepresented among poor household heads—as the household member who received the BF/AB beneficiary payment whenever possible. Targeting transfers specifically to women can partially reduce the impact of the labor market earnings gap, with positive impacts on empowerment and agency. 19 Auxilio Emergencial gave a fixed transfer to families regardless of the family size and composition. This shift in the profile of beneficiary families has been acknowledged by the new administration, which has already identified several million families to be invited for verification of the family information in Cadastro Único. 20 The studies informed the review of the exit rule from the program, the introduction of a saving component for beneficiaries that graduate from AB into a formal job, piloting of new modalities for the delivery of early childhood services, and introduction of financial rewards for excellence in education (Fietz, Katharina Maria; Tiago Falcao Silva; Maria Concepcion Steta Gandara; and Matteo Morgandi. 2021. “Balancing Stability and Transition: A First Assessment of Regra de Permanência in Bolsa Familia.” Policy Note no. 162843, World Bank Group, Washington, DC.). Page 13 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Poverty Reduction and Shared Prosperity 30. The project had a direct positive impact on increasing household income of poor beneficiaries. As noted by PDI2 (table 1 above), the benefit received among project beneficiaries amounted to an average of 53.3 percent of their reported income at the time of registration.21 In addition, as noted in the efficiency section, the AB program was well targeted and the benefits provided by the program led to a decrease in poverty in 2022, with a particular effect on extreme poverty. 31. Simulations in annex 4 suggest that the benefits provided by the AF program led to a decrease in poverty in 2022, with a particular effect on extreme poverty. More specifically, simulations show that in the absence of the program, the share of Brazilian families in extreme poverty would have been 10.3 percent, whereas extreme poverty decreases to 9.7 percent (equivalent to a 1.1 million change), owing to the core BF program (co-financed by the loan) and further drops to 6.6 percent when including the effects of the generous temporary top-up BE. As further discussed in annex 4, the benefits provided by the program achieved the outcomes efficiently, being strongly targeted to the bottom quintile. Climate Change 32. BF, and its successor AB, supported by the project helped increase the resilience of the poor to climate-related extreme events. Using data from Cadastro Único, the June 2022 AB payment list and the World Bank 2021 Disasters Risk Index by Capacity (IRDC),22 it is estimated that 29 percent (5.3 million families) of AB’s total beneficiary families live in municipalities considered high or very high risk of climate disaster. The high proportion of families’ exposure to risk denotes the important protective factor of the cash transfer program for households if enrolled in AB. Moreover, engaging with the Ministry of Social Development in the initial estimate of climate co-benefits led to a fruitful policy dialogue to further develop protocols for adaptive social protection, now part of a separate TA line.23 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 33. The project benefited from a simple design and readiness for implementation. Financing for BF/AB cash transfers was the only project component. Technical assistance, which had been a component of previous World Bank operations supporting BF (P087713 and P101504) related to the continuing Government efforts to strengthen and reform BF, was handled in a parallel program of ASA, funded by externally financed outputs (EFOs). The Results Framework focused on the main outcomes expected from the project. In the case of the human capital objective, the Results Framework intended to focus on what 21 As families who are eligible for BF can only earn less than R$210 per capita, a simple back-of-the-envelope would suggest that, on average, the benefit funded by the project was about R$112 per person (equivalent to 53.3 percent of R$210). 22 The IRDC uses Brazilian data from the Center of Natural Disaster Prevention to estimate the extent of risk exposure to natural disasters of Brazilian municipalities and the municipalities’ capacity to respond. Classified in five groups, the municipalities in classes D and E are considered at high or very high risk of disasters, mainly climate related, due to a combination of high exposure and low capacity to respond to them. In Brazil, 1,591 out of 5,570 municipalities were classified in 2021 as at high or very high risk, meaning that these areas are both at high risk of facing a climate-related distress and have a low capacity to act and protect its population before and upon the effects. 23 The World Bank organized a conference on climate and social protection in Florianopolis and prepared a policy note ‘Towards Shock Responsive Social Protection in Brazil’. https://urpartner.io/event/towards-shock-responsive-social-protection-in-brazil/. Page 14 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) the project could be held accountable for within its short implementation period—the monitoring of compliance with health and education conditions—although the policies of educational and health sector authorities had a large influence on this indicator. The Results Framework monitoring plan was based on well-established systems in the MoC. The project was ready for implementation, since BF was an existing program using robust Government systems. 34. The risks identified were accurate but skewed toward possible reductions in the BF budget and the number of BF beneficiaries. Contrary to the concerns identified in the PAD, the expansion of the BF/AB programs exceeded the PAD estimates by a very large margin, even in its higher-case scenario. The main mitigation measure identified that related to possible changes in the BF program parameters was the continuous engagement of the World Bank with the Government through the ASA/TA. A risk correctly identified was the possibility that COVID-19 could result in a reduction in the provision of health and education services for an extended period with a corresponding impact on the magnitude of the project’s impact on human capital and the ability to monitor compliance with health and education conditions. The main mitigation was trends at the time of PAD preparation: a drop in COVID-19 cases and the gradual reopening of schools in most Brazilian states. 35. The World Bank played an important role in donor coordination. Led by the World Bank, the six donors across the four main social protection programs (BF/AB, AE, BEm, and SD) agreed on a common format for reporting, using the World Bank template for the IFRs as the model, and the terms of reference for a single audit. The arrangement reduced the potential duplication of efforts and the need for the PIU to handle many different procedures and requirements. B. KEY FACTORS DURING IMPLEMENTATION Factors Subject to Government/Implementing Agency Control 36. The project benefited from established capacity and effective coordination arrangements within the MoC. The MoC, together with the MoE, had the challenge of managing financing for BF/AB from six different donors. In addition, several units within the MoC (Cadastro Único, SENARC24, Ouvidoria), as well as the Treasury, had responsibilities in the implementation of the World Bank-financed project. The units worked well together and internal communication was good, enabling speedy project preparation and timely disbursements. The PIU was well organized and possessed the required human resources, as well as the full support of the MoC management. The POM was prepared on time, with World Bank approval in December 2020. The PIU coordinated the necessary dialogue with the World Bank, on both technical and administrative topics. There were no issues with basic logistics. The legislation governing BF was in place at the beginning of the project, later replaced by a new legal framework for AB. The World Bank consistently rated Fiduciary and Environmental and Social Safeguards compliance as Satisfactory. High-quality progress reports were delivered on time and contained the required information on monitoring indicators. 24 National Secretariat for Citizen Income (Secretaria Nacional de Renda de Cidadania). Page 15 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Factors Subject to World Bank Control 37. On the side of the World Bank, an important contribution to the project was its coordination of other donors financing BF/AB and the provision of advice during implementation through complementary ASA/TA. At the stage of resource mobilization, through the leadership of the World Bank, the multiple donors financing BF/AB agreed on common standardized reporting. Despite the short duration of the project, conditions changed substantially. AB replaced the BF and in May 2022, the GoB initiated BE.25 World Bank staff adapted appropriately to these changing conditions, principally by assessing AB as equivalent to BF and therefore eligible for financing under the loan and by not constraining the faster disbursement of project funds. The same task team leader, and by and large the broader World Bank team, was in place during project implementation and provided appropriate advice and support to M&E. A parallel ASA carried out by the World Bank during the period of the project was crucial in providing evidence to inform the discussion and decisions on design changes and improvements in BF. Examples of topics included an analysis of international experience in financial incentives to stimulate education outcomes, leading to the incorporation of financial rewards to low-income students with extraordinary achievements in the AB Law. A study on mobility and exit patterns of families in BF informed the review of the exist rules for the new AB benefit (Fietz et al. 2021). A study on patterns on savings and the use of financial instruments by BF families informed parts of the AB legislation that included financial education as an objective of complementary activities for beneficiaries and opened space to design a precautionary saving instrument for graduates of AB that entered in a formal job (World Bank 2021). In addition, World Bank staff carried out simulations of the impact of possible reforms on BF on poverty, equity and targeting, ahead of the AB reform (Morgandi et al. 2019). The study on experience of AE served to inform an internal agenda of modernization of verification and controls of income declarations of AB beneficiaries (Morgandi et al. 2021). The policy note that benchmarked Brazil with other countries in adaptive social protection is now informing a dialogue with the Ministry of Social Development to review its protocols to intervene in disaster-affected areas (Ed, Malin; Alejandro Gonzalez; and Renata Ferreira. 2023. “Climate Change and the Role of Social Protection in Brazil.” Washington, DC: The World Bank Group). Factors outside the Control of Government and/or Implementing Agencies 38. The reopening of schools and the resumption of monitoring of compliance with education and health conditions were outside the control of the MoC. The resumption of monitoring of the health and education conditions of BF/AB depended on decisions of state governments to reopen schools post COVID-19 and the decisions of the national Ministries of Health and Education to again require health centers and schools to record the required information. In the case of education, additional delays were experienced due to the launch of a new information system to record school attendance information. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 25 BE is a top-up for AB recipients so that all families would receive at minimum amount. Page 16 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 39. The M&E design for the project was well aligned with its focus on financing cash transfers for the BF/AB expansion. The indicators followed directly from the two PDOs to mitigate the negative economic effects associated with COVID-19 on income and the risks of damaging human capital of poor families. The Theory of Change linked the receipt of the cash transfers with the expected objectives, supported by a large body of international evidence. While the Results Framework was concise, the indicators covered both objectives and had clear targets. Information was to be drawn from established information systems in the MoC and consolidated by the PIU. One moderate shortcoming was the use of the terminology “with project financing” since it was difficult to accurately measure the indicators only for the share of total transfers or beneficiaries that the loan financed. For the PDIs mapped to the first objective (mitigate the negative economic effects associated with COVID-19 on the income of poor families), indicators were calculated for the entire expansion of BF/AB above the 13.05 million March 2022 baseline. For the PDIs mapped to the second objective (mitigate the risks of damaging human capital of poor families), indicators were calculated for the entire BF/AF program. Another moderate shortcoming was a misunderstanding on the specification and internal government procedures related to the indicators for the grievance redress mechanism (GRM), reported in the final is(October 2022). The definition of the indicator in the PAD referred to “cases being resolved within 30 days”, while the applicable regulations required “cases to be replied to within 30 days” with additional time allowed if necessary to obtain more information or other reason. M&E Implementation 40. Monitoring data were generally of high quality and reliable. The MoC team collected and analyzed monitoring data using several information systems in the MoC related to BF/AB, including the system to follow conditionalities, the Cadastro Único, and the GRM system. Overall, they followed the robust methodology agreed upon during project preparation. The PIU effectively consolidated the required information into the progress reports. Project M&E was rated as Moderately Satisfactory in two out of four Implementation Status and Results Report (ISRs) (ISRs 2 and 3) because of the delays in reporting on the indicators related to monitoring of compliance with conditions as well as in finalizing the arrangements for the planned impact evaluation (see the following paragraph). Delays in the former were due to the issues discussed earlier in the section on the achievement of the second PDO and beyond the MoC’s control. The moderate shortcoming mentioned above on the definition of the indicator on the GRM was addressed by the Government providing information for both cases ‘resolved’ and cases ‘replied to’ within 30 days. In addition, due to a change in the information system of the unit responsible for the GRM, it was not possible to track results related to the GRM for the Quilombolas. 41. An impact evaluation survey to measure effects of AB was carried out; however, its implementation did not occur according to the intended design, limiting the ability to draw causal inference from the results. The World Bank team had planned to conduct an impact evaluation of the expansion of BF/AB as part of the parallel ASA/TA. The evaluation was meant to provide additional evidence on the impact of the project on the development objectives. The collection of such information went well beyond what was committed to be measured in the PAD and the Results Framework. The evaluation design was going to rely on a comparison of beneficiary families to a random sample of families registered in Cadastro Único but who were just above the eligibility cutoffs defined in December 2021 and expected not to receive the benefit over 2022 (that is, the control group). The survey was piloted in May and fielded and completed by November 2022, and it provided data on human development outcomes experienced by families who applied and received BF/AB, as described in the Outcomes section above. Page 17 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) However, after the sample was drawn, many sampled families shifted from control to treatment status due to their re-applications to BF after the new eligibility thresholds were announced; this greatly reduced the availability of households in the control group. Thus, the outcomes here reported for AB beneficiary families cannot be compared with similar non-beneficiary families.26 M&E Utilization 42. The ministry regularly uses the wealth of data produced by the registry and the management information system to make decisions on refining design, as well as implementation. For instance, the reform of AB was entirely simulated with the social registry data multiple times to identify the impact, and the World Bank used its own methodologies to complement on the MoC’s estimates of potential poverty incidence. The social registry data were used to identify the potential challenges related to introducing new education conditionality for age group below 6 and above 18 years. Justification of Overall Rating of Quality of M&E 43. The quality of M&E is rated as Substantial based on the ratings for the three dimensions. Based on moderate shortcomings in M&E design and implementation and good utilization of M&E data, the overall rating is Substantial. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 44. The project complied with the five applicable World Bank Environmental and Social Standards (ESS). The applicable ESS were ESS1: Assessment and Management of Environmental and Social Risks and Impacts, ESS2: Labor and Working Conditions, ESS4: Community Health and Safety, ESS7: Indigenous Peoples/Sub-Saharan African Historically Underserved Traditional Local Communities, and ESS10: Stakeholder Engagement and Information Disclosure. The PIU in the MoC fulfilled its responsibilities to report on and monitor Environmental and Social Commitment Plan measures and actions throughout project implementation, with specific authorities responsible for each ESS. For instance, the MoC Ombudsman (Ouvidoria) was responsible for ESS10, the National Secretariat of the Single Registry (Secretaria Nacional do Cadastro Único, SECAD) and SENARC were responsible for ESS4, 7, and 10, while ESS2 was the responsibility of only SENARC. During project implementation, no issues were reported by any of the aforementioned authorities, and all associated reporting obligations were completed and delivered on time. 45. The project provided financial support to expand a well-established cash transfer program, resulting in negligible negative social impacts and no adverse environmental impacts, as there were no construction activities or other services that directly or indirectly affected the environment. BF had installed capacity to promote participation by vulnerable groups. To increase coverage of the most vulnerable population, the BF operating rule prioritized certain groups to enter the program based on 26 Two factors affected the eligibility status of many families originally assigned to the ‘control group’ in the study. First, the additional expansion of AB and an increase of the program generosity raised incentives for families to recertify over the course of a few months their declared income, which was the basis to construct the sample for the study. Second, the eligibility thresholds to enter the program increased two months after the AB Law was passed, around February 2022, causing many families in the control group to become eligible and receive the benefit. These changes greatly reduced the number of families in the study’s control group. Page 18 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) their degree of vulnerability. These priority groups included indigenous peoples, Quilombolas, families with children in child labor situations, families with individuals working in situations of slavery, and recyclable material collectors, in that order. For example, once identified, indigenous families were enrolled in BF within 30 days. Furthermore, the MoC developed tailored guidance materials for traditional populations to facilitate their access to the benefit and financial inclusion through the new payment modality through banks. These materials were distributed to state coordinators who were in direct contact with these populations to support the dissemination of the program. The ministry organized ‘Auxílio Brasil Journeys’ to disseminate information on the AB program and engage with local managers of Cadastro Único, who are responsible for enrolling and following up with AB beneficiaries. The MoC also conducted two surveys to identify the special needs of indigenous peoples for enrollment and to ensure their full participation in the AB program. 46. The project surpassed, by a large margin, the disaggregated first PDI targets in the Results Framework related to vulnerable groups, mainly because of the larger-than-projected expansion of BF/AB. The disaggregated targets for the indicator “families who are maintained above extreme poverty program threshold with project financing (number)” were 1,800 indigenous women and 400 indigenous men and 1,100 women and 400 men from Quilombola communities. As of November 2022, actual values were as follows: 39,984 indigenous women, 24,190 indigenous men, and 35,365 women and 28,583 men from Quilombola communities. This overachievement was driven by the greater-than-expected expansion of BF/AB. 47. Reporting on the intermediate indicators relating to the GRM was affected by some initial confusion on the definition. As mentioned in the preceding section on M&E, there was some initial confusion on the definition of the GRM indicator, namely cases ‘resolved’. The project relied on the existing system of the MoC ombudsman (‘Ouvidoria’), regulated by Law 13.460/2017. The GRM-related target was that the Ouvidoria would resolve claims within 30 days (Cases from the BF program grievance redress mechanism resolved in 30 days.) However, as reported in the final ISR, according to the applicable legal regulation, the Ouvidoria was obligated to respond (address), but not necessarily resolve, an issue that had been raised because some cases required additional information to be provided. In addition, citizens have the right to appeal, which could prolong the time required for total resolution of a case. In addition, due to changes in the Ouvidoria information system during project implementation, it was not feasible to monitor cases specifically from the Quilombola individuals. Data on cases from the BF program grievance mechanism and their handling according to both definitions are contained in annex 1.B. 48. The project complied with all fiduciary requirements. Since there were no procurement activities under the project, fiduciary requirements covered solely financial management. Fiduciary management was rated as Satisfactory throughout the duration of the project. All IFRs submitted were satisfactory, with only occasional small delays. Audit reports were unqualified by the auditors, acceptable to the World Bank, and received with only small delays. C. BANK PERFORMANCE Quality at Entry 49. Quality at entry is rated Highly Satisfactory. The approach of the project to finance a horizontal expansion of BF by incorporating eligible families on the waiting list in the midst of the COVID-19 pandemic Page 19 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) was highly relevant for poverty reduction. The project was prepared at record speed, with Government request reaching financers in early April, a Regional Operations Committee Decision Review held in early July, and Operations Committee Decision Review in August 202027. The main delay in approval was due to protracted negotiations in the ESF documentation, introduced for the first time in Brazil with this project.28 Robust technical and economic analysis, the handling of social safeguards, and financial management all built on the history of World Bank support for the development of BF and Cadastro Único from its inception in 2003 to 2018 through two lending operations (P087713 and P101504) and thereafter through ASA/TA services. Implementation arrangements were appropriately embedded in the existing structure of the MoC. As mentioned earlier, the assessment of risks skewed toward possible reductions in the BF budget and number of BF beneficiaries, the opposite of what occurred in practice. World Bank inputs and processes made a positive contribution to quality, for example, refining the PDOs and associated indicators, including explicit mention of the flexibility of the loan to accommodate additional eligible families and the incorporation of specific mechanisms to ensure that the loan would not finance expenditures covered by other donors. Quality of Supervision 50. The quality of supervision is rated Satisfactory. The focus of supervision was on the development impact of the project through the expansion of BF/AF as well as the resumption of monitoring of compliance with health and education conditions. The World Bank team also monitored financial management and ESF aspects closely and reported on developments in the ISRs. Four ISRs were prepared, sufficient given the 24-month project duration. Given the products available in the parallel ASA/TA, supervision inputs were well above the norm. Reporting on project performance was candid. The ISRs included reporting of changes in the country context including the extension of AE. The ISRs also included the issues around the delays in the resumption of monitoring of compliance with education and health conditionalities and the misunderstanding on the definition of the GRM indicator. World Bank staff were clear in the ISRs that the reforms in BF represented by AB and the expansion of the program beyond initial estimates did not pose a risk to the PDO. Justification of Overall Rating of Bank Performance 51. The overall rating of World Bank Performance is Satisfactory. Based on the rating of Highly Satisfactory for quality at entry and Satisfactory for the quality of supervision, the overall rating for World Bank performance is Satisfactory. D. RISK TO DEVELOPMENT OUTCOME 52. The risk to development outcome is considered low. The recently elected administration launched a reformulated Bolsa Familia Program on March 2, 2023. The most important development in comparison with AB relates to the consolidation in a single budget line and a single program what used to be two benefits: Auxilio Brazil and ‘Beneficio Extraordinario’ (the top-up that significantly increased the AB value). The new BF thus resembles in structure and generosity the transfer that beneficiaries received 27 The higher than average preparation costs noted in annex were due to incorrect posting of supervision staff time expenditures for some months after approval to the preparation code. 28 This was the first investment project in Brazil since the adoption of the ESF, which was more comprehensive than the framework adopted by other donors, and this was also one of the first loans approved during the COVID-19 period. Page 20 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) in 2022, as do the institutional and implementation arrangements. Importantly, the centrality of monitoring compliance with health and education conditionalities and of following up with families experiencing difficulties remains a principle of the new Bolsa Familia. One of the priorities recognized by the Ministry of Social Development (the renamed MoC) in the context of the current program is to renew efforts with municipalities to increase the share of children for whom compliance is monitored, including by restoring the value of block grants for the municipal social assistance network closer to pre-pandemic years in the next budget. 53. The one risk, however, is that administrative budget for the new BF program had not matched its expansion in 2023. With a much larger number of beneficiaries, the new BF program reached an unprecedented scale, but the administrative budget in the last year did not grow commensurably. The 2023 budget was inherited from the previous administration and the current government recognized the importance of increasing investment in the central administrative structure, including delivery systems, Cadastro Único, and the social assistance network (this will likely pay for itself through reduced inclusion errors). V. LESSONS AND RECOMMENDATIONS 54. Investing in a solid social protection delivery system reduces the costs and time needed to roll out an efficient crisis response. Scaling up in time of need is far more feasible under those circumstances, and the prior inclusion of vulnerable families in the social registry, even if not all could be attended to, allowed to scale up BF quickly in the crisis. 55. Maintaining a robust analytical program, policy dialogue and just-in-time TA can enable a strong and speedy response by the World Bank in case of a need for financing as well as other benefits. The last World Bank project related to BF before this project had closed at the end of 2018. Nevertheless, because the World Bank had continued an active dialogue with the Government related to social protection, including BF, staff were able to respond quickly to the emergency need for financing given their strong knowledge, technical expertise, and history of cooperation with the MoC. TA can also play a strategic role by maintaining the policy dialogue across different administrations, helping to maintain the institutional memory, and contributing to the maintenance of effective programs and policies. 56. Incorporating a mechanism to assess and accommodate changes which do not affect the project rationale can reduce possible delays and time spent restructuring the project. The World Bank team correctly anticipated that changes could occur in BF during project implementation and incorporated an assessment and approval mechanism within the Loan Agreement. When the change occurred (the replacement of BF by AB), the World Bank team responded quickly, carried out an assessment in collaboration with the Government, and maintained financing, avoiding interruptions and time to restructure the project. 57. The definition and targets for indicators in the Results Framework need to be fully consistent with national regulations. As discussed in the sections on M&E and compliance with social safeguards, the initial definition and thus the targets set for cases handled in the grievance redress system were not consistent with national regulations. Page 21 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 58. Strong complementarity between World Bank-financed ASA, Government programming, and World Bank financing can serve to reduce the risk to development outcomes of the project. The diagnostics developed through ASA during the course of the project remain relevant to support implementation of the new BF benefit. With an average benefit value above R$ 600, the program is expected to continue to receive high pressure in 2023 from potential applicants to enter and for existing beneficiaries to remain. Continued analysis of benefit incentive design and monitoring of sociodemographic characteristics of beneficiaries and households will be important to identify incentive issues that could be corrected through further parametric changes. 59. The prospect of receiving high-quality tailored TA can be a major factor leading a government to seek financing by the World Bank. Clients can access a range of development financing options, such as those mobilized during the COVID-19 pandemic, at similar conditions of those offered by the World Bank. World Bank teams’ access to resources, including through trust funds and EFOs, which allows supporting borrowers via ASAs for important reforms has proven important for the World Bank to remain a partner of choice in middle-income countries. 60. Clients can coordinate donors financing for common crisis response. The Government guidance to all development banks to prepare their operation under a single project financing proposal (Carta Consulta) generated a collaborative dynamic. In this environment, the World Bank was able to take the lead in several operational aspects that generated efficiency, included IFRs, draft audit terms of reference, and a Results Framework followed by at least two other organizations. . Page 22 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Mitigate negative economic effects associated with COVID-19 on income of poor families Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Families who are maintained Number 0.00 1,200,000.00 9,959,838.00 above extreme poverty program threshold with 31-Mar-2020 30-Dec-2022 30-Dec-2022 project financing Families who are Number 0.00 1,800.00 39,984.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Indigenous Female Families who are Number 0.00 400.00 24,190.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Page 23 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Indigenous Male Families who are Number 0.00 1,100.00 35,365.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Quilombola Female Families who are Number 0.00 400.00 28,583.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Quilombola Male Families who are Number 0.00 936,300.00 7,078,828.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Other Groups Female Families who are Number 0.00 233,000.00 2,752,888.00 maintained above extreme poverty program threshold 31-Mar-2020 30-Dec-2022 30-Dec-2022 with project financing - Other Groups Male Comments (achievements against targets): Page 24 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Mean income support Percentage 0.00 55.00 53.32 received with project financing by families as a 31-Mar-2020 30-Dec-2022 30-Dec-2022 share of the income at the time of application Comments (achievements against targets): Objective/Outcome: Mitigate risks of negative economic effects of COVID-19 on human capital of poor families Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Children 6-17 benefiting from Percentage 0.00 85.00 70.56 CCT with project financing, whose school attendance is 31-Mar-2020 30-Dec-2022 30-Dec-2022 being monitored by the program Children 6-17 benefiting Percentage 0.00 85.00 63.32 from CCT with project financing, whose school 31-Mar-2020 30-Dec-2022 30-Dec-2022 attendance is being monitored by the program - Indigenous Page 25 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Children 6-17 benefiting Percentage 0.00 85.00 76.30 from CCT with project financing, whose school 31-Mar-2020 30-Dec-2022 30-Dec-2022 attendance is being monitored by the program - Quilombola Children 6-17 benefiting Percentage 0.00 85.00 70.61 from CCT with project financing, whose school 31-Mar-2020 30-Dec-2022 30-Dec-2022 attendance is being monitored by the program - Other groups Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Children under 7 years old Percentage 0.00 60.00 45.32 benefiting from CCT with project financing, who are 31-Mar-2020 30-Dec-2022 30-Dec-2022 being monitored by the program for compliance with health checkups Children under 7 years old Percentage 0.00 60.00 45.24 benefiting from CCT with project financing, who are 31-Mar-2020 30-Dec-2022 30-Dec-2022 Page 26 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) being monitored by the program for compliance with health checkups - Indigenous Children under 7 years old Percentage 0.00 60.00 59.81 benefiting from CCT with project financing, who are 31-Mar-2020 30-Dec-2022 30-Dec-2022 being monitored by the program for compliance with health checkups - Quilombola Children under 7 years old Percentage 0.00 60.00 45.18 benefiting from CCT with project financing, who are 31-Mar-2020 30-Dec-2022 30-Dec-2022 being monitored by the program for compliance with health checkups - Other groups Comments (achievements against targets): A.2 Intermediate Results Indicators Component: Conditional Cash Transfers for poor and extreme poor families Indicator Name Unit of Measure Baseline Original Target Formally Revised Actual Achieved at Page 27 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Target Completion Beneficiaries of social safety Number 0.00 2,900,000.00 20,252,936.00 net programs 31-Mar-2020 30-Dec-2022 30-Dec-2022 Beneficiaries of social safety Number 0.00 2,320,000.00 11,556,819.00 net programs - Female Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Cases from the BF program Percentage 0.00 60.00 93.98 grievance redress mechanism resolved in 30 31-Mar-2020 30-Dec-2022 30-Dec-2022 days Cases from the BF program Percentage 0.00 60.00 100.00 grievance redress mechanism resolved in 30 31-Mar-2020 30-Dec-2022 30-Dec-2022 days - Indigenous Cases from the BF program Percentage 0.00 60.00 0.00 grievance redress mechanism resolved in 30 30-Mar-2020 30-Dec-2022 30-Dec-2022 days - Quilombolas Page 28 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Cases from the BF program Percentage 0.00 60.00 93.80 grievance redress mechanism resolved in 30 31-Mar-2020 30-Dec-2022 30-Dec-2022 days - Other groups Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Adoption and improvement Yes/No No Yes Yes of communication strategies with specific channels for 31-Mar-2020 30-Dec-2022 30-Jun-2022 vulnerable groups of BF, which increase awareness of the availability of Fala BR for claims related to BF Comments (achievements against targets): Page 29 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1 Mitigate negative economic effects associated with COVID-19 on income of poor families 1. Families who are maintained above extreme poverty program threshold with project financing The December 2022 value (9,959,838) significantly exceeded the target of 1.2 million. 2. Families who are maintained above extreme poverty program threshold with project financing (indigenous female) The December 2022 value (39,984) signficantly exceeded the target of 1,800. 2. Families who are maintained above extreme poverty program threshold with project financing (indigenous male) The December 2022 value (24,190) signifcantly exceeded the target Outcome Indicators of 400. 3. Families who are maintained above extreme poverty program threshold with project financing (Quilombolas female) The December 2022 value (35,365) significantly exceeded the trget of 1,100. 4. Families who are maintained above extreme poverty program threshold with project financing (Quilombolas male) The December 2022 value (28,583) significantly exceeded the target of 400. 5. Families who are maintained above extreme poverty program threshold with project financing (other groups female) Page 30 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) The December 2022 value (7,078,828) significantly exceeded the target of 936,300. 6. Families who are maintained above extreme poverty program threshold with project financing (other groups male) The December 2022 value (2,752,888) significantly exceeded the target of 233,000. 2. Mean income support received with project financing by families as a share of the income at the time of application The December 2022 value (53.32 percent) represented a 97% achievement of the target of 55 percent. 1. Beneficiaries of social safety net programs The December 2022 value (20,252,936) surpassed by a large margin the target of 2,900,000. 2. Beneficiaries of social safety net programs (females) While the December 2022 value (11,556,819) was surpassed by a large margin the target of 2,320,000, the proportion of female beneficiaries (57 percent) is less than the share implied in the target (80 percent) Intermediate Results Indicators due to a change in eligibility criteria for the AB program. 3. Cases from the BF program grievance redress mechanism resolved in 30 days As of December 2022, 93.98 percent of cases were responded to within 30 days. As of June 2022, 41 percent of cases were resolved within 30 days. Cases from the BF program grievance redress mechanism resolved in 30 days (Indigenous) Page 31 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) As of December 2022, 100 percent of cases from BF program indigenous beneficiaries were responded to within 30 days As of June 2022, 38 percent of cases from BF program for indigenous beneficiaries were resolved within 30 days. Cases from the BF program greivance redress mechanism resolved in 30 days (Quilombolas) Because of a change in the grievance redress mechanism system it was not possible to monitor cases from Quilombolas beneficiaries. Cases from the BF program greivance redress mechanism resolved in 30 days (other groups) As of December 2022, 93.8 percent of cases were responded to in 30 days for BF beneficiaries of “other groups”. As of June 2022, 42 percent of cases were resolved within 30 days for BF beneficiaries of “other groups”. 7. Adoption and improvement of communication strategies with specific channels for vulnerable groups of BF, which increase awareness of the availability of Fala BR for claims related to BF Improved communication strategies were adopted. Component 1 1. Cash Transfers paid to BF/AB beneficiary families. 2. Applicants’ complaints and questions are answered (See Key Outputs by Component intermediate indicators 3-6 above). (linked to the achievement of the Objective/Outcome 1) 3. Improved communications strategy adapted for vulnerable groups of BF implemented. Page 32 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Objective/Outcome 2 Mitigate risks of negative economic effects of COVID-19 on human capital of poor families 1. Children 6-17 benefiting from CCT with project financing, whose school attendance is being monitored by the program The December 2022 value (referring to 3rd period 2022) was 70.56 percent, below the target of 85 percent. 2.Children 6-17 benefiting from CCT with project financing, whose school attendance is being monitored by the program (Indigenous) The December 2022 value (referring to 3rd period 2022) was 63.32 percent, below the target of 85 percent. 3. Children 6-17 benefiting from CCT with project financing, whose school attendance is being monitored by the program (Quilombolas) The December 2022 value (referring to 3rd period 2022) was 76.3 percent, below the target of 85 percent. 4. Children 6-17 benefiting from CCT with project financing, whose Outcome Indicators school attendance is being monitored by the program (Other groups) The December 2022 value (referring to 3rd period 2022) was 70.61 percent, below the target of 85 percent). 5. Children under 7 years old benefiting from CCT with project financing, who are being monitored by the program for compliance with health checkups The December 2022 value (referring to 1st period 2022) was 45.32 percent, below the target of 60 percent. 6. Children under 7 years old benefiting from CCT with project financing, who are being monitored by the program for compliance with health checkups (Indigenous) The December 2022 value (referring to 1st period 2022) was 45.24 percent, below the target of 60 percent. Page 33 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 7. Children under 7 years old benefiting from CCT with project financing, who are being monitored by the program for compliance with health checkups (Quilombolas) The December 2022 value (referring to 1st period 2022) was 59.81 percent, virtually meeting the target of 60 percent. 8. Children under 7 years old benefiting from CCT with project financing, who are being monitored by the program for compliance with health checkups (other groups) The December 2022 value (referring to 1st period 2022) was 45.18, below the target of 60 percent. Intermediate Results Indicators N/A Component 1 1. Cash transfers paid to BF/AB families Key Outputs by Component 2. Compliance monitoring system for health and education (linked to the achievement of the Objective/Outcome 2) conditionalities in operation Page 34 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION TASK TEAM MEMBERS Name Role Preparation Matteo Morgandi, Liliana Do Couto Sousa Task Team Leader(s) Sinue Aliram De Souza Procurement Specialist(s) Fernanda Balduino de Oliveira Financial Management Specialist Katharina Maria Fietz Team Member Fabio Saia Cereda Team Member Matheus De Mello de Assuncao Team Member Maria Ines Miranda Ramos Environmental Specialist Barbara Segato Monteiro Team Member Federico Antonio Beckley Gonzalez Team Member Rafael Amaral Ornelas Team Member Fabiano Silvio Colbano Team Member Gabriela Geraldes Bastos Team Member Miriam Muller Team Member Augusto Ferreira Mendonca Environmental Specialist Gabriel Lara Ibarra Team Member Pablo Ariel Acosta Team Member Rafael Munoz Moreno Team Member Graciela Sanchez Martinez Social Specialist L. Keta Ruiz Team Member Page 35 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Supervision/ICR Matteo Morgandi Task Team Leader Gabriel Lara Ibarra Task team Leader Fernanda Balduino de Oliveira Financial Management Specialist Silmara Moreira Da Silva Financial Management Specialist Alberto Coelho Gomes Costa Social Specialist Pablo Ariel Acosta Program Leader Augusto Ferreira Mendonca Environmental Specialist Tiago Falcao Silva Team Member Federico Antonio Beckley Gonzalez Team Member, Program Assistant Barbara Segato Monteiro Team Member, Program Assistant Caio Cicero De Toledo Piza Impact Evaluation lead Rafael Dantas Impact Evaluation consultant Theresa Penelope Jones ICR Lead author Tamires Amorim ICR Social Protection consultant Ricardo Campante Vale ICR Poverty consultant A. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY20 8.223 49,067.39 FY21 88.340 390,690.07 FY22 0 0.00 Total 96.56 439,757.46 Supervision/ICR FY21 11.850 49,905.11 Page 36 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) FY22 57.178 270,558.15 FY23 22.571 161,058.94 Total 91.60 481,522.20 Page 37 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 3. PROJECT COST BY COMPONENT Amount at Actual at Project Percentage of Components Approval Closing Approval (US$, millions) (US$, millions) Conditional Cash Transfers for poor and 997.50 997.50 100 extreme poor families Total 997.50 997.50 100 Page 38 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 4. ECONOMIC ANALYSIS 1. This annex presents an update of the economic analysis conducted at preparation; it estimates the impacts of the financed program on poverty reduction and assesses its targeting efficiency through a fiscal incidence analysis. 2. The BF program has long been considered one of the most progressive fiscal policies in Brazil. Higgins and Pereira29 (2014) estimated its concentration coefficient was −0.58 using Household Budget Survey (Pesquisa de Orçamentos Familiares - POF) 2008/2009 data, the most progressive policy in the country among cash and in-kind transfers, direct taxes, and indirect taxes. Silveira and Palomo30 (2023), using the more recent POF 2017/2018, reiterate the earlier finding. They also suggest that BF was one of the main factors driving the decrease in inequality in the 2000s. Finally, Lara Ibarra et al.31 (forthcoming), building upon PNADC 2019 data, estimated BF’s marginal contribution to inequality reduction was equal to 0.011 Gini points, while it was 0.4 percentage points on poverty reduction, one of the most pro-poor policies in Brazil. Figure 4.1. Share of Benefits of BF/AB by Decile of Pre- Figure 4. 2. Average Per Capita Benefit by Decile of BF Income Pre-BF Income 37% 80 70 24% 60 50 16% 40 9% 8% 30 3% 2% 20 1% 1% 0% 10 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Source: PNAD-C 2022. Note: The period of analysis was restricted to the first semester —when the World Bank’s financing was in place. Pre-BF income is defined by household’s income (without BF transfers) per capita. 3. Specific to the period covered by the project, the recently released PNADC 2022 data suggest that BF’s successor, AB, maintained very strong targeting performance, despite experiencing a rapid 29 Higgins, Sean, and Claudiney Pereira. 2014. “The Effects of Brazil’s Taxation and Social Spending on the Distribution of Household Income.” Public Finance Review 42 (3): 346–367. 30 Silveira, Fernando Gaiger, and Theo Ribas Palomo. 2023. “The Brazilian State’s Redistributive Role: Changes and Persistence at the Beginning of the 21st Century.” Discussion Paper/Institute for Applied Economic Research (IPEA). 31 Lara Ibarra, G., M. Cabrera, O. Conceição, and R. Campante Vale. Forthcoming. “Poverty and Inequality Implications of Fiscal Policies: The Case of Brazil.” Page 39 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) growth in coverage, and Bolsa Familia remains among the programs in Latin America with the best targeting performance, despite being larger than most programs. The survey data suggest that in 2022, over three-fourth of beneficiaries (76 percent) were in the bottom 30 percent of the population and 86 percent of beneficiaries were in the bottom 40 percent. This denotes only a minor deterioration in targeting performance compared to the one observed in the 2019 national household panel survey (86 percent versus 89 percent),32 though the number of beneficiaries in 2019 was about sixty five percent smaller. Moreover, when comparing Brazil with other CCT programs in Latin America, it is also important to highlight how Bolsa Familia continues to be among the programs with highest quality of targeting, according to estimates based on harmonized household surveys and methodology in the World Bank Atlas for Social Protection (ASPIRE), while also being one of the programs with the highest level of coverage of the population (Table 4.1) Table 4.1. Distribution of Conditional Cash Transfer Beneficiaries by Quintiles of Pre-Transfer Welfare and Coverage Targeting Country Program Year Coverage Q1 Q2 Q3 Q4 Q5 Panama Red de Oportunidades; SENAPAN 2019 83 13 2 2 0 6 Peru Programa Juntos 2019 65 23 8 3 1 12 Brazil Bolsa Familia 2019 64 26 8 2 0 18 El Comunidades Solidarias Rurales; 1 2019 62 23 9 5 1 Salvador Comunidades Solidarias Urbanas Paraguay Tekopora 2019 57 27 14 3 0 11 Uruguay Asignaciones Familiares 2019 56 28 11 4 1 25 Ecuador Bono de Desarrollo Humano 2019 51 25 14 7 2 20 Costa Rica Ayuda monetaria IMAS 2019 49 31 16 4 0 18 Colombia Jóvenes en Acción 2019 49 29 15 6 1 15 Mexico Prospera (Oportunidades, Progresa) 2018 47 26 16 8 2 21 Honduras Bonos PRAF; Bono Vida Mejor 2017 47 29 14 6 4 13 Subsidio Familiar (SUF); Bono de Protección Familiar y de Egreso; Bono por Chile 2017 43 28 15 8 5 17 control del niño sano; Bono por asistencia escolar; Bono por logro escolar Programme of Advancement Through Jamaica 2017 35 29 18 14 3 28 Health and Education (PATH) Bono Juancito Pinto; Bono Juana Bolivia 2019 27 25 22 18 8 57 Azurduy_madre; Bono Juana Azurduy_hijo Incentivo a la Asistencia Escolar; Bono Dominican Escolar Estudiando Progreso; Incentivo a la 2019 27 24 23 18 8 26 Republic Educacion Superior, con Tarjeta Solidaridad Source: ASPIRE - www.worldbank.org/aspire. Notes: Coverage is: (Number of individuals in a given group [i.e. total population or poorest quintile] who live in a household where at least one member receives the transfer)/(Number of individuals in the group) The poorest quintile is calculated using per capita pre-transfer welfare (income or consumption). 32 The Continuous National Household Sample Survey (Pesquisa Nacional por Amostra do Domicílios - Contínua (PNAD-C)). Page 40 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 4. Comparing this with the program eligibility rules, 78.2 percent of the reported beneficiaries had their pre-BF income below the national social registry (Cadastro Único) eligibility line (R$606 per capita in 2022). Moreover, the Cadastro Único eligibility line is very close to R$599, the value estimated (in 2022 prices) as the poverty line using the cost of basic needs approach with the most recent data available for Brazil (Lara Ibarra, Paffhausen, and Duque 202133). When translated in international terms, Cadastro Único’s eligibility line corresponds roughly to the 6.85 US dollar per day 2017 purchasing power parity (PPP) poverty line used for upper-middle-income countries. 5. Simulations using BraSIM (the same tool used at appraisal) suggest that the benefits from the program led to a decrease in poverty in 2022. The outcomes of a microsimulation based on the PNADC household data suggest that 1.1 million people would fall into extreme poverty in the absence of the children and youth benefits. Without the full program, the extreme poverty rate would increase, by 3.6 percentage points, from an estimated 6.6 percent baseline. The increase is equivalent to 7.7 million more individuals in extreme poverty. Table 4.2. Microsimulations on the Impacts of the Program Poverty Rate Extreme Poverty Rate Simulation (%) (%) Full program 27.6 6.7 Subtracting early childhood and youth components only 27.7 7.2 Subtracting BE only 29.0 9.7 Subtracting the Full Program 29.2 10.3 Source: BraSim 2.0 simulations prepared by the World Bank Poverty team. Note: Poverty rate calculated at half a minimum wage per capita (R$606) and extreme poverty at AB upper eligibility line (R$210). 6. Survey data from 2022 provide suggestive evidence that there were no negative impacts on beneficiaries’ attachment to the labor market. Among families earning less than R$210 per capita, individuals in the working-age population in BF families showed similar labor force participation rates (43.7 percent) and those in non-BF families (44.9 percent), while their employment rates were higher: 26.8 percent and 17.8 percent, respectively. Table 4.3. Employment and Unemployment Statistics among Beneficiaries and Non-Beneficiaries (%) Unemployment Rate Employment Rate Labor Force Participation Rate Other extreme poor 60.3 17.8 44.9 BF family 38.7 26.8 43.7 Total 45.0 24.3 44.1 Source: PNADC 2022. Note: The period of analysis was restricted to the first semester of 2022 —when the World Bank’s financing was in place. Extreme poor are defined as those whose household per capita income (before BF benefits) are below the program line (R$210). Statistics based on population ages 15 to 64. 33Lara Ibarra, Gabriel, Anna Luisa Paffhausen, and Daniel Duque. 2021. “Estimating a Poverty Line for Brazil Based on the 2017/18 Household Budget Survey.” Policy Research Working Paper No. 9878, World Bank, Washington, DC. Page 41 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 5. BORROWER COMPLETION REPORT ImplementationCompletio IBRD LOAN AGREEMENT 9172-BR: Brazil: Income Support for the Poor Affected by COVID-19 Project Implementation Completion Report Ministry of Development and Social Assistance, Family and Fight against Hunger Page 42 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) BRAZIL FINAL PROJECT REPORT Implementation Completion Report (ICR): Government “BRAZIL: INCOME SUPPORT FOR THE POOR AFFECTED BY COVID-19 PROJECT – 9172 – BR" Ministry of Development and Social Assistance, Family and Fight against Hunger (MDS) Brasília, February 2023. Page 43 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Summary 1. EXECUTIVE SUMMARY ................................................................................................................................. 44 2. BRAZIL: INCOME SUPPORT FOR THE POOR AFFECTED BY COVID-19 PROJECT ............................................. 45 2.1. RATIONALE FOR THE OPERATION .............................................................................................................45 3. THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT (IBRD) PROJECT ........................ 48 3.1. IBRD LOAN AGREEMENT .............................................................................................................................48 3.2. PROJECT OBJECTIVE.....................................................................................................................................48 4. ASSESSMENT OF THE OPERATION'S RESULTS .............................................................................................. 49 4.1. COMPONENT .................................................................................................................................................49 4.2. DEVELOPMENT INDICATORS MATRIX .......................................................................................................50 4.3. ESTABLISHED TARGETS ..............................................................................................................................52 4.4. NON-REIMBURSABLE TECHNICAL ADVISORY SERVICES: WORLD BANK STUDIES IN PARTNERSHIP WITH THE IDB AND AFD .........................................................................................................................................55 4.5. OTHER ACTIONS PRODUCED BY THE PARTNERSHIP ...............................................................................59 5. ASSESSMENT OF THE MINISTRY'S PERFORMANCE DURING THE PREPARATION AND IMPLEMENTATION OF THE OPERATION .................................................................................................................................................. 60 5.1. LESSONS LEARNED .......................................................................................................................................60 6. FINANCIAL MANAGEMENT OF THE IBRD PROJECT ....................................................................................... 63 7. PROJECT SUSTAINABILITY........................................................................................................................... 65 8. AUDIT.......................................................................................................................................................... 66 9. KEY TEAM: INDIVIDUALS RESPONSIBLE FOR THE PROJECT ......................................................................... 66 9.1. MINISTRY OF DEVELOPMENT AND SOCIAL ASSISTANCE, FAMILY AND FIGHT AGAINST HUNGER ......66 10. ANNEX ......................................................................................................................................................... 67 1. EXECUTIVE SUMMARY This is a Final Progress Report, known to the funder as an Implementation Completion Report (ICR): Government, provided for in Loan Agreement No. 9172-BR-IBRD between the International Bank for Reconstruction and Development (IBRD) and the Federative Republic of Brazil, within the scope of the Brazil: Income Support for the Poor Affected by COVID-19 Project, which seeks to mitigate the negative economic effects associated with COVID-19 on income and the risks of it Page 44 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) harming the human capital of poor households. The Loan Agreement signed on June 17, 2021, between the Federative Republic of Brazil, represented by the Ministry of Economy, and the International Bank for Reconstruction and Development (IBRD) was in the amount of USD 1,000,000,000 (one billion US dollars) to help finance the Programa Bolsa Família Expansion component carried out by the then Ministry of Citizenship, which has since been renamed to Ministry of Development and Social Assistance, Family and Fight against Hunger - MDS. This report encompasses the entire project implementation period until its completion. The Indicators Matrix is included as an annex to this report and covers the half-yearly reference period, with calculations using as reference the period from July to December 2022, and includes historical data from previous semesters. 2. BRAZIL: INCOME SUPPORT FOR THE POOR AFFECTED BY COVID-19 PROJECT 2.1. RATIONALE FOR THE OPERATION On March 11, 2020, the World Health Organization (WHO) categorized COVID-19 - a disease caused by the novel Coronavirus (2019-n-CoV) that affects the respiratory tract - as a pandemic, after having declared it a Public Health Emergency of International Concern (PHEIC). The necessary measures to slow down the spread of contamination by COVID-19, prevent the collapse of the health system and preserve lives directly affected the country's low-income population, as this was the segment of the population with the highest numbers of precarious and informal jobs and self-employed workers in the country. Faced with the need to socially isolate, these workers became unemployed or were prevented from working. The Brazilian Government had to enact measures to ensure the social protection for these workers so they would be able to support themselves and their families while subjected to the collective health measures imposed on the population - particularly social isolation. With the aim of affording emergency social protection to informal workers and others affected by the economic crisis prompted by the novel Coronavirus, as well as contributing to collective health measures so that these workers could engage in social isolation without losing all of their income, the President of Brazil sanctioned Law 13,982 of April 2, 2020, which amended Law 8,742 of December 7, 1993, to provide for additional parameters in defining the condition of social vulnerability for the purpose of eligibility to the Continuous Welfare Benefit for the Elderly and Disabled (Benefício de Prestação Continuada, BPC) and establish exceptional social protection measures to be put in place during the span of the public health emergency of international concern stemming from the novel Coronavirus disease (COVID-19) that caused the 2019 outbreak, as referred to in Law 13,979 of February 6, 2020”. While distancing and isolation measures were indeed needed from a public health perspective, they also had immediate economic impacts and lasting consequences on countries. The behavioral changes needed to curb the transmission of COVID-19 were strongly influenced by the decisions of governments (to close schools, cancel public events, etc.), companies and institutions (to transition to a work-from-home model, adjust production levels, etc.) and consumers (to reduce Page 45 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) social contact). Such changes led to a very significant economic contraction, with both immediate and long-term effects. In this sense, with a view to protecting the income of informal and self-employed workers during the pandemic, the government passed Law 13,982 of April 2, 2020, to create the Emergency Aid initiative, meant to protect the workers most severely impacted by the economic effects of the COVID-19 pandemic - i.e., informal workers, Individual Microentrepreneurs and individual contributors to the general pension system for private sector workers (Regime Geral da Previdência Social, RGPS). In view of the need to quickly implement actions to respond to the COVID-19 pandemic and its strong economic impact, the Brazilian Government sought credit from international institutions. This led to the creation of the Brazil: Income Support for the Poor Affected by COVID-19 Project, aimed at contributing to and ensuring minimum levels of quality of life for vulnerable people in the face of the labor market crisis caused by COVID-19. The project supported and encouraged actions put in place by the Ministry of Economy in priority areas to combat the economic effects of the pandemic. The Project was submitted for evaluation to the External Financing Commission (Comissão de Financiamento Externo, COFIEX) by means of a Consultation Letter with a request for financing entirely with external funds, without any counterpart funding from the Brazilian Government. The financing operation was approved by COFIEX on May 25, 2020, by means of Resolution 1, and published in the Federal Official Gazette on May 29, 2020, in Section 1, Page 176. The MDS, then known as the Ministry of Citizenship, was included and authorized as Program Executor by COFIEX Resolution 22 of July 29, 2020 (9003111), without prejudice to the other terms of Resolution 1. Funding was granted by the following financial institutions: • French Development Agency (AFD) • Inter-American Development Bank (IDB) • International Bank for Reconstruction and Development (IBRD) • Andean Development Corporation (CAF) • KfW Entwicklungsbank (KfW) • New Development Bank (NDB). These financing institutions were chosen for having significant funding readily available for loaning, as well as for their broad expertise in executing and supporting highly complex projects that required increased technical and institutional capacity, skills and information, in addition to international references and experience. The MDS, then known as the Ministry of Citizenship, served as the beneficiary of all the financiers, with the exception of CAF, which reimbursed the expenses of the component under the scope of the Ministry of Economy (Ministério da Economia), now called the Ministry of Finance (Ministério da Fazenda). The Program was structured into four components, in accordance with Consultation Letter No. Page 46 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 60723 (8832287): • Component 1: Emergency Basic Income • Component 2: Programa Bolsa Família Expansion • Component 3: Emergency Employment and Income Maintenance Program • Component 4: Unemployment Insurance Program Under this component structure, the Ministry of Citizenship would execute Components 1 and 2, and the Ministry of Economy would execute Components 3 and 4. Institution Component Target Amount CURRENCY Population International Bank for Bolsa Família Expansion $ 1,000,000,000.00 US DOLLAR Reconstruction and Development (IBRD); Inter-American Development Emergency Unified $ 400,000,000.00 US DOLLAR Bank (IDB) Aid Registry Bolsa Família Inventory $ 200,000,000.00 US DOLLAR New Development Bank (NDB) Emergency ExtraCad $ 1,000,000,000.00 US DOLLAR Aid (App) 2020 French Development Emergency Bolsa Família € 130,000,000.00 EURO Agency (AFD) Aid 2020 Bolsa Família Inventory € 70,000,000.00 EURO KfW Entwicklungsbank (KfW) Bolsa Família Inventory € 350,000,000.00 EURO *Inventory: Number of beneficiary households before the pandemic (Ref.: PBF Payroll - March 2020) The distribution among funders of the estimated reimbursement of expenses incurred by each component was planned carefully to ensure no overlap in fund allocation, directing them to different target audiences and different payment installments. Resources Allocation Timeline of Refinancing of Program Expenditures by Funders Allocation Page 47 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 3. THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT (IBRD) PROJECT 3.1. IBRD LOAN AGREEMENT Prior to the signing of the Agreement with the IBRD, pre-negotiations took place from September 24 to 30, 2020, and a negotiation mission occurred on October 1 and 2, 2020, with representatives of the Ministry of Citizenship (MC), the Office of the General Counsel for the National Treasury (Procuradoria-Geral da Fazenda Nacional, PGFN/ME), the National Treasury Secretariat (Secretaria do Tesouro Nacional, STN/ME) and the International Economic Affairs Secretariat (Secretaria de Assuntos Econômicos Internacionais, SAIN/ME), as well as representatives of the IBRD. From the core team of the then-called Ministry of Citizenship, the meeting was attended by representatives from the Executive Secretariat, the Directorate of International Affairs, the Legal Team, the National Secretariat of Citizenship Income (Secretaria Nacional de Renda de Cidadania, SENARC) and the General Coordination of Technical Cooperation, which at the time was part of the Secretariat for the Management of Funds and Transfers (Secretaria de Gestão de Fundos e Transferências, SGFT) under the Executive Secretariat. The Loan Agreement was signed on June 17, 2021, by and between the Federative Republic of Brazil and the International Bank for Reconstruction and Development (IBRD) of the World Bank Group, for a global amount of US$ 1 billion, entirely financed by the Bank and executed until December 31, 2022. The Project funded conditional cash transfers to at least 1.2 million households (approximately 3 million individuals in total) who were added to Programa Bolsa Família (PBF)/Programa Auxílio Brasil (PAB) in response to the COVID-19 crisis. It should be noted that, during the course of the referred project and due to the projected post- pandemic scenario, Programa Bolsa Família (PBF) was extinguished by Provisional Presidential Decree 1,061 of August 9, passed into Law 14,284, Article 1, of December 29, 2021. Programa Auxílio Brasil (PAB) was then created and replaced Programa Bolsa Família for all legal purposes and also in Loan Agreement No. 9127 – BR – IBRD. Thus, as of November 2021, PBF beneficiaries began to receive PAB benefits instead. The Project continued the Bank's long-standing partnership and operational support commitment to the Ministry regarding the development of the Unified Registry and Programa Bolsa Família until 2018, including analytical work led by the World Bank on social protection in Brazil, which included Programa Bolsa Família. From the creation of Programa Bolsa Família in 2003 until 2018, the World Bank supported the development of PBF and the delivery infrastructure of the current Social Assistance System via the First Bolsa Família Adaptable Program Loan (P087713) and, between 2012 and 2018, via the Second Bolsa Família Project (P101504). 3.2. PROJECT OBJECTIVE The Project's Development Objective was to mitigate the negative economic effects associated with COVID-19 on the income and human capital of poor households, by financing the expansion of PBF/PAB to include eligible households that had not been covered by the Program before COVID-19. Page 48 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) The health crisis caused by the COVID-19 pandemic was partially mitigated once vaccination shifted into gear. Growth had recovered by 2021, but the scenario remained uncertain. Going into 2022, it was expected that better control over the pandemic and the lifting of social distancing measures would continue to improve economic outcomes. However, the war waged by Russia in the Ukraine has affected global value chains and, by consequence, Brazil's economy. Domestic uncertainties about an upcoming - and highly polarized - election held back investment. It must be noted that the objective of the operation remained of the utmost relevance even after the project with the World Bank had come to an end, as the pandemic has left behind lessons and long-term challenges for Brazil's population. 4. ASSESSMENT OF THE OPERATION'S RESULTS 4.1. COMPONENT The External Credit Operation included as a COMPONENT the Programa Bolsa Família conditional cash transfer program (PBF), later replaced by Programa Auxílio Brasil (PAB), and financed part of the transfers to poor and extremely poor households. The Project's sole component funded the Program's countercyclical expansion in response to the COVID-19 pandemic. PBF/PAB expansion is defined as all households eligible for PBF/PAB in addition to the 13.05 million households who were already in the program prior to the expansion. By April 2020, the expansion included 1.2 million households (2.9 million individuals) amongst the poorest and most socially vulnerable in Brazil. The component also financed the inclusion of new households in the PBF/PAB, amounting to a total of 1.2 million households. Approximately 990,000 beneficiaries were children and youths who received financial support and had their school attendance and use of health services monitored as required by PBF protocols, which remained in place after the transition to PAB. These households had applied for - and had been considered eligible for - the PBF prior to the COVID-19 crisis but had not been receiving the benefit due to budget constraints that limited the program's growth. The component refers to the reimbursement of federal government expenses with part of the conditional cash transfers in Programa Bolsa Família and its successor, Programa Auxílio Brasil, under the coordination, management and operation of this Ministry. Exceptional social protection measures were instituted to mitigate the effects of the pandemic on the most vulnerable segments of the population, the main ones being the Emergency Aid (paid from April to August 2020), the Residual Emergency Aid (paid from September to December 2020) and the Emergency Aid 2021 (paid from April to October 2021). When the Emergency Aid came to an end, the government decided to establish a new Program, entitled Programa Auxílio Brasil (PAB), by means of Provisional Presidential Decree 1,061 of August 9, 2021, which was passed into Law 14,284 on December 29, 2021. Provisional Presidential Decree 1,076, of 12/07/2021, converted into Law no. 14,342 on 05/18/2022, added the Extraordinary Benefit to the basket of benefits. Intended for PAB beneficiary households, said benefit was equivalent to the amount necessary to reach R$ 400.00 Page 49 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) transferred to each beneficiary household; as such, it is calculated as the sum of BPI, BCF, BSP and BComp benefits in the reference month. Additionally, Constitutional Amendment 123/2022, Art. 5, item I expanded Programa Auxílio Brasil to beneficiary households and provided for an extraordinary monthly add-on extending to December/2022 in the amount of R$ 200.00 (two hundred reais), spanning from August 1 to December 31, 2022, thus ensuring a minimum R$600.00 transfer to each household registered in the Program. 4.2. DEVELOPMENT INDICATORS MATRIX The Ministry of Citizenship (as it was called at the time) was responsible for collecting and analyzing data about its respective indicators, in accordance with its own management and information systems for PBF and Emergency Aid beneficiaries. The achievement of the project development objective and the intermediate result indicators were monitored using administrative data at the baseline and a report was submitted to the funder, containing: (i) updated indicators; (ii) physical and financial results; (iii) information on the Project's environmental and social management, among other information on Project implementation. o Project Development Objective Indicators: Mitigate the negative economic effects associated with COVID-19 on the income of poor households 1. Households kept above the extreme poverty line, as set by the program, with project financing (number); 2. Support for the average income received from project financing by households as a share of total income at the time of application (Percentage); As for the number of households kept above the extreme poverty line, both in the general indicator (indicator 1) and in those tied to different groups and specific traditional peoples (indicators 1a to 1f), the numbers were substantially higher than the goals initially set by the project. The matrix shows two different calculations for indicator 2 in order to preserve comparability with previous measurements; one version includes the extraordinary benefit and the other does not. The extraordinary benefit was paid during the year 2022; the amount was initially equivalent to the amount necessary to reach R$ 400.00 (four hundred reais) per beneficiary household; subsequently, from August 2022 onwards, it was increased by R$ 200.00 (two hundred reais) monthly as an extraordinary measure so the total amount would reach a minimum of R$ 600.00 transferred to each household. o Project Development Objective Indicators: Mitigate the risk of negative economic effects of COVID-19 on the human capital of poor households Page 50 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 3. Children aged 6 to 17 years who received CCTs from project financing and whose school attendance was monitored by the program (Percentage); 4. Children under the age of 7 who received CCTs from project financing and whose attendance in health check-ups was monitored by the program (Percentage); As mentioned in previous reports, the COVID-19 pandemic had a profound impact on the monitoring of CCT conditionalities, especially in education, where face-to-face school activities were suspended in most state and municipal school networks, making it impossible to monitor school attendance between March 2020 and early 2022. The Ministry of Education (MEC) determined it would be impossible to monitor school attendance during that period. Monitoring of education conditionalities was scheduled to resume in early 2022; more specifically, monitoring would begin for the February/March 2022 bimester. However, due to many technical issues involving the validation of school attendance monitoring data, which had been collected with a new version of the Presença (Portuguese for “attendance”) system, consolidated results could only be obtained for the two-month period of June/July of 2022, which eventually came to be the first monitoring cycle conducted after the pandemic. These results are shown in the indicators matrix, specifically in the form of indicators 3 and 3a to 3c. In addition to the issues mentioned in the previous paragraph that explain the difficulties faced from a technical and operational perspective, it has become clear that the effects of the long suspension of face-to-face classes - namely, the significant increase in the number of school dropouts when face-to-face classes resumed - mainly in low-income households - and, in particular, among their youth, who constitute a focus group under the program - will constitute one of the major challenges for 2023 when it comes to monitoring conditionalities. In fact, overcoming this challenge will require strong mobilization by the states and municipalities, involving not only the National Citizenship Income Secretariat, but also the set of services and programs under the umbrella of Social Assistance and other public policy areas. o Intermediate Results Indicators: Component - Conditional Cash Transfers for poor and extremely poor households 5. Beneficiaries of social safety net programs (CRI, Number); 6. PBF Grievance Mechanism cases resolved within 30 days (Percentage); 7. Enactment and improvement of communication strategies with specific channels for vulnerable groups in PBF to inform of the existence of Fala BR, to be used for grievances pertaining to PBF In the realm of health conditionalities, there was no complete interruption in the monitoring of conditions related to vaccination and the collection of anthropometric measurements (weight and height) of children and the monitoring of women. This means that, although the impacts of the pandemic were significant, they were not as devastating for health as they were for education conditionalities. Impacts were felt, however, from January to June 2020 (the 1st term of 2020), which was when the Ministry of Health (MS) made health conditionality monitoring optional for Page 51 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) municipalities as the focus shifted to combating the pandemic. Despite this specific guidance, the Ministry of Health also advised municipalities to keep the identification of pregnant women and prenatal care cases a priority, so that the Variable Benefit to Pregnant Women (Benefício Variável à Gestante, BVG) could continue to be paid out. These same procedures and guidelines were repeated for the two subsequent terms: the 2nd half of 2020 and the 1st half of 2021. The Ministry of Health decided to once again make it mandatory to record monitoring data on health conditionalities in the 2nd half of 2021. The Indicators Matrix spreadsheet used in this project is included in this report as an Annex. 4.3. ESTABLISHED TARGETS Different project monitoring indicators were analyzed and, during the span of the project, those that sought to mitigate the negative economic effects of COVID-19 showed an upward trend (in the case of indicator 1 and its subcomponents) or remained stable (in the case of indicator 2), showing that the established targets had been reached and that the income of the beneficiary households had, indeed, been partially protected. For indicators associated to the mitigation of the pandemic's negative effects on the accumulation of human capital by poor households, the effects of the pandemic are substantially more negative. The trajectory over the span of the project, starting with the 1st measurement in April 2021 all the way to the current measurement relative to the 2nd half of 2022 (Graph 1), leaves no doubt as to the increase in the number of households kept above the poverty line in all partial measurements as a result of the various waves of expansion: first within the scope of Programa Bolsa Família and, lastly (and particularly), in the last expansion phase under the scope of Programa Auxílio Brasil, starting in November 2021. Graph 1 – Evolution of indicator 1 – “Households kept above the extreme poverty line, as set by the program, with project financing” Page 52 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Indicator 2 showed fluctuations over the span of the project, reaching its lowest value in June 2021, when the pandemic peaked in Brazil (chart 2). The numbers after this period seemed to be evolving positively, but then they declined during the year 2022, remaining below the established final target (55 percent) that year. Note that the numbers for June and November 2022 are the amounts without the extraordinary benefit - i.e., 54.54 percent and 53.32 percent, respectively. However, if we consider the extraordinary benefit as well, the numbers rise to 75.07 percent (June 2022) and 75.84 percent (November 2022), which exceed the established target of 55 percent. Graph 2 – Evolution of indicator 2 - “Support for the average income received from project financing by households as a share of total income at the time of application” After the interruption caused by the COVID-19 pandemic, school attendance monitoring rates for June/July 2022 (Graph 3) revolved around 70 percent (indicator 3), much lower than pre- pandemic monitoring levels, which were close to 90 percent. The indicator also showed that, even after a post-pandemic recovery, such rates have not yet reached the end target set for the project (85 percent), which is one of the challenges posed for the following year (2023). The latest monitoring results (June/July 2022) were undoubtedly strongly impacted by several factors, including 1) an interruption in monitoring for more than 2 years, which certainly disrupted the automated methods for collecting this information at the municipal level and, 2) the innovations introduced by the new Presença system, after reformulation by the MEC/INEP, and 3) the new wave of beneficiaries aged 4 - 5 and 18 - 21 years who also had to be monitored. Graph 3– Evolution of indicator 3 - “Children aged 6 to 17 years who received CCTs from project financing and whose school attendance was monitored by the program (percentage)” Page 53 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Similar to the indicator for monitoring education conditionalities (indicator 3), indicator 4, which is tied to the monitoring of beneficiary children from a health perspective, also remained below the final target of 60 percent for the duration of the project (chart 4). The biggest impact happened in June 2021 - the peak of the COVID-19 pandemic in Brazil - and, after that, despite a clear recovery, pre-pandemic levels have yet to be reached. Graph 4 – Evolution of indicator 4 - “Children under the age of 7 who received CCTs from project financing and whose attendance in health check-ups was monitored by the program (percentage)” Page 54 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Graph 5 shows that the number of beneficiaries of social safety net programs (Indicator 5) follows the growing trend of the number of beneficiaries of Programa Auxílio Brasil (formerly Programa Bolsa Família) during the project, from the first until the final measurement in November 2022. During this period, the program underwent several expansions in the number of beneficiaries, reaching more than 20 million households on the payroll for the month of November 2022, allowing it to exceed its initial target. Graph 5– Evolution of indicator 5 - “Beneficiaries of social safety net programs 4.4. NON-REIMBURSABLE TECHNICAL ADVISORY SERVICES: WORLD BANK STUDIES IN PARTNERSHIP WITH THE IDB AND AFD Partnerships with international organizations entered into by this Ministry within the scope of non-reimbursable technical advisory services have provided input for a discussion on - and review of - the policies offered to the population, such as Programa Bolsa Família/Programa Auxílio Brasil, the country's primary cash transfer program. Based on statistical evidence, academic analyses, comparisons with international experiences and input from behavioral science, the products developed were added to the technical expertise of social units and policies for citizens under the scope of Social Development. Those studies informed discussions about potential changes to several regulations and processes in the Brazilian government and were particularly relevant at times when new policies were introduced. The main products obtained by means of technical advisory services were: o Financial products to protect against income volatility Page 55 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) For this technical assistance endeavor, in addition to the involvement of World Bank teams (particularly the social protection team), specialized consultancy services were also hired at no cost to the Ministry. Studies were conducted on i) Financial Behavior, ii) Financial Education and iii) Financial Citizenship Policies. Prior to his work with the World Bank, Prof. Marcelo Neri had also worked as a consultant with the Inter-American Development Bank (IDB), on the topic of "Bolsa Família and the use of private insurance by the low-income population". This product developed by the IDB was entitled “Bolsa Família and the Use of Private Insurance by the Low - Income Population”. Such products are about offering financial services to the low-income population, and, in the case of products financed by the World Bank, they are mainly about short-term savings and financial education. Quantitative baseline analyses were carried out (POF, PNAD, Findex, Lloyds) of the risks, access and use of financial products, of the economic and behavioral barriers for low-income households and of Programa Bolsa Família itself. Based on a detailed assessment of the financial behavior of households, the products show how savings were gathered and how quality of life has smoothed over time. This work also presented lessons from past experiences in Brazil in such areas as microcredit, financial education and incentives to saving, as well as international experiences involving financial products in risk management geared towards specific target audiences, combined with a review of behavioral literature associated with the topic. Evidence of the results helped inform public policy discussions and proposals to change the design of Programa Bolsa Família, as well as potential enhancements for the new social program. Initially, the studies pointed to limited savings capacity and frequent use of more expensive products and credit, such as overdraft and credit cards. As such, a recommendation was made to expand the financial education strategy for the population covered by Programa Bolsa Família at the time, associating it with the potential for labor formalization and credit history generation. Additionally, it was recommended that the experience with virtual accounts acquired in the Emergency Aid program be used to design behavioral incentives for a savings instrument, associated with a review of the permanence rule for that Program. Improvements were also suggested to the tools used to communicate with the population, including the modernization of the Unified Registry for Social Programs. After delivering the consultancy products and having discussions with the Ministry's technical team, the World Bank developed its own note on the role of financial inclusion in mitigating the income volatility of Unified Registry households. The discussions about this Report and the consultancy products were key in improving the government's new social program. o Revision of the PBF Permanence Rule The permanence rule is a protection mechanism for Programa Bolsa Família beneficiary households that allowed them to continue to receive benefits even when their per capita income exceeded the Program's eligibility threshold. Programa Auxílio Brasil, which replaced Programa Bolsa Família, later introduced the Emancipation Rule. According to this rule, even in the event of an increase in income, the household can still remain in Programa Auxílio Brasil for an additional Page 56 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 2 years, provided that the monthly per capita household income does not exceed two and a half times the poverty line amount - i.e., R$ 525.00. If the household is cancelled because it decides to leave Programa Auxílio Brasil of its own volition (voluntary termination) or because the Emancipation Rule has run its course, the household is allowed to return to the Program as a priority if it meets the requirements for Programa Auxílio Brasil benefits. Households were guaranteed a return to the program if their cancellation from the program was rescinded, which could happen in up to 36 months after the date of cancellation. A World Bank study delivered within the scope of non-reimbursable technical advisory services looked into whether the permanence rule was successful in achieving its objectives, considering the profile of Program beneficiaries who fit the permanence rule, their insertion in the formal job market and the impacts of income volatility. The study also included comparisons with other international experiences and reached conclusions that would later inform decisions about the design of the conditional cash transfer policy. An initial recommendation by the World Bank was to keep the Permanence Rule for variable benefits in the proposal to reform Programa Bolsa Família. The incentive to invest in human capital, which is part of the rationale of variable benefits and the monitoring of conditionalities, increases the chances of entering the labor market and ensuring higher incomes for children when they become adults. With regard to encouraging participation in the labor market, the World Bank study showed that the rate of participation in the labor market among households in the conditional cash transfer program was similar to that of the general population in Brazil, despite the former group facing more restrictions and greater poverty. According to the note by the World Bank, it would be beneficial to deem any income increase within the eligibility criteria as sufficient for the household to qualify for the Permanence Rule. During the period under analysis, only households that voluntarily updated their incomes entered this group. In tandem, the Bank also recommended that the group placed under the Permanence Rule group be provided with more information about this tool – for example, they should be informed of the maximum income threshold to remain eligible for the Program or the opportunities associated with being under the Permanence Rule. The World Bank analysis also suggested that the Guaranteed Return policy be extended to all households placed under the Permanence Rule. Behavioral analyses indicated that doing so would reduce risk aversion, as a guaranteed return to the Program would be the closest thing to “unemployment insurance” available to this segment of the population and would thus mitigate the fear of exploring other possibilities in the formal labor market. Another recommendation in the study was that households with other sources of higher and more stable non-labor income - such as pensions or the Welfare Benefit for the Elderly and Disabled (Benefício de Prestação Continuada, BPC) - should spend less time under the Permanence Rule. If the household configuration changed and it slipped back into poverty or extreme poverty, the Guaranteed Return mechanism would come into effect. Regarding the maximum income for the Permanence Rule, the Bank's assessment suggested that Page 57 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) the amount be linked to Programa Bolsa Família's eligibility thresholds, and not to the minimum wage. According to the study, the Permanence Rule should strike a balance between smoothing the transition of households so that they will not have to experience a significant loss of income when the benefits are rescinded immediately after they start earning more income from work, while also preventing households from spending too much time in this condition and thus preventing other households from joining the Program. It was therefore suggested that the Rule's income threshold be set by the same managers who set other parameters for this benefit. o Studies and research on Programa Auxílio Brasil (PAB) Two studies and surveys were conducted on PAB beneficiaries, focusing on (a) beneficiaries' length of stay in the program and (b) beneficiaries' insertion in the labor market: o “Duration of Stay in Bolsa Família and Determining Factors: initial results”: this study was presented to the MC technical team in March 2022 and, among other findings, showed that between 2012 and 2019 beneficiary households had remained in the program for an average of 5 years. During that time, 49 percent of beneficiary households left the program and 51 percent either stayed or returned to it once or twice. o “Labor Profile of Low-Income Households in Brazil and Economic Inclusion Policies (Perfil de Trabalho das Famílias de Baixa Renda no Brasil e Políticas de Inclusão Econômica)”: this study was presented to the MC technical team in May 2022 and, among other findings, showed that, according to the National Survey by Continuous Household Sample (Pesquisa Nacional por Amostra Domiciliar Contínua, PNDC), 77 percent of men and 41 percent of women beneficiaries of working age were indeed working, broken down as follows: 40 percent were informal employees, 32 percent were self-employed, 20 percent were formal employees and 7 percent were self-employed and had their own formal businesses. Finally, the Note from the World Bank recommended that all the actions mentioned above be associated with initiatives that promote the formalization and financial education of this population. Productive inclusion programs, for example, could focus on this population. Once again, the recommendations about this instrument helped improve the proposal for the new social program. o Impact Assessment of the Programa Bolsa Família Expansion For this study, the World Bank team worked in partnership with the French Development Agency (Agence Française de Développement, AFD) to develop a methodology to assess the impact of Programa Bolsa Família / Programa Auxílio Brasil on the protection of human capital and the use of health and education services during the pandemic. This work can influence government decisions on whether to expand the Program to preserve human capital and avoid negative behaviors, in addition to pointing out possibilities and challenges in the reapplication of Bolsa Família conditionalities, and the impact of the Program even without conditionalities. The Methodological Note for this product was prepared by the Page 58 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) World Bank to provide greater detail about the survey format. The World Bank team prepared a questionnaire, which was originally to be applied by call center operators, via the Ombudsman's Office and within the scope of this Ministry's contract. Faced with budget constraints, however, the Ombudsman's Office stated that there was no budget available for the Ministry's own staff to make the calls or to open service orders to apply the questionnaires under the scope of the Ministry's contract. As a result, Senarc, Conjur and the World Bank team devised strategies that would allow the study to be conducted and signed a Data Licensing Agreement for specific samples and in accordance with the General Personal Data Protection Law (Lei Geral de Proteção de Dados Pessoais, LGPD) so that the World Bank itself could perform the survey with its own contract and funding. The survey results had not yet been released at the time of this writing. They are expected to be released in the 2023 financial year. o Systematization of the Emergency Aid (Auxílio Emergencial, AE) Experience During the time encompassed in this Report, the team hired by the World Bank conducted an analysis and systematization of the entire Emergency Aid process, including policy design; eligibility criteria; ways to access the benefit; eligibility checks; payment methods; error and fraud control; grievance mechanisms; communication with beneficiaries; regulation of the use of databases; privacy control; coordination with other policies. The proposal was to have a case study to disseminate the lessons learned, especially amongst partners in Latin America, the Caribbean, Africa and Asia, in a way that would be both accessible and useful. At the time, interviews were conducted with the technical team and with the Ministry's top echelons, and the consultants worked with the documents they had received and the material they had collected in their interviews. o Simulated changes to Bolsa Família After working to develop educational incentive design options for PBF based on international experience, the World Bank team provided support to the National Executing Agency in the form of simulations of the impact of potential Program benefit reforms on poverty, equity and targeting. Such studies informed the discussions about revising the PBF and drafting the proposal for the new social program enacted by the Brazilian government. 4.5. OTHER ACTIONS PRODUCED BY THE PARTNERSHIP The topic of Adaptive Social Protection was presented at a meeting on emergency and calamity scenarios held on July 18, 2022, with the World Bank. The results of the Stress Test were also presented, which was a study carried out by the Bank based on the Disasters and Calamities Risk Index (Índice de Risco à Desastres e Calamidades, IRDC). This Ministry, represented by the governance staff in office until the end of the 2022 financial year - the Executive Secretary, Deputy Secretary of the Special Secretariat for Social Development, and Page 59 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) the National Secretaries in attendance (SECAD, SEISP, SENARC and SNAS) -, outlined the priorities of the administration and actions on this topic currently underway with local governments, as well as social protection policies directed at the population. The mission helped with internal discussions and informed the Bank about the types of assistance currently offered to households affected by disasters and calamities. Such input would also be used in future discussions about the enactment of a service protocol, which could perhaps be the subject of a potential new partnership with the Bank. In future projects, this and other topics may be addressed in light of new priorities set by the MDS when supporting quick responses to disasters and calamities by local teams. 5. ASSESSMENT OF THE MINISTRY'S PERFORMANCE DURING THE PREPARATION AND IMPLEMENTATION OF THE OPERATION 5.1. LESSONS LEARNED The Ministry of Citizenship, as it was then called, in partnership with the Ministry of Economy, known today as the Ministry of Finance, were faced the atypical situation of having to prepare a credit operation with not just one financier, but rather five Development Banks. These partners came into alignment vis-à-vis the rules and guidelines that would guide the implementation of associated projects. Under the leadership of the World Bank, which played a key documentary role, each of them contributed with its share of resources to ultimately benefit a portion of the beneficiaries who were already served by the Federal Government. They also helped expand the network of beneficiaries when new households came to be assisted by social programs after the onset of the pandemic. All the Government's and the banks' project teams went above and beyond the line of duty in view of the urgency involved and the need to expedite the bureaucratic procedures inherent to pre- negotiations, negotiations and approvals by all agencies inside and outside the Ministry of Economy (as borrower) and the Ministry of Citizenship (as executor). This joint effort culminated in key products to mitigate the effects that the pandemic was causing and would still cause. Government officials met using remote work platforms to ensure deadlines were met and to draft as many minutes as possible. That was the only way to obtain all the approvals and sign all the agreements with the financiers. Throughout the implementation of the COVID-19 Emergency Program, the challenges faced, and progress achieved were gauged every six months and compiled in half-yearly progress reports submitted, during the term of the agreement, to the World Bank and to the other funders. In the first year of the pandemic, large AE transfers - along with other measures such as the expansion of then-called Programa Bolsa Família - were able to partially offset the drop in labor income. In fact, the poverty rate in 2020 was lower than in the year before the pandemic: in 2020, 13.1 percent of the population lived on less than US$ 5.50 per day (PPP); in 2019 it was 20.6 percent. The weak labor market recovery in 2021, however, was unable to compensate for the diminished role of emergency programs in income support. World Bank estimates based on data Page 60 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) recently released by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatistica, IBGE) suggest that, in 2021, 18.7 percent of the population was in poverty (US$ 5.50 PPP). The monitoring of conditionalities was made more difficult by pandemic, particularly in education, where face-to-face school activities had been suspended in most state and municipal education networks, which made it impossible to monitor school attendance. Consequently, the Ministry of Education (MEC) determined that it would not be possible to monitor school attendance during the pandemic (which spanned the entire period covered in the last report for 2021). The effects of school closures surfaced when schools reopened in 2022: namely, a significant increase in school dropouts when in-person classes resumed, mainly in low-income households and, in particular, amongst their youth (the program's target audience). The pandemic had severe effects on human capital formation. School closures had an impact on the educational trajectory of most students, especially the most vulnerable. Brazil was one of the countries that kept schools closed the longest, for a total of 78 weeks. The pandemic has undermined learning and caused more students not to return to school. Evidence shows learning loss at all levels of education. Many students dropped out of school during the pandemic: There were 579 thousand fewer enrollments in basic education in 2020 than in the previous year. The problem of school dropout tends to be more serious among older and more vulnerable students, where there is greater pressure to look for a job when household income drops. Furthermore, according to a recent telephone survey conducted by the World Bank to monitor well-being in Brazil, in the first half of 2021 households continued to experience greater food insecurity than before the pandemic, and children continued to use health services less frequently. Dropping out of school is not detrimental to just children, it also affects women in the labor market. The drop in enrollment rates and the closing of schools meant that women had to spend more time at home to care for their children, thus remaining outside the labor market. In fact, 43 percent of inactive women in the potential workforce were not looking for a job in the last quarter of 2021 due to household chores and caring for children and relatives. In 2022, education conditionalities experienced technical and operational inconsistencies in Sistema Presença. As a result, the Ministry of Education (MEC) and the Anísio Texeira Institute of Educational Studies and Research (Instituto de Estudos e Pesquisas Educacionais Anísio Texeira, Inep) decided to disregard monitoring in the first term and, therefore, the repercussions for non- compliance with conditionalities imposed in May 2022 were annulled; as there were no results to consider, it was impossible to determine whether the beneficiaries had met the conditionality criteria. In September 2022, repercussions were imposed based on information from the 3rd monitoring term (June and July) for the education conditionality, and in the 1st half of 2022 for non-compliance with health conditionalities. In September 2022, all households in which one or more members had failed to comply with the education and/or health conditionalities for reasons that caused repercussions were issued a warning. Page 61 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) In health, the monitoring of conditionalities associated to vaccination and anthropometric measurements (weight and height) of children and the monitoring of women were never completely interrupted. However, the Ministry of Health (Ministério da Saúde, MS) made it optional for municipalities to record monitoring data on health conditionalities. Meanwhile, the Ministry of Health also advised municipalities to keep the identification of pregnant women and prenatal care cases a priority, so that the Variable Benefit to Pregnant Women (Benefício Variável à Gestante, BVG) could continue to be paid out. These same procedures and guidelines were repeated in the two subsequent terms: the 2nd half of 2020 and the 1st half of 2021. The Ministry of Health decided to once again make it mandatory to record monitoring data on health conditionalities in the 2nd half of 2021. There was no change in the prioritization system in effect for the registration of indigenous households into Programa Bolsa Família and its successor, Programa Auxílio Brasil. Two surveys were conducted to collect local experiences and identify actions that could facilitate access to social assistance benefits by indigenous households. Project Supervision Missions discussed the need to implement actions to raise engagement and commitment to serving Traditional and Specific Population Groups (Grupos Populacionais, Tradicionais e Específicos, GPTE), especially indigenous peoples. This Ministry's General Ombudsman's Office, in turn, received a record amount of feedback via its four official service channels: electronic form, telephone, letter and face-to-face. The program's scope included the establishment of indicators tied to the mechanisms for responding to complaints from Poor Populations Affected by COVID-19. As agreed, the indicators covered Programa Bolsa Família (PBF) complaints resolved within 30 days, broken down into indigenous peoples, quilombolas and other groups. It should be noted that when the indicators were first set, requests were received via two systems: the Request Management System (Sistema de Gestão de Demandas, SGD) and the Integrated Ombudsman and Information Access Platform – Fala.BR. In 2022, the Fala.BR Platform became the only channel to submit requests, giving rise to new challenges in extracting and tracking data, as a given request might be processed by multiple departments before finding its way to this Ministry, which impacts the average response time. In order to better ascertain the number of services provided to people with disabilities, the General Ombudsman's Office requested that the Federal Ombudsman's Office (Ouvidoria-Geral da União, OGU) improve the Fala.BR tool and enable reports to be generated about those segments of the population. The public ombudsman function is regulated by OGU/CGU Ordinance 581 of 2021, which states that requests can be addressed in up to 30 days, extendable for an additional 30 days upon reasonable justification. Due to their complexity, many of the requests received required more than 30 days to be addressed, and this would not be treated as non-compliance with legally established deadlines. The reach of the agreed-upon indicators was strongly influenced by a more than 1,400 percent increase in the number of requests compared to 2019 and 2020. This increase was prompted by requests about the Emergency Aid, COVID-19 and the Unified Registry. Page 62 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) The new Emergency Aid scheme was confirmed by Provisional Presidential Decree 1,039 in 2021 and, once payments began, there was a 166.17 percent in the number of requests received by the Ombudsman's Office in April compared to March. There was a significant improvement in the reach of the indicators in 2022 due to the gradual drop in the number of requests at the end of 2021. As such, the availability of categorized information that can be easily measured and monitored over the period must be considered when the indicators are first set. Furthermore, such indicators must comply with the governance and regulatory structure in place for the sector - in particular the deadlines and procedures for addressing the requests sent to the Ombudsman's Office. Throughout the Project's implementation, the commitments signed in the Project's Environmental and Social Management Plan were completely fulfilled vis-à-vis the agreed-upon actions and deadlines. 6. FINANCIAL MANAGEMENT OF THE IBRD PROJECT The Loan was denominated in US dollars with percentages for the PBF Transfers component and rates provided for in accordance with the general conditions of the Agreement. Category Percentage of Expenditures to be Loan Reimbursed or Financed (including Amount taxes) Allocated (USD) (1) PBF Transfers for 100% $997,500,000 Part 1 of the Project Amount payable under Section 2.03 of $2,500,000 (2) Front-end Fee this Agreement, pursuant to Section 2.07(b) of the General Conditions (3) Interest rate premium Amount due under Section 4.05(c) of the 0 Interest rate caps and General Conditions ranges Total amount $1,000,000,000 In line with the Operational Manual (OPM) agreed upon for the Project, SENARC's General Budgetary and Financial Execution Coordination produced all Financial Reports in a timely fashion. These were then submitted with comments to the National Treasury Secretariat (Secretaria do Tesouro Nacional, STN), which, in turn, issued formal disbursement requests to the Bank. Such requests (known as IFRs) were prepared in accordance with the model agreed upon with the Bank, in Brazilian Reais and US Dollars, based on information from SIAFI and Programa Bolsa Família and Programa Auxílio Brasil monitoring reports. For financial management reasons, the Page 63 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Ministry of Economy - by means of the National Treasury Secretariat (STN) - issued half-yearly consolidated IFRs and sent them to the Bank within the period set in the OPM, in up to 60 days after the end of each semester. The difference between the number of beneficiary households in Programa Bolsa Família, March 2020 payroll, and the payroll used as a reference for requesting each disbursement was considered an eligible expense, thus comprising the totality of households that entered the program after the pandemic began. Table 1 (below) shows the budgets of Programa Bolsa Família (in effect until October 2021) and Programa Auxílio Brasil. Table 1 - Budget approved for the years 2020 to 2022 for Programa Bolsa Família and Programa Auxílio Brasil LOA 2020 R$ 29,484,920,000.00 LOA 2021 R$ 34,858,183,958.00 LOA + credit 2022 R$ 114,521,663,500.00 The Bank made 2 disbursements. The first disbursement took place on August 16, 2021, and covered expenses incurred with Programa Bolsa Família benefits paid out before the loan agreement was signed; these were deemed expenses eligible for reimbursement under the retroactivity clause of the loan agreement and were equivalent to 20 percent of the total amount of the credit operation. The Bank's second and final disbursement was against expenses incurred in the months of June 2021 to March 2022; it was made effective on May 3, 2022, in the amount of U$ 797,500,000.00 (seven hundred and ninety-seven million five hundred thousand dollars). Page 64 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Proof of the disbursement operation is available at Client Connection, as shown in the statement below. It should also be noted that the amount borrowed from the IBRD was fully repaid, as follows: U$ 200,148,858.00 on August 16, 2021, and U$ 797,500,000.00 (seven hundred and ninety-seven million and five hundred thousand dollars) on May 03, 2022. The Loan Agreement signed with IBRD, known as CFA 011395, was audited by the Federal Comptroller General (Controladoria Geral da União, CGU). The CGU assessed the key accounting, financial and managerial clauses and found the Project executor to be in compliance. Disbursement request reports were submitted to the Bank and were supported by sufficient eligible expenditures to cover the disbursed amounts. The controls were also found to be adequately compliant with the conditions laid out in the Loan Agreements. 7. PROJECT SUSTAINABILITY Now that the Income Support for the Poor Affected by COVID-19 Project has been fully implemented, the results of the actions - conducted in partnership by the World Bank's Social Protection team and specialists from this Ministry - show multiple opportunities for improvement. In terms of Project sustainability, many of PAB's new benefits and complementary programs still lack a more detailed design and proper implementation support. In addition, benefits that include Page 65 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) conditionalities for age groups will require the implementation of specific arrangements to monitor them, and a balance will need to be struck between providing incentives and maintaining coverage. The Project's notes and contributions will inform the discussions and decisions about continuing improvements and enhancements in social protection policies put in place by the Ministry. The technical assistance provided by the World Bank's Social Protection team enabled access to international experiences and specialized consultancy services funded by the Bank, at no cost to the Ministry, offered in quantities that would not have been possible unless tied to a credit operation. This led to inputs that can be used in future adaptations of the Conditional Cash Transfer Program and the Unified Registry of Social Programs used by the Federal Government to collect data and select beneficiaries. These inputs still need to be organized, monitored and assessed following the end of the Project. There was no shortage of challenges in the transition from a consolidated program like Programa Bolsa Família to a new program like Programa Auxílio Brasil. Many challenges remain for Programa Auxílio Brasil to overcome going forward. They include: maintaining zero backlog, implementing new benefits, inter-sector and federative coordination, resumption of monitoring and imposition of repercussions for missed conditionalities and the new registry project. 8. AUDIT The Federal Comptroller General (CGU) is currently auditing the expenses incurred in the 2022 fiscal year, which was when the Project's second and final disbursement took place. 9. KEY TEAM: Individuals responsible for the project 9.1 MINISTRY OF FINANCE National Treasury Secretariat Rafael Mesquita Camargo: rafael.camargo@tesouro.gov.br; Borrower's representative; Financial Management; and Disbursements. 9.1. MINISTRY OF DEVELOPMENT AND SOCIAL ASSISTANCE, FAMILY AND FIGHT AGAINST HUNGER General Coordination of the Project: Silvânia Freire de Queiroz: cgct@cidadania.gov.br National Secretariat of Citizenship Income Financial Management: Sérgio Monteiro: sergiomonteiro@cidadania.gov.br Environmental and Social Management: Maria Carolina Prado Lage; Sheila Benjuino de Carvalho: sheila.carvalho@cidadania.gov.br Page 66 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Management Monitoring and Evaluation (indicators): Hugo Miguel Pedro Nunes: hugo.nunes@cidadania.gov.br; Evaluation, Information Management and Unified Registry Secretariat Unified Registry Indicators: Jennifer Carla: jennifer.carla@cidadania.gov.br ; Environmental and Social Management Support (GPTE-Indígenas): José Roberto Frutuoso: jose.frutuoso@cidadania.gov.br General Ombudsman Grievance Mechanism Indicators and Support for Environmental and Social Management: Moises Paes Landim Placido: moises.placido@cidadania.gov.br 10. ANNEX 1) Results Indicator Matrix Marco de Resultados Brasil: Apoio à Renda dos Pobres Afetados Pela COVID-19 no Brasil Objetivo(s) de Desenvolvimento do Projeto O Objetivo do Projeto é mitigar os efeitos econômicos negativos associados ao COVID-19 sobre a renda e os riscos de prejudicar o capital humano das famílias pobres. Indicadores de Objetivo de Desenvolvimento do Projeto Relatório Final Dezembro 2022 - Relatório Julho 2021 / Ref: folha Relatório Final Fev 2022 / Ref: Relatório Final jul 2022 / Ref: Relatório Final Dezembro 2022 / Nome do Indicador Linha de base Meta Final Ref.: Folha pagto PBF ABRIL 2020 Atualização Maio 2023 / Ref: PBF junho 2021 folha PBF Dezembro 2021 folha PAB junho 2022 Ref: folha PAB novembro 2022 folha PAB dezembro 2022 Mitigar efeitos econômicos negativos associados ao COVID-19 sobre a renda de famílias pobres 1. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto (Número) 0.00 1.200.000 1.200.000 1.855.924 1.875.736 6.269.044 9.959.838 10.195.769 (1) 1.a. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Mulheres Indígenas Female 0.00 1.800 1.883 8.099 16.516 26.833 39.984 40.823 (1) (Número) 1.b. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Homens Indígenas (Número) 0.00 400 492 2.908 6.400 12.691 24.190 25.008 (1) 1.c. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Mulheres Quilombolas 0.00 1.100 1.118 5.569 13.628 24.492 35.365 35.969 (1) (Número) 1.d. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Homens Quilombolas (Número) 0.00 400 463 2.563 6.959 15.431 28.583 29.549 (1) 1.e. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Mulheres de Outros Grupos 0.00 963.300 990.557 1.461.188 1.461.676 4.701.667 7.078.828 7.221.621 (1) (Número) 1.f. Famílias que são mantidas acima da linha da extrema pobreza, conforme definida pelo programa, com financiamento do projeto- Homens de Outros Grupos 0.00 233.000 244.528 372.997 370.557 1.487.930 2.752.888 2.842.799 (1) (Número) 54,54% [exclusive benefício 53,32% [exclusive benefício 52,76% [exclusive benefício 2. Apoio à renda média recebida com financiamento do projeto por famílias como proporção da renda no momento da solicitação (Porcentagem) 0.00 55% 55% 51,74% 56,43% extraordinário] extraordinário] extraordinário] (1) 75,07% [inclusive benefício extraordinário] 75,84% [inclusive benefício extraordinário] 75,53% [inclusive benefício extraordinário] Mitigar os riscos de efeitos econômicos negativos decorrentes da COVID-19 no capital humano de famílias pobres 88,48% 70,56% 78,05% 3. Crianças entre 6 e 17 anos que se beneficiam de CCT com financiamento do projeto cuja frequência escolar está sendo monitorada pelo programa(Porcentagem) 0.00 85 - - - (2) [SICON-média anual 2019] [SICON-3º período de 2022] [SICON-5º período de 2022] 3.a. Crianças entre 6 e 17 anos que se beneficiam de CCT com financiamento do projeto cuja frequência escolar está sendo monitorada pelo programa- Indígenas 90,03% 63,32% 73,25% 0.00 85 - - - (2) (Porcentagem) [SICON-média anual 2019] [SICON-3º período de 2022] [SICON-5º período de 2022] 3.b. Crianças entre 6 e 17 anos que se beneficiam de CCT com financiamento do projeto cuja frequência escolar está sendo monitorada pelo programa- Quilombolas 90,62% 76,30% 84,34% 0.00 85 - - - (2) (Porcentagem) [SICON-média anual 2019] [SICON-3º período de 2022] [SICON-5º período de 2022] 3.c. Crianças entre 6 e 17 anos que se beneficiam de CCT com financiamento do projeto cuja frequência escolar está sendo monitorada pelo programa- Outros 91,06% 70,61% 78,06% 0.00 85 - - - (2) grupos (Porcentagem) [SICON-média anual 2019] [SICON-3º período de 2022] [SICON-5º período de 2022] 45,78% 45,78% 45,32% 54,45% 4. Crianças com menos de 7 anos de idade que se beneficiam de CCT com financiamento do projeto e cuja conformidade com checkups de saúde está sendo monitorada 65,54% [Sicon - Resultado global - 2º período de 0.00 60 21,77% [Sicon - Resultado global - 2º período de [Sicon-Resultado global - 1º período de [Sicon-Resultado global - 2º período de (3) pelo programa (Porcentagem) [SICON-média anual 2019] 2021] 2021] 2022] 2022] 44,37% 44,23% 45,24% 57,40% 4.a. Crianças com menos de 7 anos de idade que se beneficiam de CCT com financiamento do projeto e cuja conformidade com checkups de saúde está sendo 67,08% 0.00 60 27,61% [Sicon - Resultado global - 2º período de [Sicon - Resultado global - 2º período de [Sicon-Resultado global - 1º período de [Sicon-Resultado global - 2º período de (3) monitorada pelo programa - Indígenas (Porcentagem) [SICON-média anual 2019] 2021] 2021] 2022] 2022] 59,35% 59,46% 59,81% 69,67% 4.b. Crianças com menos de 7 anos de idade que se beneficiam de CCT com financiamento do projeto e cuja conformidade com checkups de saúde está sendo 63,25% 0.00 60 30,46% [Sicon - Resultado global - 2º período de [Sicon - Resultado global - 2º período de [Sicon-Resultado global - 1º período de [Sicon-Resultado global - 2º período de (3) monitorada pelo programa - Quilombolas (Porcentagem) [SICON-média anual 2019] 2021] 2021] 2022] 2022] 45,45% 45,67% 45,18% 54,26% 4.c. Crianças com menos de 7 anos de idade que se beneficiam de CCT com financiamento do projeto e cuja conformidade com checkups de saúde está sendo 64,31% 0.00 60 21,73% [Sicon - Resultado global - 2º período de [Sicon - Resultado global - 2º período de [Sicon-Resultado global - 1º período de [Sicon-Resultado global - 2º período de (3) monitorada pelo programa - Outros grupos (Porcentagem) [SICON-média anual 2019] 2021] 2021] 2022] 2022] Indicadores de Resultados Intermediários por Componentes Relatório Julho 2021 / Ref: folha Relatório Final Fev 2022 / Ref: Relatório Final jul 2022 / Ref: Relatório Final Dezembro 2022 / Relatório Final Dezembro 2022 / Nome do Indicador Linha de base Meta Final Ref.: JANEIRO/2021 PBF junho 2021 folha PBF Dezembro 2021 folha PAB junho 2022 Ref: folha PAB novembro 2022 Ref: folha PAB dezembro 2022 Transferências de Renda Condicionada para famílias na pobreza e extrema pobreza 5. Beneficiários de programas de rede de segurança social (CRI, Número) 0.00 2.900.000 2.900.000 4.394.542 4.470.161 13.680.084 20.252.936 20.671.499 (4) 5.a. Beneficiários de programas de rede de segurança social – Mulheres (CRI, Número) 0.00 2.320.000 2.320.000 2.553.867 2.597.938 7.906.624 11.556.819 11.792.912 (4) 6. Casos do mecanismo de reparação de queixas do BF resolvidos em 30 dias (Porcentagem) 0.00 60 100 100 95,28% (5) 6.a. Casos do mecanismo de reparação de queixas do BF resolvidos em 30 dias - Indígenas (Porcentagem) 0.00 60 100 100 100% (5) 6.b. Casos do mecanismo de reparação de queixas do BF resolvidos em 30 dias - Quilombolas (Porcentagem) 0.00 60 100 100 100% (5) 6.c. Casos do mecanismo de reparação de queixas do BF resolvidos em 30 dias - Outros grupos (Porcentagem) 0.00 60 100 100 95,26% (5) 7. Adoção e aperfeiçoamento de estratégias de comunicação com canais específicos para grupos vulneráveis do BF que aumentam o conhecimento da existência do Fala Não Sim Sim Sim Sim BR para queixas relacionadas com o BF (Sim/Não) (1) Conforme acordado com o Banco Mundial, foram consideradas, no cálculo dos indicadores, as famílias que entraram no Programa Bolsa Família a partir de março de 2020. Em comparação à folha de pagamentos PBF de JUNHO/2022, esse número totalizou 6.269.044 famílias. Esse aumento substancial é consequência do número de famílias na folha de Junho/2022 (18.154.897 famílias), o qual é significativamente maior que o total de famílias na folha de Março/2020 (13.058.228 famílias). E na folha de NOVEMBRO/2022 o indicador reflete outra expansão do PAB, atingindo 21.534.293 famílias, resultando em uma diferença comparativa de 9.959.838 famílias em relação a março de 2020. (2) O cálculo dos indicadores relativos às condicionalidades da educação não foi possível de ser realizado devido à suspensão do acompanhamento da frequência escolar, por parte do Ministério da Educação, causada pela pandemia do Covid-19 instalada no Brasil desde março de 2020, mês, a partir, do qual foi descontinuado o acompanhamento. Neste relatório apresenta-se os dados do primeiro acompanhamento realizado pós-pandemia, referente ao 3º bimestre de 2022 (julho/julho 2022), bem como, o último período de 2022, referente ao 5º bimestre (outubro/novembro). (3) Vale mencionar os dados de acompanhamento no período afetado pela situação pandêmica já demonstram recuperação ao longo dos dois períodos de 2022, no entanto, ainda sem antingir os patamares pré-pandemia. (4) Utilizou-se o mesmo critério do indicador 1. O indicador apresenta o número de beneficiários (pessoas) que compõem as 10.195.769 famílias descritas no indicador 1. O indicador 5a utiliza o mesmo raciocínio restrigindo o número de beneficiários apenas às mulheres. (5) O número de casos (atendimentos) e a percentagem de deles atendidos em 30 dias considera somentes aqueles recebidos pela plataforma Fala.br durante o periodo de janeiro a junho de 2022. Page 67 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 6. SUPPLEMENTARY INFORMATION ON SELECTED SOCIAL PROTECTION PROGRAMS 1. This annex provides additional information on BF, its successor programs, AB and the new BF, as well as AE and BE. Figure 6.1 summarizes the timeline of the benefits discussed in this section. Figure 6.1. Timeline of Main Social Assistance Benefits Discussed in this Annex January March April December August 2020 March 2004 2020 2021 2023 2023 Bolsa Família Bolsa Família Auxílio Brasil New BF replaces Expansion Auxílio Emergencial replaces BF, Beneficio AB and its top-up first roll out Complementar (for new poor and + Benefício (replaces Ben. top-up to BF) Extraordinário Extraordinario) top-up to AB 2. In April 2020, the GoB introduced a new benefit, the Auxílio Emergencial (AE), to mitigate the social and economic impacts caused by the COVID-19 pandemic. AE provided cash transfers to people outside formal wage employment but not eligible for social assistance, such as informal workers and self- employed individuals. The benefit covered adults not receiving a formal source of income or pension and declared to be either self-employed, unemployed, or informal in a household with earnings less than 0.5 times the minimum wage per capita or living in a household with a total income three times below the minimum wage. The AE provided five monthly transfers of R$ 600 (US$114) and four monthly payments of R$ 300 (US$57) to a maximum of two eligible adults in households without a formal income. Single mothers were eligible for a double benefit, and an exception was made to the age requirement of 18 years, so that under-age mothers in single-parent households could also qualify for the benefits. BF beneficiaries were automatically enrolled to receive a portion of AE on top of BF, up to the amount they would have received without BF (which was the case for about 90 percent of families in BF). AE ended in October 2021 after 16 months of operation, only interrupted between January and March 2021, and at its peak, it reached nearly half of Brazilian families. 3. While working on the creation of AE, in March 2020, the Government chose to expand the BF program. At the time of appraisal, BF had long-standing, well-established, and effective operational and institutional arrangements. The PAD and POM contain a detailed description of the BF delivery process.34 Since Cadastro Unico’s creation in 2001, the GoB had also invested in continuous improvements in terms of policy and legal framework, procedures, and technology, including with nearly continuous World Bank support. Before the pandemic, Cadastro Único contained information on about 26 million families and 40 34 Municipalities are responsible for the identification of families and their registration in Cadastro Único through Social Assistance Reference Centers. The MoC assesses the eligibility of the applications to the BF based on information declared in CadU and after verification in other public registries. After the benefit is authorized, the Caixa Economica Federal makes the payments, which are withdrawn by beneficiary families using a dedicated debit card. Once they are enrolled, families must comply with the education and health co-responsibilities to continue benefitting from the BF. These are verified through the ‘Siscon’, an inter- ministerial database that periodically cross-checks information on school attendance and health check compliance and shares them directly with the MoC. Cadastro Único is Brazil’s National Social Registry. Page 68 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) percent of the Brazilian population. During the pandemic, BF suspended the application of sanctions for noncompliance with conditionalities. Such suspension was already foreseen in the original 2003 law of the BF program and later reprised in the 2021 AB Law. It relieved beneficiaries from interruption of benefits in case of unexpected interruption or unavailability of health and education service provision and eliminated the financial penalties for municipalities normally in place for failing to track compliance. The suspension was renewed several times35 and ended in April 2022. 4. With the pandemic coming to an end, AE was phased out in November 2021, and, as part of this transition, the BF program was replaced in December 2021 with the creation of the AB. AB had very similar goals to BF, but it exhibited new parameters that made it larger by design and with a few design optimizations. The main changes included an upward revision in the poverty and extreme poverty threshold lines and the generosity of benefits; a simplification of the benefit structure, with the elimination of the basic benefit, absorbed by a higher ‘benefit to overcome extreme poverty’; the extension of a variable benefit for youth between 18 and 20 years with a new conditionality of being enrolled in secondary school or having completed it; and extension of a conditionality for children ages 4−5 to attend preschool if locally available. The higher eligibility threshold allowed AB to absorb many of the new poor generated by the pandemic, formerly receiving BE, in addition to carrying over all the beneficiaries of BF. Figure 6.2. Beneficiary Families in BF and its Successor Programs (January 2020 −May 2023) 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 Bolsa Família only Auxílio Brasil only Auxílio Brasil + Benefício Extraordinário Auxílio Brasil + Benefício Complementar Bolsa Família + Auxílio Emergencial Source: Authors based on data published by the Ministry of Social Development in June 2023. The data for the number of families receiving only AB but not the Beneficio Complementar top-up between July 2022 and February 2023 is not available, but this is expected to be very low, given that very few families would receive more than R$ 600 from AB without the support of Beneficio Complementar (AB average benefit is about R$ 220). 35 portaria mc nº 682, de 6 de outubro de 2021. Page 69 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) 5. In December 2021, the Provisional Measure no. 1.076 (converted into Law no. 14.342/2022) introduced the BE for families enrolled in AB. This benefit provided an additional cash payment so that at least R$ 400 would be transferred to each beneficiary family. It is worth noting that out of the 14.5 million families enrolled in the AB, 13 million previously received less than R$ 400. (BE1) A second Constitutional Amendment no. 14,284 introduced Beneficio Complementar, right at the time when BE was expiring and with a new design that would ensure all families enrolled in AB receive from August 1, 2022, to December 31, 2022, a minimum transfer of R$ 600, by providing a top-up to AB for the equivalent amount. The value of the transfer was an important benchmark for the incoming government, which set an average benefit value for the new BF of at least R$ 600 per family. 6. In March 2023, the incoming government launched the new BF program, replacing AB. The new BF has introduced guarantees that no family will receive less than R$600 in benefits and raised the income eligibility threshold to R$218 per capita. It also includes the Protection Rule (Regra de Proteção), allowing uninterrupted financial support for up to 24 months with 50 percent of the original benefit amount to beneficiaries whose income goes above the threshold up to half a minimum wage. The program also reenforces the focus on early childhood development by expanding the Early Childhood Benefit (Benefício Primeira Infância) from 0 to 3 to 0 to 6 years and increasing the benefit amount from R$130 to R$150. Table 6.1 compares the main parameters of BF, AB, and the new BF program. Table 3.1 Eligibility Criteria and Program Rules for BF, AB, and New BF Threshold Monthly Program Benefit Component (Per Capita Conditionalities Benefit in R$ Income) Children between 0 and 15 Education: 41 178 years Children between 6 and 17 years Mother of children between need to be enrolled in school. 41 178 0 and 6 months School attendance must be at Pregnant woman 41 178 least 85 percent for 6–15 years and 75 percent for 16–17 years. Health: Children under 7 years need to BF Adolescents ages 16–18 48 178 get recommended vaccines. years Weight, measure, and growth must be monitored every 6 months. Basic benefit 89 89 Gap with Families in extreme poverty Overcoming poverty extended 89 poverty line Early Childhood Benefit (BPI) Poverty Health: (R$210) or Children under 7 years need to Children between 0 and 3 extreme get recommended vaccines. 130 years old poverty Weight, measure, and growth AB (R$105) must be monitored every 6 months. Family composition benefit Poverty Education: 65 (BCF) (R$210) or Page 70 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) Threshold Monthly Program Benefit Component (Per Capita Conditionalities Benefit in R$ Income) Payments made to each extreme Children between 4 and 17 years member (up to 5) who falls poverty need to be attending basic into the following category: (R$105) education. Pregnant women, nursing Youth 18 and 20 years need to be mothers, children between enrolled in secondary school or ages 3 and 17, and youth having completed it. ages 18 to 20 Health: Pregnant woman needs prenatal care. Nursing mothers and children 0−6 need to be under health monitoring. Benefit for overcoming extreme poverty (BSP) Extreme poverty Payments made to each 65 Equal or Families in extreme poverty family member with no lower than limitation on the number of R$105 family members New BF Each family member The conditionalities were mainly 142 (made a regardless of age kept as before: commitment Additional amount for that no families with children 0−6 150 per child Education: Children 4−5 years, a family will years minimum school attendance of receive less Additional amount for 60% and 75% for beneficiaries 50 than R$600) pregnant woman ages between 6 and 18 years who 218 have not completed basic education. Additional amount for families with children and 50 Health: Prenatal care, follow the adolescents 7−18 years national vaccination calendar, monitor children’s nutrition for those younger than 7 years. Page 71 of 71 The World Bank Brazil: Income Support for the Poor affected by COVID-19 (P174197) ANNEX 7. SELECT ADVISORY SERVICES AND ANALYTICS 1. This annex provides a selection of published analytical products produced by the Social Protection and Poverty Global Practices during project execution and previous years. 2. The select studies that preceded the project on Bolsa Familia and Cadastro Único are as follows: • De la Brière, Benedicte; and Kathy A. Lindert. 2005. “Reforming Brazil’s Cadastro Único to Improve the Targeting of the Bolsa Família Program.” World Bank, Social Protection Unit and DFID. • Lindert, Kathy; Anja Linder; Jason Hobbs; and Benedict de La Briere. 2007. “The Nuts and Bolts of Brazil’s Bolsa Família Program: Implementing Conditional Cash Transfers in a Decentralized Context.” World Bank Social Protection Discussion Paper 709. • Morgandi, Matteo; Liliana D. Sousa; Alison Farias; and Fabio Cereda. 2019. “The Reform of Bolsa Familia: Evaluation of the Proposals Debated in 2019.” Social Protection and Jobs Discussion Paper 2009, World Bank Group, Washington, DC. 3. Select studies produced for the Government of Brazil during the Project: • Ed, Malín; Alejandro Gonzalez; and Renata Ferreira. 2023. Towards Shock Responsive Social Protection in Brazil: Assessing the Adaptiveness of Brazil’s Social Protection System. Washington, DC: World Bank. • Fietz, Katharina Maria; Tiago Falcao Silva; Maria Concepcion Steta Gandara; and Matteo Morgandi. 2021. “Balancing Stability and Transition: A First Assessment of Regra de Permanência in Bolsa Familia.” Policy Note no. 162843, World Bank Group, Washington, DC. • Lara De Arruda, Pedro; Marina Lazarotto De Andrade; Tiago Falcao Silva; Diana Teixeira Barbosa; and Matteo Morgandi. 2022a. Sharing Lessons Learned from Brazil's Social Protection Response to COVID- 19: The Auxílio Emergencial Program. Washington, DC: World Bank Group. • Lara De Arruda, Pedro; Marina Lazarotto De Andrade; Tiago Falcao Silva; Diana Teixeira Barbosa; and Matteo Morgandi. 2022. “Coverage Profile of Brazil’s Auxílio Emergencial and Special Design Features for Protecting Women and Other Vulnerable Groups : Lessons from Brazil’s Auxílio Emergencial”. Washington, DC: The World Bank Group. • Morgandi, Matteo; Katharina M. Fietz; Gabriel Lyrio; Luiz Henrique Superti; Tiago Falcao; and Raquel Tsukada. 2023. The Profile of Brazil’s Working Poor and Implications for Economic Inclusion Programs. Washington, DC: World Bank Group. • World Bank. 2021. Enhancing Resilience of Low Income Workers in Brazil: Financial Instruments and Innovations. Washington, DC: World Bank Group. • Morgandi, Matteo; Raquel Tsukada; Malín Ed; Asta Zviniene; Luiz Paiva; and Barbara Barbosa. Social Protection for Brazil of the Future: Preparing for Change with Inclusion and Resilience (English). World Bank Group. • Morgandi, Matteo; Katharina Fietz; Luiz Henrique Superti. 2023a. Duration of stay in Bolsa Familia and determining factors: the role of individual and community characteristics. Washington, DC: World Bank Group. Page 72 of 71