Document of The World Bank IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-47910) ON A LOAN

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-47910) Human Development Sector Unit Turkey Country Unit Europe and Central Asia Region ON A LOAN IN THE AMOUNT OF EURO 360 MILLION (US$ MILLION EQUIVALENT) TO THE REPUBLIC OF TURKEY FOR A SECOND PRIVATIZATION SOCIAL SUPPORT PROJECT May 12, 2010 Report No: ICR

2 CURRENCY EQUIVALENTS (Exchange Rate Effective as of March 2, 2010) Currency Unit = Turkish Lira (TL) TL 1.00 = US$ US$ 1.00 = TL FISCAL YEAR January 1-31 December CAS CPS EKA EU FMS FOM IBRD ISKUR ISR JLC KOSGEB LAG LRS M&E MME NGO PA PAD PCU POM QAG SBA SOE PSSP TA TAC TCEP TPSE VTSE ABBREVIATIONS AND ACRONYMS Country Assistance Strategy Country Partnership Strategy General Directorate of Economic Research (Treasury) European Union Financial Management System Field Operational Manual International Bank for Reconstruction and Development Turkish Employment Organization Implementation Status and Results Report Job Loss Compensation Small and Medium Industry Development Organization Labor Adjustment Group Labor Redeployment Services Monitoring and Evaluation Management, Monitoring and Evaluation Non-Governmental Organizations Privatization Administration Project Appraisal Document Project Coordination Unit Project Operational Manual Quality Assurance Group Small Business Assistance Services State-Owned Enterprise Privatization Social Support Project Technical Assistance Tripartite Advisory Committee Temporary Community Employment Programs Training for Prospective Self Employers Vocational Training Programs for Prospective Self Employers Vice President: Country Director: Sector Manager: Project Team Leader: ICR Team Leader: Philippe H. Le Houerou Ulrich Zachau Kathy Lindert Jesko S. Hentschel Carla Pittalis

3 TURKEY SECOND PRIVATIZATION SOCIAL SUPPORT PROJECT CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph Page 1. Project Context, Development Objectives and Design Key Factors Affecting Implementation and Outcomes Assessment of Outcomes Assessment of Risk to Development Outcome Assessment of Bank and Borrower Performance Lessons Learned Comments on Issues Raised by Borrower/Implementing Agencies/Partners Annex 1. Project Costs and Financing Annex 2. Outputs by Component Annex 3. Economic and Financial Analysis Annex 4. Bank Lending and Implementation Support/Supervision Processes Annex 5. Beneficiary Survey Results Annex 6. Stakeholder Workshop Report and Results Annex 7. Summary of Borrower s ICR and/or Comments on Draft ICR Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Annex 9. List of Supporting Documents MAP IBRD 33501

4 A. Basic Information Country: Turkey Project Name: Privatization Social Support Project 2 Project ID: P L/C/TF Number(s): IBRD ICR Date: 05/12/2010 ICR Type: Core ICR Lending Instrument: SIL Borrower: TREASURY Original Total Commitment: USD 465.4M Disbursed Amount: USD 442.2M Revised Amount: USD 421.9M Environmental Category: C Implementing Agencies: Privatization Administration Turkish Employment Organization (ISKUR) Small and Medium Industry Development Organization (KOSGEB) Cofinanciers and Other External Partners: B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 02/17/2005 Effectiveness: 01/31/2006 Appraisal: 04/04/2005 Restructuring(s): 03/28/ /13/ /27/ /29/ /20/2009 Approval: 06/14/2005 Mid-term Review: 06/29/2007 Closing: 06/30/ /31/2009 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Borrower Performance: Moderately Satisfactory Low or Negligible Moderately Satisfactory Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Moderately Satisfactory Quality of Supervision: Moderately Unsatisfactory Implementing Agency/Agencies: Moderately Satisfactory i

5 Overall Bank Performance: Overall Borrower Moderately Satisfactory Performance: Moderately Satisfactory C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Performance (if any) Potential Problem Project No at any time (Yes/No): Problem Project at any time (Yes/No): DO rating before Closing/Inactive status: No Satisfactory Quality at Entry (QEA): Quality of Supervision (QSA): Rating Moderately Satisfactory None D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 3 3 Compulsory pension and unemployment insurance General education sector 1 1 General industry and trade sector 1 1 Other social services 1 1 Theme Code (as % of total Bank financing) Education for the knowledge economy Improving labor markets Small and medium enterprise support Social analysis and monitoring Social risk mitigation E. Bank Staff Positions At ICR At Approval Vice President: Philippe H. Le Houerou Shigeo Katsu Country Director: Ulrich Zachau Andrew N. Vorkink Sector Manager: Kathy A. Lindert Hermann A. von Gersdorff Project Team Leader: Jesko S. Hentschel John A. Innes ICR Team Leader: ICR Primary Author: Carla Pittalis Carla Pittalis ii

6 F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The development objective of PSSP II was to support the Government's privatization program through mitigating the negative social and economic impact of the privatization of State-Owned-Enterprises (SOE). Revised Project Development Objectives (as approved by original approving authority) Project Development Objectives were not revised. (a) PDO Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Productivity of former State-Owned Enterprises (SOE) will increase when Indicator 1 : privatized due to reduction in labor costs. Data on sample companies suggests Value that privatization quantitative or n.a. n.a. led to efficiency Qualitative) gains in privatized SOEs. Date achieved 06/30/ /30/ /31/2009 Largely achieved. No assessment of productivity levels in privatized SOEs was Comments conducted during project implementation. The ICR team thus conducted a (incl. % background analysis of two major SOEs whose job loss compensations were achievement) financed under PSSP II. Workers displaced during privatization receive temporary income support (Job Indicator 2 : Loss Compensation) to help prevent them from falling into poverty. Value quantitative or Qualitative) 0 29,000 workers 13,614 workers received JLC payments (13,535 received Severance and Special job Loss Compensation payments; 79 received social assistance payments for civil servants). Date achieved 06/30/ /30/ /31/2009 Comments (incl. % achievement) Largely achieved. All workers displaced from the 20 SOEs privatized within the scope of PSSP II received job loss compensation (JLC) payments. The lower number of actual beneficiaries is a result of overestimation at appraisal. Workers displaced by the privatization and economic reform program receive Indicator 3 : Labor Redeployment Services (LRS) to assist them in returning to the labor force. iii

7 Value quantitative or Qualitative) 0 11,000 9,420 participants in ISKUR training programs, 8,546 completed them, 4,896 employed; 5 SBIs established, generating 1,086 jobs from existing tenants; 1,536 participants in Small Business Assistance Programs generating 210 start-ups employing 332 people. Date achieved 06/30/ /30/ /31/2009 Comments (incl. % achievement) Partially achieved. Actual beneficiaries did not include workers displaced from privatized State Owned Enterprises, nor secondary lay-offs, but comprised the unemployed population at large (see Section 3.2 in the main text). (b) Intermediate Outcome Indicator(s) Indicator Indicator 1 : Value (quantitative or Qualitative) Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Productivity of elements of industrial sector, containing about 21 former State- Owned Enterprises (SOE), is calculated to improve as a result of labor shedding during Privatization. n.a. 10% annual increase Data on sample companies suggests that privatization led to efficiency gains in privatized SOEs. Date achieved 06/30/ /30/ /31/2009 Comments (incl. % See comment under PDO Indicator 1. achievement) Indicator 2 : About 29,000 workers displaced from SOEs receive severance payments, and poverty has been mitigated as indicated through results of social impact. Value (quantitative or Qualitative) No payments. 29,000 workers displaced from SOEs receive payments, and poverty has been mitigated as indicated through 13,614 workers received JLC payments (13,535 received Severance and Special job Loss Compensation payments; 79 iv

8 results of social impact. received social assistance payments civil servants). Date achieved 06/30/ /30/ /31/2009 Comments (incl. % Largely achieved. See comment under PDO Indicator 2. achievement) Indicator 3 : A range of services are delivered to approximately 11,000 displaced workers with job placement rates as follows: Job Counseling: Job Counseling: 10%; On-the-Job 11%; Training: 70%; On-the-Job Institutional Training: 78%; Training: 50%; Institutional Vocational Training: 50%; training: 20%; Vocational training: Temporary Value 25%; Temporary Community (quantitative 0 Community Employment: or Qualitative) Employment: 16%; 10%; Small Small Business Business Assistance: 14% Assistance: 10% startup; Incubators: startup; 90% business Incubators: 90 % survival rate over business survival duration of rate over duration contract. of contract. Date achieved 06/30/ /30/ /31/2009 Comments (incl. % Fully achieved. Placement rates same or better than originally targeted. achievement) The Government evaluates the impact of privatization on key affected groups Indicator 4 : (including by sex, age, education level, region) and devises appropriate policy responses; designs criteria for LRS to focus on most effective services for different client groups Value (quantitative or Qualitative) n.a. n.a. No evaluation conducted, following an amendment to the Loan Agreement (see Comments section below) Date achieved 06/30/ /30/ /31/2009 Not achieved. The MME component, under which the impact evaluation was to Comments be conducted, was amended to exclude all the originally planned M&E studies. (incl. % As a result, these were not carried out, with the exception of a background report achievement) on SBIs. Indicator 5 : Project objectives are achieved, and disbursements are on schedule. Value n.a. 5 Project objectives v

9 (quantitative or Qualitative) were largely achieved and loan funds disbursed almost in full. Date achieved 06/30/ /30/ /31/2009 Comments Largely achieved. Overall, Euro million, corresponding to 92 percent of (incl. % loan funds, were disbursed. achievement) G. Ratings of Project Performance in ISRs No. Date ISR Archived DO IP Actual Disbursements (USD millions) 1 08/17/2005 Satisfactory Satisfactory /19/2006 Satisfactory Satisfactory /11/2007 Satisfactory Highly Satisfactory /13/2007 Satisfactory Satisfactory /26/2008 Satisfactory Satisfactory /08/2009 Moderately Satisfactory Moderately Satisfactory /02/2009 Moderately Satisfactory Moderately Satisfactory /13/2010 Moderately Satisfactory Moderately Satisfactory H. Restructuring (if any) Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO IP Amount Disbursed at Restructuring in USD millions 03/28/2006 N S S /13/2008 N S S /27/2008 N S S /29/2009 N MS MS Reason for Restructuring & Key Changes Made Amendment to Loan Agreement: Management, Monitoring & Evaluation (MME)Component revised Amendment to Loan Agreement: increase in disbursement percentage for Job Loss Compensation component POM amendment: Temporary Employment Workers with annualized monthly wages less than 1.2 gross minimum wages made eligible for JLC reimbursements under PSSP II Partial loan extension for Category 4, Consultants and Training vi

10 Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO IP Amount Disbursed at Restructuring in USD millions Reason for Restructuring & Key Changes Made 07/20/2009 N MS MS Euro 15 million cancellation I. Disbursement Profile vii

11 1. Project Context, Development Objectives and Design The Second Privatization Social Support Project (PSSP II) was approved by the Board on June 14, The Loan Agreement was signed on October 24, 2005 and the Loan became effective on January 31, The Project was the second World Bank-financed investment operation supporting the privatization process in Turkey Context at Appraisal Country and sector background: Less than five years before PSSP II was approved, the Turkish economy had suffered a serious crisis with devastating consequences, including a devaluation of the currency, a jump in nominal interest rates to 100 percent, the virtual collapse of the banking system, and the bankruptcy of scores of enterprises. In the aftermath of the 2001 crisis, authorities announced a New Economic Program, which included stabilization measures based on strong fiscal discipline, a structural reform agenda, and a renewed commitment to privatization. The combination of these interventions and the political stability that characterized Turkey since the 2002 elections helped the economy rebound from that crisis. During the period, Turkey enjoyed six years of strong economic growth, averaging 6.8 percent annually, and was set on a path of continued robust growth. Privatization revenues between 2003 and 2009 amounted to US$30 billion, four times the total revenues generated in the previous two decades (see Table 1). Supported under PSSP II, privatization included major companies of national importance, such as Petkim (petro-chemicals), Tupras (oil refinery), Tekel (tobacco and salt) and Seker (sugar). The global economic and financial crisis that began in September 2008 changed Turkey s short-term outlook considerably: output and trade fell substantially, unemployment rose, incomes fell, and poverty increased. Table 1: Privatization Revenues (US$ billion) in Turkey, Source: Privatization Administration 1 PSSP I was approved on December 21, 2000 and became effective on December 22,

12 Rationale for Bank assistance: PSSP II was prepared at the request of the Government and was directly linked with Turkey s overall economic reform program, which envisions a structural transformation of the economy through the reduction of state involvement and advancement of market reforms. In addition to enhancing competitiveness, the Government s privatization program aims to make a sizeable contribution to both fiscal management and public sector reform. Nonetheless, as in many countries, public perceptions about privatization in Turkey are frequently negative. The 2000 Country Assistance Strategy addressed the social effects of privatization as one of the political risks of the ambitious reform program initiated by the Government. In the midst of Turkey s financial crisis, which culminated in February 2001, fiscal pressure made it important for the Government to seek support for its privatization program through the financing of severance payments. A previous Bank-financed project, PSSP I, had successfully financed job loss compensation payments for more than 18,000 workers displaced from privatized State- Owned Enterprises. More importantly, the Project helped the Privatization Administration design a broader framework to mitigate the negative economic and social effects of privatization, including an awareness-raising mechanism for affected workers about post-privatization benefits and programs available to them. Such framework was maintained throughout PSSP II (see Section 3.2). 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) PSSP II substantially incorporated the mechanisms developed for the earlier PSSP I loan. The Project s development objective was to support the Government's privatization program through mitigating the negative social and economic impact of the privatization of State-Owned Enterprises (SOE). Progress towards this objective was to be measured and monitored based on the following key indicators identified in the Project Appraisal Document (PAD): The productivity of elements of the industrial sector, comprising about 21 former SOEs, improves as a result of labor shedding, workers displaced from SOEs. Workers receive severance payments, and poverty is mitigated as indicated through results of social impact assessments; Labor Redeployment Services (LRS) are delivered to both workers made redundant directly from within the SOEs and to workers suffering unemployment from the secondary employment effects. Job placement rates are equal to or better than similar programs in Turkey and the region; and The Government evaluates the impact of the privatization program and labor redeployment services, and identifies policy alternatives. Evaluation studies are undertaken and Project objectives are achieved, disbursements are on schedule, and annual Project audits are satisfactory. The Loan Agreement and the PAD also included a provision that a maximum of 210 million of the loan could be used to retroactively reimburse 81 percent of job loss compensation payments made to workers displaced from privatizations after January 1,

13 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification Neither the PDO nor the key performance indicators changed during Project implementation. The third key performance indicator, however, became partially irrelevant following a loan amendment, shortly after loan effectiveness, which excluded the originally planned evaluation studies from the Project (see Section 1.6). 1.4 Main Beneficiaries The PAD described the main beneficiaries of the Project as: (i) Workers displaced from SOEs being privatized, and their families. These workers would benefit from job loss compensations to help prevent them from falling into poverty. (ii) Workers affected by collateral displacement in firms associated with State-Owned Enterprises (SOE) where downsizing occurs, in particular in mono-enterprise communities, and the general economic reform program. Displaced workers would receive Labor Redeployment Services (LRS) to assist them in returning to the labor force. In practice, 13,614 workers, representing the totality of displaced workers from the 20 SOEs privatized within the scope of PSSP II (including workers displaced after January 1, 2003, as specified in the Loan Agreement and PAD) received job loss compensation payments. This compares with the original Project target of 29,000 workers, whereby the lower number of actual beneficiaries (47 percent of appraisal estimate) is not a reflection of weak Project performance, but rather of initial overestimations. This resulted from the Privatization Administration s double-counting of Job Loss Compensation (JLC) and Special Job Loss Compensation (SJLC) beneficiaries, whereas these payments are made to the same individuals (see Table 9). Although the PAD described beneficiaries of Labor Redeployment Services (LRS) as those primarily affected by the privatization process, including secondary lay-offs, no such targeting of services was applied. In practice, the entire adult population was eligible for LRS and more so as a result of a roll-out of ISKUR s training programs at the national level in 2007, against an initial target of 36 provinces. These had been selected on the basis of SOE privatizations that occurred since January 2003, and that were expected to occur during the implementation of PSSP II (see Section 3.2). Although the increase in the number of beneficiary provinces was in line with the PAD (which states that provinces where LRS would be delivered would be determined by the Privatization Administration) this choice had the effect of further expanding the LRS target. 1.5 Original Components (as approved) 3

14 The Project consisted of three components: (a) Job Loss Compensation (Cost at appraisal 2 : million; Actual disbursement: million): This component financed severance, special job loss compensation (JLC), and retirement social assistance payments to workers displaced by job loss due to the privatization of SOEs. The Privatization Administration (PA) was reimbursed by loan funds for the job loss compensation payments provided to: (i) SOEs making regular severance payments; (ii) the Turkish Employment Organization (ISKUR), which would make special job loss compensation payments; and (iii) the State Pension Administration, which would make retirement social assistance payments according to procedures agreed with the Bank. Severance payments also included end-of-service compensation for retiring workers. Turkish Labor laws define severance payments, applying to both public and private sectors, as payments to workers who are terminated as well as workers retiring voluntarily. A worker can choose to retire once he or she has reached the minimum requirements, or to postpone retirement until he or she has reached the maximum age. Although these end-of service payments were not specifically mentioned in the PAD, they were in line with the provisions of the Loan Agreement which refers to the Turkish Law as defining severance payments eligible for Project financing (see Box 1). Box 1: Job Loss Compensation Payments financed under PSSP II Severance payments: According to Turkish Labor Laws No and 1475, the legal definition of severance payments, which applies to both public and private sectors, includes workers who are terminated as well as workers retiring voluntarily. Severance payments are calculated on the basis of a worker s ending salary and length of service, whereby he or she receives one month of severance pay for each year in service. In the case of retirement, workers are eligible for such payments provided that they meet certain requirements, such as a minimum retirement age and a minimum number of contribution days towards the Social Security System (SGK). A worker can choose to retire once he or she has reached the minimum requirements, or to postpone retirement until he or she has reached the maximum age. Special Job Loss Compensation (SJLC) payments: In addition to regular severance payments, the Turkish system provides for SJLC payments, ranging up to eight months, depending on conditions and length of service, designed to encourage workers to leave over-staffed SOEs. Retirement Social Assistance for civil servants payments: The parameters for these payments are defined in the Privatization Law No The Law provides a 30 percent retirement bonus to those civil servants entitled to voluntary retirement before or at the time of privatization, in order to induce those employees who qualify for retirement, to actually do so. Employees are given two months from the date of privatization to choose this option. (b) Labor Redeployment Services (Cost at appraisal: 17 million; Actual disbursement: 14.9 million): This component financed labor redeployment services to workers displaced by the privatization of SOEs, including secondary layoffs, to assist them in rapidly re-entering the labor market. The component was managed by the Turkish Employment Agency (ISKUR) and by the Small and Medium Industry Development Agency (KOSGEB), as described below. 2 Appraisal costs and actual disbursements for the original components refer to IBRD financing only 4

15 Activities managed by ISKUR included: (i) Job counseling and placement services, such as pre-counseling, job/vocational and social counseling, provision of labor market information, aptitude assessments, labor exchange and placement services, and relocation services. Service providers included NGOs, private firms and labor organizations. (ii) Retraining, including vocational, general education and literacy training, and vocational training for self employment. Service providers included enterprises, private, NGO and public training institutions. (iii) Temporary community employment, such as environmental clean-up, refurbishment of public infrastructure, provision of assistance to social agencies. Organizations proposing such programs, which could account for up to 25 percent of Labor Redeployment Services disbursements, had to demonstrate that the program would not displace normal employment. Activities managed by KOSGEB included: (i) Small Business Assistance services (SBAs) for individuals wishing to start their own business. Eligible service providers included private agencies, autonomous government organizations, foundations, and statutory occupational organizations. (ii) Small Business Incubators (SBIs), including facilities/premises rental, small business consulting services for tenants, shared support services and equipment. Eligible service providers included private agencies, private and non-governmental associations and foundations. Incubators funded for less than 36 months under PSSP I were also eligible to receive continuing support for operating costs under PSSP II. Service providers had to agree to minimum job placement and start-up rates for ISKUR programs and SBAs, respectively, and, in the case of SBIs, prepare business plans to reflect increasing capacity to cover costs from their own funds. (c) Management, Monitoring and Evaluation (Cost at appraisal: 0.9 million; Actual disbursement: million): The objective of this component was to monitor the impact of Labor Redeployment Services (LRS) and manage PSSP II effectively as a whole. The component was to finance: (i) Surveys to evaluate the effectiveness of the LRS in mitigating the social costs of labor redundancies resulting from employment and privatization on selected communities. (ii) In-depth socio-economic analyses of specific communities where privatization has taken place. The component was also to finance limited TA and small goods to: (i) coordinate project execution, and manage the resources of the Project; (ii) procure all Bank-financed goods and services for implementing agencies; (iii) operate the financial management system; (iv) act as liaison between the implementing agencies and the World Bank; (v) ensure that annual audits were completed in keeping with Bank standards; and (vi) manage the MME studies with the technical support of the Directorate General of Economic Research (EKA) in Treasury. 1.6 Revised Components 5

16 Component 3 was revised by the exclusion of the M&E portion. Shortly before loan effectiveness, EKA requested that it no longer provide technical support for the ME studies, following a Treasury s management decision partly dictated by organizational priorities and partly by the potentially sensitive nature of impact assessments related to the privatization program. As a result, a Loan Amendment 3 excluded from the Management, Monitoring and Evaluation (MME) component the originally planned studies to assess the social and economic impact of the privatization program and of the effectiveness of Labor Redeployment Services as mitigating measures. The revised component, re-labeled Project Management, only included provision of technical assistance and goods for the PCU for Project coordination and implementation. Although it was agreed that specific evaluation studies would still be undertaken by the Privatization Administration during the life of the Project, only one such study on Small Business Incubators (SBI) was undertaken. To partly compensate for this lack of data, several key background notes and beneficiary surveys were conducted as part of the ICR itself (see Section 2.3). Revisions in the MME component were not accompanied by any changes of the Project s results framework, which remained in its original form, maintaining indicators on industrial productivity and poverty mitigation to be assessed through productivity and economic analysis of individual SOEs, social impact evaluations, and impact evaluation of services on employment of displaced workers. However, in practice, following the MME component revisions, these assessments were not undertaken as part of the Project (see Section 2.3). 1.7 Other significant changes During implementation, the Project underwent the following additional changes to the one listed above: (i) two amendments to the Loan Agreement; (ii) one Project Operational Manual (POM) amendment; (iii) a loan closing date extension and; (iv) two loan amount cancellations (see details below). (i) Amendments to the Loan Agreement 3 March 28, 2006: As part of the same Loan Amendment described under section 1.6 above, the Government also requested that administrative and operational support services be provided through the PCU, rather than a Technical Assistance firm initially included in the Project design. As a result, the amendment also included a new disbursement category and minor reallocations for incremental operating costs. 3 February 13, 2008: a second amendment to the Loan Agreement requested by the Government increased the disbursement percentage for the Job Loss Compensation component from 81 percent to 95 percent, with the aim of further incentivizing the ongoing privatization program. (ii) Revisions to the Project Operational Manual (POM) 3 In October 2004 the Government passed an amendment to the Privatization Law permitting workers laid-off from privatized SOEs to apply for temporary, low-wage and low-skill employment through the Temporary Employment Program (TEP). In practice, because this is not a temporary program, but results in a series of 11-month contracts within state ministries, workers enrolled in the TEP were excluded from Job Loss Compensation (JLC) 3 Dated March 28,

17 reimbursements under PSSP II, and this was explicitly stated in the original POM (see Box 2). However, in 2006, it emerged that the number of TEP applicants from privatized or liquidated SOEs was significantly higher than initially anticipated by the Privatization Administration, and that 1,876 such workers had erroneously received JLC reimbursements under PSSP II. The amount was eventually deducted from a later JLC disbursement application, but this circumstance had two important implications: - The Bank, despite some underlying concerns about the nature and incentives of the Temporary Employment Program (TEP), agreed that it was effectively a social safety-net for the poorest and most disadvantaged workers made redundant as a result of privatization. As a result, the POM was amended - in March 2008 to make all workers that joined the TEP, earning an annualized monthly wage of less than 1.2 gross minimum wages, eligible to have their JLC expenses reimbursed under PSSP II. Because TEP wages are related to the workers education levels, this definition de facto included all applicants with primary and secondary school education (i.e. an estimated 98 percent of all workers who apply to join the TEP workers), and excluded virtually all university graduates (see Table 2). - The monitoring system for JLC reimbursement applications under PSSP II was also strengthened as a result of the unintended TEP workers reimbursement prior to the POM amendment. Box 2: Turkey s Temporary Employment Program (TEP or 4C) In October 2004 the Turkish Government passed an amendment to the Privatization Law permitting workers laid-off from privatized SOEs to apply for temporary, lowwage and low-skill employment through the Temporary Employment Program (TEP, also known as 4C from the article of the Civil Servants Law governing it). Once an applicant has been placed in a government body, he/she is entitled to be employed indefinitely on an 11-month work schedule, up to the date the employee qualifies for a pension. Applicants do not need to be within a certain number of years of their retirement to be hired under the TEP, nor are they required to seek job training or attempt to find alternate employment before joining the TEP. PA data as of early 2008 indicated that roughly 93 percent of the people rehired under the TEP are under the age of 50. Only 1 percent of the people hired by the TEP had retired since starting work. Table 2: Wages and education distribution of TEP workers in Turkey Primary Education or less Secondary Education University Education Monthly Gross Wage TL (1.03 minimum wages) TL (1.17 minimum wages) TL (1.31 minimum wages) % of TEP workers 54% 44% 2% Source: PA, November 2007 (iii) Closing date extension 7

18 3 June 29, 2009: A partial loan extension for Category 4, Consultants and Training, was granted until October 31, All other Project activities were no longer eligible to finance expenditures incurred after June 30, The extension allowed the PA to complete the necessary activities related to Project closing, including submission of eligible reimbursement payments under Components 1 and 2 to the Bank, as well as preparing payment orders and processing final closure of the special account. (iv) Cancellations 3 July 20, 2009: 15 million, from portions of the loan which closed on June 30, namely Category 1 - Job Loss Compensation Payments, and Category 7-Unallocated, were cancelled at the Government s request. 3 January 12, 2010: the remaining undisbursed loan balance, amounting to 15.3 million, was cancelled as part of World Bank s loan closing procedures. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Project Design: The overall objective of the Project was clear and consistent with the Government s economic reform agenda, which was started in the aftermath of the 2001 crisis and included a renewed commitment to privatization as a means to reduce state ownership in the industrial sector. The Project design was largely based on the previous PSSP I operation, which was successfully implemented and largely achieved its intended outcomes. Project components were well aligned with the overall objective, although the actual beneficiary population, consisting largely of retirees under the Job Loss Compensation (JLC) component on one side, and virtually no displaced workers under the Labor Redeployment Services component on the other, limited the effectiveness of the Project. Strong features of Project design include: - Privatization Administration s taking charge of engagement mechanism with affected workers. Project design strengthened the interaction between the PA and affected workers by continuing, under the leadership of the Labor Adjustment Group (LAG), the in-company pre-layoff labor redeployment assessment and planning sessions established under PSSP I. This was an unprecedented approach for the PA, which, in its own view, brought, for the first time, the workers directly and early on into the process. - Minimal adverse selection. A monitoring mechanism was put in place to ensure appropriately targeted use of JLC funds. Reimbursing the Privatization Administration for JLC (i) at least six months after successful privatization (full privatization or transfer of at least 51 percent of capital or effective liquidation); and (ii) after verification that none of the workers included in the reimbursement request was rehired in the public sector or the privatized enterprise, virtually eliminated adverse selection. The inclusion, prior to the POM amendment, of ineligible Temporary Employment Program (TEP) workers as JLC beneficiaries, led to further strengthening of the monitoring system. - Accountability of service providers. Linking the Turkish Employment Organization (ISKUR) s and Small and Medium Industry Development Organization (KOSGEB) s training and business assistance programs with job placement rates made service providers more accountable and offered an incentive to provide better quality programs. 8

19 - Expansion of sustainable Small Business Incubator (SBI) model. PSSP I and II significantly contributed to the strengthening and expansion of the SBI model in Turkey, and to KOSGEB s capacity to implement it. Project design also included specific requirements to increase SBI sustainability, through the requirements that service providers submit a business plan reflecting increasing capacity to cover costs from their own funds. Tying SBI funding under PSSP II to the number of registered workers, which KOSGEB itself monitored on a regular basis, also resulted in better governance, with no unregistered workers being supported by the Project. The SBI assessment completed under PSSP II also showed that a significant majority of workers in SBIs are connected to the social security and that this number is significantly higher than in the overall economy. - Participatory process. In addition to strengthening workers participation through the pre-layoff sessions led by the Labor Adjustment Group, stakeholders engagement was envisaged through the Tripartite Advisory Committee, whose role was to advice on overall policy aspects related to Project implementation. Other design factors, however, limited the effectiveness of the overall Project: - Job Loss Compensation (JLC) largely benefitting retirees. Disbursements under the JLC component largely financed workers retirement, and thus did not promote their transition to new jobs. This was not explicitly stated in the PAD, though it was in line with the Loan Agreement which refers to the Turkish Privatization Law (where retirees are entitled to severance payments) as governing the disbursements under PSSP II (see Box 1). The ICR for PSSP I does not analyze this in detail, but makes some general reference to the close-to-retirement age of many workers displaced by the Privatization, suggesting that this aspect could have been anticipated and/or addressed more explicitly up-front in the design of the follow-up PSSP II. - Broad targeting of Labor Redeployment Services (LRS). The PAD specifically indicated that the LRS component would benefit workers displaced by the privatization of SOEs, including secondary layoffs, to assist them in rapidly re-entering the labor market. LRS activities achieved important results in terms of job placements and creation. However, because actual beneficiaries did not include workers laid-off from privatized SOEs, but the unemployed population at large (see section 3.4) LRS activities were somewhat delinked from the Project s key objective. Based on the evidence from PSSP I that the displaced workers demand for LRS was minimal and that this may have been due to the support being provided towards early retirement of workers, one of the key lessons highlighted in the PSSP I ICR was that it would be helpful, moving forward, to build in the flexibility to adjust the type of services to the level and sources of demand and, if opening the program more broadly (e.g., to generally unemployed) to do so formally rather than implicitly, so that the services could be better adjusted to the actual recipients' needs while in parallel being balanced with the needs of the workers displaced by the primary lay-offs. In practice, this did not materialize and the LRS component maintained the same design and targeting as in the previous operation. - A multi-layered implementation arrangement limited coordination effectiveness. Project activities were coordinated by a Project Coordination Unit (PCU) within the Privatization Administration (PA). The PCU was the Bank s main interlocutor and liaison with all the implementing agencies, and provided administrative and implementation support for Project activities. The Labor Redeployment Services component was implemented by the Turkish Employment Organization (ISKUR) and the Small and Medium Industry Development Organization (KOSGEB), and coordinated by the Labor Adjustment Group (LAG) within the PA. Such arrangement, with the PCU and the LAG both acting as coordinators at different levels within the PA, led to: (i) a lack of coordination and 9

20 effective cooperation between the PCU and the LAG; (ii) a perception, by the implementing agencies, of excessively long processing of administrative tasks; and (iii) somewhat fragmented flow of information from each implementing agency to the PCU and, in turn, to the Bank. Background Analysis: The ICR finds the background analysis for the preparation of PSSP II to have been overall adequate, though not all lessons were properly reflected into the project design. Background analysis was conducted towards the end of PSSP I and it included an evaluation of the economic and social impact of privatization among affected workers. However, due to Government s concerns about the sensitivity of the report, this was never made public and some of its lessons, particularly those pertaining to labor redeployment services, were not adequately reflected in the project design. Key lessons that were taken into account from PSSP I include (i) targeting and selecting service providers on a competitive basis and with pre-negotiated, resultsoriented performance criteria; (ii) a self-standing M&E component including a series of social impact studies; and (iii) a thorough monitoring system to minimize risk of misuse of funds. Specific lessons that were not incorporated into the design of PSSP II, or only partially so, include (i) the importance of formally tracking demand and use of services by SOEs laid-off workers, secondary layoffs, and the generally unemployed, to enable proper adjustments of future programs; and (ii) the importance, if broadening the scope of the program (for example to the generally unemployed) to do so formally rather than implicitly, so that such programs can be better adjusted to actual needs. Adequacy of Government s commitment: At the time of appraisal, the Government had shown full commitment to the privatization process by approving a new privatization program, which included major companies of national importance, to be implemented over the period Moreover, PSSP I was successfully being implemented, and supported the privatization of 36 SOEs, financing Job Loss Compensation to 18,261 workers. The Government continued to remain fully committed to the privatization agenda, which over the past six years has resulted in more sales revenues than in the previous two decades combined. This agenda was consistently supported by job loss compensation payments to affected workers, which PSSP II financed. Similarly, ISKUR and KOSGEB were fully committed to, and successfully implemented several labor redeployment services which resulted in significant job creation (see Section 3.2). Assessment of risks: The Project s overall risk was rated Moderate. The ICR finds that the risk assessment was overall realistic and that adequate mitigation measures were identified, with one exception related to a lack of effective institutional coordination among the Project s implementing agencies. Despite being rated Negligible, this risk did materialize and mitigating measures were not properly addressed during preparation. Instead, these were limited to stating that PCU in the PA had been adept at handling coordination so far, and that Project protocols had been signed between the PA and all implementing agencies. This particular risk was underestimated and, thus, sufficient mitigation measures were not put in place. Quality Assurance Group (QAG) assessment: A Quality-at-Entry assessment was conducted in September The assessment rated the quality at entry as Moderately Satisfactory. It identified: (i) strong features in the high strategic relevance of the Project; (ii) strong Government commitment to the privatization agenda; (iii) a more realistic estimate, compared to PSSP I, of Job Loss Compensation beneficiary workers and per/worker payments; (iv) an experienced implementing agency; (v) use of experiences from the previous project; and (v) preparation of a better quantification of expected economic benefits. The main shortcoming identified was not having reflected the findings of the first social impact study conducted under PSSP I, with 10

21 particular reference to labor redeployment services. The QAG assessment also pointed out the lack of baseline data to conduct realistic assessments of the impact of privatization in communities where employment is heavily dependent on a single enterprise. 2.2 Implementation The Project was successful in delivering most of the outputs under the three components and achieving most of its intended outcomes. The Project was not considered at risk at any time during its implementation period. Most outputs were delivered and the loan was disbursed almost in full. The loan became effective in early 2006, seven months after approval. This gap was largely due to the Privatization Administration s decision to complete the implementation of PSSP I, which closed in December 2005, prior to commencing the activities under the follow-up project. Once the loan became effective, implementation proceeded well, as reflected by steady disbursement flows. At project-end, total disbursements amounted to million or 92 percent of total loan proceeds, with cancellations totaling 30.3 million. The loan benefitted 13,614 displaced workers through job loss compensation, provided training to 9,420 unemployed (52 percent of whom were placed in jobs), supported the creation of five new Small Business Incubators which generated 1,086 jobs through existing tenants, and financed Small Business Assistance services generating 210 start-ups and 332 jobs. Overall Project ratings were consistently satisfactory. Throughout the Project s life, both Project Development Objective (PDO) and Implementation Progress (IP) ratings were mostly Satisfactory, with the exceptions of IP once being rated as Highly Satisfactory in 2007 and both IP and DO being rated Moderately Satisfactory starting in early 2009 until project closing. M&E was downgraded to Moderately Unsatisfactory in February 2009, and further downgraded to Unsatisfactory in the last two ISRs. The following section examines this aspect in more detail. Overall, the downgrading of IP and PDO ratings during the later stages of the Project s life is not a reflection of a deterioration of PSSP II performance. Rather, it resulted from intensive supervision efforts which led the team to conclude that some key issues had not been fully recognized and acknowledged in the past. At the same time, implementation was partially affected by (i) the exclusion of key evaluation studies originally part of the Project design; (ii) a loan amendment to make most Temporary Employment Program (TEP) workers eligible for Job Loss Compensation reimbursement even though the TEP was not in line with the Project s aim to support rapid re-integration into the labor market through Labor Redeployment Services (see Section 1.7); (iii) the lack of Tripartite Advisory Committee (TAC) meetings, as envisioned in the Loan Agreement and Project Appraisal Document; and (iv) the absence of a formal Mid-term Review (MTR). - M&E: M&E activities could not be implemented as originally planned as a result of the amendment to the loan agreement that excluded most of the initially envisaged evaluation studies. Some of these evaluations were undertaken as part of the ICR, but they can substitute for the comprehensive M&E conceived at appraisal only to a limited degree. The Project s Results Framework was not revised and continued to include outcome indicators, such as increased productivity of former State-Owned Enterprises and mitigation of poverty among affected workers, which proved difficult to measure. 11

22 - Tripartite Advisory Committee: The role of the TAC was described under Schedule 5 of the Loan Agreement, which explicitly indicated that it shall be responsible for reviewing the progress of Project implementation and advising on any policy issues related to the implementation of the Project. The TAC was established under PSSP I and consisted of representatives from the Government, including Treasury, Privatization Administration, State Planning Organization, State Institute of Statistics, KOSGEB, ISKUR, employers (e.g., Turkish Confederation of Employer Associations, Union of Chambers of Commerce of Turkey) and Unions. The PCU organized four annual meetings during the implementation of PSSP I, which provided a platform for sharing information and exchanging views with stakeholders on the implementation of the Project. The TAC, however, never met during the implementation of PSSP II. In late 2007, the Bank and the Government agreed that, given the smooth progress of project activities, there was no need to hold regular TAC meetings at that point in time. During project implementation, the PCU reiterated the view that the TAC was not an effective tool to discuss broader questions regarding the privatization process, and that meetings under PSSP I made sufficient contributions, with no perceived additional value-added in pursuing additional meetings under PSSP II. Given that the TAC had been prominently highlighted in the original project design as well as project documents, the Bank asked Government to hold TAC meetings during the last phase of Project implementation. The PCU subsequently informed the Bank, during the last supervision mission, about the decision of the PA not to convene the TAC for the remaining life of the Project and the Bank Team requested separate meetings with Union representatives as part of the ICR preparation. However, these meetings did not materialize either, in part due to the unavailability of Union representatives. - Mid-term Review: A MTR had been originally planned for June 2007, later scheduled for May 2008, but was not carried out, in part due to staff transition within the World Bank team. Instead, the new task team undertook a comprehensive review of all project activities through an in-depth mission conducted at the end of 2008, which constituted a de-facto mid-term review. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization M&E Design. A positive feature of the Project s original design included a self-standing M&E component, which was to finance surveys to evaluate the impact of Labor Redeployment Services and conduct socio-economic analyses of communities affected by privatization. On the other end, one shortcoming of the M&E design was a somewhat ambitious and broad Results Framework. For example, assessing the productivity of former State-Owned Enterprises (SOE) proved to be a challenge, as this data is not publicly available and the Privatization Administration does not have an incentive to track SOEs once they are privatized. Secondly, the Results Framework included indicators on poverty mitigation and increased productivity of industrial sector segments, but no proper measurements or baselines were provided. M&E Implementation. In practice, with the exception of a report on Small Business Incubators, the studies originally planned as part of project design were not conducted, as the component was revised to focus exclusively on project management aspects, and not M&E. During the in-depth November 2008 supervision mission, the World Bank team and the implementing agencies discussed the importance of undertaking a number of focused evaluation studies whose results would ideally be available before the loan closing date. The Government later decided that it 12

23 would not undertake (and the Project would not finance) such studies. This was reflected in the downgrading of the Project s M&E performance rating to Moderately Unsatisfactory in February 2009, and to a further downgrading to Unsatisfactory in the last two ISRs. As a result, some key background notes and surveys were carried out as part of the ICR analysis, in close cooperation with the implementing agencies. These included: - A quantitative and qualitative survey on a selected sample of workers benefitting from Job Loss Compensation (JLC) payments within the scope of PSSP II, to assess the social impact of privatization and the mitigating effect of JLC payments for these workers and their families. - A background note to support the assessment of the JLC component, including its contributions to SOE performance and with a specific focus on productivity. - A quantitative survey of Small Business Incubator managers, tenant and graduated companies aiming at providing a more complete picture as to whether small-business incubators can constitute a viable method for employment generation. - A pilot follow-up survey of individuals, who recently completed a training program under PSSP II, was conducted as part of a broader TA program supporting ISKUR in building its institutional capacity on M&E and performance management in order to improve the effectiveness of its Active Labor Market Programs. The objective of the survey was twofold: (i) to test the feasibility of using surveys to supplement the data already available at ISKUR and Social Security Institute; and (ii) to provide insights on the effectiveness of ISKUR training programs financed under PSSP II. Further analysis of the pilot survey will help ISKUR assess if such a tracking survey is feasible and whether it can be conducted on a regular basis and incorporated into an enhanced monitoring and evaluation system. Follow-up workshops are being planned with the relevant agencies to further discuss the key findings and lessons of the ICR and of the background surveys conducted as part of it. Although these background analyses conducted as part of the ICR are important, they can only substitute the originally planned studies to a limited extent. M&E Utilization: The utilization of M&E was limited. This largely resulted from the exclusion of key evaluation studies from the Project shortly after loan effectiveness. Also, because the Results Framework was never revised to reflect the changes in the Management, Monitoring & Evaluation component, it could not be properly utilized. 2.4 Safeguard and Fiduciary Compliance Safeguards. The Project did not trigger any safeguards policies. In the case of Small Business Incubators (SBI), environmental and occupational health and safety practices were included in the support and training given to enterprises, though SBI tenants and graduates pointed to the need of more training altogether. PSSP II, which financed job loss compensations for the privatized Tekel Tobacco SOE, was also found in compliance with the Bank's policy on tobacco in OP 4.76, on the basis that the Project did not provide direct funding to the tobacco sector, but would instead support displaced workers find employment outside of the tobacco sector. 13

24 Fiduciary. Procurement and financial management practices were defined in the Project Operational Manual (POM), and the Government received continuous guidance from countrybased Procurement and Financial Management (FM) Specialists. FM arrangements for PSSP II were found to be Satisfactory and control procedures to be in place. The accounting software used by the PCU had adequate security levels and was capable of producing the financial monitoring reports of the Project. FM reporting was prepared by the PCU as part of quarterly Project Management Reports, which included a specific section on Financial Management and Procurement. These were consistently found to be Satisfactory from a FM perspective. PSSP II was not procurement-intensive as a substantial amount of the Loan (about million) was allocated for Job Loss Compensation payments. Three Procurement post-review missions were conducted during implementation. The reviews, which included sub-contracts awarded by KOSGEB and ISKUR local offices, found that selection of equipment and service providers was made according to the procedures stipulated in the POM. 2.5 Post-completion Operation/Next Phase PSSP I was not intended to be followed by a repeater operation, but as the privatization program advanced, additional assistance was requested by the Government through the follow-up PSSP II. The same components were retained, along with the structures established within the Privatization Administration (PA) and implementing agencies. All loan activities were completed by June 30, 2009, with the exception of Category 4, Consultants and Training, which was extended to end-october 2009 to allow the completion of necessary tasks related to project closing. The Government s commitment towards privatization continues. The Ninth Development Plan, which outlines Turkey s overarching medium-term development goals, aims to reduce State-Owned Enterprises share in national income and employment as a result of privatization, in line with the EU harmonization agenda which promotes privatization of inefficient SOEs as a key ingredient to increase efficiency in production and attract Foreign Direct Investment. Job loss compensations continue to be paid to affected workers under Turkish law. However, the mechanism created under PSSP through which the Labor Adjustment Group organized pre-layoff labor redeployment assessment and planning sessions for workers is no longer in place. Though a sample of workers, who had been laid off during the privatizations of PETKIM and TUPRAS and who were contacted as part of this ICR preparation, stated that this mechanism was not fully effective, the Privatization Administration (PA) believes it had an important impact on its approach to the privatization process, by bringing workers into the process early on. Prior to PSSP I, there was no direct interaction between the PA and the affected workers. The following Labor redeployment Services activities also continue to be implemented: - Following its recently expanded mandate, the Turkish Employment Organization (ISKUR) requested World Bank s technical assistance on the delivery of public employment services and on building a performance monitoring and evaluation system for its training programs to ultimately increase their effectiveness and efficiency. This indicates strong ownership and sustainability of ISKUR s training activities. 14

25 - Small Business Incubators, with the exception of Avanos, are all operational, despite continuing to feel the negative impact of the global economic crisis. - The Small and Medium Industry Development Organization (KOSGEB) and ISKUR signed a protocol to continue small business advisory services, with the same content as the PSSP IIfunded Small Business Assistance Services. While ISKUR s primary target are the unemployed, KOSGEB offers the training to broader segments of population interested in establishing a new firm, in addition to financial support in the form of a small grant. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Project objectives remain highly relevant. The objective, design and implementation of PSSP II remain highly relevant with regard to Turkey s reform program. The Government s Ninth Development Plan ( ) indicates privatization as a key priority to increase competitiveness and reduce the burden of loss-making SOEs on the Turkish economy, and calls for a reduction in the ratio of total SOEs gross sales revenues to GDP. As a result of the global economic downturn, Turkey s public debt 4 has further increased, with a forecast of about 47 percent of GDP at end-2009, from below 40 percent at end The Government s Medium Term Program (MTP) assumes a gradual improvement in public sector balances with the contribution of privatization revenues, projected to be TL 10.4 billion in 2010, TL 9.4 billion in 2011, and TL 8.9 billion in Privatization, in the past five years, picked up considerably. Privatization revenues between 2005 and 2009 (exceeding US$28 billion) were three times as high as combined privatization revenues in the previous 19 years. The social costs of privatization also remain a key concern. Since as recently as December 2009, thousands of workers from privatized TEKEL (which benefitted from PSSP II JLC reimbursements) 5 have been protesting in Ankara. The press has been providing a broad coverage of these events, and the Government eventually responded with an extension of the working period under the Temporary Employment Program (TEP) from 10 to 11 months and with an increase of TEP workers salaries between 12 percent and 15 percent. The ongoing Country Partnership Strategy (CPS), shaped directly by Turkey s Ninth Development Plan and by the Government s Program, aims at contributing to three main development pillars, one of them being Improved Competitiveness and Employment. Continued strong privatization is highlighted as a key ingredient of a better investment climate and overall better investment environment. The backbone of this part of the World Bank-financed program is the Programmatic Competitiveness and Employment Development Policy Loans (CEDPL) series, which supports the improvement of the Turkish investment climate, including a vast program of privatization of SOEs. The CEDPL series has now been discontinued but the content of CEDPL reforms is carried over in the new Restoring Equitable Growth and Employment Programmatic DPL. 3.2 Achievement of Project Development Objectives 4 EU definition 5 Workers involved in the ongoing protests did not, themselves, benefit from PSSP II financing, as they were laid-off on January 31, 2010, when the loan had already closed. 15

26 PDO Indicator 1: Productivity of former State Owned Enterprises (SOE) will increase when privatized due to reduction in labor costs. Data suggests that privatization, through Job Loss Compensation (JLC) support, led to efficiency gains in privatized SOEs. A full analysis of privatization effects on productivity levels proved difficult, due to limited availability of data from privatized SOEs. Petkim and Tupras, both publicly traded companies subject to disclosure of information regulations, provided the only source of sufficiently available and reliable information and thus the ICR focused its analysis on these two former SOEs. JLC for these two companies, operating in the Petrochemical and Refinery sectors, were also among the largest financed under PSSP II, exceeding 63 million in disbursements and benefitting about 2,500 workers. Productivity and efficiency gains realized: The analysis found that, despite the unfavorable impact of the global economic slowdown on the sector as a whole, significant productivity and efficiency increases were realized by the two companies, thanks to the massive restructuring and investment programs carried out following their privatization. - Tupras, Turkey s largest industrial enterprise, owns 59 percent of the total petroleum products storage capacity in Turkey. In 1990 the company was handed over to the Privatization Administration. Today, 49 percent of Tupras shares are publicly traded. In late 2005 the block sale of the remaining 51 percent of state-owned Tupras shares was granted to the Koç-Shell Joint Venture Group. According to company figures, while total investments between 1989 and 2005 had been US$1.95 billion, the amount of investment undertaken after privatization has reached US$1 billion in a period of three years. The company also became the national export champion by exceeding US$4.45 billion in exports by the end of As part of the restructuring program, Koç Holding went through a significant portfolio shift, resulting in focusing on key core sectors, as well as efficiency, margin growth and shareholder value. The number of employees has decreased by 15 percent since the 2005 privatization, while exports increased by 32 percent and production, with the exception of 2008, was overall sustained. The company began implementing an operating efficiency program (OEP) to further support group s profitability just after the privatization. The program aims to generate savings of US$378 million, distributed over the period. Tupras has also undertaken various other measures to improve operations and profitability (see Table 3). C N S US O P US N I US I E US R P P E N E Source: Tupras s Annual Reports Table 3: Tupras s Pre and Post-privatization Key Indicators 16

27 - Petkim is Turkey s main petrochemicals producer and the largest producer of thermoplastics and intermediaries, satisfying approximately 25 percent of domestic demand for petrochemicals. Its products are exported to countries of the European Union, the United States, Africa, the Middle East, Eastern Europe, Asia and the Far East. After a secondary offering in the first half of 2005, the stake held by the Privatization Administration (PA) was reduced from 88.9 percent to 54.4 percent. The PA realized the block sale of its 51 percent stake in After quarters of operational losses, Petkim reported TL12.1 billion net profits in its first 2009 quarter financials compared with the net losses of TL15 billion and TL98 billion in the first and fourth quarters of 2008, respectively. Petkim succeeded in halving its operating expenses on a year to year basis thanks to its restructuring efforts leading to a decline in number of workers. Petkim s workforce has been gradually cut back over the last 8 years. At the same time, production for marketing per worker increased from about 200 tons in 2001 to about 450 tons in 2008 (see Figure 1), while operating expenses as a ratio of revenues decreased from 9 percent in 2003 to 4 percent in the first half of Petkim exported around 36 percent of its production in 2008, a much higher proportion than the 25 percent average over the last five years as the company increased its export exposure due to the increasing percentage of cheap imports in the domestic market. Figure 1: Petkim s Employees and Per Capita Efficiency, ource: Petkim 2008 Annual Report S PDO Indicator 2: Workers displaced during privatization receive temporary income support (JLC) to help prevent them from falling into poverty. 17

28 PSSP II effectively provided income support to workers affected by privatization and prevented them from falling into poverty. 13,614 workers benefitted from job loss compensation payments. The Project successfully financed temporary income support, in the form of job loss compensation payments, to 13,614 workers, corresponding to the totality of workers displaced from the 20 SOEs that were privatized under the scope of PSSP II, including those workers displaced after January , as stipulated under the Loan Agreement, whose payments were retroactively reimbursed under the loan. Although the original Project target was to benefit 29,000 workers, the lower number of actual beneficiaries (47 percent of appraisal estimate) is not a reflection of weak Project performance, but rather of initial overestimations. This resulted from the Privatization Administration s doublecounting of Job Loss Compensation (JLC) and Special Job Loss Compensation (SJLC) beneficiaries, whereas these payments are made to the same individuals (see Table 9). Beneficiaries were largely retirees but retirement date flexibility provides an incentive for early retirement. About 80 percent of beneficiary workers had either reached maximum retirement age at the time of the SOE privatization, or were eligible to continue to work for a few years but chose to retire when they first qualified. This was not explicitly stated in the PAD, though it was in line with the Loan Agreement which refers to the Turkish Privatization Law (where retirees are entitled to severance payments) as governing the disbursements under PSSP II (see Box 1). Moreover, because of the very flexibility in workers retirement date, severance payments likely provided an incentive for early retirement, reducing the liabilities of the enterprise to be privatized, thereby positively influencing the privatization likelihood itself. Of the remaining workers who did not retire, almost the totality joined the Temporary Employment Program (TEP). Severance payments constitute a safety net for most vulnerable workers. Severance payments, which constituted 99 percent of total Job Loss Compensation disbursed under PSSP II, provided income support to workers displaced from privatized SOEs, preventing them from falling into poverty. For those workers who reached the maximum retirement age, severance payments represented an additional income to the regular pension flow. For early retirees and laid-off workers, the most vulnerable category of employees, such compensations provide, respectively, a bridge to retirement or a transitional safety net to other employment, including enrollment in the TEP. Average monthly severance payments can provide affected households with incomes above standard poverty lines for several years, helping them adjust. For example, for a 5-member household, whose poverty line is calculated to be TL935, the average severance payment would provide an income 10 percent above this threshold for 46 months (see Table 4). Table 4: Average Monthly Severance vs. Poverty Line for Selected Households Sizes Household size Complete 2009 Turkey Poverty Line (TL) Total Severance Total No. of Payments under PSSP 2 Beneficiaries (Euro) Average Severance Payment per Worker (Euro) Average Severance Payment (TL) No. of Average Monthly Severance Payments 10% above Poverty Lines ,201,563 13,535 22,771 47, ,201,563 13,535 22,771 47, ,201,563 13,535 22,771 47, , ,201,563 13,535 22,771 47, Note: Complete poverty line contains both food and non-food expenditures; /US$ FX rate as of March 15, 2010 Source for Poverty Line figures: 2008 Poverty Study, TURKSTAT 18

29 Nevertheless, most workers experienced an overall decrease in their personal and household income levels, with early retirees and terminated workers feeling most of the burden. This emerged from a quantitative assessment, supported by focus groups discussions, on a sample of 201 displaced Petkim and Tekel workers. 6 Workers who had reached their maximum retirement age were almost not affected by the privatization process as they are able to sustain themselves through their pension. On the other hand, younger retirees or terminated workers claimed a more negative impact. Many workers also indicated that they used job loss compensation payments to meet ongoing expenses and honor debts, rather than invest them or spreading them over time as monthly income. This contributed to workers sense of insecurity and concerns about the future. The Temporary Employment Program (TEP) is not perceived as a temporary arrangement. Workers criticism and unmet expectations towards the TEP indicate their perception of the program as a bridge to retirement rather than a temporary arrangement to ease transition to new employment. Workers indicated their temporary status under TEP as responsible for causing a sense of job insecurity, despite the possibility to renew TEP contracts indefinitely. Criticism was also expressed about TEP salary levels and mismatches between workers qualifications and previous experiences and occupations assigned to them while enrolled in the TEP. About 25 percent of the interviewed workers had enrolled in the TEP. Of those, 61 percent are still in the program. Those who left the TEP indicated that they had remained in it for a minimum of two years. PDO Indicator 3: Workers displaced by the privatization and economic reform program receive Labor Redeployment Services (LRS) to assist them in returning to the labor force. While the Project contributed to high placement rates for ISKUR programs, creation of additional Small Business Incubators (SBI) generating jobs, and new business start-ups through Small Business Assistance Services (SBA), this objective was only partially achieved given that actual beneficiaries did not include workers displaced from privatized SOEs, nor secondary lay-offs, but comprised the unemployed population at large. (i) Labor Redeployment Services managed by ISKUR: Placement rates were higher than originally estimated. During the life of the Project, ISKUR completed 481 labor redeployment projects. Of these, over 90 percent consisted of training programs. The overall job placement rate for those who completed the training reached 57 percent, with 4,896 trainees being employed. For each service delivered, the actual placement rate either exceeded or matched the original targets (see Table 5). The employment placement link is the most important result of the training course and can well be considered as the success rate of the organized training and employment services classes. 6 Given the limited sample size, largely due to the lack of workers contact information, the survey (which was conducted as part of the ICR) was not intended to draw general conclusions but it highlights a few areas where the PA and other relevant stakeholders may wish to focus their attention and possibly undertake some further analysis. 19

30 Table 5: ISKUR s Employment Placement Rates * Temporary Community Employment Programs S T P P F T O J T TCEP C VTSE TOTAL ** Vocational Training for Prospective Self Employers Source: ISKUR P C T P E E G R PSSP R F A E P R Virtually no participation from displaced SOE workers was achieved. Only a handful of the ISKUR training participants consisted of displaced workers from privatized SOEs directly. Similarly, the case that the beneficiaries of the Labor Redeployment Services (LRS) had been affected through secondary lay-offs is very hard to make given that the LRS services were rolled out nationally (from an originally targeted 36 provinces) and did not target the areas where major privatizations occurred (see Section 1.4). The decision for the expansion, according to ISKUR, followed the inclusion of TEDAS, which had workers in all provinces across Turkey, within the scope of the Privatization Program. According to ISKUR, lack of participation from displaced SOE workers may be partly attributed to the Temporary Employment Program (TEP), which came into effect shortly before PSSP II approval, attracting several workers who might have otherwise enrolled in ISKUR training activities. However, the demand from workers displaced by privatized SOEs, as well as from the traditionally defined secondary lay-offs, was negligible under PSSP I as well. Low enrollment rates by displaced workers from privatized SOEs and views of PSSP II beneficiaries, though not conclusive, suggest an opportunity to assess (i) the attractiveness of the current set of training programs for displaced SOE workers, and incentives for them to enroll in such trainings. Because most workers benefitting from Job Loss Compensation are retirees, while the remaining ones largely opt for the TEP, the demand for LRS is expectedly negligible; (ii) the effectiveness of such trainings in placing workers in sustainable and relevant jobs; and (ii) the extent to which such workers are adequately informed about the existence of ISKUR programs, and how the information flow could be improved. The focus group of Petkim and Tupras workers, conducted as part of the ICR, indicated that (i) most workers were not aware of ISKUR trainings; (ii) chances of finding employment after attending ISKUR training programs were considered limited; and (iii) many workers believed that they were already qualified for their job, and thus would not benefit from further training. Results from a pilot survey of 561 ISKUR training beneficiaries under PSSP II, conducted within the scope of the broader TA being provided to ISKUR to increase the effectiveness of its Active Labor Market Programs, also suggest areas of improvement. (i) proactive information flow from ISKUR seems limited, with most workers coming to know about the programs from ISKUR s Employment Offices or through family and relatives; (ii) 61 percent of the participants rated the trainings as good or excellent, but only 31% believed that the skills acquired during the training helped them find a job; (iii) compared to ISKUR data, indicating an overall placement rate of 57 percent, only 39.5 percent of trainees said to have found a job following the training; and (iv) salary scales of those employed in 2009 tended to be in the low-medium range, with 74 percent declaring average monthly earnings between TL250 and TL749, and 97 percent below 1,250 TL 20

31 against Turkey s minimum wage of TL 576; (iv) training was rated relevant by half of those who were employed after the training. The capacity of ISKUR staff, especially at the local level, increased. ISKUR management and staff highlighted the PSSP series contribution to ISKUR s diversified training offers, increased geographical coverage (prior to PSSP I in-service training was offered in 25 provinces, compared to 70 provinces under PSSP II), and a strengthened implementation capacity among ISKUR staff, particularly at the provincial level. This also resulted from partial decentralization of service providers selection, whereby ISKUR as well as KOSGEB local offices were authorized to sign contracts amounting up to 30,000. (ii) Small Business Incubators (SBIs) managed by KOSGEB The Small Business Incubators (SBI) model is new in Turkey and was strengthened and expanded with the support of PSSP. The SBI model, which was first established in Zonguldak with the support of World Bank funding and technical assistance, was strengthened and expanded across Turkey under PSSP I and II. Of the 15 incubators currently operating across Turkey, 10 were established with the support of PSSP I and II 7. The implications of PSSP support go well beyond the financing per se, as it had an important impact on KOSGEB s capacity to design and implement the SBI model. The Project design also made the financing of SBIs subject to specific requirements to increase their sustainability, by requiring that service providers submit a business plan reflecting increasing capacity to cover costs from their own funds, with a view to enable sustainability. This was reflected in the phasing out of operating costs reimbursements, such as with full support for the first year, two-thirds support for the second year, and one-third support for the third year. All SBIs funded under PSSP I and II, with the exception of Avanos, which closed as a result of financial problems, are still operational through their own funding sources. SBIs helped create more than 400 new firms and more than 3,500 jobs. Over the years, KOSGEB has monitored the performance of the incubators, including employment figures and occupancy rates, but it lacked a more complete picture as to whether SBIs constitute a viable and cost-effective model for employment creation. To partly compensate for this lack of data, a background study was conducted within the scope of the ICR: 10 SBI managers, 24 graduated companies, and 71 tenant companies were surveyed. The survey also benefitted from a previous report commissioned by KOSGEB in Notwithstanding the limited sample and nonconclusive data, the survey s initial findings provide a good opportunity for KOSGEB to identify some of the key issues highlighted by SBI companies and management. Overall, incubators financed under PSSP I and II helped the creation of 436 firms which generated 3,518 jobs. Roughly 30 percent of all SBI-supported firms have graduated. Currently, PSSP-funded SBIs host 293 tenant firms which have generated 2,370 jobs. Of the existing firms, 14 percent were started by women, who hold 31 percent of the generated jobs. Graduates survival rates vary across SBIs. Though these tend to be lower beyond the second year, they compare well with international data and are an important result given the novelty 7 The remaining five incubators include Zonguldak and four other ones supported by EU funding and targeted mostly to women 8 Assessment of the World Bank-funded Incubators Under Privatization Social Support Project, by Prof. Tunc Tayanc (August 2, 2007) 21

32 of the SBI model in Turkey. Inputs from managers of PSSP I-funded SBIs (whose data was used as a proxy for PSSP II-funded SBIs as well, since the latter were established later and had no firms graduated for more than 1 year) indicate that graduates survival rate after one year is very high, with an average of 82 percent. Data on two and three-years survival rates suggest that, on average, about 65 percent of the firms manage to survive after the second year and 42 percent survive at least three years, with Adana and Eregli ranking significantly behind (see Table 6). Considering that the very concept of SBIs has been implemented in Turkey for less than a decade, these are good results and compare well with the U.S. Small Business Administration s reports that two-thirds of new employer establishments survive at least two years, and 44 percent survive at least four years. 9 Table 6: PSSP-I financed SBIs graduated companies survival rates SBI Num be r o f tenants # of graduate companies (since SBI inception) % of graduates % of graduates % of graduates survived after survived after survived after the 1 st year of the 2 nd year of the 3 rd year of graduation graduation graduation Adana % 22% 2% Eregli % 30% 12% Eskisehir % 81% 53% Tarsus % 95% 95% Note: Van % 100% 50% Data are only reported for PSSP I-funded SBIs, as most of those financed under PSSP II were established after the end of 2006 and thus had no firms graduated for more than 1 year. Data on the Diyarbakir SBI was not included because too limited. Source: KOSGEB and SBI managers PSSP Performance after graduation has been positive for half of the graduated respondents, but would welcome more training in the areas of marketing and promotion. For about 50 percent of graduated respondents production capacity increased after graduation, in the range of percent in the majority of the companies. Respondents attributed the main reason for the increase to the change of location outside the Small Business Incubators (SBIs), which helped them get closer to their customers. The average revenue created per employee increased from US$6,468 in 2008 to US$9,512 in To help them in the transition process out of the SBIs, tenant and graduated companies indicated the need for more training, particularly in the areas of marketing and promotion. Only 10 percent of the surveyed tenant SBI firms indicated that they received such training. Incentives, costs and benefits of SBIs are some areas of further analysis. About 30 percent of graduate respondents stated that they could not have started a new business without an SBI. While this cannot lead to conclusions on SBIs value-added, it gives an opportunity to assess the incentives for firms to apply to incubators, as well as the related costs and benefits, such as the quality and frequency of training courses, the extent to which networking and efficiency 9 U.S. Small Business Administration (2007 data) 22

33 opportunities are exploited, how access to financial resources and grant could be improved, and whether more can be invested in the physical infrastructure of some SBI facilities. (iii) Small Business Assistance Services (SBAs) managed by KOSGEB Start-up rates exceeded agreed rates. The Labor Redeployment Services component financed Small Business Assistance services (SBAs), implemented by KOSGEB and delivered by external service providers selected through a competitive process. Services included initial assessment of the aptitude and skills of unemployed persons to start businesses, developing business plans, advising on accounting, financial, legal, marketing and sales services issues, assistance in the dialogue with local authorities, short-term training and other consulting services to unemployed clients who intended to start their businesses. Service providers agreed to start-up rates of at least 10 percent and a minimum of 50 percent client s share in the start-up business. Within the scope of PSSP II, 14 SBAs were delivered in 13 provinces to 1,536 participants. For all the delivered SBAs, the start-up rate has exceeded the agreed rate, with notable cases such as Elazig, Osmanye, and Istanbul (see Table 7) whose start-up rates were more than double those initially agreed. Table 7: Committed and actual start-up rates for Small Business Assistance Services Source: Government s Project Closing Report Implementing KOSGEB Office No. Participants Committed startup rate (%) rate Actual start-up (%) Actual Number of start-ups No. of people employed stanbul 75 10% 16% Çorum % 11% Diyarbakûr % 15% Samsun % 11% Elazû 51 10% 24% Bartûn % 11% Giresun % 11% Mersin 92 10% 16% Gaziantep 93 10% 13% Malatya 72 10% 20% Adana 74 10% 14% Osmaniye 75 10% 23% Kastamonu 70 10% 13% 9 9 stanbul % 21% ,536 10% 14% Relatively high female start-up rates. Another important result stemming from SBAs was the relatively high participation and start-up rates of female entrepreneurs. About 37 percent of all SBA participants were female, and 33 percent of start-ups were owned by women (see Figure 2). 23

34 This compares well with entrepreneurship data across Turkey, which indicates that only 12.5 percent of all entrepreneurs are women. 10 Figure 2: Small Business Assistance Services Participation and Start-up rates by Gender Source: Government s Project Closing Report Creation, within the Privatization Administration (PA), of a consultation mechanism with workers working in soon-to-be privatized SOEs, was an overall key achievement under the Project. Such mechanism did not exist in the PA prior to the PSSP operations. The mechanism consisted in the Labor Adjustment Group (LAG) visiting each company once the decision to sell or liquidate was announced, to inform its workers about the privatization process and, more importantly, about the benefits packages available to them, including job loss compensation and labor redeployment services through ISKUR and KOSGEB programs (see Table 8). In PA s view, one of the very reasons why this activity could be carried out in what was often a very hostile environment is that PSSP provided the necessary backing to the LAG to approach workers who would have otherwise been unwilling to engage with the PA. 10 Source: World Economic Forum Gender Report 24