Document of The World Bank Report No: ICR2306 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-78630 and 80580) ON A SERIES OF TWO LOANS IN THE AMOUNT OF EURO 200 MILLION (US$285.98 MILLION EQUIVALENT) TO THE REPUBLIC OF LATVIA FOR A SAFETY NET AND SOCIAL SECTOR REFORM PROGRAM June 25, 2012 Human Development Sector Unit Central Europe and the Baltic Countries Department Europe and Central Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective May 31, 2012 US$1.00 = 0.564 Latvian Lati (LVL) FISCAL YEAR January 1 – December 31 ABBREVIATIONS AND ACRONYMS ALMP Active labor market programs LFS Labor Force Survey CDS Credit Default Swap LRSA Last resort social assistance programs CPS Country Partnership Strategy LVL Latvian Lat CSB Central Statistical Bureau MOE Ministry of Economy DO Development Objective MoES Ministry of Education and Science DPL Development Policy Loan MOF Ministry of Finance EBRD European Bank for Reconstruction and MOH Ministry of Health Development EC European Commission MOW Ministry of Welfare ECA Europe and Central Asia Region MRDLG Ministry of Regional Development and Local Government ECSHD Europe and Central Asia Human OECD Organization for Economic Development Cooperation and Development ESA European System of Accounts PDO Project Development Objectives ESF European Social Fund PER Public Expenditure Review ESSNS Emergency Social Safety Net Strategy PFM Public Finance Management EU European Union PISA Program for International Student Assessment EU12 The group of the most recent EU PSIA Poverty and Social Impact Analysis members EUR Euro QAG Quality Assurance Group FBS Fee-Based Service QEA Quality at Entry GDP Gross Domestic Product QSA Quality of Supervision GMI Guaranteed Minimum Income SAO State Audit Office GP General Practitioner SDPL Special Development Policy Loan HBS Household Budget Survey SOE State-owned Enterprises IBRD International Bank for Reconstruction USD United States Dollar and Development ICR Implementation Completion and VAT Value Added Tax Results Report IMF International Monetary Fund WWS Workplaces With Stipends program (emergency public works program) Vice President: Philippe H. Le Houerou Country Director: Peter C. Harrold Sector Manager: Roberta V. Gatti Task Team Leader: Emily Sinnott ICR Team Leader: Emily Sinnott ICR Author: Paul Marie Michel Cahu FOR OFFICIAL USE ONLY REPUBLIC OF LATVIA SAFETY NET AND SOCIAL SECTOR REFORM PROGRAM A PROGRAM OF Two SPECIAL DEVELOPMENT POLICY LOANS Table of Contents DATA SHEET ................................................................................................................. i A. Basic Information ........................................................................................................ i B. Key Dates ................................................................................................................... ii C. Ratings Summary ....................................................................................................... ii D. Sector and Theme Codes........................................................................................... iii E. Bank Staff .................................................................................................................. iv F. Results Framework Analysis ..................................................................................... iv G. Ratings of Program Performance in ISRs ............................................................... xiii H. Restructuring ........................................................................................................... xiii Executive Summary ..................................................................................................... xiv 1. Program Context, Development Objectives and Design ............................................ 1 1.1 Context at Appraisal ............................................................................................. 1 1.2 Original Program Development Objectives and Key Indicators .......................... 4 1.3 Revised Program Development Objectives and Key Indicators, and Reasons/Justification................................................................................................... 6 1.4 Original Policy Areas Supported by the Program................................................. 6 1.5 Revised Policy Areas ............................................................................................ 7 1.6 Other significant changes ...................................................................................... 7 2. Key Factors Affecting Implementation and Outcomes .............................................. 7 2.1 Program Performance ........................................................................................... 7 2.2 Major Factors Affecting Implementation ............................................................. 8 2.3 Monitoring and Evaluation Design, Implementation and Utilization ................. 10 2.4 Expected Next Phase/Follow-up Operation ........................................................ 12 3. Assessment of Outcomes .......................................................................................... 13 3.1 Relevance of Objectives, Design and Implementation ....................................... 13 3.2 Achievement of Program Development Objectives ........................................... 16 3.3 Justification of Overall Outcome Rating ............................................................ 30 3.4 Overarching Themes, Other Outcomes and Impacts .......................................... 32 3.5 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops ... 33 4. Assessment of Risk to Development Outcome ......................................................... 33 5. Assessment of Bank and Borrower Performance ..................................................... 35 5.1 Bank Performance ............................................................................................... 35 5.2 Borrower Performance ........................................................................................ 36 6. Lessons Learned........................................................................................................ 38 7. Comments on Issues Raised by Borrower ................................................................ 40 Annex 1. Bank Lending and Implementation Support/Supervision Processes............. 41 Annex 2. Summary of Borrower’s ICR and Comments ............................................... 43 Annex 3. Policy Matrix ................................................................................................. 49 Annex 4. Status of SDPL1 and SDPL2 Prior Actions from the Legal Agreements ..... 53 Annex 5: Technical Assistance Program Accompanying SDPL .................................. 55 Annex 6: Map of Latvia ................................................................................................ 56 List of Tables Table 1: Key Outcome Indicators and Targets ................................................................... 5 Table 2: Overview of the SDPL Series Operations ............................................................ 7 Table 3: Performance Indicators for Development Objectives, by Sector ....................... 16 Table 4: Efficacy of Measures in Achieving the PDO on Crisis Mitigation .................... 17 Table 5: Descriptive Statistics for GMI Spending Per Needy Person, by Municipality... 20 Table 6: Efficacy of Measures in Achieving the PDO on Mitigation of Adverse Impact of Fiscal Consolidation.......................................................................................................... 21 Table 7: State Budget for ESSNS Measures Supported by the World Bank, 2010, 2011, and 2012 ............................................................................................................................ 22 Table 8: Government Health Budget, 2008-2011 ............................................................. 23 Table 9: Spending on Education, 2008-2011 .................................................................... 24 Table 10: Consolidated Public Sector Spending on Social Protection ............................. 25 Table 11: Health Care System Inputs ............................................................................... 26 Table 12: Efficacy of Measures in Achieving the PDO on Medium-term Improvement . 27 List of Figures Figure 1: Evolution of Needy people, GMI beneficiaries (left) and GMI spending (right) ........................................................................................................................................... 19 Figure 2: Local Government Spending on Social Assistance ........................................... 25 Figure 3: Evolution of Self-Reported Health Status by Gender, 2007-2010 .................... 27 Figure 4: Average Test Score Results in Latvia ............................................................... 29 DATA SHEET A. Basic Information Program 1 First Safety Net and Country Latvia Program Name Social Sector Reform Program Program ID P115732 L/C/TF Number(s) IBRD-78630 ICR Date 06/25/2012 ICR Type Core ICR REPUBLIC OF Lending Instrument DPL Borrower LATVIA Original Total EUR 100.00M EUR 100.00M Disbursed Amount Commitment (USD 143.90M)1 (USD 134.62M) Implementing Agencies: Ministry of Finance, Ministry of Welfare, Ministry of Education and Science, Ministry of Health, Ministry of Transportation, and the Ministry of Regional Development and Local Government.2 Cofinanciers and Other External Partners: The Special Development Policy Loan (SDPL) series was part of the international crisis-response package co-financed by European Commission (EC), the International Monetary Fund (IMF), the European Bank for Reconstruction and Development (EBRD), World Bank, and bilateral donors, namely the Nordic countries, the Czech Republic, Poland, and Estonia. Program 2 Second Safety Net and Country Latvia Program Name Social Sector Reform Program Program ID P121796 L/C/TF Number(s) IBRD-80580 ICR Date 06/25/2012 ICR Type Core ICR REPUBLIC OF Lending Instrument DPL Borrower LATVIA Original Total EUR 100.00M EUR 100.00M Disbursed Amount Commitment (USD 142.09M) (USD 130.73M) Implementing Agencies: same as program 1 Cofinanciers and Other External Partners: same as program 1 1 Both SDPL amounts were in Euros. The full amount for each of the two SDPLs was disbursed, i.e. Euro 100 million per operation. 2 The Ministry of Regional Development and Local Government was restructured into the Ministry of Environmental Protection and Regional Development on January 1, 2011. i B. Key Dates Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Revised / Actual Process Date Process Original Date Date(s) Concept Review: 07/27/2009 Effectiveness: 03/31/2010 04/02/2010 Appraisal: 01/19/2010 Restructuring(s): Approval: 03/04/2010 Mid-term Review: Closing: 12/31/2010 12/31/2010 Second Safety Net and Social Sector Reform Program - P121796 Revised / Actual Process Date Process Original Date Date(s) Concept Review: 02/22/2011 Effectiveness: 09/16/2011 07/25/2011 Appraisal: 04/18/2011 Restructuring(s): Approval: 05/26/2011 Mid-term Review: 11/30/2011 Closing: 12/31/2011 12/31/2011 C. Ratings Summary C.1 Performance Rating by ICR Overall Program Rating Outcomes Satisfactory Risk to Development Outcome Moderate Bank Performance Satisfactory Borrower Performance Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Overall Program Rating Bank Ratings Borrower Ratings Quality at Entry Satisfactory Government: Satisfactory Quality of Implementing Satisfactory Satisfactory Supervision: Agency/Agencies: Overall Bank Overall Borrower Satisfactory Satisfactory Performance Performance ii C.3 Quality at Entry and Implementation Performance Indicators Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Implementation QAG Assessments (if Indicators Rating: Performance any) Potential Problem Program at any time No Quality at Entry (QEA) None (Yes/No): Problem Program at Quality of Supervision No None any time (Yes/No): (QSA) DO rating before Satisfactory Closing/Inactive status Second Safety Net and Social Sector Reform Program - P121796 Implementation QAG Assessments Indicators Rating: Performance (if any) Potential Problem Quality at Entry Program at any time No None (QEA) (Yes/No): Problem Program at any Quality of No None time (Yes/No): Supervision (QSA) DO rating before Satisfactory Closing/Inactive status D. Sector and Theme Codes Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Original Actual Sector Code (as % of total Bank financing) General education sector 27 27 Health 28 28 Other social services 45 45 Theme Code (as % of total Bank financing) Education for all 30 30 Health system performance 10 10 Social safety nets 60 60 Second Safety Net and Social Sector Reform Program - P121796 Original Actual Sector Code (as % of total Bank financing) Compulsory health finance 12 12 Compulsory pension and unemployment insurance 12 12 iii General education sector 34 34 Health 21 21 Other social services 21 21 Theme Code (as % of total Bank financing) Education for all 25 25 Health system performance 25 25 Social safety nets 50 50 E. Bank Staff Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Positions At ICR At Approval Vice President: Philippe H. Le Houerou Philippe H. Le Houerou Country Director: Peter C. Harrold Theodore O. Ahlers (acting) Sector Manager: Roberta V. Gatti Jesko S. Hentschel Task Team Leader: Emily Sinnott Truman G. Packard ICR Team Leader: Emily Sinnott ICR Primary Author: Paul Marie Michel Cahu Second Safety Net and Social Sector Reform Program - P121796 Positions At ICR At Approval Vice President: Philippe H. Le Houerou Philippe H. Le Houerou Country Director: Peter C. Harrold Peter C. Harrold Sector Manager: Roberta V. Gatti Jesko S. Hentschel Task Team Leader: Emily Sinnott Emily Sinnott ICR Team Leader: Emily Sinnott ICR Primary Author: Paul Marie Michel Cahu F. Results Framework Analysis Program Development Objectives (from Program Document) The objectives of the DPL series were to: (i) protect vulnerable groups with emergency safety net support during the economic contraction; (ii) mitigate the social costs of fiscal consolidation; and (iii) ensure that structural reforms lay a foundation for medium-term improvements in the social sectors. Given the deep impact Latvia has suffered in the wake of the global crisis, and due to the emergency nature of this program, the first operation focused mainly on the first and second objectives. The second operation expanded this focus to the third objective, seeking to support the structural reform program that the Latvian Government put in place for the social sectors. iv Revised Program Development Objectives (as approved by original approving authority) The PDOs were not revised. (a) PDO Indicator(s) These indicators are proxies for the full Prior Actions, which are set out in the text of the ICR. Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Number of rural General Practitioners (GP) and Primary Health Care Indicator 1: Providers with an extra public health nurse. Value Approximately 0 out of 668 rural 289 of all GP (quantitative or half of the 668 practices practices. Qualitative) rural GP practices Date achieved 01/29/2010 12/31/2010 12/31/2010 PARTIALLY ACHIEVED. Final value represents 87% of the targeted Comments number of additional nurses that were due to have been provided. The (incl. % initial target was set for rural GPs, but the program was expanded to also achievement) provide an additional public health nurse to GPs in urban areas in 2010. Indicator 2: Number of subsidized out-patient visits. Value (quantitative or 27,909 184,199 547,641 Qualitative) Date achieved 12/31/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED. Indicators exceeded the target by far. achievement) Enrollment rate of five- and six-year-old children in free pre-primary Indicator 3: programs. 89% for the 5 Value 93% for the 5 94% for the 5 year- year-olds, 98% (quantitative or year-olds, 98% for olds, 98% for the 6 for the 6 year- Qualitative) the 6 year-olds. year-olds. olds. Date achieved 12/31/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED achievement) Share of students who spend more than 60 minutes travelling to and from Indicator 4: primary and secondary schools (one way). Value (quantitative or 1.5% 1.2% 0.5% Qualitative) Date achieved 09/30/2008 12/31/2010 12/31/2010 Comments ACHIEVED. (incl. % v achievement) Share of the state funding for stipends of students in higher education Indicator 5: that benefit academically eligible students on the basis of poverty and social considerations in addition to merit. Value (quantitative or 39% 45% 60% Qualitative) Date achieved 12/31/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED. achievement) Percentage of unemployment benefits recipients whose benefits are Indicator 6: extended by the reform. 0 – i.e. no one Value had benefits (quantitative or extended prior to 22% 25.7% Qualitative) the change in legislation Date achieved 12/31/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED. achievement) Share of spending on the GMI financed through the ESSNS co-financing Indicator 7: mechanism. Value (quantitative or 0% 50% 50% Qualitative) Date achieved 09/30/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED. Final values represent 99.9% of the target. achievement) Number of workplaces created under the emergency public works Indicator 8: program (WWS). Value (quantitative or 16,830 34,309 51,487 Qualitative) Date achieved 12/31/2009 12/31/2010 12/31/2010 Comments (incl. % ACHIEVED. achievement) vi Second Safety Net and Social Sector Reform Program - P121796 Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1: The financing for the safety net strategy is adequate in 2010. The actual state The actual state Value budget spending Spending should budget spending for (quantitative or for the safety net be maintained at 2011 was 89.5 Qualitative) was 89.3 million the level of 2010. million LVL. LVL. Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. The actual funding for 2011 represents 100% of the (incl. % funding executed in 2010. achievement) An independent audit of the implementation of the ESSNS will be Indicator 2: carried out. The audit was performed by the Value The audit will be State Account (quantitative or carried in 2011. Office and Qualitative) disclosed to the public. Date achieved 12/31/2011 12/31/2011 Comments ACHIEVED. The audit was performed and the report was released on (incl. % October 28, 2011. achievement) Number of rural General Practitioners with an extra public health nurse Indicator 3: and development of physician advisory phone service. 289 at the end of 465 for all GP Value 2010; no 500 GPs; practices3; (quantitative or physician Physician advisory Physician advisory Qualitative) advisory phone service running. service running. service Date achieved 12/31/2010 12/31/2011 12/31/2011 PARTIALLY ACHIEVED. Final value represents 93% of the targeted Comments number of additional nurses that were due to have been provided. The (incl. % initial target was set for rural GPs, but the program was expanded to also achievement) provide an additional public health nurse to GPs in urban areas in 2010. 3 The target was initially to provide additional public health nurses only for rural GPs, but with amendments to regulations No. 1046 on August 3, 2010, it became possible to also extend extra public health nurses also to urban GPs. In total, there were 1,376 rural and urban GPs at end-2011. vii Indicator 4: Number of subsidized out-patient visits. Value (quantitative or 547,641 420,000 858,866 Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. The number of subsidized visits is more than twice as large (incl. % as the target. achievement) Enrollment rate of five- and six-year-old children in free pre-primary Indicator 5: programs. 94% for the 5 Value 93% for the 5 year 94% for the 5 year year olds, 98% (quantitative or olds, 98% for the 6 olds, 97% for the 6 for the 6 year Qualitative) year olds. year olds. olds. Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. Enrollment rates have been maintained for five- and six- (incl. % years old. achievement) Share of children from closed schools that received transportation Indicator 6: assistance. Value (quantitative or 88% 77% 95.7% Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments (incl. % ACHIEVED. achievement) Share of the state funding for stipends of students in higher education Indicator 7: that benefit academically eligible students on the basis of poverty and social considerations in addition to merit. Value 60% at the end of (quantitative or 60% 50% the school year Qualitative) 2010-2011 Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED for the school year 2010-2011. The Constitutional Court (incl. % overruled the reform that was interrupted in the third quarter of 2011. achievement) Individual contributions to the funded pension pillar have risen from 2% Indicator 8: in 2009 to 6% from January 1, 2013. Value (quantitative or 2% 6% 2% Qualitative) Date achieved 12/31/2009 12/31/2011 12/31/2011 ACHIEVED. The Government enacted an amendment to the Law on the Comments State Funded Pensions which increases the contribution rate to the (incl. % funded pension pillar to 6% on January 1, 2013. But a risk remains that achievement) the implementation of this measure may be delayed as the Government stated in its Letter of Development Policy dated April 27, 2011, that the viii contribution rate will be restored to 6% only if fiscal circumstances allow it. Share of spending on the GMI financed through the ESSNS co-financing Indicator 9: mechanism. Value (quantitative or 49.97% 50% 50.1% Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments (incl. % ACHIEVED. achievement) Number of workplaces created under the emergency public works Indicator 10: Workplaces with Stipends (WWS) program. Value (quantitative or 51,487 71,794 82,305 Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments (incl. % ACHIEVED. achievement) (b) Intermediate Outcome Indicator(s) Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Share of persons living in households whose reported income qualify as Indicator 1 : needy, that have been certified as needy. Value (quantitative or 38.6% 50% 35.1% Qualitative) Date achieved 09/30/2008 09/30/2009 12/31/2010 NOT ACHIEVED. The certification of families or individual persons as needy is based on criteria that relate to income levels and other assets Comments (e.g. properties, savings, and securities). The target value for certification (incl. % achievement) was not achieved as the proportion of persons with reported income lower than the ―needy‖ income line and who also satisfy the asset criteria was lower than was expected. ix Number of patients receiving 100% reimbursement of medicinal Indicator 2: products or medicines charged under the emergency safety social safety net. Value (quantitative or 0 34,080 21,920 Qualitative) Date achieved 09/30/2009 12/31/2010 12/31/2010 PARTIALLY ACHIEVED. The value represents 65% of the target value. The number of patients receiving 100% reimbursement of medicinal Comments products or medicine charges was lower than targeted in 2010 because (incl. % achievement) when the target was set there were not sufficient data to estimate the demand for this benefit. The target was set assuming that the number of patients requiring the benefit would be the same as the proportion of the ―needy‖ people in the population. The estimate proved too high. Number of hospitals’ hotel patients receiving a subsidy under the Indicator 3: emergency social safety net. Value (quantitative or 0 27,200 7,208 Qualitative) Date achieved 09/30/2009 12/31/2010 12/31/2010 NOT ACHIEVED. The target was reduced for SDPL2 by mutual agreement between the World Bank and the Government, according to actual needs measured in 2010. The number of hospital patients receiving subsidized hotel stays under the emergency social safety net was much lower than targeted because when the target was set there was no data on the demand for the hospitals’ hotel service. This was because it is a service which is not paid Comments by the state health care budget and the patient use of this service had not (incl. % achievement) been recorded. The hospitals’ hotel patient service usage increased over 2010, increasing from 390 patients benefitting in the first quarter of 2010 to 3521 patients in the fourth quarter of 2010. The demand for the service grew as the related day hospital services became fully reimbursed under the emergency safety net from the third quarter of 2010 and as patients and doctors became more aware of the benefit. Number of subsidized home care visits under the emergency social Indicator 4: safety net. Value 115 (in the fourth (quantitative or 200,000 82,896 quarter of 2009) Qualitative) Date achieved 09/30/2009 12/31/2010 12/31/2010 PARTIALLY ACHIEVED. The value represents 42% of the target Comments value. The deployment of that measure took more time than anticipated (incl. % achievement) but the target was fully reached in 2011. x Percentage of students receiving assistance for school transportation in Indicator 5: municipalities where schools have closed Value (quantitative or 0 72% 88% Qualitative) Date achieved 09/30/2009 12/31/2010 12/31/2010 Comments ACHIEVED. (incl. % achievement) Indicator 6: Number of people receiving GMI. 67,365 recipients on average for the Value 4 fourth quarter of (quantitative or 16,515 100,000 20105 and 120,642 Qualitative) total recipients in 20106 Date achieved 09/30/2009 12/31/2010 12/31/2010 NOT ACHIEVED. This is because the group of people whose income and assets fell below the GMI eligibility threshold was smaller than Comments expected. The coverage of people eligible for GMI reached 95% (incl. % achievement) according to Household Budget Survey Data for 2010. Full coverage would have required the number of GMI recipients to be equal to 71,000 at any one point in time. Second Safety Net and Social Sector Reform Program - P121796 Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Percentage of persons living in households whose reported income Indicator 1: qualifies as needy, that have been certified as needy. Value (quantitative or 35.1% 50% 32.6% Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 NOT ACHIEVED. The certification of families or individual persons as needy is based on criteria that relate to income levels and other assets (e.g. properties, Comments savings, and securities). The target value for certification was not (incl. % achievement) achieved as the proportion of persons with reported income lower than the ―needy‖ income line and who also satisfy the asset criteria was lower than was expected. 4 The baseline indicator gives the monthly average number of GMI beneficiaries over January 1, 2009 to September 30, 2009. 5 The indicator gives the monthly average number of GMI beneficiaries for the last quarter of the year. 6 The indicator gives the cumulative number of GMI recipients during the year. xi Number of patients receiving 100% reimbursement of medicinal Indicator 2: products or medicines charged under the emergency social safety net. Value 30,267 in total in (quantitative or 21,920 22,000 (annual) 2011. Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. (incl. % achievement) Number of hospitals' hotel patients receiving a subsidy under the Indicator 3: emergency social safety net. Value 5,658 in total in (quantitative or 7,208 7,200 (annual) 2011. Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments PARTIALLY ACHIEVED. (incl. % achievement) Number of subsidized home care visits under the emergency social Indicator 4: safety net. Value (quantitative or 82,896 200,000 223,683 Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. (incl. % achievement) Share of students who spend more than 60 minutes travelling to and from Indicator 5: primary and secondary schools (one way). Value (quantitative or 0.5% 0.8% 0.2% Qualitative) Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments ACHIEVED. (incl. % achievement) Indicator 6: Number of people receiving GMI. 62,117 recipients 67,365 recipients on average for the on average for the fourth quarter of Value fourth quarter of 20119 and (quantitative or 20107 and 100,000 121,833 total Qualitative) 120,642 total recipients in recipients in 201110 20108 7 The indicator gives the monthly average number of GMI beneficiaries for the last quarter of the year. 8 The indicator gives the cumulative number of GMI recipients during the year. 9 The indicator gives the monthly average number of GMI beneficiaries for the last quarter of the year. 10 The indicator gives the cumulative number of GMI recipients during the year. xii Date achieved 12/31/2010 12/31/2011 12/31/2011 Comments NOT ACHIEVED due to the explanation given for 2010. (incl. % achievement) G. Ratings of Program Performance in ISRs Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program - P115732 Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 06/19/2010 Satisfactory Satisfactory 0.00 Second Safety Net and Social Sector Reform Program - P121796 Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 09/21/2011 Satisfactory Satisfactory 0.00 2 01/05/2012 Satisfactory Satisfactory 130.73 H. Restructuring Not applicable xiii Executive Summary After graduating from World Bank financing in 2007, Latvia exceptionally requested renewed access to lending resources in 2009 to deal with the impacts of a severe economic crisis. An important component of the World Bank’s support was a SDPL series of two operations that supported policy actions related to protecting the most vulnerable, within the framework of a structural reform program that underpinned the fiscal adjustment in education and health. The performance of these two operations is reviewed in this document. A critical area of focus was the implementation of an Emergency Social Safety Net Strategy (ESSNS) to finance essential services and benefits. The ESSNS was designed not only to protect the poorest and those that had lost their jobs from the crisis, but also to mitigate the adverse impacts of the cuts in government spending. Measures included the maintenance of resources for pre-primary education; transportation of children from closed schools to their new schools; coverage of health services and pharmaceuticals for poor households; improved services and access to GP and primary health care services; increased coverage and pay-out period for unemployment insurance; and increased coverage and adequacy of targeted social assistance benefits. The ESSNS also put in place an emergency public works program designed to provide support for individuals who had lost their job, but were not eligible for unemployment insurance benefits. The SDPL series also assisted the Government to ensure that fiscal adjustment in the social sectors was underpinned by a longer-term structural reform agenda. In education, the government embarked on a ―funds follow the student‖ financing reform in the education sector to increase efficiency and education quality in light of the shrinking school-age population. The switch to this form of financing creates incentives to increase the low teacher-pupil ratio in Latvia and decrease school infrastructure. In the health sector, reforms aimed at improving administrative capacity, while substituting expensive inpatient services to the extent possible through increased use of outpatient surgery and alternatives to overnight hospital stays. Primary health care services were protected. The overall rating of performance is satisfactory for both SDPL operations. This rating reflects the high relevance of the development objective, the design, and the progress with regard to implementation and the achievement of the program objective. All planned policy actions were implemented, although one measure was postponed by one year and another one was reversed by the Constitutional Court. The operation was well designed with a comprehensive set of prior policy actions covering the three social sectors: health, education, and social protection. All the PDO indicators reached or exceeded their set targets. However, many intermediary outcomes were not met, mostly because the set targets were overambitious, often due to a lack of available baseline data for new services being rolled out by the Government. Implementation by the Government, particularly in the area of health, was strong. Of particular note was the Government’s efforts to put in place their own monitoring and evaluation tools, including a regular and transparent reporting to the public on the implementation of the ESSNS, the carrying out of an audit by the State Audit Office of measures supported by the SDPL, and the impact evaluation carried out for the emergency public works WWS program. xiv 1. Program Context, Development Objectives and Design 1.1 Context at Appraisal Context at appraisal of SDPL1 1. After EU accession in 2004, the Latvian economy experienced rapid economic growth, driven partly by expansionary macroeconomic policies. Average growth stayed above ten percent over 2004-2007 and was almost entirely driven by a domestic demand boom. Large real wage increases, pro-cyclical fiscal policy, capital inflows, and a rapid credit expansion fueled domestic demand. However, labor shortages and out-migration to EU15 countries soon pushed up nominal wages, causing a rapid deterioration in external competitiveness. 2. The overheating pressures led to the development of large macroeconomic imbalances. The rapid growth of public expenditures—increasing by 90 percent in real terms between 2003 and 2008—resulted in a widening of the cyclically-adjusted fiscal deficit to 6.2 percent of GDP in 2008. At the same time, vibrant domestic demand brought consumer price inflation up to 18 percent in the first half of 2008, while competitiveness deteriorated rapidly. As a result, the external current account deficit widened from nine percent in mid-2004 to almost 25 percent of GDP in mid-2007. The fixed exchange rate peg with the Euro facilitated foreign borrowing by the financial sector. By the end of 2007, this had driven external debts to 128 percent of GDP. The inflow of capital fuelled rapid credit expansion and a real estate bubble. 3. This combination of credit and housing bubbles made the Latvian economy very vulnerable to the global financial crisis in late 2008. Markets became concerned over the sustainability of the peg, and the likelihood that contingent financial sector liabilities had to be absorbed by the Government. Latvia’s Eurobond spread widened rapidly and Latvia was downgraded by all three rating agencies. Private bank deposits fell by 19 percent between end- August 2008 and end-September 2009, precipitated by the run on mostly non-resident deposits in the Parex Bank. Deposit withdrawals and declining international reserves put substantial pressure on the exchange rate peg arrangement. As a result, Latvia approached the International Monetary Fund (IMF) and the European Commission (EC) for emergency financial support in November 2008. The financial turmoil resulted in a collapse of the real economy in late 2008; GDP fell by 4.2 percent in 2008 and 18.0 percent in 2009. 4. The Latvian government responded to the crisis with a reform program supported by the EC, international financial institutions, and bilateral donors. As the Government decided to maintain the fixed exchange rate peg with the Euro, a strong internal adjustment was needed to ensure long-term external stability. This adjustment was done through cuts in both public and private sector nominal wages and a substantial fiscal consolidation. The stabilization program required large external financing from international financial institutions, with a package of EUR 7.5 billion agreed in December 2008 and provided by the EC, the IMF, the European Bank for Reconstruction and Development (EBRD) and the World Bank, and 1 bilaterally through support from several Nordic and Central European countries. The stabilization program aimed to contain the growing public debt, while also enabling Latvia to meet the Maastricht criteria in future years. 11 The World Bank agreed to support Latvia’s stabilization program with EUR 400 million in funding. As Latvia is a graduated country, the exceptional access to IBRD resources was provided on non-standard lending terms, through a Special Development Policy Loan (SDPL). Apart from the Safety Net and Social Sector Reform Program DPL series, the World Bank provided a Financial Sector Development Policy Loan of EUR 200 million to support financial sector strengthening in 2009. 5. Expenditure cuts dominated the fiscal adjustment, accounting for more than three- quarters of the fiscal savings. Public spending was cut by 14 percent of GDP during 2009-2010 as compared to projected fiscal outcomes prior to the program. This allowed the fiscal deficit to fall from 9.8 percent of GDP in 2009 to 8.2 percent of GDP in 2010. The social sectors—namely education and health—contributed substantially to the consolidation effort: from 2008 to 2010 the health budget was cut by 25 percent and the education budget was cut by 48 percent. At the same time, social insurance spending continued to rise sharply (by 17 percent) in 2009 driven by increases in pensions, before being somewhat reduced by 3.5 percent in 2010.12 6. Despite the severity and speed of the adjustment, the Government took advantage of the fiscal consolidation to undertake structural reforms in the social sectors. In all social sectors, the main medium- and long-term challenge in Latvia is adapting to a decreasing and aging population. The population in Latvia has fallen by 5.6 percent since 2000 and is expected to fall by a further 19 percent by 2060.13 At the same time, the population is getting significantly older. This demographic shift is having and will continue to have significant fiscal and design implications for social sector policy. 7. In education, the challenge is to adapt the school infrastructure to the shrinking student population. Per capita financing was introduced in the education sector in the 2009 academic year and the network of vocational schools was consolidated under the management of municipal governments in order to adjust to the marked decline in the student population in the past decade. This led to the closing of 12 percent of primary and secondary schools. 8. In health care, the challenge is to adapt to an aging population and reduce the gaps in health outcomes in Latvia compared to pre-2004 EU member states. The difference in mortality rates for men from heart disease and strokes in Latvia are particularly striking. Therefore, during the crisis the Government worked not only to gain fiscal savings in the health sector, but also to re-orient spending toward out-patient and hospital day-care procedures. To do this, the sector began to correct the overprovision of hospital beds in order to free resources to 11 According to the Maastricht criteria, the fiscal deficit should be not more than three percent of GDP. 12 All expenditure changes in this paragraph are given in nominal terms. 13 UN Population Division, World Population Prospects: The 2010 Revision. 2 protect essential preventive and outpatient care provision. The result was a 19 percent reduction in hospital beds (amounting to 3,200 beds) from 2008 to 2010. Co-payments for services were increased to control demand, and excessive costs were reduced by shifting health care toward out-patient and hospital day-care procedures. At the same time, out-patient and GP services were relatively protected from budget cuts. 9. The social consequences of the economic contraction in Latvia were deep and most severe among the poorest households, making specific interventions targeted to the most vulnerable necessary. According to Household Budget Survey data, the percentage of people below the needy line increased from 13.9 percent in 2008 to 22.5 percent in 2009 and 24 percent in 2010 due to the GDP and employment contraction.14 15 Unemployment almost tripled during the crisis, rising sharply from seven percent on average over 2007-2008 to 20.1 percent in the last quarter of 2009. Moreover, the lower-skilled segment of the labor force suffered disproportionately from the deterioration in labor market conditions. The hardest hit groups were workers with a basic education or less who experienced a 40 percent employment reduction between the beginning of 2007 and early 2010. Employment also fell sharply among workers with upper secondary school only—by 21 percent from 2007 to 2010. In contrast, workers with higher education benefitted from an increase in employment during the same period. 10. To mitigate the effects of the crisis on the most vulnerable, while supporting measures to improve efficiency in the social sectors, the Government put in place an Emergency Social Safety Net Strategy (ESSNS) in September 2009. 16 This strategy was supported by international donors supporting Latvia’s crisis response, including the World Bank through the Special Development Policy Loan (SDPL) series. The EU-IMF program included an adjustor that allowed for additional fiscal spending on social safety nets, as defined under the World Bank SDPL series or Government of Latvia Strategy (of, for example, up to LVL 35 million per quarter, or about one percent of GDP per annum, in 2010). The World Bank assisted the Government in its preparation of the ESSNS and in assessing public expenditure consolidation and structural reform options through the preparation of a Public Expenditure Review (PER) in 2010. The World Bank then contributed financing for the social safety net and social sector reforms through a program of two SDPL operations, amounting to a total of EUR 200 million. 14 A household is defined as ―needy‖ if its per capita income (before WWS stipend) is less than 90 LVL per month. Latvia has no official poverty line but the LVL 90 per capita per month is known as the ―needy‖ line. 15 The number of people living below the needy line of 90 LVL a month estimated using Household Budget Survey data is higher than the official certified needy figure. From estimates using Household Budget Survey data, it appears that only around 39.1 percent of people with income levels that make them eligible for needy status were certified in 2011. 16 The Strategy was adopted by the Cabinet of Ministers on September 8, 2009. 3 Context at appraisal of SDPL2 11. Economic activity picked up in the second half of 2010. An expansion in exports, recovering domestic demand, lower Credit Default Swap (CDS) spreads, lower interest rates and extended maturities in the domestic debt market showed that stabilization was under way in 2010. This economic recovery was projected to continue in 2011. In the first quarter of 2011, real GDP projections (about 3.3 percent for 2011 and 4 percent for 2012 and 2013) were supported by improved consumer confidence, forecasts of higher corporate profitability, and the ongoing recovery of Latvia’s trading partners. Unemployment was also seen to be declining, albeit slowly. While unemployment was on average slowly falling, there had not been a recovery in employment for the population with basic education or less. Furthermore, the situation of the most vulnerable in society had been worsening, with the number of people defined as needy and the number of GMI participants steadily increasing throughout 2010. Real income per capita continued to decline: after falling by 17.7 percent in 2009, average real disposable income per capita decreased by another 2.1 percent in 2010. The drop in the incomes of the bottom quintile was larger than average. 12. Latvia moved toward meeting medium-term fiscal targets with the continued consolidation included in the 2011 budget. In addition to the 2011 budget adopted at the end of 2010, the Parliament passed a supplementary budget on April 2011. These budgets included fiscal consolidation measures totaling around LVL350 million or 2.6 percent of the 2010 GDP, to keep the fiscal deficit 17 below 4.5 percent of GDP in 2011, with the aim of meeting the Maastricht criterion of a fiscal deficit below three percent of GDP in 2012. The consolidation program included several revenue generating measures, accounting for 1.8 percent of GDP.18 It also included a decrease in expenditure through reduced staff and wage costs in the public administration and cuts in the health, education, defense, social programs, and railway sectors. Temporary measures to achieve these goals were also put in place, such as a diversion of second pillar pension contributions. The Government demonstrated its commitment to the continued implementation of the ESSNS by securing 0.6 percent of GDP to finance it. 1.2 Original Program Development Objectives and Key Indicators 13. The Program Development Objectives (PDO) of the SDPL series were to: (i) protect vulnerable groups with emergency safety net support during the economic contraction; (ii) mitigate the social costs of fiscal consolidation; and (iii) ensure that structural reforms lay a foundation for medium-term improvements in the social sectors. Given the deep impact of the global crisis on Latvia and due to the emergency nature of this program, the first operation focused mainly on the first and second objectives. The second operation expanded the focus to 17 As computed according to the European System of Accounts 95 (ESA95). 18 The revenue measures included increases in VAT rates, employee social security contributions, taxation of vehicles, real estate taxes, excise duties, and new fees. 4 the third objective, seeking to support the structural reform program that the Latvian Government had put in place for the social sectors. Table 1: Key Outcome Indicators and Targets Objective Indicator and target values for Measurement and target values SDPL1 for SDPL2 Satisfactory implementation of the The funding planned for 2011 Emergency Social Safety Net should be similar to the executed Strategy budget of 2010. Government’s implementation of Audit of Government’s ESSNS and structural reforms is performance in implementing independently assessed and publicly Emergency Social Safety Net available Strategy completed. Strengthening of primary care Number of (rural) GP with an extra Number of (rural and urban) GPs practices public health nurse should be 300 at with an extra public health nurse is the end of 2010. at 500 at the end of 2011. A physician advisory phone service should be developed. Increased financial protection for The number of subsidized out-patient The number of subsidized out- health care for needy households visits is at least 184,199 during 2010. patient visits is at least 420,000 during 2011. Adequate budget to pre-primary Enrollment of five- and six-year-old Enrollment of five- and six-year- education and child-development children in free pre-primary programs old children in free pre-primary programs is maintained at least at 89% for 5 programs is maintained at least at year olds, and 98% for 6 year olds 93% for 5 year olds, and 98% for 6 respectively. year olds respectively. Provision of transportation to The share of students who spend The percentage of students another school for children whose more than 60 minutes travelling to receiving assistance for school schools have been closed and from primary and secondary transportation in municipalities schools (one way) is kept below where schools have closed remains 1.2%. above 77% during 2011. Improved equity of subsidies for The share of the state funding for The share of the state funding for higher education stipends of students in higher stipends of students in higher education that benefit academically education that benefit to eligible students on the basis of academically eligible students on poverty and social considerations in the basis of poverty and social addition to merit is at least 45%. considerations in addition to merit is at least 50%. Extension of the coverage of The percentage of unemployment unemployment insurance benefits recipients whose benefits are extended by the reform remains above 22%. Contribution rate to the second Individual contributions to the pillar pension system raised funded pension pillar have risen from 2% in 2009 to 6% from January 1, 2013. Adequate financing of the cost of the The share of spending on the GMI The share of spending on the GMI GMI in the 2011 budget financed through the ESSNS co- financed through the ESSNS co- financing mechanism is 50%. financing mechanism is 50%. Strengthening of active employment In Q4 2010, 34,309 workplaces have In Q4 2011, 16,500 workplaces measures been created under the WWS public have been created under the WWS works program. public works program. 5 14. The monitoring and results framework included numerous indicators to follow the implementation of the SDPL series.19 The overall 11 Project Outcome Indicators presented in Table 1 are a representative selection of the full set of indicators and are used as proxies for the full Prior Actions. 1.3 Revised Program Development Objectives and Key Indicators, and Reasons/Justification 15. No revisions were made to the original PDO. Some changes affected the Key Indicators, such as the one-year delay in the increase of the funded pension pillar contribution rate (see section 2.2 for details). 1.4 Original Policy Areas Supported by the Program 16. The SDPL series supported a large social sector reform program, including policy measures in four main policy areas in line with the PDO and outcome indicators mentioned above, to: ï‚· Support the deployment of emergency social safety net measures to mitigate the impact of the fiscal adjustment and the economic contraction on households, by: (i) ensuring adequate emergency safety net financing for education and health services, and social protection programs; and (ii) increasing public awareness, transparency, and accountability of government measures in the social sectors; ï‚· Strengthen the financing of social protection by: (i) restoring sustainable funding of second pillar pensions through an increased contribution rate; (ii) lengthening eligibility for unemployment insurance; (iii) securing adequate financing of the GMI benefit from the State budget 20 ; and (iv) providing work for the unemployed not entitled to unemployment insurance in the emergency public works WWS program; ï‚· Improve the coverage, effectiveness, and efficiency of health care by: (i) introducing for the first time a safety net to ensure access to health care, including pharmaceuticals, for the needy; and (ii) strengthening primary care services and substituting for expensive and inappropriate inpatient stays by increasing use of outpatient surgery and alternatives to overnight hospital stays via changes in reimbursement policies. These measures indeed allowed closure of overnight hospitals, including in Riga; and ï‚· Maintain access to education, by (i) providing a State subsidy to maintain an adequate level of local government financing for preschool programs for five- and six-year-old children, (ii) providing transportation to another school for students whose schools have closed; and (iii) improving the targeting of higher education state support to students from lower income backgrounds. It is to be noted that these measures accompanied the implementation of per capita financing in general education. 19 These indicators were given in Annex 4 of the SDPL1 program document and in Annex 8 of the SDPL2 program document. 20 Additional funds for GMI allowed both improving the coverage and raising the level of benefits. 6 1.5 Revised Policy Areas 17. Not applicable 1.6 Other significant changes 18. Not applicable 2. Key Factors Affecting Implementation and Outcomes 2.1 Program Performance 19. The SDPL programmatic series of two operations followed the single-operation Financial Sector DPL approved on September 22, 2009 for EUR 200 million. Together, these three DPLs represented the World Bank’s contribution of EUR 400 million of the EUR 7.5 billion international crisis-response package of the IMF, EC, EBRD, Nordic and Central European countries. The World Bank team collaborated closely with the IMF and European Commission (EC) teams, and agreed on complementarities and mutually-supportive measures. 20. In the summer of 2007, the Government asked the World Bank to conduct a Public Expenditure Review (PER) for the social sectors. This assessment served as an analytical input for the Government’s ESSNS program, revisions to the 2009 budget, and formulation of the 2010 budget. It examined public investment and outcomes in the social sectors and public administration, and identified a number of critical structural reforms. In September 2010, the World Bank updated the 2007 PER and examined options for fiscal consolidation and structural reforms in health, education, social protection, and public administration. The results of the PER process were shared with the Government during numerous visits to Latvia from March to December 2010, with the aim of informing the budget formulation process. This technical analysis formed the basis for the policy actions supported under the second SDPL operation. Table 2: Overview of the SDPL Series Operations SDPL Status Amount Expected Release Date Actual Release Date Release (EUR) First SDPL DPL fully 100,000,000 June 16, 2010 September 27, 2010 Regular disbursed* Second SDPL DPL fully 100,000,000 December 20, 2011 December 20, 2011 Regular disbursed Note: (*) The loan was fully disbursed in Euros (100 million). As the Euro-USD exchange rate varied from approval to disbursement, the actual disbursed amount in USD was lower than initially expected. 21. In addition to the 2010 PER, the SDPL lending program was accompanied by a thorough technical assistance program covering the social sector and public administration reform (see Annex 5 for a description). The World Bank provided advisory support to the Government on: (i) 7 the design and implementation of the ESSNS, and (ii) monitoring and evaluation of the impact of ESSNS measures, including help to design the evaluation program for the WWS program. 22. To support the Government’s program, the SDPL series set out ten policy actions to be undertaken prior to the first loan and eight additional policy actions triggering the second loan. Table 2 provides an overview of the SDPL series. Annex 3 contains the full policy matrix as presented at the SDPL2 Board Presentation, with outcome indicators and goals specified. 23. All SDPL1 Prior Actions were met before the Board presentation in March 2010. The first SDPL was subsequently fully disbursed on September 27, 2010 and the loan closed on its original closing date of December 31, 2010. 24. All prior conditions for the second SDPL were met on time and the second SDPL was fully disbursed on December 20, 2011 as described in Annex 4. The second loan closed on its original closing date of December 31, 2011. 2.2 Major Factors Affecting Implementation Content of the program 25. By design, the SDPL series was ambitious enough to cover all social sectors, allowing the program to address the adverse impacts of either the crisis or the fiscal consolidation. Within each sector, the measures promoted by the SDPL series also covered several sub-sectors: (i) social assistance, activation policies, pensions and other social insurance in social protection, (ii) outpatient, inpatient and pharmaceuticals in the health sector and (iii) from pre-school to university in the education sector. Ownership of the reforms by the Government 26. The need for an emergency safety net was signaled by the World Bank team during its initial visit, based on a review of existing programs and the fast-developing realization that the severity of the economic crisis was fast outstripping anyone’s predictions. However, the program was owned from the start by the Government, which designed the ESSNS through an inter- ministerial task force chaired by the Ministry of Finance. Not all the measures undertaken by the ESSNS were supported by the SDPL series as the strategy itself was wider in coverage. Besides social protection, health, and education, it also encompassed measures in the housing and transportation sectors. Moreover, the Government disclosed extensive information to the public on the implementation of the ESSNS, in addition to the indicators agreed under the World Bank- supported program. This showed the Government’s commitment to the objective of assisting the poor and vulnerable at a time of severe economic stress. Furthermore, the extension of elements of the ESSNS beyond the end of the World Bank-supported program (see section 2.4) is testimony of the Latvian Government’s ownership of the reforms. 27. The SDPL series benefited from the continuity of the Government from program conception to implementation. Prime Minister Dombrovskis remained in power, although as a 8 head of different coalitions of parties, from March 2009 to date, despite parliamentary elections (in October 2010 and September 2011). This brought continuity in the Government’s commitment to the SDPL series. The ministries of Finance, Welfare, Health, and Education and Science did see their management change following each election; fortunately, these changes did not materially affect the performance of the reforms. Sound and timely background analysis and policy dialogue 28. As noted earlier, the program benefitted from timely and extensive social sector analyses and PERs available at the time of the crisis. This enabled the Government to put in place targeted and efficient measures in a short time period. The ESSNS was designed, approved, and put in place within six months, from June to November 2009. 29. The SDPL series also benefited from continuous policy dialogue between World Bank staff and the Government. This allowed for improvement of the content of the second SDPL, especially in the health sector, from what was envisioned during the preparation of the first loan, within a limited timeframe. Policy reversal 30. The implementation of the reform on scholarships in higher education was discontinued in the third quarter of 2011. An amendment to the Cabinet Regulation on scholarships was passed on June 2, 2009 (No. 511), adding social considerations, in addition to merit, as a criterion for awarding State support to students in higher education. This policy action was a prior condition to be met for the first SDPL. This measure had positive consequences as the share of eligible students receiving stipends based on social considerations exceeded expectations, reaching 60 percent during 2010 and the first semester of 2011, far above the baseline of 39 percent in 2009. However, this legislation was overruled by the Constitutional Court of Latvia on August 9, 2011. According to the ruling, stipends should be allocated first on the basis of academic achievement; social considerations only apply as a secondary criterion, to allocate stipends among students of equal academic achievement. The implementation of the reform was therefore halted in the second half of 2011 in compliance with the Constitutional Court ruling. Cooperation with other partners of Latvia 31. The commitment of the Government to pursue the reforms was helped by the support of the EC, the IMF, and bilateral donors. Coordination among lenders allowed the country to dedicate a substantial share of its budget to the ESSNS in a time of high fiscal pressures. The EU-IMF stabilization program strongly backed these temporary measures and included an adjustor to fiscal targets allowing for additional safety net spending if necessary. The Bank team provided economic expertise during the 2009, 2010, and 2011 budget cycles and through the PER in 2010 that supplemented the macroeconomic focus of the other lenders and of the government during the crisis. The Bank’s engagement was considered a necessity by these 9 partners on social sector expenditure issues that went well beyond the safety net supported by the Bank’s loans. 2.3 Monitoring and Evaluation Design, Implementation and Utilization 32. Monitoring and Evaluation (M&E) and implementation arrangements were continuous and contributed to reform implementation, the achievement of the development objectives, and the improvement of the ESSNS. Design 33. The indicators chosen to monitor the outcomes were adequate, covering all three social sectors and the three PDOs in detail. The fact that many intermediate indicators were not or only partially achieved, while the outcome indicators were achieved, suggests that they were not good indicators of interim progress towards the PDIs. Perhaps unique in this operation, and reflecting an effort by the team to depend on normal Borrower systems to the degree possible in implementing the agreements, the Bank team sought guidance from the State Audit Office of the Republic of Latvia (SAO) to specify the indicators and the PDOs in such a way that they would lend themselves to normal auditing procedures by the Government. Implementation 34. Procedures and adequate financing were incorporated into the ESSNS to allow close monitoring of its implementation and evaluation of its impact. The Government regularly monitored around 35 indicators pertaining to the ESSNS, which they then updated and published every quarter on the internet. 21 Baseline and updated data were provided by the respective specialized agencies for the pertinent functions and tracked according to the indicators and outcome measures. 35. Moreover, the SAO, which reports independently to Parliament and is a key institution for public sector accountability, evaluated the Government’s performance in implementing the ESSNS under SDPL1, with the auditor’s report being made available to the Government prior to the end of the first year of implementation of the program.22 This report was also used by the Bank team appraising SDPL2 in its discussions with each ministry and the Ministry of Finance, and the team reviewed with the SAO the results of the audit in detail. The fundamental problem found by the audit was that the legal process for the certification of needy people was not adequately followed by local authorities and that benefits were paid to non-eligible households. 21 Available in Latvian at http://www.lm.gov.lv/text/1843. The last update was made on March 3, 2012. 22 Actions of national and municipal authorities to ensure the fulfillment of the obligations under the Loan Agreement ―Safety Net and Social Sector Reform Program‖ between the Republic of Latvia and the World Bank and the World Bank Program Document, Audit Report 28.10.2010 No. 5.1-2-3/2010. Available at http://www.lrvk.gov.lv/upload/PB_soc_drosiba_28Oct2010_EN.pdf. 10 This shortcoming affected both the social assistance and health portions of the program; although the leakage was relatively small, the audit revealed simple ways public records could have been used by local authorities to verify eligibility. Overall, 24 recommendations were formulated by the SAO, consisting of three categories, namely: (i) changes to regulations/legal acts; (ii) improvements in monitoring and internal control systems; and (iii) recommendations regarding measures under the ESSNS that expired on December 31, 2011. As part of the normal audit process, each government agency was required to file a response to the audit and an action plan to correct the deficiencies within one year of the report, or in this case, October 2011. These action plans are monitored by the SAO. The audit of the ESSNS was widely disseminated by the SAO, including being made publicly-available on the internet. Action plans with the relevant ministries were actually agreed prior to publication, so the action plans by the ministries were in place prior to appraisal of the second loan. The audit made extensive use of country databases and information to cross-check the implementation of the ESSNS at the national and municipal levels against the stated policies and aims of the program. It used the data collected through the program to assess the targeting of social assistance and identify leakages. All necessary amendments to regulations were implemented. Regarding the system of controls, the Ministry of Welfare improved the system of checks and balances applied to GMI monitoring and provided training seminars to social workers of local authorities. The Ministry produced a report on the follow-up implementation of the SAO’s recommendations. However, no assessment has yet been undertaken by the SAO of the implementation of the program subsequent to October 2011. 36. The Government carried out an impact evaluation of the emergency public works WWS program with technical assistance from the World Bank. The evaluation was based on a unique household survey administered to 3,000 households (9,054 respondents, including 2,309 WWS participants and 1,016 persons from the waiting list) conducted between December 2010 and March 2011. An evaluation of the WWS intervention based on the survey data was done by the World Bank (see Ajwad, Azam, and Ferré, 2012) and also by local researchers Latvijas LauksaimniecÄ«bas un Meža ZinÄ?tņu AkadÄ“mija (Latvian Academy of Agricultural and Forestry Sciences, 2011)23. The World Bank evaluation found that the WWS self-targeting mechanism worked well, with most WWS beneficiaries belonging to the bottom 40 percent of the income/welfare distribution (poor and vulnerable), with the income quintile two highly represented. Similarly, the evaluation found that administration and implementation of the program were satisfactory, with payments being made accurately and on time. The WWS program had long waiting lists and large numbers of workers re-registered. This demonstrates 23 Latvijas LauksaimniecÄ«bas un Meža ZinÄ?tņu AkadÄ“mija (Latvian Academy of Agricultural and Forestry Sciences), 2011. AktÄ«vÄ? nodarbinÄ?tÄ«bas pasÄ?kuma 'ApmÄ?cÄ«ba darba iemaņu iegÅ«Å¡anai un uzturÄ“Å¡anai, ja darba devÄ“js ir paÅ¡valdÄ«ba' rezultÄ?tu novÄ“rtÄ“jums. Gala ziņojums. (Evaluation of the results of active employment measure: Training for acquiring and maintaining work skills when employer is a municipality. Final Report). http://www.nva.gov.lv/docs/17_4e1431090ba7a3.03485280.pdfhttp://www.nva.gov.lv/docs/17_4e1431090ba7a3.03 485280.pdf 11 that people valued the WWS program and also that the program was too small given the devastating impact of the crisis. In addition, most program beneficiaries had positive views about the program administration and implementation. The Government showed strong ownership of the impact evaluation and conducted a peer learning event with 10 other EU countries, Croatia, and Norway on April 26-27, 2012 to discuss the lessons from the emergency public works program and share the results of the evaluation. Utilization 37. The audit and its use by the Government represent best practice in the use of country systems to monitor and evaluate a World Bank-supported program, and to use the information to improve implementation and results. Given the complex nature of structural reforms and emergency assistance supported by the SDPL program, it would have been impossible for the World Bank team to conduct such a comprehensive and thorough fiduciary assessment of performance of the various systems involved down to the local government level. By coordinating with the SAO and involving that office at the design stage of the loan, both the Government of Latvia and the Bank benefited from having a timely, on-the-ground assessment of program implementation after the first year of the program. Having action plans in each ministry in place prior to the second year immeasurably improved confidence on all sides that program implementation would be improved in the second year. 2.4 Expected Next Phase/Follow-up Operation 38. The ESSNS was initially foreseen to be effective from October 1, 2009 (Cabinet of Ministers adopted the Strategy on September 8, 2009) until December 31, 2011. The continuation of some strategy measures into 2012 was discussed in the Cabinet of Ministers’ meetings on November 4 and 8, 2011. The decision was made to continue implementing the measures related to social assistance, health, and transportation. Consequently, only three measures supported by the SDPL series were stopped on December 2011: (i) the state financing for the transportation of students from closed to new schools, which was taken over by local governments; (ii) the emergency public works WWS program, which was replaced by a smaller program that built on the findings of the WWS evaluation; and (iii) the measure providing for social conditions in addition to merit as criteria for eligibility of stipends in higher education (overruled by the Constitutional Court). Further continuation of the Strategy measures in 2013 are to be discussed in July 2012 during the preparation process of the State Budget for 2013. 39. After the closing of the SDPL series, the Ministry of Welfare expressed interest in engaging the World Bank to carry out an evaluation of the employment and social protection situation in Latvia, especially with respect to the situation of the long-term unemployed and the inactive. The commitment of the Ministry is underlined by the fact that this activity is a Fee- Based Service (FBS). The study will analyze the unemployment, tax, and benefit systems to support the re-integration of the long-run unemployed into the labor market, with a focus on: (i) the evolution of the Latvian labor market emerging from the crisis; (ii) profiles of people with no 12 or limited labor-market attachment; (iii) expenditure and performance of welfare benefits and employment in Latvia; (iv) financial incentives of the tax and benefit system in Latvia; (v) the impact of employment policies; (vi) other countries’ experiences in linking social benefit programs to employment activation measures; and (vii) options for the design of income and work support measures to promote labor force participation, whilst providing adequate social protection. The Cabinet of Ministers met on April 3, 2012 and approved the FBS terms of reference and the Legal Agreement was signed on May 25, 2012. This work will assist the Government in motivating future reforms. It could also inform the 2013 budget planning. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Relevance of objectives 40. As Latvia graduated from World Bank financing in 2007, the last Country Partnership Strategy (CPS) expired in 2005 and was focused on accession to the EU, an objective that was reached in 2004. The SDPL series drew, however, from a series of successful projects with the World Bank undertaken in the social sectors prior to graduating. 24 Moreover, a PER was prepared by the World Bank in 2007 at the request of the Ministry of Finance and updated in 2010, helping to put in place a swift program to respond to the crisis. 41. The objectives of the SDPL were, and still are, highly relevant to Latvia’s social sector reform program (see paragraph 13 for a description of the PDOs). The three objectives were in line with the policy actions of the Government’s stabilization program signed on December 19, 2008 with the EU, the IMF, and bilateral and multilateral donors. 42. The SDPL series supported the protection of vulnerable groups with emergency safety net support during the economic contraction. Given the severity of the economic crisis, its deep social consequences on the labor force and the fact that low-skilled workers were the most affected, the interventions supported by the programmatic SDPL series to protect the vulnerable groups were appropriate and timely. The crisis induced a sharp increase in unemployment: the unemployment rate more than tripled in two years, from 5.4 percent in late 2007 to 20.1 percent in late 2009. But the safety nets in place at the time of the crisis were not adequate: (i) the coverage of unemployment insurance was around 40 percent of the newly unemployed in 2008; (ii) the coverage of last resort social assistance program was small (22.8 percent of those certified as needy were receiving the GMI in 2008); and (iii) its adequacy was 24 The Latvia Welfare Reform Project (P035807) that supported structural reforms in social protection during the 1990s closed with a highly satisfactory rating in 2003. The Latvia Health Reform Project (P058520) closed in 2001 with a satisfactory rating and the Education Improvement Project (P049172) closed in 2004 with a satisfactory rating. 13 low, as the GMI benefit was set to 27 LVL per month per person in 2008, 25 representing less than 15 percent of median per capita monthly income. The first SDPL objective was therefore highly relevant. 43. The SDPL series assisted in mitigating the social costs of fiscal consolidation. As the Government introduced large cuts in social spending to contain the public deficit, the SDPL series was relevant to minimize potential adverse effects and maintain access to public services, especially for the poorest households, while improving efficiency in the social sectors. Fiscal pressure on the budget was high, with a public deficit reaching 7.5 percent in 2008 and 7.8 percent, leading to fiscal consolidation. Efficiency of spending was low, both in the health sector and the education sector, suggesting that the social sectors could not be spared from that consolidation. Real spending decreased by 20.8 percent between 2008 and 2009 in health26, 21.7 percent in education,27 and 8.9 percent in social assistance28. In the context of crisis, the ESSNS was highly relevant to protect the most vulnerable households from this severe contraction in public spending and the second objective of the series was highly relevant. 44. The SDPL series sought to support reforms that lay a foundation for medium-term improvements in the social sectors. The project documentation clearly suggests that this objective focuses on structural reforms that might improve social spending efficiency, although its formulation was rather generic. Efficiency of social spending was an issue prior to the crisis. The education system contained an excessive number of teaching staff, schools were numerous and small, while the quality of education as measured by the Program for International Student Assessment (PISA) remained well below EU average. The structure of health care financing raised costs, with little impact on health outcomes that were among the worst in the region. And, as previously mentioned, the coverage and adequacy of social assistance programs remained meager, with spending on the last resort poverty benefit only about 0.2 percent of GDP, one of the lowest shares of spending among the countries in the EU. The third objective was therefore highly relevant as improvement in the social sectors was needed. Relevance of design and implementation 45. The design of the SDPL series was also relevant. The loans were based on a comprehensive set of analytical studies including an evaluation of the likely poverty and social impacts due to the impact of the crisis on the labor market, reviews of revenue and public expenditure, and technical notes on international best practice (see Annex 5). This informed the Government and the team in selecting the key areas where significant efficiency gains could be 25 The GMI benefit was increased to 37 LVL per month per person in 2009 and from October 2009 was set to 40 LVL per adult and 45 LVL per child. 26 See Table 8, which shows total Health Spending from the Budget deflated by the consumer price index. 27 Total public education spending deflated by the consumer price index. 28 Spending on social assistance at the local level deflated by the consumer price index. 14 achieved in a short time frame. Initial triggers for the second loan and prior actions were designed accordingly. 46. The program was designed as two successive operations because the overall program was ambitious and included measures that would require some time to be adequately implemented, and it covered three budget cycles, including the supplemental budget in 2009, the 2010 budget, and the 2011 budget. Splitting the program assured that the Bank team would be appraising the program as the budgets were developed and approved, which also coincided with high-profile IMF and EC discussions with the government. The first SDPL focused especially on the mitigation of the crisis and the fiscal consolidation, as those objectives needed to be addressed in the short term. Then one year later, the second operation continued to support emergency crisis measures and built on the measures of the first operation focusing on medium-term improvements, particularly in the health sector. 47. The indicators chosen were proxies for the prior actions, selected from the larger set of indicators in the results framework. Overall, the indicators chosen to measure the achievement of the PDOs were relevant as they covered the three objectives in each of the three sub-sectors. However, the targets set for several intermediary outcomes were ambitious, given how the potential number of beneficiaries emerged and they did not provide for an accurate assessment of progress towards the PDIs. 48. Each of the three parts of the PDO was covered by at least one policy action in each subsector, as seen in Table 3. First, transfers to poor households during the crisis were monitored in the areas of social protection, health, and education. Second, the main mitigation measures, that guaranteed a minimal level of spending in the three social subsectors, were directly captured by the indicators. Third, the remaining indicators describe the medium-term reforms, namely the enhanced transparency in the implementation of reforms, the shift from inpatient to preventive and outpatient care in the health sector, and the measures taken to accompany the implementation of the per capita financing in education. 49. A majority of the Development Objective indicators chosen were output indicators covering the three subsectors. In health, the number of the GPs with an extra nurse and the number of subsidized health visits are clear outputs of the program. In education, pre-school enrollment rates, travel time to the school, the share of students receiving assistance for transportation and through stipends are also quantitative outputs of the program. In social protection, beneficiaries of the extension of the unemployment benefit duration or the WWS program are also outputs of the program. 50. The remaining indicators were process-oriented and allowed monitoring the adequacy of the funding of the program as well as evaluation of the strategy. 15 Table 3: Performance Indicators for Development Objectives, by Sector PDO \ Sub-sectors Health Education Social Protection (i) Protect vulnerable Number of patients Share of the state funding Percentage of groups during the receiving 100% for stipends of students in unemployment benefits economic contraction reimbursement of higher education that recipients whose benefits medicine; number of benefit academically are extended by the hospitals’ hotel patients eligible students on the reform. receiving a subsidy; basis of poverty and social Number of workplaces number of subsidized considerations in addition created under the home care visits for needy to merit. emergency public works persons with serious program (WWS). diseases. (ii) Mitigation of the Adequate financing for safety net in 2011 social costs of fiscal consolidation Number of GPs with an Enrollment rates of five- Share of the spending on extra public health nurse. and six-year-old children GMI co-financed by the in free pre-primary State. programs (iii) Foundations for Completion of the Audit of Government’s performance in implementing Emergency medium-term Social Safety Net Strategy. improvements The number of subsidized Time students spend Individual contributions to out-patient visits. travelling to and from the funded pension pillar. primary and secondary Status of the Physician schools. Telephone advisory service. Percentage of students receiving assistance for school transportation in municipalities where schools have closed. 3.2 Achievement of Program Development Objectives Rating: Satisfactory PDO (i): Protection of the vulnerable households during the crisis 51. The efficacy of the SDPL series in the achievement of the first objective is substantial. Protection of poor households during the crisis has been achieved through targeted interventions in health, education, and social protection. As displayed in Table 4, the targets set for the performance indicators were mostly met and in some cases exceeded. 16 Table 4: Efficacy of Measures in Achieving the PDO on Crisis Mitigation Sub- SDPL policy Expected outcomes Medium term policy sector actions outcome Indicators Baseline Target Health Elimination of out- Number of patients 99 patients 22,000 The target was met. of-pocket receiving 100% in the fourth patients in There were 30,267 fully payments for reimbursement of quarter of 2011 covered patients in health visits, tests, medicinal products or 2009 2011. and medicines charges pharmaceuticals for needy patients; provision of package of benefits for patients as their incomes exceed Number of hospitals’ 551 7,200 during The target was partially the needy line, and hotel patients beneficiaries 2011 met as 5,658 patients access to receiving a subsidy in the fourth benefitted from this outpatient under the emergency quarter of subsidy. psychiatric and social safety net 2009 home care for chronic diseases. Number of 115 visits in 200,000 The target was subsidized home care the fourth during 2011 exceeded as 223,683 visits for needy quarter of subsidized visits persons with serious 2009 occurred in 2011. diseases Education Social Share of stipends 39% in 2009 50% in late The target was considerations attributed on social 2011 exceeded and reached added in addition criteria 60%. The equity of to merit as criteria subsidies for higher for the attribution education improved of stipends in during the program, but higher education this reform was challenged on legal grounds and halted in 2011. Social Extension of the Percentage of 0% in 2009 22% during The target was achieved protection coverage of unemployment 2010-2011 during the program. unemployment benefit recipients, insurance whose benefits are extended by the measure Public works Cumulative number 16,830 in 82,305 The target was program (WWS) of workplaces created Dec. 2009 workplaces exceeded. The public under the emergency created at the works (WWS) program public works end of 2011 provided 82,305 jobs program for the unemployed not eligible for unemployment benefits, covering around 17% of the eligible 17 Sub- SDPL policy Expected outcomes Medium term policy sector actions outcome Indicators Baseline Target beneficiaries. Social Extension of the Percentage of needy 22.8% in 50% in 2011 The target was not met. protection coverage of GMI persons receiving 2008 While total spending for benefits GMI benefit GMI was expanded more than threefold over 2009-2011, coverage reached 39.7% on average per month or 45.9% on average in 2011. 29 This was due to the lower than expected share of the certified needy having incomes below the GMI eligibility threshold. 52. In health: The social safety nets deployed allowed not only the protection of the poor from budget cuts but also improvements to both their financial protection and their access to care. Indeed, the number of subsidized outpatient visits increased by 89 percent from 2009 to 2011. At the same time, the number of inpatient subsidized visits decreased by 6 percent from 2009 to 2011. This reduction in inpatient care was more than compensated for by the provision of subsidized hotel accommodation for hospital patients and 222,683 subsidized home care visits in 2011. The access to mental care for the needy was also largely expanded as the number of visits increased nine fold. Moreover, during the program, around 20,000 needy patients also benefited from free medicines.30 53. In education: The reform of stipends in tertiary education improved significantly the equity of aid to students, as the share of stipends attributed to social considerations grew from 39 percent in 2009 to 60 percent in 2011. This measure resulted in additional monetary support targeted to poor students. However, as this reform was overruled by the Constitutional Court, its impact has not been sustained. 54. In social protection, three distinct interventions were set up to mitigate the effects of the crisis on households. First, the duration of the unemployment insurance was extended; second, a 29 The rate of coverage can be computed either as (i) the monthly average number of GMI beneficiaries divided by the monthly number of needy people; or (ii) the total distinct number of GMI beneficiaries throughout the year divided by the total distinct number of people certified as needy during the same year. The first method is consistent with the way the targets were set out in the two program documents for the SDPL series, where targets were set out on a quarterly basis. The second method gives a better account of the number of people who benefitted from the program at some point during the year. Although the indicators for the rate of coverage calculated by both methods differ, both indicators remain below the expected target of 50 percent. 30 The number of patients who benefitted from fully free medicines varied from one quarter to another, from 22,000 in the second quarter of 2011 to 17,000 in the last quarter of the same year. 18 public works program, the WWS, was implemented under the ESSNS; and third, the state co- financing for social assistance benefits was introduced. However, the percentage of newly- registered unemployed eligible for unemployment insurance benefit remained below the target of 60 percent. This target was too ambitious given the severity of the crisis and the dramatic increase in unemployment that ensued. 55. According to the evaluation undertaken by the World Bank, the public works program WWS implemented under the ESSNS was successful31: (i) targeting performance was good as most WWS beneficiaries were poor and vulnerable people; (ii) administration and implementation were satisfactory as payments were accurate and on time, supervision was good and beneficiaries rated the program positively; and (iii) the program performed as a safety net by mitigating the impact of income shocks and protecting households from adopting harmful coping strategies. It was also a substitute for other safety nets. However, the scale of WWS was small relative to the demand, as between 20,000 and 30,000 persons were on the waiting list, compared with the 82,000 workplaces created from 2010 to 2011. Figure 1: Evolution of Needy people, GMI beneficiaries (left) and GMI spending (right) 200 25 180 160 20 140 Thousands Millions Lats 120 15 # certified as needy 100 # GMI beneficiaries 80 10 GMI spending 60 40 5 20 0 0 2009 2010 2011 Source: Ministry of Welfare. 56. Despite a very rapid growth in GMI benefits—beneficiary numbers increased three- fold over 2009-2011—the program never reached the target of covering 50 percent of needy people (see Figure 1). The eligibility threshold for the GMI program was too low to allow the target of 50 percent coverage of the needy to be reached. The needy line was set at 90 LVL per 31 The Impact of Latvia’s Public Works Program During the Great Recession, Céline Ferré, Mehtabul Azam and Mohamed Ihsan Ajwad, The World Bank, Mimeo, November 2011. 19 month per person, while the GMI eligibility threshold was much lower at 40 LVL per adult and 45 LVL per child per month. The share of the needy with incomes low enough to qualify for GMI was smaller than the 50 percent of the needy target set in the SDPL program. Therefore, meeting the 50 percent target would have required a much more precipitous drop in incomes of the poorest segment of the population than occurred. 57. However, according to municipal level data, the spatial inequity of GMI spending across municipalities narrowed following the introduction of the SDPL series. As displayed in Table 5, the average GMI spending per needy person at the municipal level increased by 78 percent between 2010 and 2009, and increased further by about 26 percent in 2011. At the same time, inequalities between municipalities, as measured by the standard deviation of the average GMI spending by needy person decreased drastically in relative terms. Finally, the negative correlation between the share of the needy in the total population and average GMI spending by needy person dropped from 2009 to 2011, when co-financing from the state budget was in place. This negative correlation illustrates that in municipalities with larger needy populations, spending per needy person is lower. The fact that the negative correlation diminished shows that the co-financing of GMI spending by the central government was effective in equalizing GMI spending for potential beneficiaries across municipalities over a period when local government budgets were constrained. Table 5: Descriptive Statistics for GMI Spending Per Needy Person, by Municipality 2009 2010 2011 Average 37.3 65.5 82.9 Minimum 0.0 2.5 5.3 Maximum 194.3 120.3 145.2 Standard deviation as % of national average 68.5 40.9 40.2 Correlation between share of needy in total population and per -0.34 -0.34 -0.10 needy GMI spending Sources: World Bank Staff calculations using Ministry of Welfare data. Note: The table presents the average spending on GMI in each municipality divided by the number of people certified as needy in that same municipality. Its purpose is to show to what extent the actual spending on GMI benefits are linked to local needs. PDO (ii): Mitigation of the impacts of fiscal consolidation 58. The efficacy of the SDPL series in the achievement of the second objective is substantial. The targets were met for the performance indicators (see Table 6), and the program managed to preserve an adequate funding level targeted on the vulnerable and poor households, despite large cuts in the budgets of all the social sectors. 20 Table 6: Efficacy of Measures in Achieving the PDO on Mitigation of Adverse Impact of Fiscal Consolidation Sub-sector SDPL policy Expected outcomes Medium term policy and actions Indicators Baseline Target outcome Overall ESSNS Maintenance The financing Emergency safety net funding satisfactorily of an adequate for the safety net measures were deployed and implemented budget for the strategy is remained in place through strategy in adequate. 2010-2011 2010 and 2011 Health Strengthening Share of GPs 0 in 2009 The target 465 practices benefitted from primary care with an coverage of the a public nurse, representing practices by additional measure was 500 93% of the target. providing extra nurse practices at the The provision of primary additional end of 2011. health care services remained public health stable since 2009, while nurse and subsidized visits increased by preventive 89% during the program. examinations of patients targeted on the poor Education Adequate Enrollment 89% for 93% for five Enrollment rate increased to budget allocated rate of five- five year year olds, 98% 94% for 5-year olds and to cover for and six-year olds, 98% for six year olds remained (97%) stable for 6- five- and six- olds. for six year olds as intended. year olds year olds education Social Increase of the Share of 47.1% 50% The share of co-financing of protection GMI and co- spending on GMI by the State remained financing by the GMI financed around the set target of 50% State through throughout the program. ESSNS co- However, the share of needy financing households receiving GMI mechanism, % was below the target set of 50%. 59. In order to mitigate the effects of the fiscal consolidation in the social sectors, the SDPL series aimed first at guaranteeing adequate funding for the implementation of the ESSNS. Overall, the budget allocated for the ESSNS was adequate both in 2010 and 2011 (see Table 7). The actual spending for the ESSNS increased by 34 percent compared to the spending planned in the initial budget in 2010: the planned spending was LVL 56.3 million and the actual spending increased to LVL 77.3 million. The fiscal consolidation process was difficult, particularly during the 2010 discussions on the 2011 budget cuts. In the version of the 2011 budget, which was sent to Parliament in November 2010, the plan was to cut resources for the WWS program and state co-financing for the GMI program by about 40 percent overall in 2011. The World Bank team advocated for a larger budget allocation for the strategy. Following discussions with the EU, IMF, and World Bank, the Latvian Government decided to maintain the social safety net spending at 2010 levels in 2011. For GMI, it was possible to make the 2011 budget modification prior to the final passing of the Budget Law 2011 by the Saeima in December 2010. The WWS 21 funding contained in the 2011 budget was later supplemented by additional resources from the European Social Fund (ESF) in 2011.32 60. Actual state budget spending for the ESSNS measures in 2011 was about 100 percent of the 2010 executed budget. Spending on the health safety net increased by 28 percent in nominal terms in 2011 compared to 2010. Funding was cut for the transportation of students due to lower than expected demand and the WWS program funding was reduced by 30 percent. Given the continued high demand for the WWS program, it would have been desirable to increase or maintain funding in 2011 at 2010 levels. However, given the considerable fiscal consolidation pressures, the Government kept to their plan of winding down the program in 2011. Table 7: State Budget for ESSNS Measures Supported by the World Bank, 2010, 2011, and 2012 (in millions of Lats, nominal) 2010 2010 2010 2011 2011 2011 2012 plan revised actual plan revised actual plan Total 56.3 77.3 76.2 68.6 75.9 74.7 43.5 GMI State co-financing 7.0 9.7 8.7 9.1 12.0 11.1 11.2 Housing benefit State co-financing 2.8 3.5 3.0 3.1 4.0 4.0 4.0 Pre-primary and primary school curriculum 0.6 0.2 0.2 0.9 0.9 0.9 0.0 improvement Local government schools, teacher salaries for 5- 12.3 12.3 12.3 12.3 12.3 12.3 12.3 and 6-year old education Transporting students from closed to new schools 5.9 4.2 3.5 1.0 0.6 0.6 0.0 ESSNS health care measures 24.2 24.2 24.2 30.3 30.9 30.9 19.9 Workplaces with Stipend program 6.4 26.6 27.2 14.0 19.1 18.9 0.0 Memo item: Total ESSNS Spending (as a 0.3 0.6 0.6 0.5 0.6 0.6 0.3 percentage of GDP) Source: Ministry of Finance Note: Table includes all ESSNS measures supported by the World Bank. 61. In health: The Government introduced substantial cuts in the health sector (see Table 8). However, spending cuts have been focused strategically on the least efficient areas, such as inpatient spending (fell by 50 percent in real terms), while critical services (such as emergency care which rose by one percent in real terms), were spared. About seven percent of the total health expenditure was directed toward putting in place the social safety net (amounting to 30 million LVL in 2011), which focused on needy and low income households. 32 The funding of the WWS from the state budget was cut by 45 percent compared to 2010 in the Budget Law 2011 passed by the Saeima in December 2010. This was supplemented by additional resources from European Social Fund (ESF) in 2011. Following the passing of the 2011 Budget Law, the Government requested a reallocation of LVL 5 million of ESF resources into the WWS program. The EU accepted the proposal in March 2011 and the Cabinet of Ministers approved the proposal on April 12, 2011. 22 Table 8: Government Health Budget, 2008-2011 (in millions of Lats, nominal) 2008 2009 2010 2011 Nominal Real Change Change 2008-2011 2008-2011 (percent) (percent) Payments Center 463.0 379.0 381.0 382.0 -18 -22 Inpatient 242.0 182.0 155.0 130.0 -46 -49 Outpatient 138.0 119.0 118.0 129.0 -6 -12 Secondary Outpatient 89.0 72.0 73.0 86.0 -3 -9 General Practitioner Practices 43.0 42.0 45.0 38.0 -11 -17 Dentistry 6.0 6.0 6.0 6.0 -13 -18 Medicines, Material and vaccines 76.0 72.0 78.0 85.0 12 5 Social Safety Net 0.0 0.0 24.0 31.0 - - Other 7.0 5.0 6.0 7.0 -7 -13 Ministry of Health 101.0 82.0 81.0 77.0 -24 -29 Communicable Disease 24.0 20.0 15.0 14.0 -41 -44 Blood Supply 9.0 7.0 6.0 5.0 -37 -40 Emergency Medical Assistance 27.0 26.0 31.0 30.0 13 6 Other 42.0 29.0 29.0 27.0 -36 -40 Total Health Spending from -19 -24 564.0 461.0 462.0 458.0 Budget Price index 100.0 103.0 102.0 106.0 6 - Sources: Ministry of Health and Latvia’s CSB. 62. In education: Public education funding was reduced as part of the fiscal adjustment that followed the crisis (see Table 9). Total government spending on education fell by 36 percent in nominal terms and 38 percent in real terms from 2008 to 2010. 33 Although nominal spending grew by one percent in 2011, it continued to shrink in real terms by three percent. Rapidly declining enrollment since 2008 compensated to some extent for this fall in funding. However, the per student allocation still decreased in real terms by about 20 percent in general education, 30 percent in pre-school education, and 51 percent in higher education. 63. In 2005, the Government made pre-school for five- and six-year-olds free and compulsory. However, local governments are responsible for education spending. Because of the tight budget constraints faced by local authorities, the Government sought to ensure that the funding of pre-primary school should be secured for five- and six-year-old children. This measure was effective and enrollment rates for both five- and six- year olds remained stable in 2010 and 2011. By guaranteeing two years of pre-school, this measure prevented an entire cohort from being less prepared for school. However, funding should be improved in the medium run to improve overall access of children to pre-school, including three- and four-year-old children. The 33 Source: World Bank Staff calculations using Budget Documents, Ministry of Education and Science and Treasury’s Annual Report on Local Governments. 23 Ministry of Education and Science (MoES) is considering extending pre-school education coverage to children from the age of 18 months upwards. Table 9: Spending on Education, 2008-2011 (in millions of Lats, nominal) 2008 2009 2010 2011 Nominal Real Real Change Change Change per student 2008-2011 2008- 2008-2011 (percent) 2011 (percent) (percent) General Education (1-12) 490.0 397.0 350.0 350.0 -29 -33 -22 Preschool Education 116.0 106.0 86.0 85.0 -27 -31 -29 Special Education 49.0 49.0 42.0 42.0 -14 -19 -20 Vocational Education 66.0 52.0 46.0 55.0 -17 -22 -13 Other Education 68.0 53.0 48.0 48.0 -29 -34 -10 Higher Education 127.0 54.0 43.0 45.0 -65 -67 -51 Science 43.0 22.0 17.0 16.0 -63 -65 Education spending 992.0 803.0 639.0 647.0 -35 -39 - Consumer Price Index 100.0 103.3 102.0 106.3 6 - - Sources: Budget Documents, Ministry of Education and Science and Treasury’s Annual Report on Local Governments Consolidated Budget Execution. For 2010 and 2011 local government spending for each category is calculated to correspond to the average spending structure during 2008 and 2009; Consumer Price Index data taken from the IMF’s International Financial Statistics. Note: The figures represent public spending on education from domestic financing sources. 64. To accommodate the per capita financing reform, which led to a reduction in the number of schools by 128 from 2008 to 2009, the Government subsidized transportation for the children from closed schools. These transportation measures exceeded their targets in spite of a reduced budget, reaching almost universal coverage or 97 percent of the children whose school closed. 65. Total spending in social protection continued to grow quickly following the onset of the economic crisis in Latvia (see Table 10). Old age spending, including pensions, accounted for most of the growth. The rise in pensions was driven by the high notional interest rates applied to pensions reflecting the lagged growth of the wage and benefit bill of the insured population, and also by pre-crisis discretionary policy measures. Unemployment benefits also grew; pushed up both by massive unemployment and the extension of the benefits coverage under the ESSNS. Other categories such as sickness and family were reduced. Spending on GMI and WWS emergency public works programs rose substantially over the period. GMI spending increased in 116 out of 117 municipalities between 2009 and 2011. Moreover, total social assistance spending at the local level increased by 70 percent between 2009 and 2011 (see Figure 2). The state co- financing of 50 percent of GMI and 20 percent of housing benefits was effective in preserving local governments’ social assistance spending. 24 Table 10: Consolidated Public Sector Spending on Social Protection (in millions Lats, nominal) Change 2008-2010 2008 2009 2010 (in percent) Total 1,568.0 1,870.0 1,802.0 14.9 Old age and survivor (including pensions) 914.0 1,096.0 1,128.0 23.5 Unemployment 60.0 146.0 107.0 77.7 Sickness 334.0 386.0 319.0 -4.3 Family and children 125.0 128.0 117.0 -6.5 Housing and exclusion 72.0 57.0 67.0 -6.8 Other 63.0 57.0 63.0 -0.2 Memo items: GMI spending 4.0 14.0 37.0 802.4 WWS public works 0.0 0.0 27.0 … Sources: Latvia’s CSB. Notes: The CSB data amalgamates data from the state budget and local governments. 66. Measures taken under the ESSNS mitigated the impact of the fiscal consolidation on the poorest households. That the large increase in GMI spending occurred in the context of substantial fiscal pressures to consolidate municipal budgets was a result of the provision of 50 percent co-financing by the central government. Additional funds for GMI were used to increase the amount of the benefits and also to expand coverage, which increased from 22.8 percent in 2008 to 45.6 percent in 2011(if measured as the share of the people classified as needy who received GMI at some point during 2011). Figure 2: Local Government Spending on Social Assistance 100 90 Others Housing Current Millions of LVL 80 70 GMI 60 50 40 30 20 10 0 2008* 2009 2010 2011 Source: Local government reporting of social assistance spending. Note: (*) Housing benefits are not displayed separately in 2008 local spending accounts. 25 PDO (iii): Foundation for medium-term improvement 67. Despite a limited budget, the SDPL series managed to support structural reforms that are very likely to induce solid improvements in the medium-term in all three sectors. In addition, the monitoring and evaluation process put in place for the ESSNS enhanced the independence and the transparency of the governmental administrations involved. 68. In health, the Government seized the opportunity of the fiscal consolidation to undertake a major reform of the health care system, improving efficiency through a large shift from inpatient to outpatient services (see Table 11). Between 2008 and 2010, half of the contracted hospitals were either closed or their inpatient contract cancelled, thus being converted into daily care centers. This allowed a reduction in the number of hospital beds by 28 percent during the same period. The Government also improved the quality of the medical personnel by replacing those with secondary education with nurses with higher education and assigned to primary care. This allowed reinforcement of the general practices and improvement in their coverage through an extension of their opening hours and by deploying a family physician advisory telephone service. This service has been operating from May 1, 2011 from 7 p.m. to 8 a.m. on weekdays and around the clock during weekends and holidays. This service has been receiving around 2,000 calls a month since its creation, but its usage increased to 5,913 calls a month by March 2012 after a public awareness campaign was carried out at the beginning of 2012. It is designed to provide medical advice when general practitioner offices are closed, preventing unnecessary hospital visits and providing good primary care advice during off hours. Table 11: Health Care System Inputs 2008 2009 2010 2011 Change 2008-2010 (in percent) Physicians of all specialties (excluding in 8,437 7,964 7,951 7,987 -6 service training and post-graduates) Nurses with higher education 692 831 1,090 1,480 58 Medical personnel with secondary 15,197 13,492 13,217 12,494 -13 medical education Hospitals 88 69 67 70 -20 Hospitals Contracted 62 43 40 40 -35 Beds (in all hospitals) 16,907 14,434 11,920 12,111 -28 Beds (in hospitals who provide state 16,042 12,604 10,219 10,909 -32 funded in-patient health care services) Sources: Ministry of Health and Latvia’s CSB. 69. The health reforms were accompanied by an improvement in self-reported health status following the onset of the crisis (see Figure 3). In addition, the infant mortality rate dropped from 8.7 to 5.7 per thousand live births between 2007 and 2010, a 65 percent drop; and the crude death rate dropped from 14.5 to 13.4 per thousand even while hospitalizations dropped from 234 26 to 170 per thousand. This progress coincided with the reforms supported by SDPL series. Although we cannot attribute cause and effect, despite the safety net for the poor and the intended improvements to emergency services, it should be noted that there were strong fears in 2008-09 that the shift of emphasis from inpatient to outpatient care was an extremely risky reform and would result in many additional deaths. Figure 3: Evolution of Self-Reported Health Status by Gender, 2007-2010 8.0% 7.0% Change in the share of respondant by status 6.0% 5.0% (compared to 2007) 4.0% 3.0% 2.0% 1.0% 0.0% 2008 2009 2010 2008 2009 2010 -1.0% Males Females -2.0% -3.0% Very good Good Bad Very bad Source: World Bank staff calculations based on Health Interview Survey for Latvia (Eurostat). Table 12: Efficacy of Measures in Achieving the PDO on Medium-term Improvement Sub-sector SDPL policy Expected outcomes Medium term policy and actions Indicators Baseline Target outcome Health Shift to Number of 27,909 out- 420,000 Total in-patients visits were outpatient subsidized out- patient and out-patient reduced by 10.3% in 2010 services patient and in- 2,140 in- and 22,680 while subsidized visits patient visits patient in-patient decreased only by 6%. The subsidized subsidized number of beneficiaries visits in the visits exempted from medicine co- last quarter of during payments increased fourfold 2009 2011 since 2009. 27 Sub-sector SDPL policy Expected outcomes Medium term policy and actions Indicators Baseline Target outcome Education Adequate Share of 70% 77% in 96% of students from closed budget allocated children from late 2011 schools received to cover the cost closed schools transportation assistance of transportation receiving 0.8% in for pupils from assistance for late 2011 closed schools transportation; Travel time to 1.5% of Travel time exceeded one school below students travel hour for only 0.4% of the one hour more than one students. hour Social Financing of the Contribution It was reduced It will be The rise in contributions will protection second pillar rate to 2% during raised to contribute to increasing pensions the crisis 6% in replacement rates in the January future. 2013 70. In education, the margin for efficiency improvement was large on the eve of the crisis. Latvia did not undertake major reforms to compensate for its rapidly declining student population and unit costs grew by 38 percent in real terms between 2000 and 2009. Therefore, efficiency was very low prior to the crisis with too many teaching staff and small schools.34 The implementation of the per capita financing system allowed a reduction in the number of schools by 124 in general education (a 13 percent decrease) and as a result of optimization by nine in vocational education (a 10 percent decrease) between 2009 and 2011. The number of schools continued its decline, although at a slower pace in 2012 in general education as the number of schools decreased by four. In 2010/2011 the number of vocational education institutions has been reduced from 54 to 38. 71. The measures taken under the ESSNS to ensure transportation of students whose school had closed to new schools prevented this rapid school consolidation from having an adverse effect on access. Indeed, enrollment rates in general schools did not decrease from 2008 to 2011. Moreover, the overall quality of education did not suffer, as average test scores tended to be higher in 2009-2011 than before the crisis for grades 3 and 6, and similar for grade 9 (see Figure 4).35 In grade 12, the results clearly improved in language, but deteriorated in mathematics. At the same time, inequalities in school performance neither grew nor declined. Moreover, according to a statistics produced by the MoES, pupils’ performance in the schools that were reorganized between 2009 and 2011 was maintained overall.36 34 Pupil teacher ratios were as low as 10.1 in the secondary level. 35 All students are tested in grades 3, 6, 9 and 12 in the Latvian language and Mathematics every year. 36 These statistics give a comparison of average mathematics performance in ninth grade in a random selection of half the schools that were reorganized between 2009 and 2011. 28 Figure 4: Average Test Score Results in Latvia 35% Difference with 2006-2007, average 30% 25% 20% scores 15% 2008 10% 2009 5% 2010 0% 2011 Math Math Math Math Language Language Language Language -5% -10% Grade Grade Grade Grade Grade Grade Grade Grade 3 3 6 6 9 9 12 12 Source: State Education Center. 72. In social protection, the diversion of the second pillar fund contributions to finance the first pillar commitments reduced the long-term sustainability of the pensions system. From January 2009, the contribution rate to the private pillar was reduced from eight percent of a worker’s salary to two percent and diverted to fund the current social insurance budget. This allowed for the partial closing of an unexpected funding gap in the pay-as-you-go system. Indeed, following strict pension benefit calculation rules, newly-assigned pensions were growing even after the growth in pension contribution revenues had turned negative. This further added to pension spending in the short and medium term. While the second pillar contribution rate of eight percent was consistent with economic prospects before the crisis, the rate was not affordable following the onset of the crisis. 73. Initially, in the first SDPL the Government committed to increase the second pillar pension contribution rate to six percent on January 1, 2012. A six percent contribution rate, even though less ambitious than the eight percent in existence prior to the crisis, was still considered sizable and in line with what some of the other Eastern European countries have in place. But in 2010, the economic crisis was much deeper than expected. Therefore, in the second SDPL the Government shifted its commitment to increase the contribution rate to six percent by one year to January 1, 2013. Postponing this measure by one year was not unreasonable given the funding gap in the pay-as-you-go system to be financed in the short-term. The drawback of this approach was that it may have eroded some credibility in the Government's commitment to individual pension savings accounts. 74. The SAO report illustrated that the rapid expansion of spending on GMI benefits was accompanied by problems in implementation regarding the targeting and the administration of 29 social assistance.37 The implementation of the action plan in response to the recommendations of the SAO may have positive impacts on the efficiency of social assistance spending in the near future. In addition, during this period, there remained a shortfall in the certification of those eligible for ―needy‖ status by local governments. The percentage of persons eligible for needy status that have been certified as needy has remained below the target of 50 percent. 75. Finally, the process used for monitoring and evaluation has fostered government transparency in implementing the policies supported by the SDPL series. First, an independent audit of the ESSNS was completed on October 2011 by the SAO and published on the Internet. The report underlined several weaknesses surrounding the targeting of social assistance. The SAO’s recommendations led the Ministry of Welfare and other involved governmental agencies to agree on an action plan that has been since implemented (see paragraph 35). Second, the ESSNS has been monitored on a quarterly basis and the indicators tracked have been published on the Internet, ensuring transparency on the implementation and results of the strategy. 3.3 Justification of Overall Outcome Rating 76. The overall rating is satisfactory for both SDPLs. The rating reflects the high relevance of the development objective, the relevance of the design and implementation and the achievement of the program objective. All planned policy actions were implemented, although one measure was postponed by one year and another one was reversed by the Constitutional Court. The operation was well designed with a comprehensive set of prior policy actions covering the three social sectors: health, education, and social protection. All the PDO indicators reached or exceeded their set targets, although many intermediary outcomes were not met as the targets set by the Government were overambitious given the brief time between startup of the ESSNS and measurement of targets. Finally, the operation occurred in a constrained fiscal environment, but still the Government honored its commitment to the ESSNS. 77. Outcomes in the social sectors have continued to improve during the programs. In Social Protection, first, the share of needy people in the total population remained consistently below expectations, reaching 12.8 percent in 2011 compared to 7.9 percent in 2009. Second, self-reported health status continued improving during 2010 for both males and females. Third, neither enrollment nor performance dropped in the education sector, in spite of severe spending and inputs adjustment. These positive trends underline that the program was protecting the population, especially the most vulnerable households, from the effects of the economic downturn and the fiscal consolidation. 78. The targets set for the overall implementation were fully met on time, enabling: (i) an adequate funding of the strategy, (ii) full transparency during implementation, and (iii) an 37 The SAO’s evaluation identified three problems: (i) qualifying criteria not sufficiently clear, (ii) an incorrect application of those criteria by the municipal social assistance offices, and (iii) insufficient or inadequate controls. 30 independent and relevant evaluation that led to improvements in the program. The extension of most of the measures after the end of the World Bank’s financial support to the program is further proof of the effectiveness and relevance of the ESSNS. 79. In health, the results of the key PDO indicators exceeded the targets set. The number of subsidized visits for: (i) inpatient, (ii) outpatient, (iii) home care, and (iv) mental care were significantly higher than the targets set for 2010 and 2011. Moreover, at the closure of the SDPL, the number of GP practices benefitting from an additional public health nurse reached 93 percent of the target, providing 465 additional nurses to mitigate the negative risks of cutbacks to inpatient care. The number of patients receiving 100 percent reimbursement of medicinal products or medicines charges exceeded what was expected in 2011, with almost a half more patients receiving reimbursements compared to targets (30,267 patients received reimbursements versus the target of 22,000). One gap was that no day care centers were built although two should have been according to the results framework. Demand for subsidized places in hotels for hospital patients did not materialize as expected and so the target set for both 2010 and 2011 were missed. As noted in the Government’s response to this ICR, the Government will stop the subsidy for hotel patients in 2012 and increase availability of day care and home care to provide a more appropriate solution for these patients (such as dialysis patients or oncology patients that previously had inpatient stays for their regular therapy), so it continues to execute the underlying strategy. However, the number of needy patients who benefitted from home care and mental care respectively exceeded the target set by 340 percent and 80 percent. This is evidence of the Ministry of Health’s efficiency and being able to provide the needed care without additional infrastructure. The overall health outcomes continued improving during the implementation of the reform while inputs in the sector were drastically reduced and real spending was cut by 25 percent between 2008 and 2011. This shows that the SDPL series clearly had a very positive impact on efficiency. Eventually, the numerous hospital38 closures represent a strong step toward a medium-term rationalization of the health system. In the health sector, the program achieved the three PDO objectives beyond expectations and is therefore rated Highly Satisfactory. 80. In education, all indicators met or exceeded their targets. However, because of a Constitutional Court ruling, the measure promoting social diversity among beneficiaries of higher education stipends was terminated. Although its temporary implementation led to some short term gains for students of low socioeconomic background, this measure did not lay the foundations for improvement in the medium-run. The indicators monitored showed that enrollment of the five and six year olds did not suffer from the cut in funding. Finally, measures taken for the transportation of students whose schools have closed were very successful and 38 In most of the cases, the hospitals were not shut down but their operating contract was modified to make them operate only on a daily basis as outpatient centers. 31 exceeded their targets in reducing excessive travel time. The outcomes in education are rated Satisfactory. 81. In social protection, all four key PDO indicators were met: (i) the extension of the duration of unemployment benefits by nine months benefited more than 22 percent 39 of the unemployment insurance beneficiaries, but the share of beneficiaries declined progressively; (ii) the temporary diversion of the second pillar pension was halted; (iii) the State funded half of the spending for the GMI; and (iv) the results exceeded the targets set for the public works program. The evaluation of the public works WWS program showed that this measure was very effective at mitigating economic shocks. Together with the extension of unemployment benefits and the increase of the GMI, this measure allowed poverty to be contained during a very severe crisis and in a tight fiscal environment as social assistance spending decreased by 6.8 percent between 2008 and 2010. The increase of the second pillar contribution and the implementation of the action plan to better target the benefits to the needy constituted a solid step for the rationalization of social protection in the medium run. However, as the coverage of both unemployment insurance and GMI did not reach the set targets, outcomes in social protection are rated Moderately Satisfactory. 3.4 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 82. The operation was clearly intended to benefit the poor directly in terms of mitigating the impacts on the poorest households from the economic crisis and the fiscal adjustment that followed. The program increased the direct monetary support to the poorest households through (i) a 70 percent increase of social assistance spending at the local level from 2009 to 2011, (ii) more targeted direct transfers to the needy households due to GMI spending multiplied in real terms more than threefold from 2009 to 2011, (iii) the extension of the duration of unemployment benefits by nine months, and (iv) the provision of public works to those unemployed who were not eligible for benefits. 83. In the area of health care, the program clearly increased the financial coverage of the poorest households. The strengthening of primary care allowed the subsidies coverage to be extended to outpatient services. The exemption of low-income households from any health copayments substantially reduced the financial burden of health care. As a result of the new Ministry of Health direct subsidies system to the care providers and pharmacies, low-income households did not need to pay upfront for care and benefited from free services. Under the previous system they had to pay first, then seek reimbursement from local authorities. In addition, the switch from in-patient toward more out-patient visits is likely to have contributed to better health outcomes by expanding preventive care. 39 The target of 22 percent was met except throughout 2010-2011 except in the last quarter of 2011. 32 84. In the education sector, the program interventions were clearly targeted to the poorest and most vulnerable and therefore were able to mitigate the impact of school and fiscal consolidation. 85. The project design did not incorporate or intend to have specific impacts on gender, environment or social development. Government data did not disaggregate the impact by gender. Simulations made by World Bank staff showed that the impact of the crisis may have been similar on both males and females.40 (b) Institutional Change/Strengthening 86. The SDPL series had a positive impact on the Latvian institutional environment, by laying the foundations for medium-term change. The intense policy dialogue supported by several technical assistance outputs that accompanied the SDPL series helped institutional strengthening, especially by: (i) clarifying and improving the efficiency of the social assistance administration at the local level,41 (ii) supporting the creation of a centrally operated health care safety net for the needy, (iii) supporting local authorities in organizing transport for the students. Furthermore, the program promoted transparency of the Government’s institutions in the implementation of reforms, through (i) regular monitoring of the Safety Net Strategy and the quarterly release of the monitoring indicators to the general public on the internet and (ii) an independent audit of the ESSNS by the SAO, whose results were also disclosed on the internet. (c) Other Unintended Outcomes and Impacts 87. None 3.5 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 88. Not applicable 4. Assessment of Risk to Development Outcome Rating: Moderate Macroeconomic risk: 89. The macroeconomic risk is moderate. The economic recovery was faster and stronger than initially expected. Real GDP growth has been positive and accelerating since the second half of 2010. The authorities’ commitment to maintaining the exchange rate peg to the Euro involved achieving substantial wage and price cuts combined with a growth in productivity with the aim of improving competitiveness and reducing external imbalances. This entailed costs in terms of increased unemployment, which reached a peak of 20.7 percent in 2010. As the 40 Ajwad, M.I., F. Haimovich and M. Azam (2012), ―Simulating the impact of the 2009 financial crisis on welfare in Latvia‖, World Bank Policy Research Working Paper; no. WPS 5960, The World Bank: Washington DC. 41 Thanks for instance to the numerous recommendations of the SAO about the way municipalities determine the eligibility of needy households to social assistance. 33 economy recovers, unemployment has been decreasing, albeit slowly, and it still remains high at above 14 percent at end-December 2011. Inflation has fallen from previous peaks and is expected to continue to decline as international food and energy prices fall further. The large fiscal adjustment is expected to result in a 2012 fiscal deficit of 2.1 percent of GDP (ESA95), below the fiscal deficit ceiling of three percent according to Maastricht criteria. The financial sector has strengthened as indicated by an improvement in prudential indicators for the banking sector. Deposits are now above pre-crisis levels. 90. The EU-IMF stabilization program concluded at end-2011. The Government of Latvia successfully completed the Fifth Review under the Stand-By Arrangement and the financing assurances review on December 21, 2011. The IMF Staff Report concluded that Latvia’s program had achieved many of its objectives given that international reserves had recovered to pre-crisis levels, the exchange rate peg was maintained, the financial sector strengthened, competitiveness improved, and a large fiscal consolidation achieved bringing the country on track to meet the Maastricht fiscal criterion in 2012.42 However, IMF staff noted the cost of the adjustment in terms of high unemployment and poverty, and cautioned on a number of unresolved vulnerabilities, namely the increased external debt ratio entailing short-term financing requirements and the high share of non-performing loans in the banking sector that will continue to constrain credit availability and domestic demand. 91. The improvement of economic conditions has been reflected in ratings upgrades and facilitated the placement of a Eurobond in June 2011. In 2012, Latvia received a further upgrade to investment grade from all three credit rating agencies. Nevertheless, the worsening global outlook is likely to impact adversely on growth in 2012 and the euro area debt crisis may constrain the availability of funding from international capital markets. Political risk: 92. The political risk is moderate. Although the Prime Minister Dombrovskis’s party, Unity, lost seats in the Parliament in the September 2011 elections, the Prime Minister formed a coalition with the Reform Party and National Alliance and remains in power for a third term. This has helped ensure the continuity of the fiscal and structural reforms, including those undertaken in the social sectors. Government ownership/commitment risk: 93. The risk is moderate. Throughout the crisis and the reform program that followed, the Government exhibited strong ownership and a commitment to reforms. Although examples from 42 Fifth Review Under the Stand-By Arrangement and Financing Assurances Review, IMF Country Report No. 12/31, February 2012, IMF: Washington D.C. 34 other countries are not favorable,43 the Government has shown its commitment to reverse the decrease of the contribution rate to the second pillar pensions at the beginning of 2013. Finally, the Government’s continued collaboration with the World Bank on how to combat long-term unemployment and labor market inactivity is an additional sign that they are willing to build on short-term measures taken in response to the crisis and build on these measures to put in place sustainable reforms. 94. However, the Government does face a considerable challenge in continuing with structural reforms, particularly as the fiscal situation eases and political pressures for spending increases emerge. The country managed to implement large reforms in a short time, but the structural reforms to be tackled in the coming years remain large in health, education, and social protection. In the health sector, the country has achieved a major shift to a better mix of inpatient and outpatient services while also lowering costs as a consequence. The potential gains from this strategy are far from exhausted. However, as the hospitals that were closed or converted into day care centers remain under the management of local authorities, there is still a risk that such facilities might be reopened or reconverted back to inpatient care centers in the future, especially if economic growth were to accelerate. The school network consolidation that was achieved during the crisis remains insufficient to compensate for the past decline in student population. Further adjustment will be needed as the enrolled population will continue to shrink in the coming decades. To these quantitative changes there is a need to further develop and improve both pre-primary and higher education. In both these areas, education spending remains critical, not only to foster growth in the long-term, but also to attract and retain the young workers that the country need to finance its social system. Public spending in higher education remains very low in Latvia, as per student spending represents only 58 percent of what the older EU member states spend. 44 Finally, it is to be noted that the diversion of the funded pillar contributions to the public scheme, even temporarily, is likely to result in lower pensions for future retirees. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Satisfactory 95. The performance of both project teams during identification, preparation, and appraisal is rated satisfactory. First, the SDPL series supports the Government’s Employment and Social Safety Net Strategy, approved on October 1, 2009. The SDPL was been accompanied by an extensive and timely technical assistance program to help the Government in the design and monitoring of the ESSNS. The components of the SDPL series were designed in accordance with 43 Second pillar pensions system has been seized by governments during downturns but never replenished afterwards in Argentina and France for instance. 44 Here, the comparison is the 15 older members of the EU (EU15). 35 the findings of a very detailed PER undertaken by the World Bank in 2007 and updated in 2010, which examined options for fiscal consolidation and structural reforms in the social sectors and public administration. The design of the SDPL series also benefitted from the collaboration with the EC, the IMF, and other international donors allowing the policy actions to be aligned with the reform program covering the financial sector supported by the Government’s partners. Moreover, the SDPL series supplemented a Financial Sector DPL that was disbursed by the World Bank in 2010 to support the financial stabilization of the country after the 2008-2009 crisis. In addition, the World Bank team has accompanied government consultations with non- government organizations representing constituents affected by the fiscal consolidation and structural reforms. (b) Quality of Supervision Rating: Satisfactory 96. The operation was well supervised. Both operations were delivered on time, all prior actions were met allowing the second SDPL to start quickly and both disbursements occurred within the year. The two operations were also closed on schedule, at the end of fiscal years 2010 for the first loan and 2011 for the second. 97. The two SDPL series teams consisted of social sector economists, most of whom participated in the design and supervision of both SDPLs, which guaranteed consistency across the two phases of the program. Additionally, team members were involved in the update of the social sectors PER so that the supervision of the program benefited from their up to date knowledge of the details of expenditure in each sector. The dialogue with the Government during the program was continuous, allowing adjustments to the programs in response to the rapid changes in the economic prospects of Latvia from 2009 through 2011. The quality of the technical capacity of the World Bank team has been acknowledged by the Government, which now wishes to continue receiving technical assistance from the World Bank team on a FBS, in order to motivate reforms to policies targeted at the long-term unemployed and inactive in Latvia. 98. In addition, both teams were dedicated to liaison and collaboration with the IMF and the EC to ensure the communication consistency with the Government, to adjust the design of the program to the structural reforms that were supported by the other donors, and to share data and evaluations on the implementation of the SDPL series. (c) Justification of Rating for Overall Bank Performance 99. Overall, the World Bank team was well prepared and provided timely and relevant policy advice under the operations. The program benefitted from the expertise of the team, allowing the design, implementation, and monitoring of relevant and quantifiable results. 5.2 Borrower Performance (a) Government Performance Rating: Satisfactory 36 100. Performance of the Borrower during the preparation phase was satisfactory. Government bodies involved in the SDPL series included the Ministry of Finance, Ministry of Welfare, Ministry of Education and Science, Ministry of Health, Ministry of Transportation, and the Ministry of Regional Development and Local Government. 45 Other government agencies participating were the Local Governments Association and state limited liability company ―The Road Transport Administration‖, which were all involved in the SDPL series dialogue from the beginning. 101. The preparation and implementation of the SDPL series was facilitated by the strong commitment of the Government to push for reforms in the social sector. The ownership of the Government was a key factor of the success of the SDPL series from the start. Although the sector ministers changed during the program, the continuity of commitment to the program was ensured as the current Prime Minister remained in office throughout the program. 102. The Government also consulted with a large number of non-governmental organizations and civil society organizations on the content and implementation of the reforms. It is to be noted that the Government ensured transparency around the design and implementation of the ESSNS, first, by allowing the SAO to perform an independent audit of the Strategy, and second, by closely monitoring the results of the program and disclosing the related information to the public on the internet on a quarterly basis. All line ministries also published budget, strategy and monitoring documents on their webpage as well as details of consultations with stakeholders. (b) Implementing Agency or Agencies Performance Implementing Agency Performance Comments 1. Ministry of Finance Satisfactory The Ministry of Finance was in charge of the overall funding and the coordination of the SDPL series. All operations and prior conditions were met on time, and adequate and timely funding was allocated to the ESSNS. 2. Ministry of Welfare Satisfactory The Ministry of Welfare was responsible for the implementation of the Social Protection measures, along with the local authorities. All measures were implemented in time. It also had responsibility for monitoring the implementation of the ESSNS starting January 1, 2011. This monitoring was highly satisfactory. 3. Ministry of Education and Satisfactory The Ministry of Education and Science implemented enormous Science changes, such as per capita financing and creating incentives for localities to make tough allocation decisions in education, against strong opposition. However, consolidation of the school network has slowed down after this upfront adjustment. The role of the Ministry in the implementation of the education measures was limited as the local authorities are in charge of managing the funds for pre-primary education. The transportation measures to 45 This latter Ministry was restructured into the Ministry of Environmental Protection and Regional Development on January 1, 2011 37 Implementing Agency Performance Comments mitigate the effects of the per capita financing reform implementation were the responsibility of the Ministry of Regional Development and Local Government (from January 1, 2011, the Ministry of Environmental Protection and Regional Development). The Ministry of Education and Science was in charge of managing the ESSNS funds put aside to improve the curriculum and materials for pre-primary and primary schools. 4. Ministry of Health Highly The Ministry was responsible for the implementation of the Satisfactory measures in the health sector. Among all of the ministries, it was quickest to respond to the opportunity for structural reforms and considered the safety net an integral part of its reform program. Once the reforms were agreed, it also pursued them relentlessly and effectively. Initially surprised by the slow take-up of the program, the Ministry reacted fast by reallocating available funds along the way to pro-poor measures such as dropping the deductible for covered pharmaceuticals or by fully covering the cost of hospital treatment (not just the co-payments) for the poor. As is clear in the response of the government to this ICR, the MOH continues to develop the programs initiated during this crisis period because they were part of a larger strategy. 5. Ministry of Regional Satisfactory The Ministry was responsible for the implementation of the Development and Local transportation measures aiming at reducing the cost and time of Government (from January travel for the pupils whose schools have closed. Both objectives 1, 2011, the Ministry of of compensating the cost of transportation and keeping the travel Environmental Protection time below one hour were reached. and Regional Development) 6. Ministry of Regional Satisfactory The Ministry was in charge of the monitoring of the ESSNS prior Development and Local to a restructuring of government ministries that occurred on Government January 1, 2011. This monitoring was highly satisfactory. 7. Ministry of The Ministry was responsible for giving compensation to public Transportation transportation services providers for carrying passengers free of charge along regional intercity, regional local and urban routes according to the regulation of the Cabinet of Ministers No. 872. (c) Justification of Rating for Overall Borrower Performance Rating: Satisfactory 103. The Government owned the reform agenda from the start through playing a leading role in the design of the Social Sector Strategy. It managed to implement politically sensitive reforms, such as the consolidation of the schools and hospitals networks, in a tough economic environment. 6. Lessons Learned 104. Prioritizing and making trade-offs in social sector spending enabled the Government to take advantage of the need for consolidation to improve the long-term efficiency of public services. Although the government implemented across-the-board cuts in 2009, it also took the opportunity to target structural reforms with the high efficiency gains over the longer term, particularly in education and health. It also developed a successful set of emergency safety net measures that staved off disaster for the most vulnerable members of 38 society and gave the government experience in how such countercyclical programs can operate. These choices may have been necessitated by the depth of the crisis and a national consensus around the necessity to prepare for the adoption of the Euro, but the safety net may also have contributed to the government’s ability to maintain the consensus and to be reelected to office midway through the crisis. As social spending was relatively low as a share of GDP in Latvia, with coverage gaps in education, health, and social protection prior to the crisis, large cuts in these sectors were extremely risky. Nevertheless, improving the quality of expenditure and developing the safety net allowed some of the worst gaps to be reduced despite the lower spending. 105. Regular reviews of spending and outcomes in the social sectors enabled the program to be tailored and implemented very rapidly. In Latvia, the comprehensive spending review undertaken jointly by the World Bank and IMF allowed the Government to rapidly design a relevant program to respond to the crisis. An important lesson from Latvia was that investing in economic and social data analysis in good times and on a regular basis is all the more relevant in a time of crisis, especially given that the cost of such practice remains low. 106. The World Bank’s support through the SDPL series provided a framework for the Government’s policy reforms in the social sectors during a time of deep fiscal cuts. While the financial resources from the SDPL series were utilized by the Government, other available international and bilateral financing resources were not drawn upon. Hence, the Government did not face a large financing gap on the fiscal side and, in this regard, the SDPL series played a limited role in assisting the Government. The value of the SDPL series was the World Bank engagement it entailed in providing technical assistance on implementing the emergency safety net measures and in placing fiscal cuts and emergency measures in the context of the wider structural reform agenda in education, health, and social protection. 107. Reforming social insurance, especially pensions, during a crisis is very difficult. The reform experience in Latvia has shown that pension expenditures exhibit very high inertia in times of recessions and even extremely reform-minded governments are not able to quickly reduce spending on such programs. Pension spending actually increased by 27 percent in real terms over 2008-2010. Although the Constitutional Court ruling about the pension reform was very specific and related to the lack of consultation on allowing cuts in pensions, the Government did not pursue their proposed reform. 108. Reforming the social sectors during good times creates fiscal space to react during economic downturns. Contrary to Latvia, some countries among the EU12 started earlier to implement policies aiming at counteracting the demographic decline, which is large and certain: per capita financing and school consolidation in education, improved targeting of social assistance, increased retirement age, hospital consolidation and increased focus on primary care provision, among others. Taking early steps: (i) allows limiting the cuts needed when a crisis occurs as poorly targeted social insurance expenditure can boom during economic contraction, (ii) accelerates fiscal consolidation as structural reforms consume time and political capital 39 before actually providing fiscal savings, (iii) provides fiscal space through deficit and debt reduction allowing for counter-cyclical policy through discretionary policies, and (iv) prevents massive lay-offs of public workers at a time of crisis when increased unemployment would command the exact opposite policy. 109. Budget rules should take into account economic cycles. It is important to consider the trajectory of social protection expenditures—particularly unemployment benefits and social safety net support programs—across the economic cycle in designing fiscal rules and buffer- stock savings mechanisms. 110. Targets should not be overly-ambitious in projects designed during a period of high uncertainty. That certain intermediate targets were not met in health and social assistance suggests that the targets were overambitious given the fiscal space and the pace of implementation of health reforms. These targets were based on tentative projections, which happened to be too optimistic as often baseline data was not available. Setting such specific targets at the beginning of the program was risky given the context of large uncertainty as: (i) the project was implemented in the midst of a severe economic crisis which triggered a larger than expected fiscal contraction; and (ii) interventions were both new and never piloted before or involved the large scaling up from small programs. In such an environment, the targets for the outputs should have been defined as absolute figures, not as ratios, and using more conservative projections. 7. Comments on Issues Raised by Borrower 111. See Annex 2 40 Annex 1. Bank Lending and Implementation Support/Supervision Processes (a) Task Team Members P115732 - Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program Responsibility/ Names Title Unit Specialty Lending Mohamed Ihsan Ajwad Senior Economist ECSH4 Social assistance Charles C. Griffin Senior Adviser ECAVP Health Paulina Ewa Holda Country Economist ECSPE Country economist Kathy Lalazarian Senior Public Sector Specialist LCSPS Public sector reform Michael Mertaugh Consultant Lead Education Specialist ECSH2 Education Truman G. Packard Lead Economist EASHS TTL Stanislav Polak Country Economist ECSPE Country economist Alberto Rodriguez Country Sector Coordinator ECSH2 Overall Emilia Skrok Country Economist ECSPE Country economist Asta Zviniene Senior Social Protection Specialist ECSH3 Social insurance Anarkan Akerova Counsel LEGEM Lawyer Joseph Paul Formoso Senior Finance Officer CTRFC Finance Supervision Truman G. Packard Lead Economist EASHS TTL P121796 - Second Safety Net and Social Sector Reform Program Responsibility/ Names Title Unit Specialty Lending Anarkan Akerova Counsel LEGEM Lawyer Mohamed Ihsan Ajwad Senior Economist ECSH4 Social assistance Charles C. Griffin Senior Adviser ECAVP Health, education Joseph Paul Formoso Senior Finance Officer CTRFC Finance Carmen F. Laurente Senior Program Assistant ECSHD Support Nadezhda Lepeshko Junior Professional Associate ECSH4 Fiscal analysis Truman G. Packard Lead Economist EASHS Social protection Clelia Rontoyanni Public Sector Specialist ECSPS Public sector reform Emily Sinnott Senior Economist ECSH4 TTL Emilia Skrok Country Economist ECSPE Country economist Asta Zviniene Senior Social Protection Specialist ECSH3 Social insurance Supervision Emily Sinnott Senior Economist ECSH4 TTL 41 (b) Staff Time and Cost P115732 - Latvia First Special Development Policy Loan: Safety Net and Social Sector Reform Program USD (including travel and Stage No of staff weeks consultant costs) Lending 63.5 521,771 Supervision 0.1 27,558 P121796 - Latvia Second Safety Net and Social Sector Reform Program USD (including travel and Stage No of staff weeks consultant costs) Lending 40.7 213,894 Supervision 646 31,57847 46 This figure does not reflect the final cost of supervision. 47 This figure does not reflect the final cost of supervision. 42 Annex 2. Summary of Borrower’s ICR and Comments General Comments The Safety Net and Social Sector Reform Program First Development Policy Loan (hereinafter - First Social DPL) entered into force on March 12, 2010. The Safety Net and Social Sector Reform Program Second Development Policy Loan (hereinafter - Second Social DPL) entered into force on June 17, 2011. Both were part of an international financial support package of EUR 7.5 billion. Their objectives were to protect vulnerable groups with emergency safety net support during the economic contraction and fiscal consolidation measures and to ensure that structural reforms lay a foundation for medium-term improvements in the social sector. In this regard, the Government approved the Emergency Social Safety Net Strategy, which came into the force on October 1, 2009. The reforms were implemented by the Ministry of Finance (MoF), the Ministry of Welfare (MoW), the Ministry of Education and Science (MoES), the Ministry of Health (MoH), the Ministry of Transport (MoT), the Ministry of Regional Development and Local Governments (MoRDLG), which was restructured in the Ministry of Environmental Protection and Regional Development (MoEPRD) on January 1, 2011, as well as the Local Governments Association and the state limited liability company ―The Road Transport Administration‖. Activities to strengthen the social safety nets To deal with rising unemployment and a deteriorated social situation, the Government of the Republic of Latvia in its meeting on September 8, 2009 approved a Social Safety Net Strategy (hereinafter – the Strategy). Its purpose was to develop a batch of safety net measures to mitigate adverse effects resulting from: (1) the financial and economic crisis and (2) the structural reforms realized. We were committed to implement the social safety net activities into four areas: welfare, education, health and transportation. The Strategy was to begin on October 1, 2009 and to end on December 31, 2011, but could be extended if: (1) the situation in the social sector was to worsen by the end of 2011 or (2) the context in which the Strategy was implemented was to change considerably. The extension of the Strategy measures in 2012 was discussed by the Cabinet of Ministers in early November 2011, who decided to pursue the implementation of the measures in welfare, health and transportation. A further extension of the Strategy measures in 2013 will be discussed during the preparation process of the State Budget for 2013. Social safety net activities in the welfare domain - Municipal social assistance The GMI level was increased as of October 1, 2009, setting its level at LVL 45 per child and LVL 40 per adult per month. Moreover, in 2010, the criteria of eligibility of the status of needy person were change and some restrictions were removed. 43 The number of persons certified as needy increased from 176.1 thousands (7.8% of the population) in 2009 to 282.1 thousands (12.5% of pop.) in 2010 but decreased to 265.3 48 thousands (12.8% of pop.) in 2011. The total number of GMI benefit receivers increased from 62.8 thousands (2.8% of total population) in 2009 to 120.6 thousands (5.4% of total population) in 2010 and after- decreased to 121.8 thousands (5.9% of total population) in 2011. Although it was forecasted that the number of needy persons and GMI receivers will continue to increase in 2011, reports from municipalities show that those statistics started decreasing in 2011. As local government revenue seemed still insufficient to finance GMI and housing benefits, the state co-financed 50% of the GMI benefits and 20% of the housing benefit from 2009 until the end of 2011. As a result, the municipal spending for social assistance increased 49 from 31.1 million LVL in 2008 up to 45.4 million LVL in 2010 and to 52.7 million LVL in 2011. The government decided to prolong the state co-financing of the GMI until the end of 2012 and of the housing benefit until the end of April 2012. To improve social assistance targeting and reduce poverty traps, the Ministry of Welfare has started working on a comprehensive reform of social assistance and social support system which will come into force in 2013. - Unemployment insurance From July 1, 2009 the period during which unemployment benefits were paid was extended to nine month on a temporary basis. In addition, for persons with less than 20 years of insurance record, unemployment benefit was set to a fixed amount equalling 45 LVL per month. In 2010 extended unemployment benefit were granted and paid to 17,008 persons in average per month or 28.2% from the total number of unemployment benefit receivers. This percentage decreased to 22% in 2011. This decrease was influenced by the overall decrease of unemployment benefit receivers among the total number of registered unemployed persons. In 2010 the average duration of unemployment benefit payment was 5.32 months, which is approximately 60% from the period of unemployment benefit payment prescribed in the legislation, i.e. 9 months. This represents an increase by 0.88 months from 2009. In 2011 this duration decreased to 4.82 months (54% of the maximum legal). The extension of the unemployment benefit payment period ceased at the end of 2011. - Workplace with Stipend Emergency Public Works (WWS) program The Workplaces with Stipend program (WWS) has been managed by the State Employment Agency (SEA) and implemented by municipalities that were responsible for creating the workplaces. The stipend amounted to LVL 100 per month, although in the second half of 2011, in order to ensure a gradual transition to long-term measures and lower a financial attractiveness of the measure, the amount of the stipend was reduced till 80 LVL, thus 48 The 2011 data are estimates. 49 The source of data is the Annual statistical reports of municipality. 44 motivating the unemployed to search for a job and return to the labour market or to decide to take part in other active labour market measures. In total, 74,037 unemployed people took part in the WWS program during its implementation while 122,937 participations positions were created (45% of the beneficiaries participated more than once). 21.6% of all the participants were able to find a job within 6 months after leaving the WWS program. A new public works program has been launched taking into account the results of the impact evaluation of the WWS program. Its main features are: non-commercial lower qualified activities (state and municipal enterprises exempted from participation), the allowance is 100 LVL net per month, the non-allowance costs have to be covered by municipalities (including equipment, materials), two days per month (allowance not suspended) can be used for active job-search (if afterwards evidence is submitted of this) or short courses organized by the State Employment Agency. The general target mechanism is self-selecting (first-come first-served); municipalities have the right to set clear and non-discriminatory priority criteria if necessary, the maximum participation period is 4 months. With work creation remaining slow, in the private sector, especially for lower qualified jobs, priority will be given to improved and better targeted wage subsidy programs in the private sector to reduce the risks of replacement of existing employees and to channel the available resources to those most in need of assistance rather than to those who can find a new job without or with little assistance from the State Employment Agency. Wage subsidy schemes are under review. Matching and profiling activities of unemployed will be further developed at the SEA. Discussions are in progress to assess the needs and possibilities of organizing services to assist the inactive who cannot participate because of dependants, regional mobility or alcohol and drug addiction. - Second Pillar Pension Policy The Government has completed a concept paper on the long-run sustainability of social security forming the basis for a comprehensive pension reform. The Government reiterates its commitment to restore the contribution rate to the second pillar pension to 6% by 2013, if the budgetary situation improves as planned. Social safety net activities in education As of September 1, 2009, local governments did not have the resources to ensure pre-school education for five- and six-year-old children, the state covered through the strategy the cost of such mandatory free of charge pre-school education. Pre-school attendance rates have been maintained over the crisis, equalling 94% for five-year-old children and 97% for six-year-old children at the end of 2011. Simultaneously, efficient per capita financing reforms allowed reducing the financial burden. The financing has also been provided for the on-going improvement of the curriculum for pupils of age five to ten years. The Ministry of Education and Science has continued reforms that lead to efficiency and quality gains in education system and provided sufficient state subsidy for pre- school education in 2012. 45 Education system reform has led to school network optimization. During the period from 2009 to 2011, 77 schools were confirmed to be closed, 130 were reorganized and 14 new schools were founded. According to the data received from the Ministry of Education and Science in 2011/2012 school year 31 schools were closed or reorganized. To mitigate the effects of school closure, each local government has evaluated the possibilities and costs of transporting students to schools. Four different methods have been used to assist in the transportation of children: (1) purchase of school buses; (2) compensation of fares in public transportation, (3) compensation of transportation expenditure of local governments, or (4) purchase of transportation services from private providers. 209 school busses were purchased and delivered to 88 local governments. In 2010, 76% of students from reorganized and closed schools received transportation assistance, but this proportion increased to 88% in the 4th Quarter of 2010 and reached 91% in 2011. The time spent travelling to and from school exceeded 60 minutes (one way) for only 0.46% of the students in 2010 and 0.38% in 2011. From 2012 transportation of relocated is financed from the local government budget. The government added into the regulation of Cabinet of Ministers No. 740 related to the awarding of student’s stipends, social considerations, in addition to merit by issuing the Regulation of the Cabinet of Ministers No. 511 dated June 2, 2009. In accordance with the Constitutional court decision on August 9, 2011 the regulation on stipends was amended to prescribe that if two or more students have equivalent academic performance, the stipend first shall be granted to the persons who conforms to certain criteria, taking into account person`s social, family and financial situation. Social safety net activities in health Prior to October 1, 2009 patients who were certified as needy paid only 50% of the patient co- payments for health care visits. In 2010, needy patients were exempted from co-payments and from the second semester of 2010 payments were expanded to cover the entire cost of the provided service and not uniquely the co-payment. The full subsidies for the needy patients in outpatient and inpatient health care will remain in effect in 2012. These measures will be financed from the general health care budget. From February 3, 2010, the 50 LVL co-payments for medicine and medical devices have also been lifted for the patients whose per capita income did not exceed 120 LVL. In 2011 those compensations benefitted 30,267 patients. This measure is effective in 2012. Starting from January 1, 2012 expenses for hotel-type beds in hospitals will not be covered for needy and low-income persons anymore. Instead, free home care services are expanded: the chronically ill patients are treated at home, thus decreasing the need for in-patient and emergency care. Community-based mental health care has been developed in day care centers. This allows a more efficient use of resources by reducing the need of inpatient services and by taking care of patients 46 with mental disability closer to their living place ensures their integration in the society and decreases stigmatization. To alleviate the workload of general practitioners that resulted from the shift from in-patient to out-patient medical care, additional nurses have been hired covering approximately half of the primary health care practices. From the second semester of 2010, nurses were recruited for family doctors’ practices in cities as well. In 2010, a physician advisory telephone service was developed, to connect patients to a doctor after working hours and on weekends. This method extends the access to primary care services 24 hours a day across the country at a low cost. It is aimed at the chronically ill and emergency situation. Implementation started in the second quarter of 2011. Mental health care in day centers, home care, the additional nurses in family doctor’s practice and the physician advisory telephone service will be financed beyond 2011. Social safety net activities in the transportation domain The Regulation of the Cabinet of Ministers No. 872 ―Regulations on Passenger Categories to which Reduced Tariffs are Extended when Using Public Transport Services in the Rout Network‖ determines the categories of passengers, who have right to use public transport free of charge along regional intercity, regional local and urban routes. The said categories of persons are: (a) children of preschool age, (b) orphans and children without parental care, (c) first and second group of disabled persons, children with disabilities and persons who accompany first group disabled persons or disabled children, (d) politically repressed persons and members of National resistance movement. The Law on Disability defines that disabled persons with serious disabilities have the right to use free of charge all types of public transport in the territory of Latvia, except taxi services, aviation transport and transport of inland waterways. The passenger categories that are foreseen in the regulation do not pay for tickets in the public transport and they can utilise the said benefit by presenting special identity cards. The providers of the public transportation in accordance with the normative acts that regulate public transport services are entitled to claim the incurred losses from providing the aforementioned service from the state budget. Compensation of the said losses is foreseen as one of the activities of the strengthened social safety net for 2010 and 2011 to foster free movement of the said socially vulnerable groups of people. Road Transport Administration has summarized the information about passengers carried free of charge in regional intercity, regional local and urban routes and caused losses to public transport service providers in the year of 2011. The total sum of losses is calculated according to real losses monitoring the number of voyages of all the categories of passengers with fare reliefs in public transport. The total amount of financing assigned in the Strategy to public transport 47 service providers was 10.8 million LVL in 2011 and it covered 98 per cent of losses to public transport service providers in regional intercity, regional local and in urban route network. The largest group of passengers carried with fare reliefs in public transport was disabled persons of the first and second group of disability (60 per cent). According to statistics the number of voyages of disabled persons increased. In 2011 the number of passengers carried has increased for 3 per cent in regional intercity and regional local traffic, but in city traffic increase was 16 per cent in comparison to year 2010. Explanation is that in first half-year of 2010 there was not in place precise passenger registration in republic city transport. Monitoring and implementation of the social safety net activities The Government of the Republic of Latvia designated the Ministry of Regional Development and Local Governments as the institution responsible for monitoring of the Strategy implementation. Due to the reorganization of central government ministries the Ministry of Welfare was responsible for the supervision and co-ordination of the Strategy implementation starting from January 1, 2011. Monitoring reports were submitted to the Cabinet of Ministers every quarter. The responsible institutions carried out the assessment of the Strategy’s activities and the proposals concerning the Strategy measures continuation in 2012. This evaluation was discussed in the meeting of Cabinet of Ministers on November 2011. The Government of the Republic of Latvia highly appreciated the help and support of the World Bank in mitigating the impact of the financial and economic crisis and the structural reforms on the most vulnerable part of society. We are very happy that in June 2010 we have been honoured by the award from the World Bank for achieving outstanding results in improving lives of people in Europe and Central Asia. 48 Annex 3. Policy Matrix (Program of Two Special Development Policy Loans)50 Objective Loan 1 Loan 2 Outcome Prior Actions (bold) & Supporting Measures Prior Actions (bold) (Completed) Ensure adequate The Borrower’s Cabinet of Ministers The Borrower, through the national and local Emergency safety net emergency safety net approved an Emergency Social Safety Net government agencies, has been satisfactorily measures are deployed and financing for education Strategy for the period from October 1, 2009 implementing the Emergency Social Safety remain in place through and health services, and through December 31, 2011 on September 8, Net Strategy as indicated by the maintenance 2011. social protection programs 2009 (Protocol No. 56, 78§). of adequate budget financing in 2010 and Information on to mitigate the impact of 2011, and evidenced by regular performance government’s fiscal adjustment and the monitoring reports submitted to the Cabinet implementation of ESSNS economic contraction on of Ministers and publishing them on the web and structural reforms is households. page of the Ministry of Welfare. independently assessed and publicly available. Monitor performance of The Borrower has published the Emergency The Borrower, through the Auditor General, The independent Audit national and local Social Safety Net Strategy on the internet and has conducted on October 28, 2010 an Report on ESSNS government agencies in in local media on December 21, 2009. independent audit of the Cabinet of Minister's implementation is implementing ESSNS & The Borrower has provided adequate resources performance in implementing the Emergency completed and published increase public awareness, for and put in place a performance monitoring Social Safety Net Strategy. on the internet in October transparency and and impact evaluation plan, making use of 2010, and the action plan accountability of existing monitoring mechanisms, administrative of audit recommendations government measures in data, and surveys. is put in place by October the social sectors. 2011. 50 This is the policy matrix as it had evolved at the time of the SDPL2 Board presentation (April 2011). 49 Social protection Objective Loan 1 Loan 2 Outcome Prior Actions (bold) & Supporting Measures Prior Actions (bold) (Completed) W1. Ensure social The Borrower enacted an amendment to the The Borrower has enacted an amendment to Percentage of insurance system is Law “On Insurance in Case of the Law on the State Funded Pensions which unemployment benefit responsive to households Unemploymentâ€? effective as of July 1, 2009, increased the contribution rate to the funded recipients whose benefits needs which extends the pay-out period of pension pillar to 6 percent of a worker's salary are extended by the reform unemployment benefits to 9 months, effective starting from January 1, equals 22% in 2011. until December 31, 2011, which, inter alia, 2013. Individual contributions to provides for a decrease in the eligibility the funded pension pillar requirement for unemployment benefits to 9 have risen from 2% in 2009 months of contributions in the previous 12 to 6% from January 1, months. 2013. W2. Strengthen The Borrower: (a) enacted an amendment to The Borrower has allocated at least 50 percent Municipalities have mandatory, municipally the Regulations of the Cabinet Ministers No. of the cost of the guaranteed minimum income sufficient financial administered, mean-tested 1070 dated October 1, 2009 and No. 1489 in the 2011 budget, with local governments resources to make timely social assistance to dated December 1, 2009 to increase the funding the other 50 percent. payment of GMI benefit to mitigate the impact of the Guaranteed Minimum Income (GMI) benefit all eligible individuals. crisis on the poorest amount to LVL 40 per adult and LVL 45 per households. child in a household; and (b) allocated financing in the 2010 budget of, at least, 50 percent of the cost of the GMI from state funds, with local governments funding the other 50 percent W3. Strengthen active The Borrower introduced amendments to the In 2011, on average 16,500 employment measures to Cabinet of Ministers Regulation No. 166, people will be enrolled in respond to increased dated July 14, 2009 to provide temporary the emergency public demand for assistance employment for the unemployed not covered works program. from unemployed people by unemployment insurance effective as of who do not get August 8, 2009 unemployment insurance 50 Health Objective Loan 1 Loan 2 Outcome Prior Actions (bold) & Supporting Measures Prior Actions (bold) (Completed) H1. Reduce the risks The Borrower provided for an additional The Borrower has strengthened primary Increased hours and associated with fiscal public health nurse for approximately half of health care services program in the 2010 and outreach services at the consolidation in the health the general practitioners and primary health 2011 budget by expanding the additional Primary Health Care level sector by strengthening care providers in the emergency social safety public health nurse or physician’s assistant to compensate in part for General net and financed it fully in the 2010 budget program to primary care practices and reduced financing of Practitioner/Primary pursuant to the Cabinet of Ministers developing a family physician advisory specialist and hospital care. Health Care to partially Regulation No. 1630 dated December 22, telephone service. The share of GPs with an substitute for reduced 2009. extra public health nurse access to specialist and increased to 36% in 2011 hospital-based care. (from 0% in 2009). H2. Augment financial The Borrower allocated in the 2010 budget The Borrower has implemented measures to: People from households protection from the cost of funds for households with per capita income (a) strengthen the exemption program for certified as needy are health care, and maintain less than half the minimum wage, at adequate health payments by eliminating all out-of exempt from health service critical health services for levels, to finance: (a) an exemption from co- pocket health payments for visits, tests, and co-payments; are receiving needy households. payments for general practitioner, outpatient, pharmaceuticals for patients classified as subsidized prescriptions; specialist, and inpatient services; (b) an "needy"; (b) provide a limited package of and retain access to critical exemption from pharmaceutical charges benefits for patients as their incomes exceed health care services. above LVL 50 per year per person; and (c) an the "needy" line; and (c) provide outpatient exemption from the cost of overnight hotel psychiatric and home care for chronic diseases stays in hospitals, home care for those with without cost to all the population of Latvia. serious diseases and day-care centers for those with mental diseases. 51 Education Objective Loan 1 Loan 2 Outcome Prior Actions (bold) & Supporting Measures Prior Actions (bold) (Completed) E1. Ensure access to The Borrower allocated funds in the 2010 The Borrower has allocated in the 2011 budget Enrollment of five- and preprimary education and budget for pre-primary education/child adequate financing for pre-primary six-year-old children in child development development programs for children starting 5 education/child development programs for free pre-primary programs programs is maintained years of age. children starting from the age of 5 years old. is maintained at least at 94% for 5 years old, and 98% for six year olds in 2011. E2. Ensure access of The Borrower allocated in the 2010 budget The Borrower has allocated in the 2011 budget The time students spend primary and secondary funds to cover the cost of transportation of adequate financing to cover the cost of travelling to and from school-aged children to students whose schools have closed to transporting to replacement schools students primary and secondary education facilities is replacement schools. whose schools have closed. school is kept within 60 maintained during minutes (one way). implementation of the per Share of children from capita financing model closed schools receiving assistance for school transportation under the State-subsidized scheme reaches at least 77% in 2011. E3. Improve equity of The Borrower added into the Regulation of Not less than 50% of state subsidies for higher the Cabinet of Ministers No. 740, dated subsidies to cover the education August 24, 2004 social considerations, in stipends of students in addition to merit, to the awarding of state higher education are support to students in higher education by awarded to academically issuing a Regulation of the Cabinet of eligible students on the Ministers No.511 dated June 2, 2009.49 basis of poverty and social considerations in addition to merit in 2011. 52 Annex 4. Status of SDPL1 and SDPL2 Prior Actions from the Legal Agreements SDPL1 Prior Actions51 SDPL1 SDLP2 Prior Actions52 SDPL2 status status Policy action 1 Policy action 1 An Emergency Social Safety Net Strategy has MET The Borrower, through the national and MET been approved for the period October 1, 2009 On local government agencies, had been The indicators through December 31, 2011. September Satisfactorily implementing the Emergency monitoring the 8, 2009 Social Safety Net Strategy as indicated by implementation (Protocol the maintenance of adequate budget of the strategy No. 56, financing in 2010 and 2011, and evidenced have been 78§). by regular performance monitoring reports released on a submitted to the Cabinet of Ministers and quarterly basis. publishing them on the web page of the Ministry of Welfare. Policy action 2 Policy action 5 The Borrower provided for an additional MET The Borrower has strengthened the primary MET. public health nurse for approximately half of the health care services program in the 2010 This service the general practitioners and primary health Cabinet of and 2011 budget by expanding the started care providers in the emergency social safety Ministers additional public health nurse or physician’s operating in net and financed it fully in the 2010 budget. Regulation assistant program to primary care practices May 2011 and No. 1630 and developing a family physician advisory received 5,913 dated telephone service. calls in March December 2012. 22, 2009. Policy action 3 Policy action 6 The Borrower allocated in the 2010 budget MET The Borrower has implemented the MET funds for households with per capita income measures to: (a) strengthen the exemption Amendment of less than half the minimum wage, at adequate program for health payments by eliminating the Cabinet of levels, to finance: (a) an exemption from co- all out of-pocket health payments for visits, Minister's payments for general practitioner, outpatient, tests, and pharmaceuticals for patients Regulations specialist, and inpatient services; (b) an classified as "needy"; (b) provide a limited #1046, dated exemption from pharmaceutical charges package of benefits for patients as their December 19, above LVL 50 per year per person; and incomes exceed the "needy" line; 2006. (c) an exemption from the cost of overnight MET and (c) provide outpatient psychiatric and MET hotel stays in hospitals, home care for those home care for chronic diseases without cost Amendment of with serious diseases and day-care centers for to all the population of Latvia. the Cabinet of those with mental diseases. Minister's Regulations #1046, dated December 19, 2006. Policy action 4 Policy action 7 The Borrower allocated funds in the 2010 MET In the 2011 budget, the Borrower has MET budget for pre-primary education/child allocated adequate financing for pre- development programs for children starting 5 primary education/child development years of age. programs for children from the age of 5 years old. 51 As displayed in the SDPL1 loan agreement. 52 As displayed in the SDPL2 loan agreement. 53 SDPL1 Prior Actions SDPL1 SDLP2 Prior Actions SDPL2 status status Policy action 5 Policy action 8 The Borrower allocated in the 2010 budget MET In the 2011 budget, the Borrower has MET funds to cover the cost of transportation of allocated adequate financing to cover the students whose schools have closed to cost of transporting students whose schools replacement schools. have closed to replacement schools. Policy action 6 The Borrower added into the Regulation of MET the Cabinet of Ministers No. 740, dated Regulation August 24, 2004, social considerations, in of the addition to merit, to the awarding of state Cabinet of support to students in higher education. Ministers No. 511 dated June 2, 2009. Policy action 7 Policy action 3 The Borrower enacted an amendment to the MET The Borrower has enacted an amendment to MET Law "On Insurance in Case of On July 1, the Law on the State Funded Pensions On December Unemployment" effective as of July 1, 2009, 2009. ("Grozijumi Valsts fondcto pensiju likuma" 30, 2010. which extends the pay-out period of dated December 30, 2010 ("LV", 206 unemployment benefits to 9 months, effective (4398), 30.12.2010)) which increased the until December 31, 2011, which , inter alia, contribution rate to the funded pension provides for a decrease in the eligibility pillar to 6 percent of a worker's salary requirement for unemployment benefits to 9 starting from January 1, 2013. months of contributions in the previous 12 months. Policy action 8 Policy action 4 The Borrower: (a) enacted an amendment to MET In the 2011 budget, the Borrower has MET the Regulations of the Cabinet Ministers No. financed at least 50% of the cost of the GMI 1070 dated October 1, 2009 and No. 1489 from State funds, with local governments dated December 1, 2009 to increase the GMI funding the other 50%. benefit amount to LVL 40 per adult and LVL 45 per child in a household; and (b) allocated financing in the 2010 budget of, at least, 50% of the cost of the GMI from state funds, with local governments funding the other 50 %. Policy action 9 The Borrower introduced amendments to the MET Cabinet of Ministers Regulation No. 166, On August dated July 14, 2009 to provide temporary 8, 2009. employment for the unemployed not covered by unemployment insurance effective as of August 8, 2009. Policy action 10 Policy action 2 The Borrower has published the Emergency MET The Borrower, through the Auditor General, MET Social Safety Net Strategy on the internet and On has conducted an independent audit of the On October 28, in local media. December Cabinet of Minister's performance in 2010. 21, 2009. implementing the Emergency Social Safety Net Strategy. 54 Annex 5: Technical Assistance Program Accompanying SDPL53 Technical Assistance activity Deliverable Poverty and social impact (PSIA) of Latvia’s economic contraction: Simulation PSIA for SDPLs 1 analysis of selected macroeconomic and social policy scenarios, taking into account and 2;Technical government responses and household coping strategies using the crisis response survey note module added into the quarterly Labor Force Survey. World Bank Staff initiated contact with the Central Bureau of Statistics (CBS) to request the addition of a Crisis Response Survey module to the continuous Labor Force Survey, based on other Crisis Response Surveys that the World Bank (Human Development ECA region) has led in other crisis- affected countries. Analysis of the impact of the economic contraction on the labor market: The note PSIA for SDPLs 1 focuses on the victims of the adjustment and the channels by which they were affected. and 2 Review of revenue and social spending at the municipal level: including a review of the Technical note social sector spending at the local level and a review of Latvia’s municipal revenue equalization structures, covering municipal spending on wages and social service delivery. Review of the International best practice for design, roll out and monitoring of Technical note emergency safety net programs: to ensure essential social services during crises (all social sectors). The note includes international best practice in extension of General Practitioner and primary health care services, including costing of several proposals. Comparative analysis of the earnings of teachers and medical workers: using Technical paper European and international benchmarks. PER update on the education sector: Review of budget reductions; assessment of local PER-style and government school network rationalization decisions under new financing model; format technical development of new career concept, performance standards, and compensation scheme for note the teaching profession consistent with ―output-based‖ financing model; support for developing a more modern results-based approach for vocational education; greater progressivity in distribution of university student financing. PER update on the health sector: Review of budget reductions; assessment of PER-style and implementation of hospital bed/network consolidation and other health sector structural format technical reforms to treatment incentives. note PER update on social protection: Social insurance and social assistance review of PER-style and budget reductions; simulations of pension reforms to date, and likely further measures format technical (specifically, retirement age increase) and restoration of contributions to the funded pillar note of individual accounts; spending, distribution and administration of universal social assistance benefits and income-tested social assistance benefits since increase safety net reforms (updated survey analysis); assessment of targeting errors (type 1 and 2) and administrative issues in municipally administered social assistance. PER update on public administration: Review/assessment of public sector wage PER-style and reductions and dismissals; consolidation of administrative functions; likely impact of format technical establishment control and civil service compensation reform measures. note 53 Source: SDPL1 Program Document. 55 22°E 23°E 24°E 25°E 26°E L AT V IA ESTONIA MAIN CITIES AND TOWNS REPUBLICAN CITIES LATVIA NATIONAL CAPITAL To RIVERS Nomme MAIN ROADS To To Viljandi RAILROADS Haademeeste 58°N MUNICIPALITY (NOVADI) BOUNDARIES Rujiena To Ainazi Tartu INTERNATIONAL BOUNDARIES Salacgriva Valka Kolka Aloja Mazirbe Renceni To Pskov 28°E Gulf of Dunte Dundaga Ape Roja Riga Limbazi Valmiera Baltic Ventspils Pope Smiltene Aluksne Liepa Sea Ugali Mersrags R US S I AN Cesis Piltene Talsi Saulkrasti Engure Vilaka F E D. Stende Jaunpiebalga Gulbene Balvi Sigulda 57°N Kandava Bolderaja Vangazi To 57°N Alsunga Kuldiga Tukums Jurmala Malpils Ostrov Pavilosta Ropazi Ergli RIGA Salaspils To Gaizinkalns Madona Ogre Lubanas Karsava Krasnogorodskoye Olaine (312 m) Aizpute Kalnciems Ezers Skrunda Broceni Baldone Lielvarde Barkava D Jelgava au Saldus ga Skiveri Plavinas Dobele Vecumnieku va Varaklani Liepaja Grobina Iecava Jaunjelgava Aizkraukle Ludza Atasiene Vilanu Auce Jekabpils Rezeknes Priekule Vainode Ezere Bene To Nica Eleja Skaistkaine Livani Raznas Zilupe Idritsa Bauska Ezers Da To Viesite Malta ugav Telsiai Preili a To Rucava Siauliai To Nereta Panevezys Akniste To Aglona Birzai Dagda 56°N To 56°N Kretinga Subata Ilukste To Kraslava Rokiskis To Novopolotsk Daugavpils Piedruja To Dusetos LITHUANIA This map was produced by the Map Design Unit of The World Bank. To BELARUS The boundaries, colors, denominations and any other information Dukstas shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any 0 20 40 60 Kilometers IBRD 33432R endorsement or acceptance of such boundaries. APRIL 2012 21°E 0 20 40 Miles 22°E 23°E 24°E 25°E 26°E