Document of The World Bank FOR OFFICIAL USE ONLY Report No. 55657-PE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT PROGRAM DOCUMENT FOR A PROPOSED THIRD RESULTS AND ACCOUNTABILITY (REACT) DEVELOPMENT POLICY LOAN IN THE AMOUNT OF US$50.0 MILLION TO THE REPUBLIC OF PERU October 29, 2010 Human Development Sector Management Unit Bolivia, Chile, Ecuador, Peru and Venezuela Country Management Unit Latin America and the Caribbean Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. PERU - GOVERNMENT FISCAL YEAR January 1 ­ December 31 CURRENCY EQUIVALENTS Exchange Rate Effective as of September 22, 2010 Currency Unit = Nuevo Sol US$1.00 = S/.2.79 ABBREVIATION AND ACRONYMS AAA Analytical and Advisory Activities APEC Asia-Pacific Economic Cooperation BCRP The Central Reserve Bank of Peru CAS Country Assistance Strategy CCT Conditional Cash Transfer (Programa de Transferencias Condicionadas) CENAN National Center for Food and Nutrition (Centro Nacional de Alimentación y Nutrición) CFAA Country Financial Accountability Assessment CFH Community Health Fund CGR Office of the Comptroller General CIAS Inter-ministerial Commission for Social Affairs (Comisión Interministerial de Asuntos Sociales) CLAS Local Health Administration Committees (Comités Locales de Administración de Salud) CONEACE Council for the Evaluation, Accreditation, and Certification of Non-University Higher Education (Consejo de Evaluación, Acreditación y Certificación de la Calidad de la Educación Superior No Universitaria) CONEAU National Commission for University Evaluation and Accreditation (Comisión Nacional de Evaluación y Acreditación Universitaria) CONEI Institutional Education Council (Consejo Educativo Institucional) CPAR Country Procurement Assessment Report CPS Country Partnership Strategy CRED Child Growth and Development (Sub Programa de Control de Crecimiento y Desarrollo) CUI Unique Ppersonal Iidentity Nnumber (Código Unico de Identidad) CUS Committee of Health Beneficiaries (Comités de Usuarios de Salud) DFID Great Britain's Department for International Development DIRESA Regional Health Directorates (Direcciones Regionales de Salud) DNI National Identity Number (Documento Nacional de Identidad) i DNPP National Public Budget Directorate (Dirección Nacional de Presupuesto Público) DPL Development Policy Loan FAO Food and Agriculture Organization of the United Nations FDI Foreign Direct Investment Network FONCODES Fondo de Cooperación para el Desarrollo Social (Social Funds) FONIPREL Fondo de Promoción a la Inversión Pública Regional y Local GDP Gross Domestic Product GOP Government of Peru GTZ German Technical Cooperation Agency HIS Health Information System IB Institutional Births IBRD International Bank for Reconstruction and Development IDB Inter-American Development Bank IFC International Finance Corporation IMF International Monetary Fund INEI National Statistical Institute (Instituto Nacional de Estadística e Informática) IPEBA Peruvian Institute for the Evaluation, Accreditation, and Certification of Basic Education (Instituto Peruano de Evaluación, Acreditación y Certificación de la Calidad de la Educación Básica) JSDF Japan Social Development Fund LAC Latin America and the Caribbean Region MCLCP Citizen Roundtables to Fight Poverty (Mesas de Concertación para la Lucha Contra la Pobreza) MEF Ministry of Economy and Finance (Ministerio de Economía y Finanzas) MIMDES Ministry of Women and Social Development (Ministerio de la Mujer y Desarrollo Social) MINEDU Ministry of Education (Ministerio de Educación) MINSA Ministry of Health (Ministerio de Salud) NGO Non-Governmental Organization OPIC Overseas Private Investment Corporation PAHO Pan-American Health Organization PAN Articulated Nutrition Program (Programa Articulado Nutricional) PAS Regional Performance Agreements PBL Public Budget Law PBM Participatory Budgeting and Monitoring (Presupuesto y Monitoreo Participativo) PCM Presidency of the Council of Ministers PEAS Basic Health Insurance Plan (Plan Esencial de Aseguramiento en Salud) PER Public Expenditure Review PFM Public Financial Management ii PHRD Japan Policy and Human Resources Development Fund PIA Initial Budget (Presupuesto Inicial de Apertura) PIM Modified Budget (Presupuesto Inicial Modificado) PpR Results Based Budget (Presupuesto por Resultados) PRONAA The National Food Assistance Program (Programa Nacional de Asistencia Alimentaria) PSIA Poverty & Social Impact Assessment RBB Results Based Budget REACT Results and Accountability RENIEC National Registry for Identification and Civil Status (Registro Nacional de Identificación y Estado Civil) ROSC Report on the Observance of Standards and Codes RUB Single Beneficiary Registry (Registro Único de Beneficiarios) SDR Special Drawing Rights SIAF Integrated Financial Administration System (Sistema Integrado de Administración Financiera) SINEACE National System for Evaluation, Accreditation and Certification of Education Quality (Sistema Nacional de Evaluación, Acreditación y Certificación de la Calidad Educativa) SIS Integrated Health Insurance (Seguro Integral de Salud) SISFOH Household Targeting System (Sistema de Focalización de Hogares) TSA Treasury Single Account UGEL Local Educational Management Units (Unidades de Gestión Educativa Local) UMC Quality Measurement Unit (Unidad de Medición de la Calidad) UNDP United Nations Development Program UNICEF United Nations Children's Fund USAID United States Agency for International Development UST Universal Standard WHO World Health Organization Vice President: Pamela Cox Country Director: Carlos Felipe Jaramillo Sector Director: Keith Hansen Sector Manager: Helena Ribe Task Team Leader: Ian Walker iii PERU THIRD RESULTS AND ACCOUNTABILITY (REACT) DEVELOPMENT POLICY LOAN TABLE OF CONTENTS LOAN AND PROGRAM SUMMARY ............................................................................................................. 6 I. INTRODUCTION ................................................................................................................................ 8 II. COUNTRY CONTEXT ....................................................................................................................... 9 ECONOMIC DEVELOPMENTS .......................................................................................................... 9 MACROECONOMIC OUTLOOK AND DEBT SUSTAINABILITY ............................................... 13 III. THE GOVERNMENT'S PROGRAM AND PARTICIPATORY PROCESSES ......................... 14 IV. BANK SUPPORT TO THE GOVERNMENT'S STRATEGY ...................................................... 42 LINK TO CPS ...................................................................................................................................... 42 COLLABORATION WITH THE IMF AND OTHER DONORS ....................................................... 44 RELATIONSHIP TO OTHER BANK OPERATIONS ....................................................................... 45 LESSONS LEARNED ......................................................................................................................... 46 ANALYTICAL UNDERPINNINGS ................................................................................................... 48 V. THE PROPOSED THIRD RESULTS AND ACCOUNTABILITY (REACT) DPL .................... 48 POLICY AREAS ................................................................................................................................. 51 VI. OPERATION IMPLEMENTATION ............................................................................................... 56 POVERTY AND SOCIAL IMPACTS ................................................................................................ 56 PARTICIPATION ................................................................................................................................ 61 ENVIRONMENTAL ASPECTS ......................................................................................................... 62 IMPLEMENTATION, MONITORING AND EVALUATION .......................................................... 62 FIDUCIARY ARRANGEMENTS ...................................................................................................... 63 DISBURSEMENT AND AUDITING ................................................................................................. 64 RISKS AND RISK MITIGATION ...................................................................................................... 65 ANNEXES ANNEX 1: LETTER OF DEVELOPMENT POLICY....................................................................................... 67 ANNEX 2: PERU REACT III - OPERATION POLICY MATRIX ................................................................. 82 ANNEX 3: FUND RELATIONS NOTE............................................................................................................ 90 ANNEX 4: FISCAL AND EXTERNAL DEBT SUSTAINABILITY ANALYSIS .......................................... 91 ANNEX 5: COUNTRY AT A GLANCE........................................................................................................... 94 ANNEX 6: MAP IBRD 33465 ........................................................................................................................... 97 FIGURES FIGURE 1: REAL GDP GROWTH ................................................................................................................... 10 FIGURE 2: GENERAL GOVERNMENT BUDGET BALANCE .................................................................... 10 FIGURE 3: MONETARY POLICY REFERENCE RATE AND INFLATION ................................................ 12 FIGURE 4: EXCHANGE RATE AND INTERNATIONAL RESERVES ........................................................ 12 FIGURE 5: NUMBER OF SISFOH'S LOCAL TARGETING UNITS BY REGION, 2010 ............................ 17 FIGURE 6: CHRONIC MALNUTRITION AND ALLOCATED BUDGET .................................................... 23 FIGURE 7: INSTITUTIONAL BIRTHS COVERAGE AT NATIONAL, RURAL & URBAN LEVELS ....... 33 FIGURE 8: STRUCTURE OF INSURANCE BY QUINTILE, 2003 AND 2008 ............................................ 36 FIGURE 9: HEALTH INSURANCE BENEFICIARIES, BY QUINTILE, 2008 .............................................. 36 iv TABLES TABLE 1: KEY ECONOMIC INDICATORS (PERCENT UNLESS OTHERWISE NOTED) ....................... 11 TABLE 2: POVERTY RATES, 2004-2009 (PERCENT) ................................................................................. 13 TABLE 3: PPR STRATEGIC PROGRAMS ..................................................................................................... 22 TABLE 4: INSTITUTIONAL BIRTHS COVERAGE BY INCOME QUINTILE............................................ 33 TABLE 5: TYPE OF DELIVERY BY EDUCATION AND MOTHER TONGUE .......................................... 35 TABLE 6: RECENT TRENDS IN CHRONIC MALNUTRITION RATES IN PERU ..................................... 39 TABLE 7: ANALYTICAL AND ADVISORY WORK UNDERPINNING THE REACT DPL SERIES ....... 50 TABLE 8: SUMMARY OF REVISED PRIOR ACTIONS AND TRIGGERS FOR THE PROGRAMMATIC PERU REACT DPL SERIES ............................................................................................................................. 52 TABLE 9: CONSULTATION AND PARTICIPATORY ACTIVITIES RELATED TO THE SOCIAL SECTORS REFORM PROGRAM IMPLEMENTATION DURING 2009-2010 ................................................................ 61 TEXT BOXES BOX 1: SYSTEMATIZING IDENTIFICATION DATA IN BIRTH CERTIFICATES ................................... 20 BOX 2: PARTICIPATORY BUDGETING AND ITS RELATION TO RESULTS-BASED - BUDGETING 26 BOX 3: APPLICATION OF GOOD PRACTICE PRINCIPLES FOR CONDITIONALITY IN REACT DPL 54 BOX 4: QUALITATIVE EVIDENCE ON INDIGENOUS WOMEN'S DEMAND FOR INSTITUTIONAL BIRTHS .............................................................................................................................................................. 60 The PERU REACT DPL III Loan was prepared by an IBRD team consisting of Ian Walker (LCSHS) Task Team Leader, Alessandra Marini and Nelson Gutierrez (LCSHS), Fernando Lavadenz (LCSHH), Ines Kudó (LCSSE), Marcelo Becerra (LCSHE), Oscar Calvo Gonzalez (LCSPE), Lourdes Linares and Nelly Ikeda (LCSFM), Patricia Hoyes (CTRFC), Elisa Seguin (JPA, LCSHH), Solange Alliali and Teresa Genta-Fons (LEGLA), William Reuben (consultant), Melissa Zumaeta (consultant), Ana Maria Muñoz Boudet (consultant), Lerick Kebeck, Patricia Orna, and Sara Burga (LCSHD). v LOAN AND PROGRAM SUMMARY PERU THIRD RESULTS AND ACCOUNTABILITY (REACT) DEVELOPMENT POLICY LOAN Borrower Republic of Peru Implementing MINISTRY OF FINANCE Agency Financing Data Fixed-spread loan of US$50,000,000 United States Dollars with a repayment schedule linked to commitments and with all the conversion options. Fee equal to one quarter of one percent (.25 percent) of the loan amount, payable up front from the Borrower's own proceeds. Loan payable in 12.5 years, including an 11 year grace period with a custom repayment schedule with repayments on each 15th April and 15th October respectively. Operation type DPL Main Policy Areas This is the third in a planned series of three DPLs (programmed for FY 07, 09 and 11) to support the strengthening of the Results and Accountability (REACT) framework in Peru's social sectors. The program builds on the recommendations of the "RECURSO" Analytical and Supervisory Activity (AAA) report ("A New Social Contract for Peru"). To help basic education, health and nutrition services break out of prevailing "low quality equilibria", the REACT DPL promotes transparent performance standards and encourages parents and local governments (whose role is growing due to decentralization) to demand improvements. The ministries of Education and Health have set standards for children's learning, health and nutrition outcomes, and established goals for the proportion of children achieving them. They are developing monitoring systems to track the performance of health posts and schools, and to provide reports for parents on the health, nutrition and learning status of their children. The operation also supports complementary actions to strengthen the links between budgeting and outcomes (through the roll-out of performance budgeting in the social sectors), to improve service quality, reduce exclusion, improve targeting and increase the participation of poor communities in budgeting and program monitoring. The REACT DPL complements a parallel series of fiscal DPLs, which is supporting the development of the umbrella system of results-based management in MEF. Key Outcome Improved outcomes of priority social programs, as follows: Identity: Indicators Peruvian children born after 2008 will automatically receive a unique 6 identification number (CUI) when their birth is registered. Education: Improved 2nd grade reading comprehension reflected in an increase in students registered as fully proficient from 16 percent in 2007 to 35 percent by 2011; and improved 2nd grade math performance reflected in an increase in students registered as fully proficient from 7 percent in 2007 to 30 percent by 2011. Health: Rural Institutional Birth coverage increases from 49.4 percent in 2007 to 70 percent in 2011. Maternal mortality is reduced from 164 per 100,000 live births in 2007 to 120 in 2011; and Neo-natal mortality is reduced from 15.2 per 1,000 live births in 2007 to 14.6 in 2011. Nutrition: Average frequency of nutritional monitoring of children reaches 4.0 CRED controls per year in the 1st year of life; and 2.0 per year in the 2nd year of life in 2009. Stunting rates for children under-5 are reduced from the baseline of 28.5 percent in 2007 to 23 percent by 2011, on the new WHO standard. Program The REACT DPL series supports policies which are expected to lead Development to: (i) improved parental understanding of expected outcomes in Objectives and education, health and nutrition; (ii) to improved outcomes in second Contributions to grade literacy (especially in rural schools); (iii) reduced maternal and CAS/ Country neo-natal mortality; and (iv) better nutrition outcomes. These are all Partnership Strategy highlighted as key development outcomes of the CPS. Risks and Risk Political risks: The Garcia administration which has promoted the Mitigation reforms supported by REACT will come to the end of its term in 2011. The Bank team will work with the incoming administration, through the Policy Notes process, to ensure that they understand the purpose and achievements to date of the reforms, and to offer continuing Bank support for this agenda. Social and environmental risks: In education, the fear that universal standard testing of education outcomes might lead to conflicts with the teachers has proved unfounded; the vast majority of schools nationwide are now participating. It remains important to institutionalize the use of the results in a way that is perceived as supportive (and not threatening) by the teachers. In health, there is a risk that birth services might be culturally inappropriate, leading to inequity the exclusion of indigenous groups. A PSIA study has been undertaken to identify cultural obstacles to increased coverage of institutional births and based on the findings, compensating measures were introduced into the policy matrix. Economic risks: The Country's strong macro- fiscal position has allowed it to weather the global economic and liquidity crisis without a major domestic contraction, but it now faces the need to rebalance fiscal accounts as the recovery develops. World Bank support through this operation and other DPLs linked to fiscal and environmental policy have helped to increase Peru's access to budgetary support to tackle fiscal contingencies. Operation ID Number PE-P116264-LEN-BB 7 NEGOTIATIONS VERSION IBRD PROGRAM DOCUMENT FOR A PROPOSED THIRD RESULTS AND ACCOUNTABILITY (REACT) DEVELOPMENT POLICY LOAN TO PERU I. INTRODUCTION 1. The Peru Results and Accountability (REACT) DPL is the third in a planned series of three operations to support the Government of Peru (GoP) in strengthening the accountability framework and improving results in health, nutrition and education. 2. Over the last 15 years, Peru has expanded the coverage of basic social programs, but the quality of services remains poor. There are many causes of poor quality, but a critical factor preventing change is the absence of clearly defined standards and results frameworks, across the social sectors. The consequence is inadequate human capital formation, as reflected in the health, nutrition and education status of children. If these issues are not addressed, the resulting labor supply deficiencies will constitute a major obstacle to the country's medium-term economic competitiveness strategy, undermining efforts to diversify away from traditional exports towards activities with more local value-added, which require a more sophisticated labor force. 3. Although Peru is a middle-income country, with near-universal coverage of primary education, at the start of the REACT program only 15 percent of second grade students reached full sufficiency in literacy and almost half were unable to read at all. In health, 25 percent of births nationwide were still not institutionally supported ­ rising to 45 percent in the 10 poorest departments ­ hampering efforts to reduce maternal and peri natal mortality. Thirty percent of five year old children in Peru were chronically malnourished (stunted). In each case, outcomes were particularly bad in rural areas and among indigenous communities. For these reasons, service quality, results and accountability were placed at center-stage in the Garcia administration's economic development and poverty reduction strategy. 4. The Peru REACT DPL series is supporting actions in primary education, health and nutrition to define standards and set goals for the outcomes which families should expect for their children. This is coupled with the development of robust monitoring systems which provide up-to-date information on the performance of health posts and schools and individualized data for parents on the health, nutrition and learning status of their children. The operation also supports actions to reduce exclusion, rationalize programs, to improve targeting and to increase the participation of poor communities in budget processes and program monitoring. The REACT DPL series is expected to lead to improved outcomes in second grade literacy (especially in rural schools); increased access to institutional birth services; and increased coverage of individualized growth monitoring and counseling services for children under 24 months of age in areas with a high incidence of chronic malnutrition (stunting), leading to better nutrition outcomes. 5. The GoP has requested Bank financing for three main reasons. First, while Peru has achieved strong growth and poverty reduction in recent years, it still faces a high poverty rate as well as other important development challenges in areas such as infrastructure and human capital formation. This operation supports implementation of key policy reforms that can help increase 8 the impact of public spending on health, nutrition, and education. Second, this is the third and last operation in a programmatic series agreed with the government under the current CPS and which has been included in the government's financing plan. As such, this operation has been programmed to finance upcoming expenditures during 2010. Third, while global financial markets have improved, there remain uncertainties and risks that the authorities seek to mitigate with further Bank financing. The GoP seeks to strike an appropriate balance between protecting the outstanding results in debt management and maintaining flexibility to respond to a volatile global environment. In this regard, the authorities made a strategic choice of maintaining DPL/DDOs as contingent lines of credit to meet their financing needs in 2011 and 2012 while drawing on traditional DPLs for immediate budget support during 2010. The Peruvian authorities have informed the Bank that they are currently planning to disburse over 50 percent of the balance of DPL/DDOs during CY2011. II. COUNTRY CONTEXT ECONOMIC DEVELOPMENTS 6. Prior to the global economic crisis, Peru enjoyed a period of broad-based rapid economic growth. In the run-up to the global economic crisis Peru displayed a strong economic performance, with growth accelerating from 6.4 percent in 2005 to 9.8 percent in 2008. Rising commodity prices fueled export growth but the economic expansion from 2006 to 2008 was also based on buoyant domestic demand and was led by the private sector. Consumption and investment contributed roughly equally to economic growth. Private investment rose from 15 percent of Gross Domestic Product (GDP) in 2005 to around 22 percent of GDP in 2008, supported by a positive outlook and sound macroeconomic management. Strong capital inflows led to a reserve build up and put appreciation pressure on the Nuevo Sol. FDI inflows surged from 3.2 percent of GDP in 2005 to around 5 percent of GDP in 2008. Bond spreads declined and Peru's sovereign foreign currency rating was upgraded to investment grade first by Fitch and Standard and Poor's in 2008 and then by Moody's in 2009. 7. Sound macroeconomic management during the boom years created the fiscal space needed for countercyclical policies to soften the impact of the global economic crisis. The GoP maintained a prudent fiscal policy through the commodity boom period, running fiscal surpluses in the three years to 2008, with the overall budget surplus peaking at 2 percent of GDP in 2008. The authorities saved resources into a fiscal stabilization fund, which saw its balance increase from US$314 million in 2005 to US$1.8 billion in 2008 (around 1.4 percent of GDP). The authorities also brought public external debt down from 28 percent of GDP in 2005 to 15 percent in 2008. In parallel, the Central Bank (BCRP) accumulated net international reserves which increased from US$14 billion in 2005 to US$31 billion in 2008--close to five times the stock of short-term external debt and 11 months of imports of goods and services. 8. Despite the magnitude of the challenges generated by the global economic crisis, financial and exchange rate stability were preserved. The sharp increase in risk aversion in global financial markets in mid-September 2008 was quickly felt in Peru. From mid-September to late October 2008 net international reserves fell by US$3.3 billion, or close to 10 percent of holdings. However, the central bank avoided sharp fluctuations of the exchange rate. The global economic 9 crisis also had a relatively limited impact on the Peruvian financial sector, which did not rely heavily on short-term external borrowing. In addition, there was no credit crunch as the growth of credit to the private sector remained positive, slowing down gradually to an annual rate of 9 percent at its trough in December 2009. 9. Economic growth decelerated sharply in 2009 but remained positive, as external demand and an anti-cyclical public spending program propped up growth. GDP growth slowed down to 0.9 percent in 2009 driven largely by a fall in domestic demand, and in particular by a sharp drop in private investment. In contrast, the external demand had a positive contribution to growth as a 15 percent decline in exports was offset by a 26 percent decline in imports, as demand for imported capital and intermediate goods dropped sharply on account of the virtual standstill from private investment. Figure 1: Real GDP growth Figure 2: General government budget balance Annual percentage change Percent of GDP 16 8 14 6 12 4 10 2 8 0 6 2 4 4 2 6 0 8 -2 10 -4 1Q2008 2Q2008 3Q2008 4Q2008 1Q2009 2Q2009 3Q2009 4Q2009 1Q2010 2Q2010 Sep-08 Sep-09 Jan-08 Jan-09 Jan-10 May-08 May-09 May-10 Source: INEI. Source: Bloomberg. 10. In response to the economic slowdown the authorities launched in 2009 a two-year economic stimulus plan amounting to US$4.8 billion, or about 3.5 percent of GDP. The plan focused primarily on increased public expenditures and attempts to balance the need for a stimulus in the short-term which reflected also medium-term priorities. Infrastructure projects accounted for around 51 percent of planned expenditures. In particular, about US$412 million were allocated for priority projects and US$567 million have been allocated to projects which were already in execution but had undisbursed budget balances in 2008--effectively funding the `carry over' of those projects into 2009. An additional US$912 million was earmarked to infrastructure projects to be chosen and implemented by regional governments. As of end-July 2010 the disbursement ratio of the total economic stimulus plan was around 94 percent. 11. The fiscal stimulus plan combined with lower fiscal revenues caused a fiscal deficit of 1.9 percent of GDP in 2009. Peru's budget balance turned negative in 2009 due to both lower revenues and increased public expenditure associated with the economic stimulus plan. The existing Fiscal Responsibility and Transparency Law sets two limits: (i) the fiscal deficit is not to 10 exceed 1 percent of GDP, and (ii) the growth of current expenditures of the central government is not to exceed 3 percent p.a. in real terms. Article 5 of the Law states, however, that in cases of national emergency or international crisis the legislature can, at the request of the executive, grant exceptions to the fiscal rules for a period of up to three years. The administration proposed legislation--which the Congress approved-- to increase the limit on the fiscal deficit to 2 percent of GDP in both 2009 and 2010. In addition, the limit on the real growth of current expenditures by the central government--set at 3 percent bylaw--was also temporarily lifted to 10 percent in 2009 and 8 percent in 2010. Table 1: Key economic indicators (percent unless otherwise noted) 2007 2008 2009 2010 2011 2012 2013 GDP growth rate 8.9 9.8 0.9 8.0 5.5 5.5 5.5 Growth of real domestic demand 11.9 12.1 -2.9 10.9 6.2 6.2 6.2 Inflation rate, CPI (period average) 1.8 5.8 2.9 1.4 3.0 2.0 1.8 Non-financial public sector balance/GDP 3.3 2.0 -1.9 -1.5 -1.0 -0.4 0.4 Public debt/GDP 29.7 24.0 26.7 23.4 22.3 21.1 19.8 Public external debt/GDP 18.7 15.1 16.3 14.2 13.6 12.6 11.6 Exports (FOB), change p.a. 17.5 13.1 -14.7 21.5 6.1 6.5 7.6 Imports (CIF), change p.a. 31.8 45.1 -26.1 22.5 9.7 10.0 10.0 Trade balance/GDP 7.7 2.4 4.6 4.6 3.9 3.3 2.9 External Current Account/GDP 1.4 -3.3 0.2 -0.4 -0.9 -1.4 -1.9 Terms of Trade (deterioration -); change 2.4 -2.9 -5.5 11.0 -0.4 -0.1 0.7 Net international reserves (NIR), US$ billion 28.6 32.2 33.1 42.5 44.2 46.0 48.2 NIR, in months of G&S imports 13.9 11.4 15.4 16.3 15.4 14.7 14.0 NIR, as % of public external debt 143 167 161 198 196 205 216 NIR, as % of Money Stock 80 89 88 95 90 87 85 Source: Banco Central de Reserva del Perú (BCRP), Ministry of Economy and Finance (MEF) --Marco Macroeconómico Multianual 2011-2013, World Bank staff, IMF staff. 12. In parallel, the Central Bank eased monetary policy as inflation decreased and exchange rate appreciation pressures resumed. The central bank cut its policy rate every month from February to August 2009, bringing the rate to 1.25 percent for a cumulative easing since the beginning of 2009 of 525 basis points. In addition, the central bank adopted a number of measures aimed at supporting the flow of credit, such as reducing reserve requirements, accepting new instruments as collateral for liquidity operations, and the extension of the term of liquidity management operations. 11 Figure 3: Monetary policy reference rate Figure 4: Exchange rate and international and inflation reserves Percent Nuevos soles per US$ (left scale), and US$ billion (right scale) 8 2.5 50 Monetary policy 7 reference rate Inflation 2.7 40 6 5 2.9 30 4 3 3.1 20 2 1 3.3 10 Exchange rate (left axis, inverted scale) 0 Net international reserves (right axis) Oct08 Oct09 Apr08 Apr09 Apr10 Jan08 Jan09 Jan10 Jul08 Jul09 Jul10 3.5 0 2Jan08 2Jan09 2Jan10 Source: INEI. Source: BCRP. 13. In 2010, as the economic recovery continues to gather pace, the authorities have started to withdraw the expansionary stance of fiscal and monetary policies. Economic activity has recovered since 2009, with growth reaching 8.2 percent in the first semester of 2010 (Figure 1). Growth was fuelled by domestic demand ­ which increased by 11.1 percent. Private investment has picked up sharply, increasing by 28 percent in the first semester of 2010. In response, the authorities announced in May 2010 a series of measures aimed at controlling the pace of public spending. This new measure more than fully reverses the loosening of the constraint on growth of public expenditure that had been temporarily set for 2010 at 8 percent growth in real terms. In addition new measures have been put in place to limit the increase in public investment which had grown by 26 percent in real terms in 2009. With regard to monetary policy, the central bank raised the reference rate, which had been held at 1.25 percent through April 2010 at each of its meetings since, bringing it to 3 percent as of September 2010 (see Figure 3). In parallel, the central bank has stepped back the extent of intervention in the foreign exchange market on the back of strong capital inflows. In the three months to end-August 2010 net international reserves increased by US$5.6 billion (Figure 4). 14. Fast economic growth in recent years has been accompanied by poverty reduction and job creation. Between 2005 and 2009 poverty fell from 48.7 percent to 34.8 percent, while extreme poverty was reduced from 17.4 percent to 11.5 percent (Table 4). The continued reduction in poverty is remarkable given that 2008 saw high food price inflation and that economic growth slowed down in 2009 as a result of the global economic crisis. The reduction of poverty in 2009 was concentrated mainly in urban areas, especially in Lima, and took place at a lower rate than in previous years. Poverty incidence remains highly unequal in geographic terms, with some regions posting poverty rates above 60 percent. 12 Table 2: Poverty rates, 2004-2009 (percent) Total poverty Extreme poverty 09-04 09-04 2004 2005 2006 2007 2008 2009 (ppts) 2004 2005 2006 2007 2008 2009 (ppts) National 48.6 48.7 44.5 39.3 36.2 34.8 -13.8 17.1 17.4 16.1 13.7 12.6 11.5 -5.6 Coast urban 37.1 32.3 29.9 25.1 23.4 21.4 -15.7 5.6 4.0 3.0 2.1 2.4 2.3 -3.3 Coast rural 51.2 50.0 49.0 38.1 34.8 40.6 -10.6 13.8 13.4 14.4 10.5 7.9 9.2 -4.6 Sierra urban 44.8 44.4 40.2 36.3 33.5 31.3 -13.5 13.6 11.6 10.3 8.5 9.2 6.8 -6.8 Sierra rural 75.8 77.3 76.5 73.2 68.8 65.6 -10.2 44.0 46.6 46.5 40.8 37.4 33.2 -10.8 Selva urban 50.4 53.9 49.9 40.3 31.3 32.5 -17.9 18.7 22.5 18.1 11.0 7.2 8.8 -9.9 Selva rural 63.8 65.6 62.3 55.3 49.1 57.4 -6.4 30.4 28.0 24.6 23.4 20.7 23.8 -6.6 Metropolitan Lima 30.9 32.6 24.2 18.5 17.7 14.1 -16.8 1.3 2.0 0.9 0.5 0.7 0.2 -1.1 Source: Instituto Nacional de Estadística e Informática (INEI). MACROECONOMIC OUTLOOK AND DEBT SUSTAINABILITY 15. While uncertainty about the near-term outlook remains elevated due to risks to the global economy, prospects for Peru continue to be positive provided that a sound policy framework is kept in place. Following a rapid pick-up in economic activity from the end of 2009 driven by both domestic demand and a recovery of commodity prices, the most likely scenario for 2010 is one of substantial recovery. Growth is expected to increase to around 8 percent in 2010 and to hover around 5.5 percent over the medium term. Downside risks to this baseline scenario include a weaker than expected recovery of the world economy which could soften both demand and prices for key commodities of the Peruvian export sector such as copper, gold, and zinc. A weaker than anticipated global economy could also affect domestic confidence and investment, as it did in 2009, slowing down the rebound in domestic demand which is once again expected to be the key driver for overall growth in Peru. 16. Peru's debt servicing indicators show a robust position, helped by an active liability management policy. Peru's total public debt in 2009 stood at US$34 billion, or 26.4 percent of GDP. Net international reserves closed 2009 at US$33 billion, about 15 months of imports of goods and services and near 200 percent of foreign currency deposits. The country's debt service profile and its currency and interest risks exposures continued to improve in 2009 as a result of active debt management policies. In particular, in the third quarter of 2009, the authorities pre- paid debt with the Paris Club for US$952 million while in May 2010 the authorities engaged in a pre-purchase and exchange of bonds totaling US$1.8 billion that resulted in a lengthening of maturities and a rebalancing towards domestic-currency denominated debt. Finally, in parallel with withdrawing the fiscal and monetary stimulus, Peru returned to the international bond markets, raising US$1,261 million in April 2010. 17. A debt sustainability analysis suggests that the public debt position is resilient to a range of simulated negative shocks. The Article IV consultation completed by the IMF in April 2010 included an updated debt sustainability analysis. The results of such debt sustainability analysis, which was conducted under similar assumptions as those reflected in the macroeconomic outlook provided in Table 1, confirm the results of earlier analyses provided in previous Program Appraisal Documents for this series of DPLs. In particular, public debt is expected to decline under the baseline scenario to below 20 percent of GDP by 2015. Public external debt is 13 expected to decline to below 11 percent of GDP by 2015. A series of sensitivity analyses based on historical standard deviations of key variables suggest that the public debt profile is unlikely to differ significantly from the baseline and support the assessment that public debt sustainability is not a major concern in the medium term. 18. Notwithstanding the uncertainty about the near-term outlook, Peru's macroeconomic policies are deemed adequate for the proposed loan. Fiscal policy remains prudent as the expansion of Government expenditure in 2009 and 2010 does not threaten fiscal sustainability. In line with the temporary nature of the economic stimulus plan, the Government has already started to take actions to curb the expansionary stance of fiscal policy, which is particularly noteworthy in light of the upcoming presidential elections to take place in April 2011. Similarly, monetary and exchange rate policy is also supportive of macroeconomic and financial stability. III. THE GOVERNMENT'S PROGRAM AND PARTICIPATORY PROCESSES 19. The Government's twin strategic objectives are to accelerate economic growth; and to accelerate poverty reduction and improve social outcomes. On the economy, as outlined in Section II, the Government aims to maintain macro-economic stability, increase the institutional capacity of the state and strengthen Peru's physical and social infrastructure. It is promoting private sector-led growth, strengthening regulation, facilitating integration to the global economy and took advantage of favorable external conditions to greatly reduce vulnerability to domestic and external shocks. This proved prescient, in the light of recent global events. It is now concentrating on phasing out the counter cyclical policies which were successfully used in 2009- 2010 to minimize the impact of the global downturn in Peru. 20. In the social sectors, the Government is promoting human development, social justice and the reduction of extreme poverty. Since 1990, Peru has expanded the coverage of basic social programs, but the quality of services remained poor- especially in rural areas. As detailed in paragraph 3, above, although Peru is a middle-income country, with high service coverage, developmental outcomes, such as children's learning attainment and health and nutritional states, have remained disappointing. For these reasons, the issues of service quality, results and accountability occupy center-stage in the government's economic development and poverty reduction strategy. 21. There are many causes of poor quality, but a critical factor preventing change is the absence of clearly defined standards and results frameworks, across the social sectors. To address this issue, the Government's Social Sector Reform Program1 is working to reorganize programs and establish robust accountability frameworks, embedded in the performance budgeting system. 22. In this context, the Peru REACT DPL has supported actions in primary education, health and nutrition to define standards and set goals for the outcomes which families should expect for their children. This is coupled with the development of monitoring systems providing up-to-date information on the performance of health posts and schools and individualized data for parents 1 Decreto Supremo 029-2007-PCM, dated 30, March 2007. 14 on the health, nutrition and learning status of their children. The operation also supports complementary actions to strengthen the links between budgeting and outcomes (through the roll-out of performance budgeting in the social sectors), to improve service quality, reduce exclusion, improve targeting and increase the participation of poor communities in budgeting and program monitoring. In terms of development outcomes, the policies supported by the REACT DPL series are expected to lead to improved second grade literacy (especially in rural schools); increased access to institutional birth services; and increased coverage of individualized growth monitoring and counseling services for children under 24 months of age in areas with a high incidence of chronic malnutrition (stunting), leading in turn to improved nutrition outcomes. 23. The strategy adopted by the GoP was described in detail in the PD for REACT I (Report 39673-PE) which was approved by the Board in June, 2007 and was updated in the PD for REACT II (Report 47104-PE) which was approved by the Board in April, 2009. The strategy includes actions and goals in four broad policy areas specific to the REACT DPL: the strengthening of governance, equity and accountability in social programs; education; health; and nutrition. The following sections detail the development of the Government's strategies over the subsequent 18 months (to September 2010), commenting on the advances achieved, explaining the changes made to strengthen the strategy and documenting the outcomes which are observable at this point. Governance, equity and accountability in the social sectors 24. Fusion, integration and articulation of social programs. Between 2006-2008, the Government implemented a program of integration and rationalization of social programs which aimed to fuse programs which overlap on both objectives and beneficiaries; to integrate those which pursue similar objectives in different target groups of the population; and to improve coordination between programs which pursue related, but different, objectives. The goal is to improve coverage, reduce leakage of resources to unintended beneficiaries, reduce administrative costs, improve transparency and optimize impacts. 25. In 2008, the Government requested the Bank to carry out a mid-term evaluation of the social program reform process. The evaluation recommended that in the next phase, the emphasis should be placed on improving the targeting and effectiveness of the major programs, which account for over 80 percent of social spending (outside health and education). These include the "Juntos" CCT program, the social fund (FONCODES), the new investment fund, FONIPREL, the workfare program (Construyendo Peru, previously A Trabajar Urbano), and the major food programs (Vaso de Leche and PRONAA). It also recommended that the reform process should engage sub-national governments by using budgetary incentives, linked to the performance budget system, to encourage improved targeting and effectiveness in programs run by local government, such as the Vaso de Leche food program. 26. Based on these findings, the social program reform process in 2009-2010 has focused on improving the performance of the major programs with a focus on results and targeting. This has included continuing work to strengthen the urban household targeting system, SISFOH; and the establishment of a birth certificate data-base and identity campaigns for children under-five years of age. These policies are expected to improve targeting by reducing leakage of resources to unintended beneficiaries and also by tackling exclusion errors (see next section). 15 27. Results orientation has been promoted in social programs by requiring administrators to link new funding requests to a logic model that defines goals and specifies indicators to monitor results. This facilitates cost-effectiveness evaluations of social expenditure and has guided the articulation of institutions and goals under the strategic programs of the Results Based Budget (RBB) system, which was adopted by the MEF in 2008. The RBB's strategic plan for nutrition, PAN (Articulated Nutrition Plan), constitutes an example of this programmatic articulation through RBB. Several agencies contribute to achieving the PAN's goals by strengthening supply of services (SIS and MINSA) and by incentivizing demand for nutrition services (Juntos). 28. Targeting of social programs. The Government has taken two related initiatives on targeting. A household targeting system, the Sistema de Focalización de Hogares SISFOH, was created in the 30 major cities. SISFOH uses a proxy means test approach to identify households in the bottom two quintiles of the income distribution, as a first filter to help program administrators avoid leakage to non-poor beneficiaries. SISFOH now covers 376 districts and 1.8 million households (almost 7 million individuals). The register includes the unique personal identity number (Código Unico de Identificacion, CUI) which appears in official identification documents such as DNI and in birth certificates (see below). SISFOH's effort is complemented in rural areas by the targeting system implemented by Juntos, Peru's CCT program, which transfers NS/100 per month to beneficiary households and has adapted the targeting algorithm to the rural context. Its targeting system involves three stages: (i) geographic targeting, which has identified 638 priority districts (out of a total of 1303) in the poorest 14 departments of Peru, based on a selection criteria that combines basic needs, and estimates of the poverty gap, extreme poverty and malnutrition indexes, and also takes into account the percentage of town centers affected by political violence; (ii) household targeting, which relies on a proxy means test administered by INEI; and (iii) community validation. As of December 2009, there were 450,110 beneficiaries on the Juntos program. 29. The Government has also advanced with work to determine which households are actually covered by existing programs, to provide a basis for correcting errors of inclusion and exclusion. The creation of a Beneficiary Register (Registro Unico de Beneficiarios, RUB) for Vaso de Leche has been undertaken by the statistical institute, INEI, under contract to MEF. Vaso de Leche has an estimated three million beneficiaries nationwide. The register includes information on the identity number (DNI), the name and address of the claimant and the number of people in the household. By the end of 2008, INEI had completed a register of the 1.6 million beneficiaries in the 30 main cities of Peru. 30. SISFOH's current strategy integrates these two initiatives in a 3-stage plan to be implemented between 2010 and 2012. The first stage focuses on strengthening the beneficiary register, the second on improving the quality and use of household census information, and the third on identifying poor beneficiaries using an improved household database. Over the last year SISFOH has focused on improving the registry of beneficiaries, in alliance with MINDES' RUB. 31. The implementation of targeting reforms is being supported by a new Municipal Incentive Plan (PI for its Spanish acronym) was established in January 2010 through DS Nº 002- 2010-EF and is intended to become a permanent mechanism for resource allocation to local 16 governments. In 2010, S/.700 million (US$250 millions) have been allocated through the PIM (this represents about 11 percent of all budgeted transfers to municipal governments and about 4 percent of total budgeted municipal expenditure for 2010). The PI creates incentives for the use of SISFOH's algorithm in targeting beneficiaries of the two major social protection programs managed by local governments, Vaso de Leche (VdL) and Comedores Populares (CP). It also provides incentives to update VdL and CP beneficiary registries using the RUB. Every six months, SISFOH verifies municipalities' compliance, based on an updated RUB, remitted through the SIAF financial reporting system, and clears PI-related transfers accordingly. In September 2010 the PI was further enhanced with the inclusion of additional incentives which include the processing of statistical information about birth certification. 32. SISFOH has also advanced in improving household information by crossing databases and establishing local targeting units. Databases are cross-checked with the Government's payroll and pension data; credit rating information from the Bank and Insurance Oversight Agency (SBS- for its Spanish acronym); tax data from SUNAT (Superintendencia Nacional de Administracion Tributaria) and with the affiliation registers of the contributive insurance system (EsSalud). SISFOH has established 237 local units in all regions of Peru (see Figure 5). Local officers' main task is to update local household information and send it to SISFOH central targeting office in Lima. 33. In 2011, MEF has budgeted funds for updating SISFOH's to prepare for its adoption as the common targeting system for urban social programs, including SIS. There is also growing interest in connecting the SISFOH database with RENIEC's birth certificate database, which would allow it to automatically update household information to support better targeting of social programs focused on children under 5 year old (see next section). Figure 5: Number of SISFOH's local targeting units by region, 2010 Source: http://sisfoh.mef.gob.pe. 34. Identity. The lack of identification documents underlies the continuing socio-economic exclusion of many Peruvians. Citizen identification is central to ensuring access to social 17 programs. Without good data on the distribution of the population, it is difficult to identify coverage gaps and plan the expansion of programs. And without legal status as a citizen, it is difficult for individuals to exercise their rights to program access. As was detailed in the PD for REACT 1, at the start of the DPL series, almost 10 percent of the adult population was undocumented: an estimated 1.5 million adults had no DNI identity number. The undocumented population lives largely in rural and urban-marginal areas. The latter include immigrants from the Sierra and the Amazon who fled the armed conflict or to seek economic opportunities. 35. The Government has made citizens' identity documentation a high priority. The "National Plan for the Restitution of Undocumented Persons" was incorporated into the 2007 Public Budget Law (PBL). Identity was included as one of the strategic programs to be monitored under the 2008 and 2009 Performance Budgets, and for the first time, general taxation funds have been assigned to RENIEC to implement its programs. Among the identity-related goals being monitored are: the number of birth certificates issued at the national and regional levels; and the number of adult and infant DNIs issued national and regionally. 36. Since 2006, all birth certificates issued by RENIEC or by the Municipal Civil Registers (OREC for its Spanish acronym) have included a unique personal identification number (CUI), which will remain with the individuals through their lives and will be incorporated into their National Identification Document (DNI). RENIEC modified the registry books so that all new birth certificates include a CUI and the barcode associated with it. All birth certificates issued for people under 18 years of age now include a CUI, including those issued more than 60 days after their birth date (so-called "extemporary" registers). Coverage of birth certificates in Peru for the present generation of children is high (reaching 98 percent of children aged five according to the 2007 Census), so if every child with a birth certificate has a CUI, the basis will be established for near universal coverage of children with birth certificate information linked to a unique personal identification number. Since birth certificates (unlike DNIs) are free, there is no economic exclusion factor.2 37. With RBB support, RENIEC has advanced in the issuance of personal identification documents (DNI) to children. In June 2010 the GOP issued an Urgency Decree (DU No. 044- 2010) assigning 20 million Soles to RENIEC to attain a significant reduction of the gap during FY 2010/11. However, it will be very difficult to expand DNI emission to young children sufficiently to establish universal cover. Based on 2007 census data and RENIEC's 2007/2009 administrative data, it is estimated that only 8 percent of children under 2 years of age have a DNI (see Box 1). 38. A more sustainable and lower-cost alternative for systematizing personal identification data for young children is the promotion of timely birth registration and the development of a central register of data in the birth certificates issued by OREC and RENIEC's offices and auxiliary agencies. The GOP is implementing policies to: (i) ensure that birth certificates are issued in a timely way to all children (at present 8 percent of children do not get their certificates 2 However, a charge is levied when births are registered late (ex-temporal). For rural households the travel costs they must incur to process birth certificates in register offices are also an important obstacle. These can increase considerably when the families migrate, because there is no national database that would allow other registries or RENIEC to issue the document, so they must return to the original registry office. 18 until they start school, so they may miss out on early childhood programs); (ii) establish a national electronic register of all the birth certificates issued, including the CUI; and (iii) ensure that social programs have access to information contained in the electronic register. Given the high coverage of birth certificates and the incorporation of the CUI in the certificates, this is a viable strategy to reduce the exclusion errors of programs focused on young children as it will facilitate beneficiary identification. 39. A World Bank-supported study of coverage and exclusion from birth certification (W.Reuben and R. Cuenca, 2009) found that the lack of documentation is an obstacle to accessing jobs and social services. It highlighted problems of access to registry offices as a key factor in delaying the emission of certificates. It recommended: focusing identification campaigns in populations with pockets of undocumented children (which were identified by the study); using health posts (which deal with babies in remote areas before other public offices) to ensure timely emission of birth certificates; and creating incentives to encourage the OREC to reach remote areas through outdoor registry campaigns. 19 Box 1: Systematizing identification data in birth certificates The lack of identification documents is a serious obstacle to accessing social programs and underlies the continuing socio-economic Figure A: Percentage of children under 2 with birth exclusion of many Peruvians. For this certification/DNI issued by ORECs or RENIE C, 2007 and reason, RENIEC ­ with RBB support - 2009, national average has promoted campaigns to reduce the 100 87.7 number of undocumented people in 90 81.4 Peru. 80 70 To date, child documentation 60 campaigns have focused on DNIs, 50 which can only be issued by RENIEC's 40 2007 2009: 6.3% increase own offices. However, RENIEC is 30 directly responsible for documenting 20 8.0 only 8 percent of children under two 10 1.7 years of age (Figure A). Registry 0 Offices at the municipalities (OREC- OREC 07 OREC 09 RENIEC 07 RENIEC 09 Oficina de Registro Civil) issue over 80 percent of birth certificates. Figure B: Information flow for birth certificates database A comprehensive birth certificate Municipal OREC Municipal OREC database that supports targeting of with PC/internet without No Municipal OREC RENIEC agencies PC/internet social programs, through SISFOH, thus connection connection needs to integrate information from OREC issues electronic Agency issues OREC issues BC OREC issues BC with both RENIEC and ORECs. As shown in birth certificates (BC) with CUI with CUI CUI electronic BC with CUI Figure B, different actors will need to OREC officer sends BC feed this database. OREC sends BC database to Municipal BC's hard copy is sent to municipal hard copy to municipality and Information enters RENIEC coord. points receives monthly system Budget Office (BO) payments for this service RENIEC agencies and those ORECs Coord. Points send that are located in relatively well- OB sends information via SIAF hard copy to local RENIEC Municipality sends hard RENIEC centralizes information copy to local RENIEC equipped municipalities (with access to PC and internet connection) can easily SIAF sends database RENIEC digitalizes RENIEC digitalizes BC RENIEC sends SISFOH BC information and information and information to find ways to share digitalized to RENIEC/ SISFOH sends it to SISFOH sends it to SISFOH information. However, the capacities of OREC offices vary. They are SISFOH receives information, process database and sends to social programs particularly low in Amazon "Selva" districts. ORECs located in poor municipalities and the community-run civil registry annexes which operate in many remote indigenous communities will need further support to generate information for this database. 40. RENIEC has begun centralizing data from new birth certificates into a single electronic database for consultation and use by social programs. In addition to making registry data available to social programs, this will allow families who have migrated to process copies of the document in any municipal register or RENIEC office. Physical copies of birth certificates remitted to RENIEC are scanned and the personal information contained in them is digitized. RENIEC is also exploring the possibility of having its regional and auxiliary offices directly issue digitalized certificates and remit the information electronically to the central register. This 20 computer-based interface will reduce birth registration errors. Such a system could be extended to ORECs in provincial capitals. Through the Executive Decree 183-2010-EF, the Municipal Incentive Plan for 2011 includes a specific goal linked to the regular and timely delivery of information from municipal registries to the database in RENIEC. This will ensure a sustained and regular flow of information of birth certificates that will make this database a reliable source for social targeting. 41. The next step will be to ensure social programs' access to the electronic database for their planning and targeting purposes. SISFOH is naturally placed to become the interface between the electronic register and the social programs. This will require: (i) an agreement with RENIEC, to ensure SISFOH's access to the register, and (ii) asking SISFOH to develop a simple platform to distribute the information to the social programs. 42. Transparency in the budget process. As was outlined in the PD of REACT 1, the GoP has progressively strengthened the legislative and institutional framework to make social spending more transparent, equitable and effective. Under the 2002 Law of Transparency and Access to Public Information, all 16 ministries implemented transparency portals with links to respond to citizen questions. The MEF established the transparency portal "Consulta Amigable" ("Friendly Consultation") of the Integrated Financial Administration System (Sistema Integrado de Administración Financiera - SIAF). This portal is among the most advanced in the LAC region (http://ofi.mef.gob.pe/ transparencia/default.aspx). It provides information on economic indicators and projections, summaries of the strategic plans of different sectors, central and regional government budgets and expenditures, financial statements of all public agencies, and information on internal and external public debt. Consulta Amigable has recently been enhanced to include a direct link to the RBB sub-system. 43. In 2008, a further important step was taken, when the GoP began publishing on Consulta Amigable the modified budget (Presupuesto Inicial Modificado, PIM), which is used by government agencies during budget execution. An additional advance in 2007 was the publication of PIM for local governments through the SIAF GL (Local Government) system. All 1,834 municipalities' budgetary information can now be found on Consulta Amigable. For civil society agencies working on budget monitoring, access to PIM has allowed users to compare executed spending with the actual budgetary assignments, increasing accountability of social spending. In a further development of this line of work, in 2009 MEF developed a new application focused on RBB, which includes the same features as Consulta Amigable and will be uploaded on the Ministry portal within the next months. A pending issue is to create a user- friendly manual on how to access and use the information in Consulta Amigable. 44. Performance budgeting. The Garcia administration is committed to linking the budget process to performance evaluation. A results-based budget (RBB) office has been created in MEF. The World Bank's Human Development team has been supporting the development of the performance budgeting system in education, health, nutrition, and citizen identification. Since 2008 the annual Budget Law has defined the RBB's Strategic Programs (PE for its Spanish acronym), setting expected outcomes and allocating resources, some of which come from pre- existing budget lines regrouped under the PE, others from budget reassignments, and some from incremental resources. The number of Strategic Programs and the proportion of resources 21 allocated to them have steadily increased. It is expected that in FY2011, the RBB will represent 21 percent of the total budget, excluding financial and provisional expenditures, and will include a total of 24 PEs, covering 44 budget holders (Table 3). The strategic programs related to the social sectors are: Nutrition; Maternal and Neonatal Health; Educational Attainment; and Identity. Table 3: PPR strategic programs 2009 2010 2011 o N of Strategic Programs (SP) 9 15 24 No of budget holders under SPs 33 40 44 % Total Budget in SP* 8.3% 13.2% 21.0% * Not including local governments, pensions and financial expenditure. Source: 2010 Budget Law, p. 81. 45. For each strategic program, the budgetary system has been analyzed to determine the relevant financing sources and service providers. These include: MINSA, SIS, MINEDU, the Presidency of the Council of Ministers, MIMDES, the National Health Institute, RENIEC, Housing, MTC and the Regional Governments. The PPR system identifies the corresponding budget line items and physical output targets have been established at national, regional and local level (for each budgetary executing unit or Unidad Ejecutora). A baseline has been established for the high-level outcomes of each program and progress on expenditures and intermediate physical outputs is monitored quarterly for each executing agency of 31 health and nutrition programs, including institutional births and nutrition consultations.3 46. During 2009/10 MEF has analyzed the relation of outputs to outcomes for the strategic programs in nutrition and education and has assigned additional resources for the attainment of goals. In nutrition, the focus is on improving the supply chain for nutrition services; and, in education, it is on increasing the number and effectiveness of support teachers. The Government has also intensified progress monitoring for these strategic programs. 47. PPR has promoted the use of the SIGA budget planning tool by regional health authorities (DIRESAs) in their nutrition and institutional birth programs, for analyzing the supply chain between budget executing units and health networks. In 2010 a SIGA analysis justified budget increments to PAN, which increased the per capita (child) budget for CRED and vaccination by 331 percent and 150 percent, respectively, to fund CRED services in remote locations (Sierra and Selva), improved awareness campaigns and vaccinations against respiratory and diarrheic diseases. 48. In 2010, for the first time, PAN's additional budget for regional governments (12 percent of total PAN budget) was assigned based on malnutrition indicators, with bigger increases for regions with higher rates of chronic malnutrition. As shown in Figure 6, the previous allocation of the nutrition budget did not correspond well with the geographic distribution of chronic malnutrition.4 3 See http://www.mef.gob.pe/DNPP/PpR/seguimiento.php and http://www.mef.gob.pe/DNPP/PpR/ seguimiento.php. 4 In 2009, MEF received a grant from the European Union to strengthen PAN in 3 prioritized regions (Apurimac, Ayacucho and Huancavelica ­ all among those showing higher chronic malnutrition indicators, see Figure 6). The grant, called EUROPAN, will fund additional PAN products such as CRED and vaccinations. Regions receiving 22 49. In education, a detailed cost analysis for the support teachers program led to a budget increase. In 2010, regional governments received almost S/.111 million to fund this program - five times more than in 2009. This was justified by evidence of the effectiveness of support teachers in improving students' performance (e.g. shown by projects such as "Aprendes" in San Martin). This expanded budget aimed to finance support teachers for approximately 26,700 teachers in 6,327 education institutions (about half in pre-school and half in primary schools), together with the supplies and expenses budget needed to make them effective. Figure 6: Chronic malnutrition and allocated budget Chronic malnutrition PIA 2009: 1,052 M PIA 2010: 1,594 M (%) Increase 51% Targeted allocation Rate % Budget per child under 36 months Source: 2010 Budget Law, p. 103. 50. The 2010 budget law reports positive trends in the high-level outcomes supported by these efforts. After many years without significant improvement, between 2007 and 2009, chronic malnutrition in children under five years of age was reduced by 3.6 percentage points and the proportion of second grade students with sufficient reading comprehension and math skills increased by 1 and 2 percentage points, respectably. Although data limitations mean that these advances cannot be directly attributed to RBB, they have strengthened positive perceptions of the strategy. This contributed to RBB's institutionalization (as the Thematic Directorate) when the MEF's Budget Directorate (DGPP) was restructured in 2010. 51. DGPP is planning the incorporation of local governments within the RBB system in 2011. It is also working on improving SIGA, to better inform budget allocations by strengthening physical output targets for institutional births and neo natal health services and taking account of staff time requirements in cost calculations and budget allocations. these funds have signed results agreements that include detailed processes indicators to monitor the supply chain for nutrition services. 23 52. Areas for future development and improvement of performance budgeting in Peru include: (i) Strengthening the high-level strategic planning process and reinforcing its links to the macro-budget allocation decisions that are relevant to financing strategic goals; (ii) Enhancing the dialogue with the sectors and regions to improve the alignment of the actors with national and sector priorities; (iii) Moving progressively towards a more decentralized performance management system which creates "budget headroom", beyond the pre-committed line items, to allow budget holders (line ministries and regional governments) greater discretion in resource allocation, in pursuit of the targeted results, based on performance information; (iv) Relying less on ring-fencing; (v) Improving the quality and relevance of cost analysis and identifying critical bottlenecks in the process; and (vi) Aligning the incentives of service producing units (below the level of budget executing units) with the high-level goals of the system. 5 53. Participatory budget monitoring (PBM) at sub national level. In the context of decentralization to regions and municipalities, the Government has promoted participatory budgeting and monitoring at district and regional level. The aim is to make services more accountable to citizens and users, through the publication of timely and reliable budget information. Government-Civil Society forum such as the Mesa de Concertación para la Lucha Contra la Pobreza together with NGOs, research centers, church and social organizations have participated in this process. Participatory budgeting and monitoring is expected to lead to more equitable resource allocation, to raise public awareness of the use of funds and help limit corruption by increasing transparency throughout the budgeting process. 54. Participatory budgeting (PB) was initiated in Peru in 2002. The 2003 Budget Law mandated its application by all regional governments and provincial and district municipalities.6 After five years of implementation, PB in Peru involved the participation of approximately 180,000 accredited civil society and grass roots representatives in local and regional governments. 55. Responding to a MEF request, the World Bank conducted an evaluation of PB in 2009. The study found that PB has become an important space to harmonize local demands. In 2007, 36 percent of municipal districts' capital expenditure (approx S./1,174 Millions or US$391 Millions) was prioritized through PB. The study provides evidence that the PB favors pro-poor allocation of investments in social infrastructure and empowers grass-roots groups in budgeting decision-making. However, there is a need to strengthen the participation of the poorest sectors of the population, and the prioritization of investments needs to be informed by a more strategic and longer term vision. Box 2 summarizes main finding of this evaluation. 56. The Government's agenda has focused increasingly on promoting the participatory monitoring of budget implementation and results. Participatory monitoring aims to ensure that funds are used where and how they are supposed to be used and that the allocations produce the 5 For an in-depth analysis of performance budgeting in the health sector see: "Peru ­ Improving health outcomes by strengthening users' entitlements and reinforcing public sector management" (June 2010). This report was produced by the Bank's RECURSO AAA program. 6 Reuben, William and Belsky, Leah. 2006. Voice in the Accountability of Social Policy. In A New Social Contract for Peru. The World Bank, Washington DC. 24 hoped-for results. This is a critical challenge for the transparency and accountability dimension of the results-based management strategy. 57. MEF has broadened the scope of the local participatory budget monitoring strategy beyond municipal programs and made it an intrinsic part of the Performance Budgeting system. It is promoting a role for municipalities and civil society in tracking budget execution of national and regional government programs within their jurisdiction. The 2007 Budget Law announced a participatory monitoring (PM) process for the products and results prioritized by the Performance Budgeting system. 25 Box 2: Participatory budgeting and its relation to results-based - budgeting In 2009 the World Bank conducted an evaluation of Participatory Budgeting (PB), and its interface with results- based budgeting. The evaluation measures impacts at the district level by comparing treatment and comparison groups that were defined using a matching technique. The degree of PB implementation is measured as the share of capital expenditure allocated through PB, relative to total capital expenditure. The main findings were as follows: Regarding process: PB participants represent a wide range of organizations and population groups. Grass roots entities constitute the biggest representation, but there is relatively little participation by women. Participation quality is negatively affected by insufficient disclosure of information and poor knowledge of PB's implementation process. Civil society surveillance committees are weak, due to difficulties in getting access to relevant information, poor training, lack of resources to pay for transportation costs and members' high rotation. This facilitates PB's capture by traditionally powerful local groups. PB rarely informs the general budgeting process, due to delays in process implementation. Some projects prioritized by PB do not comply with SNIP requirements and are not in local development plans. Regarding results: Participatory budget agents are more likely Figure A: PB relative prioritization and need on basic to prioritize social infrastructure projects infrastructure (water, electricity and sanitation), according to the needs of the poor than are by quintile of income, 2007 the elected authorities' (figure i). When PB is strong, spending on basic infrastructure is higher and more is spent on water and electricity (figure ii). There is no statistical evidence that PB fragments public investment. PB has no effect on local tax collection; both treatment and comparison group showed low levels of tax collection. Figure B: Capital expenditure structure, by group Recommendations: Incorporate a results orientation in PB by defining national strategic targets, Treatment Group Comparison Group Basic establishing participatory monitoring and Basic Infrastructure Infrastructure creating municipal incentives linked to 40% 27% results. Create guidelines for regional PB. Regional governments could lead strategic budgeting processes within their jurisdictions. Pilot a two-year PB process so that Other Other 60% participatory agreements can inform 73% municipal budget allocations; Improve quality of participation by increasing training and organizing PB workshops that don't interfere with productive activities and using native languages; Strengthen Surveillance Committees by creating minimum standards for information disclosure and evaluating alternatives to fund their costs. Integrate the PB database (Aplicativo) with SIAF and SNIP. 58. The 2008 Budget Law called for MEF to implement PM pilots. Pilots are now ready to be implemented in the nutrition program in Amazonas, Cajamarca and Huanuco regions, with the support of a Japanese Grant (JSDF), supervised by the World Bank. Similar pilots will be implemented in Ayacucho, Huancavelica and Apurimac with the support of the EUROPAN. Stakeholder groups, called Social Monitors, run by local governments with the participation of 26 local civil society organizations, media, users and service providers, will oversee nutrition results and the quality of primary health and nutrition services at district level. Monitoring results will be linked to the award of non-monetary incentives7 such as public recognition or "branding" of the municipality. 59. In 2009, participation in public expenditure planning was further strengthened by the publication of a new format of the Participatory Budget Guidelines, called "Results based Participatory Budgeting". The new orientation of PB is emphasized at the identification and prioritization stage, which centers on results defined in RBB strategic plans. Education 60. As was detailed in the PDs for REACT 1 and REACT 2, Peru has near universal primary education coverage and is well above LAC regional averages in secondary and tertiary cover. The challenge now is to address quality issues, in the context of a rapid decentralization process in which regional governments are becoming important actors in the Peruvian education system. 61. Standards for educational quality improvement and the introduction of a universal standardized test. Education quality in Peru lags behind comparable countries in the region and worldwide, based on PISA 2000 (an international assessment of education quality as measured by the attainment of 15-year olds enrolled in school) and other similar assessments. Since 1996, the Education Ministry's Quality Measurement Unit (Unidad de Medición de la Calidad, UMC) has carried out rigorous sample survey-based national assessments in various subject areas and grades in primary and secondary schools. These assessments provided rich, comprehensive information on what subject matters were weakest and what student populations were most at risk, but school level data were not available. This made it hard for schools and teachers to know how they were doing and to target improvements. 62. To overcome this obstacle, in 2006 the government established a Universal Standard Testing (UST) system to provide feedback on outcomes to the schools, communities and parents on how well their children are doing and to provide the basis for setting school-specific performance goals. The first assessment was carried out in December 2006. It was organized on a tight timeline and was resisted by some sections of the teachers' union, SUTEP. Nevertheless, it reached around 55 percent of the target population of schools, and 44 percent of students. 63. In 2007, the exercise was repeated, with much better coverage, reaching 91 percent of schools and 80 percent of children. The tests were extended to cover bilingual schools and a mathematical test was added. In November 2008 and in December 2009 the tests were repeated for the third and fourth time, maintaining the same high level of coverage. The design of the questions ("items") in the tests was refined to improve the continuity in the difficulty level, which in turn will facilitate a more accurate analysis of findings.8 In 2008 and 2009, the UMC 7 The FY2010 National Budget will allocate an additional S/.1,300 million (approximately US$400 million) to be transferred to the most vulnerable municipalities against verifiable results related to nutrition outcomes and improvement in tax collection. This initiative provides an additional incentive for local governments to hold local primary health providers accountable for their performance. 8 Analysis of the raw reading scores is undertaken using standard psychometric transformations to produce Rasch scores. Students are then categorized by attainment level. For a detailed analysis of this process, see Kudó and Bazan, (2007). 27 built into the UST a rigorously controlled sub-sample to facilitate quality control of the findings. This permits a precise estimation of outcomes at national and regional level, weighted according to the population distribution.9 64. The UST classifies second grade students into three groups, based on two attainment levels. Children who reach Level 2 are those that perform at grade-level, that is, they have learned or acquired the reading or math skills that should be taught in second grade. Reaching attainment at Level 1 (L1) implies the child has developed basic skills but is not at grade-level, and so is falling behind. A student in the group below Level 1 (