Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004310 IMPLEMENTATION COMPLETION AND RESULTS REPORT (LN8107-MX) ON A LOAN IN THE AMOUNT OF US$ 100 MILLION TO THE UNITED MEXICAN STATES FOR THE SAVINGS AND CREDIT SECTOR CONSOLIDATION AND FINANCIAL INCLUSION PROJECT ( P123367 ) January 12, 2018 Finance, Competitiveness, and Innovation Global Practice Latin America And Caribbean Region CURRENCY EQUIVALENTS (Exchange Rate Effective July 21, 2017) Currency Unit = Mexican Peso (MX$) MX$ 17.56= US$ 1 US$ 0.06= MX$ 1 FISCAL YEAR July 1 - June 30 Regional Vice President: Jorge Familiar Acting Country Director: Jutta U. Kern Senior Global Practice Director: Ceyla Pazarbasioglu-Dutz Practice Manager: Zafer Mustafaoglu Task Team Leaders: Patricia Caraballo, Daniel Ortiz del Salto ICR Contributor: Jose Isaac Rutman ABBREVIATIONS AND ACRONYMS AML Anti-money Laundering BANSEFI National Bank for Savings and Financial Services (Banco del Ahorro Nacional y Servicios Financieros, Sociedad Nacional de Crédito, Institución de Banca de Desarrollo) BTP BANSEFI Technological Platform CAR Capital asset Ratio CNBV National Banking and Securities Commission (Comisión Nacional Bancaria y de Valores) CPS Country Partnership Strategy ENIF National Survey of Financial Inclusion (Encuesta Nacional de Inclusión Financiera) GDP Gross Domestic Product FM Financial Management GoM Government of Mexico IBRD International Bank for Reconstruction and Development ICR Implementation Completion and Results Report IFRs Interim unaudited Financial Report ISR Implementation Status and Results report IT Information Technology LACP Popular Savings and Credit Law (Ley de Ahorro y Crédito Popular) LRASCAP Law that regulates the activities of Savings and Loan Cooperatives (Ley para Regular las Actividades de las Sociedades y Cooperativas de Ahorro y Préstamo) M&E Monitoring and Evaluation NPL Non-performing loan PAD Project Appraisal Document PATMIR Regional Rural Microfinance Technical Assistance Program (Programa de Asistencia Técnica a las Microfinanzas Rurales) PATyC TA and Training Program (Programa de Asistencia Técnica y Capacitación) PDO Program Development Objective PROIIF Comprehensive Financial Inclusion Program (Programa Integral de Inclusión Financiera) PROSPERA Social Inclusion Program (Programa de Inclusión Social) SCEs Saving and Credit Entities SCIs Saving and Credit Institutions SHCP Ministry of Finance (Secretaría de Hacienda y Crédito Público) SOCAP Financial Cooperatives (Sociedades Cooperativas de Ahorro y Préstamo) SOFINCOS Community Financial Societies (Sociedades Financieras Comunitarias) SOFIPOS Popular Financial Societies (Sociedades Financieras Populares) TA Technical Assistance WB World Bank TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 5 A. CONTEXT AT APPRAISAL .........................................................................................................5 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) .......................................8 II. OUTCOME ...................................................................................................................... 9 A. RELEVANCE OF PDOs ..............................................................................................................9 B. ACHIEVEMENT OF PDOs (EFFICACY) ........................................................................................9 C. EFFICIENCY ........................................................................................................................... 14 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 15 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 15 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 17 A. KEY FACTORS DURING PREPARATION ................................................................................... 17 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 17 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 19 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 19 M&E Implementation ............................................................................................................... 19 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 20 C. BANK PERFORMANCE ........................................................................................................... 20 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 21 V. LESSONS AND RECOMMENDATIONS ............................................................................. 22 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 24 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 35 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 37 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 39 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 41 ANNEX 6. BIBLIOGRAPHY ..................................................................................................... 42 ANNEX 7. SUPPORTING INFORMATION ................................................................................ 43 ANNEX 8. THEORY OF CHANGE ............................................................................................ 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name SAVINGS AND CREDIT SECTOR CONSOLIDATION AND P123367 FINANCIAL INCLUSION PROJECT ( P123367 ) Country Financing Instrument Mexico Specific Investment Loan Original EA Category Revised EA Category Not Required (C) Not Required (C) Organizations Borrower Implementing Agency Government of Mexico, Secretaria de Hacienda y BANSEFI Credito Publico Project Development Objective (PDO) Original PDO The objective of the project is to support the: (a) consolidation of the savings and credit sector institutions (SCI); and (b) deepening of financial inclusion in selected areas and selected locations in Mexico. Page 1 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 100,000,000 69,570,296 69,570,296 IBRD-81070 Total 100,000,000 69,570,296 69,570,296 Non-World Bank Financing Borrower 94,320,000 0 64,088,425 LOCAL: BENEFICIARIES 15,300,000 0 21,000,000 Total 109,620,000 0 85,088,425 Total Project Cost 209,620,000 69,570,296 154,658,721 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 01-Dec-2011 02-Apr-2012 18-Nov-2013 31-Jul-2015 21-Jul-2017 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 31-Jul-2014 35.43 Change in Results Framework Change in Loan Closing Date(s) 13-Dec-2016 60.71 Change in Results Framework Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Moderately Satisfactory Substantial Page 2 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 28-Mar-2012 Satisfactory Satisfactory 0 02 09-Nov-2012 Satisfactory Satisfactory 1.86 03 22-Jun-2013 Satisfactory Satisfactory 11.57 04 30-Dec-2013 Satisfactory Moderately Satisfactory 26.98 05 04-Jul-2014 Satisfactory Moderately Satisfactory 35.68 06 25-Jan-2015 Satisfactory Moderately Satisfactory 42.11 07 18-Sep-2015 Satisfactory Moderately Satisfactory 50.05 08 04-Apr-2016 Satisfactory Moderately Satisfactory 54.64 09 06-Oct-2016 Satisfactory Moderately Satisfactory 60.96 10 24-Apr-2017 Satisfactory Moderately Satisfactory 64.23 11 20-Jul-2017 Satisfactory Moderately Satisfactory 69.57 SECTORS AND THEMES Sectors Major Sector/Sector (%) Financial Sector 100 Banking Institutions 55 Other Non-bank Financial Institutions 18 Public Administration - Financial Sector 27 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Private Sector Development 17 Enterprise Development 17 MSME Development 17 Page 3 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Finance 17 Financial Infrastructure and Access 17 MSME Finance 17 Social Development and Protection 17 Social Inclusion 17 Indigenous People and Ethnic Minorities 17 Human Development and Gender 17 Gender 17 Urban and Rural Development 33 Rural Development 33 Rural Markets 16 Rural Non-farm Income Generation 17 ADM STAFF Role At Approval At ICR Regional Vice President: Pamela Cox Jorge Familiar Calderon Country Director: Gloria M. Grandolini Jutta Ursula Kern Senior Global Practice Director: Ede Jorge Ijjasz-Vasquez Ceyla Pazarbasioglu-Dutz Practice Manager: Ethel Sennhauser Zafer Mustafaoglu Patricia Caraballo, Daniel Ortiz Task Team Leader(s): Pierre Olivier Colleye del Salto ICR Contributing Author: Jose Isaac Rutman Page 4 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context 1. Financial inclusion, as a key enabler to reducing poverty and boosting prosperity, has been one of the top issues on the development agenda for Mexico. Overall, in 2011, the year the Project was approved, of the 112 million population, about 71 million (65 percent) was estimated to have no access to formal financial services. The Mexican financial sector was relatively small with credit to the private sector at 19.6 percent of Gross Domestic Product (GDP) compared to the regional average of 33.6 percent, and the penetration rate was low, with only 65 percent of municipalities with at least one point of service. Lack of access to financial services was a particularly serious issue in rural municipalities with only 28 percent of municipalities with at least one point of service. Given the wide dispersion of the Savings and Credit Entities (SCEs)1, and their deep focus on rural (and marginal) areas, authorities wanted to provide a stimulus to financial inclusion in rural areas through this segment, in order to support overall economic growth and employment. 2. One of the main challenges to financial inclusion was that a significant number of -mostly rural- clients (around 2.4 million) were operating with unauthorized non-regulated and supervised entities (with no deposit insurance), with their savings at risk and limited access to proper and sound financial services. As of December 2011, only 6.1 million clients were members of an authorized SCE (covered by a deposit insurance scheme), representing around 72 percent of total of assets and 74 percent of clients of the sector. Additionally, financial inclusion in the rural areas was very limited, with only 13.6 percent of the adult population having a formal saving account compared to the regional average of 33.6 percent2. Under this context, it was deemed necessary to support the consolidation of the SCEs through several activities that enable the strengthening and authorization of the SCEs, as well as to support the implementation of policies that facilitate the access to financial services –in a sound and viable manner- for certain segments of the population and regions. 3. It is important to highlight that for over 15 years the World Bank (WB) has been working closely with the Government of Mexico for the improvement of the savings and credit sector with significant results. The National Bank for Savings and Financial Services (BANSEFI), the Mexico’s development bank for improving access of households to financial services, was selected by the Government to be the implementing agency. BANSEFI I project (P070108), approved in June 2002, took a dual approach of strengthening the SCE´ sector and directly supporting its growth in rural areas to foster financial inclusion. Through supporting the development of an inventory and assessment of the sector, providing a first round of training and strengthening to the sector’s institutions, the development of a first blueprint for the information system, and carrying out dissemination activities, it laid the ground for a second Bank project. The BANSEFI II project (P087152) had a narrower focus, providing additional funding for strengthening and consolidating 1 The SCEs consist of popular financial societies (SOFIPOs), community financial societies (SOFINCOs) and financial cooperatives (SOCAPs), with the latter accounting for almost 85 percent of the member’s sector and operating mostly in rural areas through a single branch or a multiple- branch network. These types of institutions offer a natural comparative advantage in expanding outreach due to their wide geographical dispersion, particularly in marginal areas. These entities have 60 percent of their branches in small, semi-urban and rural localities of up to 50,000 inhabitants playing a key role for the provision of financial services. 2 See Annex 7.1 for a benchmark with peer countries. Page 5 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) the sector to bring it under regulation, and supporting the implementation and refinement of a technical platform that would support the harmonization within the sector and the development of joint products and services. Finally, in December 2011, BANSEFI III (the project considered under this Implementation Completion and Results Report, ICR) was approved to provide additional support for the sector’s strengthening and consolidation, and fostering the outreach of regulated SCEs into unbanked rural areas.3 4. At the time of appraisal, the Project fit well with Mexico’s Country Partnership Strategy (CPS) 2008-20134 and the CPS Progress Report 2008-20105. This Project met the CPS objectives by supporting two of the strategic pillars: a) competitiveness; and b) social inclusion and poverty reduction. Theory of Change (Results Chain) Project Development Objectives (PDOs) 5. The objectives of the Project were to support the: (a) consolidation of the savings and credit institutions (SCIs6); and (b) deepening of financial inclusion in selected areas and selected locations 7. The PDO was not revised during project implementation. 3 The series of BANSEFI projects consisted of: BANSEFI I -Savings and Credit Sector Strengthening and Rural Microfinance Capacity Building, US$ 79 million (2001); BANSEFI II - Mexico Savings and Rural Finance Project Phase II, US$ 75 million (2004), with two additional financing projects for US$ 29 million (2007) and US$ 50 million (2009); and BANSEFI III – Savings and Credit Consolidation and Financial Inclusion (2011). 4 Report No. 42846-MX 5 Report No. 52776-MX 6 Savings and Credit Institutions (SCIs) institutions consist of savings and credit entities (SCEs), the Confederation, Federations, Self-Regulatory Body, Auxiliary Supervision Committee, and Deposit Protection Funds. 7 Selected Area means any marginal area with population of less than 100,000 inhabitants. Page 6 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Key Expected Outcomes and Outcome Indicators 6. The key outcomes and PDO indicators to be used to assess the achievement of the Project’s development objectives are: Outcome 1: “Support the consolidation of the savings and credit institutions (SCIs)” a) Number of entities certified by CNBV, the regulator. b) Number of entities merged or liquidated. c) Number of reporting certified institutions with a capital adequacy ratio 150 percent of minimum requirement or above. d) Average of non-performing loans for certified institutions. Outcome 2: “Support the deepening of financial inclusion in selected areas and selected locations” a) Clients mainstreamed into formal financial sector. b) Number of service points providing financial services. c) Number of members financially included through BANSEFI’s Regional Rural Microfinance Technical Assistance Program (PATMIR) (by gender). 7. The results framework (in Annex 1) also included a range of intermediate outcome indicators, which were directly linked to the expected outcomes under individual subcomponents. Components 8. The savings and credit sector consolidation and financial inclusion project included four main components8: 9. Component 1: Consolidation and strengthening of savings and credit institutions (SCIs). Under this component, the Project contemplated the provision of support to: (i) facilitate the transition of the SCEs from not certified to certified entities by the National Banking and Securities Commission (CNBV); help an orderly exit of those financial institutions that would not be certified by the authorities; and finance technical assistance and training to help the SCIs overall institutional strengthening and/or services improvement; (ii) finance goods, technical assistance and training to strengthen the sector's second and third tier institutions (i.e. Confederation, Federations and Auxiliary Supervision Committee) to perform their roles; (iii) carry out sector-related studies and evaluations to establish a consolidated database on the sector; (iv) carry out dissemination activities related to certification and reporting requirements. 10. Component 2: Broadening access to financial services and products. Under this component, the Project considered the provision of support to: (i) expand the membership base in existing entities (PATMIR); (ii) expand the membership base through agency banking in unserved or underserved municipalities (Corresponsales); (iii) expand the range of financial services to entities and clients (l@ Red de la Gente9), and (iv) finance a financial education initiative. 11. Component 3: Strengthening BANSEFI capacity. Under this component, the Project considered the provision of support to: (i) expand the technological capacity of BANSEFI, including its information technology platform's mainframe, 8See Annex 7.2 for additional information about activities by component. 9L@Red de la Gente, is a network of entities that offers a range of financial services, such as international and national remittances, third-party payments, micro-insurance, among others. The number of people served by the network grew from 9 million in 2011 to more than 19 million people in 2017 (4.7 million SCEs’ clients and 14 million BANSEFI’s clients). There are 180 entities in the network serving cl ients through 2,300 branches located in 933 municipalities. Page 7 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) to enhance its processing and storage capacity and efficiency and provide technical assistance (TA) and training to SCEs for adopting BANSEFI's information technology platform; (ii) train BANSEFI's staff on topics related to the Project; and, (iii) disseminate the functional capabilities of BANSEFI's information technology (IT) platform and on BANSEFI's brand as a development bank. 12. Component 4: Project management. The objective of this component was to assist in the implementation of the Project by providing technical assistance, training and material support to BANSEFI's staff, consultants and providers of non-consulting services. 13. By the end of the Project, US$ 172.9 million was invested compared to the US$ 209.6 million estimated during project preparation. The estimated budget for the project included a US$ 100 million loan from the International Bank for Reconstruction and Development (IBRD), US$ 94.3 million funding from the Government, and US$ 15.3 million from project entities. The reasons for the difference between the estimated and actual project investments include: (i) budget allocation through 2015 -2017 was less than expected; (ii) BANSEFI did not fully execute its allocated budget in 2016 and 2017, (iii) implementation delays due to changes in BANSEFI’s authorities; and (iv) the devaluation of the Mexican peso (MX$) against the US dollar (US$)10. The estimated and actual resource allocation for each component is described in the table 1 below and in Annex 3. Table 1: Estimated and Actual Costs by Component (US$ million) Total IBRD Gov. of Mexico Project Entities Component Component Name Estimated Actual Estimated Actual Estimated Actual* Estimated Actual Consolidation and strengthening Component 1 25.75 18.4 16.0 8.8 4.8 7.1 5.0 2.5 of savings and credit institutions Broadening access to financial Component 2 140.1 114.3 56.8 39.7 73.3 56.1 10.0 18.5 services and products Component 3 Strengthening BANSEFI capacity 26.07 34.9 16.0 18.7 9.8 16.2 0.3 0.0 Component 4 Project management 17.5 5.1 11.0 2.1 6.5 3.0 0.0 0.0 Front-end fee 0.25 0.25 0.25 0.25 0.0 0.0 0.0 0.0 Total 209.6 172.9 100 69.6 94.3 82.3 15.3 21.0 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) 14. The project was restructured twice to extend the closing date. The restructurings also fine-tuned and adjusted the results framework (see Annex 1). On July 31, 2014, the Project extended 17 months the Closing Date, from July 31, 2015 to December 31, 2016, and made necessary adjustments to the results framework to better reflect the activities undertaken under the Project towards the achievement of the PDO. The results framework was updated to (i) clarify or improve some definitions, (ii) take into consideration changes to the regulatory and supervisory framework of the sector, (iii) correct one of the baselines and (iv) make other minor changes. The second restructuring approved on December 13, 2016, had the main objective of extending the Closing Date from December 31, 2016 to July 21, 2017, bringing the total extension of the Project to below two years at the request of the Ministry of Finance (SHCP). It also adjusted the final targets of some indicators of the Result Framework in light of what had been accomplished so far and what was expected over the additional seven months of implementation, especially taking into account that total 10 Data are compiled in local currency and converted to U.S. dollars using the annual average exchange rate (see Annex 3). Page 8 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) disbursement was going to be significantly below the levels originally envisaged (additional discussion about these issues is included in the monitoring and evaluation (M&E) section of this report). Other Changes 15. The Project cancelled about US$ 31 million on November 21, 2017. The total IBRD disbursement was US$ 69.6 million, 69.6 percent of the total amount allocated for the project loan. Rationale for Changes and Their Implication on the Original Theory of Change 16. The changes abovementioned did not affect the PDOs or the reforms to be implemented under the components. Thus, the underlying theory of change remained valid. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 17. The objective of strengthening the SCIs in Mexico and of fostering financial inclusion remains highly relevant. The CPS 2014-201911 for Mexico considers the issue of increasing access to finance and improving financial inclusion a key aspect under Theme 1 “Unleashing Productivity” section on “Fostering sound financial sector development”. Additionally, under the proposed gender specific targets of the CPS, “Outcome 1: Increased access to finance and improved financial inclusion” specifically addressed the number of new clients mainstreamed into the formal financial sector using financial services, including the gender target of 1,600,000 new clients of which 800,000 should be –at least- female. 18. The relevance of the PDOs is rated as high as the Project provided clear evidence of the alignment of the PDOs to the current CPS objectives. The project PDOs and components were fully aligned with the government’s reform strategy and ongoing WB support. The two restructurings above mentioned did not include any changes in the PDOs. B. ACHIEVEMENT OF PDOs (EFFICACY) Outcome: “Support the consolidation of the savings and credit institutions (SCIs)” 19. The project reached its development objective of supporting the consolidation of the SCIs. The vast majority of the sector’s assets and clients is now formally managed by regulated and supervised SCEs. 91 percent of total assets and 87 percent of total clients in this sector are managed by 202 authorized SCEs, compared to around 73 percent of assets and clients managed by 104 authorized entities in October 2011 (in the period 98 additional SCEs were certified). As a result, 9.5 million members of SCEs are now protected by a deposit insurance (compared to 6.1 million in December 2011). 20. The TA and training sessions provided under the Project generated a significant and positive impact. The TA and training sessions provided to SCEs is associated to an increase of assets in about 33 percent, an increase of the number of members in about 28 percent, and to an increase of deposits in about 31 percent12. During the life of the 11 Report No. 80800-MX 12 Component 1 formal evaluation. September, 2016. https://www.gob.mx/cms/uploads/attachment/file/187680/Reporte_final_eval_impacto_ATyC.pdf Page 9 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Project, over 1,500 specialized TAs and training sessions13 were provided to more than 300 SCEs, accounting for about MX$132 million, to help them comply with the applicable regulatory framework, improve their financial and operational indicators, and to strength the capacity of their managerial and operative staff. 96 of the 98 SCEs that were authorized during project implementation benefited from more than 600 specialized TAs provided by the Project to help them strengthen their operations, compliance and obtain the certification. During the project, BANSEFI also contracted specific training activities in coordination with financial authorities (i.e. CNBV) and the confederation, that attended needs shared by the certified SCEs. More than 60 percent of the support was oriented to improve risk management, credit process, corporate governance and internal control; the rest of the support was related with strategic planning (14 percent), cost of certification (8 percent), compliance of Anti-Money Laundering (AML) regulation (7 percent), cost of auxiliary supervision (5 percent), and others (6 percent) (see Annex 7.3). 21. The solid progress towards the consolidation and strengthening of the sector is also confirmed by the improvement of the certified SCEs’ financial performance indicators. During project implementation, the average non-performing loan (NPL) ratio decreased from 8.8 percent in October 2011 (baseline) to 5.7 percent in July 2017; financial cooperatives´ Return over Equity (ROE) and coverage ratios increased from 9.5 percent and 106.7 respectively in December 2011 to 14.6 percent and 126.1 in June 2017. Despite the growth in assets (+118 percent since 2011 for the newly authorized entities), the SCEs could maintain a higher capital asset ratio (CAR) than required by CNBV, with 174 entities having a CAR above 150 percent. This shows that they are sufficiently capitalized to mitigate potential risks arising from this rapid growth. A formal evaluation contracted under the project shows that SCEs beneficiaries perceived the TAs and training as useful for improving their performance, and that the support provided had value added as the entities did not have sufficient resources available to fully pay for the TAs themselves. 14 22. The Project also supported an organized exit process of the unviable SCEs. During project implementation 27 SCEs were merged, liquidated or their assets and liabilities transferred, 15 of them directly supported by the Project (representing 85,282 members and MX$ 1,263 million of deposits). The number of SCEs merged or liquidated increased from 15 in October 2011 to 42 in July 2017. During the same period, 56 due-diligence “consolidation” assessments (for the same number of SCEs), were prepared, with most of the recommendations in favor of transferring assets and liabilities to certified SCEs (see Table 2). The low implementation rate of the due-diligences (27 percent of the total due-diligences done), with some cases taking years after a clear diagnosis has been made, was consequence of the difficulties in reaching a sound SCE interested in acquiring assets and liabilities from an unviable one, as well as the extended times demanded by the legal processes for the dissolution and liquidation of SCEs and the requirement for the participation of the State Government for the implementation of the process of liquidation and consolidation15. Table 2: Due-diligence Consolidation Reports Consolidations Reports SCEs Members Deposits Total 56 466,771 3,498 Fully implemented 15 85,282 1,263 Recent report / in process of implementation 14 153,206 533 Not implemented 27 228,283 1,702 Recommendations for transferring assets and liabilities 44 354,920 3,284 Source: BANSEFI 13 Training sessions were directed to groups of SCEs. Specialized TAs were directed to specific needs of individual SCEs. 14 The evaluation also shows that the TA provided did not generate a significant improvement in the case of financial indicators. However, the evaluation also mentions that those results are not conclusive due to the relative small number of observations. 15 FIPAGO law establishes a monetary contribution that the government of the State where the SOCAP is to be consolidated must carry out. Given that, in some cases, the State in question did not make its contribution, the consolidation process was not perfected (there are 7 cases of due diligence that were not impeded for this reason). Page 10 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Outcome: “Support the deepening of financial inclusion in selected areas and selected locations” 23. The project reached its development objective of supporting the deepening of financial inclusion in selected areas and selected locations. 3.4 million new clients were mainstreamed into formal financial sector through authorized SCEs (56 percent increase since 2011); more than 1.0 million new members were financially included in marginal/rural areas (of which 58.8 percent are women); BANSEFI expanded 3 times the number of service points providing financial services, from 1,827 in October 2011 to 5,649 in July 2017; about 1.8 million people received financial education; and people in rural areas increased the number of transactions (deposits, withdrawals, balance checks) performed in the formal financial sector. 24. As of July 2017, the 9.5 million clients operating with certified SCEs (which have been authorized, regulated and supervised by CNBV, as well as covered by the deposit insurance) represented 17.6 percent of the economically active population (this ratio might be higher if considered at the rural area, where the SCEs have relatively higher presence)16. The figure surpassed the end target of 8.0 million and compares well with the baseline of 6.1 million (October 2011). The actions taken during the Project that positively affected this performance were: • Increase in the number of new authorized SCEs and the initiatives to strengthen the sector (improving the public confidence on it), as described in the previous outcome. • PATMIR program, which provided support for expanding services to unbanked areas of the country and through this incorporate new members.17 • Financial education and outreach programs directed to SCEs and their members. 25. The authorized SCEs had a significant increase–from end 2011 until mid-2017- in the number of branches and the financial intermediation, measured through the performance of deposits, credits and assets (see Table 3). This performance can be attributed –even partially- to the Project, since of the 202 authorized entities as of July 2017, 173 received some type of TA or training under the Technical Assistance and Training Program (“Programa de Asistencia Técnica y Capacitación” PATyC). Table 3: Branches and Financial Intermediation Cumulative Growth Total SCEs Dec-11 Jun-17* Nominal Percentage Branches 1,925 3,040 1,115 58% Deposits 58,525 121,322 62,797 107% Credits 55,924 99,004 43,080 77% Assets 74,631 156,225 81,594 109% Source: CNBV * Information of entities LACP (SOFIPO) are as of March, 2017 26. During project implementation, PATMIR, a program that specifically focused on expanding the financial inclusion in rural areas, proved to be very successful, with 1,027,449 new members that complied with the eligibility 16The information of the economically active population comes from INEGI (Instituto Nacional de Estrategia y Geografía). 17 The PATMIR program was launched by the Ministry of Agriculture, Livestock, Rural Development, Fisheries and Food (SAGARPA) in 2001, with the aim of promoting access to comprehensive financial services, adapted to the rural population in conditions of marginalization. The Program was transferred to BANSEFI on November 1, 2010. Thus far, PATMIR has had three phases: PATMIR I (from 2001 to 2005), PATMIR II (from 2008 to 2010) and PATMIR III (from 2011 to 2016). As a result of the PATMIR I and II during a seven-year period, the incorporation of more than 697,000 people to the use of formal financial services was achieved, in localities of less than 15,000 inhabitants and municipalities classified as high, very high and medium marginality. Phase III of PATMIR was supported by this project, and achieved impressive results as explained in paragraph 26. Page 11 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) criteria (location, average balance, among others). From 2012 to 2016, 213 entities -78 percent financial cooperatives (SOCAPs), 17 percent popular financial societies (SOFIPOs), and 5 percent community financial societies (SOFINCOs)- participated in this initiative18 through their 2,358 branches in 2,341 municipalities within 32 states. As a result, more than 1 million new members were financially included through the third phase of PATMIR from which 59 percent are women (exceeding the PDO target of 50 percent), 29 percent come from indigenous regions, 57 percent are located in high and very marginalized areas, and 50 percent live in rural towns with less than 2,500 inhabitants.19 From 2012 to 2016, with an investment of MX$ 1,216 million, PATMIR III generated MX$ 6,141 million in savings, MX$ 6,054 million in credits, and MX$ 1,978 million in insurance. 27. An impressive result of PATMIR III is the fact that people that previously were not banked have continued to put money aside for emergencies or other future needs in saving accounts for at least 6 months, which has helped them get into a good habit of savings (in fact, the savings accounts under the program were much higher than the credits). The average balance of the saving accounts was around MX$ 5,800, totalizing MX$ 6,141 million of average savings, while other financial products such as credits (MX$ 6,054 million) and insurance (MX$ 1,979 million) were also demanded. An impact evaluation financed under the Project20 confirmed the positive impact of PATMIR III on the people financially included through this program compared with the people that were not part of the program. The people under PATMIR III: (a) tend to acquire more loans to the SCEs member of PATMIR; (b) present a higher percentage of their loans directed for economic development; (c) have better financial behavior (e.g. preparing a budget); and (d) present a greater percentage that have acquired an insurance product. The qualitative section of the evaluation report reflected the positive impact of PATMIR on the capacity of the SCEs to expand their geographical coverage and their membership, as well as to improve their sustainability and performance. 28. As a result of Project support, the number of service points providing financial services for BANSEFI’s clients increased 3 times, from 1,827 in October 2011 to 5,649 in July 2017, from which 2,401 were new banking agents (corresponsales). Despite a slow start to this activity, a partnership with the private network of corresponsales, Yastas, was the main factor to increase the number of banking agents from 95 in 2011 to 2,496 in July 2017, covering 951 municipalities, 69 percent of them located in areas with no presence of BANSEFI branches and 58 percent with no presence of “L@ Red de la Gente” partners. Moreover, during the second half of 2016, 25,984 transactions were completed through these banking agents, supporting a growth of transactional volume 3.5 times the same period of 2015. Particularly encouraging is the initial distribution of transactions since July 2016, with banking agents serving not just as a deposit point (44 percent of transactions), but also for other financial transactions that indicate financial inclusion, such as balance consultations, withdrawals and payments (30 percent, 21 percent and 5 percent of transactions respectively) (see Annex 7.4). By the end of the Project, about 4.5 million people have available new points of access to make deposits, withdraw their money, check their balances and make payments, 51 percent from municipalities with less than 100 thousand inhabitants and 2 percent from municipalities that did not have any access point to financial services before. The number of transactions performed by this segment of the population through 18 In December 2016 PATMIR was instrumented through a total of 159 SCEs. To have a sound expansion of the membership of the SCEs and to provide deposit protection to the financially included people it was required for the authorized ones to comply with the minimum CAR and for those non-authorized SCEs it was required to be rated A or B, and to implement an improvement plan defined by the technical agent (with the aim to be authorized in the near future). As of July 2017, 4 (non-authorized) SCEs participating in PATMIR (with 23,142 members and MXN 227 million of deposits) were under a consolidation process. 19 82 percent live in towns with less than 15,000 inhabitants and 100 percent in areas with less than 100,000 inhabitants. 20 Berumen and BANSEFI (September 2016). “Informe de Resultados de la Evaluación de Impacto de PATMIR en sus Vertientes Cualitativa y Cuantitativa”. Page 12 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) banking agents experienced an exponential growth, from 2,500 transactions per month in 2016 to an average of 20,000 transactions per month in the first semester of 2017.21 29. Regarding mobile banking activities, the indicator (to reach 250,000 customers) was dropped in the first restructuring of 2014, as BANSEFI prioritized the certification of SCEs and TAs and the implementation of PATMIR in the middle of tight fiscal constraints. A pilot test was held in 2015, concluding that the lack of penetration of smart phones among the targeted population and, at that time, the limitations from the regulatory framework to use SMS, suggested not advancing on it. 30. Complementing the financial inclusion initiatives directed to expand the membership or the SCEs and the number of access points, BANSEFI has implemented a series of initiatives on financial education that helped achieved the outcomes of the Project. Financial education activities were directed to: (a) SCEs and its members; (b) general public; and (c) beneficiaries of the governmental Social Inclusion Program (PROSPERA), the latter mostly of them women. These activities directly benefited around 1.8 million people, of which: 139,000 where channeled through training representatives of the SCEs that afterwards replicated to their members; 114,000 were directed to students of 332 schools in 10 states which generated a short-term positive impact on the financial culture and entrepreneurship of elementary and secondary school children22; and 1.6 million were directed to the beneficiaries of PROSPERA through the Financial Inclusion Program (PROIIF)23. Some of the financial education programs included short-term impact and feedback studies, which afterwards trigger their redesign (e.g. PROSPERA, school financial education). Other financial education initiatives of BANSEFI were the promotion of digital financial education (e.g. Massive Open Online Course) and the graduation of BANSEFI as a Certification and Evaluation Entity in financial education; both initiatives were in line with the objective to include more cost-effective financial education programs (see Annex 7.5). 31. The Project also supported the debit cards for PROSPERA beneficiaries. About 1.6 million (out of 7.0 million) debit cards with chip were issued. This new debit cards enabled the beneficiaries to have direct access to basic financial services such as credit, savings and microinsurance, as well as other benefits such as funeral insurance. A 2017 study from Bachas et al.24 confirms that debit cards provided not only easier access to savings, but also a mechanism to monitor bank account balances and thereby build trust in a financial institution. The study also reveals that prior to receiving the debit card beneficiaries of the program did not save in these accounts. Beneficiaries begin to increase their savings when trust is established after 1 to 2 years the debit card causes savings rate to increase by 3 to 5 percent of income. Financial education programs about the use of the new debit card with chip were also implemented (as described in Annex 7.5). 32. Finally, Project activities contributed to technological support and workshop and dissemination activities for L@ Red de la Gente, an alliance between BANSEFI and SCEs. The activities from Component 3 (“Strengthening BANSEFI capacity) indirectly supported the achievement of this outcome, through the strengthening of the BANSEFI 21 In addition, two months after the closure of the project (October 2017), BANSEFI signed another memorandum of understanding with one of the largest networks of rural corresponding agencies (Telecomm). 22 Impact Evaluation for Financial Literacy for Kinds and Youth population – December, 2016 23 A program that started operations in 2015 to provide customized financial services and financial education to PROSPERA beneficiaries. An Impact Evaluation of the Financial Inclusion Program (PROIIF) - December, 2016, determined that as a result of financial education the majority of PROSPERA beneficiaries understand the costs, term, and conditions of the savings and credits products offered under PROIIF. 24 Pierre Bachas, Paul Gertler, Sean Higgins, and Enrique Seira. Banking on Trust: How Debit Cards Enable the Poor to Save More. NBER Working Paper No. 23252, March 2017. http://www.nber.org/papers/w23252. The study used administrative data on over 340,000 bank accounts over four years. Page 13 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Technological Platform –BTP-, which helped BANSEFI in developing and performing its activity as a bank, but had limited impact in terms of the number of SCEs that currently are using the BPT. Although the BTP was used -at a given point- by more than 106 SCEs, while a total of 75 entities have received technical support on it, it quickly became obsolete as of July 2017, only 7 SCEs were using the core banking and 7 the AML module. This limited usage was consequence of some weaknesses of the BTP related with the capacity to: incorporate the changes of the regulatory framework, comply with the requirements from the Tax Agency and capacity for the integration with the accounting system of the SCEs. 33. Overall, the efficacy of the project is rated as substantial. The Project reached its development objectives (intended outcomes) by: (i) supporting the savings and credit entities, operating mostly in rural/marginal areas, to improve their capacity for running their business under a formally regulated and supervised environment; (ii) helping increase the share of total assets managed by authorized SCEs from 72 percent in 2011 to 91 percent in 2017; (iii) helping increase the share of total clients managed by authorized SCEs from 74 percent in 2011 to 87 percent in 2017, 3.4 million clients mainstreamed into the formal financial sector; (iv) supporting the inclusion into the financial sector of about 600,000 women and 400,000 men in rural/marginal areas; (v) creating the conditions to generate an exponential increase on the usage of basic financial services in the rural population, 3,822 new service points providing financial services were created; and, (vi) providing financial education to about 1.8 million people. C. EFFICIENCY Assessment of Efficiency and Rating 34. The economic analysis of technical assistance interventions is particularly difficult to quantify due to problems of attribution, multiple-generation effects, and the lack of financial flows generated from project investments, among others. Although the Project Appraisal Document (PAD) for the project included a calculation for an Economic Rate of Return (ERR) Analysis, the methodology used for its calculation is not clear enough to allow its replication with the final results of the Project.25 35. As part of this ICR, an effort was made to provide an approximate notion of Project efficiency by conducting a qualitative assessment and contrasting Project achievements with associated costs. For instance, the support provided under component 1 to SCEs is associated to an increase of their assets in about 33 percent, increase of the number of members in about 28 percent, and to an increase of deposits in about 31 percent. The combined effects of TA, financial education programs and capacity building activities contributed to enhancing the institutional and financial performance of SCEs and BANSEFI. 36. On the other side, relatively high investment in technology and IT support with very limited usage by the SCEs of the IT facilities provided by BANSEFI hinder the Project’s efficiency. The technological platform supported the improvement of BANSEFI’s operations, particularly in branches located in rural areas; however, it was less successful in supporting SCEs. The IT platform experienced cost-overruns due to challenges in implementation and changes in the business strategy. The actual amount invested under Component 3 (US$ 34.9 million), about 20 percent of project costs, was US$ 8,8 million higher than the amount originally estimated, yet the technological platform had a limited use by the SCEs. The limited use was mainly caused by issues of design and implementation of the core banking tool, which presented deficiencies in its functionality, in order to have a solid Platform that allowed SCEs to use it efficiently.26 25 Annex 4 “Efficiency Analysis” provides some examples of positive and negative impacts to efficiency. 26 Certain updates were applied according to the regulations; however, not all could be concluded due to the requirements of the different user Page 14 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) 37. From a macroeconomic perspective, the benefits of a well-regulated and supervised SCEs sector and the increased access to finance for the poor are highly relevant. From that perspective, the improvements in the regulatory and supervisory framework and the training provided to SCEs to comply with it has important and positive social implications, mainly by preventing potential financial problems and preventing potential impact in the low- income population that uses these services. As described in the Efficacy section, the Project supported growth of the SCE sector and its outreach to under-banked populations in rural areas, which contributes to access to finance for the poor. Increased access to reliable financial services for low-income population enables them to benefit from adequate and secure savings. There is also positive efficiency impact to highlight from the implementation of PATMIR III, in terms of number and type of people financially included as well as on the results from the demand of financial services (for each MX$ 1 spent, there were MX$ 5 of savings, another MX$ 5 of credits and MX$ 2 of amount insured). 38. Overall, the precedent of two previous projects (BANSEFI I and II), with some similarities to this project in their design, provided BANSEFI with the experience to operationalize the Project in an efficient way. The project implementation benefited from very good teams in the executing agency and the Bank that were adequately staffed by experienced personnel. 39. Considering the factors described above, the Efficiency rating is rated as Substantial. D. JUSTIFICATION OF OVERALL OUTCOME RATING 40. Considering the rates assigned to Relevance of PDO (High); Efficacy (Substantial) and Efficiency (Substantial), the Overall Outcome Rating is rated as Satisfactory. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 41. The project had a positive impact on women’s access to financial services. While the data on the share of women in the overall membership of SCEs is not available to assess evolution over time, PATMIR and PROIIF had a strong focus on women. Under PATMIR, more than 604,000 women were linked to financial services (58 percent of total new members), while under PROIIF (through the first phase -2015- of financial inclusion of PROSPERA beneficiaries) around 500,000 of women (almost 50 percent of total) received a basic loan and 670,000 (2/3 of total) have contracted a “programmed” deposit. Finally, 90 percent of the 1.8 million people receiving financial education from BANSEFI were women. 42. PATMIR and PROIIF have contributed to the relatively higher reduction in the gender gap 27 observed in the rural sector. The last National Survey of Financial Inclusion (ENIF, 2015) –compared with ENIF 2012- has shown a decrease in the gender gap of 8 percentage points (from 12 percent to 4 percent), while for the rural area, the gap has not only decreased (in 8 percentage points) but also reversed (higher percentage of women -38 percent- than men -32 percent- with a saving account) (see Table 4). entities benefiting from the technological platform. Nevertheless, 58 SCEs were benefited with the implementation of the AML application, which allowed them to comply with the normative prudential provisions issued by the CNBV. 27 The gender gap is defined as the difference between the percentage for male and female with a saving account. Page 15 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Table 4: Percentage of Adult Population with a Formal Savings Account Year Rural Area Urban + Rural Men Women Gap Total Men Women Gap Total 2012 26 19 7 22 42 30 12 36 2015 32 38 -6 35 46 42 4 44 Difference 6 19 -13 13 4 12 -8 8 Source: National Survey of Financial Inclusion (ENIF) 2012 and 2015 Institutional Strengthening 43. The project contributed to the institutional strengthening of BANSEFI. Throughout the years, BANSEFI has benefited from the capacity building of staff on the best practices for supporting the SCE sector, and on the implementation of financial inclusion and financial education initiatives. The project directly contributed to establish an institutional mechanism for ongoing capacity building (including quality control), particularly for the SCE sector. Moreover, component 3 of the project focused specifically on improving BANSEFI’s technological capacity, providing training to BANSEFI staff on the information technology platform and supporting the dissemination of the functional capabilities of BANSEFI’s. Overall, these activities somewhat contributed to strengthening the operating environment of the institution. Mobilizing Private Sector Funding – N/A Poverty Reduction and Shared Prosperity 44. The Project has positive impact on poverty reduction and shared prosperity through the increased access to financial services in selected areas and locations, as well as the provision of financial education that enable the lower income population to improve their financial administration and increase their productivity, as well as to take responsible decision on financial matters. Some examples are: • Access to financial services to people in selected areas and locations, predominantly rural and marginalized: low income people were able to save in certified SCEs (covered by deposit insurance) with potential access to other financial services such as loan, insurance, and remittances. The PATMIR programmed was strongly focused on reaching the indigenous regions (29 percent of new saving accounts were located in an indigenous region) as well as in towns less populated or with very high, high or medium marginality. • Increased access points for financial services: reduction in the cost (transportation, security) of making financial transactions through correspondent banks located in areas where there are no BANSEFI branches. • Financial education: 80 percent of the total people that received financial education from BANSEFI were beneficiaries of social programs (PROSPERA). Other Unintended Outcomes and Impacts 45. As a free perk, debit cards distributed for the PROSPERA social program included accidental death insurance funeral coverage and other benefits. These benefits were not intentional when considering the selection of the debit card issuer, but are an important contribution for the millions of beneficiaries in the lower income bracket who receive them. Page 16 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 46. The key factors during preparation influencing the operation’s achievements include: 47. Project design: Overall, the Project was well designed. Project components were fully aligned with the government’s reform strategy and ongoing WB support. The Project benefited from a strong partner in BANSEFI, which had the experience from the two prior WB financed projects to strengthen the sector and increase outreach. 48. Results framework: In general terms, the PDO indicators were realistic and aligned to the PDO, considering the uncertainties at the moment of preparation. As reported in the mid-term review, the original results framework presented some weaknesses in its baseline calculations, its definitions and its relevance to some of the activities of the project, making it difficult to ensure a consistent monitoring of these indicators. 49. Assessment of risks affecting the implementation of the Project: certain risks were not considered during the preparation, although they eventually materialized during project implementation and contributed to delays in implementation of reforms. These included (a) reduction of funds from the Federal Government, limiting the disbursement of the loan; (b) postponement of the entry in to force of the new legal framework for the sector; (c) frequent changes in the senior management of BANSEFI; (d) difficulties and challenges in the implementation of the resolution (liquidation, merge, or transfer of assets and liabilities) of SCEs; and (e) changes in BANSEFI’s strategy. B. KEY FACTORS DURING IMPLEMENTATION 50. Despite some challenges, most components produced a rich output and numerous achievements, as discussed in Section II.B. Successful implementation can be attributed to the commitment of authorities to execute an ambitious Project agenda with numerous complex tasks, and the close Bank supervision during implementation. However, the overall Implementation Progress (IP) was downgraded from Satisfactory to Moderately Satisfactory (MS) in December 2013 in reflection of the slow disbursements, keeping that rating until the closing of the Project. Factors Subject to Client Control Coordination and Engagement 51. There was a reasonable coordination among the stakeholders (e.g. financial sector authorities, BANSEFI, SCEs, Federations, Confederation, Technical Agents, FIPAGO, and consulting firms, among others) that facilitated the implementation. Having the same staff on the BANSEFI side coordinating TAs under Component 1 and Component 2 allowed for appropriate oversight, to ensure that there was no overlap on the TA provided but also to allow for some flexibility for implementation. There were good coordination efforts with CNBV (especially during the latest stages of the Project) via frequent meetings to discuss the sector support programs, regulatory priorities, follow-up on issues of orderly liquidations, etc. Commitment and Leadership 52. Budget cuts from the Federal Government limited the implementation speed: Some components and subcomponents were affected in their implementation by the reductions in the funds availability from the Page 17 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Government for implementing TA as well as financial inclusion policies. These cuts were consequence of external shock (e.g. fall of oil price) with impact on the incomes to the Government and the need of budget cuts. As an example, the 2013 fiscal cuts had a considerable impact on the Project, since a large number of institutions seeking help with the certification process or general strengthening had to be turned away (277 request for technical assistance by 138 entities of the sector were denied) at a time when this assistance was needed the most. Several strategic studies were put on hold, while the IT-related activities (which had mostly been contracted) were relatively untouched. This factor was one of the main drivers for the two restructurings of the Project, extending their deadline and allowing the achievement of the outcomes. 53. Commitment from staff and new authorities of BANSEFI: a multidisciplinary team from BANSEFI was in charge of the design, development, implementation, execution and monitoring of the work program. The implementation teams were qualified and strongly committed with the implementation and monitoring of the Project. However, during the life of the Project, there were five changes in the BANSEFI´s high level authorities involved in implementation, with some impact on the speed of implementation (which were reflected in delays in the implementation until the new authorities became familiar with the project and authorized its continuity) and on prioritization of the activities per component. 54. BANSEFI new strategy: The development of the new strategy of BANSEFI in 2013/2014 moved the institution into a broader scope, deflecting its focus on being a second-tier entity for the SCEs concentrated in the savings component. This also impacted the reassignment and prioritization of certain components and sub-components throughout the Project cycle. An appropriate assessment of the risks and potential mitigation measures for changes in BANSEFI´s priorities as well as for the changes on its senior management could potentially have avoided some implementation delays. Factors Subject to WB Control Adequacy of Supervision and Reporting 55. World Bank Supervision: A multidisciplinary team composed of F&M and Agriculture Global Practices provided implementation support, including close and frequent interaction with BANSEFI´s relevant Departments and, during field visits, with other key stakeholders. There was a low turnover rate of WB staff supervising the project during the implementation stage. Reporting was appropriate, but there were some missed opportunities to assess the challenges of implementation of some components of the project (i.e. Component 3, IT platform) that could have been anticipated by the mid-term review of the project or during supervision missions. Factors Outside of Government and/or Implementing Entities 56. Legal processes and the requirement for the participation of the State Government affecting the enforcement and implementation of the resolution of SCEs: These are external factors, outside of the project scope. The legal framework establishes several steps until a SCE is effectively resolved (liquidated, merged or assets and liabilities transferred) which cause significant delays in the effective implementation of the corresponding resolution process or the dissolution and liquidation of the SCEs. This situation is worsened by the fact that there is no relevant requirement for the creation of new “basic” financial cooperatives. Page 18 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 57. Overall the M&E framework was sufficient to assess the achievement of the objectives and test the links in the result chain, but it appears that there was scope for improvement in the design and rationale for some indicators. As clarifications on the definitions of some indicators was deemed necessary, the project rightly revised the results framework in the first restructuring of the Project in 2014 (as explained above). In the second restructuring of the Project, in 2016, the targets of the results framework were revised. This adjustment of indicators could be considered unnecessary since the final results reached by the end of the Project compared with their baselines are the ones used to evaluate the level of contribution to the achievement of the intended outcomes. M&E Implementation 58. The implementing agency and the Bank systematically reported on key indicators. BANSEFI monitored the progress of the Project and prepared semi-annual Project reports that were shared with the WB each semester. The reports included a summary of the activities performed during the period per each component of the Project, the updated table with the performance of the PDO indicators and Intermediate Results Indicators, the different type of tenders during the period, financial information of the loan and future actions. Also, in addition to the official results matrix, BANSEFI followed about 38 additional indicators to monitor progress. The collection of information for calculating the indicators was benefited by a sector database built up by BANSEFI, as well as by specific informative regime (e.g. the one applicable for new members under PATMIR III). A comprehensive M&E arrangement was in place for the training and TA and the outreach components, consisting of consultant and providers of non-consulting services presentations and reports, discussions by BANSEFI with entity managements, and group discussions with members of entities. Periodic financial and other performance indicators were also aggregated based on submissions of consultants and providers of non-consulting services to measure performance for the assisted entities as a group. There were also annual workshops with participating consultants and providers of non-consulting services discussing their approaches, results, success factors, constraints, and recommendations to improve the approach to providing assistance to the sector. It is likely that BANSEFI will keep collecting information and calculating the indicators used for the Project. M&E Utilization 59. The M&E utilization was used to inform project management and decision-making. The M&E data on performance and results progress periodically reported were used and incorporated in the Implementation Status and Results report (ISR). Also, the indicators fed into management’s strategic and implementation decisions at BANSEFI. Moreover, after each implementation support and supervision mission by the WB, an Aide-Memoire was prepared and shared with BANSEFI and SHCP. The indicators were monitored against their target levels as a way to measure the accomplishment of objectives, nonetheless the M&E continuously assessed the relevance of the indicators and the potential need to restructure them to better reflect the outcomes of the Project (as it was the case after each of the two restructurings). 60. In addition to monitoring progress on implementation and towards development objectives, BANSEFI would regularly conduct studies and evaluations to inform future activities. Some of the evaluations included the impact Page 19 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) evaluation of Component 1, the impact evaluation of PATMIR, the impact evaluation of the PROIIF, and the impact evaluation of the pilot of Financial Literacy for Kids and Youth population (see Annex 6). Justification of Overall Rating of Quality of M&E 61. The Overall Rating of the Quality of M&E is substantial. Despite minor shortcomings at design of the results framework, the M&E system’s implementation and utilization were adequate. The M&E system was sufficient to assess the achievement of the objectives and to test the links in the results chain. M&E findings were disseminated and used to inform the direction, evolution and perspectives of the project. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 62. Overall compliance with safeguard policies was satisfactory. The Project was provided a C classification from an Environmental Assessment standpoint (OP/BP 4.01), implying that it was likely to have no adverse environmental impacts. The Project triggered the Indigenous Peoples (OP/BP4.10) safeguard, and followed a framework approach.28 To this end, an improved social strategy was designed building on the experience of the previous Projects, and taking into account the evaluation conducted during preparation, which provided lessons learned and recommendations. The compliance of the social safeguards was rated as satisfactory during the life of the Project. 63. On Financial Management (FM), through the Project implementation financial management performance remained mostly satisfactory. Project’s Interim Unaudited Financial Reports (IFRs) were generally submitted in due time, with some minor delays, in acceptable terms; audited financial statements were also generally submitted in due time and included unmodified (clean) opinions.29 However, towards the end of the implementation stage, weaknesses were observed in the budget execution process. From 2014 through 2016, already lower than expected allocated budget was not executed as planned, additional budget reductions and under execution were observed during this period, and disbursement progress was consequently lower than anticipated, reaching only 69.6 percent of total loan proceeds at the closing of the Project. As a result, the final FM performance for the Project was deemed moderately satisfactory. 64. On Procurement, BANSEFI managed to consolidate a solid and well-trained procurement team within international area of BANSEFI which helped the procurement processes to be carried out in an orderly manner and in accordance with the Bank's Guidelines. No major procurement issues arose during Project implementation However, because of the lack of clarity in the budget assigned to the Project and the constant changes in the administration and it priorities, a large number of processes were frequently cancelled or postponed and the Procurement Plan was constantly modified losing its sense of planning instrument. C. BANK PERFORMANCE Quality at Entry 65. The Bank identified, facilitated the preparation of, and appraised the operation appropriately for the achievement of the PDOs, but envisaged relatively short implementation time below the average of similar projects. The explanation of the strategic relevance of the SCI sector and the importance to support its consolidation as well as on the importance to deep financial inclusion in selected areas and location was well addressed. The technical issues (components, activities, etc.) as well as the financial and economic aspects were well covered. Because of the type of 28 There were numerous materials on communication and education that were translated in indigenous languages to ensure inclusivity. 29 The last audit for the Project covering the period from January 1 to July 21, 2017 (closing period) is due by January 21, 2018. Page 20 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) project there were no environmental aspects included, while the poverty, gender and social development aspects were considered in the design of the PDO indicator as well as safeguard. However, the needed timeframe for the implementation of activities and achievement of significant outcomes was underestimated. This was originally a 3 years and 7 months Project, that was extended twice for a total of two additional years in response to delays of implementation and low disbursements, and to allow additional time to support the achievement of Project objectives. 66. The design of the M&E framework had some shortcomings as explained in paragraphs 48 and 57. Quality of Supervision 67. The Bank team actively supervised Project implementation and was responsive to client needs. During Project implementation, the WB was proactive in identifying the challenges and risks to the achievement of the PDO. The ISRs and the Aide-Memoires reflected well the discussions and the exchange of ideas between the WB team and BANSEFI staff. The missions that visited Mexico during the Project had meetings with the relevant Departments at BANSEFI and with the rest of the stakeholders (supervisor, beneficiaries, other public sector agencies, SCEs, etc.) to directly assess and identify the shortcomings and discuss the best ways to tackle them. The Aide-Memoires were well written, and considered clear and useful by BANSEFI´s staff. The project presented several situations where funds had to be reallocated (e.g. delay in the entry into force of the legal framework, budget cuts, new activities prioritized, etc.), which were discussed and agreed between BANSEFI and the Bank. 68. Nevertheless, there were unwarranted changes to some target indicators near the closing date of the project. Some target indicators were reduced and increased in the second restructuring of December 2016 when the project closing date was extended 7 months. As it became clear that resources would not be fully disbursed, two indicators targets were revised downward. For example, the PDO indicator related to the number of entities certified by CNBV was reduced from 224 to 202 entities after it became clear that a natural consolidation of the sector took place during the life of the project, but also because a number of non-certified entities had total assets too small to eventually trigger the certification requirement. In this case, it is important to highlight that at the time of the reduction of the target the project had already achieved 80 percent of the original objective. One indicator (the number of service points providing financial services) was revised upward to reflect the rapid progress with the roll out of the banking agents which reached over 2,100 units as of November 2016, from barely 255 units in the summer of 2016. This revision aimed to continue pushing the financial inclusion agenda for the country (see Annex 1). Justification of Overall Rating of Bank Performance 69. Based on the assessment of Quality at Entry and Quality of Supervision, the rating of the Bank Performance is Moderately Satisfactory. D. RISK TO DEVELOPMENT OUTCOME 70. BANSEFI is an important player in the field of access to finance and is a key source of support to the sector. Stakeholders (both from the private and the public sector) are clear in terms of the need to continue working on the activities and initiatives for keeping and strengthening the achievement of the PDOs. The SCE sector will face increased prudential regulatory requirements, as well as increasing competition that would require further improvements in their governance, policies and procedures, strategic planning, risk management as well as investment in technology. With respect to the effective implementation of the resolution (liquidation, merge, transfer of assets and liabilities) of “failed” SCEs, the Government is considering legal amendments to clarify and increase the enforcement capacity Page 21 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) of the process; the draft would also include minimum entry standards for the creation of new “basic” financial cooperatives. 71. While the large entities are now subject to supervision and most of the assets and members are covered, a number of entities are still pending liquidation / merger or are classified as "basic" (societies with assets below UDIS 2.5 million - USD 750,000). There are currently 76 SCEs (33 classified in D and 43 registered for consolidation process) pending to be liquidated, merged, or its assets and liabilities transferred, as they have not yet received authorization from CNBV or are basic ones that have been classified as D (the worst rating) during the review process. The basic ones are only subject to limited reporting requirements as well as simplified capital requirements reported to the Auxiliary Supervisory Committee, who rate and report them to CNBV. Furthermore, the number of non-certified SCEs increased from an estimated of around 400 in early 2012 to 658 in July 2017, although their importance in terms of assets of the sector has diminished. The growth in basics SCEs reflects on the one hand deficiencies in minimum requirements for the creation of new financial cooperative (e.g. the ex-ante verification of their viability) and on the other hand deficiencies in the process for liquidating unviable entities, demanding lot of time and supervisory efforts, with a potential negative impact in the reputation of the sector. The remaining consolidation agenda is not expected to derail the reform process, but should be addressed to prevent potential reputational risks in the medium term. The TAs should continue, focused on the specific areas that SCEs need support (for which a strong coordination with CNBV would be needed), for continue strengthening their performance. 72. Regarding financial inclusion, the challenges could come not only in expanding the membership of SCEs but also how to consolidate the new members and ensure they have access to adequate financial services in an appropriate way. Under PATMIR, BANSEFI expects to keep the strategic alliances with the technical agents and trust on the capacity building of people trained by the technical agents within each SCEs. PATMIR will also start splitting its priorities to attend migrant populations. In terms of financial education, BANSEFI has been active in the recently approved Comprehensive Financial Education Strategy, and expects to keep providing financial education to different segments of the population (e.g. personnel for SCEs, teachers, students, parents, migrants and other vulnerable people). 73. The main risks to maintain and reinforce the outcomes achieved may derive from: Risk Likelihood Impact Obtain support from the Federal Government High High Reducing, in relative terms, the direct support to the SCI sector in favor of other initiatives Medium Medium more focused on BANSEFI´s direct role on financial inclusion / education New Gov’t administration delaying and/or changing the priorities Medium Low Changes in the implementation procedures that deteriorates the quality of the activities Low Medium (since there is no obligation to keep following the same standards due to the end of the engagement with the Bank) related with both outcomes V. LESSONS AND RECOMMENDATIONS 74. A more strategic view on the role of BANSEFI could have been considered during project design. At the time of preparation, the project took into consideration CPS priorities and it likely reflected client demand. The Bank provided substantial technical support for project preparation in the context of other ongoing operations and strong Page 22 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) analytical work. However, in retrospective, issues regarding strategic guidance to help BANSEFI evolve its role overtime as it faced competitive pressures should have been considered during the design of the operation. 75. As indicated in the lessons learned of the first and second BANSEFI projects, reforms that involve hundreds of heterogeneous institutions with bottom-up governance require long-term support and commitment. International experience and prior projects show that the need to training of SCE officials is permanent as these small institutions rotate key positions through an election process. Thus, it is necessary to plan a medium-to long term initiative to ensure sustainability. Such plan would need to determine the content of capacity building support necessary post-certification of SCEs, its financing, and the roles of the various institutions involved in the sector, including BANSEFI, the co-federation and federations to avoid overlap between institutions. A self-financed system for continuing education of the sector would be desirable. 76. Despite the relevance of the IT component to support Project objectives, efficiency assessments and other feedback mechanisms should be reinforced during project implementation to ensure that expenditures in IT components are being directed to have the desired outcomes. Projects that include the support of a technological component should be designed and implemented considering the changing environment as well as be continuously monitor for progress. The changing regulatory requirements and the quick evolution of technology requires an adequate design for the investment in technology (which, generally, tend to be significant in terms of amount), assuring it is flexible enough to adapt to changing contexts and well aligned with the requirements from the demand side. Most importantly, continuous monitoring during implementation is needed to ensure an efficient use of resources towards the achievement of the development outcomes. For example, in the case of this project, an IT audit would have been desirable during the mid-term review to assess efficiency and efficacy of the technological platform. 77. Undoubtedly, indicators provide useful insight into the achievement of PDOs but they do not provide a complete account of all outcomes. Projects should include a relevant and measurable set of performance indicators, consistent with the PDO and project expected outputs. Timely revisions of monitoring indicators to reflect the evolving realities on the ground should be considered, while unwarranted changes to target indicators is not advisable near Project closing. 78. Strong government and industry commitment to a project agenda are a precondition for success of Bank projects. Particularly, strong coordination among public and private sector stakeholders prove to be successful when designing solutions for financial sector development. For this project, the agreement signed with private network of banking agents, Yastas, was a key factor to significantly increase the number of correspondent banking agents in a short period of time. Recommendations 79. Given the critical role played by BANSEFI in expanding financial services and inclusion in Mexico, it would be important to conduct a thorough review on BANSEFI and redefine its role and strategy going forward to ensure sustainability of the institution and improve its impact. After such review, BANSEFI could reinforce its strategy for the sector through supporting the use of innovative channels for the provision of financial services. . Page 23 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Support the consolidation of the savings and credit institutions Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of entities certified Number 104.00 224.00 202.00 202.00 by CNBV, the regulator 31-Oct-2011 15-Jul-2014 12-Dec-2016 21-Jul-2017 Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "Additional Entities certified by CNBV, the regulatory agency (target 120) / resolved (target 15)" to make it clearer and to monitor the total number of certified entities (not only the additional ones). In December 2016, the final target was reduced from 224 to 202 in light of what was already accomplished and what was expected to be achieved over the additional seven months of implementation. 96 of 98 new certifed SCEs are attributable to the Project support. Compared with the original target of 120 additional certified SCEs (224 in total), 80 percent of the target was achieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of entities merged or Number 15.00 40.00 42.00 liquidated 31-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "Additional Entities certified by CNBV, the regulatory agency (target 120) / resolved (target 15)" to make it clearer and to proportionally increase the final target from 15 to 25 Page 24 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) additional entities merged or liquidated (due to the increase of Project support to those entities no able to obtain certification). During Project implementation 27 entities were merged, liquidated or their assets and liabilities transferred, 15 of them directly supported by the Project. Compared with the original target of 15 additional resolved SCEs, the project reached 100 percent of the target. Compared to the adjusted target of 25 additional resolved entities (40 in total), 60 percent of the target was achieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of reporting certified Text 79/92 193 170 174 institutions with a capital adequacy ratio 150% of 30-Oct-2011 15-Jul-2014 12-Dec-2016 21-Jul-2017 minimun requirement or above */ Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "New Entities with improved performance on financial and operational indicators" to make it specific and measurable. In December 2016, the final target was reduced from 193 to 170 in light of what was already accomplished and what was expected to be achieved over the additional seven months of implementation. Compared with the original target of 193, the project reached 90 percent of the target. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Average of non-performing Percentage 8.75 7.25 5.69 loans for certified institutions 30-Oct-2011 15-Jul-2014 30-Jun-2017 Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "New Entities with improved performance on financial and operational indicators" to make it specific and measurable. The target was achieved. Page 25 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Clients mainstreamed into Number 6.12 8.00 9.50 formal financial sector (number in millions) 30-Oct-2011 15-Jul-2014 30-Jun-2017 Comments (achievements against targets): This target was achieved and surpassed. With the Project support, 3.4 million new clients were mainstreamed into formal financial sector Objective/Outcome: Support the deepening of financial inclusion in selected areas and selected locations Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Clients mainstreamed into Number 6.12 8.00 9.50 formal financial sector (number in millions) 30-Oct-2011 15-Jul-2014 30-Jun-2017 Comments (achievements against targets): This target was achieved and surpassed. With the Project support, 3.4 million new clients were mainstreamed into formal financial sector Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of service points Number 1827.00 3750.00 5400.00 5631.00 providing financial services 30-Oct-2011 15-Jul-2014 12-Dec-2016 30-Jun-2017 Comments (achievements against targets): This indicator was included as a PDO indicator in the first restructuring to replaced two intermediate resutlts Page 26 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) indicators "2,000 Diconsa and 2,000 Pemex stores providing Government conditional cash transfer services". By the time the restructing happened, the target was already reached, with over 15,000 locations providing these services. The Project included this indicator to monitor the number of access poins that function as banking correspondent agent, which better reflected the Project's strategy. The indicator also removed the mention of Diconsa and Pemex as BANSEFI was planning to add other private and public partners. In December 2016, the final target was increased from 3,750 to 5,400 in light of what was already accomplished and what was expected to be achieved over the additional seven months of implementation. The target was achieved and surpassed compared with its orginal target and after it was increased. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of members Number 0.00 825000.00 977950.00 1027449.00 financially included through PATMIR 30-Oct-2011 15-Jul-2014 12-Dec-2016 30-Jun-2017 of which women Percentage 0.00 0.00 50.00 58.80 30-Oct-2011 15-Jul-2014 12-Dec-2016 30-Jun-2017 Comments (achievements against targets): This indicator was included in the first restructuring as a PDO Indicator. Upon the extension of the Project the target of new members financially included in rural and marginal areas was increased from the current 825,000 to 1 million. The target was achieved and surpassed in both the number of members and the percentage of women. A.2 Intermediate Results Indicators Component: Component 1: Consolidation and strengthening of savings and credit institutions Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Entities receiving training Number 0.00 180.00 170.00 168.00 Page 27 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) and technical assistance for 31-Oct-2011 15-Jul-2014 12-Dec-2016 21-Jul-2017 certification Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "Entities receiving training and technical assistance for certification / resolution" to differenciate between the ones receiving support for certification and for resolution.In December 2016, the final target was reduced from 180 to 170 in light of what was already accomplished and what was expected to be achieved over the additional seven months of implementation. Compared with the original target of 180, the project reached 93 percent of the target. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Entities receiving training Number 0.00 130.00 171.00 179.00 and technical assistance for operational improvement 31-Oct-2011 31-Oct-2011 12-Dec-2016 21-Jul-2017 Comments (achievements against targets): In July 2014, the target was increased from 130 to 150. In December 2016, the final target was increased from 150 to 171 in light of what was already accomplished and what was expected to be achieved over the additional. The target was achieved and surpassed compared with its orginal target and after it was increased. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of entities assessed Number 9.00 42.00 60.00 61.00 for mergers, assets/liability transfers or liquidation 31-Oct-2011 15-Jul-2014 12-Dec-2016 21-Jul-2017 Comments (achievements against targets): This indicator was included in the first restructuring to replaced the original indicator "Entities receiving training and technical assistance for certification / resolution" to differenciate between the ones receiving support for certification and for resolution. In December 2016, the final target was increased from 42 to 60 in light of what was already accomplished and what was expected to be achieved over the Page 28 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) additional seven months of implementation. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of institutions that Number 0.00 0.00 176.00 176.00 improved their performance towards certification (to be 31-Oct-2011 31-Oct-2011 15-Jul-2014 21-Jul-2017 monitored until certification deadline) Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of reporting Text 90/92 210 180 185 certified institutions with a capital adequacy ratio 100% 31-Oct-2011 15-Jul-2014 12-Dec-2016 21-Jul-2017 of minimum requirement or above Comments (achievements against targets): This indicator was included in the first restructuring to measure the financial performance of supported SCIs. In December 2016, the final target was reduced from 210 to 180 in light of what was already accomplished and what was expected to be achieved over the additional seven months of implementation. Compared with the original target of 210, the project reached 88 percent of the target. Component: Component 2: Broadening access to financial services and products. Indicator Name Unit of Measure Baseline Original Target Formally Revised Actual Achieved at Page 29 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Target Completion Mobile phone pilot Text Initial pilot design 0 Lessons learned and N.A strategy developed 31-Oct-2011 31-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): This indicator, which initially was hoping to reach 250,000 customers, was dropped in the first restructuring. As BANSEFI gave priority to the certification and technical assistance activities to strengthen the entities and the implementation of PATMIR in the middle of tight fiscal constraints, the roll out of the mobile banking pilot was postponed until more funds were made available. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of access points for Number 95.00 2300.00 2496.00 correspondent agents 30-Oct-2011 12-Dec-2016 30-Jun-2017 Comments (achievements against targets): This indicator was monitored by BANSEFI during the life of the Project and it was included as intermediate result indicator in the second restructuring to reflect the importance of corresponded banking agents in the PDO indicator that measures the number of service points providing financial services. The private network of banking agents, Yastas, was the main factor to increase the number of correspondent banking agents from 95 in 2011 to 2,496 in July 2017, covering 951 municipalities, 69 percent of them located in areas with no presence of BANSEFI branches and 58 percent with no presence of “L@ Red de la Gente” partners. Component: Component 3: Strengthening BANSEFI capacity Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion IT Platform Uptime Percentage 99.86 99.90 100.00 Page 30 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) 31-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): Indicator included in the first restructuring. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Processor Utilization Text 82% Less than 70% 50% 31-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): Indicator included in the first restructuring. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Bandwidth Utilization Text 9.6 Mbps 20 Mbps 13.5 Mbps 31-Oct-2011 31-Oct-2011 21-Jul-2017 Comments (achievements against targets): Indicator included in the first restructuring. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of entities receiving Number 0.00 25.00 40.00 75.00 technical support to use IT Platform 31-Oct-2011 31-Oct-2011 15-Jul-2014 21-Jul-2017 Page 31 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Comments (achievements against targets): This indicator increased its final target in July 2014 from 25 to 40 entities receiving technical support to use IT Platform. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Reduction in man-hours Hours 120.00 40.00 3.60 spent on resolving errors 30-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): Indicator included in the first restructuring. This target was achieved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Reduction in time required Text 7.8 Hours Less than 6 Hours 7.9 for end-of-day batch runs 30-Oct-2011 15-Jul-2014 21-Jul-2017 Comments (achievements against targets): Indicator included in the first restructuring. Page 32 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1 1. Number of entities certified by CNBV, the regulator 2. Number of entities merged or liquidated 3. Financial Performance expressed as: a) Number of reporting Outcome Indicators certified institutions with a capital adequacy ratio 150% of minimum requirement or above */ b) Average of non-performing loans for certified institutions 1. Entities receiving training and technical assistance for certification 2. Entities receiving training and technical assistance for operational Intermediate Results Indicators improvement 3. Number of entities assessed for mergers, assets/liability transfers or liquidation Component 1. Consolidation and strengthening of savings and credit institutions 1. TAs and trainings (PATyC) for supporting certification and strengthening of the SCI sector. Individual demand driven approach among the topics and TA suppliers selected by BANSEFI. Key Outputs by Component 2. Due diligence reports –prepared by consulting firms from the (linked to the achievement of the Objective/Outcome 1) BANSEFI list- for assessing and recommending the type of consolidation (liquidation, merge or transfer of assets and liabilities) of selected SCEs 3.Hire of technical agents to provide support for the strengthening of SCEs participating in PATMIR III program Objective/Outcome 2 1. Clients mainstreamed into formal financial sector Outcome Indicators 2. Number of service points providing financial services Page 33 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) 3. Number of members financially included through PATMIR of which women 1. Number of access points for correspondent agents 2. Mobile phone pilot (dropped) 3. IT Platform Uptime 4. Processor Utilization Intermediate Results Indicators 5. Bandwidth Utilization 6. Number of entities receiving technical support to use IT Platform 7. Reduction in man-hours spent on resolving errors 8. Reduction in time required for end-of-day batch runs Component 2. Broadening access to financial services and products 1. TAs to SCEs for supporting the expansion of their membership base in selected areas and locations 2. Expansion of the correspondent agents in municipalities where – mostly- BANSEFI has no branch presence 3. Implementation of financial education programs directed to SCEs, beneficiaries of conditional governmental transfers (PROSPERA) and Key Outputs by Component general public (linked to the achievement of the Objective/Outcome 2) 4. Replacement of debit cards (with new ones with chips) of PROSPERA beneficiaries, which include the provision of financial services under preferred conditions Component 3. Strengthening BANSEFI capacity 1. Improvements in the technological infrastructure capacity of the BANSEFI Technology Platform, including support and dissemination to SCEs Page 34 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Supervision/ICR Patricia Caraballo, Daniel Ortiz del Salto Task Team Leaders Gabriel Penaloza Procurement Specialist Luis Barajas Gonzalez Financial Management Specialist Pierre Olivier Colleye Former Task Team Leader (from 2011-2016) Renan Alberto Poveda Environmental Safeguards Specialist Rekha Reddy Former Task Team Leader (from January 2017-July 2017) Martin Henry Lenihan Social Safeguards Specialist B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY11 19.328 126,470.96 FY12 11.914 81,952.55 Total 31.24 208,423.51 Supervision/ICR FY12 1.062 19,202.98 FY13 18.122 86,192.89 FY14 9.398 81,500.51 FY15 10.712 75,337.20 FY16 13.017 109,612.42 FY17 23.969 138,543.92 Page 35 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) FY18 12.588 74,920.91 Total 88.87 585,310.83 Page 36 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 3. PROJECT COST BY COMPONENT 3.1 Project Costs Amount at Actual at Project Actual / planned Component Component Name Approval Closing US$ million US$ million Percentage Consolidation and strengthening of savings Component 1 and credit institutions (SCIs) 25.75 18.4 71% Broadening access to financial services and Component 2 products 140.05 114.3 82% Component 3 Strengthening BANSEFI capacity 26.07 34.9 134% Component 4 Project management 17.5 5.1 29% Front-end fee 0.25 0.3 100% Total 209.6 172.9 82% 3. 2 IBRD Funding Amount at Actual at Project Actual / planned Component Component Name Approval Closing US$ million US$ million Percentage Consolidation and strengthening of savings Component 1 and credit institutions (SCIs) 16.0 8.8 55% Broadening access to financial services and Component 2 products 56.8 39.7 70% Component 3 Strengthening BANSEFI capacity 16.0 18.7 117% Component 4 Project management 11.0 2.1 19% Front-end fee 0.25 0.25 100% Total 100.0 69.6 69.6% 3. 3 GoM Funding Amount at Actual at Project Actual / planned Component Component Name Approval Closing* US$ million US$ million Percentage Consolidation and strengthening of savings Component 1 and credit institutions (SCIs) 4.8 7.1 149% Broadening access to financial services and Component 2 products 73.3 56.1 77% Component 3 Strengthening BANSEFI capacity 9.8 16.2 166% Component 4 Project management 6.5 3.0 46% Front-end fee 0.0 0.0 0% Total 94.32 82.3 87% * The GoM funding in U.S dollars decreased to US$ 68,66 million when it is used the exchange rate at completion (MX$ 17.56 – 1 US$) as described in annex 3 of the ICR guidelines. Page 37 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) 3.4 Actual counterpart funds used during Project implementation (MX$ million) Component 2011 2012 2013 2014 2015 2016 2017 Total 1 - 19.1 19.5 23.8 14.7 16.5 9.2 102.8 2 63.1 130.0 118.1 141.7 152.7 219.0 8.7 833.3 3 - 26.9 129.6 22.0 24.6 14.6 - 217.6 4 - - - - 21.7 25.2 5.2 52.1 Total 63.1 176.1 267.2 187.5 213.7 275.2 23.1 1,205.8 3.5 GoM Funding Amount at Actual at Project Actual / planned Component Component Name Approval* Closing MX$ million MX$ million Percentage Consolidation and strengthening of savings Component 1 and credit institutions (SCIs) 59.5 102.8 173% Broadening access to financial services and Component 2 products 917.7 833.3 91% Component 3 Strengthening BANSEFI capacity 122.3 217.6 178% Component 4 Project management 81.4 52.1 64% Front-end fee 0.0 0.0 0% Total 1,180.9 1,205.8 102% * Amount calculated using the exchange rate at approval (PAD): MX$ 12.52 – 1 US$ 3. 6 SCEs Funding Actual at Project Amount at Approval Actual / planned Component Component Name Closing US$ million US$ million Percentage Consolidation and strengthening of savings Component 1 and credit institutions (SCIs) 5 2.5 50% Broadening access to financial services and Component 2 products 10 18.5 185% Component 3 Strengthening BANSEFI capacity 0.3 0 0% Component 4 Project management 0 0 0% Front-end fee 0.0 0.0 0% Total 15.3 21.0 137% Page 38 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 4. EFFICIENCY ANALYSIS Below is a list of the tasks/activities of the Project where positive or negative efficiency impact has been detected: a. High overall efficiency of the TAs: considering the number, topics and SCEs benefited through the TAs provided during the project, as well as the fact that they were helpful for the authorization and strengthening of most of the SCEs (that benefited directly or indirectly 9.5 million members/clients with total deposit of MX$ 121,322 million, US$ 6,240 million), the total cost of them (MX$ 130 million, around US$ 8.8 million) was relatively low. b. Possible efficiency gains that would have arisen from performing centralized TAs: despite some group training, most of the money spent for providing TAs and training was channeled through individual contracts between the service provider (consulting firm or federation) and the SCE, with the intervention of BANSEFI. Although the design was demand driven, and probed to be efficient in terms of supporting the SCEs for their strengthening taking into account the different stages of development and challenges they face, additional TAs and training on certain topics could have been provided through group training, with a gain in efficiency from economy of scale. c. Limited final impact of the “consolidation” due-diligence reports: the process and type of consulting firms hired was appropriate and the reports prepared were useful tool for the consolidation of SCEs. Nevertheless, the reduced rate of implementation of the recommendations (only 36 percent of them) negatively affected their efficiency. d. Efficiency improvement from doing transactions through correspondent agents: the effective opening of almost 2,500 new access points of correspondent agents, and considering that almost 69 percent are located in places where BANSEFI has no branches, has reduced the cost of BANSEFI´s clients for doing financial transactions (the evidence of this preference was the significant increase in the number of transactions performed through agent in the last months). Additionally, from BANSEFI point of view, it was much more efficient (and cost effective) in certain locations to have presence through the access point of the correspondent agent instead of opening a new branch (although is clear that certain operations, like opening an account or getting a credit, cannot be done through a correspondent agent). e. Efficiency weaknesses in the process of expanding correspondent agents: the continuous effort, money and time spent in trying to expand the access points of correspondent agents through individual firms (e.g. DICONSA and PEMEX) might have a negative impact on efficiency, since the expansion came finally from contracting an Agent Manager which would have been possible to implement the first year of the project (with the implicit reduction in the cost of the projects and on BANSEFI´s clients. f. Positive efficiency impact from the implementation of PATMIR III, in terms of number and type of people financially included as well as on the results from the demand of financial services (saving, loans, insurance): the program (with a total support from the Project of MX$ 1,216 million, the highest of all the components and subcomponents) had a significant impact in terms of financial inclusion in selected areas and locations, as well as strengthening the SCEs and increasing their viability. The program maintained its nature oriented to the sustainable financial inclusion of predominantly rural population, marginalized and with little access to formal financial services. The financial inclusion of 1,027,449 members and clients that met the standards to determine the use of financial services was achieved, of which 604,037 are women (58.79% of the total), which exceeded the Page 39 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) two goals contemplated in the Project. The percentage of population included in indigenous regions that was served exceeded the original goal of 10%, given to the great attention to this segment and the follow-up of the indigenous participation plans. Now, more people in marginalized areas will have tools to stabilize their income and consumption, take advantage of investment opportunities, face economic adversities and reduce vulnerability to basic life risks, and generate greater economic development and wellbeing. The approach of using technical agents (and the selection of them, through a bidding process among the ones technically fit) for assisting the participating SCEs combined with and approach mechanisms tailor made for the target people, were key for the results obtained. The process has proved to be efficient. In terms of the value of the money spent, for each MX$ 1 spent, there were MX$ 5 of savings, another MX$ 5 of credits and MX$ 2 of amount insured. g. Efficiency impact from the delays in the implementation of the activities due to several factors: the delays in the implementation coming from the fiscal cuts, as well as from the development of the new BANSEFI strategy and the frequent changes in key top management of BANSEFI might affect the efficiency in the implementation. The TAs that were not being provided could have further strengthen the SCEs, improving their governance, risk management, compliance with the regulatory framework and performance. h. Relatively high investment in technology and IT support with very limited usage by the SCEs of the IT facilities provided by BANSEFI: although the BTP was used -at a given point- by more than 106 SCEs, while a total of 75 entities have received technical support on it, as of July 2017, only 7 SCEs were using the banking core and 7 the AML module. This limited usage was consequence of some weaknesses of the BTP related with the capacity to: incorporate the changes in the regulatory framework, comply with the requirements from the Tax Agency and capacity for the integration with the accounting system of the SCEs. Although the number of SECs using the BTP currently is reduced, the TAs received by the SCEs, as well as the experience of using the BTP, provided the SCEs with the expertise and experience to be better prepared for acquiring and implementing their own IT systems either for their core banking or for AML risk management. Taking into account these results, the amount of money disbursed under Component 3 involving this activity presented low levels of efficiency. Page 40 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS Government comments were included in the ICR document. Page 41 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 6. BIBLIOGRAPHY BANSEFI (2017). Informe Final de Implementación; Préstamo 8107-MX; Mexico City. Berumen (2016). Impact Evaluation for the Regional Rural Microfinance Technical Assistance Program (PATMIR); Mexico City. Comisión Nacional Bancaria y de Valores (2012). Reporte Nacional de Inclusión Financiera. Reporte No.4; México. Comisión Nacional Bancaria y de Valores (2017). Reporte Nacional de Inclusión Financiera. Reporte No.8; México. EASE (2016). Evaluación de Resultados del Programa Integral de Inclusión Financiera (PROIIF), México. Pierre Bachas, Paul Gertler, Sean Higgins, and Enrique Seira. Banking on Trust: How Debit Cards Enable the Poor to Save More. NBER Working Paper No. 23252, March 2017. http://www.nber.org/papers/w23252 Spectron Desarrollo (2016). Evaluación del Impacto del Componente de Asistencia Técnica y Capacitación; Préstamo 8107; México D.F. Spectron Desarrollo (2016). Evaluación de Impacto del Piloto Ampliado del Programa para el Desarrollo de la Inteligencia Económica Y Financiera de Niños y Jóvenes de BANSEFI; México D.F. World Bank (2008). Country Partnership Strategy for the United Mexican States; Report No.42846-MX; Washington D.C. World Bank (2010). Country Partnership Strategy Progress Report for the United Mexican States; Report No.52776- MX; Washington D.C. World Bank (2011). Project Appraisal Document; Report No. 63733-MX; Washington, D.C. World Bank (2011). Loan Agreement; Loan Number 8107-MX; Mexico City. World Bank (2011). Savings and Credit Sector Strengthening and Rural Microfinance Capacity Building Project; Project Documents; Retrieved from http://projects.worldbank.org/P070108/savings-credit-sector-strengthening-rural- microfinance-capacity-building?lang=en World Bank (2012). Savings and Rural Finance Project; Project Documents; Retrieved from http://projects.worldbank.org/P087152/mexico-savings-rural-finance-bansefi-project-phase- ii?lang=en&tab=overview World Bank (2014). Country Partnership Strategy for the United Mexican States; Report No.83496-MX; Washington D.C. World Bank (2014). Restructuring Paper. Saving and Credit Sector Consolidation and Financial Inclusion Project; Report No. RES14690; Washington D.C. World Bank (2016). Restructuring Paper. Saving and Credit Sector Consolidation and Financial Inclusion Project; Report No. RES123246; Washington D.C. World Bank (2017). Saving and Credit Sector Consolidation and Financial Inclusion Project; Project Documents; Retrieved from http://projects.worldbank.org/P123367/rural-savings-credit-sector- consolidation?lang=en&tab=results World Bank Global Findex (2017). Financial Data. Retrieved from http://www.worldbank.org/en/programs/globalfindex Page 42 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 7. SUPPORTING INFORMATION Annex 7.1: Account Penetration Comparison with Peer Countries (Paragraphs 1 and 2) Percentage of Adults (15+) with an account at a Formal Institution in 2011 55.9 53.6 48.2 33.1 32.0 30.4 27.4 Argentina Brazil Colombia Mexico Russian South Africa LAC Average Federation Source: WB Global Findex Percentage of Adults (income, poorest 40%) with an account at a Financial Institution in 2011 39.4 39 38.8 25.3 19.2 13.3 11.9 Argentina Brazil Colombia Mexico Russian South Africa LAC Average Federation Source: WB Global Findex Percentage of Adults (15+) in rural areas with an account at a Financial Institution in 54.0 2011 41.8 43.6 38.0 33.6 26.9 13.6 Argentina Brazil Colombia Mexico Russian South Africa LAC Average Source: WB Global Findex Federation Page 43 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Annex 7.2: Activities performed per component Comp. Summary of each component and activities involved 1 • Support to carry out assessments of selected SCEs for certification, strengthening or consolidation, as well as provision of TA and training to selected SCEs for their institutional strengthening and service improvement; • Provision of goods, technical assistance and training to strengthen the sector’s second and third tier institutions (i.e. Confederation, Federations and Auxiliary Supervision Committee) to perform their roles; • Carrying out of sector-related studies and evaluations to establish a consolidated database on the sector; • Carrying out of dissemination activities related to certification and reporting requirements. 2 • TA and training to selected SCEs for expanding their membership in selected areas and locations and to support the implementation of improvement programs; • Expansion of BANSEFI´s clients base in selected locations through the provision of support from banking agents; • Expansion of the range of financial services to be provided to SCEs and individual clients through: TA to improve the provision of lines of credits from BANSEFI to SCEs; strengthening of the L@ Red de la Gente Network to offer an expanded range of financial services to SCEs; carrying out of a pilot program and scaling up for mobile phone banking operations targeted to BANSEFI clients; and support for dissemination of the expanded services; • Provision of financial education through: TA to civil society and sector organizations to build capacity for dissemination of financial information; direct support of measures to educate the general public on financial services; the development of materials to be distributed at outreach events; and carrying out of evaluation studies to measure impact. 3 • Provision of services to expand the technological capacity of BANSEFI, including its information technology platform’s mainframe, to enhance its processing and storage capacity and efficiency and provide TA and training to SCEs for adopting BANSEFI’s information technology platform; • Provision of training for BANSEFI’s staff on topics related to the Project ; • Provision of support for the dissemination of the functional capabilities of BANSEFI’s information technology platform and on BANSEFI’s brand as a development bank. 4 • Provision of TA, training and material support to BANSEFI’s staff, consultants and providers of non- consulting services in order to assist them in the implementation of the Project. Page 44 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Annex 7.3: Technical Assistance provided with Project support (paragraph 20) Total number and amount paid for TAs (per year) Technical Assistance Amount Paid Year Number Percentage MS$ million Percentage 2012 293 19% 21,818 16% 2013 322 21% 23,291 18% 2014 243 16% 20,646 16% 2015 287 18% 26,211 20% 2016 255 16% 25,504 19% 2017 153 10% 14,828 11% Total 1,553 100% 132,298 100% Source: BANSEFI Distribution per type of Technical Assistance Cost of Others Supervision 6.50% 4.80% AML 6.90% Cost of Certification 7.60% Risk Mgt, Corporate Gov. Strategic Credit process, Planning Internal Control 13.80% 60.40% Source: BANSEFI Annex 7.4 Evolution of transactions performed through correspondent agents (paragraph 28) Transaction / Year 2012 2013 2014 2015 2016 2017* Balance inquiry 2,449 2,795 1,477 1,022 9,064 58,819 Deposit 27,297 14,572 15,208 8,864 13,637 21,957 Payment of utilities 5,217 5,950 4,704 4,120 1,671 564 Withdraw 3,637 4,286 1,563 1,165 6,331 40,335 Total 38,600 27,603 22,952 15,171 30,703 121,675 Monthly average 3,217 2,300 1,913 1,264 2,559 17,382 Source: BANSEFI * Jan-July 2017 Page 45 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Annex 7.5: Financial education initiatives SCEs • 2012-2016 95,000 people financially educated in marginal areas, with 1,000 trainers • Redefinition of the program (2016), considering inputs from previous programs: 202 trainers were trained, with 30,000 people receiving financial education in 1,178 workshops • In conjunction with SCEs member of “L@ Red de la Gente”, a financial education pr ogram called "The value of closeness" was carried out, with almost 14,000 people received sessions. PROIIF • 2014-2017 350 facilitators trained who provided financial education in 27,000 events to 1.6 million of beneficiaries of PROSPERA (2.0 million of financial education material kits were distributed) at the moment their debit cards were replaced to new ones with chips. The program was suspended at the end of 2016, due to budget constraints that limited the support and logistic activities from the Social Development Secretary. General • Nov 2015-Nov 2016: financial education to 114,000 students (preschool, primary and secondary) in Public 332 schools (this activity had an external evaluation, which showed more impact on students at primary school, followed by high school and preschool) • BANSEFI as a Certification and Evaluation Entity in financial education by the National Council for Standardization and Certification of Labor Competencies, which allows it to professionalize its personnel and the other public and private institutions. Since February 2016, more than 110 people have been certified in labor competency standards in the area of financial education; there are 34 independent evaluators that will be able to replicate these processes. • Promotion of a digital financial education training module (called “MOOC”) • Pilot program for financial education and inclusion targeted to women (2,300 persons) attending it; • Participation in different financial education events, such as the National Financial Education Week Annex 7.6: Importance of the Authorized SCEs in terms of Number, Assets and Clients of the Sector (Paragraph 70) Distribution of the number of SCEs Others Authorized 20% 23% Basic Level Source: BANSEFI 57% Page 46 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) Distribution of Clients of the SCEs Others Basic Level 11% 3% Authorized 87% Source: BANSEFI Distribution of Assets of the SCEs Others Basic Level 8% 1% Authorized Source: BANSEFI 91% Page 47 of 48 The World Bank Savings and Credit Sector Consolidation and Financial Inclusion Project ( P123367 ) ANNEX 8. THEORY OF CHANGE Page 48 of 48