IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-79420) ON A DEVELOPMENT POLICY LOAN FISCAL CONSOLIDATION FOR EFFICIENCY AND GROWTH

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank Report No: ICR2992 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-79420) ON A DEVELOPMENT POLICY LOAN FISCAL CONSOLIDATION FOR EFFICIENCY AND GROWTH IN THE AMOUNT OF US$1.045 BILLION TO THE MUNICIPALITY OF RIO DE JANEIRO, BRAZIL December 17, 2013 Poverty Reduction and Economic Management Department Brazil Country Management Unit Latin America and the Caribbean Region

2 CURRENCY EQUIVALENTS Exchange Rate Effective: December 17, 2013 Currency Unit = Real (R$) US$ 1.00 = R$ 2.33 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS CPS DPL EDI FHC FUNPREVI IBRD IDEB IDE-Rio ISR ISS LDO LDP LRF M&E MTEF MRJ NCR NFE NPV PD PDO PPP PSIA RAS RBM REGIN SMF SME SMS TAL UPA Country Partnership Strategy Development Policy Loan Integrated Preschool Program Family Health Center Pension Fund for the Municipality of Rio de Janeiro employees International Bank for Reconstruction and Development Basic Education Indicator Education Indicator for Rio Implementation Status Report Tax on Services Law of Budgetary Guidelines Letter of Development Policy Fiscal Responsibility Law Monitoring and Evaluation Medium Term Expenditure Framework Municipality of Rio de Janeiro Net Current Revenue Electronic Fiscal Invoice Net Present Value Program Document Project Development Objective Public-Private Partnership Poverty and Social Impact Assessment Reimbursable Advisory Service Results Based Management Integrated Registration System Municipal Secretariat of Finance Municipal Secretariat of Education Municipal Secretariat of Health Technical Assistance Loan Emergency Care Clinics Vice President: Country Director: Sector Manager: Task Team Leader: ICR Team Leader: ICR Primary Author: Hasan A. Tuluy Deborah L. Wetzel Auguste T. Kouame Edith Kikoni Edith Kikoni Robert Keyfitz

3 BRAZIL FISCAL CONSOLIDATION FOR EFFICIENCY AND GROWTH DEVELOPMENT POLICY LOAN CONTENTS Data Sheet A. Basic Information... i B. Key Dates... i C. Ratings Summary... i D. Sector and Theme Codes... ii E. Bank Staff... ii F. Results Framework Analysis... ii G. Ratings of Program Performance in ISRs... vi H. Restructuring... vi 1. PROGRAM CONTEXT, DEVELOPMENT OBJECTIVES, AND DESIGN KEY FACTORS AFFECTING IMPLEMENTATION AND OUTCOMES ASSESSMENT OF OUTCOMES ASSESSMENT OF RISK TO DEVELOPMENT OUTCOME ASSESSMENT OF BANK AND BORROWER PERFORMANCE LESSONS LEARNED Annex 1 Bank Lending and Implementation Support/Supervision Processes Annex 2. Summary of Borrower's ICR Annex 3. List of Supporting Documents MAP... 40

4

5 A. Basic Information Country: Brazil Program Name: Rio de Janeiro Municipality Fiscal Consolidation for Efficiency and Growth DPL Program ID: P L/C/TF Number(s): IBRD ICR Date: 12/17/2013 ICR Type: Core ICR Lending Instrument: DPL Borrower: MUNICIPALITY OF RIO DE JANEIRO Original Total Commitment: USD 1,045.00M Disbursed Amount: USD 1,045.00M Revised Amount: USD 1,045.00M Implementing Agencies: Municipal Secretariat of Finance of the City of Rio de Janeiro Cofinanciers and Other External Partners: B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 12/16/2009 Effectiveness: 08/27/2010 Appraisal: 02/09/2010 Restructuring(s): Approval: 07/01/2010 Mid-term Review: Closing: 06/30/ /30/2013 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Borrower Performance: Highly Satisfactory Low Satisfactory Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Satisfactory Quality of Supervision: Highly Satisfactory Implementing Agency/Agencies: Highly Satisfactory Overall Bank Performance: Satisfactory Overall Borrower Performance: Satisfactory i

6 C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments (if Indicators Performance any) Potential Problem Program Quality at Entry No at any time (Yes/No): (QEA): Problem Program at any Quality of Supervision No time (Yes/No): (QSA): DO rating before Satisfactory Closing/Inactive status: None None Rating: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration Compulsory pension and unemployment insurance General education sector General industry and trade sector Health Theme Code (as % of total Bank financing) Education for all Health system performance Municipal finance Public expenditure, financial management and procurement Regulation and competition policy E. Bank Staff Positions At ICR At Approval Vice President: Hasan A. Tuluy Pamela Cox Country Director: Deborah L. Wetzel Makhtar Diop Sector Manager: Auguste Tano Kouame Rodrigo A. Chaves Program Team Leader: Edith Kikoni Yaye Seynabou Sakho ICR Team Leader: Edith Kikoni ICR Primary Author: Robert Keyfitz F. Results Framework Analysis Program Development Objectives (from Project Appraisal Document) The Program Development Objective is to assist the municipal government of the city of Rio de Janeiro in creating fiscal space which will be used for investments to improve the quality and ii

7 efficiency of public service delivery, especially in poor areas, through innovative programs in health, education and private sector development. Other investments will strengthen the institutional framework for efficient service delivery through the implementation of a medium term expenditure framework, results-based management with the public sector, and PPPs with the private sector. Revised Program Development Objectives (if any, as approved by original approving authority) (a) PDO Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Reduced NPV of the pension system's actuarial deficit over the next 50 years of 17% Indicator 1 : (from a current baseline of R$36 billion) or R$6.3 billion of which R$3.1 billion from parametric reforms and R$3.2Billion from the recapitalization measures. Value (quantitative or R$ 36 billion R$ 29.7 billion R$ 0.7 billion Qualitative) Date achieved 01/01/ /30/ /31/2012 Comments (incl. % achievement) Indicator 2 : Partially achieved. Parametric reforms not implemented, though actuarial deficit almost eliminated through recapitalization measures. Discrepancy between Government and PD baseline data. Improved revenue collection, evidenced by a 4 percent increase in ISS tax collections through the NFEs (Electronic Fiscal Invoice system). Value (quantitative or 0% 4% 6% Qualitative) Date achieved 01/01/ /30/ /30/2013 Comments (incl. % achievement) Fully achieved. Indicator 3 : Rationalized current expenditures as evidenced by a reduction in the growth of personnel expenses. Value (quantitative or Qualitative) Growth of personnel expenses in 2008 was 13.2% and averaged 6% between 2005 and N/A Growth of personnel expenses in 2012 was 5.5% and averaged 3% between 2009 and Date achieved 12/31/ /31/ /31/2012 Comments (incl. % Fully achieved. Growth of personnel expenses has declined since achievement) Indicator 4 : Reduction in the number of days required to obtain a municipal business license in Rio from 20 days in 2009 to 12.5 days in Value (quantitative or 20 days 12.5 days 3 days iii

8 Qualitative) Date achieved 01/01/ /31/ /30/2013 Comments (incl. % achievement) Fully achieved. The end target value of the indicator was met in 2010 and has since been exceeded. Indicator 5 : The coverage of family health care will double from a January 2009 baseline of 6 percent to a projected 12 percent in December 2011 Value (quantitative or 6% 12% 40% Qualitative) Date achieved 01/01/ /31/ /30/2013 Comments (incl. % achievement) Indicator 6 : Value (quantitative or Qualitative) Fully achieved. The original target of 12% was met in January 2011 and has since been exceeded. A) Annual reductions in within year and between-year dropout rates and annual improvement in IDEB/IDE-Rio for Escolas do Amanhã; B) annual expansion by 3000 new openings in enrollments in ECD centers and pre-schools targeted to low-income communities. Dropout rate: 2.05% Dropout rate: 3.6% IDE-Rio: 3.7 for Grades 1-5 and 3.0 for Grades 6-9 IDEB: 4.6 for Grades 1-5 and 3.1 for Grades 6-9. Crèches: 29,921 Pre-schools: 77,845 Crèches: 39,921 Pre-schools: 86,845 iv IDE-Rio: 4.6 for Grades 1-5 and 4.2 for Grades 6-9 IDEB: 5.0 for Grades 1-5 and 4.0 for Grades 6-9. Crèches: 51,647 Pre-schools: N/A Date achieved 01/01/ /30/ /30/2013 Comments (incl. % achievement) Fully achieved. Part B assessed using ECD outcomes only as age limit for pre-school attendance lowered (6 year olds now enroll in grade 1), making current enrollment data incomparable with baseline. The share of municipal Grade 2 students meeting national reading proficiency Indicator 7 : standards on Provinha Brasil rises by at least 1 percentage point annually from 73 percent in Value (quantitative or 73% 76% 90% Qualitative) Date achieved 01/01/ /31/ /31/2012 Comments (incl. % achievement) Indicator 8 : Fully achieved. Assessed using results of Alfabetiza Rio which the Government has substituted for Provinha Brasil. The share of students meeting reading proficiency standards, measured using the Alfabetiza Rio, rose from 80 percent in 2010 to 90 percent in The share of children more than two years behind grade level at grade 8 or 9 (atrasados) decreases by at least 1 percentage point annually from 14.3 percent in Value 14.3% 11.3% Grade 8: 1.92%

9 (quantitative or Grade 9: 0.63% Qualitative) Date achieved 01/01/ /31/ /31/2012 Comments (incl. % achievement) Fully achieved. The share of atrasados at grade 8 declined from 5.5% in 2009 to 1.9% in At grade 9, the share of atrasados also declined from 4.2 to Baseline data mismeasured. Implementation of the initial year of a 3-5 years strategy with associated expenditures Indicator 9 : plan and financing plan used for the preparation of the budget guideline law of 2012 LDO (Lei das Diretrizes Orçamentárias). Value (quantitative or Qualitative) No medium-term strategy with expenditures and financing plans in place Implementation of the initial year of a 3-5 year strategy with associated expenditures plan and financing plan used for the preparation of the budget guideline law (LDO) of v Initial elements of an MTEF have been implemented. A medium term costing of planned expenditures has been prepared including a financing plan. This exercise has been repeated on an annual basis to inform budget preparation. Date achieved 01/01/ /30/ /30/2011 Comments (incl. % achievement) Fully achieved. Indicator 10 : Implementation of RBM, evidenced by annual evaluation of the performance contracts. Value No performance contracts 16 performance 42 performance (quantitative or Qualitative) signed; as such none evaluated. contracts signed and evaluated. contracts signed and evaluated. Date achieved 01/01/ /30/ /30/2013 Comments (incl. % achievement) Fully achieved. Indicator 11 : Establishment of a framework to enable PPPs to invest in infrastructure and service delivery projects, evidenced by at least one project prepared by the PPP unit. Value (quantitative or Qualitative) No PPP project prepared and under implementation. 1 PPP project prepared by the PPP unit and under implementation 4 PPP projects prepared by the PPP unit and under implementation. Date achieved 01/01/ /30/ /31/2012 Comments (incl. % achievement) Fully achieved. Indicator 12 : Average savings of 20 percent through centralized procurement of general consumption goods and services. Value (quantitative or 0% 20% 38.5%

10 Qualitative) Date achieved 01/01/ /30/ /30/2013 Comments (incl. % achievement) Fully achieved. Indicator 13 : Improved efficiency in stock management allows reducing losses in the stock of goods in the health sector from 20% to 10%. Value (quantitative or 20% 10% 0.01% Qualitative) Date achieved 01/01/ /30/ /31/2012 Comments (incl. % achievement) Fully achieved. Baseline data mismeasured. Between 2006 and 2009, losses in the stock of goods in the health sector averaged 4.9%. G. Ratings of Program Performance in ISRs No. Date ISR Actual Disbursements DO IP Archived (USD millions) 1 02/26/2011 Satisfactory Satisfactory /17/2011 Satisfactory Satisfactory /13/2011 Highly Satisfactory Highly Satisfactory /02/2012 Highly Satisfactory Highly Satisfactory /11/2013 Satisfactory Satisfactory /10/2013 Satisfactory Satisfactory H. Restructuring Not Applicable vi

11 1. PROGRAM CONTEXT, DEVELOPMENT OBJECTIVES, AND DESIGN 1.1 Context at appraisal Brazil is an upper middle income country with a population of 199 million and a per capita income of $11,630. Since the Plano Real brought inflation under control in the mid-1990s, the macro economy has stabilized and growth has accelerated. Rio de Janeiro with a population of 6.3 million is the second largest city and the capital of Rio de Janeiro State. As macroeconomic performance improved at a national level, Rio suffered a lengthy decline. Its economy contracted at an average annual rate of 1.1 percent from , and subsequently growth continued to lag other major capital cities. Rio is affected by social and economic inequalities, which were exacerbated by slow growth. At the time of appraisal, the poor comprised about 13 percent of the population and nearly 19 percent were living in neighborhoods often characterized as slums called favelas with inadequate infrastructure, schools and health facilities. After a fiscal crisis in the early 1990s Rio, along with many other municipalities, renegotiated its debts with the National Treasury. Since 2000, strict controls have been imposed under Brazil s Fiscal Responsibility Law (LRF). The LRF limits personnel spending to 60 percent of total expenditure, and puts a ceiling on public debt of 120 percent of current revenue. A golden rule limits net borrowing to the capital budget and outlaws lending and borrowing between state and municipal governments. Through adherence to the LRF, Rio s fiscal situation improved and was in compliance with the LRF s prudential limits. In 2009, personnel expenditures were below 50 percent of net current revenue, while net consolidated debt was 25.4 percent of net current revenue, down from 79 percent in Nevertheless, the capital budget contributed disproportionately to the adjustment, as investment fell from 11 percent of expenditure in 2004 to 9 percent in Moreover, fiscal constraints contributed to a deterioration in the quality of municipal service delivery and limited provision of services to poor and underserved areas. Meanwhile, the private sector continued to underperform. A new Government was elected in 2008 on a reformist platform and after taking office in January 2009 put forward a multiyear plan focused on laying the foundations for faster growth and improved public service delivery. The global financial crisis added further to the impetus for reform. The Municipal Government sought support from the Bank and the National Treasury for a debt restructuring operation that would decrease interest costs and create fiscal space for investment as well as support key reforms to strengthen fiscal management and government efficiency, and improve service delivery in poor and high-conflict areas. Among the main elements of the Government s program were: Fiscal adjustment: Creating fiscal space for investment through debt restructuring, improved revenue collection and expenditure management, and pension reform. 1

12 Service delivery innovations: Increasing access to public services and improving their quality and efficiency by simplifying business registration and licensing; expanding and restructuring health care delivery; and strengthening education through curriculum reform, remedial learning programs, and better performance monitoring. Public sector management: Raising public sector efficiency by implementing MTEF and Results Based Management, reforming procurement and promoting PPPs. 1.2 Rationale for Bank involvement The Bank has been a long standing partner in Brazil s development, valued for its technical assistance and knowledge services. However, Brazil is a large and sophisticated borrower and the Bank s lending envelope is small in relation to the federal budget. Thus, with the encouragement of the Federal Government, the Bank has increasingly sought to maximize its value added and developmental impact by engaging with subnational jurisdictions where its technical and financial support and knowledge services are more valuable. The Bank especially tries to engage in innovative operations which have demonstration effects and can be replicated elsewhere. The Rio Municipal DPL exemplified this approach. Though the Bank has worked with states such as Minas Gerais, Ceará and Rio de Janeiro, the Rio Municipal DPL represented the first development policy operation with a city government and extended the Bank s reach to a new level. Under Brazil s 1988 constitution, municipalities are substantially independent entities, similar to states, and are the main providers of many essential public services. This makes them natural partners for the Bank in pursuing Country Partnership Strategy (CPS) objectives of strengthening growth, reducing poverty, and improving social outcomes. The Rio Municipal DPL was seen as an innovation (requiring a waiver of OP/BP 8.60) that could be replicated in other municipalities. Rio s institutional strength and technical capacity made it a good choice to pilot such an operation. The Federal Government was supportive because the Bank s engagement would further implementation of the LRF. Beyond that, states and municipalities have overlapping jurisdictions in sectors such as health and education, requiring policy coordination. 1 The Rio Municipal DPL was prepared in conjunction with a Rio State DPL which covered similar areas, presenting a unique opportunity to support policy coordination to generate efficiency gains at both levels. Finally, the Bank s lending solved a difficult problem in that a number of jurisdictions that had renegotiated their debts with the Federal Government following the mid-1990s fiscal crisis, had found their (renegotiated) debts rising uncontrollably. Restructuring Rio s debt by using the 1 Indeed, Rio s situation is even more complicated as it was the national capital until 1960 when it became a state and then later the capital of a larger state with its own government. In areas such as health and education, it retains unique institutional arrangements reflecting this legacy. 2

13 proceeds to prepay the Federal loan and reduce debt servicing costs was expected to enable the loans to be paid off more quickly. 1.3 Original Program Development Objective and Key Indicators The objective of the operation was: to assist the municipality of Rio de Janeiro in creating fiscal space which will be used for investments to improve the quality and efficiency of public service delivery, especially in poor areas, through innovative programs in health, education and private sector development. Other investments will strengthen the institutional framework for efficient service delivery through the implementation of a medium term expenditures framework, resultsbased management with the public sector, and PPPs with the private sector. Outcome Indicators for tracking progress towards these objectives were the following: 1. Reduced NPV of the pension system s actuarial deficit over the next 50 years by 17 percent or R$6.3 billion of which R$3.1 billion from parametric reforms (first tranche) and R$3.2 billion from the recapitalization measures (second tranche) 2. Improved revenue collection, evidenced by a 4 percent increase in ISS tax collections through the introduction of electronic fiscal invoice system (Nota Fiscal Electrônica) 3. Reduction in the number of days required to obtain a municipal business license in Rio from 20 days in 2009 to 12.5 days in Doubling of family health care coverage from a January 2009 baseline of 6 percent to a projected 12 percent in December Annual reductions in student within-year and between-year dropout rates and annual improvement in the national and municipal education indices IDEB and IDE Rio for Escolas do Amanhã from the 2009 baseline and annual expansion in enrollments in ECD centers and preschools targeted to low income communities from 29,921 (crèches) and 77,845 (preschools) in 2009 baseline by 3000 new openings a year 6. Implementation of the initial year of a 3-5 years MTEF strategy with associated expenditures plan and financing plan 7. Implementation of Results Based Management, evidenced by annual evaluation of performance contracts signed by the mayor with line secretariats 8. Improved efficiency in stock management reduces losses in the stock of goods in the health sector from 20 percent to 10 percent 9. Establishment of a framework to enable PPPs to invest in infrastructure and service delivery projects, evidenced by at least one project prepared by the PPP Unit 10. Rationalized current expenditures as evidenced by a reduction in the growth of personnel expenses. 11. The share of municipal Grade 2 students meeting national reading proficiency standards on Provinha Brasil rises by at least 1 percentage point annually from 73 percent in The share of children more than two years behind grade level at grade 8 or 9 (atrasados) decreases by at least 1 percentage point annually from 14.3 percent in

14 13. Average savings of 20 percent through centralized procurement of general consumption goods and services. 1.4 Original Policy Areas Supported by the Program The operation comprised 3 pillars, each with specific objectives. Pillar 1: Creating Fiscal Space. Specific objective: Creating fiscal space to expand public investment and lay the foundation for municipal growth. Policy measures focused on tax administration and management of the municipal pension plan. However, the PD also discussed personnel costs and included milestones (i.e. not required for disbursement) on Human Resource management in the results matrix, bringing the expenditure side into the fiscal space equation. Debt restructuring, implicitly an element of the program, rounded out the program. Revenue collection: One of Rio s main revenue sources is the Imposto Sobre Serviços (ISS), a tax on services. DPL policy measures sought to increase its yield by: (a) implementing a system of electronic transactions recording, the Nota Fiscal Electrônica (NFE) and creation of a database (Sistema de Inteligencia Fiscal) to facilitate the identification of ISS tax evasion; and (b) submitting to the legislative assembly a draft law expediting recourse to judicial measures to recover tax arrears. An additional milestone called for approval by the legislative assembly of a user fee for public lighting. Outcome indicator: increased ISS tax revenue collection due to the NFE, with a target of 4 percent. Pension reform: The operation supported two types of reforms: (a) "Parametric reforms" entailing: (1) Calculation of pension benefits based on average rather than last wage; (2) Indexation of benefits on inflation rather than wage growth; and (3) Reducing the replacement rate of survivors benefits from 100 percent to 70 percent for pension benefits above the plan ceiling. In addition to the financial savings, the measures were intended to bring Rio s municipal pension plan closer in line with Constitutional Amendment 41. (b) Recapitalization of the pension fund, FUNPREVI, by dedicating various municipal revenue streams, including future oil and gas royalties and earnings on the municipal real estate portfolio (or equivalent sources). Outcome indicator: reduction of NPV of the pension system s actuarial deficit over the next fifty years by R$3.1 billion (8.5 percent) from the parametric reforms, and R$3.2 billion (8.8 percent) from the recapitalization. Human resource management: After taking office, the Government had initiated steps to bring operating costs and personnel expenditures under control. Amongst the various measures were reducing political appointments and limiting credit from prior employment toward pension benefits, which were included as first tranche milestones. A second tranche milestone called for undertaking a payroll audit, including active employees, pensioners and survivors, in order to identify fraud or clerical errors and 4

15 improve workforce planning. Outcome indicator: reduction in the growth of personnel expenses (no target specified). Debt restructuring: Proceeds of the DPL were used to pay down the municipality s debt to the National Treasury indeed the money was disbursed directly to the Treasury resulting in significant interest savings. Under the terms of the Treasury debt, a 20 percent prepayment would reduce the interest rate on the remaining balance by 3 percent. Twenty percent of Rio s debt of R$7.25 billion in 2009 represented around 80 percent of the amount of the DPL. As explained in the Letter of Development Policy (LDP), the Municipality planned to use the money for this purpose. Outcome indicator: not specified. However, a reduction in debt service costs of around 2 percent of net current revenue (NCR) in was anticipated (see Annex A5 of the PD). Pillar 2: Innovations in public services delivery. Specific objective: Reforming public service delivery with (a) improved government processes for registering businesses; (b) increased access to quality family health care and emergency care services through a new model for subcontracting social organizations to manage health facilities; and (c) improved early development of poor children, better quality of primary schools in high-conflict slum areas and system wide improvements in student learning outcomes. Business registration: The Government had initiated a project called Alvará Já, to improve the business environment by simplifying registration and issuance of business licenses for activities with low sanitary and environmental risk and providing for online application. The Government further planned to merge its business registration system with those at the state and national levels by joining the Integrated Registration System (REGIN). A DPL prior action required approval of Alvará Já s enabling legislation. Outcome indicator: Reduction in the average time to start a business from 20 days to 12.5 days. Health: Reforms in the health sector aimed at expanding access to primary care in underserved neighborhoods; improving the management and efficiency of hospitals and health facilities; and rationalizing provision of emergency services. The approach, adapted from other jurisdictions, was to establish a network of Family Health Clinics managed by Social Organizations private nonprofit entities having managerial and financial autonomy. Prior actions included approval of the legislation to transfer public services management to social organizations, and signing of management contracts for 20 family health clinics and 7 emergency care clinics. Outcome indicator: Doubling family health care coverage from 6 percent of the population in 2009 to 12 percent in Education: The Government s education reforms addressed core challenges related to (a) poor learning outcomes, (b) difficulties faced by schools in high-crime neighborhoods, and (c) large, unmet demands for preschool and day care services. The DPL supported specific reforms in all three areas. Prior actions included issuing decrees establishing innovative schools, Escolas do Amanhã, in 150 high-conflict slum neighborhoods; and 5

16 expanding provision of early childhood care and preschool services, with establishment of 10 integrated facilities in targeted low-income areas. Additional milestones called for provision of remedial literacy classes; upgrading teachers skills in literacy training; and setting up a system of bonus pay for schools attaining annual performance targets. Outcome indicators: annual improvement in test scores and reductions in dropout rates (no targets specified); annual expansion of 3,000 new openings in early childhood development centers and preschools targeted to low income communities; annual increase in the share of students meeting national reading proficiency standards by at least 1 percent. Pillar 3: Public Sector Management. Specific Objective: developing a framework to improve the efficiency of service delivery (a) in the public sector through introduction of results-based management (RBM) tools and the implementation of the first steps of a Medium Term Expenditure Framework (MTEF) and (b) in the private sector through public-private partnerships (PPP) in priority areas. Results Based Management and Medium Term Expenditure Framework: The Government sought to modernize planning and budgeting techniques in order to sharpen the focus on strategic priorities, increase efficiency and accountability, and improve the effectiveness of public service delivery. DPL supported measures included developing a Public Investment Framework and incorporating medium term projections into the budget process; and establishing a results-based monitoring and evaluation system and utilizing it for results agreements signed with city agencies. Outcome indicator: Implementation of initial year of a 3-5 year investment strategy with associated expenditure and financing plans incorporated into the 2012 budget; signing of 16 results agreements and publication of annual evaluations on the web. Public Private Partnerships (PPP): PPPs, if well managed, are a means of increasing both efficiency and the availability of resources for public investment. DPL prior actions entailed establishing a framework for PPPs and approval of the enabling legislation by the Legislative Assembly, and issuance of a decree establishing the by-laws of the Municipal Guarantee Fund for PPPs. Outcome indicator: preparation of at least one project by the PPP Unit. Procurement: Fragmented and inefficient procurement was seen as a further impediment to public sector efficiency. DPL prior actions required issuance of a decree institutionalizing yearly planning and centralized procurement of common goods and services; and piloting a stock management system for the health sector together with approval of a schedule for a municipality-wide roll-out. Outcome indicator: Average 20 percent saving on the cost of general consumption goods and services through centralized procurement; reduction in stock losses of goods in the health sector to 10 percent from a baseline of 20 percent. 6

17 Table 1 below shows a complete list of prior actions and indicative milestones from the PD results matrix (Annex 2). Table 1 Prior actions and second tranche release conditions Pillar 1 Creating Fiscal Space Second tranche First tranche Submitting to the legislative assembly a draft law consistent with the federal Constitutional Amendments No. 20 of 1998 and No. 41 of 2003 that mandate for new public servants: (i) the calculation of pension benefits based on the average wage rather than the last wage; (ii) the indexation of pension benefits on inflation rather than wage growth; (iii) the reduction by 30 percent of survivors benefits for pensioners above the RGPS ceiling. Decree No (01/01/09) mandating the reduction of 30 percent of commissioned positions in the direct and indirect municipal administration and the reduction of the incorporation of benefits earned in the private sector. Approval of the draft law to introduce a user charge tax on public illumination by the legislative assembly. Approval of Decree No /2009, which simplifies the registration process for municipal business licenses for activities with low environmental risk. Approval of the Law No. 5026/2009 and Decrees No /2009, No /2009 and No /2009 to transfer public services management to social organizations and implementation of the law with the Municipal Health Secretariat signing three Pillar 2 Innovation in public service delivery 7 Approval by the legislative assembly of the draft law on pensions that was submitted as a prior action for the first tranche. Submitting to the Legislative Assembly a draft law and/or adopting the adequate legislative framework to allow the following resources transfers to recapitalize FUNPREVI including through: (a) future royalties revenues from the municipality; (b) returns (amortizations and interests) from the portfolio of real estate loans of FUNPREVI; (c) real estate assets from FUNPREVI; (d) real estate assets from the municipality; or (e) other equivalent measures. Implementing a set of legal and administrative measures to improve the corporate governance of the pension fund and the institutional capacity to manage the pension fund s assets Issuing an administrative decree (Decreto Interno) setting up the organizational restructuring and the institutionalizing of the cash management function performed by the Treasury. Implementing the Electronic Fiscal Invoice (Nota Fiscal Eletrônica) for the collection of the municipal ISS and creating a database system (Sistema de Inteligencia Fiscal) to facilitate the identification of ISS tax evasion. Submitting to the legislative assembly a draft law to expedite the recourse to judicial measures to recover tax arrears. Completion of the municipality payroll audit including compensation for active employees, pensioners, and survivors benefits. REGIN (one stop shop for business registration) is implemented at the municipal level. Continued implementation of the new management model with the Municipal Health Secretariat signing 5 new management contracts with social organizations to implement services delivery in 10 additional family health care clinics (clínicas da

18 management contracts with qualified social organizations to deliver services in ten family health care clinics (clínica da família) and three emergency care clinics (UPA). Issuing of: (a) Decrees No /2009 and No /2009, establishing the design and budget for innovative schools (Bairro Educador/Escolas do Amanhã) in 150 high-conflict slum neighborhoods; and (b) Resolution No. 1057/2010 (published in the official gazette on January 25, 2010) that establishes a new model of integrated early child care and preschool services called EDIs (Espaços de Desenvolvimento Infantil), with the first set of centers in targeted low-income areas. família) and 4 additional emergency care clinics (UPA). First 10 EDIs (Integrated Early Child Development Centers) opened and fully operational in targeted low-income neighborhoods. Issuance of Decree No /2009, establishing a system of bonus pay for schools that attain annual targets for improvements in student learning and student flows. Establishment of a Partnership agreement with Ayrton Senna Institute for delivery of remedial classes during 2009 for 21,000 students diagnosed as functionally illiterate and training of 1,736 teachers in literacy teaching skills. Approval of Decree No /2010 and the signing of 16 results agreements with said entities and secretariats for the implementation of a monitoring and evaluation system with results based management tools. Development of an IT system for results monitoring. Establishment of a Project Unit to monitor a portfolio of 46 priority projects linked to the Strategic Plan of the Municipality. Establishment of a framework to enable PPPs to invest in infrastructure and service delivery projects, including: (a) approval of Law No. 105/2009 on municipal PPP programs by the Legislative Assembly; and (b) the issuance of Decree No /2010 establishing the by-laws of the Municipal Guarantee Fund for PPPs. Policy Pillar 3: Public Sector Management Initiating the implementation of a medium term expenditures framework as evidenced through a report presented by the Borrower s Secretariat of Finance confirming consistency between said framework and the Borrower s budget guideline draft law of 2012 in form and substance satisfactory to the Bank Monitoring and evaluation system for results agreements in operation as evidenced by (a) the publishing of the annual evaluation of the 16 results agreements signed in 2010; (b) the updating of the IT system with information on progress towards achievement of targets for elaboration of monthly status reports; (c) the maintenance of quarterly meetings on the progress towards achieving results agreements targets; (d) preparation of report on lessons learned in the first year of implementation of results agreements and their monitoring system. Implementation of a Public Investment Framework for the municipality to evaluate and select capital investment projects. The PPP unit has been created (decree issued and published in the Official Journal), is staffed and operational, and has issued processes approved by the PPP council for the preparation and procurement of PPP projects. Decree Nº (of March 17, 2009) institutionalizing yearly procurement planning for the centralized purchase of general consumption goods and services. Publication in the Official Journal of the call for expressions of interest for one PPP project. Piloting a stock-management system for the health sector and approval of the schedule for municipalitywide rollout implementation within 15 months. 8

19 1.5 Revised policy areas There were no revisions or modifications during implementation. 2. KEY FACTORS AFFECTING IMPLEMENTATION AND OUTCOMES 2.1 Program performance All required policy actions were satisfied in a timely manner, with no delays in effectiveness or disbursement. Table 1 above lists prior actions for the first tranche (first column) and second tranche release conditions (second column), together with milestones (not bolded). 2.2 Key factors affecting implementation Adequacy of the government s commitment: The DPL s policy areas were drawn from an integrated and mutually reinforcing package of reforms supported by all levels of government. A new Municipal Government under Mayor Eduardo Paes was elected in 2008 on a reformist platform and subsequently demonstrated a strong commitment to the overarching objectives, supported by the DPL, of private sector development; improving public services, especially to the poor; and strengthening public sector performance. In addition, the Municipal Government is politically aligned with the Rio State and Federal governments and DPL supported policies were reinforced by federal and state initiatives including Fiscal Responsibility Law and pension reforms mandated by constitutional amendments and federal law. Specific policy actions in the DPL were drawn from the Municipal Government s Multiyear Plan, which in turn was developed through a consultative process and enjoyed widespread popular support. Interest in RBM and MTEF, and better management of the capital budget reflected a general recognition of the need to improve public sector efficiency, especially in preparation for the World Cup and Olympics. Finally, the Bank was at the same time supporting complementary reforms at the State level, and the municipal DPL was coordinated with the State of Rio de Janeiro Fiscal Sustainability, Human Development and Competitiveness DPL. Soundness of background analysis The DPL built on strong analytical foundations, with contributions from the Municipal Government, World Bank and other sources, including: Studies undertaken specifically for the DPL on fiscal sustainability, public sector management and procurement, debt restructuring, social security, and a Poverty and Social Impact Analysis. 9

20 Technical studies on taxation and revenue collection, and fiscal projections prepared by the Municipal Finance Secretariat (SMF). Experience of other states on Results Based Management and partnerships with nongovernmental organizations to manage provision of public services. Doing Business indicators and studies from the SMF, McKinsey and Company, and various national organizations on the business environment, infrastructure, the regulatory environment and PPPs. Analysis by the municipal education secretariat (SME) in partnership with the Ayrton Senna foundation on remedial literacy training, and the Escolas do Amanhã and integrated early childhood development programs. Assessment of the operation s design The operation was jointly requested by the Municipal and Federal Governments and was complemented by other initiatives, especially a Rio de Janeiro State DPL. Thus, it enjoyed strong support from all levels of government. Policy areas were central to the Municipal Government s strategic vision and budget law that in turn had been prepared through a participatory and consultative process. In addition to a solid political commitment, the design incorporated lessons from extensive experience of lending at the State and Federal levels and took advantage of the Municipality s strong institutional and technical capacity. The two tranche structure allowed a more sustained commitment by the Bank while clarifying expectations about required policy actions and the timing of disbursements. The alternative of a programmatic series was deemed unsuitable because of higher transactions costs and uncertainty stemming from both the Government s and the Bank s complicated fiduciary requirements for subnational loans. 2 The DPL was designed as the focal point of an integrated program of support. It was packaged together with two additional Technical Assistance (TA) operations 3 to reinforce the objectives of improving service delivery in health and education, strengthening public sector management, and establishing an institutional framework for PPPs. These operations facilitated an extensive and ongoing dialogue which formed part of the DPL s supervision. Relevance of the risks identified Various plausible risks were identified and the overall level of risk was assessed as moderate to substantial. Potential problems were anticipated from: 2 Under the Fiscal Responsibility Law, the Government required a rigorous process of reviews and approvals at the Federal level, while the Bank needed to grant an OP/BP 8.60 waiver in addition to the normal process of preparation and Board approval. 3 Rio de Janeiro Strengthening Public Sector Management Technical Assistance Project (P127245) and Rio de Janeiro Fee-based Analytical and Advisory Services (P126815). 10

21 Institutional constraints on a new and inexperienced government s ability to implement and manage change. Capacity limitations in coping with the complexities of PPPs, or designing and overseeing reforms at the same time as preparing for the Olympics. Political risks, including those stemming from State and Federal elections which might change the political environment. In fact, none of these risks materialized. However, one area where problems occurred was in pension reform. It was thought that resistance to the parametric reforms would be contained because they applied only to new hires and therefore would not be opposed by an organized constituency. Nevertheless, though the legislation was presented to the Municipal Chamber it has not yet been passed and remains the subject of a legal challenge (see fn. 9). 2.3 M&E Design, Implementation and Utilization Design: The results framework comprised 13 indicators which were generally well selected and provided adequate coverage. However, including an indicator pertaining to investment spending would have been useful in tracking progress toward the program objectives, as would have been indicators to capture the quality and efficiency of health service provision. The health sector indicator in the results framework percentage of the population covered by family health care is a measure of access to family health care, which is only one aspect of the objective. In addition, baseline data were incorrect for two indicators. The 2009 pension plan deficit was R$22 billion rather than R$36 billion as given in the PD; and the initial rate of stock losses in the health service was 4.9 percent rather than 20 percent. Implementation: The Secretary of Finance had overall responsibility for M&E, with support from the office of the chief of staff (Casa Civil) and a Project Implementation Committee comprising officials from agencies directly involved with DPL-supported reforms. M&E was carried out in a satisfactory and timely manner, and results were shared with the Bank. Utilization: The M&E framework was used directly to provide evidence on satisfaction of the second tranche release conditions. Beyond that, attention to measuring outcomes stimulated further awareness and appreciation of the importance of M&E for the Government s reform strategy more generally, especially with regard to implementing Results Based Management and monitoring public service delivery in health and education. Support for development of M&E capacity and managing for results at a subnational level are elements of the Country Partnership Strategy. 2.4 Expected Next Phase/Follow-up Operation A follow-up operation, the Rio de Janeiro Strengthening Public Sector Management Technical Assistance Project (P127245), was prepared in , during the DPL s active phase. The 11

22 PDO is: to support institutional capacity strengthening in the Municipality of Rio de Janeiro to enhance public service delivery, including in health, education and environmental management. Following recent Board approval, the project is expected to become effective in early 2014 and run until Policy areas substantially overlap with the DPL though coverage is not identical. 4 Specific elements of the project pertinent to the DPL include: (a) Improving revenue forecasting and analysis of fiscal risks, and institutionalizing MTEF; (b) Building project selection and appraisal capacity to strengthen public investment management; (c) Enhancing the collection and use of electronic medical records and patient referral data to improve system-wide resource allocation and cost effectiveness in the health service; (d) Strengthening testing and monitoring, and evidence based policy making in education, especially for early childhood development programs. In addition to such specific components, the TA focuses on cross-cutting themes of M&E, results based management, and IT investment, all of which will increase the efficacy of the DPL reforms. 3. ASSESSMENT OF OUTCOMES 3.1 Relevance of Objectives, Design and Implementation Relevance of objectives: The objectives and program areas remain central to the Bank s current Country Partnership Strategy (CPS, FY ) whose objectives include strengthening fiscal management, improving the efficiency of public investment, and pursuing innovative approaches to Government programs and service delivery, especially in health and education. Creation of fiscal space in itself is less prominent, reflecting the extent of success already achieved. However, other program areas from the DPL by and large retain their importance. Some of the more relevant entries in the CPS results matrix include: Fiscal space: Strengthening tax administration and compliance, fiscal control systems, and debt management. Public sector efficiency: Improving budget management, through MTEF and RBM, better HR management, and PPPs. Innovative service delivery: Raising the quality and coverage of early childhood development, especially for low income households; improving learning outcomes and completion rates in primary and secondary education; expanding access to quality primary health care In terms of overall strategy, the CPS continues to emphasize engagement with sub-national Governments to facilitate learning and policy coordination, especially to develop innovative approaches that can be replicated. Thus, it is notable that various initiatives from the Rio Municipal DPL in areas such as early childhood development programs and teacher evaluation 4 The TA includes components on social governance, security and environmental management. 12

23 appear in operations planned for other states and municipalities. The reverse was also true as the Rio Municipal DPL was being prepared. Relevance of objectives: High Relevance of design: The policy measures were central to the Government s reform priorities and were in areas where the Bank was well positioned to contribute expertise and knowledge services. Prior actions and tranche release conditions were clearly delineated with an evident rationale. In terms of overall structure, the operation featured a compact and elegant synergy between the creation and use of fiscal space. Relevance of design: High 3.2 Achievement of Program Development Objectives The PDO refers to the creation of fiscal space, promotion of investment, and improvement of public sector performance. To assess the achievement of objectives, it is useful to break the statement into four components 5 : The objective of the proposed operation is to i. assist the municipality of Rio de Janeiro in creating fiscal space ii. which will be used for investments iii. to improve the quality and efficiency of public service delivery, especially in poor areas, through innovative programs in health, education, and private sector development. iv. Other investments will strengthen the institutional framework for efficient service delivery through the implementation of a medium term expenditure framework, resultsbased management with the public sector, and PPPs with the private sector. Measured in terms of the results, the operation was very successful. Of the 13 results indicators, 12 fully succeeded in achieving their targets, most by wide margins, and one partially succeeded. A few aspects of the PDO were not adequately captured by the indicators, but results there also appear to have been highly satisfactory. (i) Assist the municipality of Rio de Janeiro in creating fiscal space: Four subcomponents were designed to contribute to fiscal space: a) Improvements in tax administration; b) Better 5 This breakdown of the PDO splits the first pillar into twin objectives of creating fiscal space and increasing investment. Since this was a DPL, resources did not directly finance investment expenditure. Resources supported the development of policy reforms which would create an enabling environment to provide a much needed boost to investment. The PD and the Government s Letter of Development Policy (LDP) stressed that investment had contributed disproportionately to fiscal consolidation since the Fiscal Responsibility Law. As a result, they saw an imperative need to increase spending on physical and human capital in order to close the infrastructure deficit and lay the foundations for faster growth 13

24 control of personnel expenses; c) Reduction in the pension plan deficit; d) Debt restructuring utilizing proceeds of the DPL. a) Tax administration: The Nota Fiscal Electrônica was implemented (a first tranche prior action) and factored into buoyant ISS tax collections in , with real growth averaging 10 percent. The authorities estimate the NFE contributed 6 percent of the increase or R$ 686 million, exceeding the DPL target of 4 percent. 6 Measures to increase recovery of tax arrears (the PPI - Incentive Payment Program) were passed in 2012, mainly comprising amnesty against fines and penalties, and yielded a one-time inflow of R$972 million in Finally, user charges for public lighting, established by Law No of December 21, 2009, have been collected by the electric utility and transferred to the municipality, raising R$580 million to date. Fully achieved. b) Reduction in the pension plan deficit: Law 5.300, passed in 2011, incorporated a number of recapitalization measures to eliminate the actuarial deficit on the pension plan, which the authorities estimated at R$22 billion. 7 First, the Municipality committed to pay a 35 percent supplementary contribution on the wages of current employees for a period of 35 years, with a NPV of R$18 billion. Second, various assets were transferred to the pension plan: (1) Oil and gas royalties accruing to the municipality over the period , with an expected NPV of R$2.5 billion. (2) Real estate valued at R$1.2 billion, which had previously been managed jointly with, but not owned by, the pension plan. 8 (3) Amortization on mortgage loans offered as part of the benefit plans for current employees of the municipality with a NPV of R$0.5 billion. In addition to recapitalization, parametric reforms were proposed to reduce benefits to new hires, which would also bring Rio into compliance with Federal laws. A draft law on parametric reforms ( ) was submitted to the Municipal Chamber in So far, however, it has not been passed and its prospects are uncertain. As a result, no savings from parametric reforms have been realized. 9 6 ISS revenue collection was modeled statistically and the impact of the NFE inferred from the difference between predicted and actual outcomes. See SMF technical note, Nota Carioca: Impactos Financeiros Diretos Após 3 anos de Implantação. 7 The PD gives the shortfall as R$36 billion. Different actuarial assumptions likely explain the discrepancy between the two estimates. 8 Technically, these assets belonged to a municipal corporation in which current employees were the beneficial owners, and were managed by PREVIRIO, which also manages the pension plan. Thus, in effect, this measure increases employee contributions. 9 The Municipal Government has not fully implemented, the federally mandated pension reforms laid out in Constitutional Amendment 41 of 2003, arguing that it has funded its pension plan with its own revenues, a position upheld by the Brazilian courts. While the Government remains committed to the parametric reforms, it prefers a consensual approach and has sought the acquiescence of civil servants to the changes. An attempt to fast track the 14

25 Despite some risks primarily concerning oil and gas prices and changes to revenue sharing arrangements which could affect revenue from royalties these actions have put the pension plan on a sound basis. However, two facts should be noted. First, the parametric reforms have not been completed. Second, the net impact on fiscal space is relatively small because most of the resolution transfers assets already on the Municipality s books and therefore has no net impact on the consolidated accounts. Partially achieved. c) Personnel expenses. The Government decreed a 30 percent reduction in commissioned employment and cuts to benefits in early 2009 (a first tranche milestone), sharply slowing wage growth that year. Subsequently an external audit of the payroll was carried out (a second tranche milestone), though relatively few problems were found which may not be surprising as similar audits had been done in the recent past. As a result, growth in personnel costs slowed in both real and nominal terms (Table 2). The average 2.8 percent slowdown in nominal growth translates into a cumulative saving of R$0.7 billion on the municipal wage bill over the DPL period. Fully achieved. Table 2 Municipal wage bill growth Annual average, percent Real Nominal Source: SMF d) Debt restructuring: Debt restructuring meaning paying down Rio s debt to the National Treasury with proceeds of the DPL 10 had a twofold impact on the municipal finances. (1) On the first US$1.045 billion, the saving was equal to the difference between terms on the debt to the Treasury and the Bank; (2) prepayment of over 20 percent of the outstanding debt reduced the interest charges on the remaining balance by 3 percentage points. The combined effect has been almost a halving of debt service costs and a saving of R$951 million to date. 11 Fully achieved. To summarize, from 2010 through the first half of 2013 measures under this pillar created R$3.9 billion of fiscal space around 7.7 percent of Net Current Revenue or 43.6 percent of investment spending over the period (see Table 3). Around a quarter of the total came from using the proceeds of the DPL to prepay Rio s loan to the National Treasury, the rest from reforms through the legislative process raised suspicions and may be partly responsible for the subsequent delay as the Government has tried to improve communication with stakeholders. 10 The DPL money was disbursed directly to the National Treasury and used in its entirety to pay down Rio s debt. 11 As a percentage of net current revenues, debt service costs have been reduced from 10.5 percent in 2008 to 5.4 percent in 2012, exceeding the 2 percentage point increase that was anticipated. 15

26 tax measures and slower growth of personnel costs. Three results indicators were associated with this pillar, two of which were fully achieved (increase in ISS tax collections and slowing the growth of the wage bill), and one was partially achieved (reduction in the pension plan deficit through recapitalization and parametric reforms). Considering the overall outcome, a satisfactory rating is justified. Table 3 Fiscal space created by DPL supported measures 12 R$ millions, except as noted 2010 June 2013 Debt Restructuring Tax Administration Reform 1,657.7 Electronic Fiscal Invoice System PPI-Incentive Payment Program Public lighting user charges Personnel expenses Total 3,908.3 Total (% of Net Current Revenue) 7.7 Total (% of investment budget) 43.6 Source: SMF; World Bank staff calculations Outcome rating: Satisfactory which will be used for investments There are no indicators in the results framework pertaining to investment. However, Government data show that municipal public sector investment surged in , rising from R$400 million to R$3.3 billion, or from 3.7 percent to 15.9 percent of total expenditure (see Table 4). Education and health accounted for R$341 million of the increase, which included investment geared toward improving public service delivery and supported by the DPL through policy measures (day care facilities, schools and family health clinics). The bulk of the investment has been on transportation systems, mainly four Bus Rapid Transit routes, which will enhance Rio s long run growth potential, as well as some upgrading of infrastructure (wastewater, roads) in the favelas. The investment is expected to benefit low income, more remote neighborhoods by improving labor market access, though the entire population of the city will benefit from reduced road congestion. Meanwhile, PPPs added a further R$1.1 billion to public investment. The investment cycle is expected to continue for several more years before tapering off. It is important to note that virtually none of the increase in investment is related to preparation for the World Cup or Olympics, though some demands on the city s capital budget are anticipated over the next two years. 12 Estimates by the SMF. Estimates also available in the fiscal transparency reports published by the SMF. 16

27 Table 4 Municipal Investment R$ millions except as noted Investment 400 1,560 3,347 3,321 Of which: World Cup/Olympic facilities Investment in social sectors Education Health Investment spending on PPPs ,175 Of which Government s share Memo item: Investment/total expenditure (%) Source: SMF Outcome rating: Highly satisfactory (iii) to improve the quality and efficiency of public service delivery, especially in poor areas, through innovative programs in health, education, and private sector development. Health: Since 2009, the government has carried out a major restructuring and expansion of health care provision based on integrated networks of family and urgent care clinics. Group practice Family Health Centers (FHCs) have been established to serve mainly low income populations without access to private health insurance. Each center houses a number of Family Health Teams and provides access to a comprehensive range of primary care services, including prenatal and pediatric, dental, and radiology (X-rays and ultrasound) services along with exercise facilities and a pharmacy dispensing commonly prescribed drugs. Community outreach and educational programs focus on preventive care. Patients with more advanced needs may be referred to other centers for specialty care. Urgent care centers operating 24/7 provide emergency services. The networks are financed jointly by the Federal and Municipal Governments and all services are provided free of charge. A rapid buildup since 2009 has established 71 FHCs with 825 Family Health Teams, all managed by private, non-profit Social Organizations which are subject to close monitoring by the Municipal Health Secretariat. Implementation is not yet complete some neighborhoods have not yet been reached and there are long waits for some specialty treatments, but access to health care services has expanded impressively from 6 percent of the population in 2009 to 40 percent currently, 13 Corresponds to investments made through concessions, and not the municipal budget. Investments included counterpart funding from the MRJ only in the case of the Olympic Park. Projects: Porto Maravilha, Olympic Park, West Zone Sanitation. 17

28 almost entirely low income populations. 14 This greatly exceeds the DPL target of 12 percent and the plan is to reach 70 percent in the next year or two. Meanwhile, service quality indicators such as immunization rates, availability of physicians, community outreach and emergency wait times have also registered gains, while the percentage of unnecessary hospital admissions has fallen, pointing to greater efficiency. Health service efficiency and effectiveness is one of the areas to be addressed in the follow up TA Project (see Section 2.4) 15. In a recent survey, a remarkable 92 percent of patients expressed satisfaction with the service (Table 5). Fully achieved Table 5 Health Care Quality Indicators Immunization rate 16 (%) Family health care physicians per 10, population Coverage of dengue control home inspections 17 (%) Patient satisfaction score 18 (%) 92 Emergency wait time (minutes) ICSAB Source: SMS Education: Wide-ranging reforms have been implemented to improve overall educational effectiveness and learning outcomes, which include increasing classroom time, strengthening monitoring and evaluation of student performance, expanding programs for teacher training and mentoring, and promoting community engagement, especially in the 152 innovative, Escolas do Amanhã (Schools of Tomorrow). DPL policy actions aimed at specific factors affecting learning outcomes, especially in low-income, dangerous neighborhoods, and remedial efforts to reduce the backlog of older students performing below grade level in reading and math. Four results indicators focused on keeping students in school and improving their performance. Despite some technical difficulties, the indicators point to highly successful outcomes: 14 Coverage is defined in terms of access to a primary care provider. Family Health Teams serve an average of 3,450 people, hence 2.8 million persons are covered by the 825 FHTs. 15 The TAL (P127245) will also support evaluations of primary health care reforms, with a view to further improve their design. The findings of the evaluation will be used to support decisions in respect to the continued roll-out and adjustments of the primary health care delivery model 16 Coverage of 3 vaccines (Tetra, Penta and DTP). Decline in coverage in 2010 due to the focus of resources on treatment and management of the H1N1 influenza pandemic. 17 Proportion of residential houses located in dengue infested areas that are inspected by agents of endemic disease control. 18 Survey conducted in February and March 2012 in Family Clinics by the Institute of Research, MAPEAR. 19 The ICSAB (Indice de condições sensíveis à atenção básica) is a key indicator used by the Municipal Secretariat of Health to evaluate the quality of the primary care system. The indicator measures the proportion of hospital admissions for conditions that should be effectively managed/ treated at the primary care level. 18

29 Annual reduction in student within year and between-year dropout rates and annual improvement in IDEB/IDE-Rio 20 for Escolas do Amanhã from 2009 baseline. Student dropout rates were reduced from 3.6 percent in 2009 to 2.1 percent in 2012, while Escolas do Amanhã improved their reading performance scores on IDEB/IDE-Rio between 2009 and 2012 for all grades (see Table 6). Fully achieved. Annual expansion in enrollments in ECD centers and preschools targeted to low income communities from 29,921 (crèches) and 77,845 (pré-escolas) in 2009 baseline by 3,000 new openings a year. In 2012, 47,906 children were attending public crèches representing an annual increase of nearly 6,000 places, while an additional 16,430 children were attending semi-public crèches (conveniadas). Space in these facilities is limited and children are selected through a lottery which takes socio-economic status into consideration. Most or all of the new places have been allocated to children from low income families (mostly Bolsa Familia recipients). Measuring the increase in preschool attendance is problematic because of a change in the eligible age range. In 2009, preschool programs covered children aged 4-6, but since 2010, 6 year olds have enrolled in first grade. Dropping the second indicator and evaluating the outcome based on ECD enrollment alone, the outcome is assessed as fully achieved. The share of municipal Grade 2 students meeting national reading proficiency standards on Provinha Brasil rises by at least 1 percentage point annually from 73 percent in Due to issues with the Provinha Brasil, Rio has substituted its own reading proficiency indicator, the Alfabetiza Rio, which has been developed rigorously and is thought to be more suitable. The Provinha Brasil implicitly reflects a standard that children should achieve literacy by the end of second grade, while Rio Municipality's own standard is that children should achieve literacy by the end of first grade. In addition, a change in the test design between 2010 and 2011 makes it difficult to compare results over the DPL implementation period. Assessing the outcome using the Alfabetiza Rio shows that there has been annual progress since the baseline year, The percentage of children tested with an adequate level of reading skills at the end of grade 1 improved from 80 percent in 2010 to 82 percent in 2011 and 90 percent in 2012 (Table 6). Fully achieved. The share of children more than two years behind grade level at grade 8 or 9 (atrasados) decreases by at least 1 percentage point annually from 14.3 percent in Two data series are available for this indicator. Neither one reproduces the 14.3 percent baseline figure in the results framework. 21 Table 6 shows a somewhat irregular pattern of decreasing shares due to termination of the automatic promotion system (Progressão Continuada) in 2009, which impacted repetition rates and consequently the number of 20 IDEB (Índice de Desenvolvimento da Educação Básica) and IDE-Rio (Indice de Desenvolvimento da Educação Rio) are respectively national and municipal indices of education outcomes. 21 At the time of appraisal, the SME's initial data from school census reports in late 2009 was that the share of "atrasados" was 14.3%. Over the past two years, the SME has worked with the Federal Ministry of Education to review and refine its data on age-grade distortion. As a result, the SME has now published the time series in table 6. 19

30 atrasados in However, shares have declined by more than 1 percentage point each year from 2010 as the result of the set of DPL supported policies introduced by the SME to reduce age-grade distortion. 23 Fully achieved. Table 6 Education outcome indicators Student dropout rates (%) IDEB average score: grades 1-5 (Escolas do Amanhã) IDEB average score: grades 6-9 (Escolas do Amanhã) IDE Rio average score: grades 1-3 (Escolas do Amanhã) IDE Rio average score: grades 4-7 (Escolas do Amanhã) ECD Enrollments 29,921 33,386 39,386 47,906 Literacy pass rates: grade 2 (Alfabetiza-Rio) (%) Atrasados: grade 8 (%) Atrasados: grade 9 (%) Source: SME Business licensing and registration: The burden of starting a business has been eased by allowing online license applications and simplifying bureaucratic requirements, including through Rio s Alvará Já program specially tailored to businesses with low environmental and sanitation impacts. Participation in the state-run REGIN system has integrated business registration procedures at the municipal, state and federal levels. As a result, in September percent of licenses for firms with low environmental and sanitation impacts were made within 3 days. The municipality receives around 525 queries per day on the web, a 310 percent increase from 2010 and grants an average of 105 licenses, a 500 percent increase. Fully achieved. In summary, the Government implemented extensive reforms in all three program areas and fully achieved all results targets, most by a wide margin. Outcome rating: Highly satisfactory (iv) Other investments will strengthen the institutional framework for efficient service delivery through the implementation of a medium term expenditure framework, resultsbased management with the public sector, and PPPs with the private sector. 22 Progressão Continuada is a program in Brazil s public education system that allows for automatic promotion between school years irrespective of performance to reduce discouragement from repetition and prevent high dropout rates. 23 Those policies include: i) beginning of the school year reinforcement classes for students not quite on grade level; ii) after school tutoring for students falling behind during the year, so as to avoid repetition at the end of the year; iii) accelerated learning classes to help "atrasados" catch up with their peers; and iv) a school bonus program which incentivizes schools to lower repetition at the same time as they seek to increase student learning. 24 Unlike IDE Rio, IDEB is computed only in odd years. Both indices range on a scale from 0 to

31 Results Based Management: RBM was launched with a small pilot involving a few schools in In 2010 the first Results Agreements were signed, providing significant incentives up to four times (monthly) base salary for managers achieving exceptional performance. Since then the program has been extended to all but one city agency and covers over 80 percent of the Municipality s 110,000 workers. 25 Targets are set in an open and consultative fashion and managers who feel they have not been given budgetary and other resources needed to achieve the results can request a review by the mayor s office. Both targets and results are published on the web and are externally audited to ensure transparency, for instance, to discourage setting targets at too easy a level in order to ensure a bonus will be paid. After some initial suspicion and skepticism the program has gained acceptance and anecdotal evidence suggests possible early signs of a change in the culture of the civil service. The DPL s target of 16 Results Agreements signed in 2010 has been substantially exceeded. Fully achieved. Medium Term Expenditure Framework: Since 2011, draft MTEF budget documents covering percent of expenditure have been prepared in parallel to the traditional budget process and published on the Municipality s web site. They describe policies and programs, indicate planned investments and recurrent expenditures, and estimate fiscal risks for a rolling four year period. Traditionally, forecasts of current revenue have been quite accurate. However, forecasting expenditure and capital revenue, mainly comprising transfers from the Federal Government, has been more difficult. The TA Loan will support efforts to strengthen performance in these areas. While the MTEF is not fully operational, the steps that have been taken represent important progress and a move up the learning curve, hence are considered as meeting the target of implementing the first steps of a 3-5 year MTEF. Fully achieved. Public Private Partnerships: Using existing federal law for PPPs, the city has created a PPP Council, defined project criteria and created a guarantee fund for PPPs, strengthening the institutional capacity to design and manage PPP projects. The Municipal Guarantee Fund was set up in accord with the DPL s prior action for the first tranche, though the municipality has adopted a different approach of providing private partners with additional real estate assets sufficient to guarantee an adequate return. As of June 2013, five projects at a total value of R$15 billion were under implementation, four of them prepared by the PPP Unit, in sanitation, transport (light rail and road), and Olympic park facilities. Fully achieved. Procurement reform: In 2009 a centralized, electronic auction system was introduced for standardized purchases utilizing the Federal Government portal and subsequently The exception is COMLURB, which provides sanitation services and collects garbage, because of special circumstances, but is expected to be brought in to the scheme shortly. 21

32 auctions have been held. Small and medium enterprises are eligible to bid on contracts up to R$80,000, and the more open bidding process also potentially spreads benefits to local economies and entrepreneurs across the country. The Municipal Secretariat of Administration estimates average savings of 39 percent in the first half of Complementing the procurement reforms, new stock management procedures piloted in the health system have reduced losses from 4.9 percent in to 0.01 percent in Fully achieved. Outcome rating: Highly Satisfactory 3.3 Justification of Overall Outcome Rating The operation was highly relevant to the Municipal Government s reform program and simultaneously supported the Federal Government s objectives of fiscal consolidation and policy coordination. From the Bank s standpoint, it was aligned with the Country Partnership Strategy objectives of poverty reduction, private sector development and fiscal consolidation, and it was consistent with the Country Partnership Strategy s approach of working with subnational governments. The design was highly relevant, with program areas and policy actions supportive of the program development objectives. Implementation was satisfactory with no delays in completing prior actions and tranche release conditions. The achievement of objectives in three of the four components of the PDO was highly satisfactory, while the remaining one, creation of fiscal space, was satisfactory. Even there, the bottom line fully met expectations in terms of fiscal space created, since the shortcoming parametric pension reform had only a limited impact on the accomplishment of the program objectives, in the short term. 27 The operation contributed to many aspects of the Government s program institutional, technical, financial and social and has had some broader impact within Brazil and across the region. 28 Overall, the outcome is rated as highly satisfactory. Overall outcome rating: Highly Satisfactory 3.4 Overarching Themes, Other Outcomes and Impacts (a) Poverty impacts, gender impacts and social development A preliminary Poverty and Social Impact Assessment (PSIA) during preparation anticipated that DPL supported elements of the Government s program would have significant beneficial 26 The baseline of 4.9 percent is far below the 20 percent figure in the PD. Authorities could not explain the discrepancy, but thought the 20 percent figure was implausibly large. 27 Parametric reforms supported applied only to new public servants. 28 Rio s Escolas do Amanhã, the first program of its kind in Brazil, has inspired the development of similar programs. One such example is the Escuela amiga program in Lima s high conflict areas, which directly adapted elements of the Escolas do Amanhã. Specific structural aspects of Rio s health care delivery model (the inbuilt incentive mechanism and the technical monitoring system) have also inspired other sub-national governments within Brazil. 22

33 impacts. Building on the PSIA s analysis, a more precise look at how the policies were actually implemented suggests strongly that the overall impacts were pro-poor. Education: Poverty is closely correlated with poor educational outcomes which limit economic mobility and perpetuate low socio-economic achievement. Education measures supported by the DPL aimed specifically at breaking this cycle. Early childhood development is a robust predictor of later academic success and the 18,000 additional places in crèches were almost entirely reserved for low income families. The Escolas do Amanhã program focusing on the 150 worst performing schools in dangerous and violent neighborhoods, and remedial programs for older children performing behind grade level also specifically targeted low income populations. Measured improvements in learning outcomes and declining attrition rates almost certainly predominantly reflect gains in these communities. It takes a generation for education to impact on socio-economic outcomes, but maintaining these initiatives, as the Government appears committed to doing, should contribute to narrowing the gap in socio-economic outcomes. Health: As with education, health outcomes too are strongly correlated with poverty. The city s wealthier residents typically purchase private health insurance and are well served by private health providers, while in poorer areas access to primary care had been limited or non-existent. Family Health Centers are located in previously underserved, low income areas and exclusively serve their immediate neighborhoods. Pre-natal, infant and youth programs ensure attention to women s and children s health needs, while community outreach encourages wider use of the services. Penetration remains low in some of the more remote areas in the Western Zone of the city, but for the 40 percent of the population now covered, acceptance of the new management model and a patient satisfaction rating of 92 percent in an early 2012 survey (see Table 5, above) attests to the program s success. Private Sector Development: Poverty and social impacts of easier business licensing and registration are less clear cut, but DPL supported reforms have likely been a factor in the three-fold rise in the rate of business registration. Shifting activity out of the informal sector and creating formal sector employment with better wages and working conditions should benefit low income households, especially in conjunction with investments in public transport to increase access to those better jobs. More generally, improving the business climate, where Rio has lagged other regions, leads to faster private sector growth, stronger labor markets and declining poverty. (b) Institutional Change/Strengthening Institutional development explicitly or implicitly underpinned reforms in many of the DPL s program areas. With regard to the third pillar on public sector management, the PDO refers 23

34 directly to strengthening the institutional framework. For PPPs, the DPL together with associated TA sponsored an intensive workshop, prepared manuals and provided a resident advisor leading to development of an institutional framework and establishment of a functioning, operational PPP Unit. Institutional development is inherent in moving to MTEF, which comprises a framework for aligning resource allocation to policy priorities, and incorporating future implications of current spending into the budget process. Results Based Management may be the most far reaching institutional reform of all and, anecdotally, has begun to impact on the culture of the civil service. Under the service delivery pillar, a massive expansion of health service provision entailed the development of an extensive network of clinics under the management of Social Organizations, as well as setting up IT systems for electronic medical record keeping and online referrals. In education, reforms included major attention to M&E, teacher training and incentives, and community engagement in low performing, dangerous schools. All three service delivery components featured IT investment, and the forthcoming TA project intends to develop further data management and analytical capacities for evidence based policymaking. (c) Other Unintended Outcomes and Impacts The operation was the first ever DPL to a municipality and required a waiver of OP/BP The operation triggered a substantive revision of Bank-wide DPL policy and in 2011, the Board agreed to amend OP/BP 8.60 and allow DPOs with any political subdivisions having the budgetary and legislative authority to enter into an agreement with the Bank. Waivers are therefore no longer needed. Subsequently, one more Brazilian municipal DPL has been approved the Belo Horizonte Municipal DPL (P126749) and one more is under preparation. The instrument is particularly suited to Brazil because the Federal Government is interested in having the Bank engage directly with cities and the constitutional and legislative framework grants them substantial autonomy. However, there is considerable potential elsewhere as well. 4. ASSESSMENT OF RISK TO DEVELOPMENT OUTCOME While there are risks to the development outcome, overall they appear to be low. Brazil s current slowdown implies macroeconomic and fiscal risks that might lead the Government to relax its fiscal controls. Current indicators are well within FRL limits and there is space to ease back on fiscal consolidation, for instance in areas such as HR management. Nevertheless, preparations for the World Cup and Olympics are likely to insulate Rio de Janeiro from national trends until At least in the medium term this risk appears low. Recent unrest indicates possible policy risks. Public opinion in Brazil has been volatile and resistance to the reforms could build rapidly, causing implementation to slow. So far protests have not been directed against DPL supported reforms. The Government remains popular 24

35 and a moderate level of resistance might even serve as an incentive to accelerate the pace. The risk here is low to moderate. The PD anticipated operational risks for a relatively new and inexperienced city administration in managing its reform program and eliciting cooperation from civil servants and political risks due to opposition from entrenched political interests. However, successful implementation of the program and strong support from the electorate for a second mandate in the 2012 municipal election suggests such risks are negligible. Institutional capacity may be lacking in some areas, though so far the municipality appears to have coped well with the complex challenges of preparing for the Olympics, managing PPPs and introducing large scale reforms in health and education. The recently launched Rio TA project further mitigates this risk. Rating: Low 5. ASSESSMENT OF BANK AND BORROWER PERFORMANCE 5.1 Bank performance Quality at entry: There is no doubt that this was a well-conceived and designed operation, that responded effectively to a joint request from the Municipal and Federal authorities. Both the objectives and policy areas were clear and highly relevant to the Government s reform strategy and the Bank s CPS and built on substantial analytical foundations. Although this was the first municipal DPO in Brazil (and indeed in the world), it incorporated ideas and lessons from previous operations with various State Governments. Concurrent preparation and implementation of Rio Municipal and State DPLs facilitated policy coordination at the two levels and avoided duplication in areas where there were overlapping jurisdictions. At the same time, a few aspects of the operation might have been strengthened. First, a more thorough due diligence could have ascertained whether there was an adequate consensus for parametric pension reforms. Elsewhere, the results framework should have been broadened to include indicators in a few areas that were relevant to the project objectives. Specifically, there were no indicators to cover the important aspects of public sector service quality and efficiency, and investment. In addition, there were a few discrepancies in baseline figures, though most of these can be explained. Rating: Satisfactory Quality of supervision: Supervision was given close attention because of both the size of the loan and the innovative aspect of municipal lending. Careful monitoring of results indicators contributed to assessing 25

36 fulfillment of second tranche release conditions and ISRs were regularly prepared and filed. No problems were encountered that required management attention, though in the case of pension reform where implementation slipped, the Bank responded proactively by arranging a mission for a pension specialist to intensify the dialogue about the Government s approach to recapitalization and parametric reforms. In addition, two Technical Assistance projects coordinated closely with the DPL and contributed to supervision. The first, a Reimbursable Advisory Service (RAS) project on PPPs enabled the Bank to collaborate closely with the authorities on all aspects of PPP management from setting up the institutional framework to building operational capacity for project preparation, bidding and oversight, and contingent liability management. The second, a TA Loan on Strengthening Public Sector Performance, contributed to supervision of most of the remaining elements of the DPL while it was under preparation in , including health, education, public investment management, MTEF and RBM, and procurement. A series of preparation and appraisal missions maintained a close policy dialogue with the authorities about progress on the DPL program and the need for additional follow up. Rating: Highly Satisfactory Bank performance rating: Satisfactory 5.2 Borrower Performance The Government s performance was Satisfactory, exhibiting strong political leadership and a high level of commitment during preparation and implementation. Without this leadership the operation could not have succeeded to the extent it did, especially in the more innovative and challenging components such as Results Based Management and the rollout of the new health care model. Required actions were satisfied in a timely and exemplary manner, and the loan disbursed fully without any delays. Apart from parametric pension reform, expected outcomes were achieved in all policy areas and results indicators met or exceeded their targets. Without the pension setback, the Government s performance would easily have been rated Highly Satisfactory. Implementing Agency performance was Highly Satisfactory. The Secretariat of Finance had overall responsibility with support from the office of the chief of staff. A Project Implementation Committee comprised of officials from agencies directly affected by the reforms was in charge of program surveillance and M&E. Officials and agencies overseeing the implementation demonstrated commitment, as well as the technical and institutional capacity to manage the complex and demanding program. Communication with Bank staff was excellent. Borrower performance rating: Satisfactory 26

37 6. LESSONS LEARNED The operation reinforces a number of well-known lessons: Strong and effective political leadership is essential: The Government s ownership and commitment, backed up by institutional capacity, were critical factors in the DPL s success. In no small measure, the DPL achieved its objectives because it was aligned to the priorities of the Government s reform strategy. Political economy matters: The experience with pension reform is a reminder of the need for realistic expectations where policymakers may not be able to control the political economy. Stiff resistance from civil servants who represent a powerful political bloc made it difficult to reconcile the Government s consensual approach with pushing through controversial reforms in the DPL s time frame. The Bank should be cautious about backing reforms where the consensus is inadequate. Flexibility opens up new possibilities: The Bank is at its best when it is flexible and responsive and willing to adapt to clients needs, especially with middle income countries. What began as an experiment with municipal lending that required a waiver quickly led to a change in policy and may in time develop into an important new line in the Bank s portfolio. In addition, there are some less familiar lessons concerning subnational lending, where there is substantial growth potential: Political alignment lowers risk and improves effectiveness: Along with the Municipal Government s commitment, political alignment and shared vision at all levels of government Federal, State, and Municipal reduced risks and contributed to the successful outcome. The operation garnered political support by addressing issues of mutual interest in such areas as fiscal consolidation, health, education and public sector management. Interventions targeted at the right level can be highly effective: Direct collaboration at the municipal level was a fruitful strategy for both the Bank and the Brazilian authorities. Local governments in Brazil are responsible for managing education, health and other social services, and the Bank s resources and technical know-how were especially valuable at the Municipal level. Smaller scale projects can generate useful demonstration effects and attribution may be easier than at a national level, enhancing the credibility of the Bank s policy advice. Projects should seek to exploit externalities from knowledge sharing: As a long-time partner, the Bank has become an important repository of Brazil s development experience and many of the DPL s policy areas made use of prior or concurrent work in other Brazilian jurisdictions. Knowledge sharing was a major element of the DPL s value added in such areas as Results Based Management which borrowed from Minas Gerais 27

38 State, and the use of Social Organizations to manage health service delivery which had been pioneered in Sao Paulo. In health and education, where jurisdictions overlap at the State and Municipal levels, the Bank s concurrent engagement in a State DPL was also helpful. The Bank even acted as a knowledge broker more generally for South-South exchanges, as it put the Government of Peru itself dealing with the problem of schools in violent and troubled neighborhoods in touch with Rio about the Escolas do Amanhã program. 28

39 Annex 1 Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Responsibility/ Specialty Lending Yaye Seynabou Sakho Advisor MDI Task Team Leader Pablo Fajnzylber Sector Manager AFTP2 Roland Clarke Sector Leader LCSPR Ngoc-Bich Tran Research Analyst LCSPE Fernando Blanco Senior Economist AFTP4 Cristian Quijada Torres Private Sector Development Specialist LCSPF Barbara Bruns Lead Education Economist LCSHE William Dillinger Lead Economist ECSP4 Andre Medici Senior Health Economist LCSHH David Evans Senior Economist AFRCE Jose Guilherme Reis Sector Coordinator ECSPF Evelyn Levy Consultant Juliana Wenceslau Biriba Tarsila Velloso Public Sector Specialist LCSPS Marcelo Caetano Consultant Mario Rattes Consultant Regis Cunningham Senior Financial Management Specialist EASFM Luciano Wuerzius Procurement Specialist LCSPT Antonio Paulo Vogel de Medeiros Miguel Navarro Martin Lead Financial Officer FABBK Anderson Caputo Silva Lead Securities Market Specialist FCMSM Armando Roselli Ann Tereza Pereira Erica Amorim Zelia Brandt de Oliveira Program Assistant LCC5C Miguel Santiago Senior Finance Officer CTRLN Andresa Lagerborg Research Analyst LCSPE Ricardo Rocha Silveira Kai Kaiser Senior Economist EASPP Joao Pedro Wagner de Azevedo Senior Economist ECSP3 Supervision Edith Kikoni Economist LCSPE Task Team Leader Jorge Thompson Araujo Lead Economist LCSPE Cindy Audiguier Consultant LCSPE Marcelo Caetano Consultant LCSPP Marta Riveira Analyst LCSPE 29

40 (b) Staff Time and Cost Stage Lending Supervision/ICR Staff Time and Cost (Bank Budget Only) USD Thousands (including No. of staff weeks travel and consultant costs) Total: $472, Total: $296,

41 Annex 2. Summary of Borrower's ICR The Development Policy Loan for the Municipality of Rio de Janeiro (MRJ) was the basis of a successful program which, with financial and technical support from the World Bank, achieved various public policy objectives including reforms in the municipal public administration. These goals rested on three pillars: (1) creating fiscal space for investment; (2) improving the quality of public service delivery, especially for the poorer population of the city, through the implementation of innovative programs in the areas of health and education and assisting new entrepreneurs located in the city; and (3) municipal public management. The following are the main results achieved. Pillar 1: Creating Fiscal Space for Investments The implementation of the Electronic Tax Invoices System (Nota Carioca) played a key role in creating fiscal space for municipal investment. A report by the Municipal Finance Secretariat, published on its official website, indicates that the financial yield arising from the introduction of the Nota Carioca was substantial - R$686 million - representing an increase of 6% of the total collection of the Tax on Services (ISS) between July 2010 and June 2013, and exceeding the end target of 4% set in the Loan Agreement. Note that these gains represent a permanent change in the level of ISS collection. As for the pensions system, we achieved a reduction in the actuarial deficit of the municipal pension fund (FUNRPEVI) from R$22 billion (at current values) recorded in 2009 to R$650 million in December 2012, due to the implementation of the FUNPREVI capitalization plan approved by the Municipal Chamber (Law 5.300/2011). Regarding expenditures reduction, Net Personnel Expenditure (referring only to the Executive) represented % of Net Current Revenue in 2012, well below the limit set by Brazil s Fiscal Responsibility Law (54%) and the percentage of NCR recorded in December 2008 (48.69%). Pillar 2: Innovation in Public Services Delivery In the health care area, family health care coverage expanded from 6% of the total population served to 34.63% at the end of 2011, nearly three times higher than the target established in the Loan Agreement (12% in December 2011). By August 2013, family health coverage had reached 39.14% of the city s population. All the results indicators for the education sector were also achieved or exceeded, such as that which provided for an annual increase of 3,000 enrollments in public crèches and pre-schools, from the 2009 baseline. Between 2009 and 2013 an average of 31

42 4,000 new places per year were created in the crèches and pre-schools. Also worth noting is the reduced dropout rate in the Escolas do Amanhã scheme 29, from 5.1% in 2009 to 3.75% in With regard to improving the environment for opening new business ventures in Rio de Janeiro, the average time taken in 2009 to obtain a municipal operating license for business activities with low environmental impact was 20 days. Currently, 95.2% of these applications are dealt with in up to 3 working days. Pillar 3: Municipal Public Management Given the great challenges facing Rio de Janeiro in 2009, we envisaged attracting private partners. This has since become a strategic goal, especially with regard to expanding the delivery of public services and revitalizing specific urban areas of the city. The establishment of the institutional framework for the creation of public-private partnerships began with the approval by the City Chamber, in December 2009, of the law setting up the Municipal Program for Public- Private Partnerships (PROPAR RIO), and its Management Council. Subsequently, in April 2011, a PPP Unit was created in the city, now called the Sub-Secretariat for Strategic Projects and Public Services Concessions and Public-Private Partnerships (CVL/SUBPPP). The basic sanitation concession in the west of the city, signed in January 2012, is the largest of its kind in Brazil. A number of other PPPs and concessions are also important: (i) the concession to build the Ligação Transolímpica, an express transit corridor that will connect Barra da Tijuca to Diodorus, two of the main centers for the 2016 Olympics; (ii) the PPP signed in 2012 for the construction of the Olympic Park over an area of 1.2 million m 2, where 14 Olympic and 9 Paralympic events will take place; and (iii) the concession signed in 2013 for the construction of the Light Rail System (Veículo Leve sobre Trilhos) which will run 28 km from the downtown and port areas of the city. With technical assistance support from the World Bank (Fee Based Service), the City Government has prepared manuals aimed at standardizing procedures and working methods with a view to facilitating liaison between the various agencies and entities involved in the concessions and PPP processes, as well as for strengthening the relationship between the MRJ authorities and the private sector. These manuals are now being used in new projects being developed by CVL/SUBPPP. In 2009, the Strategic Plan of the Municipality of Rio de Janeiro was launched. To ensure good monitoring of this Plan, performance contracts were negotiated across municipal 29 Created in 2009, the Escolas do Amanhã program aims to reduce absenteeism in the schools and improve the performance of students living in problematic areas of the city. It aims to attract 105,000 students in 151 school facilities in 10 pacified communities. The Program provides full-time education, learning support, sporting activities, science laboratories, and classes in healthcare, reading and computer technology. 32

43 agencies and entities. These are a management tool now employed by the public administration to: (i) ensure greater commitment from all municipal agencies to achieving the outcomes desired by the City Government; (ii) institutionalize a new culture that prioritizes planning with clear goals; (iii) foster participation by public employees with the use of a meritocratic model for appraising and rewarding high achievers; and (iv) anticipate problems and find solutions through formal results monitoring. For 2013, 43 official municipal agencies and entities have signed contracts with the Mayor of Rio de Janeiro, embracing 196 goals and involving around 83% of all municipal employees. The Strategic Plan was revised in 2012, setting targets for the period Finally, we should mention the benefits accruing to the management of municipal finances generated by the transfer of US$1.045 billion of funds from the World Bank to the MRJ. The funds were released in two tranches. The first tranche (US$545 million), released in August 2010, allowed the reduction of the interest rate charged by the Federal Treasury on the 1999 municipal debt refinancing agreement from 9% to 7.5% above the IGP-DI. The second release of funds (US$500 million), in November 2011, led to a second reduction in the costs of the refinanced debt, down to IGP-DI (General Price Index Domestic Availability) inflation rate plus 6%. The following graph illustrates the improvement of the MRJ s debt profile, to which the loan agreement signed with the World Bank made an outstanding contribution. It can be seen that in the years leading up to 2010, the city expended around 13% of its Net Real Revenue (NRR) and 10% of its Net Current Revenue (NCR) on servicing its financial debt burden. The two descending curves of the graph reflect the sharp fall in 2011 of NRR and NCR to 8% and 6% respectively, stabilizing at under 10% of NRR and 8% of NCR in the following years up to 2029, by when the debt swap with the Federal Treasury will have completely matured, thereby leading to a further major reduction in the city s annual debt repayments. The Municipal Finance Secretariat estimates savings of up to US$2 billion in debt expenses, due both to the fall in interest rates charged by Federal Government (from 9% to 6% above IGP-DI), and to the longer maturity period (from 2029 to 2040) of the portion of the debt with the Federal Treasury which was amortized thanks to World Bank funding. By September 2013 we had already recorded a reduction of debt expenses amounting to R$ billion. 33

44 Ratios of Debt Expenses (Principal and Interest) compared to Revenues 16.0% Debt Expenses / RLR 14.0% 12.0% 13.2% 2010/2011: World Bank Loan Debt Expenses / RCL 10.0% 10.1% 8.0% 6.0% 4.0% 2.0% 0.0% % While the MRJ was within all the limits of the LRF (Fiscal Responsibility Law) in December 2008, the result improved considerably thereafter, as evidenced in the table below. Indicator Limit June, 2029: maturity of debt with Fed Treasury MRJ OBS - In 2010 and 2011, the accounting effects of the World Bank loans (debt swap with Federal Treasury) were excluded. RLR = Receita Líquida Real (Provisional Measure ); RCL = Receita Corrente Líquida (Fiscal Responsibility Law): measures of current revenues. MRJ 2012 Personnel expenditures (Executive)/Net current revenue (NCR) Gross Debt/ Annual net real revenue (NRR). (MP indicator, not the LRF indicator) Net debt/ Annual net current revenue (NCR) Average annual debt service up to 2027 (NCR) Education expenditures (constitutional definition) 54% Maximum 100% Maximum 120% Maximum 11.5% Maximum 25% Minimum 48.69% 40.69% 118% 75% 58% / 79% * 54% 10.49% % 32.99% ** Health expenditures 15% Minimum 15.73% 23.26% * In 2008, the index was published using pension system funds to reduce the net debt, resulting in 58%. Using the current method (not including pension system funding), the indicator in 2008 would be 79%. ** Percentage obtained using the 2008 calculation method (since amended). The official 2012 figure was 25.68%. 34

45 The improvement of the MRJ s situation, largely due to the decisive structural reforms implemented and the loan agreement with the IBRD, can also be illustrated by comparing MRJ spending allocations between 2009 and The two following graphs show the percentage shares of expenditures of the MRJ in years 2009 and 2012, grouped into Payroll Expenditures, Debt (amortization plus charges), Investment and Financial Investments, and Other Current Expenditure. Municipality of Rio de Janeiro: Expenditure by sector in 2009 and % 0.8% 27.9% 57.8% PERSONNEL & SOC. CHARGES DEBT OTHER CURRENT EXP. INVESTMENT 15.9% 0.8% 45.1% 34.4% PERSONNEL & SOC. CHARGES DEBT OTHER CURRENT EXP. INVESTMENT 9.8% FINANCIAL INVESTMENTS 3.7% FINANC. INVESTMENTS It is clear that there was a healthy trade off in the allocation of municipal resources between the beginning and end of the period of government analyzed, given that in 2009 approximately 68% of expenditures were related to disbursements committed to personnel, social charges and debt service, leaving 31.6% to be used for spending (priority) and investment (residual). In 2012, the reductions in personnel expenditures and municipal debt largely allowed an expansion of public investment, accounting for the significant 15.9% share of total spending, and leading to the aggregated "investments plus expenses" element to emerge as representing the largest share of the MRJ s spending allocation. Since 2011, the MRJ has enjoyed the highest investment rate of all the main (State and Municipal) governments, as shown in the tables below. 35