IEG Report Number: ICRR14890 ICR Review Independent Evaluation Group 1. Project Data: Date Posted: 02/16/2016 Country: Brazil Project ID: P101324 Appraisal Actual Project Name: Minas Gerais Project Costs (US$M): 6,442.93 N/A Partnership Ii Swap L/C Number: Loan/Credit (US$M): 976.00 1,434.88 Sector Board: Public Sector Cofinancing (US$M): Governance Cofinanciers: Board Approval Date : 05/01/2008 Closing Date: 06/30/2011 10/31/2014 Sector(s): Sub-national government administration (35%); Rural and Inter-Urban Roads and Highways (25%); General education sector (15%); Health (15%); General industry and trade sector (10%) Theme(s): Public expenditure; financial management and procurement (25%); Administrative and civil service reform (25%); Managing for development results (24%); Other economic management (13%); Regulation and competition policy (13%) Prepared by: Reviewed by: ICR Review Group: Coordinator: Stefano Migliorisi Robert Mark Lacey Lourdes N. Pagaran IEGPS2 2. Project Objectives and Components: a. Objectives: The Project Appraisal Document ( PAD, p. 9-10) states that the objectives of the project are: (1) “to help the Government of Minas Gerais to improve the efficiency of public resource use (getting more out of each real spent) and the effectiveness of public resource allocation”; (2) “to support the adoption of innovations in public management of the State” and (3) “to support the Government of Minas Gerais in strengthening the system of monitoring and evaluation of results.” According to the Loan Agreement – LA (p. 6), “The objectives of the Project are: (a) to help the Borrower improve the efficiency and effectiveness of public resource use and allocation for economic and social development; (b) to support the adoption of innovations in public management by the Borrower; and (c) to support the Borrower in strengthening its results-based management system of monitoring and evaluation of results.” This Review is based on the statement of objectives in the Loan Agreement. Additional Financing was approved on April 6, 2010, in the amount of US$461 million. This was in order to withstand the unanticipated deep global and domestic recession and fiscal challenges resulting from significant drops in state and federal revenues. The project development objectives remained unchanged. b.Were the project objectives/key associated outcome targets revised during implementation? No c. Components: The project had two components:  ·Component 1: SWAp (Original Cost: US$6,424.43 million; Revised Costs: US$7,846.18; Actual Expenditure: N/A).This component was designed to disburse against Eligible Expenditure Programs (EEP) in five sectors (i.e., public sector, private sector development; health; education; and infrastructure – transport). In addition to these five sectors, there were indicators for strengthening implementation of environmental and social capacity building. Disbursement was dependent on the achievement of 24 Disbursement-Linked Indicators (DLIs) measuring the progress of the EEPs.  ·Component 2: Technical Assistance (Cost: US$18.5 million; Actual Expenditure: US$17.81). This component supported the implementation of EEPs and the achievement of the DLIs Revised Project The original components were maintained, but a sixth sector (rural poverty) was added to Component 1. A Rural Poverty Reduction Program was included as an EEP and linked to five new DLIs to measure rural poverty reduction. Three other DLIs were added for other five sectors bringing the total to 32. d. Comments on Project Cost, Financing, Borrower Contribution, and Dates: Cost Total project costs at appraisal were US$6,442.93 million. Total actual costs were US$7,864.68 million. No actual cost data are available for the activities funded by the Borrower or on expenditures by sector. Financing The Bank loan was increased to US$1,437 million by the Additional Financing of US$461 million. This covered those project costs financed by the Bank plus IBRD’s front-end fees of US$3.59 million. A total of US$ 1,434.88 million was disbursed (99.9% of commitment). Borrower Contribution The Borrower Contribution was estimated at US$5,469.37 at appraisal and revised to US$6,431.27 million when the Additional Financing was approved. No data is available on the actual contribution from the Borrower. Dates The project closing date was extended five times from June 30, 2011 to March 2012, March 2013, December 2013, October 2014 and finally to October 31, 2014, when the project closed. The first extension was made at the approval of the Additional Financing to allow for disbursement and implementation of the scaled up activities, while the other four were due to delays in the procurement of consultants and implementation of key technical assistance activities. 3. Relevance of Objectives & Design: a. Relevance of Objectives: Substantial. The objectives are relevant to the World Bank Group’s Country Partnership Strategy (CPS) for Brazil covering the Fiscal Years 2012-2015. At least two out of four objectives of the strategy were pursued through the project: objective 1 (i.e., increase the efficiency of public and private investments – with a particular focus on State-level results as shown in the CPS Results Matrix) and objective 2 (i.e., improve the quality of public services for low income households, and expand their provision through public and private channels). The objectives were also consistent with the CPS at appraisal (Fiscal Years 2008-2011), which highlighted the need for an increased Bank’s role at State level, and emphasized the importance of addressing complex development challenges in an integrated and multi-sector fashion, as attempted by the project. The objectives were relevant to the second phase of the Minas Gerais Government’s reform program “Estado para Resultados - State for Results” (2007-2010) - stressing efficiency and results; with its third phase “Estado em Rede - Management for Citizenship” (2011-2014) - increasing efficiency, building a permanent constituency through citizen participation, and decentralizing performance management to individual secretariats; and with the Government Action Plan (PPAG) for 2012-2015. However, the new government elected in January 2015 was against the reform program while in opposition. At project closure, it had not yet formulated a clear strategy on the next phase of reform, although health and education are likely to remain priority areas. b. Relevance of Design: Substantial. Shortcomings in the articulation of the project development objectives affected the quality of the project’s monitoring and evaluation framework as discussed in Section 10 below. The second and third objectives were intermediate objectives towards the achievement of the first. The objectives were also ambitious given State conditions (i.e., the transition to results-based management had just started, and the previously bankrupt state had been allowed to borrow again only two years before project approval), as the project tried to address several key issues simultaneously across six sectors. Notwithstanding these limitations, the project presented a convincing causal chain between inputs, outputs and outcomes with a series of relevant indicators across six sectors. The implementation of results agreement, for example, was linked to evidence-based decision-making in eight strategic projects that were used to monitor performance in the results agreements, as well as to impact evaluations, electronic contract management, and the recertification of public sector positions. The choice of lending instrument (Sector Investment Loan blending programmatic support for Eligible Expenditure Programs and investment lending for capacity building) was appropriate. 4. Achievement of Objectives (Efficacy): 1. Help the Borrower improve the efficiency and effectiveness of public resource use and allocation for economic and social development . Substantial.  Efficiency of public resource use for economic and social development was improved through better measurement and tracking of costs in both health and education, and extensive procurement reforms, that were completed under the project and rolled out to all State purchasing units by 2014. The procurement reforms carried out under the project (i.e., rationalization of purchase and contracting procedures; introduction of pre-purchasing procedures; online publication of a unified procurement calendar) were part of a wider program started in 2004, and supported by the Bank through policy lending and technical assistance. These reforms led to savings of US$ 180 million per year due to increased competition, better contract management, and a strategic approach to procurement with a reduction of direct contracts from almost 13,000 in 2004 to less than 1,200 in 2009. In addition, the average percentage delay of paving works concluded under the Pro-Acesso road program (a State government program that became an Eligible Expenditure Program supported by the project) decreased from 40 percent in 2007 to around 10 percent by project closing, showing the positive impact of the project on the programs it supported.  The share of “strategic projects” over total budget expenditures increased from less than 7 percent in 2007 to almost 17 percent at project closure . This was considered to be an indicator of enhanced efficiency of public resource allocation, and was to stem from the adoption of project-supported results-based management systems. However, as the ICR (page 29) acknowledges, the concept of “strategic projects” was not clearly defined, thereby facilitating manipulation of the indicator. In addition, there was high staff turnover in the State civil service, and the method for calculating the indicator was not consistent over time.  The State’s strong fiscal-management system, as well as greater transparency in the award of contracts, led Standard & Poor’s in 2012 to raise the state’s credit rating to investment-grade. The accompanying press release emphasized that “Minas Gerais' financial management compares favorably with those of other states in terms of medium-term financial planning, forward-looking perspective on its cash flows, and debt management.”  As noted in the review of State-level SWAps carried out as part of the IEG Country Program Evaluation for Brazil (p. 83), “The work on local government finances and fiscal reforms for growth represented a good example of informal analytical work that encouraged a candid exchange of views in some critical areas.” Attribution to Bank projects could only be “catalytic either through engaging in sustained dialogue over a long period or fostering replication and dissemination of good practices” as was the case in this operation. 2. Support the adoption of innovations in public management by the Borrower . Substantial  In Public Sector Management , the project supported the introduction of results-based management through the adoption of first stage and second stage Results Agreements, the impact of which on economic and social development is discussed in the following paragraphs. First-stage results agreements were signed between the Governor and Line Secretaries where the Governor committed to provide a certain amount of resources and the Secretaries to achieve at least 70 percent of the targets included in their agreement. Second-phase results agreements were signed between the heads of Secretaries, and their subordinates, grouped in teams. By project end, 28 Secretaries (against an original target of 7 and a baseline of 4) had implemented first-phase results agreement agreements and achieved a performance of at least 70 percent, and 61 second-phase results agreement had been implemented (against a baseline of 10 and a target of 11).  The introduction of results-based management through first stage and second stage Results Agreements had a measurable positive impact only in education. Regression analysis carried out for the ICR concluded that states like Minas Gerais that introduced performance agreements and a bonus for teachers and school staff improved their Index of Basic Education Development (IDEB) score for public secondary schools by 0.26 additional points compared to states that did not do so. Similar results were obtained for security expenditure. In contrast, improvements in health outcomes, as measured by changes in child mortality, were very similar among states, and the difference between Minas Gerais and other states remained fairly constant.  The project supported the introduction of Public-Private Partnerships (PPPs) through the establishment of a PPP Unit in the Secretariat of Economic Development and provided training to its staff members. PPPs completed included, inter alia, a prison, a highway and a soccer stadium, and their number increased from one in 2007 to six in 2013 (against a target of three), making Minas Gerais the most advanced Brazilian State in this field. IEG’s Brazil Country Program Evaluation FY2004-11 concluded that the project “supported steps to implement PPPs and achieved important progress” (p xvi), and that “the shift toward multisector operations at the subnational level contributed to increasing private investment in infrastructure and reducing the cost of doing business in such places as the state of Minas Gerais.” (p. 75).  The project supported the introduction of results-based maintenance and rehabilitation contracts for State paved highway roads. Such contracts surpassed its target of 42 percent of the total value awarded, while at the same time the proportion of highway roads in good conditions grew from 43 percent in 2007 to 72 percent in 2011. 3. Support the Borrower in strengthening its results -based management system of monitoring and evaluation of results. Substantial. The project supported the establishment and operation of the monitoring and evaluation system, as well as three impact evaluations. In particular, the project was instrumental in the successful introduction of the first and second generation of results agreements through technical assistance activities and the engagement of Bank specialists with the Government team. The combination of better information on both costs and results in key sectors, and the use of results agreements with specific indicators to measure institutional and team performance and release funds, provided both the information and incentives for an effective monitoring and evaluation of results. The ICR provides extensive information and evidence of the significance of the results-based management reforms in an Annex. However, there is little evidence of significant evaluation activities that could have improved learning in the health sector, as the impact evaluation of Viva Vida Centers was excluded from the project in 2011 while they were completed for project activities in transport and education. 5. Efficiency: Modest. The project appraisal included an economic analysis determining an NPV and ERR for education and transport, and the ICR carried out a simplified ex-post economic analysis for the transport sector, showing higher economic rates of return than those estimated at appraisal. Their exact values of the latter were not disclosed. The ICR (page 43) states that “conducting a rigorous economic analysis is not viable due to the lack of an appropriate counterfactual. However, considering the achievement of most of the PDO and Intermediate indicators, the overall efficiency of the project can be considered rather high. ” Achievement of development objectives and attainment of indicator targets refers to efficacy, not efficiency. Assessing efficiency means comparing the benefits with the cost of achieving them. In this case, the ICR does not provide any cost comparisons with similar operations or any other way of assessing cost effectiveness. Consequently, "value for money" for this project cannot be established. Although some of the 40 month extension of the closing date could be explained by activities supported by the Additional Financing, there were also delays related to technical assistance contracts and safeguards compliance (see Sections 8b and 9b below). a. If available, enter the Economic Rate of Return (ERR)/Financial Rate of Return (FRR) at appraisal and the re-estimated value at evaluation : Rate Available? Point Value Coverage/Scope* Appraisal Yes 26.4% 23% ICR estimate No * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome: Relevance of objectives and of design was substantial. The project substantially achieved two of its three objectives. It contributed to stronger and more transparent public resource use and allocation, and supported relevant innovations in service delivery across several sectors. Efficacy of the three objectives was, substantial. Efficiency is rated modest given the limited evidence of “value for money.” Overall outcome is assessed as moderately satisfactory. a. Outcome Rating: Moderately Satisfactory 7. Rationale for Risk to Development Outcome Rating: While the main achievements of the project (e.g., a culture and systems of results-based management) would be difficult to reverse, there are significant risks of policy-reversal by the new government, and of a deepening of the macroeconomic difficulties that had already affected the State There could be a consequent weakening of discipline in public resource management. a. Risk to Development Outcome Rating : Significant 8. Assessment of Bank Performance: a. Quality at entry: The project’s scope was very wide as it covered all of the State’s borrowing space, and addressed six different sectors in a State whose economy was the size of Peru. The SWAp modality was an innovation for Minas Gerias. However, the approach followed in the SWAp component was relatively simple: disbursement occurred at three to six month intervals, depending on the achievement of 24 Disbursement-Linked Indicators (DLIs) (later increased to 32) measuring the progress of the EEPs. Most indicators were clear and the size of the operation ensured that the fiscal impact of missing a disbursement, because not all DLIs were met, would have been substantial. that the impact would have been smaller in the case of separate loans. The choice of a single lending instrument was intended to facilitate rapid progress across the board, while fostering strong coordination among, and overcoming resistance in, Line Secretariats through peer pressure. IEG’s Country Program Evaluation that concluded that this approach was appropriate: “because of their multisectoral nature, the loans proved ideal for addressing institutional reforms on state public finances that cut across sectors, particularly those involving difficult trade-offs and, hence, consensus across different agencies and stakeholders.” (p. 238). The project’s results framework included some outcome indicators on service delivery, and was focused on improving the State’s capacity for results management. The project was prepared in close collaboration with the government counterparts. While the multi-sector nature of the project made it challenging to establish and measure an overarching outcome, there was an attempt to overcome this difficulty by breaking the development objectives down into three intermediate objectives. Some of the outcome level indicators measured outputs rather than outcomes and the number of indicators measuring sector achievements did not reflect the relative share of the sectorial EEPs, leading to a disconnect between the scope of project activities and PDO. Quality-at-Entry Rating: Satisfactory b. Quality of supervision: There were a total of thirteen supervision missions over 6.5 years or once every six months on average. For the period of disbursement of the SWAp component (2008-2011), one of the co-TTLs was based in the Brazil Country Office. The Bank was quick in replying to the State’s request for Additional Financing once the financial crisis reduced its revenues as well as its ability to fund key programs: it took about five months from Borrower request to Board approval. Financial management issues were resolved by close Bank supervision. However, once the SWAp component had been fully disbursed in 2011 and only TA activities remained to be carried out, the supervision effort diminished. The technical assistance component was much slower in disbursing, and the Borrower complained in its comments to the ICR that the Bank was sometimes tardy in sending letters of no objection thus delaying the hiring of personnel. As noted in the ICR, the quality of the supervision reports varied over the implementation period and sometimes lacked information on the results framework. During the extension periods, implementation of safeguards was not followed as closely as previously, with the result that the delays on completion of the compensation processes was not flagged as early as it could have been. Finally, a number of indicators appeared in the ISRs under different names compared to the PAD or were included only in later years. Quality of Supervision Rating : Moderately Satisfactory Overall Bank Performance Rating : Moderately Satisfactory 9. Assessment of Borrower Performance: a. Government Performance: The Government was fully committed to the project. It created the pre-conditions for its launch through the first phase of its public sector reform efforts (e.g., reduction in the number of secretariats and redistribution of responsibilities, civil service pay reform, introduction of management by results, procurement reform) that improved adherence to the Fiscal Responsibility law and opened up the necessary borrowing space. It assigned a qualified team, which participated in project design and supported its conclusions. Strong support was provided during implementation. The Government requested an additional financing rather than cutting programs included in the original design it could no longer fund due to the global financial crisis. Government Performance Rating Satisfactory b. Implementing Agency Performance: The implementing agency monitored performance of the various programs, managed technical assistance, and coordinated the work of Secretariats across several sectors. There were some weaknesses on fiduciary management due to high staff turnover, and on compliance with social safeguards due to the limited attention paid to the speed of compensation for involuntary resettlement (see Section 11 below). However, staff turnover would have been difficult to avoid given the length of the implementation period. Implementing Agency Performance Rating : Satisfactory Overall Borrower Performance Rating : Satisfactory 10. M&E Design, Implementation, & Utilization: a. M&E Design: Some of the PDO level indicators measured outputs rather than outcomes and the number of indicators measuring sector achievements were not reflecting the relative share of the sectorial EEPs, leading to a slight disconnect between the scope of project activities and PDO. Some indicators were repeated as both outcome and intermediate (for example, the number of accredited hospitals). Most indicators had a baseline and a clear timing of targets. As most indicators were already monitored by Government, the project did not create duplication of efforts in M&E. b. M&E Implementation: Day-to-day monitoring was carried out by the various Secretariats under the supervision of the Results Management Unit of the Planning and Management Secretariat. The implementation of the M&E Framework was weak on evaluation, as discussed in Section 4. In addition, as noted in the ICR (p. 8), “a number of indicators appear in the ISRs under different names as those in the original project documents, and some were included in the ISRs only in later years.” c. M&E Utilization: Data gathered through the M&E framework were used to monitor progress. The Bank used supervision missions and interviews to fill any gaps. M&E Quality Rating: Modest 11. Other Issues a. Safeguards: The project was classified as environmental category B. Other than Environmental Assessment (OP 4.01), the original project triggered three safeguards policies – Physical Cultural Resources (OP 4.11), Involuntary Resettlement (OP 4.12), and Indigenous Peoples (OP 4.10). When the Additional Financing was approved, two further safeguards policies were triggered: Pest Management (OP 4.09) and Natural Habitats (OP 4.04). According to the ICR (page 9), project activities followed the Environmental Management Framework agreed with the World Bank and there were no compliance issues. However, implementation of social safeguards had some shortcomings: “Guidance provided by the Resettlement Policy Framework (RPF) in general ensured compliance of the policy until the stage of compensations to affected people. However, at the time of the preparation of the ICR, about 18 percent (or 500 of 2,776 processes) of the compensation processes were not yet concluded and were in an administrative completion stage. The Government has elaborated an action plan to finalize these processes. The Bank will continue to closely monitor the implementation of that action plan until the pending compensation processes are fully and satisfactorily completed” (ICR, page 8). No information is provided in the ICR concerning compliance with the other safeguard policies triggered (OP 4.11, OP 4.10,, OP 4.09 and OP 4.04. b. Fiduciary Compliance: The ICR reports no significant fiduciary issues during the project’s implementation. Procurement arrangements were considered satisfactory throughout project implementation. Financial management was more challenging, due to the complex reporting rules on compliance with EEPs, and the need to adapt the State’s existing financial reporting system to specific Bank requirements. As noted in the ICR (p. 9), “Initial weaknesses in the reporting system, the design of IFRs, and the use of spreadsheets to prepare financial statements led to data inconsistences, which were resolved by close Bank support during implementation.” In addition, “the complex reporting rules on the compliance with the eligible expenditure programs also posed challenges during implementation and sometimes caused delays in disbursements.” The ICR (page 9) reports that “the lack of familiarity with the Bank’s auditing arrangements during project preparation was overcome during implementation. The Tribunal de Contas do Estado de Minas Gerais (TCE-MG, responsible for external audits of the project) was an important partner in strengthening executing agencies’ capacity and producing high quality auditing reports,” although the Borrower’s comments on the ICR (ICR, page 49 state that “the TCE’s audit process was significantly hampered by the high turnover rate of State employees. “ No information is provided in the ICR as to whether the auditors’ opinions were qualified. c. Unintended Impacts (positive or negative): None. d. Other: 12. Ratings: ICR IEG Review Reason for Disagreement/Comments Outcome: Satisfactory Moderately Relevance of both objectives and Satisfactory design is rated substantial, as is that of the efficacy of all three objectives. Efficiency, however, is rated modest given the absence of any evidence of “value for money.” Risk to Development Moderate Significant There are significant policy reversal and Outcome: macroeconomic risks. Bank Performance: Satisfactory Moderately There were moderate shortcomings in Satisfactory the Quality of Supervision. Borrower Performance : Highly Satisfactory Satisfactory There were minor shortcomings in implementing agency performance. Quality of ICR: Satisfactory NOTES: - When insufficient information is provided by the Bank for IEG to arrive at a clear rating, IEG will downgrade the relevant ratings as warranted beginning July 1, 2006. - The "Reason for Disagreement/Comments" column could cross-reference other sections of the ICR Review, as appropriate. 13. Lessons: There are four main lessons that can be drawn from the experience of this project:  Political commitment is key in a project that aims at supporting public resource management reform , where some degree of resistance from line ministries is to be expected. However support from top management at key Ministries is as crucial, particularly for institutional reform projects where the timeframe of implementation is usually very long. Political commitment can in fact vary with changes in government, but top management support ensures deeper rooting of reforms, and makes policy reversals more difficult.  Large multi-sector SWAPs with disbursement -linked indicators can offer stronger incentives for reform , ensuring that a wider array of policy changes can be implemented more smoothly and expeditiously , as noted in the ICR. Sector-specific SWAPs cannot address as well issues that require cooperation of different line ministries, some of which may not have any financial stake in the successful implementation of the reforms supported by the project.  Impact evaluations are key for learning and they should be properly timed . Impact evaluations in multi-sector SWAPs are particularly challenging, but should not be replaced by partial evaluation carried out immediately after the SWAP closes. The presence of a Technical Assistance component offers the opportunity for diversified completion schedules. The Bank’s know-how can be as important as its ability to mobilize significant financial resources . The Government of Minas Gerais – see ICR p. 22 - made clear that one of the key factors for choosing the Bank as the project’s financier was its ability to provide valuable inputs on the design of the program, and advice during its implementation. 14. Assessment Recommended? Yes No 15. Comments on Quality of ICR: The ICR is candid and provides useful information to assess project performance. However, it could have attempted a more quantitative assessment of efficiency such as that made at appraisal or provided some other means of judging cost effectiveness. In its analysis of efficacy, the ICR relies too much on indicators instead of using all available data to evaluate outcomes. There is no discussion of compliance with four of the safeguards policies triggered, and no information regarding whether the external auditors’ opinions were qualified or not. a.Quality of ICR Rating : Satisfactory