SP DISCUSSION PAPER NO.0113 Regulating Private Pension Funds' Structure, Performance and Investments: Cross-country Evidence P.S. Srinivas, Edward Whitehouse and Juan Yermo July 2000 ionia LABOR MARKETS, PENSIONS, SOCIAL ASSISTANCE T H E W O R L D B A N K Regulating private pension funds' structure, performance and investments: cross-country evidence P.S. Srinivas, Edward Whitehouse and Juan Yermo* July 2000 0 0 Srinivas is a Financial Economist with the Finance, Private Sector and Infrastructure Sector Management Unit of the Latin America and Caribbean Regional Office of the World Bank. Whitehouse is Director of Axia Economics, London. Yermo, formerly with the World Bank, is now in the Directorate for Financial, Fiscal and Enterprise affairs at the OECD. We are grateful to participants at two World Bank seminars for their comments. The views and opinions in the paper are the authors' own, and do not reflect those of any of the World Bank or any of its members. Comments and suggestions: Srinivas: +1 202 473 8939, e-mail psrinivas@worldbank.org; Whitehouse: +44 171 274 3025, e-mail edward.whitehouse@axiaecon.com; Yermo: e-mail: juan.yermo@oecd.org. This paper is part of the World Bank's pension reform primer: a comprehensive, up-to-date resource for people designing and implementing pension reforms around the world. The series is edited by Robert Palacios of the World Bank and Edward Whitehouse. For more information, please contact Social Protection, Human Development Network, World Bank, 1818 H Street NW, Washington, D.C. 20433; telephone +1 202 458 5267; fax +1 202 614 0471; e-mail socialprotection@worldbank.org. All Pension Reform Primer material is available on the internet at www.worldbank.org/pensions. Abstract Because defined-contribution systems expose pensions to a number of risks, reforming governments have often strictly regulated the pension fund industry's structure, performance, and investments. This paper compares the rules in the new systems of Latin America and eastern Europe with richer OECD countries. The authors argue that the benefits of competing pension funds and individual choice can only be achieved if regulations are loosened in the medium term. 2 Table of Contents 1. Pension funds' supermision and regulation ................................................. 7 2. Risks in pensionfunds ................................................. . 8 2.1 Systematic market risk ................................................. 9 2.2 Systemic risks ................................................. 10 2.3 Agency nsks ................................................. 11 3. Regulating industy structure .................................................. 11 3.1 Rationale ................................................. 13 3.2 Adverse effects of structural regulations ................................................. 14 3.3 Issues in m e mber choice of investments ................................................. 15 3.4 Empirical evidence of concentration in fund management ................................................. 17 4. Regulatingnperformance .................................................. 19 4.1 Rationale ................................................. 21 4.2 Adverse effects of performance regulation ................................................. 21 4.3 Empirical evidence ................................................. 22 4.4 Performance regulation and herding ................................................. 23 5. Regulating asset allocation ................................................... 25 5.1 Rationale of asset allocation regulations .................................................. 29 5.2 Adverse effects of asset allocation regulations ............................................ 30 5.3 Empirical evidence of portfolio limits and asset allocations . ................................ 32 5.3.1. Latin American countries ................................................. 33 5.3.2. OECD countries ................................................. 35 5.4 Empirical evidence of pension fund returns ................................................. 36 5.4.1 Latin American countries ................................................. 37 5.4.2 OECD countries: cross-national comparisons ................................................. 38 5.4.3 United Kingdom and United States ................................................. 39 6. Concluseions andpoliy implications .40 7. Bibliograph...47 3 List of Tables and Figures Table 1. A taxonomy of investment risks in pension funds ................................................................... 10 Figure 1. Asset allocation in member-directed 401 (k) pension plans ................................................... 17 Figure 2. Concentration curves for fund managers in Latin America and the United Kingdom ... 19 Table 4. Pension fund performance regulations and government guarantees in Latin America ..... 20 Table 5. The herding effect in Chile (per cent of assets of pension funds) ......................................... 22 Table 6. Correlation of pension fund returns ....................................................................... 23 Table 7. Peru: Average pension fund portfolio and standard deviation, 1995-8 ................................ 24 Figure 3. Limits on foreign investments in OECD countries ................................................................ 26 Figure 4. Limits on domestic investments in OECD countries ............................................................ 27 Figure 5. Pension fund portfol io limits, 1998 ....................................................................... 28 Figure 6. Evolution of portfol io liits in Chile, 1982-1998 ................................................................... 29 Figure 7. Pension funds' equity holdings as a percentage of total stockmarket capitalisation, 1997 ....................................................................... 32 Figure 8. Equity investments as a percentage of total pension-fund portfolios .................................. 33 Table 13. Pension-fund portfolios and limits in Argentina, Chile and Peru ........................................ 34 Table 14. Asset allocation of funds in Chile, 1981-97 ....................................................................... 35 Table 15. Portfolios relative to regulations in eight OECD countries ................................................. 36 Table 16. Returns on pension funds and balanced portfolios: Latin America .................................... 38 Figure 9. Returns on pension funds and balanced portfolios: OECD countries ................................ 39 Table A.1. Asset allocation in member-directed 401(k) pension plans ................................................ 42 Table A.2. Concentration of fund managers in Latin America and the United Kingdom ............... 42 Table A.3. Pension asset regulations in OECD countries ...................................................................... 43 Table A.4. Pension fund portfolio limits, 1998 ....................................................................... 44 Table A.5. Evolution of portfolio limits in Chile, 1981-1998 ................................................................ 45 Table A.6. Pension fund portfolios, selected countries ....................................................................... 45 Table A.7. Returns on pension funds and balanced portfolios: OECD countries ............................. 46 4 Regulating private pension funds' structure, performance and investments: cross-country evidence P.S. Srinivas, Edward Whitehouse and Juan Yermo 'Risk is risk. It cannot be legislated away. It can only be diversified away.' George Russell, financier, quoted in de Ryck (1998) A number of countries have implemented or proposed fundamental reforms of their pension systems, including eight in Latin America and five in Europe'. These reforms emphasise the role of individual, privately managed defined-contribution accounts, where the value of the pension benefit will depend on accumulated contributions and investment retums. They are, by definition, fully funded. The new pension plans substitute for the old, public, defined benefit schemes where the pension depended on some measure of earnings and years of coverage. Public schemes are usually financed on a pay-as-you-go basis, where current workers' contributions pay for current pensioners' benefits. The new defined contribution systems expose workers' future pension benefits to a number of different risks. To try to mitigate these risks, reforming goveruments have often strictly regulated the pension fund management industry's structure, performance, and asset allocation. Often, a new fund management industry has been established, consisting of multiple competing pension funds, separated from other financial institutions. In the majority there are restrictions on the type of investments that can be made and sometimes regulations specify the retums that the funds should earn. These fundamental reforms of pension systems aim to: * enhance individual choice and responsibility through the freedom to select a fund manager; Chile (1981), Peru (1993), Argentina (1994), Colombia (1994), Uruguay (1995), Bolivia (1997), Mexico (1998), El Salvador (1998), Czech Repubhc (1998), Hungary (1998), Poland (1999), Sweden (1999) and the United Kingdom (1988). Schwarz and Demirguy-Kunt (1999) provide a global survey of pension reforms of the last six years. 5 * ensure good service and performance through competition between fund managers and so deliver reasonable pension benefits; and * limit risk through competition and investment restrictions. However, in practice, 'Draconian' regulation of pension funds has prevented the achievement of many of these objectives. Regulations have generally focussed on three aspects: industry structure, asset allocation, and performance. Structural regulations force workers to choose only one manager and one fund. So, workers are unable to diversify investments across funds, exposing them to aberrant behaviour by fund managers, and preventing portfolio adjustments according to the individual's age, household characteristics, career profile and attitude to risk. Strict asset-allocation rules and relative performance criteria mean that pension funds often invest and perform almost identically, removing any substantive choice for workers over the allocation of their pension fund's assets and the portfolio's risk and returns. This paper provides evidence for some of the effects of structural, investment and performance regulation of pension funds in emerging economies and compares them with evidence from more developed OECD countries. Concentration in the pension fund management industry is found to be higher in the new pension systems of Latin America and Eastern Europe than in most OECD countries. Concentration might be because the new pension markets are smaller than in countries with more established funded pension systems, but it could also be because of restrictions on industry structure. In Latin America, asset allocation and performance is nearly identical across pension funds. So-called 'herding' behaviour is almost a defining characteristics of these pension regimes. Again, this reflects, at least in part, asset allocation restrictions and strict performance regulation. There is also evidence that pension funds have often under-performed simple portfolios composed of market indices of stocks and bonds. All the rules imposed in the new systems of Latin American and Eastern Europe2 seem to be more stringent than in the OECD, with one exception: portfolio limits. Some OECD countries have a tighter investment regime than countries such as Argentina, Chile, Colombia, Peru and Poland. But OECD countries tend to have fewer barriers to entry and impose fewer constraints on performance than Latin American and Eastern European countries. 2 The countries in Europe considered are Poland, Hungary and the Czech Republic. These countries are actually part of the OECD, but for the purposes of the study they are discussed together with Latin American countries, because thev have established very similar private pension industries. 6 The rest of the paper is structured as follows. Section 1 reviews investment supervision and regulation in practice. The subsequent section looks at risks in pension funds. Sections 3, 4, and 5 review the adverse effects of structural, performance and portfolio restrictions respectively. Section 6 concludes. 1. Pension funds, supervision and regulation Pension funds have shown an impressive growth pattern. In Chile, which reformed its system in 1981, pension funds are the leading institutional investors, managing a total of $32 billion at the end of 1997, worth some 44 per cent of GDP. Only five countries have proportionally larger pension fund sectors - Ireland, the Netherlands, Switzerland, the United IKingdom and the United States - where funds average 75 per cent of GDP. In these five countries, the value of funds has been growing rapidly: by 56 per cent between 1987 and 1996.3 Intersec, a financial data firm, expects world pension fund assets, currently $11,000 billion, to grow by 40 per cent over the next five years. In other Latin American countries and in Eastern Europe, reforms were more recent, and so funds are much smaller. The next largest system after Chile is Argentina, where assets are worth 3 per cent of GDP. But funds in other countries are forecast to grow rapidly. Goldman Sachs, an American investment bank, expects the value of Argentine funds to increase from $8.8bn in 1997 to $33bn in 2003, or 6.4 per cent of GDP.4 Mexico has the largest number of workers covered by the new plans - about 14 million. Around 6 million workers each in Chile and Argentina, 21/2 million in Colombia, just over 1 mnillion m Peru and fewer than half a million in Bolivia and Uruguay are covered.5 In the United Kingdom, 5.7 million workers (28 per cent of total employees) are covered by the new personal pensions. A further 10 million are covered by longer-established employer-provided plans (of which more than 90 per cent are defined benefit). In Hungary, 800,000 workers have so far announced their intention to switch to the new funds. Poland and Croatia will implement their reforms during 1999. In Hungary, Poland and most of the Latin American countries, a new agency was established to supervise the new pension funds. The exceptions are in Colombia and Uruguay, where this 3 OECD (1998), table VA. 4 Mariscal (1998a) 5 However, only 54 per cent on average of these members actually contribute to the schemes, ranging from 44 per cent in Peru to 65 per cent in -Mexico. See Queisser (1998b). 7 responsibility falls on the Central Bank.6 These agencies ensure compliance with regulations on capital, disclosure and reporting, comnmissions, transfers between funds, rates of return and investment allocation. In other countries, such as Australia, Switzerland and the United Kingdom, existing financial regulators expanded to cover pension funds. 2. Risks in pension funds Government intervention in markets can be justified by market failures. In financial systems, externalities, asymmetric information and monopoly are the three main types of market failure. Pension funds pose a different set of risks than other financial institutions, such as banks. Pensions are long-term contracts and they involve a sizeable proportion of the individual's wealth. However, the existence of assets in pension funds avoids the danger of the t,pe of runs that can occur in banking crises (i.e., externalities). Monopoly, too, is likely to be less of a problem in the pension fund industry, as barriers to entry are low compared with banking.7 Asymmetric information -the fact that it is cosdly for the buyer of financial services to obtain sufficient information to assess the quality of that service - is likely to be the most senrous problem for pension funds. Lack of information means that the buyer is vulnerable to fraud, negligence, incompetence and unfair treatment by the provider. Clearly, the desire for providers to maintain a good reputation offers a high degree of protection, but there remain three risks in pension funds: e Systematic (undiversifiable) market risk: current generations cannot trade with unborn, ones, so efficient intergenerational risk sharing cannot take place * Systemic risk: Asymmetric information problems in banking systems can lead to bank runs, and make financial systems fragile * Agency risks: in financial markets, trading often takes place between parties with different information, creating problems of moral hazard and adverse selection8 6 The issue of supervision is covered in Demarco, Rofman and Whitehouse (1 998). 7 See, however, section 5.1 below for evidence of high concentration in pension fund management in Latin America. Also, Altman (1992) shows the monopoly problem that arises with employer-provided plans. 8 There is some overlap between the first and the other two forms of market failure risks. Whenever there are systemic and agency risks, systematic market risk for the investor is created. 8 Table 1 gives a taxonomy of these investment risks as they affect funded, defined contribution pensions plans. It also shows the mechanisms to reduce risks that might be used and the new risks that might be created. We describe these risks in turn. 2.1 Systematic market risk Once market-based ways of reducing systematic risks (such as diversification and risk pooling) are exhausted, investors are left with some rate-of-return uncertainty. This systematic market risk can only be reduced further through intergenerational risk sharing, pooling returns of investors across time. Example policies include issuing indexed bonds or offering government guarantees.9 Some observers (such as Heller, 1998) have argued that the mandatory nature of the new pension systems means that governments retain a responsibility for ensuring adequate pensions beyond the guarantees specified by legislation, producing 'contingent' or 'conjectural' public-sector liabilities. 9 Defined-benefit pensions might also reduce this kind of risk, but exposes the worker to other forms of uncertainty over, for example, job tenure and earnings profiles (see Disney and Whitehouse, 1994 and 1996 and Bodie, Marcus and Mierton, 1988). 9 Table 1. A taxonomy of investment risks in pension funds Type of risk Example Risk reduction Example New risk created Non-systematic Management Portfolio Diversification market risk inefficiency or diversification across countries or inexperience (fund intermediaries or industry specific) Systematic market Global asset price Government Inflation-indexed Policy risk risk volatility provides risk- bonds reducing instruments Government Minimum pension Agency risk'moral guarantee or real rate of hazard return guarantee Policy risk Systemic risk Banking crisis Prudential Capital adequacy Policy risk regulation Government Deposit insurance Agency risk/moral guarantee hazard Policy risk Agency risk Moral hazard: Prudential Diversification, Policy risk fraud, pension mis- regulation limits on self- selling, excessively investment risky investments Government Compensation Agency risklnmoral Adverse selection: guarantee hazard reasonably priced insurance not universally available Governtnent intervention in the form of guarantees may not necessarily be a panacea for risk. Guarantees create a moral-hazard problem: for example, a pension guarantee creates an incentive for informal sector workers to contribute to the system for the minimum number of years to qualify for the minimum pension. Investment managers may take excessive risks knowing that the member's pension is underwritten by the government. In general, guarantees reduce one t pe of risk but may increase others. 2.2 Systemic risks Investment in capital markets depends crucially on the option to exit into the safe-haven of liquid money markets. If banks take excessive risks, impairing their solvency, the solidity of the whole financial system is put at risk by the potential for a run on the banks. Hence, a sound banking system and a secure pension system go hand in hand. The regulatory framework should ensure that the moral hazard from deposit guarantees is mitigated. Latin American countries are still trying to make accounting and supervisory standards 10 stringent enough to evaluate risks more effectively than in the past (Rojas-Suarez and Weisbrod, 1996). 2.3 Agency risks Intervention to limit agency risks takes the form of prudential rules and guarantees applied to financial markets and intermediaries generally, not just to pension funds. This framework includes aims to * avoid fraud through setting accounting and auditing standards, information disclosure and insider trading rules * reduce overexposure to specific risks by requiring minimum levels of diversification by issuer and security * mnitigate conflicts of interest through limits on self-investment * limit market power by restricting concentration of share ownership Government might also choose to go further and guarantee individuals against these risks. The contrast between the regulatory regire for pension funds and other financial intermediaries in many developing countries is startling. While pension funds are subject to strict prudential controls, such as capital, disclosure, fiduciary and diversification standards requirements, the regulatory and supervisory framework of other financial institutions is often weak. Valuation is also a widespread problem. The strengthening of prudential controls is a basic precondition for the successful development of financial markets and expanding the investment universe of pension funds. 3. Regulating industry structure In Latin America and Eastern Europe, reforming countries restricted the industry structure in three ways * investment was limnited to one instrument, the specially created private pension accounts * administration of funds was restricted to companies exclusively dedicated to pension fund management and managers were restricted to one fund each 11 * ownership of pension fund managers was not open to existing financial institutions in some countries (Bolivia, Chile, Mexico, Peru) The structure of the industry in many reforming countries is limited to specially created pension fund managers, which must be independent of other financial institutions. Colombia is one exception: severance funds were allowed to manage pensions as long as this activity was kept separate from other businesses. But in other countries, too, there are strong economic ties between pension fund managers and other companies. For example, Maxima, the largest fund in Argentina, has Banco Quilmes, the Argentine subsidiaries of Deutsche Bank and HSBC (two of the world' largest banks) and New York Life as shareholders. Pension fund managers are usually restricted to pensions-related activities, such as collecting contributions, asset management, reporting results, and benefit payments. Associated activities - such as custody of assets, provision of life and disability insurance, etc. - are often carried out by separate institutions for economic or prudential financial reasons. In Eastem Europe and Latin America, each manager may usually administer only one fund. In Poland, the regulations allow managers to offer two funds from 2005: one with a relatively liberal investment regime, the other restricted to fixed-income securities. In Mexico too, the regulations contemplate allowing morc than one fund some time in the future. In most OECD countries, in contrast, pension plans are offered by a variety of different providers. In some, employers play an important role. In Ireland, the United Kingdom and the United States, employer schemes are a mnix of defined benefit and defined contribution. Larger schemes tend to be managed 'in-house', while smaller plans contract out fund management ; specialist financial institutions. The investment of defined-benefit schemes is, of course, of less concem to members than defined-contribution. Other countries with predominantly defined- benefit coverage include Belgium, Finland, France and Germany. In the United States, around half of employer-provided pension coverage is now defined- contribution. So-called 401(k) schemes (named after the relevant clause of the income tax legislation) cover 37 million workers. They now account for 39 per cent of the total of pension fund members, 29 per cent of assets and 53 per cent of new contributions.'( Typically, the employer selects the range of investment options in 401(k)s, but they are generally broad, including equity, bond and money-market funds. 10 VanDerhei (1999). 12 In Denmark and the Netherlands, the pension system is based on industry-wide schemes. There are 35 funds in Denmark, and the number of single-employer schemes has now declined to around 100. There are 65 compulsory industry-wide funds in the Netherlands, of which 95 per cent are defined-benefit. In contrast, pensions in Denmark are defined-contribution. Dutch companies are free to opt out of these plans if they offer their own scheme with equivalent benefits. There are around 1,000 of these single-employer plans. Australia's new superannuation system is based around compulsory employer-provided defined-contribution schemes.'1 Initially, the employer decided where contributions were invested. However, the government is proposing that employers be required to offer a minimum of five different funds. Already, 15 of the 24 largest funds offer a menu of investment strategies. The market for individual pension accounts in OECD countries usually involves a wide range of financial intermediaries. In the United Kingdom, for example, there are around 90 providers of personal pensions, including most life insurers and banks. They offer an average of around 8 funds each, and individuals are free to divide their assets between different funds.12 3.1 Rationale The restrictions in the Latin American and Eastern European regimes are designed to keep the regulation and supervision of the industry simple, avoiding the complexity of multiple instruments and funds. The poor performance of some existing financial intermediaries led to the decision to establish a new industry. But this poor performance could only result either from poor market or economic performance, or from inadequate regulation. In the first case, there is no a pnron reason to expect the new pension funds to perform any better. The second case justifies improvements in the existing regulatory framework, not necessarily the creation of another. Moreover, if the previous regulatory failure resulted from some systematic, cultural failure of governance, there is no reason to expect the new regime to be any better. In addition to being simpler to regulate, restrictions on the structure of the pension market makes the system easier for participants to understand. This is probably an advantage initially, as the new regime offers people new choices. However, as people become accustomed to the new system, this simplicity is less important. 1 See Flanagan (1999) and Edey and Simon (1996). 12 Dilnot et al (1994). 13 Limiting managers to one fund avoids the moral-hazard risk generated by minimum pension guarantees. If a manager were able to 'stream' low-income workers into one fund, they could then take 'wild bets' in high risk/high return assets knowing that the government insures the worker. Excluding existing financial intermediaries, such as mutual funds and banking conglomerates, from the new pensions industry is common in countries with weak banking systems or poor past mutual-fund performance. The aim of the restrictions was to protect retirement savings from deficiencies in existing financial institutions, often in the form of agency risks that were not checked by the existing regulatory and supervisory system. In some countries, these restrictions were also designed to reduce the market power of these intermediaries. The mandatory nature of pension contributions in many countries increases the government's responsibility for the safety of pension assets. Some OECD countries which also have mandatory private pension pillars also impose a single instrument requirement. In France, Switzerland, Finland and Australia employets are obliged to set-up pension plans for their employees. There is, however, more flexibility, because asset management may be carried by a variety of financial institutions. In the reforming countries, only licensed pension fund administrators are allowed to manage the funds. 3.2 Adverse effects of structural regulations The most important adverse effect of structural regulation is that it prevents diversification. Workers are unable to spread retirement savings across different financial intermediaries and different financial products. Hence, non-systematic market risk (the risk of aberrant behaviour by a specific fund manager or investment instrument) is not pooled away. Such risk could be easily diversified away if workers were able to invest in various funds at the same time, though this may raise administrative costs significantly. Also, to the extent that governments impose relative performance rules, and guarantee such performance, these constraints may not be worrying. Some countries, however, do not have performance rules, and in some cases require investors to remain with a specific fund for up to six months before they can transfer to a new one. In these cases, governments will probably be forced to bear the responsibility for funds which consistently underperform the industry. Another adverse effect arises because excluding existing financial intermedianres precludes the use of existing infrastructure and the potential benefits of economies of scale, raising 14 administrative costs. Instead, investors have to finance the set-up costs of the new industry through fees and commissions (Shah, 1997). The restriction of one fund per administrator also has significant costs. Workers cannot choose the optimal portfolio that best suits their age, career earnings path, and risk aversion. For example, younger workers have few assets other than their human capital (i.e. their future earnings). It is optimal for them to hold assets with a low correlation with their projected wages.13 It may also be better for younger workers to weight their portfolio towards equities, which have a higher tong- run return but also a higher short-term risk, whereas older workers prefer a less nrsky, bond- weighted portfolio.14 Furthermore, workers of a given age will also vary in a range of characteristics, such as occupation and industry and family type, which affect their attitudes to risk. They will also differ in the types of other assets that they hold: housing, durable goods and liquid assets, such as equities, bonds or deposits. The 'one-size-fits-all' portfolio that results from these restrictions means workers are unable to reap the benefits of diversification. Finally, the structural constraints can behave as barriers to entry in the pension fund industry, limiting competition, and raising administrative costs. This, however, is a very controversial effect, since industry competition and administrative costs is affected by many factors, like the size of the industry, the stage of development of capital markets, and the ability of workers to switch between funds. 3.3 Issues in member choice ofinvestments The structural constraint that has received most attention is the limit of one fund per admninistrator. In order to ensure an adequate degree of matching between investor preferences and the portfolio chosen by the funds, the solution would be to liberalise the investment market to give employees choice over how their pension fund is invested. The main counter-argument is one of cost and complexity. Dividing individual pension contributions between different funds (even when they are offered by the same manager) and 13 See Jagannathan and Kocherlakota (1996). This is a key attraction of defined-contribution plans over defined- benefit which also tie the worker's pension to future earnings. See Disney and Whitehouse (1994, 1996) and Bodie, MIarcus and Merton (1988). 14 Constantinides, Donaldson and Mehra (1998) suggest that liquidity constraints prevent younger workers from investing as much as they should in equities. This behaviour in turn may help explain the 'equity premium' or the excess risk-adjusted return observed on equities compared with short-term government bonds. A defined-contribution pension could alleviate this problem, if workers have some control over their portfolio. See also Blanchard (1993), Jagannathan and Kocherlakota. (1996) and Mehra and Prescott (1985). 15 transferring ilvestments between funds on members' request adds significantly to the administrative burden. Providing information on different investment options and educating workers about investment choice would also be expensive. There is also the risk that workers make the 'wrong' choices. Many studies of member- directed investment in 401(k) plans in the United States have found evidence for 'reckless conservatism', with people investing the majority of their fund in low-risk, low-return instruments.15 Figure 1 (and Table A.I in the Appendix) show the allocation of 401 (k) investments from a large survey covering 18 per cent of 401(k) members.16 Overall, nearly 70 per cent of funds are invested in equities, with 15 per cent in bond or money-market funds and 15 per cent in guaranteed investment contracts. The pattern with age seems prudent. Older workers tend to reduce the proportion in equities and increase the allocation to bond and money-market funds and guaranteed investment contracts. These contracts, provided by insurance companies, provide for a 'holding period' during which a fixed rate of return is paid, guaranteed for the life of the contract. Withdrawals can be made at book value to provide benefits. There are, however, some important divergences from prudent investment. First, the large allocation to the stock of the employer: 28 per cent of the total invested in equities or 19 per cent of the total fund. A more diverse portfolio would be more sensible. Indeed, given individuals' future employment and wages are already dependent on the performance of their employer, any investment in the employer's stock seems imprudent. There is also evidence that a substantial minority are very conservative. Fifteen per cent of people have no equity investments at all, even though balanced funds or their own employer's stock. Although this may be a rational strategy for people in their 60s (25 per cent of whom have no equity investments), it certainly is not for people in their 20s (of whom 15 per cent avoid equity investments). 15 Regulations protect plans and sponsoring employers from fiduciary responsibilities if members are allowed a sufficiently broad choice of investments with different risk and return characteristics. The vast majority of plans intend to comply with these regulations, allowing members to choose investments (94 per cent of schemes covering 92 per cent of members according to survey data: KPMG Peat Marwick, 1998). 16 VanDerhei et al (1999). Earlier studies used much smaller data sets. These include Yaboboski and VanD)erhei (1996), who looked at 180,000 members with three large employers. Goodfellow and Schieber (1997) analysed 36,000 participants in 24 schemes. Other papers have investigated investment choices in the Thrift Savings Plan (a defined- contribution scheme for federal employees) - Hinz, McCarthy and Turner (1997) - and in TLAA-CREF (a plan for teachers and college professors) - Ameriks, King and Warshawsky (1997). 16 Figure 1. Asset allocation in member-directed 401(k) pension plans 100%- 90% 'GCs .2 80%- X 70%- Bonds/ ~~~. 80%- ~~~~~~~money market ° 50% 40 Cs2 30%3 Equiti 0) 10) 0% 20s 30s 40s 5Os 60s age Note: investment in balanced funds is allocated 60 per cent to equities and 40 per cent to bonds, in line with the Investment Company Institute's data for the average balanced mutual fund Source: VanDerhei et al. (1999) In all, however, it is likely that workers would benefit from some degree of choice, like the two funds of the Polish system, where one fund is invested in a 'balanced' manner, and the other is more conservative. The need for at least two portfolios becomes more apparent when one looks into the future. As the new pension systems mature, older workers that are close to their retirement have a high preference for a conservative portfolio. 3.4 Empirical evidence of concentration in fund management Figure 2 shows the degree of concentration in the pension fund industry in Latin Amenrca and, for comparison, in the liberalised fund management market of the United Kingdom. The curves show the curnulative percentage of funds under management moving downwards from the largest fund. (Appendix Table A.2 gives detailed data.) The pattern in Latin America is remarkably similar, particularly between Chile and Argentina. The largest firm in Argentina, Chile and Mexico accounts for around 20-25 per cent of total assets, with the top three holding over half of funds, and the top five, around three-quarters. The situation is similar in Colombia, Peru and Uruguay (not shown in the Figure), where the largest three firms cover 60-75 per cent of total members.17 Bolivia has licensed only two funds. The situation is very similar in Hungary, although 45 funds were licensed initially. The three- firm concentration ratio for mandatory funds is 57 per cent, and the five-firm ratio, 71 per cent. 17 The voluntary pension sector is a little less concentrated. The three-firm ratio is 46 per cent and the five-firm ratio, 66 per cent. These ratios are exactly the same for voluntary funds in the Czech Republic. The fund management industry in the United Kingdom is significantly less concentrated than in Latin America. Prudential takes just 8 per cent of the market, with under a quarter of funds for the top three and a little over a third for the top five. Even the top 15 only accounts only for around three-quarters of funds. These funds include both individual's personal pensions and externally managed accounts for employer-provided pension plans. The largest employer fund managing its own assets - Hermes, which runs the pension schemes for the Post Office and British Teleconimunications - would rank 15-20th. Other sectors of the pension market in the United Kingdom are more concentrated. Employer-provided plans where funds are managed externally rely mainly on just five fund managers. A recent Pensions and Investments survey in the United States found a five-firm concentration ratio of 20 per cent and a 20-firm ratio of 40 per cent, significantly below even the United Kingdom figures of 36 and 72 per cent respectively. In both Chile and Argentina, there has been substantial recent consolidation in the pension funds industry. In 1994, there were 26 funds in Argentina, falling to 18 at the beginning of 1998 and 15 after three recent mergers (see the notes to the Appendix Table). In Chile, there were 21 funds in 1994, 13 at the beginning of 1998 and 10 now. Mexico has also experienced substantial consolidation, despite the relative infancy of its private pension fund industry. The number of fund managers has fallen from 17 in 1997 to 13 at present and some more mergers are expected soon. In other Latin American countries, reforms were more recent and there were fewer funds initially (e.,g., nine in Colombia, five in Peru, six in Uruguay, and two in Bolivia). Hence, it is not surprising that there has been little consolidation in these counties. Consolidation has already begun in Hungary, where the majority of the 45 funds are very small. Hungaria has already absorbed five of the smallest funds. Poland expects to have 10-12 funds after two years, although regulators expect to license mote funds initially. An important policy question is to whether the concentration in reforming countries is due to entry restrictions and structural regulations or is a natural consequence of the size of the market, the efficiency of capital markets, and the ability of workers to switch between fund managers. In 17 See Queisser (1998), chapter 4. 18 addition, the impact of concentration on industry competition, administrative costs, quality of service and capital markets should be explored. Figure 2. Concentration curves for fund managers in Latin America and the United Kingdom 100 Chile _. 90 80 70 60 *.. Mexico 50 40 .:rgnia 30 : ntdKndom 20 1 - 1 2 3 4 5 6 7 8 9 10 rank Source: Pension fund regulators in Latin America, HSBC James Capel for United Kingdom 4. Regulating performance Some countries - Chile, Argentina, Peru, Uruguay, and Colombia - require pension funds to achieve rates of return above a prescribed minimum, typically related to the industry average ('Table 4). Argentina and Chile define their profitability band in relative terms: the miniimum of 2 percentage points and 50 per cent (Chile) or 70 per cent (Argentina) above or below the average annual return of the industry18. The supervisory agency monitors compliance with the minimum on a monthly basis. All fund managers have to establish a reserve fund with their own capital (invested in the same way as the pension fund). If the reserve is insufficient to top up the fund's return to the minimum, the government guarantees the minimum. In Peru the minimum return is calculated in the same way as Argentina and Chile, but is not guaranteed by the government. There is no maximum return: the ceiling was eliminated in November 1996. There are also plans to move to a rate-of-return rule based on performance over five years. In Uruguay, the guarantee is expressed in both absolute and relative terms. The state- managed fund guarantees a minimum real return of 2 per cent a year, while private pension 18 Chile is considering changing the application of the rule to a 36-month rolling basis. 19 managers have to create a guarantee fund (similar to the reserve fund in Argentina and Chile). This fund is drawn down if the return falls below the average of the industry by more than 2 percentage points. There is also a lirit on the maximum return that funds can earn. Because the state managed fund -Repuiblica - dorninates the market average (56 per cent of total assets in May 1998), other pension funds are also forced to reach the 2 per cent real return. In Colombia, the minimum return is calculated as the arithmetic average of the return of the pension fund industry over three years and the return over three years of a market portfolio19. No ceiling is placed on the returns. The regulator checks compliance with the stipulated minimum return on a three-month basis. Table 4. Pension fund performance regulations and government guarantees in Latin America Minimum rate of return Maximum rate of return Government guarantee Argentina relative to average relative to average yes Bolivia no Chile relative to average relative to average yes Colombia relative to markets yes El Salvador relative to average relative to average yes Mexico no Peru (relative to average) no 2 per cent for Republica Uruguay relative to average relative to average yes Note: Maximum removed in Peru in November 1996. Minimum legislated but regulations not yet issued Source: Pension fund regulators Poland will place a lower limit of 50 per cent of the pension funds' average returns or four percentage points below the average. There will not be an upper lirmit. Hungary regulates the pension funds' performance relative to benchmark indices.20 Unlike asset restrictions, performance regulation is rare outside the privatised Latin American pension systems. In Brazil, non-occupational private pensions must deliver a minimmum real return of 6 per cent. In Singapore and Switzerland, minimum nominal returns of 4 and 2'/2 per cent respectively are imposed. But these are all absolute not relative limits, and are likely to be more damaging, since they encourage fixed-income investments, particularly when the guarantee applies to a short period. 19 From 1 July 1995, the composition of the market portfolio is (percentage of total pension industry assets invested in shares x 90 per cent of the average rate of return of the three stock exchanges in the country) + (percentage of total industry assets not invested in shares X 95 per cent of rate of return of a fixed-income index). As of June 1998, only 5 per cent of industry assets were invested in equities, so the market portfolio is mainly, a fixed-income index. 20 Chlon, Gora and Rutkowski (1998) and Palacios and Rocha (1998). 20 4.1 Rationale Performance regulation is normally encountered in non-competitive industries, such as utilities. Asset management, in contrast, is a competitive business and barriers to entry are fairly low. Investors are tvpically able to diversify away fund manager risk by investing in various funds. In the pension systems of Latin America, however, affiliates may only invest in a single fund managed by a specific fund manager. Since investment in pension funds is mandatory, individuals can neither avoid nor diversify away fund manager risk. In some countries, workers may not even transfer between funds within a specified period, which can be as long as six months. Performance rules ensure that the worker does not suffer from the exposure to this diversifiable, non-systematic risk. 4.2 Adverse effects ofperformance regulation The main adverse effect of performance regulation is to exacerbate 'herding' behaviour (Vittas, 1998b and Queisser, 1998a). Smaller fund managers behave like Stackelberg followers (Tirole, 1988), choosing portfolios similar to the larger funds, which have a greater weight in the industrv average return. Free from intense rate-of-return competition, the larger funds have an incentive to opt for lower risk-return assets, such as deposits and bonds. Return ceilings (as in Argentina, Chile, El Salvador and Uruguay) generate moral hazard in fund managers. At a given level of risk, there is no incentive to achieve a return above the ceiling and so the optimal point in the portfolio efficiency frontier might not be reached. Since returns no longer serve as a benchmark for comparing schemes, funds compete through advertising and marketing campaigns. The costs are passed on to consumers in the form of higher commissions. Portfolio homogeneity can be explained by other factors. First, the limit of one fund per manager forces them all to have a similarly balanced portfolio. Secondly, illiquidity of markets also encourages concentration of asset choice, as funds cannot easily take advantage of buying or selling opportunities. Thirdly, 'yardstick' competition, where managers measure their performance relative to their competitors, is entrenched even in countries with liberal regimes. Fourthly, an institution's trading decisions have informational content, which can be observed by its competitors and inferences drawn. Fifthly, fund managers tend to react in the same way to market news (e.g., the issue of macroeconomic data). Finally, the prudent-person legislation seems to be worded in a way that encourages herding. The United States rules say managers must invest "with the care, skill, prudence and diligence under the circumstances then prevailing that a prudent person, acting in a 21 like capacity and familiar with such mattes would use in the conduct of an enterprise of a like character and with like aims". The Employee Retirement Income Security Act of 1974 goes further than common law. It is not sufficient to be a careful amateur: managers must act as a prudent professional, experienced and educated in financial matters. Trustees of employer-provided pension plans surveyed in the United Kingdom reported that they took four main factors into account when determining investment policy: historic returns of different assets, the financial position of the scheme (the relationship between assets and defined- benefit liabilities), the scheme's maturity. Finally, and most important for our purposes, trustees said they took into account the asset allocation of other schemes. Indeed, the majority said they remained close to the average portfolios measured by WMNI (World Markets) or Combined Actuarial Performance Services (CAPS).21 However, compared with countries with prudent-person regulations, the degree of similarity in Latin American portfolios is much greater. Workers end up with practically identical portfolios, whichever their choice of manager. 4.3 Empirical evidence of herding Herding has become almost a defining characteristic of the pension fund industry in Latin America. Table 5 shows the mean and standard deviations of portfolio weightings of different assets in Chile. In equities, for example, the mean share of the portfolio is 29.4 per cent and the standard deviation is just 1.6. Table 5. The herding effect in Chile (per cent of assets of pension funds) Asset Average weighting Standard deviation Government bonds 39.4 4.3 Equities 29.4 1.6 Mortgage credit bills 16.8 3.9 Bank instruments 5.3 2.9 Corporate bonds 5.1 1.7 Source: Queisser (1998) The principal effect of herding is to generate very similar returns between different funds. Table 6 summarises the correlation in returns across pension funds in Argentina, Chile and Peru 21 Pratten and Satchell (1998). See also Bunt, Winterbotham and Williams (1998). 22 from the inception of their systems until May 1998. The average correlation between pairs of funds is exceptionally high: 0.98 in Chile, 0.93 in Peru and 0.87 in Argentina. Since these countries have the most flexible regimes, the figures for the other countries are unlikely to be very different. Table 6. Correlation of pension fund returns Country Mean Range Argentina 0.94 0.72-0.94 Chile 0.98 0.97-0.99 Peru 0.93 0.88-0.96 Note: Based on annualised monthly returns. Includes only companies operating throughout the period Source: Authors' calculations based on data from Superintendencias de Administradoras de Fondos de Pensiones. Studies of other countries include Lakonishok et al. (1991) on the United States and Blake, Lehman and Timmerman (1997) on the United Kingdom. Pension funds in the United States invest mainly in the equities of large companies: they own 25 per cent of the stockmarket as a whole, but 55 per cent of the largest 100 companies.22 4.4 Performance regulation and herding The link between performance regulation and herding is controversial. Ramirez Tomic (1997) found that herding by Chilean pension funds had (perversely) decreased slightly after the fluctuation band around the minimum rate of return was narrowed. Valdes-Prieto and Ramirez (1999) revised Ramirez Tomic's figures, showing that the width of the band caused a statistically significant but very small increase in the degree of herding among Chilean pension funds. To investigate the impact of return ceiling on herding, we take a closer look at the case of Peru, which eliminated its upper band in November 1996. Until then, the constraint on the return was 50 per cent above or below the industry average. The removal of the upper limit might be expected to lead to greater dispersion of investment across asset classes, as a wider range of risks can now be taken. However, Table 7 shows that the opposite occurred. After the regulation changed, the squared deviations from the industry averages for the largest asset classes, such as equities and government bonds, fell. This suggests that removing upper limits on performance does not provide adequate incentive for taking greater risks than the industry average. A more definitive analysis will 22 Mlonks (1992) and Brancato (1994). 23 be possible when data from countries without portfolio limits, such as Bolivia and Mexico, becomes available. Table 7. Peru: Average pension fund portfolio and standard deviation, 1995-8 Industry Average Standard deviation 1995-96 1997-98 1995-96 1997-98 Government bonds 27.5 3.8 2.5 1.1 Corporate bonds 6.2 14.8 1.7 1.7 Bank securities 21.9 16.5 1.8 1.3 Time deposits 27.5 27.1 1.4 1.4 Shares 16.1 37.3 1.5 0.7 Mortgage-backed 0.9 0.5 0.6 0.2 Average 1.2 0.8 Note: Data are squared deviation from quarterly industry average, averaged over the periods (March 1995-March 1996) and March 1997- March 1998) and square-rooted. Source: Authors' calculations based on data from Superintendencia de Administrados de Fondos de Pensiones It is possible that other regulations (such as the limit of one fund per manager) and the structure of capital markets (for example, the supply of liquid investments) are more important than performance regulation in explaining herding and the lack of portfolio diversity. What is certain is that performance regulations have reduced - and indeed almost eliminated - the risk of below industry-average performance by specific fund managers to the point where all workers obtain a similar return, irrespective of their choice of pension fund. The result is that there is no real choice between different asset managers, and no performance reason for transferring between managers. Despite this, transfers in many reforming countries have been running at very high rates. In Chile, for example, 29 per cent of members transferred in 1997. In Argentina, regulations designed to reduce transfers have been introduced, which cut the annualised transfer rate from 18 per cent in December 1997 to 5 per cent in January 1998. Since then, however, the rate has increased again, but only to 7/2 per cent. Chile is currently looking at reducing transfers by allowing funds to cut charges for long stayers. Poland has adopted such a policy as a way of liniting transfers. With little difference in portfolios between funds, this transfer process is, at least in part, wasteful. And the marketing costs of wooing and keeping new members, including, in Chile, now-banned practices such as gifts and promotions, an indication of the degree of waste. Sweden is to adopt a 'clearing- house' system to try and limit direct marketing. Contributions will be collected centrally and 24 allocated to chosen fund managers, but the managers will not know the identity of their members. This will not, however, preclude indirect marketing and promotional expenditure. 5. Regulating asset allocation Pension fund investments in all countries in Latin America are tightly controlled. Almost all countries' regulations include five types of limits * by asset class (a ceiling on the proportion of specific assets classes in a fund's portfolio); * by concentration of ownership (a ceiling on the proportion of the issue of a company that a given fund can hold); * by issuer (a ceihng on the proportion of assets in a fund's portfolio issued by the same institution); * by security (a ceiling on the proportion of individual securities im a fund's portfolio); * by risk (a mninimum acceptable risk rating of securities). The last four types of controls are a form of prudential regulation, similar to those of other institutional investors, like mutual funds. All countries impose restrictions on concentration by ownership, by issuer and by security.23 In addition, most reforming countries have restricted the securities eligible for investment to those that have been risk rated. In Chile, the mninimum acceptable nrsk category for fixed-income securities is BBB or equivalent. The law requires all investments - not just fixed-income securities - to be rated. This rating system for stocks has meant that only 30, mainly blue chip companies, out of a total of approximately 300 listed were eligible for pension fund investment until 1997. The new capital market reform bill, approved in 1997, extended coverage to more than 200 companies with smaller capitalisation and to other financial instruments, such as project financing, securitised bonds and venture capital. Concentration of ownership is limited in Chile through ceilings on the proportion of a firm's bond or share issue that any fund can hold, currently 20 and 7 per cent respectively. Minimum diversification requirements are also imposed, limiting funds to 7 per cent of fixed-income securities and 5 per cent of shares from the same issuer. To avoid conflicts of interest, the limits are set lower 23 Exceptions include the large balances invested by 401(k) participants in the United States in their employer's stock (see Table 2 above). Reserve funding systems, such as those in Germany, Japan and Luxembourg are equivalent to investing all of the fund in the sponsoring employer's equity. 25 for issuers that have financial interests in the pension fund managing companies. There are similar prudential rules elsewhere. In addition to these prudential rules, some countries also impose direct constraints on asset allocation. Countries tend to take two approaches to regulation of asset allocation, which VTittas (1996) describes as 'Draconian' and 'relaxed'. The latter refers to countries that apply the 'prudent- person' principle as described in section 4.2. (Countries with few or no restrictions on investments are listed at the top of Appendix Table A.3.)24 Secondly, countries which impose limits, usually either a mninimum investment in public bonds (between 15 and 50 per cent of total assets) or a maximum in equities (between 20 and 30 per cent of total assets), including Denmark, France, Germany, Japan, Norway, Portugal and Switzerland. Other countries have quantitative limits on investments in particular assets or asset classes. ('These are listed countries in the lower panel of Appendix Table A.3.) For example, around half of OECD countries have limits on foreign investments, averaging around 16 per cent of total funds (Figure 3). Belgium, Denmark, Portugal and France impose a minimum investment in bonds (Figure 4). Six countries limit equity holdings (Figure 4, again) and eight, investment in property. Figure 3. Limits on foreign investments in OECD countries Portugal Switzerland Japan Greece I Germany -: Finland Denmark Sweden Canada Poland I France Czech Republic average 16% 0 5 10 15 20 25 30 35 40 45 ceiling as a percentage of portfolio Source: Laboul (1992), Davis (1998), EFRP (1996), Watson Wyatt (1997), Chlon, Gora and Rutkowski (1998) 24 See Blommenstein (1998), Davis (1995) and OECD (1998), chapter V. 26 Figure 4. Limits on domestic investments in OECD countries Belgium, Portugal - France- Poland Denmark Switzerland Japan Germany 4 Minimum in public bonds Norway: Italy _Maximum in equities - 0 20 40 60 80 100 percentage of portfolio Source: Laboul (1992), Davis (1998), EFRP (1996), Watson Wyatt (1997) The portfolio restrictions imposed by regulators in May 1998 in seven Latin American countries are summarised in Figure 5 and shown in detail in Appendix Table A.4. In some countries, although legislation allows a more liberal investment regime, regulators have imposed tighter restrictions. In Chile and Bolivia, the law establishes a band for the ceiling by asset class. The regulator must then fix the ceiling within the value of the band. In Argentina, the law only sets out portfolio maxima. For example, the ceilng on equities is 50 per cent by law, but the regulator permits only 35 per cent of the fund to be invested in this asset class. All countries have tight portfolio limits, but the most flexible svstems currently are Chile, Argentina, Colombia, and Peru (probably in that order). They are the only countries that permit equity and foreign investment (the highest limit on shares is Peru's of 40 per cent, and on foreign assets, Chile's of 12 per cent). In Bolivia, although the legislated limits on shares and foreign assets have been set at relatively high levels (50-90 and 10-50 per cent, respectively), funds have to invest a minimum amount in government bonds. In the first few months of the system, this was set at $180m per annum, only just below the actual flow of funds into the funds. In general, the limits encourage government debt holdings at the expense of equity and foreign assets. 27 Figure 5. Pension fund portfolio limits, 1998 Equities Corporate bonds Foreign Chile _ Argentina Colombia Peru Uruguay _ Mexico 3 0 20 40 60 80 100 limit, per cent of total portfolio Source: Pension fund regulators Uruguay and Mexico have the most restrictive regimes, although, as in Bolivia, they are supposed to be only temporary. In Uruguay, pension funds are subject to both minimum and maximum limits on investments in government securities. The band is expressed as a percenrage of the portfolio, and there is a phased program in which the band is to fall from 80-100 per cent in 1996 to 40-60 per cent in 2000. The laws allow the amount above the band to be invested in any security, but only time deposits have so far been approved. In Mexico, the regulator has so far only approved fixed-income instruments (largely government securities).25 Investment guidelines for pension funds have tended to become more liberal over time, peirmtting and extending investments in equities, foreign assets and less liquid assets, such as real estate and venture capital. The development of the regime in Chile, which has the longest experience, is shown in Figure 6. (Details are in Appendix Table A.5.) In general, the domestic investment regime currently in place in Chile, Argentina, Peru, Colombia and the new regime to be implemented in Poland is more liberal than in most of the OECD countries with statutory portfolio limits. On the other hand, these same OECD countries allow a higher share of the portfolio to be invested in foreign securities, and some also permit direct investment in property and lending to affiliates (at least employer pension plans). 25 The Mexican pensions law also requires that funds must invest in securities that encourage national productive activity, create infrastructure, generate employment, housing investment, and regional development (article 43). 28 Figure 6. Evolution of portfolio limits in Chile, 1982-1998 90 8q Venture capital o 70 Hedging t 6 Foreign 0. - 50 g i i i 31 Property 0 40 Mutual funds 0. 2q | | | | | | Equities 1982 1985 1990 1992 1995 1996 1998 Source: Superintendencia de Administradoras de Fondos de Pensiones 5.1 Rationale of asset allocation regulations Two common arguments for controls on international investment26 are first, that they limit volatile capital flows and hence achieve monetary sovereignty and macro-economic stability (Fontaine, 1997) and secondly, that they reduce capital flight and deepen domestic financial markets (Reisen, 1997). These are problems that are particularly relevant for developing countries, which would explain why in general the ceiling on investment in foreign securities is lower in these countries than in the OECD area. Five main arguments have been used to justify domestic portfolio limits * lack of experience in fund management and, in particular, the absence of adequate risk assessment models mean pension funds take 'excessive' risks * capital markets lack liquidity and transparency * fragile financial markets might jeopardise the sustainability of the pension reform * limiting the fund's overall risk can alleviate the moral-hazard problem caused by govemment pension guarantees 26 See Candia (1998) for a summary. 29 * the transition cost to a funded pension system may be prohibitively high for countries with large explicit debt burdens and so can be eased by requiring investment in government bonds As with restrictions on industry structure, asset-allocation limits are a way of isolating pension assets from agency and systemic risks in capital markets. The prudent-person nrle may not be viable where capital-market infrastructure is underdeveloped and prudential controls are not properly in place. Theoretical models, such as that of Corsetti and Schmidt-Hebbel (1996), support the government-debt argument to an extent. But they provide a case for floors on investment in government securities, not for ceilings. If the new pension funds were unwilling to hold the explicit debt burden created by the transition from pay-as-you-go to funded financing of pensions, interest rates would rise. This would, in turn, worsen government finances and crowd out private investment.27 All of these arguments apply only temporarily. Over time, the efficiency and effectiveness of fund managers should improve with experience and as prudential standards are adopted and the costs of the transition amortised. Regimes should therefore be relaxed over time and, eventually, move towards prudent-person rules. 5.2 Adverse effects of asset allocation regulations LTimits on asset classes have three main adverse effects: * constraints on portfolio diversification create systematic market risk, meaning that higher returns can only be achieved at higher relative risk * pension funds are more likely to control large shares of the markets in which they can invest, creating liquidity problems * capital market development might be hindered Modem portfolio theory provides the most critical perspective on portfolio limits. Shah (1997) uses a capital-asset-pricing model to show that asset restrictions hamper the ability of fund managers to earn the highest possible risk-adjusted return. Returns as high those in an unconstrained system can only be reached with greater risk. Or, for a given degree of risk, retirement income will be lower. This argument is particularly relevant for developing countries, 30 because the range of investment products is typically very limited when the new pension system was set-up. Further restricting portfolios can therefore have adverse consequence on the degree of risk diversification that can be achieved. Market power (the second adverse effect) has become more of a problem as systems develop. Figure 7 shows the percentage of the stockmarket owned by pension funds in a selection of OECD and Latin American countries. Chile comes top among Latin American countries, with 11 per cent of equities owned by pension funds. Pension funds account for a third of stocks in the United Kingdom, and a quarter in the United States. In the Netherlands, although pension funds are very large, only a quarter is invested in shares (compared with over three-quarters in the United Kingdom, for example). In contrast, Belgian funds' portfolio is the most heavily weighted in equities after Ireland and the United Kingdom, but the pension funds overall assets are relatively small. In other countries, both the funds' assets and their equity proportions are small. The concentration of equity ownership in pension funds' hands raises a number of issues. Eirst, liquidity problems. Coupled with the herding effect of performance regulation (see above), when shifts in asset allocation involve the majority of pension funds buying or selling at the same time, market prices can shift strongly (in an adverse direction). When the Chilean investment regime was partially liberalised in 1985, pension funds found it difficult to close their fixed-income positions without adversely affecting prices. Pension funds moved only gradually from fixed-income instruments into stocks.28 As a result, asset allocations become ossified, and changed only slowly in response to liberalisation of the investment regime. Walker (1993a, 1993b) looks at differences in risk-adjusted returns between Chilean funds and finds that smaller funds' variable income portfolios perform better than those of larger ones. He attributes this to the 7-per-cent limit of each company's shares that funds can hold. In fixed-income portfolios, he finds no significant differences. A second important issue anrsing from the concentration of ownership is corporate governance: whether pension funds make effective owners of stocks. This has been hotly debated in the United Kingdom and the United States, where strong movements for 'shareholder activism'; have developed. 27 See also Holzmann (1998b) on the issue of debt financing of the transition to a funded system. 28 The jump in the share of the proportion allocated to equities between 1990 and 1991 (from 11 to 24 per cent) is largely due to an extraordinary stock market real return of nearly 90 per cent that year. 31 Figure 7. Pension funds' equity holdings as a percentage of total stockmarket capitalisation, 1997 Luxembourg 0 Austna Do Spain 0 Colombia 1 Italy 1 Germany 1 average France 2 Belgium 2 Greece Vl 2 Portugal 3 Finland 3 Argentina Fg' 3 Peru ; --S - 4 Brazil 6 Sweden * 8 Denmark 11 .M .= Chile 11 Netherlands 13 Ireland C24 US eM 25 UK * _____ 33 0 5 10 15 20 25 30 35 pension fund holdings, per cent of stockmarket Source: De Ryck (1998); pension fund regulators 5.3 Empirical evidence ofportfolio limits and asset allocations Asset allocation vanres widely across countries. Appendix Table A.6 compares the portfolios of five Latin American countries with a range of OECD countries and two from Asia. Figure 8 focuses on the proportion invested in equities. With the exception of Mexico and Uruguay, the Latin American countries all invest above the average (24 per cent) proportion in equities. The highest proportion of funds are invested in equities in the English-speaking countries. In Australia, Ireland, the United Kingdom and the United States, the average equity holding is 60 per cent of the fund. At the other end of the scale are Mexico and Uruguay, which have only recently reformed their systems, Singapore, where the provident fund invests mainly in bonds, and a number of continental European countries. The first contributions have only just begun to flow into Hungarian pension funds, so most of the assets are currently invested in short-term deposits. 32 Figure 8. Equity investments as a percentage of total pension-fund portfolios United Kingdom ' United States . Ireland t. Australiar Belgium Brazil , Peru Canada . Chile G>_2: Sweden I Argentina I Denmark l - Netherlands - Luxembourg Malaysia . Switzerland average France Hungary . Austna Greece Finland Portugal Germany Italy Spain Singapore Mexico Uruguay 0 10 20 30 40 50 60 70 80 90 100 equities, per cent of total portfolio Source: De Ryck (1998), Manrscal (1998a,b,c,d), Asher (1998) 5.3.1. Latin American countries Table 13 shows the structure of portfolios in Argentina, Chile and Peru in June 1997 along with the legal maxima by type of instrument. For some instruments, restrictions have been binding.29 Table 14 shows how the relaxation of portfolio restrictions in Chile over time has led to 29 Information refers to aggregate portfolios. Restrictions do not necessarily require the aggregated amount to coincide with the legal upper limit. Also, individual funds usually establish lower-than-legal upper limits of their own, to avoid incurring the costs of asset hquidation when changes in the portfolio are required. Another reason for lower-than-legal limits in Argentina is that the supervisor values the funds, and, in exceptional cases, this may result in differences between official prices and those assumed by the pension-fund managers. 33 changes in portfolio composition. Pension funds have taken advantage of the ehmination of the ban on investment in equities. By 1997, they had invested nearly one quarter of their portfolio in stocks. The lowering of the limit on mortgage investments (from 70 per cent of the portfolio in 1981 to 50 per cent in 1990) had the opposite effect. However, the dramatic reduction in their portfolio share (from 51 per cent in 1983 to 17 per cent in 1997) is largely a consequence of a supply constraint. In 1997, pension funds owned over one half of all mortgages. The Table does not indicate the full extent of the impact of regulations on portfolio allocation. Other regulatory controls, such as limits on the concentration of ownership, can create a discrepancy between the effective hmit to which the funds are subject and the one stipulated in legislation. In Chile, for example, the 7-per-cent limit of a company's shares that a pension fund can own becomes binding for larger funds long before the overall equities limit of 37 per cent (Walker, 1993b). Iglesias (1990) calculated that, because of the 7-per-cent constraint, the effective =n-ut on equities for the largest Chilean funds was around 14.8 per cent, compared with the overall maximum of 30 per cent at that time. Pension funds in Latin America have so far only dipped their toes in the water of international markets. Foreign investment has been permitted in Chile since 1992, but only 1 per cent of the portfolio is now invested overseas, mainly via mutual funds. Table 13. Pension-fund portfolios and limits in Argentina, Chile and Peru Assets Argentina Chile Peru r/. fund) Actual Maximum Actual Maximum Actual Maximum Public-sector bonds 49 50 38 35/50 12 40 Private-sector bonds 5 28 4 30/50 16 35 Certificate of deposit 18 28 8 30/50 34 50 Equites 22 35 29 35/50 35 30 Mortgages 0 28 17 35/50 1 40 Others 6 - 4 - 3 - Total 100 100 100 Note: Data relate to June 1997 Source: Pension fund regulators 34 Table 14. Asset allocation of funds in Chile, 1981-97 percent of funds 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Government bonds 28 26 45 42 42 47 41 35 42 44 38 41 39 40 39 42 40 Mortgages 9 47 51 43 35 26 21 21 18 16 13 14 13 14 16 18 17 Deposits 62 27 3 13 21 23 29 30 22 18 13 11 8 6 7 7 14 Shares 4 6 8 10 11 24 24 32 32 30 25 23 Funds 1 3 3 3 Foreign 1 1 1 1 Corporate bonds 1 1 2 2 1 1 3 6 9 11 11 10 7 6 5 5 3 Source: Superintendencia de Administrados de Fondos de Pensiones 5.3.2. OECD counrifes Table 15 shows portfolios relative to limits for eight OECD countries with quantitative investment restricnons (Table 8 above). In most cases, the limits again do not seem to be binding, with the exception of the (soon to be abolished) equity limit in Japan and the (informal) equity limit in the Netherlands. In effect, fund managers in, for example, Germany and Switzerland have been far more conservative than the regulations would allow. This is also the case in international investment. Even countries with no restrictions invest very few assets abroad. In Belgium, Ireland and the United Kingdom this proportion exceeds 30 per cent. In the United States, the proportion is just 10 per cent. This effect is termed home bias, and there are a number of likely explanations.30 First, overseas investments imply additional exchange-rate, settlement and liquidity risks. While it is possible to hedge such risks, this can be costly and, as recent experience has shown, can be difficult in periods of extreme volatility, lack of liquidity or where historic relationships between markets break down. Secondly, pension funds' liabilities are almost wholly domestic, so it seems prudent to match them mainly with domestic assets. Thirdly, the type of benchmark or yardstick orientation of fund managers outlined in section 5.1 may play a role. Fourthly, the world market portfolio, as suggested for pension-fund investment by Kotlikoff (1994), may not be optimal if markets are inefficient.31 Moreover, there is also evidence that adverse, downward movements in world markets are more correlated than upward.32 Finally, some have argued that increased 30 See Adler and Jorion (1992), French and Poterba (1991), Solnik (1991), Nowakowskic and Ralli (1987) and Candia (1998). 31 Beltratti (1998) and Huel and Cozzini (1990). 32 Solnik, Boucle and Le Fur (1996). 35 integration of global capital markets mean that the benefits of diversification are decreasing.33 The correlation of returns between a broad United States equity index (the Standard and Poors 500) and returns in emerging markets was 0.41 in the period 1990-95, compared with 0.27 in 1975-95.34 A simnilar effect can be observed between the United States and Latin American markets: the correlations were 0.38 in 1990-95 and 0.24 in 1975-95. Investment returns among the major industrial economies are stronger: between Germany, the United Kingdom and the United States, the correlations are between 0.54 and 0.62. One exception is Japan: the correlation wvith return s in the United Kingdom and the United States is around 0.05.35 Table 15. Portfolios relative to regulations in eight OECD countries Equities Bonds/loans Property Deposits Foreign Belgium +32 -33 0 Canada -11 Denmark -18 +5 -13 Germany -25 -14 -14 Japan -3 +11 -18 -23 Netherlands -7 -4 Portugal -15 -47 -11 Switzerland -17 -35 -21 Source: EFRP (1996) 5.4 Empirical evidence ofpension fund retums In section 4.2 above we established that individual pension funds in Latin America perform very close to the industry average. We now assess performnance of pension funds relative to alternative investrnents. Funded pension systems of the type introduced in Latin America impose considerable fiduciary duties on governments. First, because governtnent mandates contributions. Secondly, because governments set investment allocation limits, and empirical evidence suggests that 90 per cent of individual funds' returns in Latin America can be explained by the investment regime, with only 1I0 per cent attributable to investment managers' performance. In this section, we compare pension fund investment performance with various market benchmark indices. While market benchmark comparison is common in the pension fund industry 33 Kessler (1996), Blommensten (1998) and OECD (1998). 34 Source: ICFA. 35 Holzmann (1998a), Table A.2. 36 in developed countries (especially in defined-benefit schemes), they are as yet rare in Latin America. Absolute returns are often quoted to demonstrate the 'success' of the new systems but returns can only be judged against alternatives. The Colombian supervisory agency has established its own market index that makes up half of the stipulated pension fund return. In Bolivia, the contract between the government and the pension funds requires a benchmark to be established, and permits funds would to raise commissions by 10 per cent if they reach the benchmark. But the government has not so far decided what the benchmark should be. 5.4.1 Latin American countries Table 16 evaluates performance in Chile, Argentina, and Peru against domestic market indices (to May 1998). The IFC index of equity returns comprises 60 per cent of the balanced portfolio, with 40 per cent from an index of bond returns. The Table gives the average annual real returns before fees and the standard deviation of returns, a simple measure of volatility. Pension funds only appear to have performed better than the benchmark in Argentina. However, it is important to note that around 25 per cent of the assets of Argentine pension funds are in an 'investment account'. This account, created after the Mexican peso devaluation in 1994, allows funds to avoid marking to market fixed income securities that lost significant value during the crisis. Hence, 'retum' figures for the Argentine pension fund industry should be interpreted with caution, since they are likely to be significantly overstated. Pension fund returns in Peru were only half the return of the balanced portfolio and three- quarters in Chile. However, the volatility of pension fund returns was much less than the variance in the balanced portfolio in all three countries. It must be remembered, however, that these three are the countries with the most liberal investment regimes. In countries with more stringent regimes pension funds can be expected to have performed relatively worse36. 36 An adequate evaluation of perfornance in countries like Bolivia, Uruguay, and MIexico cannot be carried out, however, because the tine period is too short. 37 Table 16. Returns on pension funds and balanced portfolios: Latin America (/0) Period Actual return Balanced Bond index Equity index portfolio Argentina 1994-97 11.9 (5.0) 11.5 (15.2) 8.9 (13.1) 12.8 (18.7) Chile 1981-97 11.2 (9.0) 15.4 (25.7) 7.6 (1.2) 17.9 (43.3) Peru 1993-97 7.7 (3.8) 14.6 (18.9) n/a 14.6 (18.9) Note: Balanced domestic portfolio is 40 per cent bonds, 60 per cent equities. Standard deviation in parentheses Source: Pension Fund Regulators, National Securities Commission, Central Banks, IFC 5.4.2 OECD countries: cross-national comparisons Earlier in this section, we showed that OECD countries' policies can broadly be divided between those with prudent-person rules and those with asset limnits. Comparing pension fund perfornance between the two groups of countries can provide some useful evidence on the effect of investment regulations. Figure 9 gives data for ten countries. Four - Australia, Ireland, the Netherlands, the United Kingdom and the United States - have systems best described as prudent person. The other six have some form of portfolio regulation (although the degree, of course, varies). The bars show actual returns for pension funds, the lines, the returns on a balanced portfolio (50 per cent bonds, 50 per cent equities). Table A.7 in the Appendix gives more details, and some data for another five countries. On the surface, the prudent-person countries perform significantly better, earning 9/2 per cent a year, compared with 6/2-7 per cent a year in the countries with asset limits. But this analysis is rather superficial for a number of reasons. First, it ignores risk. Funds in prudent-person countries have larger equity portfolios. Davis (1998) constructs a synthetic rate of return for pension funds over the period 1967-90. He couples data on the portfolio structure of funds in different countries with aggregate indices of the return on different asset classes to estimate pension funds returns. (Actual returns of pension funds will differ from this synthetic return.) Over this period, the standard deviation of returns in prudent-person countries was 11.1, compared with 8.1 in asset-limits countries. Thus, some of the higher return is bought at the price of higher volatility. Secondly, there may be many other correlated factors that explain the difference in returns between the two groups of countries, including other types of regulations, macroeconomic policies, taxation, structural factors etc. But the lines on the Figure, however, show that market returns on a balanced portfolio were somewhat lower in prudent-person countries. Thus, it was pension-fund rather than market performance that differs between the two sets of countries. 38 Figure 9. Returns on pension funds and balanced portfolios: OECD countries Prudent person balanced portfolio actual Ireland United Kingdom United States Netherlands _wau Asset limits Belgium Sweden Germany f- Japan _ __ Denmark Switzerland 0 2 4 6 8 10 annual average real return (%) Source: OECD (1998), Tables V.2 and V.3, based on EFRP (1996), Pragma Consulting, Davis (1998) 5.4.3 United Kingdom and United States Section 5.3.1 showed the rate of return to pension funds in Latin America relative to market returns. A comparable analysis for the United Kingdom and the United States, both of which have prudent-person rules rather than asset limits, is instructive. Lakonishok, Shleifer and Vishny (1992) investigated the performance of defined-benefit pension funds relative to the Standard & Poots 500 over the period 1983-89. Weighting each funds return equally, the average return fell 1.3 percentage points below the index return of 19 per cent. Weighting funds by value, the undet-performance was 2.6 percentage points. Over the same period, other institutional investors, such as mutual funds, outperformed the market. Since there are no asset limits, this under-performance should arise from some other structural factors such as market failure. A similar analysis for the United Kingdom shows marginal underperformance of pension funds' investmnents in domestic equities of around 0.3 percentage points over the period 1981-91.37 Government bond investments also performed at about the market average. The only significant 37 Dilnot el al. (1994), section 5.4 and Figure 5.4, based on data from Combined Actuarial Performance Services (1993). 39 underperformance was in investments in overseas equities, which were three percentage points below market indices, reflecting a conservative strategy with foreign investtnents. The lesson of these analyses is that it is too simplistic to attribute the whole of underperformance to investment regulations. Even in countries with prudent-person rules, there is some evidence that pension funds do not achieve market levels of returns. 6. Conclusions and policy implications Along with housing, pensions will be the largest asset most workers (at least in developing economies) own. Governments that have mandated pension contributions have a fiduciary responsibility and a financial interest (through inplicit and explicit guarantees) in ensuring that this important component of workers' savings provides the best possible returns. Governments have used this responsibility to justify Draconian regulations of pension funds' structure, performance, and investment allocation. The result of these regulations is that pension funds' portfolios are very similar and their returns practically indistinguishable. Such regulations provide little incentive for improved efficiencies in investment management. They also fail to offer workers significant portfolio choice. Although workers have their individual accounts, they have no real choice over how their contributions are invested. They have little real responsibility for determining their own financial future. A policy implication of the evidence presented in the paper is that investment regimes should be liberalised to allow diversification. Funds should be able to compete in offering different risk-return strategies, to allow workers with different degrees of risk aversion and at different points in their lifecycle to choose different portfolios. Restrictions when a reform is first introduced are probably necessary to bolster confidence in the system. Much of the risk at this point comes not from market volatility, but from systemic risk that could lead to the collapse of one or more of the private funds, or indeed, of the whole system. If new financial intermediaries and the restriction of a single fund manager per investor are deemed desirable, then performance regulations may also be required to ensure that investors in mandatory systems are not exposed to fund manager risk that they cannot diversify away. The key policy question then becomes how quickly should the system be liberalised? In Chile, which pioneered this type of reform, the answer was probably fairly slowly. In countries that have reformed more recently, the success of other countries' models should allow for far more rapid relaxation of investrnent restrictions. A medium-term goal should be to allow managers to offer 40 different types of funds. The long-term goal should be to move towatds a 'prudent-person' rule. This kind of regulation also has its faults, but is still preferable to a long-term policy of quantitative investment restrictions. Governments have a responsibility to ensure that mandatorv pension funds are managed well. It is therefore not unnatural that developing countries with a historv of poor performance of financial institutions err on the side of caution. Draconian regulations are designed to protect pension funds from fragile and underdeveloped financial systems, both in Latin America and in the transition economies of Eastern Europe and Central Asia. These regulations are not cost free, however, and it is critical that governments evaluate carefully the impact of the regulations they impose, since they can undermine many of the objectives of pension reforrn. 41 Appendix. Detailed data tables Table A.1. Asset allocation in member-directed 401(k) pension plans Equity of which, own Bond/money funds Guaranteed employer's stock investment contracts 20 77 22 14 8 30 76 26 14 9 40 72 29 14 12 50 67 29 15 16 60 53 28 18 26 Total 68 28 15 15 Note: investment in balanced funds is allocated 60 per c- -t to equities and 40 per cent to bonds, in line with the Investment Company Institute's data for the average balanced mutual fund Source: VanDerhei et a!. (1999) Table A.2. Concentration of fund managers in Latin America and the United Kingdom United Kingdom Chile Mexico Argentina Prudential 8.2 ProvidalUnion 23.2 Bancomer 25.1 Consolidar/Fecunda 18.9 Mercury 15.8 Habitat 42.5 Banamex 44.8 Origines/Clardad 36.3 Schroder 23.2 Cuprum 59.8 Inbursa 54.4 Maxima 53.1 Commercial Union 29.8 Santamaria 72.3 Bital 62.5 Siembra 67.8 Morgan Grenfell 35.9 Summa/Bansander 84.7 Profuturo 70.2 Previnta 78.0 Fleming 41.4 Proteccion 94.3 Garante 77.3 Nacion 84.7 PDFM 46.6 Planvital 96.9 Santander 82.7 Generar 88.8 Standard Life 51.0 Magister/Qualitas 98.7 XXI 87.9 Arauca Bil 91.8 INVSECO 55.2 Aporta 99.5 Banorte 92.4 Previsol 94.0 Norwich Union 59.3 Fomenta 100.0 Bancrecer/Dresdner 95.1 Prorenta/San Jose 96.1 Legal & General 62.8 Previnter 96.7 Future 98.0 BZW 66.3 Atlantico-Promex 97.7 Ethika 98.7 Threadneedle 69.2 Confia-Principal 98.4 Unidos 99.3 Hill Samuel 72.1 Tepeyac 98.8 Profesion + Auge 99.7 Note: Columns show the cumulative percentage of total funds under management. Figures for Argentina includes three recently announced mergers (Consolidar/Fecunda, Origines/Claridad, Prorenta/San Jose), as does Chile (Provida/Union, Summa/Bansander, Magister/Qualitas) Source: Pension fund regulators in Latin America, HSBC James Capel for United Kingdom 42 Table A.3. Pension asset regulations in OECD countries Domestic Intemational Prudent person Austria - no limits Australia no limits no limits Iceland - no foreign investments by public-sector funds (e.g. civil servants and fishermen) Ireland no limits no limits Netherlands no limits (informal 30% limit on equities) no limits New Zealand no limits no limits Spain no limits in other OECD countries United Kingdom no limits no limits United States no limits no limits Asset limits Belgium Minimum 15% in public bonds, maximum no foreign investments 40% in property, 10% in deposits Canada 7% maximum on property tax on foreign assets over 10% Czech Republic - no foreign investments Denmark Minimum 60% in domestic debt; property, 20% limit equities and mutual funds maximum 40% Finland - 20% limit in other EU states (lower limit on property, higher on government bonds) France minimum 50% in EU public bonds no foreign assets (insured funds) (AGI RC/ARRCO) minimum 34% in public bonds, 40% limit on property and 15% Treasury deposits (insured funds) Germany guidelines: 30% limit on EU equities, 25% 20% limit on foreign assets overall; 6% EU property limit on non-EU equities, 6% on non-EU bonds Greece - 20% limit on domestically based mutual funds, which can invest abroad Italy limited to public bonds, deposits, property, no limits mortgages, investment funds (insured funds) Japan guidelines (being phased out): 30% limit on 30% limit on foreign assets; 10% limit in equities, 20% property; minimum 50% bonds any one country Norway 20% limit on equities, 30% on private bonds no limits or loans Portugal minimum 30% in public bonds, 50% limit on 40% limit property Poland 20% limit on bank deposits or securities, 5% limit on foreign assets 40% in listed equities, 15% in open-ended investment funds, 5% in closed-end funds, 15% in publicly traded municipal bonds, 5% in untraded bonds; property, commodity and derivatives investments prohibited Sweden majority of investments in listed bonds and 5-10% limit, depending on type of fund loans Switzerland 30% limit on equities, 55% on property 30% total limit, 30% in foreign bonds, 25% in foreign equities, property 5% Source: Laboul (1992), Davis (1998), EFRP (1996), Watson Wyatt (1997), Chlon, Gora and Rutkowski (1998) Note: - indicates data are unavailable 43 Table A.4. Pension fund portfolio limits, 1998 Asset Argentina Chile Colombia Mexico Peru Uruguay Government securities (total) 65 50 50 100 30 75-85 Federal 50 Provincial and municipal 15 Central bank 30 Corporate bonds (total) 40 45 20 35 35 25 Long term 28 Short term 14 10 Convertibles 28 10 Privatised firms 14 Bank bonds 50 10 25 25 Mortgage-backed securities 28 50 30 30 Letters of credit 50 Fixed-term deposits 28 50 30 30 Short-term margin loans 10 Repurchase agreements 15 Shares, public companies 35 37 30 0 20 25 Shares, workers' shares 20 Preference shares 10 Shares, privatised companies 14 Stock index instruments 5 Securitised instruments 20 Primary issues, new ventures 10 Mutual funds 14 5 5 10 0 Real estate funds 10 Venture capital funds 5 Securitised credit funds 5 Direct investment funds 10 Foreign securities (total) 10 12 10 0 5 Government securities 10 Corporate bonds/shares 7 0 Fixed income 12 10 Variable income 6 Hedging instruments 2 9 10 Note: Argentina: The Nacion pension fund must invest between 20 and 50 per cent (or $300m) in provincial and municipal bonds to finance regional projects. Colombia: a limit of 15 per cent is imposed on investment securitised instruments backed by non-admitted assets, real estate and infrastructure Source: Pension fund regulators 44 Table A.5. Evolution of porffolio limits in Chile, 1981-1998 Asset 1981 1982 1985 1990 1992 1995 1996 1997 1998 Government securities 100 100 50 45 45 50 50 50 50 Corporate bonds 60 60 40 40 40 40 45 45 45 Convertible 10 10 10 10 10 10 10 Mortgage-backed securities 70 40 40 50 50 50 50 50 50 Letters of credit 70 40 40 50 50 50 50 50 50 Fixed term deposits 70 40 40 50 50 50 50 50 50 Shares, public companies 30 30 30 37 37 37 37 Mutual funds 10 10 10 5 5 5 Real estate funds 10 10 10 10 10 10 Venture capital funds 5 5 5 Securitised credit funds 5 5 5 Foreign securities 3 9 9 12 12 Fixed income 9 9 12 12 Variable income 41/2 41/2 6 6 Hedging instruments 9 9 9 12 Source: Superintendencia de Administradoras de Fondos de Pensiones Table A.6. Pension fund portfolios, selected countries (% of portfolio) Equities Fixed interest United Kingdom 78 14 United States 62 27 Ireland 58 30 Australia 41 15 Belgium 40 46 Brazil 38 38 Peru 35 60 Canada 28 48 Chile 28 68 Sweden 28 62 Argentina 27 70 Denmark 27 63 Netherlands 26 63 Average 24 56 Luxembourg 21 61 Malaysia 16 55 Switzerland 14 69 France 14 38 Hungary 14 19 Austria 13 71 Greece 10 53 Finland 9 61 Portugal 9 27 Germany 8 74 Italy 8 63 Spain 5 76 Singapore 0 70 Mexico 0 96 Uruguay 0 100 Source: De Ryck (1998), Mariscal (1998a,b,c,d), Asher (1998) 45 Table A.7. Returns on pension funds and balanced portfolios: OECD countries Annual average, Actual retums Balanced domestic real, % 1984-96 1984-93 portfolio Prudent person 9.5 9.5 3.4 Australia 2.7 Ireland 11.0 10.3 3.8 Netherlands 8.0 7.7 4.5 United Kingdom 10.0 10.2 3.8 United States 9.0 9.7 2.1 Asset limits 6.5 6.9 4.0 Belgium 9.0 8.8 4.2 Canada 2.2 Denmark 6.0 6.3 5.3 France 5.2 Germany 7.0 7.2 6.1 Italy 1.9 Japan 6.5 5.5 Spain 7.0 Sweden 8.1 3.8 Switzerland 4.0 4.4 2.0 Note: Balanced domestic portfolio is 50 per cent bonds, 50 per cent equities. Source: OECD (1998), Tables V.2 and V.3, based on EFRP (1996), Pragma Consulting, Davis (1998) 46 7. Bibliography Adler, MI. and Jorion, P. (1992), 'Foreign portfolio investment', in New Palgrave Dictionary of Money and Finance, Macmillan, London. Altman, N. (1992), 'Govemment regulation: enhancing the equity, adequacy and security of pension benefits', in OECD, Private Pensions and Public Polig', Paris. Ameriks, J, King, F.P. and Warshawsky, M. (1997), 'Premium allocations and accumulations in TIAA-CREF - trends in participant choices among asset classes and investment accounts', TL7A-CR-EF Research Dialogues no. 51. Asher, M.G. (1998), 'The financial crisis and its inplications for pension funds in South-East Asia', presentation', presented at EDI/World Bank conference, Hangzhou, China, April 1998. Bajtelsmit, V.L. and VanDerhei, J.L. (1997), 'Risk aversion and pension investment choices' in Gordon, M., Mitchell, O.S. and Twinney, M. (eds), Positioning Pensions for the Twen!y-First Century, University of Pennsylvania Press, Philadelphia, Penn. Ball, R. Kothari, S.P. Shanken, J. (1995), Problems in measuring portfolio performance: an application to contrarian investment strategies', Journal of FinancialEconomics, vol. 38, pp. 79-107. Bassett, W.F., Fleming, M.J. and Rodrigues, A.P. (1998), 'How workers use 401(k) plans: the participation, contribution and withdrawal decisions', National TaxJournal, vol. 51, no. 2, pp. 263-289. Baxter, M. and Jermann, UJ. (1997), 'The intemational diversification puzzle is worse than you think', American Economic Review, vol. 87, no. 1. Beltratti, A. (1998), Asset allocation of pension funds: identification of benchmarks', in Blommenstein, H.J. and Funke, N. (eds), Institutional Investors in the New Financial Landscape, OECD, Paris. Benston, GJ. (1988), Regulating Financial Markets: A Critique and Some Proposals, Hobart Paper no. 135, Institute of Economic Affairs, London. Blake, D., Lehmann, B.N. and Timmerman, A. (1997), Performance measurement using multiple asset class portfolio data: a study of UK pension funds', Discussion Paper no. 1618, Centre for Economic Policy Research, London. Blanchard, OJ. (1993), 'The vanishing equity premium', in O'Brien, R. (ed.), Finance and the International Economy 7, Oxford University Press. Blommenstein, HJ. (1997), 'Institutional investors, pension reform and emerging securities markets', Working Paper no. 359, Office of the Chief Economist, Inter-American Development Bank, Washington, DC. -(1998), 'Impact of institutional investors on financial markets', in Blommenstein, H.J. and Funke, N. (eds), Institutional Investors in the New Financial Landscape, OECD, Paris. Bodie, Z. (1990), Pensions as retirement income insurance', Journal ofEconomic Literature, vol. 28, pp. 28-49. -(1996), 'W7hat the pension benefit guarantee corporation can learn from the federal savings and loans insurance', Journal of Financial Services Research, no. 10, pp. 83-100. Marcus, AJ. and Merton, R.C. (1988), 'Defined-benefit versus defined-contribution pension plans: what are the real trade-offs?', in Z. Bodie, J.B. Shoven and D.A. Wise (eds), Pensions in the US Economy, University of Chicago Press for National Bureau of Economic Research. -and Merton, R.C. (1994), Pension benefit guarantees in the United States: a functional analysis', in Schmitt, R. (ed.), The Future of Pensions in the United States, University of Pennsylvania Press. -, - and Samuelson, W.F. (1 992), 'Labour-supply flexibility and portfolio choice in a life-cycle model', Joumnal of Economic Dynamics and Control, vol. 16, no. 3, pp. 427-449. Brancato, C.K. (1994), The Brancato Report on Institutional Investment, the Conference Board, Seminar on Trends in U.S. Institutional Investment and Implications for Corporate Governance, New York. Brown, S.J. and Goetzmann, W.N. (1995), Performancepersistence, Journal of Finance, vol. 50, pp. 679-698. -, Ibbotson, R.G. and Ross, S.A. (1992), 'Survivorship bias in performance studies', Review of Financial Studies, vol. 5, pp. 553-580. 47 -, Ibbotson, R.G. and Ross, S.A. (1992), 'Survivorship bias in performance studies', Review of Financial Studies, voL 5, pp. 553-580. -,-and Ross, S.A. (1995), Survival,journal of Finance, vol. 50, pp. 853-873. Bunt, K., Winterbotham, N. and Williams, R. (1998), The Role of Pension Scheme Trustees, Research Report no. 81, Department of Social Security, London. Campbell, J.Y., Cocco, J.F., Gomes, F.J. and Maenhout, P.J. (1999), 'Investing retirement wealth: a life-cycle model', Working Paper no. 7029, National Bureau of Economic Research, Cambridge, Mass. Candia, F. (1998), 'International diversification for LAC funds: why or why not?', presented at INID/EDI World Bank conference, LAC Pension Systemns: Investino for the 21st Centu9', Washington, DC., July 1998. Carhart, M.M. (1997), 'On persistence in mutual fund perforrnance', Journal ofFinance, vol. 52, pp. 57-82. Chen, Z., and Knez, PJ. (1995), 'Portfolio performance measurement: theory and applications', Review of Financial Studies, vol. 9, pp. 511-555. Chisari, 0. and Dal B6, P. (1996), 'Las regulaciones a la composici6n de cartera y las inversiones de los fondos de jubilaciones y pensiones: un ejercicio de simulaci6n para el caso Argentino', Revista de And/isis Economico, vol 11, no 1. Clark, R. Goodfellow, G.P., Schieber, S.J. and Warwick, D. (1998), 'Making the most of 401(k) plans: who's choosing what and why?' Working Paper no. 98-12, Pensions Research Council, Philadelphia, Penn. Combined Actuarial Performance Services (1993), Pension Fund Investment Performance: General Report, CAPS, Leeds. Conistanrinides, G., Donaldson, J. and Mehra, R. (1998), "'Junior can't borrow." A new perspective on the equity premium puzzle', Working Paper no. 6617, National Bureau of Economic Research, Cambridge. Corsetti, G. and Schmidt Hebbel, K (1997), Tension reform and growth', in Valdes-Prieto, S. (ed.); The Economics of Pensions: Principles, Policies, and International E,xperience, Cambridge University Press. Da-is, E.P. (1998), 'Regulation of pension fund assets', in Blommenstein, H. J. and Funke, N. (eds), Institutional Investors in the New Financial Landiscape, OECD, Paris. Demarco, G., Rofman, R., with Whitehouse, E.R. (1998), 'Supervising mandatory funded pension systems: issue and challenges', Pension Reform Primer Series, Social Protection Discussion Paper no. 9817, World Bank, Washington, D.C. Demirgu,-Kunt, A. and Levine, R. (1996), 'Stock-market development and financial intermediaries: s-ylised facts', World Bank Economic Review, vol. 10, no. 2. De Ryck, K. (1998), 'Asset allocation, financial market behaviour and impact of EU pension funds on European capital markets', in Blommenstein, H.J and Funke, N. (eds), Institutiional Investors in the New Financial Landscape, OECD, Paris. Dilnot, A.W., Disney R.F., Johnson, P.G. and Whitehouse, E.R. (1994), Pensions Poliy in the UK- An Economic Analysis, Institute for Fiscal Studies, London. Disney, R.F. and Whitehouse, E.R. (1994), 'Choice of private pension and pension benefits in Britain', Institute for Fiscal Studies, Working Paper no. 94/2. -and - (1996), 'What are pension plan entitlements worth in Britain?', Economica, vol. 63, pp. 213-238 Dooley, M.P. (1995), 'A survey of academic literature on controls over international capital transactions', Working Paper no. 5352, National Bureau of Economic Research, Cambridge. Duval, D. (1999), 'Institutional Arrangements for mandatory funded pensions', Pension Reform Primer series, Social Protection Discussion Paper, World Bank, Washington, D.C., forthcoming. Edey. M and Simon, J. (1996), 'Australia's retirement income system: implications for saving and capital markets', Working Paper no. 5799, National Bureau of Economic Research, Cambridge, Mass. Edwards, S. (1996), 'Chile: los fondos mutuos', Regional Studies Report no. 39e, Latin America and Caribbean Technical Department. 48 Elton, E.J., Gruber, M.J. and Blake, C.R. (1996a), Survivorship bias and mutual fund performance', Review of FinancialStudies, vol. 10, pp. 1097-1120. -, - and- (1996b), 'The persistence of risk-adjusted mutual fund performnance', Joumnal of Business, vol. 69, pp. 133-157. -, Das, S. and I-lavka, WI. (1993), 'Efficiencv with costly information: a reinterpretation of evidence for managed portfolios', Review of Finanial Studies, vol. 6, pp. 1-22. European Federation for Retirement Provision (1996), European Pension Funds: Their Impact on European Capital Markets and Competitiveness, EFRP, London. Fama, E.F., and French K.R., (1993), 'Common risk factors in the return on bonds and stocks', Journal of FinantialEconomics, vol. 33, pp 3-53. Ferson, W. and Schadt, R. (1996), 'Mseasuring fund strategy and performance in changing economic conditions', Journal of Finance, vol. 51, pp. 425-462. -and Warther, V.A. (1996), 'Evaluating fund performance in a dynamic market', Finania/l4na/ysts Journal, pp. 20-28. Flanagan, K. (1999), 'Pension reform in Australia', Pension Reform Primer series, Social Protection Discussion Paper, World Bank, Washington, D.C., forthcoming. French, K.R. and Poterba, J.M (1991), 'Investor diversification and international equity markets', 4merican Economic Revziew, vol. 81, pp. 222-226. Fontaine, J. A. (1997), 'Are there good macroeconomic reasons for limiting external investments by pension funds? The Chilean experience', in Valdes-Prieto, S. (ed.), The Economics of Pensions: Principles, Policies and International Experience, Cambridge University Press. Frijns, J. and Petersen, C. (1992), 'Financing, administration and portfolio management: how secure is the pension promise?' in OECD, Private Pensions and Public Poliy, Paris. Goodfellow, G.P. and Schieber, S.J. (1997), 'Investment of assets in self-directed retirement plans', in Gordon, M., Mitchell, O.S. and Twinney, M. (eds), Positioning Pensions for the Twen.t'-First Centuy, University of Pennsylvania Press, Philadelphia, Penn. Chlon, A., G6ra, M. and Rutkowski, M. (1999), 'Shaping pension reform in Poland: Security through Diversity', Pension Reform Primer series, Social Protection Discussion Paper no. 9923, World Bank, Washington, D.C. Heller, P.S. (1998), 'Rethinking public pension reform initiatives', Working Paper no. 98/61, Fiscal Affairs department, International Monetary Fund, Washington, D.C. Hinz, R., McCarthy, D. and Turner, J. (1997), 'Are women conservative investors? Gender differences in participant-directed pension investrnents', in Gordon, M., Mitchell, O.S. and Twinney, M. (eds), Positioning Pensions for the Tweny-First Century, University of Pennsylvania Press, Philadelphia, Penn. Holzmann, R. (1998a), 'A World Bank perspective on pension reform', Pension Reform Priner series, Social Protection Discussion Paper no. 9807, World Bank, Washington, D.C.. - (1998h) 'Financing the transition to multipillar', Pension Reform Primer series, Social Protection Discussion Paper no. 9809, World Bank, Washongton, D.C.. Howell, M. and Cozzini, A. (1990), International Equiy Flowus, Salomon Brothers. Iglesias, A. (1990), 'Estrategias de Inversi6n de los Fondos de Pensiones', presented at Third Congess of Finance, Escuela de Administraci6n, Pontificia Universidad Cat6lica de Chile, August 1990. Jagannathan, R. and Kocherlakota, N. (1996), 'Why should older people invest less in stocks than younger people?', Federal Reserve Bank of Minneapolis Quarterly Review, vol. 20, no. 3, Summner. Kessler, C. (1996), 'Diversification - Is it still alive?' Economic and Financial Prospects, no. 6, Swiss Bank Corporation. Kotlikoff, L.J. (1994), 'A critical review of the World Bank's social security analysis', Education and Social Policy department, World Bank. KPMG Peat Marwick (1 998), Retirement Benefits in the 1990s: 1998 Survgey Data, Washington, D.C. 49 Laboul, A. (1998), 'Private pension systems: regulatory policies', Ageing Working Paper no. 2.2, OECD, Paris. Lakonishok, J. Shleifer, A. ar Vishny, R.W. (1992), 'The structure and performance of the monev management industry', &ings Papers on Economic Activi¸: Microeconomics, pp. 339-93. -, -, Thaler, R and Vishny, k . (1991), 'Window-dressing by pension fund managers', American? Economic Review. Leibowitz, M. L., Bader, L. and Kogelman, S. (1998), Return Targets and Shortfall Risks: Studies in StraPgic Asset AManagemenA Irwin. Llewellyn, D. (1999), 'The economic rationale for financial regulation', Occasional Paper no. 1, Financial Services Authority, London. Mariscal, OJ. (1998a), 'Argentine Pension and mutual funds and their impact on stocks', Goldman Sachs Investment Research, New York. Mlariscal, OJ. (1998b), 'Brazilian Pension and mutual funds and their impact on stocks', Goldman Sachs Investment Research, New York. Manrscal, OJ. (1998c), 'Chilean Pension and mutual funds and their impact on stocks', Goldman Sachs Investment Research, New York. Mariscal, OJ. (1998a), 'Mexican Pension and mutual funds and their impact on stocks', Goldman Sachs Investment Research, New York. Mehra, R. and Prescott, E.C. (1985), 'The equity premium: a puzzle', Journal ofMonetay Economics, vol. 15, pp. 145-161. Merton, R.C. and Bodie, Z. (1992), 'On the management of financial guarantees', Financial Maragement, Winter, pp. 87-109. Monks, R.A.G. (1992), 'Reckless "prudence": investment of pension fund assets in the United States of America', Lens. Musumeci, J., 'Investing for a distant goal: optimal asset allocation and attitudes toward risk', TL7AI-CREF Research Dialogues, no. 56. Nowakowski, C. and Ralli, P. (1987), 'Intemational investment, diversification and global markets', in Altrnan, E.I. (ed.), Handbook of FinancialMarkets and Institutions, John Wiley, New York. OECD (1998), Maintaing Prosperi¸y in an Ageing Socie~', Paris. Otermin, M. and Zablotzky, E. (1996), Anilisis de la Performance de los Fondos de Jubilaciones y Pensiones en Argentina, Centro de Estudios Macroecon6micos de Argentina. Palacios, R. and Rocha, R. (1998), 'The Hungarian pension system in transition', Pension Reformn Primer series, Social Protection Discussion paper no. 9805, World Bank, Washington, D.C. -and Whitehouse, E.R. (1998), 'Individual choice and the transition to a funded pension system', Pension Reform Primer series, Social Protection Discussion Paper no. 9812, World Bank, Washington, D.C. Papke, L.E. (1998), 'How are participants invesdting their accounts in participant-directed individual account pension plans?' American Economic Review, vol. 88, no. 2, pp. 212-216. Parisi, F. and Parisi, A. (1997), 'Desempenio y ranking de los fondos mutuos de renta vanrable en Chile desde 1992 a 1995', Revista deAndlisis Econdmico, vol. 12, no. 2, pp. 101-123. Poterba, J.M and Wise, D.A. (1998), 'Individual financial decisions in retirement savings plans', in Feldstein, M. (ed.), Privatising Social Securi!y, University of Chicago Press for National Bureau of Economic Research. Pratten, C. and Stachell, S. (1998), Pension Scheme Investment Policies, Research Report no. 82, Department of Social Security, London. Queisser, M. (1998a), 'Regulation and supervision of pension funds: principles and practices', International Social Security Review, vol. 51, no. 2. -(1998b), 'The second-generation pension reforms in Latin America', Ageing Working Paper no. 5.4, OECD, Paris. 50 Ramirez Tomic, M. (1997), 'Efecto manada en la administraci6n de los fondos de pensiones Chilenos: relevancia de la regulaci6n vigente', mimeo., Seminario de Titulo Microeconomia, Santiago. Reisen, H. (1997), 'Liberalising foreign investment by pension funds: positive and normnative aspects', World Development, July. - (1998a), 'Liberalising foreign investment by pension funds: positive and normative aspects', Ageing Working Paper no. 5.3, OECD, Paris. - (1998b), 'Warning: past pension fund performance is no guarantee for future performance', in Blommenstein, HJ. and Funke, N. (eds), Institutional Intestors in the Nex Financial Landscape, OECD, Paris. - and Williamson, J. (1997), 'Pension Funds, capital controls, and macroeconomic stability', in Valdes- Prieto, S. (ed.), The Economics of Pensions. Principles, Policies and International Experience, Cambridge University Press. Ro;as-Suarez and Weisbrod, S. (1996), 'Building stability in Latin American financial markets', pp. 139-166 in Hausmann, R. and Reisen, H. (eds), Securins Stability and Growth in Latin Awmerica: Polig Issues and Proipects for Shock-Prone Economies, OECD, Paris. Samuelson, P.A. (1989a), The judgement of economic science on rational portfolio management: indexing, timing and long-horizon effects', Journal of Portfolio Management, pp. 4-12, Fall. -(1 989b), 'A case at last for age-phased reduction in equity', Proceedings of the National Academy of Sciences, vol. 86, pp. 9048-9051. Schwarz, A.M. and Demirguc-Kunt, A. (1999), 'Taking stock of pension reforms around the world', Pension Reform Primer series, Social Protection Discussion Paper no. 9917, World Bank, Washington, D.C. Shah, H. (1997), 'Toward better regulation of private pension funds', Policy Research Working Paper no. 1791, World Bank. Shoven, J.B. (1999), 'The location and allocation of assets in pension and conventional savings accounts', Working Paper no. 7007, National Bureau of Economnic Research, Cambridge, Mass. Singh, A. (1993), 'The stock-market and economic development: should developing countries encourage stock markets?', UNCT1D Review, no. 4, United Nations Conference on Trade and Development, Geneva. -(1996), Pension reform, the stock market, capital formation and economic growth: a critical commentary on the World Bank's proposals', International Social Securit Review, vol. 49, no. 3. Solnik, B.H. (1991), International Investments, Addison Wesley, Reading, Mass. -,Boucle, C and Le Fur, Y., 'Intemational market correlation and volatility', Financial-4na/ysts Journal, 9/10. Sunden, A.E. and Surette, B.J. (1998), 'Gender differences in the allocation of assets in retirement savings plans', American Economic Review, vol. 88, no. 2, pp. 207-211. Srinivas, P.S. and Yermo. J. (1998), 'Do Investment Regulations Compromise Pension Fund Performance - Evidence from Latin America', forthcoming in The Viewpoints Series of the Latin American and Caribbean Region, World Bank. Tirole, J. (1988) The Theogy of Industrial Orqganisation, MIT Press, Cambridge, Mass. United Kingdom, Department of Social Security (1998), 4 New Contractfor Welfare: Partnership in Pensions, London. United Kingdom, Her Majesty's Treasury (1999), 'Helping to deliver stakeholder pensions: flexibility in pension investment', London. Uthoff, A. W. (1997), 'Reformas a los sistemas de pensiones, mercado de capitales y ahorro', Revista de la CEPA4L, no 63, December. Vald6s-Prieto, S. (1997), 'Libertad de precios para las AFPs: aun insuficiente', Estudios Pziblicos, no. 68, Primavera, Santiago. -and Ramirez, M. (1999), 'Herding in DC pension funds: regulation or market forces?', rnimeo. 51 VanDerhei, J.L., Galer, R., Quick, C. and Rea, J. (1999), 401(k) Plan AssetAllocation, Account Balances, and Loan Activity, Employee Benefit Research Institute, Washington, D.C. Vittas, D. (1996), Tension funds and capital markets', Financial Sector Development Note no. -'1, World Bank. - (1998a), 'The role of non-bank financial intermediaries', Policy Research Working Paper no. 1892, World Bank. - (1998b), 'Regulatory controversies of private pension funds', Policy Research Working Paper no. 1893, World Bank. - (1998c), 'Investment rules and state guarantees for mandatory private pension funds', in Blommen,-tein, H. J. and Funke, N. (eds), Institutional Investors in the New Financial Landscape, OECD, Paris. Walker, E. (1991), 'Reflexiones en torno a politicas de inversi6n adecuadas para las AFP', Cuadernos de Economia, no. 85, December. - (1993a), 'Desempeiio financiero de las carteras de renta fija de los fondos de pensiones en Chile. ha tenido desventajas ser grandes?', Cuadernos de Economia, no. 89, April. -(1993b), 'Desempenio financiero de las carteras accionarias de los fondos de pensiones en Chile. ha tenido desventajas ser grandes?', Cuadernos de Economia, no. 89, April. World Bank (1994), Avertineg the Old Afge Crisis: Policies to Protect the Old and Promote Growth, Oxford Uriversity Press. Yaboboski, PJ. and VanDerhei, J.L. (1996), 'Worker investmnent decisions: an analysis of large 401 (k) plan data', Issue Brief no. 176, Employee Benefit Research Institute, Washington, D.C. Zheng, L. (1998), 'Is money smart? A survey of mutual fund investors' fund selection ability', Journal of Finance. 52 Social Protection Discussion Paper Series No. Title 0113 Regulating Private Pension Funds' Structure, Performance and Investments: Cross-country Evidence by P.S. Srinivas, Edward Whitehouse and Juan Yermo 0112 The World Bank and the Provision of Assistance to Redundant Workers: Experience with Enterprise Restructuring and Future Directions by Yi Chen 0111 Labor Markets in Transition Economies: Recent Developments and Future Challenges by Mansoora Rashid and Jan Rutkowski 0110 A Review of Social Investment Fund Operations Manuals by Juliana Weissman 0109 Risk and Vulnerability: The Forward Looking Role of Social Protection in a Globalizing World by Robert Holzmann 0108 Australia's Mandatory Retirement Saving Policy: A View from the New Millennium by Hazel Bateman and John Piggott 0107 Annuity Markets and Benefit Design in Multipillar Pension Schemes: Experience and Lessons from Four Latin American Countries by Robert Palacios and Rafael Rofman 0106 Guide for Task Teams on Procurement Procedures Used in Social Funds by Jorge A. Cavero Uriona 0105 Programmes Actifs Pour Le Marche Du Travail: Un Apervu General Des Evidences Resultant Des Evaluations by Zafiris Tzannatos and Amit Dar 0104 Kazakhstan: An Ambitious Pension Reform by Emily S. Andrews 0103 Long-term Consequences of an Innovative Redundancy-retraining Project: The Austrian Steel Foundation by Rudolf Winter-Ebmer 0102 Community Based Targeting Mechanisms for Social Safety Nets by Jonathan Conning and Michael Kevane Social Protection Discussion Paper Series continued No. Title 0101 Disability and Work in Poland by Tom Hoopengardner 0024 Do Market Wages Influence Child Labor and Child Schooling? by Jackline Wahba 0023 Including the Most Vulnerable: Social Funds and People with Disabilities by Pamela Dudzik and Dinah McLeod 0022 Promoting Good Local Governance through Social Funds and Decentralization by Andrew Parker and Rodrigo Serrano 0021 Creating Partnerships with Working Children and Youth by Per Miljeteig 0020 Contractual Savings or Stock Market Development. Which Leads? by Mario Catalan, Gregorio Impavido and Alberto R. Musalem 0019 Pension Reform and Public Information in Poland by Agnieszka Chlon 0018 Worker Reallocation During Estonia's Transition to Market: How Efficient and How Equitable? by Milan Vodopivec 0017 How Poor are the Old? A Survey of Evidence from 44 Countries by Edward Whitehouse 0016 Administrative Charges for Funded Pensions: An International Comparison and Assessment by Edward Whitehouse 0015 The Pension System in Argentina: Six years after the Reform by Rafael Rofman and Buenos Aires 0014 Pension Systems in East Asia and the Pacific: Challenges and Opportunities by Robert Holzmann, Ian W. Mac Arthur and Yvonne Sin 0013 Survey of Disability Projects. The Experience of SHIA, Swedish International Aid for Solidarity and Humanity by Kaj Nordquist Social Protection Discussion Paper Series continued No. Title 0012 The Swedish Pension Reform Model: Framework and Issues by Edward Palmer 0011 Ratcheting Labor Standards: Regulation for continuous Improvement in the Global Workplace by Charles Sabel, Dara O'Rourke and Archon Fung 0010 Can Investments in Emerging Markets Help to Solve the Aging problem? by Robert Holzmann 0009 International Patterns of Pension Provision by Robert Palacios and Montserrat Pallares-Miralles 0008 Regulation of Withdrawals in Individual Account Systems by Jan Walliser 0007 Disability Issues, Trends and Recommendations for the World Bank by Robert L. Metts 0006 Social Risk Management: A New Conceptual Framework for Social Protection and Beyond by Robert Holzmann and Steen Jorgensen 0005 Active Labor Market Programs: Policy Issues for East Asia by Gordon Betcherman, Amit Dar, Amy Luinstra, and Makoto Ogawa 0004 Pension Reform, Financial Literacy and Public Information: A Case Study of the United Kingdom by Edward Whitehouse 0003 Managing Public Pension Reserves Part I: Evidence from the International Experience by Augusto Iglesias and Robert J. Palacios 0002 Extending Coverage in Multi-Pillar Pension Systems: Constraints and Hypotheses, Preliminary Evidence and Future Research Agenda by Robert Holzmann, Truman Packard and Jose Cuesta 0001 Contribution pour une Strategie de Protection Sociale au Benin by Maurizia Tovo and Regina Bendokat Social Protection Discussion Paper Series continued No. Title * The papers below (No. 9801-9818 and 9901-9934) are no longer being printed, but are available for download from our website at wvw.worldbank.org/sp 9934 Helping the Poor Manage Risk Better: The Role of Social Funds by Steen Jorgensen and Julie Van Domelen 9933 Coordinating Poverty Alleviation Programs with Regional and Local Governments: The Experience of the Chilean Social Fund - FOSIS by Jorge C. Barrientos 9932 Poverty and Disability: A Survey of the Literature by Ann Elwan 9931 Uncertainty About Children's Survival and Fertility: A Test Using Indian Microdata by Vincenzo Atella and Furio Camillo Rosati 9930 Beneficiary Assessment of Social Funds by Lawrence F. Salmen 9929 Improving the Regulation and Supervision of Pension Funds: Are there Lessons from the Banking Sector? by Roberto Rocha, Richard Hinz, and Joaquin Gutierrez 9928 Notional Accounts as a Pension Reform Strategy: An Evaluation By Richard Disney 9927 Parametric Reforms to Pay-As-You-Go Pension Systems by Sheetal K. Chand and Albert Jaeger 9926 An Asset-Based Approach to Social Risk Management: A Conceptual Framework by Paul Siegel and Jeffrey Alwang 9925 Migration from the Russian North During the Transition Period by Timothy Heleniak 9924 Pension Plans and Retirement Incentives by Richard Disney and Edward Whitehouse 9923 Shaping Pension Reform in Poland: Security Through Diversity by Agnieszka Chlon, Marek G6ra and Michal Rutkowski Social Protection Discussion Paper Series continued No. Title 9922 Latvian Pension Reform by Louise Fox and Edward Palmer 9921 OECD Public Pension Programmes in Crisis: An Evaluation of the Reform Options by Richard Disney 9920 A Social Protection Strategy for Togo by Regina Bendokat and Amit Dar 9919 The Pension System in Singapore by Mukul G. Asher 9918 Labor Markets and Poverty in Bulgaria by Jan J. Rutkowski 9917 Taking Stock of Pension Reforms Around the World by Anita M. Schwarz and Asli Demirguc-Kunt 9916 Child Labor and Schooling in Africa: A Comparative Study by Sudharshan Canagarajah and Helena Skyt Nielsen 9915 Evaluating the Impact of Active Labor Programs: Results of Cross Country Studies in Europe and Central Asia by David H. Fretwell, Jacob Benus, and Christopher J. O'Leary 9914 Safety Nets in Transition Economies: Toward a Reform Strategy by Emily S. Andrews and Dena Ringold 9913 Public Service Employment: A Review of Programs in Selected OECD Countries and Transition Economies by Sandra Wilson and David Fretwell 9912 The Role of NPOs in Policies to Combat Social Exclusion by Christoph Badelt 9911 Unemployment and Unemployment Protection in Three Groups of Countries by Wayne Vroman 9910 The Tax Treatment of Funded Pensions by Edward Whitehouse Social Protection Discussion Paper Series continued No. Title 9909 Russia's Social Protection Malaise: Key Reform Priorities as a Response to the Present Crisis edited by Michal Rutkowski 9908 Causalities Between Social Capital and Social Funds by Jesper Kammersgaard 9907 Collecting and Transferring Pension Contributions by Rafael Rofman and Gustavo Demarco 9906 Optimal Unemployment Insurance: A Guide to the Literature by Edi Karni 9905 The Effects of Legislative Change on Female Labour Supply: Marriage and Divorce, Child and Spousal Support, Property Division and Pension Splitting by Antony Dnes 9904 Social Protection as Social Risk Management: Conceptual Underpinnings for the Social Protection Sector Strategy Paper by Robert Holzmann and Steen Jorgensen 9903 A Bundle of Joy or an Expensive Luxury: A Comparative Analysis of the Economic Environment for Family Formation in Western Europe by Pierella Paci 9902 World Bank Lending for Labor Markets: 1991 to 1998 by Amit Dar and Zafiris Tzannatos 9901 Active Labor Market Programs: A Review of the Evidence from Evaluations by Amit Dar and Zafiris Tzannatos 9818 Child Labor and School Enrollment in Thailand in the 1990s By Zafiris Tzannatos 9817 Supervising Mandatory Funded Pension Systems: Issues and Challenges by Gustavo Demarco and Rafael Rofman 9816 Getting an Earful: A Review of Beneficiary Assessments of Social Funds by Daniel Owen and Julie Van Domelen 981-5 This paper has been revised, see Discussion Paper No. 9923 Social Protection Discussion Paper Series continued No. Title 9814 Family Allowances by Suzanne Roddis and Zafiris Tzannatos 9813 Unemployment Benefits by Zafiris Tzannatos and Suzanne Roddis 9812 The Role of Choice in the Transition to a Funded Pension System by Robert Palacios and Edward Whitehouse 9811 An Alternative Technical Education System: A Case Study of Mexico by Kye Woo Lee 9810 Pension Reform in Britain by Edward Whitehouse 9809 Financing the Transition to Multipillar by Robert Holzmann 9808 Women and Labor Market Changes in the Global Economy: Growth Helps, Inequalities Hurt and Public Policy Matters by Zafiris Tzannatos 9807 The World Bank Approach to Pension Reform by Robert Holzmann 9806 Government Guarantees on Pension Fund Returns by George Pennacchi 9805 The Hungarian Pension System in Transition by Robert Palacios and Roberto Rocha 9804 Risks in Pensions and Annuities: Efficient Designs by Salvador Valdes-Prieto 9803 Building an Environment for Pension Reform in Developing Countries by Olivia S. Mitchell 9802 Export Processing Zones: A Review in Need of Update by Takayoshi Kusago and Zafiris Tzannatos 9801 World Bank Lending for Labor Markets: 1991 to 1996 by Amit Dar and Zafiris Tzannatos Summary Findings Because defined-contribution systems expose pensions to a number of risks, reforming governments have often strictly regulated the pension fund industry's structure, performance, and investments. This paper compares the rules in the new systems of Latin America and eastern Europe with richer OECD countries. The authors argue that the henefits of competing pension funds and individual choice can only be achieved if regulations are loosenecd in the mdclium term. HUMAN DEVELOPMENT NETWORK About this series... The World Bank Pension Reform Primer aims to provide a timely and comprehensive resource for those engaged in the design and implementation of pension reforms around the world. Policymakers and those who advise them will find useful information on other reform experiences, the current thinking of pension specialists and a vast array of cross-country evidence. A flexible and dynamic format ensure that key developments are updated as they occur. The World Bank set out a conceptual framework for fundamental pension reform in Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth. This study, published in 1994, helped shape the global debate about the impact of population ageing on pension systems. The Pension Reform Primer builds on this pioneering work and on the experience of the World Bank and other international institutions in the last five years. It focuses on practical questions. For more information, please contact Social Protection, Human Development Network, World Bank, 1818 H Street NW, Washington, D.C. 20433; telephone +1202 458 5267; fax +1 202 614 0471; e-mail socialprotection@worldbank.org. All Pension Reform Primer material is available on the internet at www.worldbank.orgipensions