89249 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The World Bank does not guarantee the accuracy of the data included in this work. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The material in this publication is copyrighted. FINANCIAL SECTOR ASSESSMENT PROGRAM ALBANIA DEVELOPING PRIVATE PENSIONS IN ALBANIA TECHNICAL NOTE FEBRUARY 2014 This Technical Note was prepared in the context of a joint IMF-World Bank Financial Sector Assessment Program mission in Albania during October - November 2013. The note contains technical analysis and detailed information underpinning the FSAP assessment’s findings and recommendations. Further information on the FSAP program can be found at http://www.imf.org/external/np/fsap/fssa.aspx, and www.worldbank.org/fsap. THE WORLD BANK FINANCIAL AND PRIVATE SECTOR DEVELOPMENT VICE PRESIDENCY EUROPE AND CENTRAL ASIA REGIONAL VICE PRESIDENCY i Contents Page EXECUTIVE SUMMARY 1 I: CONTEXT AND CURRENT OUTCOMES 4 A: Introduction 4 B: Political and Economic Environment 4 C: Retirement Income System 6 D: An Outcomes Based Assessment 9 II: SOLUTIONS 11 A: Lessons from Successful Implementation 11 B: Setting out the Vision 11 C: Costs 13 D: Investments and Capital Markets 15 E: Coverage and Informality 18 F: Security and Supervision 20 III: RECOMMENDATIONS AND A POTENTIAL TIMELINE 22 ANNEX: Evidence from Brazil – Special programs for formalization and pension coverage. 27 Boxes Box 1: The experience in Kosovo in establishing a diversified system Figures Figure 1: Financial Inclusion (Coverage) % age 15+ with a financial account Figure 2: Diversified Systems have Multiple Income Sources to Offset Multiple Risks Figure 3: Gross Replacement Rates from First and Second pillars at 0.5 and 1.5 times Average Earnings Figure 4: Market Structure: Design Options to Improve Cost and Investment Outcomes Figure 5: Percentage of Working Age Population, Ages 15-64, Contributing to the Pension System Figure 6: Gains in Coverage of Pensions in Brazil by Type of Contributor Figure 7: A Spectrum of Coverage from Mandatory to Voluntary Figure 8: A Potential Timeline for Implementation to Build a Diversified Pension System Tables Table 1: Recommendations Table 2: Global real investment returns in 19 Countries 1900 – 2012 ii GLOSSARY EC European Commission ECA Europe and Central Asia EU European Union FSA Albania Financial Services Authority IFC International Finance Corporation INSTAT Albania Statistical Agency IORP Institutions for Occupational Retirement Provision LAC Latin America and Caribbean Region MIFID Markets in Financial Instruments MIGA Multilateral Investment Guarantee Association OECD Organisation for Economic Cooperation and Development OSCE Organisation for Security and Cooperation in Europe SII Social Security Institute UCITS Undertakings for Collective Investment in Transferable Securities 1 EXECUTIVE SUMMARY 1. This note sets out options to develop private pensions in Albania. It complements analysis by the World Bank of reform options for state pensions (World Bank 2013a). In combination they set out options for the crucial decisions on pensions required in the short, medium and long term. The diversified approach and options suggested draw on international lessons1 targeted to Albania’s particular characteristics (including high levels of labor market informality). But they also support recommendations in the 2012 EU White Paper on Aging to the existing EU members (European Commission 2012a). So there is a good fit with the future for which Albania is preparing. 2. Albania is a relatively young country but is rapidly aging. This is compounded by emigration of key working age groups. The 2011 census shows an 8% fall in overall population since 2011 from 3.1m to 2.8m. The average age has risen from 30.6 to 35.3. The proportion of the population aged over 65 rose from 8% to 11%. This is still low, particularly relative to EU neighbors. But by 2050 this is forecast to rise to 25% - above the expected regional average of 23%. Low birth rates, improved longevity and significant emigration of core working age groups all act to worsen dependency ratios. 3. Fortunately, over the past decade Albania has been one of the fastest growing countries in Europe. Between 2002 and 2008 poverty in the country fell by half and extreme poverty affects less than 2% of the population (World Bank 2013b). However, these positive factors will raise expectations about future pension provision, as well as raise life expectancy. So it makes it important to plan now for how to deliver a decent retirement income that eradicates poverty and boosts shared prosperity in a sustainable way. 4. The current system in Albania is dominated by an unsustainable state pension system (first pillar). Poor incentives and informal labor markets mean contributions are vastly below required levels. Some positive reforms such as increases in the retirement age have been offset by reduced contributions and increased generosity. 5. First pillar reform is urgent (and is currently being considered by the government). Sustainability must be improved. Fiscal sustainability longer-term will be helped by automatically linking future increases in the retirement age to increased longevity – building in evidence-led adaptability in policy. Identifying non-wage labor sources of funding will improve the prospects for employment and boost formality. It will help if these decisions needed for fiscal sustainability are set out alongside a commitment to a diversified pension system so people can see a viable path to good pensions in the future. 6. A critical part of building and sustaining a long term vision for pensions is to develop government, political and public understanding and agreement. Progress should be started immediately by establishing a Pension Commission. This can build on past efforts so that the strategic and technical issues can be substantially advanced before the next election in 2017 with final decisions soon after. Without focused attention in the next few years the introduction of a second pillar / auto-enrolment to diversify the system by 2020 would not be attainable – pushing off even further the achievement of a sustainable system. 1 OECD (2012) Pensions Outlook, World Bank (2012c) International Patterns of Private Pension Provision, Price and Rudolph (2013) Reversal and Reduction, Resolution and Reform, World Bank. 2 7. The Government should seize the opportunity to reap the benefits of the ‘science of delivery’. It should develop a deep understanding of data and attitudes behind the dynamics in the formal and informal labor market. It should pilot solutions so the ultimate design works in the Albania context. The proposed Pension Commission should also set clear objectives for the new system through an outcomes based assessment. One example is aiming for best in class investment at a long term cost of 0.6% a year or lower. This has been achieved in countries from Kosovo to Sweden. They, and other countries, provide lessons on how to create scale, expertise, member-focused governance and cut distribution costs through an intelligent market structure and a robust administration platform. 8. The governance, potential use of guarantees, and modeling of future investment strategies must balance the evidence on maximizing efficiency and long term returns with the need to build trust and confidence. Albania currently has relatively low financial inclusion and faces the continuing legacy of the 1990s pyramid scandal on trust and confidence. 9. Creating the necessary foundations for a robust and diversified pension system will enhance and build upon the efforts being made as part of the EU accession process . A number of the recommendations pick up on themes identified in the latest EU progress report. They also fit well with the EU’s own agenda on pension reform as set out in their 2012 White Paper on Aging. The recommendations also address issues identified in the 2005 FSAP and in a range of expert reviews submitted to the government. 10. A central issue to meet international standards for security is to ensure the independence and resourcing of the FSA and the development of robust risk based capacity. Planned amendments to the 2009 Pension Act to make it fully compliant with EU and international standards should be passed and the barriers to providing tax incentives should be removed. Fixing the independence and resourcing of FSA is vital to allow effective supervision now and preparation for an expansion in private pensions. Other key issues include: developing the payout phase with intelligent phased withdrawal, inflation-linked bonds in the longer term and developing joint-life annuities for the future to improve gender equality; and the progressive development of domestic investment opportunities, building on innovative IFC and MIGA projects in a range of sectors. 11. The timelines for pension reforms are long-term. This can obscure the need and possibility for action now. The recommendations have been set into a potential timeline to help breakdown the challenge into practical steps. The reform process could follow a ‘launch-design-build-operate’ program starting in 2014 and leading to the start of the new system in 2020. This is just one option – but one in which the World Bank and other partners stand ready to support the leadership of the Albanian authorities. 3 Pension Reform Recommendations and Potential Timeline Term Priority Lead Take decisions to put first pillar pensions on fiscally sustainable path – High Min Welfare including options to fund from general revenue rather than wage based 2013 contributions (Ministry of Welfare and Youth). Establish cross government and external Pension Reform Commission to build High Min Welfare a diversified pension system and start design of data collection and pilots to support reform. - Present vision for public and private system with long term goal of efficient, High Min Welfare sustainable, secure pensions with broad coverage and decent adequacy. 2014 Ensure FSA has independence and adequate resources in line with High Min Welfare international standards for current and future role. Ensure tax relief for third pillar (voluntary pensions) delivered simply and Med Min Finance quickly to workers and employers and amend pension act for EU compliance. Implement pilots to test incentives and enrolment mechanisms in formal and Med Min Welfare informal workplaces to support 2nd pillar / auto-enrolment design. 2015 Implement detailed baseline survey of employment patterns and employer and MOW/FSA/ employee attitudes to support 2nd pillar design – run by Institute of Statistics Med INSTAT (INSTAT) or Ministry of Welfare and Youth. Pension Reform Commission outcome based assessment of 2nd pillar models - High Min Welfare for governance, administration and investment. Conduct feasibility test on Social Security Institute (SII) as platform for 2nd High SII pillar administration. 2016 Deliver Capital Market Enhancement Program. Med MOF / FSA Draft legislation following pilots and testing to deliver shared vision for High Min Welfare pensions in 2050 for introduction after 2017 election. Develop FSA Risk Based Supervision capacity for Pensions. Med FSA Pass new 2nd pillar legislation setting out contribution rates, coverage, market High Min Welfare 2017 - 2019 2020 structure, investments and supervision. Competitive tender/licensing for administration and investment services. High MOW/FSA Build operations (in public and private sector). High SII/ Private Providers Communications plan and data collection for employers and employees. Med Min Welfare Continue capital market development program. Med MOF / FSA st nd Welfare/Fin Reformed 1 Pillar joined by new 2 pillar ance/FSA/ with continuing 3rd pillar High Providers/ Employers 4 I. CONTEXT AND CURRENT OUTCOMES2 A. Introduction 12. This note sets out options to develop private pensions in Albania. It has been requested by the authorities to help create a more diversified pension system that can deliver improved outcomes for Albania now in decades to come. As set out below the current system in Albania is dominated by an unsustainable state pension system (first pillar). Even without the challenges it faces, international experience shows that diversified systems with a mix of public and private provision provide the best outcomes3. This note on how to develop the private pension side of a diversified system has been developed to complement analysis by the World Bank of reform options for state pensions (World Bank 2013a). 13. A diversified pension system mixes state and private pensions as well as other sources of retirement income. State provision comes in the form of direct transfers (a zero pillar), contributory state pensions – often with very significant redistribution in benefit formulas – (a first pillar), mandatory/quasi mandatory private savings (second pillar) and voluntary pensions (third pillar)4. In addition, retirement income can come from other forms of saving and assets including saving accounts, family transfers and housing (fourth pillar). Finally, of course, some people continue to work to provide income in ‘retirement’ to supplement a pension. 14. The analysis, diagnosis and recommendations use an outcome-based assessment framework. This Technical Note looks at how the pension system performs relative to 5 key outcomes – efficiency, sustainability, coverage, adequacy and security. These outcomes depend on the political and economic environment in which a pension system operates, and its overall design including regulatory framework, market structure and conduct and supervisory approach. B. Political and Economic Environment 15. Albania is a relatively young country but is rapidly aging. This is compounded by emigration of key working age groups. The 2011 census shows an 8% fall in overall population since 2011 from 3.1m to 2.8m. The average age has risen from 30.6 to 35.3. The proportion of the population aged over 65 rose from 8% to 11%. This is still low, particularly relative to EU neighbors. But by 2050 this is forecast to rise to 25% - above the expected regional average of 23%. Low birth rates, improved longevity and significant emigration of core working age groups all act to worsen dependency ratios. 2 This Technical Note has been prepared by Will Price, Senior Financial Sector Specialist, World Bank. It was prepared in the context of an FSAP assessment mission led by Michael Edwards, World Bank and S. Erik Oppers, IMF which are respectively overseen by the Financial and Private Sector Development Vice Presidency, World Bank and the Monetary and Capital Markets Department, IMF. 3 OECD (2012) Pensions Outlook, World Bank (2012) International Patterns of Private Pension Provision, Price and Rudolph (2013) 4 In this report third pillar is used to describe the pure individual voluntary system based on individual choice that currently exists in Albania and most other countries in the region. When options are being discussed for adding a second pillar, or moving to a more diversified system, they are focusing on using either compulsory or mandatory provision or techniques like auto-enrolment into employer provided plans. 5 16. There have been many positive developments in the Albanian economy over the past 20 years. Over the past decade Albania has been one of the fastest growing countries in Europe. Between 2002 and 2008 poverty in the country fell by half and extreme poverty affects less than 2% of the population (World Bank 2013b). These positive factors will raise expectations about future pension provision, as well as raise life expectancy, so make it important to plan now for how to deliver a decent retirement income that eradicates poverty and boosts shared prosperity in a sustainable way. The model that is suggested of a diversified pension system is what is also recommended in the 2012 EU White Paper on Aging to the existing EU members (European Commission 2012a). So there is a good fit with the future for which Albania is preparing. 17. There are a range of challenges in the business and legal environment. The EU accession process will provide a structured engagement to make progress on the key pre-conditions. The EU Accession process focuses, among other things, on progress in adopting EU laws and standards (the ‘acquis’) across 33 Chapters. They include all the key elements required for a well-functioning second pillar private pension system – from Company Law and Financial Services to Social Policy and Employment and the Judiciary. Accession will not need to be achieved to introduce a second pillar pension. But as the political, design and development elements of pension reform develop, they can be joined-up with elements of the Accession process to aid efficiency and political focus. 18. Overall employment rates are reasonable but there are important challenges in gender and for older workers that will affect pension outcomes. The employment rate of the population aged 20-64 is close to 60 %. Unemployment is close to 13 %. There is a significant gender gap with the employment rate for men 73 % and women 49% (European Commission (2012c). The employment rate for older workers – aged 55- 64 - is closer to 50 %. Improved employment for women and older workers is an important part of ensuring the best outcomes from the pension system. Gender inequality in saving for pensions can be mitigated through good product design – for example through joint annuities that continue to pay to a spouse on the death of the main account holder. 19. Much of the employment is in the informal labor market and over 50% of employment in the agricultural sector. This has led to very low contribution rates to the state pension (first pillar). Informality has also increased due to weak tax and expenditure policies (European Commission 2012b). Collection issues in the state pension system have been exacerbated by poor incentives to contribute, poor data collection and poor historic capacity in the Social Security Institute (SII). A set of reforms are attempting to deal with these issues, including the successful digitization of all the historic records. But the issue of informality will be one of the greatest challenges to pension reforms for all parts of the system. 20. The country has relatively low levels of financial inclusion. This is partly a legacy of distrust for financial services resulting from the pyramid selling scandal of the late 1990s. Fewer than 30% choose to have access to a formal financial account compared to 50% - 80% for the closest neighbors. This presents a challenge in terms of widening access. But pension reform presents a major opportunity to extend financial 6 inclusion. Final policy proposals should be formulated based on significant attempts to understand the views of employers and employees and develop options which are likely to gain the most consent – which is important even if a mandatory saving system is to be introduced. Figure 1: Financial Inclusion: % age 15+ with an account at a financial institution 21. The political environment has been challenging but with positive developments. Albania was recommended for EU Candidate Status in late 2013. The 2013 elections marked a peaceful transition in power. The OSCE has identified significant progress, whilst a number of key challenges still remain. Given the legacy of public distrust of financial institutions and political conflicts, it will be important to build government, political and public understanding of the need for change and sufficient support for eventual solutions. This is a process that the World Bank would be well placed to support, with other partners, to provide independent, objective and technical input into national dialogue. But because it may take many years to get to a final reform, it is important to start this process sooner rather than later. C. The Retirement Income System in Albania 22. As in many countries, pension reform has been a longstanding issue for Albania. Reforms to the state system (first pillar) and the development of the private system were both recommendations in the 2005 FSAP (IMF 2005). Some reforms to the state system were undertaken, and a voluntary pension system (third pillar) was established. But the creation of a diversified pension system that will deliver good outcomes for the coming decades is still substantially incomplete. 23. Pension income in Albania currently comes almost exclusively from the state pension. The state pension is a traditional ‘Pay As You Go’ system, where current contributions pay for current pensions. It has a split between urban and rural workers with different contribution levels and benefit structures. Since only 20% of the working 7 age population contribute to the first pillar (the lowest in the region) the system is financially unsustainable. Despite the country being young, the system is already has an annual deficit of close to 2% of GDP. And this deficit is despite many workers not being covered (because they have not contributed sufficiently). 24. Reform of the first pillar is an urgent issue that cannot wait for a new private second pillar to be introduced. But the short term and longer term reforms for public and private pensions can be developed in an integrated way . There have been some reforms to the state pension system over the years – for example contributions have been reduced from 29.9% in 2005 to 21.6% of covered earnings by 2009. The retirement age was raised from 55 for women and 60 for men to 60 and 65 respectively in stages between 2002 and 2012. But these were counteracted by other changes, meaning the situation is still critical. The analysis in this paper has been developed to work in the context of the reform currently being considered to the first pillar. 25. There is no second pillar (mandatory/quasi mandatory private pension saving) in Albania, but there is a small voluntary pension system (third pillar) . The current voluntary system is based on the 2009 Law on Voluntary Pension Funds (Law Number 10 197). There are 3 licensed operators – Raiffeisen, SiCRED and Sigal5. By 2013Q2 there were 7,784 individual customers and US$3.4m in total assets. The new system replaced the pre-existing system operated by pension institutes. Assets were transferred into the new system. 26. The providers follow a Pension Fund Management Company structure with a vertically integrated service offering including all administration and investment management – but with separate custodians. They can offer pension plans to individuals directly, or via setting up a plan for an employer. The employer market is more attractive given the costs of customer acquisition for each individual. But the plans are essentially the same in terms of asset allocation and structure. Employers can only change the level of contributions and can negotiate on charges. But unlike in some other countries they have no role in overall governance. 27. The market structure relies on the distribution network of other companies in the financial group – with direct sales and marketing as well as cross sales to existing customers. The reliance on a sales network increases costs and can lead to a situation where high fees do not translate into better investment results or services but merely serve to reduce investment returns. 28. The current fees charged are 3 per cent of assets under management a year. This is high by international standards – but not necessarily for a system in its early years. There are no additional entry fees or on-going fees. There is a fee of 0.5% to transfer to another provider. More importantly, there is a system of penalty fees for members who want to withdraw their money from their pension. These are very significant, with 15% of total assets charged for a withdrawal within the first 5 years, 10% for withdrawals between 5 and 10 years, 5% between 10 and 20 years and 2% after 20 years (if before eligible retirement). There is some possible justification for these fees 5 FSA (2012) Annual Report 2012, Republic of Albanania Financial Supervisory Authority 8 to help offset start-up costs compared to other systems which had higher costs in the future (although there is no necessary reason why the first contributors should bear the start-up costs). There is less justification if they cover sales and marketing. But there is no available data on rates of commission paid to help identify where the fees are going. In general it would be more effective to control early access in better ways, e.g. with complete bans, allowing up to 25% withdrawal over a full life time or having access to a rolling window of 2 years-worth of contributions. Certainly any second pillar or broader occupational pension coverage should not have such fees. Restrictions on withdrawals would be preferable to exit fees which in Turkey at least have not reduced exits and have provided a very significant source of fee income for companies. 29. FSA Regulations set the percentages allowed domestically and abroad and ranges for the different asset classes. Despite being allowed some limited investment abroad all three companies invest exclusively in Albanian assets – the majority of which are government bonds. This appears to be a combination of: lack of any clear long-term strategy for the pension assets that would distinguish them from an investment fund; very low yields in euro denominated markets; a desire to be cautious in the early stages of the market; and the fact that the assets are currently too small to justify doing very much more than the basics for investment. 30. Investment returns after costs in 2011 were 6.6% in nominal terms and 3.0% real. In 2012 they were 6.0% in nominal terms and 3.6% in real terms. These returns are around 3% above a 1 year term deposit and comparable to those from a newly launched investment fund (see below). Thus far there is no real track record to assess returns, but with a 3% fee the pension funds will not be providing any long term benefit for locking away assets and taking on (in theory) more investment risk compared to putting money in a bank account. Fees are however already under pressure following negotiations from one employer setting up a company pension plan. For that plan at least fees have fallen to 1.5%. For the second pillar to be viable a fee of well under 1% should be the target – and in the long term 0.6% or better given the evidence from near neighbors such as Kosovo and global best practice figure. 31. The Albanian Financial Supervisory Authority is the supervisor of private pensions. The 2009 Law has many positive features and will be further improved when recommendations to make it compliant with various EU directives and international standards are introduced. An extensive recent review (Ashcroft 2013) of the 2009 legislation has compared its consistency with the EU acquis relevant to voluntary pension funds (the IORP, UCITS, MiFID directives) and the Principles of the International Organization of Pension Supervisors (IOPS). The report provided suggestions for changes to legislation and FSA regulations. These have been accepted by FSA. FSA consulted the industry on the draft legislation in 2013. These changes should be passed. 32. Whilst the legislation will soon be a good platform for supervision, resource constraints in FSA mean that it is not possible to implement the legislation and associated supervisory approach fully. FSA has a limited number of supervisors and staff set by a detailed statute. It is facing a further cut of 30% in staffing. It also does not 9 have the independence needed by financial regulators. This means that it cannot currently supervise as it would choose. 33. Financial inclusion is low, as are saving accounts in addition to bank deposits and there are few if any life annuity products. However, 2 new investment funds have attracted considerable inflows. Both are operated by Raiffeisen. Total assets at end October 2013 stood at €317m (US$234m) spread across some 22,000 customers. It is not clear to what extent any of this provides savings that will be used to supplement retirement income and the exact nature of the customer base. 34. Homeownership rates are high - representing an important source of financial security for retirement and reducing the need for pension income relative to those having to pay rent. 71% of private dwellings are already owned without a mortgage. 90% of private dwellings are in the process of being owned. Homeownership without a mortgage when retirement is reached is a considerable financial advantage. Seen as a financial product, home ownership is like an annuity that pays out until death a value similar to the costs of renting. This has very desirable properties to reduce risk – whilst also leaving an asset for bequests. Set against this is the illiquidity of the asset and propensity of property markets to experience regular and sometimes severe cycles. D. An Outcomes Based Assessment 35. The success of a pension system can be judged against 5 outcomes: i) efficiency (particularly of costs and investment); ii) sustainability and affordability (for government, workers and employers); iii) coverage; iv) adequacy; and v) security (of assets and supervisory controls). Well-designed pension systems typically have multiple components including, for example, public and private, both flat rate and linked to earnings, and voluntary and mandatory components. Reforms to one part of the pension system should consider how they affect the combination of the different outcomes for the system as a whole. Important trade-offs exist, for example, between the adequacy and sustainability. 36. The main conclusions on each outcome are set out below. This sets out the challenges that need to be faced to maximize the different outcomes. Going forward the design of the new system could examine the options against the ultimate outcomes they produce. Efficiency - The new voluntary pension market clearly has small scale. But given that Albania is a relatively small country, scale will always be an issue. There are a lack of domestic investment opportunities and yet the potential concern of a high bias to domestic investments since none of the current providers use even the limited flexibility to invest abroad. There is no annuity market, the labor market is informal and there is a history of very poor contribution collection in the first pillar. Costs of the private voluntary pensions are high at 3% of assets under management per year (but falling to 1.5% potentially). They are, however, actually lower than many other pension systems at a similar stage of development. Asset returns are around 3% a year above bank deposits so far. So members are getting more than in bank accounts – but the majority of the excess return is going in fees and assets are locked away. Although the third pillar is 10 small it is a useful development and incorporates many of the key features required in the future. Hence it can help to test design and operational features that could be used in an expanded private pension system in the coming years. Sustainability – The pension system is still effectively reliant on a first pillar that is itself unsustainable. So far there is no political consensus or process to create a pension system that will need to operate for many years to come. The contributions as a share of wages are still high despite cuts in recent years. There is little or no public understanding or appreciation of the need for change. But changes to the first pillar – particularly if there is a move to more general revenue finance – will create the space for a contributory pension. Coverage – Ultimately only the state system can provide universal coverage but the current system is not doing that. The third pillar will see coverage grow but the experience of other countries suggests that this will remain below 10%. But even this level will not be reached unless the tax incentives offered are clarified and simplified to make them effective. Second pillar pensions, or mandatory provision with an employee opt out are needed to provide significant coverage on top of the state system. But it will need to deal with the issue of extensive labor markets informality. Improving the current collection infrastructure and the enforcement of contributions would need to be fixed going forward. Financial inclusion is low – not so much due to lack of supply but a history of mistrust due to pyramid selling scheme. Adequacy – The future level of pensions from the state pension will depend on the reforms chosen. An affordable replacement rate will probably see higher replacement rates for the lower income workers and falling rates as income rises. This reflects the current system already, where benefits are very compressed and similar nominal pension amounts produce lower replacement rates as wages rise. The reforms will determine how much the second pillar will need to generate and for whom. It will not need to cover workers with the lowest incomes – not only because their replacement rates from the first pillar will be higher but also because they have less money to contribute and are likely to be most difficult to include in a contributory pension. Current labor market trends show gender inequality in employment that will affect savings achieved. But good product design with a future of joint annuities can offset this – at least for couples. High levels of home ownership reduce risks somewhat. If the future system has a Defined Contribution design then the potential variability in ultimate income is an issue – and some consideration may be needed to guarantees if they can be inexpensively and intelligently designed. Security – There has been general progress on a range of the pre-conditions for a private system. The new 3rd pillar shows that both the legislation and the institutions can be created. But it is currently very small and an important part of the development of options would be to ensure it is well supervised and resources and capacity can be significantly scaled up. The EU membership process provides a positive dynamic. The critical issue is with supervisory independence and resourcing. Without fixing these a second pillar should not be introduced. Core data (like IDs) and mortality tables seem to be in place as does an increasing supply of people with professional finance qualifications and 11 international experience – coupled with openness to external companies that helps provide capacity. II. SOLUTIONS A. Lessons from Successful Implementation 37. The process of developing a diversified pension system can draw lessons from reform success and failures. This helps to design the approach now so that it builds in mechanisms to increase the chances of a good outcome. A central feature of the lessons learnt is the importance of having a clear link between the project and the government’s key strategic priorities, including agreed measures of success. So an important first step is making pension reform a priority – and using a focus on the ultimate outcomes to help set the goals. In addition there needs to be clear leadership and ownership from senior management and heads of the key agencies and ministries (including the Ministries of Welfare and Youth and Finance, the Social Security Institute and the Albanian Financial Services Authority). 38. Effective engagement with stakeholders is also a key element, in addition to the central importance of a clear vision and political and technical leadership. Contributory pensions are built over decades not weeks. So this means not only other political parties and the pension industry, but also employers and the people need to be progressively involved and engaged. The reformers need to have a very good understanding of the target market. Finally a central lesson is to ensure that there is the experience and capacity to lead major reforms. This is why the sequencing of potential reforms will be important – and making sure there are a set of clear practical steps that can be taken year by year that will make the ultimate objective of a good retirement for all Albanians a possibility. And as conditions change over the years the system can be made more adaptive – for example linking retirement ages to growing longevity automatically. So the sooner these changes are made the better. 39. The recommendations try to build in these lessons from the ‘science of delivery’. Firstly a pension commission is suggested with a mandate to build a diversified system and a framework that uses the ultimate outcomes as a way to judge the viability of different options. Secondly a significant increase in data and understanding of the specific types of workers and firms that may be included. Thirdly a focus on what works in Albania through the use of pilots. Capacity also needs to be tested at appropriate points and most importantly built in the FSA so that it will be able to play the necessary role if the private pension system were expanded. B. Setting out the Vision 40. A key part of the reform process is the vision to guide the reforms. Building a diversified system is suggested as a core part of this vision. This is based on the view that no one part of a pension system – public or private, flat rate or contributory – can achieve all the objectives that a pension system needs to achieve. Public systems are necessary to ensure broad coverage – even if only of a basic minimum. Private 12 contributory systems are needed to reduce the burdens on the (unsustainable) public system and to provide an adequate level of future benefits. 41. The timelines in pensions are so long that the system will face a wide range of risks. These risks can be specific to a particular pension pillar or shared. A key lesson is that if retirement income is drawn from a range of sources there is a better chance of dealing with these risks over the long-term. The state and private pensions help. But so does owning other assets like housing, having strong family links with mutual support and the ability to work for longer or gradually reduce work intensity but not have an all or nothing retirement date. Figure 2: Diversified systems have multiple income sources to offset multiple risks 42. In addition to the basic nature of the system, a critical issue is the retirement income objective. What is the target level of income and for whom? This issue needs to be at the heart of the design and then assessment of any system. The objective can be set in terms of total replacement rates, a target level of assets in the second pillar to match state provision or a level of retirement income (from an annuity). 43. The future replacement rates in different systems vary. For those countries that do have a clear objective a figure of 60% - 70% is often chosen (although minimum pensions often mean that rate for those on lower incomes is much higher and those on the highest incomes typically need additionally voluntary pension provision to reach the target). Figure 3 shows a range of expected replacement rates from across the world. It shows gross replacement rates at 0.5 (GRR low) and 1.5 (GRR high) multiples of average earnings. It includes both state and mandatory private pensions (first and second pillars). 13 44. If there are no reforms in Albania to introduce a diversified system then future pensioners will face outcomes at the very bottom of this range. The objective should not be the top of the range as many of those systems are unsustainable. In other words there is a clear adequacy/sustainability trade off. Figure 3: Gross Replacement Rates from first and second pillars at 0.5 and 1.5 times average earnings. Source: OECD 45. The level of contributions needed to achieve a given replacement rate will vary depending on costs and the rate of return. So a strong focus on the efficiency of the system is vital. (see below). The first pillar reform will have an important impact on how any contributions will be funded. In countries with informal labor markets taking contributions as a share of wages can be difficult. One option for the first pillar is to progressively transfer funding to non-wage sources. This would open up space for a contributory element in the second pillar. This would not face the same problems as the first pillar because individuals would have clearer incentives to contribute if they extra contributions led to higher pensions. However, finding other sources of finance is clearly a challenge and changes to the benefit structure of the first pillar will certainly be needed. C. Costs 46. Increasing the net of fee investment returns is a critical way to improve pension outcomes. One part of the story is to control costs. Cutting costs in a way that reduces investment returns is self-defeating. But a range of approaches in different systems across the world combine low costs with access to high quality investment. 47. To reduce costs the underlying cost dynamic as well as market dynamic must be addressed. The main cost drivers are i) customer acquisition (marketing, running a sales force); ii) administration (record keeping, account management and 14 communication); and iii) investment management. The importance of scale to deliver low cost is one of the clearest findings in the pension literature. 48. Ensuring the market genuinely has a robust demand and supply side is vital. The shortcomings of a number of systems come from failures in design – particularly when a model of many individual buyers from many sellers is used. This may work well for many products but does not perform well for mass market private pension products. 49. The table below sets out a comparison of some different options for the determination of investment strategy, choice of provider and nature of the provision of administration and investment management. The examples of best practice range from large occupational or employer funds in a number of countries (e.g., ABP in the Netherlands); institutions covering the whole market such as those in Kosovo (KPST), Sweden (PPM) and Denmark (ATP); specialized institutions covering segments of a pension market such as NEST in the UK and the Thrift Saving Plan (TSP) in the US; and Provident Funds such as those in Malaysia and Singapore. Some countries are in the second tier of 0.5% to 1% costs per year (e.g. Chile, Australia, some UK plans); many in the third tier of 1% - 2% (including Mexico, Peru, Hong Kong and smaller UK plans) and some countries have fees even above 3% a year (Turkey’s 3rd pillar in 2012). Figure 4: Different models for improving cost and investment outcomes in pension* DEMAND SIDE OF DISTRIBUTION and SUPPLY SIDE OF Costs MARKET MARKET Who decides strategy, How demand and Administration and investment chooses provider and supply are management negotiates costs? matched Individual Sales Agents Admin and Inv Management High (LAC/ECA) Individual Default Sales agents Admin and Inv Management Med – High member AFTER initial e.g. Mexico allocation rule allocation Individual Auction Sales agents Admin and Inv Management Med e.g. AFTER initial Chile allocation Member-focused Member Admin – Investment Low and governance organization automatically competitive Management - Med inc with scale and expertise – added via bulk tender competitive bulk ‘NEST’ UK bulk purchases for employer tender model and members Kosovo Member-focused Member Admin Investment Low – e.g. governance organization automatically competitive Management – Swedish with scale & expertise: bulk added via bulk tender in-house and large purchases for members employer or in-house provision & occ. Plans external choice Member-focused Member In-house In-house invest. Low – e.g. governance organization automatically Management inc 15 with scale & expertise: bulk added via admin outsourcing ATP model purchases for members employer *Fee caps can be used in any model but any positive impact will depend on good design Note table shows ways to improve costs and investment, not all possible ways to deliver pensions. 50. Setting a goal for future fees in 5, 10 or 20 years could help guide policies in this regard. This was one of the approaches adopted in Kosovo. But it is critical that the market structure is right so that the cost structure is achievable. A model that has extensive use of sales agents and marketing for the ‘customer acquisition’ is inherently expensive. In such systems high costs can also be accompanied by low profits for providers. 51. There is no one solution for costs and fees. The Swedish system provides an example of cost reduction through multiple policies. A centralized administration platform for all providers of investment management reduces costs through economies of scale. In addition, external providers have to accept mandated fee reductions on their normal retail prices to be allowed into the system. These lead to an average price reduction of 0.5%, from 0.8% to 0.3%. When the reforms were introduced providers said they would not be viable and providers would not enter the market. There are currently 100 fund management companies offering 800 funds. In addition there is a central default fund. In the Swedish system most members ‘select’ the centrally provided default fund, which at 0.12% is significantly cheaper than the externally provided average of 0.32%. 52. For Albania a critical issue will be whether there is a viable administrative platform or whether a new one needs to be built. The tax administration system does not appear to be the right platform – certainly at the moment. The Social Security Institute has had issues in the past but is going through a process of upgrading – including recently digitizing all of its records (World Bank 2013 d,e). This means it could be a viable option for the future. But this would need to be tested in the years that follow to ensure it could deliver scale and efficiency. In Kosovo a new system was built. D. Investments and Capital Markets 53. To serve the best interests of members of private pension funds their assets should be invested into the combination of international and domestic assets that will have the best risk-return trade off in delivering retirement income. The right level of investment in the home country will depend on whether there are good investment opportunities available. The proactive development of the capital market is needed to ensure the most effective matching of the holders of pools of savings and individuals and firms seeking capital for investment. 54. For Albania the governance, potential use of guarantees, and modeling of future investment strategies, must balance the evidence on maximizing long term returns with the need to build trust and confidence. Albania currently has relatively low financial inclusion and faces the continuing legacy of the 1990s pyramid scandal on trust and confidence. Guarantees – for example if they only apply at retirement and guarantee contributions but not a real rate of return – could be a useful feature. But 16 guarantees can also be expensive, and if badly designed they can impact the asset allocation of providers in a negative way. The critical issue in the coming years will be to robustly examine the different options and test with individuals the issues about which they are most concerned. For example the security of the assets might be the biggest fear – which is why supervision is such a critical issue (see below). 55. International evidence typically shows that equity markets offer a superior long term return compared to other investments. But the time period and country are important. Modeling will need to be done for the specific Albanian situation. But the table shows how returns have differed over the years in different markets. It is important to have a robust process to ensure that the investment strategy is built using the best available data for the country in question. And it is important to realize that returns will be variable – as seen in the Kosovo example (see box). Table 2: Global Real Investment Returns in 19 countries 1900 – 2012. Year 1900 - 2012 1963 - 2012 2000 - 2012 Assets Equity Bonds T Equity Bonds T Equity Bonds T Bills Bills Bills World 5 1.8 0.9 5.2 4.3 1.0 0.1 6.1 -0.3 World 4.4 1.4 0.9 5.4 4.9 1.0 0.5 5.8 -0.3 – US South 7.3 1.8 1.0 8.3 1.6 1.9 9.6 6.1 2.6 Africa Spain 3.4 1.3 0.3 3.8 1.4 0.6 0.4 2.2 -0.3 Source: Credit Suisse Global Investment Returns 2013 (Dimson and others 2013). 56. The issue of domestic versus international investments can be politically difficult. But forcing a new second pillar to invest all the assets in Albania would be counter-productive. Equally, having no investment in the country can be difficult. The right approach should be guided by a trustee or governing body that has the expertise and member-focus to determine the best strategy for Albania’s future pensioners. In the early years of the system many countries have been cautious and had tight international investment limits. Whilst this could make sense, particularly when there are not high levels of assets, it is essential to relax these constraints. 57. Without a proactive program of capital market development the development of private savings are not likely to benefit the domestic economy as they could (Price and Rudolph, 2013). Developing the domestic capital market also helps provide better options to solve the international v domestic issue (although in a small market 100% domestic investment will always be risky). 58. There are fortunately some good projects on which to build a more holistic capital market development strategy. The World Bank Group has been increasing private investment in infrastructure and fostering public-private partnerships. These include providing risk guarantees for investments in power distribution and energy, development of the regulatory framework for renewable energy and small hydropower 17 plants and investments in specific hydro-electric companies, and equity and loan investments in onshore oil exploration (World Bank 2013b). BOX 1: THE EXPERIENCE IN KOSOVO OF ESTABLISHING A DIVERSIFIED SYSTEM Kosovo faced many of the same problems as Albania when designing its new pension system – as well as some that were unique. The country was small and there was little local capital market. The labor market had a mixture of formal and informal sectors and there was limited capacity among government, industry and the people to tackle pension issues. The model adopted had a number of good features – as well as elements that may not be suitable for Albania. In the first pillar an old age basic pension, and disability pension, are funded from general revenue. In the second pillar employers and employees each contribute 5% of salary. Contributions go to the Kosovo Pension Saving Trust – an independent body with a legal duty to focus on the best interests of members. By January 2014 it had €912m in assets under management. The Trust has broad freedom in investment strategy – with investments made by major professional asset managers. Collections are centralized to minimize fees, which fell to 0.6% of assets under management in 2012. This puts Kosovo Pension Saving Trust close to market leading systems where fees are 0.5% or lower. The Trust is run by a mixture of local and international expert trustees to enhance the overall quality of governance. The key area of challenge has been on investment strategy. In Kosovo there is a relatively high allocation to equities – which makes sense in a long term perspective but clearly mean that the funds suffered losses during the global financial crisis. It may be that an Albanian equivalent starts more cautiously given the levels of distrust in financial services in Albania. The right approach will be determined by a combination of rigorous modeling and gaining a detailed understanding of the views of the target market – both suggested as part of the plan in this note. The Kosovo experience has been helpfully documented in Gubbels and others 2007 which sets out the detailed practical steps that are required to establish a new system – from overall design to the development of collection and IT systems. It also sets out the important financial and technical assistance received. This helps to inform the timeline for action set out in this Technical Note. 18 E. Coverage and Informality 59. Informal labor markets create challenges for contributory systems in both the public and private sectors. The issue is critical for options in Albania where informality is high and there is in particular a very large agriculture sector. There is a range of international experience on which to reflect that will help future design. But a key conclusion is the universal coverage of a pension will require payments from the state. Or in other words private pension pillars are vital to a diversified system but they do not achieve universal coverage. As the figure shows, the percentage of workers contributing to the first pillar in Albania is very low compared to others internationally and hence shows how important are the issues of coverage, informality and incentives. Figure 5: Percentage of Working Age Population, Ages 15-64, Contributing to the Pension System 100.0% 90.0% 80.0% 70.0% % of Working Age Population 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Source: World Bank 2013a 60. There are good lessons coming from recent work in Latin America. Brazil is considered to have had a lot of success compared to countries like Mexico where coverage has been less broad. A detailed examination of the Brazilian programs is included in the annex. These include a range of targeted programs directed to Small and Medium sized businesses or rural workers covering both pensions and general 19 formalization experiences. As the figure shows there have been some impressive increases in coverage achieved as a result. Figure 6: Gains in coverage of pensions in Brazil by type of contributor Proportion of contributors to the RGPS (private sector workers) by type of worker, in 2000 and 2010 100 +12.6% 90 +2.4% 78.6 80 69.8 72.5 74.3 70 60 Percent +12.1% 50 39.3 +20.4% 40 35.0 30 26.4 21.9 20 10 0 Employees Domestic workers Self-employed Employers workers 2000 2010 Source: Brazil’s Ministry of Social Security 61. Mandatory and voluntary private pensions are not really totally distinct systems so much as either ends of a spectrum of how to deliver private pensions (see Figure 7). So the terminology of second versus third pillar can sometimes be used by different people to describe different systems. In this report third pillar is used to describe the pure individual voluntary system based on individual choice that currently exists in Albania and most other countries in the region. When options are being discussed for adding a second pillar, or moving to a more diversified system, they are focusing on using either compulsory or mandatory provision or techniques like auto-enrolment. The UK case for example now includes the mandatory provision of a pension by an employer with workers automatically put into the pension – but with the option to opt out for the members. Some people may call this a voluntary system but throughout this report it would be classed as a second pillar. 62. The spectrum of options from third pillar voluntary to second pillar mandatory and quasi mandatory is shown in Figure 7. The options in this report are about moving towards the end of the spectrum that will increase coverage. Future coverage is likely to be lower than in countries with more formal markets. But adding some form of second pillar/quasi mandatory element to the system will increase coverage very significantly above the level obtained in a purely voluntary system. Figure 7: A Spectrum of Coverage from Mandatory to Voluntary Mandatory Quasi-mandatory auto-enrollment Employer/union based Personal voluntary choice Higher coverage Lower Coverage Low levels of informality High levels of informality 20 63. There are also lessons from more developed countries about the importance of incentives and the role of behavioral economics. The UK has one of the world’s largest voluntary private pension sectors by assets. Despite many advantages the level of coverage – particularly among private sector employers – had been falling for many years. In particular there was very low coverage among small and medium employers and for low and middle income workers. After a long debate the UK introduced mandatory requirements for employers to offer a pension and auto-enrol workers in the pension. In addition the UK created ‘NEST’ as a not-for-profit independent pension fund to ensure all workers would have access to a pension at low cost (0.5% a year initially falling to 0.3% in 20 years), as the for-profit providers did not cover the full work force and often had higher charges for the sectors they did cover. The UK also shows how long it can take from starting a pension reform process to workers and employers actually beginning to make contributions. The Turner Commission was formed in 2002 but the new system it recommended in 2005 did not start until October 2012. However, within a year one million more workers were in a pension – starting a process of significant expansion in coverage. 64. To ensure that the international experiences can be best tailored to Albania it is recommended that pilots be introduced to test different approaches. As part of the process of developing evidence and options pilots could be run – for example with employers – to look at what works to encourage workers to contribute to pensions, using the voluntary pension market as a real-life test bed. Pilots would help to test the communications, collections, target population and governance that could work most effectively if private pension provision were expanded. Surveys and focus groups would also help to expand the evidence base to ensure better policies to expand provision. F. Security and Supervision 65. Ensuring that contributions to pension funds are safe is one of the most important parts of a pension system. The role of the supervisor and the legislative framework are crucial. The use, for example, of custodians or depositaries as in the current third pillar can be an effective mechanism – and also reduces the need for supervisory intervention when resources are scarce. 66. The existing pension legislation for the third pillar was passed in 2009. It has already incorporated many good features. Following an extensive review new amendments are being discussed to ensure the legislation meets all the EU directives. The extensive review covered the IORP, MIFID and UCITS directives. The FSA have prepared changes to the legislation as a result and these are being reviewed for comments by the industry. These changes should be passed. 67. But this review and the previous one have both identified a lack of independence and resourcing for the FSA as important weaknesses. These need to be fixed to allow a fully effective supervisor now. But no second pillar should be introduced without a fully resourced supervisor capable of robust risk based supervision. 21 68. Many of the legal and environmental pre-conditions for a second pillar are developing – particularly due to the EU accession process. This is a helpful mechanism to ensure the continued developments that will underpin a successful second pillar. EU membership is not needed before a second pillar is introduced. But the process towards membership should achieve many of the developments that need to be made. 69. The suggestions for a Pension Commission are designed to protect against future pension ‘reversals’. An important part of the story of pension reforms that have been reversed is that there was not broad political agreement (Price and Rudolph 2013). So that is why so much attention has been placed on this area in this report. 70. Finally, the tax incentives in the third pillar are not operating effectively because of a conflict between the pension law and tax operations. This issue should be resolved by making the approach set out in the pension legislation work as intended. One possibility is to allow pension contributions to be paid from gross salary monthly and then taxing the remainder, rather than needing to claim back a refund of tax in the April following the year the contributions are made. Another issue is that the tax law puts a lower limit on total contributions than the pension law. This adds to confusion and complexity and means that the incentives will not work as well as they could. Since the assets in the system are very small this does not affect the fiscal position. III. RECOMMENDATIONS AND A POTENTIAL TIMELINE 71. This note sets out options to develop private pensions in Albania. It complements analysis by the World Bank of reform options for state pensions (World Bank 2013a). In combination they set out options for the crucial decisions on pensions required in the short, medium and long term to improve outcomes for Albania. The diversified approach and options suggested target Albania’s particular characteristics (like high levels of labor market informality). They support recommendations in the 2012 EU White Paper on Aging to the existing EU members (European Commission 2012a). So there is a good fit with the future for which Albania is preparing. 72. The current system in Albania is dominated by an unsustainable state pension system (first pillar). Sustainability must be improved. Poor incentives and informal labor markets mean contributions are vastly below required levels. Some positive reforms such as increases in the retirement age have been offset by reduced contributions and increased generosity. Fiscal sustainability longer-term will be helped by automatically linking future increases in the retirement age to rising longevity– building in evidence-led policy adaptability. 73. First pillar reform is urgent (and is currently being considered by the government). It will help if these decisions needed for fiscal sustainability are set out alongside a commitment to a diversified pension system so people can see a viable path to good pensions in the future. Identifying non-wage labor sources of funding will improve the prospects for employment and boosting formality. 22 74. Political sustainability and protection against future reversals can be enhanced through building government, political and public understanding and agreement. Progress should be started immediately by establishing a Pension Commission. This can build on past efforts so that the strategic and technical issues can be substantially advanced before the next election in 2017 with final decisions soon after. Without focused attention in the next few years the introduction of a second pillar to diversify the system by 2020 would not be attainable – pushing off even further the achievement of a sustainable system. 75. The Government should seize the opportunity to reap the benefits of the ‘science of delivery’. It should develop a deep understanding of data and attitudes behind the dynamics in the formal and informal labor market. It should pilot solutions so the ultimate design works in the Albania context. The proposed Pension Commission should also set clear objectives for the new system through an outcomes based assessment. One example is aiming for best in class investment at a long term cost of 0.6% a year or lower. This has been achieved in countries from Kosovo to Sweden. They, and other countries, provide lessons on how to create scale, expertise, member-focused governance and cut distribution costs through an intelligent market structure and a robust administration platform. 76. The governance, potential use of guarantees, and modeling of future investment strategies, must balance the evidence on maximizing efficiency and long term returns with the need to build trust and confidence. Albania currently has relatively low financial inclusion and faces the continuing legacy of the 1990s pyramid scandal on trust and confidence. 77. Creating the necessary foundations for a robust and diversified pension system will enhance and build upon the efforts being made as part of the EU accession process. A number of the recommendations pick up on themes identified in the latest EU progress report. They also fit well with the EU’s own agenda on pension reform as set out in their 2012 White Paper on Aging. The recommendations also address issues identified in the 2005 FSAP and in a range of expert reviews provided to the government. 78. A central issue to meet international standards for security is to ensure the independence and resourcing of the FSA and the development of robust risk based capacity. Planned amendments to the 2009 Pension Act to make it fully compliant with EU and international standards should be passed and the barriers to providing tax incentives should be removed. Fixing the independence and resourcing of FSA is vital to allow effective supervision now and preparation for a second pillar. Other key issues include: developing the payout phase with intelligent phased withdrawal, inflation-linked bonds in the longer term and developing joint-life annuities for the future to improve gender equality; and the progressive development of domestic investment opportunities, building on innovative IFC and MIGA projects in a range of sectors. 79. The timelines for pension reforms are inherently long-term. This can obscure the need and possibility for action now. The recommendations have been set into a potential timeline to help breakdown the challenge into practical steps. The reform process could follow a ‘launch-design-build-operate’ program starting in 2014 and 23 leading to the start of the new system in 2020. This is just one option – but one in which the World Bank and other partners stand ready to support the leadership of the Albanian authorities. 24 Figure 8: A Potential Timeline to Build a Diversified Pension System Phase 1: Launch Phase 2: Design Phase 3: Build Phase 4: Operate 2013-2014 2015-2016 2017-2019 2020 onward •Prepare options for 1st pillar •Test delivery models against •Pass new 2nd pillar legislation •Start 2nd pillar (with phasing reform potential outcomes with future contribution if needed) •Take urgent decisions on 1st •Feasibility study of SII as rates, coverage, market •Contributions start pillar and fiscal potential admin platform structure, investments and •Investments start •Prepare FSAP note on 2nd •Collect baseline data and supervision. •On-going supervision and pillar reform attidudes •Determine phasing-in of enforcement •Establish Pension •Pilot incentives with contributions linked to ability •Capital Market Development Commission and agree 3 employers and employees to reduce contributions to 1st continues year work program pillar •Model potential investment •On-going monitoring of •Begin design of data strategies •Competitive tender/licensing domestic v international collection and pilots for administration and •Deliver capital market asset allocation investment management. •Public statement of future development plan •Development of joint annuity vision of diversified system •Build and test operations (in •Decide delivery model and market public & private sector) •Ensure FSA independence coverage after testing •Take part in EU Aging and resources cf FSAP •Continued data collection •Develop FSA Risk Based debates recommendations and attitude testing Supervision capacity for •Resolve tax relief block in Pensions •Renew capital market 3nd pillar and amend pension development program if •Draft legislation for act for EU compliance successful introduction after 2017 •Design holistic capital market election. •Communciations with development plan affected employers and •Public statement on shared employees vision for diversified pension system 25 REFERENCES Aschcroft, John (2013) “Report of the Expert Mission in fully aligning the voluntary private pension act (third pillar) and associated by-laws with the EU directives and relevant international pension standards”, Funded by World Bank IDA Credit 3513 ALB. European Commission (2012a) “An Agenda for Adequate, Safe, and Sustainable Pensions.” COM (2012) 55 final, European Commission, Brussels, February 16 http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2012:0055:FIN:EN:PDF European Commission (2012b) Albania 2012 Progress Report, EC, SWD (2012) 334 Final, European Commission, Brussels European Commission (2012c) Albania 2012 Progress Report Statistical Annex, EC, SWD (2012) 334 Final, European Commission, Brussels Gubbels, J. Snelbeker, D. and Zezulin, L (2007) “The Kosovo Pension Reform: Achievements and Lessons, Social Protection Discussion Paper, No 0707, World Bank, Washington DC. Kosovo Pension Saving Trust (2103) ‘Kosovo Pension Saving Trust Annual Report 2012’, Pristina, 2013 IMF (2005) Albania: Financial System Stability Assessment, including Reports on the Observance of Standards and Codes on the following topics: Banking Supervision and Payment Systems IMF Country Report No. 05/274 August 2005, IMF, Washington DC IMF (2013) Republic of Kosovo: Financial System Stability Assessment IMF Country Report No. 13/99, April 2013, IMF, Washington DC INSTAT (2012) Albania Population and Housing Census 2011: Main Results Part 1 http://www.instat.gov.al/media/178070/rezultatet_kryesore_t__censusit_t__popullsis__d he_banesave_2011_n__shqip_ri.pdf OECD (2012) Pensions Outlook, OECD, Paris Pension Fund Monitoring Center (2013) Individual Pension System Progress Report 2012, Pension Fund Monitoring Center, Turkey www.egm.org.tr/bes2012gr.htm Price, W and Rudolph, H (2013) “Reversal and Reduction, Resolution and Reform: Lessons from the Financial Crisis in Europe and Central Asia to improve outcomes from mandatory private pensions. World Bank (2013a) “Albania: Pension Policy Note: Key Challenges and Directions for Reform, Human Development Sector Unit, South East Europe Country Unit, Europe and Central Asia Region, World Bank, Washington, DC 26 World Bank (2013b) ‘Albania Overview’, World Bank, Washington DC www.worldbank.org/en/country/albania/overview.print World Bank (2012c) International Patterns of Private Pension Provision, World Bank, Washington DC World Bank (2013d) World Bank Group Partnership Program Snapshot, March 2013, World Bank, Washington DC World Bank (2013e) ‘Years of Crucial Information Now Seconds Away for Albania’s Pensioners’, World Bank, Washington DC 27 Annex: Evidence from Brazil Special programs for formalization and pension coverage for SMEs, domestic workers, and rural populations Brazil has implemented several programs since 1996 to increase pension coverage and contributions, in particular for SMEs, self-employed, domestic, and rural workers. They were complemented by simplified tax payment and collection. The Simples program (1996) targeted at micro and small enterprises, introduced changes to tax and social security contributions. Simples unified several taxation and social security contribution charges and made it dependent on total revenues, introducing generally more favorable rates to the eligible businesses. The rates were progressive and thus more favorable for micro and small enterprises (Delgado et al. 2007; Rangel et al. 2009; Ansiliero 2013). The program was subsequently amended and expanded to include micro and small enterprises with annual revenues of up to R$0.24 million and R$2.4 million, respectively6. The number of businesses opting for Simples increased from 0.9 million in 1999 to almost 1.6 million in 2005. In the same period, the number of employees covered increased from 3.6 million to 5.8 million (Rangel et al. 2009), representing 6.7% of total employment (up from 4.8% in 1999). The Supersimples/Simples Nacional program (2007), also targeted at micro and small enterprises, replaced the Simples program and attempted to further its goals. Legislated in 2006 and effective in 2007, it continued the process started by Simples by further unifying the charges. In addition, it generally reduced the tax burden for the participating businesses (Rangel et al. 2009) in an attempt to increase their rate of formalization. In 2011, the program was expanded to include small companies with annual revenues of up to R$3.6 million/year7. The Microempreendedor Individual (MEI) program (2009), targeted at individual micro- entrepreneurs (revenues