Document of The World Bank Report No: 36024 IMPLEMENTATION COMPLETION REPORT (TF-25752 IDA-32410) ON A CREDIT IN THE AMOUNT OF SDR 3.7 MILLION (US$5 MILLION EQUIVALENT) TO THE PEOPLE'S REPUBLIC OF CHINA FOR A PENSION REFORM PROJECT July 10, 2006 HUMAN DEVELOPMENT SECTOR UNIT EAST ASIA AND PACIFIC REGION CURRENCY EQUIVALENTS (Exchange Rate Effective December 29, 2005) Currency Unit = Yuan (Y) Y1 = US$ 0.12 US$ 1.00 = Y8.1 FISCAL YEAR JANUARY 1 DECEMBER 31 ABBREVIATIONS AND ACRONYMS CAS Country Assistance Strategy COEs Compensation of Employees ERR Economic Rate of Return FRR Financial Rate of Return ICB International Competitive Bidding IT Information Technology MOLSS Ministry of Labor and Social Security NPV Net Present Value PIU Project Implementation Unit PMO Project Management Office PRC People's Republic of China PROST Pension Reform Options Simulation Toolkit PSCPF Projection System of China Pension Funds SARS Severe Acute Respiratory Syndrome SOEs State Owned Enterprises Vice President: Jeffrey Gutman (EAPVP) Country Director: David Dollar (EACCF) Social Protection Sector Manager: Ana Revenga (EASHD) Task Team Leader: Yee Mun Sin (HDNSP) CHINA Pension Reform Project CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 5 5. Major Factors Affecting Implementation and Outcome 12 6. Sustainability 12 7. Bank and Borrower Performance 13 8. Lessons Learned 14 9. Partner Comments 15 10. Additional Information 15 Annex 1. Key Performance Indicators/Log Frame Matrix 16 Annex 2. Project Costs and Financing 20 Annex 3. Economic Costs and Benefits 22 Annex 4. Bank Inputs 23 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 25 Annex 6. Ratings of Bank and Borrower Performance 26 Annex 7. List of Supporting Documents 27 Annex 8. Appendices: Coverage and Contribution Data for Heilongjiang and Qingdao 28 Project ID: P058308 Project Name: China Pension Reform Project Team Leader: Yee Mun Sin TL Unit: HDNSP ICR Type: Core ICR Report Date: July 11, 2006 1. Project Data Name: China Pension Reform Project L/C/TF Number: TF-25752; IDA-32410 Country/Department: CHINA Region: East Asia and Pacific Region Sector/subsector: Compulsory pension and unemployment insurance (96%); Non-compulsory pensions, insurance and contractual savings (4%) Theme: Social risk mitigation (P); Public expenditure, financial management and procurement (S) KEY DATES Original Revised/Actual PCD: 01/06/1999 Effective: 10/25/2000 Appraisal: 04/14/1999 MTR: 09/23/2002 Approval: 06/04/1999 Closing: 12/31/2002 12/31/2005 Borrower/Implementing Agency: PEOPLE'S REPUBLIC OF CHINA/MINISTRY OF LABOR AND SOCIAL SECURITY Other Partners: STAFF Current At Appraisal Vice President: Jeffrey S. Gutman Jean-Michel Severino Country Director: David R. Dollar Yukon Huang Sector Director: Tamar Manuelyan Atinc Alan Ruby Team Leader at ICR: Yee Mun Sin ICR Primary Author: Yasue Pai 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objectives and the design of the China Pension Reform Project are assessed in the context of: (i) their contributions to the Bank's Country Assistance Strategy (CAS) objective of social protection by assisting the government in developing financially sustainable parameter for pension policies; and (ii) their support of the government in overcoming the barriers to establishing a national unified, multi-pillar pension system, and strengthening pension oversight institutions at the provincial and municipal level. In 1997, the government introduced a new pension reform framework under State Council Document No. 26 with the goal of adopting a unified pension system for all urban workers. Specifically, Document No. 26 mandated the following: l implement a nationally unified pension system with multiple pillars, including a pay-as-you-go basic pillar which aims to provide a basic benefit, a fully funded individual account pillar which links benefits to an individual's lifetime contributions, and a voluntary supplementary pillar; l gradually achieve unification of system design in terms of contribution rates and benefits; l shift pension administration responsibilities from enterprises to local social insurance bureaus to strengthen financial control and improve service standards; and l ultimately move from pooling of the basic pension at the municipal to the provincial level. At the time of this project inception, although progress had been made in implementing the new framework, there were numerous barriers that needed to be resolved in order to promote speedier reform: l Contribution rates continued to vary by province, as well as between municipalities within most provinces. l Contribution rates amongst non-state enterprises varied between State Owned Enterprises (SOEs) and Compensation of Employees (COEs). Those enterprises with younger workforces were reluctant to participate in a system where they would be subsidizing enterprises with older workforces. l Individual pension accounts in many localities were essentially notional because contributions were being used to pay current beneficiaries. l Funds accumulated in individual accounts were invested in deposits and government bonds yielding returns that were not commensurate with wage growth. 3.2 Revised Objective: The objective of the project was to assist the Borrower to reform its pension system through the formulation of a sustainable funding strategy and the improvement of pension systems in a pilot province and a pilot city, in order to obtain experience and develop a template for the establishment of a national unified pension system. - 2 - Alignment with CAS and government priorities. These project objectives reflect the Bank's CAS objective of social protection by assisting the government in developing financially sustainable parameters for pension policies, and are consistent with the policy priorities of the government to establish a unified multi-pillar pension system that is sustainable in the long term. Human development, comprising health, education and social protection, is one of five major themes of the Bank's CAS for China. This project contributes to the CAS objective of social protection by assisting the government in developing financially sustainable parameters for pension policies that lead to an improved pension system. Given the state of the pension reforms and the government's strategy to strengthen the pension system, the project objectives were very much responsive to the Borrower's circumstances and priorities. The provincial and municipal pilots were chosen because of their distinct circumstances. Heilongjiang was chosen because it was a province whose demographic and economic circumstances made it particularly difficult to find policy parameters resulting in financial sustainability. Heilongjiang's economy was a proportionately large percentage of SOEs and almost all of such enterprises are covered in the existing pension system. Moreover, Heilongjiang had limited automation and recognized weaknesses in administrative systems at both the municipal and the provincial levels. Qingdao had a large SOE presence with relatively high dependency rates, but also had rapid economic growth including a large growth in joint ventures and private sector enterprises not covered by the pension system. By focusing this project on reforms in Heilongjiang and Qingdao, the central government aimed to learn from experiences in these localities and apply the lessons to the rest of the country. This is consistent with the policy making process in the People's Republic of China (PRC) whereby policy reforms are often piloted on a trial basis before being implemented on a national level. Given the vast population size, the disparity between the provinces and municipalities, and the stage of pension system development, reform can most sensibly take place by implementing pilot programs in select cities or provinces as this project did. Limiting the project scope to Qingdao municipality and Heilongjiang province brought focus and clarity to the project objectives and made objectives realistic and within reach. 3.3 Original Components: The project consisted of four components: (i) Policy Development; (ii) Information Systems; (iii) Training; and (iv) Project Administration. Component 1: Policy Development (US$0.7 million). This component aimed to support policy studies at the central, provincial and municipal levels. Long-term financial projections of expected contributions and other accrued income (interest income on assets held) and benefit payouts would be carried out at the central, provincial (Heilongjiang) and municipal (Qingdao) levels. This would include projected cashflows for all four pension system benefits - basic, transitional, mandatory individual accounts and voluntary supplementary accounts. Sensitivity analysis would test modifications to, including: (i) effective contribution rates; (ii) coverage rates and labor force growth; (iii) compliance rates; (iv) rates of real wage growth; (v) rates of real returns accrued to pension accounts (interest rate assumption); (vi) adjustments to retirement ages; (vii) adjustments to the transition formula applied; (viii) changes to the period of distribution of individual account accumulations to more closely approximate municipal or provincial life expectancies; and (ix) modification on the basis of indexation of pension benefits (wages to prices). Multiple iterations of such projections aimed to indicate the feasibility of achieving a sustainable, albeit partially-funded position over a 70-year period. These iterations would provide policy-makers with a sense of the modifications or clarifications of policy necessary to achieve such viability and sustainability. - 3 - Parallel to the above financial sustainability analysis, studies and training would be undertaken with support of project funds. Component 2: Information Systems Development (US$4.2 million). This component aimed to support the development of updated and revised information and related administrative systems in the Ministry of Labor and Social Security (MOLSS), as well as in the Heilongjiang and Qingdao labour and social security bureaus, respectively. The aim of such administrative reform and automation is to create greater capacity, increase processing efficiency, strengthen control provisions and establish record-keeping and processing capabilities. Component 3: Training (US$0.9 million). This component aimed to support the identification by MOLSS and by the Heilongjiang and Qingdao labour and social security bureaus of personnel skills requirements and, thereafter, development and implementation of training programs to improve the skills of its personnel, including in: (i) pension management and oversight; (ii) pension reform policy; (iii) financial projections and modeling; and (iv) pension asset management. Training would only be supported provided they were part of staff development plans at the Heilongjiang and Qingdao Labor and Social Security Bureaus and MOLSS, respectively. Component 4: Project Administration (US$0.5 million). This component would provide assistance in project administration at the level of the MOLSS project implementation unit (PIU) and the project management offices (PMO) in Heilongjiang and Qingdao, respectively. Financing would include support for: (i) personnel costs associated with the PIU and PMO; (ii) costs associated with counterpart meetings; (iii) printing, communications, postage and translation costs; (iv) internal travel costs for coordination between PIU and PMOs; (v) office operating costs (electricity, miscellaneous office supplies); and (vi) costs of special hired assistants. All four components were clearly linked to the objectives described above. l The policy component which undertook to make financial projections and simulations based on expected contributions and benefits in Heilongjiang and Qingdao aimed to provide central policy makers with a sense of necessary policy modifications to achieve long-term sustainability. l By implementing and strengthening record-keeping and automating the administration process, the information development component aimed to contribute to the objective of strengthening institutional oversight at the provincial and municipal level. l The training component was necessary for staff at the central, provincial and municipal level to: (i) broaden and deepen their knowledge of international and domestic pension development issues, and thus be able to make more accurate assessment of domestic policy development needs; and (ii) learn to use the financial projection and modeling tools to support policy development decisions. Staff training is also necessary to help them efficiently use the technology acquired with this project. l Efficient project administration was obviously necessary for achieving all the targeted objectives. - 4 - 3.4 Revised Components: There was no major restructuring of the project during implementation. During implementation, a portion of the credit that was allocated to Heilongjiang was reallocated to the MOLSS for technical assistance in developing a short to medium term actuarial forecasting model, the Projection System of China Pension Funds (PSCPF). These reallocated funds resulted from savings from hardware procurement and software development in Heilongjiang. Support to the PSCPF was seen as consistent with the project's objectives as the PSCPF would complement the long term financial projection analysis that the government has been undertaking using the Bank's Pension Reform Options Simulation Toolkit (PROST) model. The project originally envisaged technical assistance and studies to focus on the development of voluntary supplementary schemes in Heilongjiang and Qingdao. During project implementation, the central government issued regulations for a national supplementary plan, "Enterprise Annuities" to be developed. Project support for such a voluntary scheme at the provincial or municipal level was therefore abandoned so as not to duplicate efforts. 3.5 Quality at Entry: Quality at Entry is rated satisfactory, based on: l the consistency of the project objectives with the government policy objective and with the Bank's CAS of strengthening social protection. With the rapid growth in China's old-age population which will continue over the next 15-20 years, the government recognizes that it is in the interest of the country's overall stability to speedily reform its current pension system to manage government pension liabilities while providing social security for its elderly; l the project design which included project support at the central, provincial and municipal levels of government. The project's design focusing on pilot locales took into consideration how policy is usually developed in the PRC; and l The government's explicit support of the project both at a central level and at the level of the participating province and municipality, as indicated by its explicit role in project design and provision of administrative support at each level to coordinate efficient execution of project tasks. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: Satisfactory. The project achieved most of its major relevant objectives and has achieved or is expected to achieve satisfactory development results. During project implementation, study tours and a training program were held to train staff at the central, provincial and municipal levels of government on sustainable pension development issues. Overseas and domestic study tours provided the opportunity for relevant staff to learn about international and regional pension systems and their administration practices which can be applied to PRC system developments. Pension administration system hardware was procured through international competitive bidding (ICB) procurement procedures. Installation of the systems was carried out smoothly and administrators were given necessary training to efficiently use the systems. These training sessions, overseas and domestic study tours and the successful procurement and installation of the administrative information systems all contributed to the project objective of sustainable pension policy - 5 - development. The project also supported training for relevant staff to use forecast tools to estimate the long term sustainability of the Qingdao municipal pension system and the Heilongjiang provincial pension system. Based on this training process, the project was able to provide input to national policy makers on pension policy parameters. Since the quality of the forecasts using actuarial models is directly linked to data quality used for inputs, training programs were also held on data collection. Although training seminars for data collection were held, the quality of pension related data used for long term forecasting is still questionable which impacts the integrity of the forecasts. Data collection in China is a bottom-up process that involves participation at the county and municipal level. The difficulty in implementing a systematic method of data collection with set criteria is that conditions can differ vastly from one city or county to the next even when they are within the same province. Unsystematic data collection hinders policy-makers' faith in the forecast outcomes and therefore their reliance on the forecast as a policy input. Achieving systematic data quality is a major challenge of its own, and addressing this issue falls outside the realm of this project. The outcomes achieved during this project are expected to have a lasting impact on the development of the municipal and provincial and to some degree national pension systems. This is evident in the policies issued in Qingdao, Heilongjiang and by the MOLSS in the years since the project was launched. Since this project is closely inline with the government's pension development objectives, it is at times difficult to differentiate whether the final achievements stem from the project or the government, vice versa or both. But this project undoubtedly contributed to the positive direction of many of these policy initiatives. The development human resource capacity in terms of staff who have improved their skills through training supported by the project, while difficult to measure, will also have a lasting impact on future developments. 4.2 Outputs by components: Component 1: Policy development (US$0.7 million). This component is rated satisfactory. This purpose of this component was to forecast the financial viability and sustainability of the municipal and provincial pilots' pension systems and provide input based on these forecasts to central policy-makers on key pension parameters for policies applied to China more broadly. Training for the Bank's PROST forecasting model was carried out at the central government level and in Heilongjiang and Qingdao. In addition and in parallel to project activities, the initial momentum gained through such initial training resulted in the training of officials in PROST from additional provinces and municipalities. There were various issues that hindered the accuracy and thus further reliance of PROST by those who received training to use it in Qingdao and Heilongjiang, including: (i) PROST is a long-term forecasting model (in excess of 50 years) requiring accurate data inputs, yet the data quality was questionable and required substantial efforts at improvement; and (ii) PROST is a highly sophisticated model requiring users with multi-disciplinary knowledge in order to properly navigate and use the model properly. Project funds did support a short to medium term forecasting model developed by the MOLSS's PSCPF. Initially developed by MOLSS' own resources, project funds were allocated to support modifications to PSCPF to incorporate a wider set of pension parameters. With modifications, the model has the flexibility to run forecasts based on a varied set of parameters, a useful function given the disparity in pension systems across the country. The PSCPF will be decentralized for use in provinces and municipalities. - 6 - Parallel to the sustainability analysis using forecasting models, the project also supported research and studies of developments in other provinces and municipalities, leading to the following developments in Heilongjiang, Qingdao and the central government level and ultimately resulting in improved financial prospect of the system. MOLSS l A conference was held in Chengdu in 2000 to discuss the future of pension developments, resulting in a research document titled "10th Five-Year Plan for China's Pension Insurance Development and Targets for 2010". The document provided direction on a national level for how the country's pension system should develop in the next 5 - 10 years. l The project provided training on a national level on supplementary pension system reform, leading to the issuance of policy documents issued by the MOLSS: Document No. 20 "Trial Measures for Enterprise Annuity" and No. 23 "Trial Measures for Managing Enterprise Annuity Funds" which sets the framework for a regulated voluntary pension scheme, "Enterprise Annuity" [to be] subsequently launched nationally. l The project initiated a process to explore options for establishing an integrated national social insurance information management system linked to provincial and municipal systems using a common template. l Via training sessions, conference and study tours, the project contributed to the development of State Council Document No. 38 "Decisions on Improving the Basic Old-age Insurance System for Enterprise Employees", which seeks to implement on a national level the policy lessons of experiments that were deemed successful in the three pilot provinces of Liaoning, Jilin and Heilongjiang. Heilongjiang l Coverage grew in Heilongjiang during project implementation as a result of increased participation by the self-employed, private enterprise employees and civil servants. l Differences in contribution rates between cities and between types of enterprises within Heilongjiang were narrowed and average rates reduced and individual account contributions were raised. o For civil servants, the average contribution rate was reduced from 35.4% in 2000 to 26.12% in 2003 while the average individual account contribution rate was increased from 1.17% in 2000 to 4.57% in 2003. o For enterprises, the average contribution rate was reduced from 25.11% in 1999 to 23.2% in 2003 while the average individual account contribution rate was increased from 4.24% in 2000 to 7.95% in 2003. l Rise in contribution compliance rate. o Contribution collection process was improved (see Annex 8, Appendices A3 and B3). o Allocation of manpower towards compliance monitoring was increased. - 7 - l A separate study on developing designs for voluntary retirement savings accounts was not carried out because during the course of project implementation, the central government issued regulations for "Enterprise Annuities". Appendix A1 contains further details of changes in coverage and contribution rates. Qingdao l During project implementation, the participation rate of self-employed, private enterprise employees, civil servants and rural resident employees of foreign funded enterprises all increased. l Contribution compliance rates also increased, mostly due to the city's healthy economic growth. l Enterprise contribution rates decreased while individual contribution rates increased. l While absolute benefits increased over the period, average replacement rates declined. Appendix B1 contains further details of changes in coverage and contribution rates. Component 2: Information Systems Development (US$4.2 million). This component is rated satisfactory. The purpose of this component was system design, hardware procurement, software development for Heilongjiang, and site preparation for servers and other hardware, operating procedure development and personnel training to use the new systems. The satisfactory implementation of this component has automated information collection, management and dissemination as well as expenditure tracking in the pilot municipality and province. By digitizing the information and providing access to via wide area networks, the systems are now in place for more informed policymaking, including providing a clearer understanding to policy-makers of the state of the current pension system and the parameters that need change. MOLSS l Compiled "Labour and Social Insurance Information Management System Training Handbook" in 2000. l Compiled "Social Insurance Information System Management Index" in 2000. l Facilitated the preparation for ICB procurement across project pilots. Heilongjiang l All hardware was procured via the Bank's competitive bidding procedures, resulting in substantial savings, transparency and fairness amongst vendors. l Original data at a municipal and at the provincial level were successfully keyed in to convert to an electronic form. Such data provided the input from which four types of databases were built: (i) a database backup to store raw data; (ii) a database to support daily operations; (iii) a database supporting customer service; and (iv) a database supporting policy decisionmaking. During the process of inputting non-electronic data, all data were screened for quality and consistency, ensuring a higher quality of data entered into the databases. Application software for the four databases was developed to process data for each respective purpose. - 8 - l A wide area network was built to allow system integration and data aggregation. A centralized provincial level pension insurance database management system has been constructed. Networks were built for pension administration on the provincial level as well as at the municipal level in the city of Jiamusi. l A systems operating site, including space for mainframe and server storage, was renovated and put in use. l Training sessions were carried out for staff responsible for transferring historical data to the digitized platform, as well as financial management and network and software maintenance. Qingdao l All hardware was procured via the Bank's open bidding procedures, resulting in substantial savings, transparency and fairness amongst vendors. l All data stored on the historical data storage system decentralized to municipal offices were transferred to the newly created data management systems at the district and city level. l Non-electronic, original data in cities [and at the provincial level] were successfully converted to digitized form, from which four types of databases were built: (i) a database backup to store raw data; (ii) a database to support daily operations; (iii) a database supporting customer service; and (iv) a database supporting policy decision making. During the process of inputting non-electronic data, all data were screened for quality and consistency, ensuring a higher quality of data entered into the databases. Application software for the four databases was developed to process data for each respective purpose. l Created an information management network system that linked district and sub-district offices to the Qingdao municipal labor and social security bureau. With the support of the project, the system is now centralized at the municipal level, and allows access by all levels of pension system participants. For example, the labor and social security bureau can access the system to determine which enterprises are lagging in contribution remittances, employment units can access the system to input changes in the status of individual employees, and individual pension system members can access the network to view personal pension contribution and benefit data. l Network systems were developed following a national standard protocol so that it can be integrated with the provincial and national level social insurance information system in the future. l Training sessions were carried out for relevant staff to understand and operate the new systems network and to maintain the newly developed software. Component 3: Training (US$0.9 million). This component is rated satisfactory. The purpose of this component was to support the identification by MOLSS and by the Heilongjiang and Qingdao labor and social security bureaus of personnel skills requirements and, thereafter, development and implementation of training programs to improve the skills of its personnel. - 9 - MOLSS In addition to the training discussed under Policy Development Component, MOLSS: l carried out 10 training seminars focusing on old-age pensions, Enterprise Annuities and information system maintenance to support the development of human resources; l took the lead in organizing overseas training in the Information Technology (IT) area with a 20-day study tour to Australia and Malaysia to study social insurance IT system management techniques; and l took the lead in organizing overseas study tours for relevant MOLSS, Heilongjiang and Qingdao staff to Poland and the United Kingdom to study their respective experiences in pension policy planning. Heilongjiang Domestic and overseas training were organized for bureau directors, division chiefs and section heads to better understand administrative and policy developments in other cities and provinces and overseas. Other more focused workshops included: l A week-long basic pension policy training workshop was held in 2001 for bureau staff responsible for pension insurance work in various cities. Training aimed to better communicate to managers the elements of the national unified pension system and emphasized the measures to improve pooling at the municipal level and ensure the provision of basic pension benefits. l A week long training course was held in 2002 to train staff on core issues of the basic pension system. Training content focused on the following policy documents: o "Circular of Heilongjiang Provincial Labor & Social Security Bureau on the Calculation of Basic Pension Benefits for Retirement of Employees of Special State Industries"; o "Regulations on the Basic Pension Insurance for Urban Enterprise Employees in Heilongjiang"; o "Regulations of the Basic Pension Insurance for Urban Self Employed Workers in Heilongjiang";and o "Methods for Contributing to Social Insurance in Heilongjiang". l A two week workshop was held in 2004 to study the policies for adjusting and improving the basic pension system in Heilongjiang. Contents of the workshop focused on: o The State Council Decision: "Approval of Heilongjiang as a Pilot Site for Implementing Improvements to the Urban Social Insurance System"; o "Methods for Calculation of Basic Pension Benefits for Enterprise Workers"; and o Other relevant regulations on improving the basic pension insurance system for urban workers in Heilongjiang. - 10 - Qingdao Domestic and overseas training sessions were organized for relevant staff to better understand administrative and policy developments in other Chinese cities and provinces and overseas. Programs included: l Two domestic study tours in 2002 to learn about experiences and measures in other cities to expand coverage and address enterprise contribution arrears. l Two conferences in 2002 to discuss domestic pension policy and draft policy documents. Attendees included academics and relevant persons involved in Qingdao's social insurance operation. The conferences focused on pension policy and regulations in Germany, the United Kingdom, the United States and Japan as reference points for the discussions. l A series of seminars were organized in 2002 for a total of 500 relevant staff to discuss and exchange ideas on how to enforce current pension policies. Participants included those involved in policy decisions, those in hands-on operational positions and staff from branch offices interfacing with the public. l A 10 day seminar in 2003 to train 100 relevant staff on pension regulations, pension operations and to exchange ideas on how to resolve certain operational issues. Component 4: Project Administration (US$0.5million). This component is rated satisfactory. This component aimed to provide assistance in project administration at the level of the MOLSS' PIU and PMOs in Heilongjiang and Qingdao, respectively. The PIU and PMOs all effectively carried out their organization and financial management responsibilities. The most important reason contributing to the satisfactory rating of this component was the allocation of sufficient and appropriate staffing to these units. The project was completed within the established budget, with substantial savings realized from the procurement processes employed. The project completion date was extended twice due to unforeseeable reasons including Severe Acute Respiratory Syndrome (SARS) which delayed domestic travel and visa restrictions set by international governments which delayed overseas study tours. Project completion was also delayed by reallocation of savings from the allocation to Heilongjiang to the MOLSS to make improvements to the MOLSS' actuarial model. The latter delay contributed to realizing the overall project objectives. 4.3 Net Present Value/Economic rate of return: A Net present Value (NPV) or Economic rate of return (ERR) is not quantifiable for this project. Components supported by the Bank credit include policy development, information system development, training and project implementation. While all components were executed satisfactorily and will contribute positively to the financial sustainability of the pension system, the potentially direct financial benefits are not quantifiable. 4.4 Financial rate of return: A financial rate of return (FRR) was not calculated for this project. 4.5 Institutional development impact: The project's institutional development impact is rated as highly satisfactory. The most lasting impact of the project is the training provided to relevant staff involved in operations and policy-making for pension insurance. Training sessions supported by the project deepened and broadened the understanding of - 11 - participants of issues facing China's pension system. These training sessions and domestic and overseas study tours exposed participants to new ideas for pension reform that can be adopted to their own jurisdictions. The process of designing the information systems and ultimately the procurement of computer hardware procurement and software development [yet software was not procured] to automate and network the pension insurance information management systems in Qingdao and Heilongjiang will have a sustainable impact. These measures have and will continue to improve the quality of data recorded into the system, cut down data processing time, and provide a more complete and accurate picture to administrators and policy-makers of participation and compliance rates as well as the systems' financial status. These information systems were established in line with the broad specifications provided by the MOLSS with the plan that they will eventually be linked to the national system, thus serving as a tool for central policy decisions. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: International and local hardware vendors have a strong interest in gaining market share in China and were therefore very aggressive in their bidding to supply project financed systems hardware. This, along with using the Bank's procurement process led to substantial cost savings. The enthusiastic attitude of staff involved in this project, especially in Heilongjiang and Qingdao contributed to the implementation and favorable outcome. The managers at the PIU and PMOs were mindful of the project requirements as specified by the project objectives and implemented tasks accordingly. 5.2 Factors generally subject to government control: N/A 5.3 Factors generally subject to implementing agency control: N/A. 5.4 Costs and financing: The project was completed within the original budget of US$6.2mn. Where there were savings as in Heilongjiang and MOLSS components, Bank approval was sought and received to re-allocate the funds towards technical assistance to the MOLSS to develop an actuarial model. 6. Sustainability 6.1 Rationale for sustainability rating: Sustainability of the project's achievements is highly likely given that the thrust of these achievements is consistent with the government's pension reform directions. The integration of the project design with the government's own development objectives is essential to achieving such sustainability of results. Since the project launch in 1999, the government has consistently pursued reforms towards a sustainable pension system in parallel with the activities of this project. The establishment of the National Social Security Fund in 2000 as a future source of funding for local governments with difficulty paying legacy pension benefits and the pilot reforms from 2001 to 2005 in the northeastern provinces of Liaoning, Jilin and Heilongjiang are clear examples of the government's commitment to further building a sustainable pension system. Regulations for a voluntary supplementary pension, Enterprise Annuities,which were announced in 2004 is another action reinforcing the government's intention to make policy decisions - 12 - supportive of a sustainable and diversified pension system. State Council Document No. 38 which calls for extending the successful parametric changes realized in the northeastern pilot provinces to the rest of China during 2006 is the government's latest initiative towards pension reform. Information systems development supported by the project is consistent with the national plan to upgrade pension insurance information management to be included as part of the Jinbao national systems development project. The Jinbao project was approved by the State Council in 2003 and, upon completion in five years, is intended to support program management, service delivery, fund supervision and management, as well as policy analysis and evaluation of social insurance programs at the central, provincial and municipal levels of government. 6.2 Transition arrangement to regular operations: For the same reasons cited under the Sustainability section above, it is highly likely that the government will continue to provide the appropriate technical, financial, and institutional resources necessary to ensure the continuity of the policy development and implementation and information system supported by this project. The recently issued State Council Document No. 38 discussed above is just one example of the government's continued commitment to pension policy reform. The hardware and software procured with project funds to build information systems in Heilongjiang and Qingdao are in accordance with the central government plans to build a system that will eventually be networked at the national level. While the project provided support, the government has been making its own investments in this area and is expected to continue to do so to maintain and upgrade the systems as necessary. With respect to the continued development of the short to medium term actuarial forecast model, PSCPF, while the model is capable of accepting and processing parameters defined in regulations issued up to 2004, it will need further programming changes in order to accept input parameters as defined by later regulations. The developers of this model have indicated that further financial support would be necessary for these programming changes to occur. 7. Bank and Borrower Performance Bank 7.1 Lending: The Bank's performance in identification, preparation and appraisal of the project was satisfactory. As discussed in the previous sections, the objectives of this project are consistent with the Bank's country assistance strategy as well as with the government's pension system development strategy. The appraisal document was thorough in providing the background and rational for the objectives of this project, and provided clear performance indicators as a guideline for project implementation. Most importantly, the project was prepared jointly with the participation and expressed interest of the MOLSS, Heilongjiang provincial and Qingdao municipal representatives resulting in substantial ownership and commitment to project implementation. 7.2 Supervision: The Bank's performance in supervising this project was satisfactory. Since the execution period for this project spanned over 5 years, the project leader was changed twice although there was consistency in the makeup of the project team from identification through implementation. However this did not seem to have - 13 - any negative impact on the supervision and monitoring of the project. Supervision missions were in accordance with the Bank's requirements and reports following the missions gave useful assessments of project status as well as useful feedback and suggestions for resolving issues and `next-steps'. 7.3 Overall Bank performance: The Bank's overall performance on this project is satisfactory. Borrower 7.4 Preparation: The PRC government's performance at preparation of the project was satisfactory. 7.5 Government implementation performance: The PRC government's implementation performance was satisfactory. As discussed previously, the objectives of this project are consistent with that of the government's and it is therefore in the interest of the government to be supportive by providing appropriate staff and counterpart funds. 7.6 Implementing Agency: The implementing agency's performance was satisfactory. The staff members assigned to this project at the central level, in Heilongjiang and Qingdao were all capable and enthusiastic in participating in this project and implemented their respective tasks earnestly. They were mindful in following guidelines such as those set by the Bank for the procurement process through competitive bidding and took pride in the results of the project as well as what they learned in the process. 7.7 Overall Borrower performance: The Borrower's overall performance on this project is satisfactory. 8. Lessons Learned The most significant lessons learned from this project are: l The importance of aligning project objectives with that of the Borrower's. With the objectives of both sides aligned, the project received strong administrative support from the government representatives in the MOLSS, Heilongjiang and Qingdao, in the PIU and the PMOs. There was a clear and shared sense of purpose in moving the project forward. l The practicality of focusing the project on one province and one municipality as pilots with oversight from the central government. Given the differences between provinces and municipalities and the varying issues they face in the pension development, piloting the project components in Heilongjiang and Qingdao kept the administration manageable. l In terms of project management, a lesson learned from this project is that there should be better clarity as to the responsibilities and repayment obligations at the onset of project launch. Project implementation was delayed in Jiamusi municipality due to unclear agreement regarding responsibility for repayment of the subloan between the central government and the municipality which is the common practice for projects of this sort in China. The Heilongjiang PMO reported that a large amount of Bank funds were kept in the Finance Bureau of Jiamusi for over half a year because of a lack of agreement between the municipal Finance Bureau and the Labor and Social Security Bureau regarding the repayment obligations. As a result, the suppliers could not get payment after considerable delivery. This delay could perhaps be avoided in the future by having better clarity prior to project launch. - 14 - 9. Partner Comments (a) Borrower/implementing agency: MOLSS as the borrowing agency concurred with this ICR and agreed that the project provided valuable lessons in many aspects which were subsequently applied in other activities. (b) Cofinanciers: (c) Other partners (NGOs/private sector): 10. Additional Information - 15 - Annex 1. Key Performance Indicators/Log Frame Matrix Output/Impact Indicators: Indicator Projected in SAR/PAD Actual/Latest Estimate Project the financial viability and Same as indicator. Directly or indirectly provided sustainability of the Qingdao municipal input to the following national pension system and the Heilongjiang policies: provincial pension system and from such projections, test the implementation of key l State Council Document 35 policy parameters embodied in State Council and 36 on the Decision to Document 26. Following this projection, Pilot Social Insurance System provide input to national policy-makers on key Improvements in Jilin and parameters of policy and implementation. Heilongjiang, respectively. l MOLSS Document 20 on Trial Measures for Enterprise Annuity. l MOLSS Document 23 on Measures on Management of Enterprise Annuity Funds for Trial Implementation. l State Council Document 38 on Improving the Basic Social Insurance System for Enterprise Workers. Install, test and review wider applicability of a Same as indicator Improved operational efficiencies. new model of pension data collection, Examples of some of the processing and management. improvements are listed in Note 1 and Note 2 below. - 16 - Indicator Projected in Actual/Latest Estimate SAR/PAD Develop and carry out Same as indicator Training for Heilongjiang, training programs for Qingdao and MOLSS staff on MOLSS, provincial and pension system development and municipal staff. information management related issues including: l Knowledge of regulations regarding municipal and provincial pension systems. l Timely implementation methods. l Experience of other cities, provinces and countries in drafting pension regulations. l Statistical methods. l Design and administration of IT systems, database management and application of new software. l Improve service efficiency of county and district level staff in dealing with pension participants. l Using the Bank's PROST. - 17 - Note 1: Heilongjiang No. Key indicator Efficiency Before After Changing rate (%) 1 Transacting insurance participation 25 100 400 affairs by one clerk 1 (person-time /hour/person) Collecting contribution (person/hour) 8,000 17,000 212 2 Benefits inspection (person-time /hour) 2 30 60 200 3 Inspection (case/hour) 3 1,200 38,000 3,167 Benefits distribution (person/hour) 10,000 30,000 300 4 Accounting (hours taken to transact a full 7 0.5 -7 account affair) 5 Periodic statistic report (working days 4 56 5 -8.5 taken per month (hour*person)) 6 Decision support analysis (working days 52 16 -30 taken per month [hour*person]) 7 Working hours taken for every one 300 30 -10 hundred new participants (person*hour) 1Participating procedure refers to accepting the application of the new entrants to the social insurance, establishing the social insurance documents for them, registering and issuing individual account manuals. It also refers to annulling the social insurance for those who terminate their social insurance, transforming or continuing the social insurance for those who exit or enter the pooling area. 2Benefit auditing includes auditing and inspecting basic pension benefit which consist of checking and ratifying applicants' retired benefit, benefit adjustment, one time paid benefit, benefit for direct dependants and etc. Only by calculating these checking and ratifying the pension benefit amount of the applicant, the checking and ratifying opinion and the benefit amount can be noted down. 3Benefit inspection includes auditing and inspecting whether the cardinal number and contribution rate of the social basic pension is complied with the concerning laws and regulations, whether the declared amount is consistent with his or her situation and whether the document of declaration is all ready. 4Periodic statistic report refers to fixed report forms filled out regularly (monthly, seasonally or annually) to the supervisor or authorities. - 18 - Note 2: Qingdao No. Key indicator Efficiency Before After Changing rate (%) 1 Transacting pension participation 30 120 400 procedure (increase, decrease, transfer leave, terminate and set up file, etc) (person/time/hour/person) Collecting contribution (person/hour) 10,000 20,000 200 2 Benefits auditing (person/hour) 30 60 200 3 Inspection (case/hour) 1,000 40,000 4,000 Benefits distribution (person/hour) 12,000 3,000 250 4 Fund accounting (hours taken to transact 8 0.5 -6 a full account affair) 5 Periodic statistic report (working days 5 72 8 -11 taken per month [hour*person]) Decision support analysis (working days 48 12 -25 taken per month [hour*person]) 7 Working hours taken for every one 20 2 -10 hundred new participants (person*hour) 5Periodic statistic report refers to fixed report forms filled out regularly (monthly, seasonally or annually) to the supervisor or authorities. - 19 - Annex 2. Project Costs and Financing Project Cost by Component (Appraisal Estimate) Component Category Cost Incl. % of Total Bank-financing % of Contingencies (US$M) Bank-financing (US$M) Policy Institutional 0.7 11.1 0.7 14.0 Development Development Information Institutional 4.2 66.7 3.5 70.0 Systems Development Training Institutional 0.9 14.3 0.8 16.0 Development Project Institutional 0.5 7.90 Administration Development Total 6.3 100.0 5.0 100.0 Project Costs by Component (Actual/Latest Estimate) Component Bank-financing(US$M) Government Total (US$M) Policy Development 0.47 0.02 0.49 Information Systems 3.14 0.60 3.74 Training 0.75 0.03 0.78 Project Administration 0.10 0.19 0.29 Total 4.46 0.84 5.30 - 20 - Project Cost by Procurement Arrangements Component Procurement Method Total ICB NCB Other N.B.F. Goods 2.60 0.45 0.35 0.60 4.00 Consulting 0.90 0.90 Services Training 0.80 0.10 0.90 Operating Cost 0.50 0.50 Total 2.60 0.45 2.05 1.20 6.30 Project Cost by Procurement Arrangements (Actual/latest Estimate) Component Procurement Method Total ICB NCB Other N.B.F. Goods 1.95 0.36 0.83 0.60 3.74 Consulting 0.47 0.02 0.49 Services Training 0.75 0.03 0.78 Operating Cost 0.10 0.19 0.29 Total 1.95 0.36 2.15 0.84 5.30 - 21 - Annex 3. Economic Costs and Benefits - 22 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 10/7/1998 2 TEAM LEADER (1); SOCIAL PROTECTION SPEC (1) Appraisal/Negotiation 3/1/1999 7 TEAM LEADER (1); SOCIAL PROTECTION SPEC (3); ECONOMIST (1); FM SPECIALIST (2) Supervision 07/16/1999 3 TEAM LEADER (1); PROST S S SPECIALIST (1); SOCIAL PROTECTION SPEC (1) 03/30/2000 5 TASK MANAGER (1); S S PENSIONS MODELING SPEC (1); SOCIAL PROTECTION SPEC (1); SOCIAL PROT IT SPEC (1); TASK ASSISTANT (1) 03/21/2001 2 TASK MANAGER (1); S S SECTOR COORDINATOR (1) 03/30/2002 5 ECONOMIST-TASK S S MANAGER (1); ECONOMIST (1); PROCUREMENT SPECIALIST (1); FIN MGMT SPECIALIST (1); TASK ASST/TRANSLATOR (1) 04/23/2004 3 TASK MANAGER (1); S S PROCUREMENT SPECIALIST (1); FM SPECIALIST (1) 09/27/2004 2 TTL (1); PROGRAM S S ASSISTANT (1) 1/19/2005 3 TASK MANAGER (1); S S PROCUREMENT SPECIALIST (1); FM SPECIALIST (1) ICR 9/26/2005 2 TTL (1); PROGRAM S S ASSISTANT (1) - 23 - (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation NA Appraisal/Negotiation NA 177,755 Supervision NA 521,894 ICR NA Total 699,640 - 24 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA No other social related objectives are applicable here. Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA No other objectives are applicable. - 25 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 26 - Annex 7. List of Supporting Documents Aide-Memoires, Back-to-Office Reports, Project Status Reports, and Provincial Reports. 10th Five-Year Plan for China's Pension Insurance Development and Targets for 2010. Document No. 26: State Council Decision on Establishment of Unified Basic Old-age Insurance System for Enterprise Employees Document No. 20: Trial Measures for Enterprise Annuity. Document No. 23: Trial Measures for Managing Enterprise Annuity Funds. Document No. 36: State Council Decision on Refining the Basic Old-age Insurance System in Heilongjiang Document No. 38: State Council Decision on Improving the Basic Old-age Insurance System for Enterprise Employees. Labour and Social Insurance Information Management System Training Handbook. Circular of Heilongjiang Provincial Labor & Social Security Bureau on the Calculation of Basic Pension Benefits for Retirement of Employees of Special State Industries. Regulations on the Basic Pension Insurance for Urban Enterprise Employees in Heilongjiang. Regulations of the Basic Pension Insurance for Urban Self Employed Workers in Heilongjiang. Methods for Contributing to Social Insurance in Heilongjiang. Approval of Heilongjiang as a Pilot Site for Implementing Improvements to the Urban Social Insurance System. Methods for Calculation of Basic Pension Benefits for Enterprise Workers. - 27 - Additional Annex 8. Appendices: Coverage and Contribution Data for Heilongjiang and Qingdao a. Coverage and contribution data for Heilongjiang. Appendix A1: Basic Pension Coverage in Heilongjiang Province, 1999-2003 Active workers who are participants Active workers who should be participants (1,000 persons) (1,000 persons) Coverage (%) Enterprises, Enterprises, Enterprises, individuals and individuals and individuals and Year Government others Government others Government others Total- organizations organizations organizations and public Total- and public Total and public Sub institutions Sub institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individuals 1999 4,662 48.0 4,564 50 2000 4,664 66.0 4,511 87 2001 5,140 532.0 4,469 139 5,211 536 4,513 162 986 99.3 99.0 85.8 2002 5,024 478.0 4,278 268 5,121 531 4,310 280 981 90.0 99.3 95.7 2003 5,183 503.0 4,277 403 5,278 559 4,305 414 982 90.0 99.3 97.3 Note: Profit making public institutions included under enterprises. Appendix A2: Number of Basic Pension Participants in Heilongjiang Province, 1999-2003 Total pension participants (active Active workers who are Retired pension participants workers + retirees)(1,000 persons) participants (1,000 persons) (1,000 persons) Enterprises, Enterprises, Enterprises, individuals individuals individuals and Year Government and others Government and others Government others Total- organizations organizations organizations and public Total- and public ls Total and public Sub institutions Sub institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individua 1999 617.1 64 6,057 50 4,662 48 4,564 50 1,509 16 1,493 2000 624.9 83 6,079 87 4,664 66 4,511 87 1,585 17 1,568 2001 692.4 639 6,145 140 5,140 532 4,469 139 1,784 107 1,676 1 2002 689.8 594 6,030 274 5,024 478 4,278 268 1,874 116 1,752 6 2003 714.2 631 6,097 414 5,183 503 4,277 403 1,959 128 1,820 11 Note: Profit making public institutions included under enterprises. - 28 - AppendixA3: AverageContributionRateto BasicPensioninHeilongjiangProvince,1999-2003 Employeraveragecontribution Individualaveragecontribution rate(%) rate(%) Pensioncollectionrate(%) Enterprises, Enterprises, Enterprises, individualsand individualsand individualsand YearGovernment others Government others Government others organizations organizations organizationsand andpublic andpublic Total public institutions institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individuals 1999 - 25.1 0 - 4.2 16 86.5 92.0 86.2 - 2000 35.4 25.5 0 1.2 5.4 16 87.6 95.8 87.5 - 2001 30.7 24.1 0 5.7 6.2 20 91.3 95.0 90.9 - 2002 23.8 22.9 0 7.1 6.9 20 95.5 95.6 94.9 - 2003 26.1 23.2 0 4.6 7.9 20 93.5 95.8 93.1 - Note: Profitmakingpublicinstitutionsincludedunderenterprises. Appendix A4: Pension Benefit Level of Retired Persons in Heilongjiang Province, 1999-2003 Number of retired persons (1,000 persons) Average benefits (RMB) Substitution rate of pension (%) Enterprises, Enterprises, Enterprises, individuals and individuals and individuals and Government others Government others Government others Year Total- organizations organizations organizations and public Total- and public Total and public Sub institutions Sub institutions institutions erprises Enterprises Individuals Ent Individuals Enterprises Individuals 1999 1,509 16 1,493 - 445 - 443 - 84.7 - 86.2 - 2000 1,585 17 1,568 - 487 - 489 - 89.1 - 89.3 - 2001 1,784 107 1,676 1 511 666 500 - 88.1 82.8 89.3 - 2002 1,874 116 1,752 6 550 805 523 - 85.5 83.3 90.8 - 2003 1,959 128 1,820 11 564 847 545 - 83.2 83.2 83.2 - Note: Profit making public institutions included under enterprises. - 29 - Appendix B1: Basic Pension Coverage in Qingdao Municipality, 1999-2003 Active workers who are participants Active workers who should be (1,000 persons) participants (1,000 persons) Coverage (%) Enterprises, Enterprises, Enterprises, Year individuals and individuals and individuals and Government others Government others Government others Total- organizations organizations organizations and public Total- and public Total and public Sub institutions Sub institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individuals 1999 803 80 715 8 830 80 740 10 96.7 100 96.6 80.0 2000 904 110 779 15 910 110 780 20 99.4 100 99.9 75.0 2001 916 120 773 23 970 120 821 29 94.4 100 94.2 79.3 2002 959 120 784 55 990 120 810 60 96.9 100 96.8 91.7 2003 1,065 160 821 85 1,150 160 902 88 92.6 100 91.0 96.6 Note: Profit making public institutions included under enterprises. Appendix B2: Number of Basic Pension Participants in Qingdao Municipality, 1999-2003 Total participants (active workers + retirees) Active workers who are participants Retired pension participants (1,000 persons) (1,000 persons) (1,000 persons) Enterprises, Enterprises, Enterprises, individuals and individuals and individuals and Year Government others Government others Government others total- organizations organizations organizations and public total- and public ls Total and public Sub institutions Sub institutions institutions Enterprises Individuals Enterprises Individua Enterprises Individuals 1999 1,071 90 973 8 803 80 715 8 268 10 258 0 2000 1,202 130 1,057 15 904 110 779 15 298 20 278 0 2001 1,242 140 1,079 23 916 120 773 23 326 20 306 0 2002 1,302 140 1,107 55 959 120 784 55 343 20 323 0 2003 1,446 200 1,161 85 1,065 160 821 85 380 40 340 0 Note: Profit making public institutions included under enterprises. - 30 - AppendixB3: Average Contribution Rate of Basic PensioninQingdao Municipality,1999-2003 Employer average contribution rate Individualaveragecontributionrate Pension collection rate (%) (%) (%) Enterprises, Enterprises, Enterprises, Government individualsand individualsand individuals and Year organizations others Government others Government others total- and total- organizations organizations and public Total andpublic uals Sub publicinstituti Sub ons institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individ 1999 23.4 22 23.6 0 7.6 2 7 18 96.6 98 96.0 100 2000 23.5 23 23.6 0 8.4 2 8 18 96.6 97 96.2 100 2001 23.6 24 23.6 0 8.4 2 8 18 96.8 98 96.2 100 2002 23.8 24 23.8 0 8.6 2 8 20 97.2 99 96.6 100 2003 22.7 27 22.2 0 8.6 2 8 20 97.1 99 96.5 100 Note: Profit makingpublic institutions included under enterprises. Appendix B4: Pension Benefit Levels of Qingdao Municipality Retirees, 1999-2003 Number of retired persons Personal average benefit Substitute rate of pension (%) (1,000 persons) (RMB Yuan) Enterprises, Enterprises, Enterprises, individuals individuals individuals Year Government and others Government and others Government and others total- organizations organizations organizations and public total- and public Total and public Sub institutions Sub institutions institutions Enterprises Individuals Enterprises Individuals Enterprises Individuals 1999 268 10 258 0 574.4 1360 544 0 91.9 105 90.4 - 2000 298 20 278 0 638.2 1420 582 0 90.4 106 88.7 - 2001 326 20 306 0 644.6 1480 590 0 85.4 105 83.2 - 2002 343 20 323 0 693.8 1530 642 0 84.6 106 82.2 - 2003 380 40 340 0 773.8 1640 672 0 83.3 106 80.8 - Note: Among enterprises, include public institutions that are under the management of enterprises. - 31 - - 32 -