33385 World Bank Pension Reform Primer Public management Part I: How well do governments invest pension reserves? Public pension fund managers around the world largest in terms of membership is the Employees' control retirement savings worth more than Provident Fund (EPF) in India with more than 20 three trillion US dollars. Their performance directly million members while the Central Provident Fund affects millions of workers forced to contribute to (CPF) of Singapore is the best known, with only one these schemes. The indirect impact is much million contributors. Both prescribe yields on broader; if long-term savings are channelled individual accounts and invest mostly in government effectively, good projects are easier to finance and bonds. this leads to higher economic growth. Poor management of these savings not only reduces Public versus private management future pensions, it also misallocates capital. The The government influences the investment policies stakes are especially high for developing countries of all pension schemes. Private, voluntary pension where capital is scarce and workers depend on funds rely on favorable tax treatment and are often pensions to keep them out of poverty. subject to certain restrictions on investments and withdrawals. Private managers in mandatory Public pension schemes with some degree of schemes sometimes face intrusive regulators and prefunding can be grouped into two categories. The often must obey strict limits on the type of assets most common form is the partially funded, defined that they can purchase. Taken to an extreme, benefit (DB) scheme. These are often found in portfolio limits can severely limit competition and younger countries where pension schemes are still may even be manipulated to force private funds to immature, as in Francophone Africa, some parts of be invested exactly as the government wishes. the Middle East and East Asia. However, a few older countries such as Sweden and Japan also fall At the same time, a centrally run government into this category, having chosen to accumulate monopoly could use private managers and apply surpluses and offset intergenerational transfers. The market based criteria to their selection and extent to which these schemes are actually funded compensation. In theory, a centrally managed fund varies across countries and over time. However, to that incorporates market-based criteria could face a our knowledge, a fully funded, defined benefit plan less restrictive environment than a privately- run by a national government does not exist. managed scheme subject to onerous regulations. In practice however, regulatory constraints are rarely so A second type is the centrally-managed, defined extreme as to eliminate the discretion of private contribution arrangement or provident fund. These managers while most publicly-managed funds are are mostly found in the former British colonies of subject to restrictions and mandates. Africa and Asia. About a dozen countries continue to use this model and the number is shrinking. The This briefing 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. 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 2 Public management Restrictions and mandates In short, public pension funds do not hold the type of portfolio that would be recommended by private The investment policies of public pension fund pension fund managers interested in maximizing risk monopolies are determined by boards that include adjusted returns. Instead, their investments reflect representatives from government, unions and public policy objectives other than pension employers. These tripartite boards have limited provision. It should be noted of course, that discretion however, due to restrictions and mandates governments may also interfere in the investment that force them to invest in certain ways. For choices of private pension funds. But the impact is example, Japan, Korea and the United States are typically much more benign since private managers among the countries that force their public pension are accountable for their performance. funds to loan surpluses directly to the government - outside the normal channels of government Limited investment options at home borrowing. Illiquid or non-existent bond and stock markets can More common are mandates determined by be another impediment to public (and private) politicians under the general heading of pension fund managers. This is especially true in `development policy'. These include social developing countries where the public fund may be investments like housing bonds in Sweden or the single largest investor in the country. For economically targeted investments (ETIs) like the example, Sri Lanka's Employee Provident Fund purchase of state enterprise bonds. Since 1987 for (EPF) had assets that were roughly equivalent to the example, Iran's pension reserves have been used size of the stock market in the late 1990s. Alone, its among other things to provide "financial facilities to investment decisions could move the markets. industrial units of the country, to buy raw materials Meanwhile, a large equity position would make it the and machinery and improving business activities". biggest shareholder in the country. And in Japan, the pension fund has invested 17 percent of its massive reserves in health, recreation In some countries, the rule of law, definition of and other facilities under the heading "welfare". property rights and basic financial infrastructure do not meet even minimal standards. The riskless But aside from government bonds, housing was the benchmark does not exist since the government may most popular investment for pension funds. In a default on its bonds or fail to ensure the solvency of recent study of a dozen Anglophone African the banking system. Unfortunately, the case of countries, the International Social Security Cameroon, described in the box below, is not Association reports that the public pension funds in unique. Gambia, Ghana, Kenya, Mauritius, Nigeria, Swaziland, Tanzania, Uganda and Zambia all played While underdeveloped local capital markets are a role in housing finance and/or construction. In clearly a problem for any type of funding, this does Tanzania and Zambia for example, the fund itself not appear to explain the kind of investment constructs housing which is then rented to patterns described above. But even in countries individuals. where capital markets do provide potential investment outlets for large public pension funds, at The result of these practices is a pattern of least three problems arise: First, ownership of a investments that reflects the priorities of the large proportion of the shares by the government government but that are not in line with the would at the very least, raise important questions objectives of prefunding pensions. Most public regarding corporate governance. Second, the pension funds invest heavily in government bonds government may be tempted to use pension funds to and bank deposits. Many make social investments support the stock market or specific firms with or make loans to state-owned firms and even political influence. Finally, the government may find individuals. Very little is invested in shares and itself in the awkward position of regulator and foreign assets are almost never part of the portfolio. owner of certain industries, creating other possible of conflicts of interest. World Bank Pension Reform Primer Cameroon ­ no safe haven for public pension funds In the 1980s, the public pension fund in Cameroon (CNPS) had a difficult time finding safe places to invest its growing funds. There were few private securities available domestically. It was forced to purchase medium term, low interest bonds from the National Investment Company. Bank deposits were the second largest item in the portfolio but these deposits turned out to be quite risky. The accounts were frozen when several banks were closed in the 1980s. Only the small part of the portfolio in real estate maintained its real value over time. The largest part of the portfolio was held in the form of government bonds. Naturally, it would be safe there. Or would it? In the mid-1980s, the government of Cameroon covered part of its burgeoning deficit by borrowing large sums from the CNPS. More loans were mandated to state enterprises like the Cameroon Sugar Corporation and the Cameroon Banana Organization. Social housing was financed through loans to the Cameroon Housing Company. These loans were made upon the instruction of the President of the Republic or the Minister of Finance while the CNPS management was restricted to carrying out instructions. The duration of the loans was 10 or 20 years and the interest rates were often lower than those given on bank deposits. Some state companies were unable to repay the loans while others simply chose not to repay. One official expressed his dismay, "The managers of the CNPS have no real means of exerting pressure on the government since the Fund is a public establishment... This situation justifies the fears of those who worry that the temptation of the abundant reserves accumulated by social security schemes will be too strong for public authorities, which will draw on them for more dubious purposes." Frustrated managers at the CNPS even advocated depositing some of the pension fund money abroad but this was rejected by the government. In the end, there seems to be no safe haven for the savings of Cameroon's workers. Investment returns and volatility Perhaps the most important single indicator of how defined benefit schemes. Countries with data for public pension reserves are being managed is the rate at least eight consecutive years were included in of return. Unfortunately, data are not readily the sample available for most countries and where they are, may not be strictly comparable. For example, valuation Figure 1 below plots the annual compounded rates methods vary, with most countries using book values of return and standard deviations for publicly rather than marking to market. In other countries, managed schemes in a sample of 22 countries and returns are simply not published. The dearth of good the Chilean private pension sector. Averages are information in this area is itself symptomatic of the based on at least 8 and as much as 35 years of lack of accountability and transparency that returns. Half of the sample or 10 countries failed characterize many of the schemes. to generate positive real returns and most were between 1-2 percent. Only members of provident With these caveats in mind, this subsection looks at funds in Malaysia and the partially funded, defined two types of returns for a small sample of countries. benefit scheme in Korea earned real returns in The first type is the return credited to the accounts of excess of three percent. The Korean performance provident fund members. The second refers to the was the best at 5.4 percent per annum. The worst reported return on investments of partially funded, returns were found in Uganda (-50.5%), Peru (-46.6%), and Zambia (-30.5%). 4 Public management Annual returns and volatility 1 basis points, respectively. This is less impressive however, if we note that returns are about the same as on one year, government bonds. Most of 30% the sample posted returns well below the short- term interest rate. In the worst cases, the funds 25% would have earned 400-1000 basis points more per oni annum by keeping all of their money in bank at 20% vi deposits. Worse yet, workers in these countries de d 15% were not compensated for these poor returns by lower risk; the volatility of bank deposit interest ndara 10% Chile St rates was the same as or lower than the volatility of 5% pension fund returns. 0% Figure 2 suggests that returns in most of our -55% -45% -35% -25% -15% -5% 5% 15% sample are not even as high as what could be Compounded real return obtained in banks by individual savers. But pension funds handle long term savings that The striking feature of Figure 1 is the inverse should be able to earn higher returns in exchange correlation between risk and returns. Normally, more for lower liquidity. Rather than the riskless, short volatility would be associated with higher long run term rate of return, they should be able to achieve rates of return, but the opposite seems true here. something closer to the long run return on capital. Figure 1 is difficult to interpret. Unless public Returns minus bank deposit rates 2 pension fund managers had full access to an globally diversified portfolio, each will face very different investment opportunities. They are further Average -1.8% constrained by mandates and restrictions as we have Uganda seen above. Finally, country-specific crises during the Zambia periods covered here could skew the results. In Venezuela short, while the data confirm a pattern of low or Egypt negative returns, it does not reveal whether the Ecuador returns experienced were reasonable given prevailing Sri Lanka conditions. Guatemala Kenya In order to get a clearer picture, we can normalize Jamaica returns relative to bank deposit rates in the same Canada countries during the same periods. This should help Singapore control for some of the country-specific factors and Morocco provide a crude benchmark for the available Costa Rica investment alternatives. The results of this exercise India are shown in Figure 2. Malaysia US Sweden There is significant variation across the sample. Philippines Using a simple average, returns are 180 basis points Korea lower than bank deposit rates. However, four Japan countries - Sweden, Philippines, Korea and Japan - show returns that are more than 100 basis points -12% -10% -8% -6% -4% -2% 0% 2% 4% higher than deposit rates. Japan and Korea stand out gross returns minus bank deposit rate with returns exceeding deposit rates by 300 and 250 World Bank Pension Reform Primer In assessing the gap between actual and potential Returns minus income growth returns we can look to the growth of incomes as a 3 rough benchmark. In a dynamically efficient economy, returns to capital are higher than the growth in incomes over long time horizons. Thus, Average Peru the returns available to publicly-managed pension Uganda funds from a diversified portfolio should be greater Zambia than income growth. Maintaining this relationship is Venezuela also important if the pension scheme is to generate Egypt reasonable replacement rates. Tanzania Ecuador Costa Rica Figure 3 shows that only two publicly-managed Guatemala pension schemes earned returns that were greater Kenya than the growth of income per capita during the Singapore Sri Lanka periods in question. Only in the Philippines and Jamaica Morocco were returns greater than income growth Korea and then only by a small margin. In most countries, Japan returns trailed income growth. In half the sample, the India differential in favor of income growth was greater Canada Malaysia than 300 basis points. Sweden US The results for Singapore and Malaysia merit further Morocco discussion. Both provident funds exhibit low but Philippines positive real rates of returns for their members. -50% -40% -30% -20% -10% 0% 10% However, incomes in both countries were among the gross returns minus income per capita fastest growing in the world over the last three decades. Another similarity is that prescribed yields to members have been based on short-term interest Private pension fund returns in contrast, almost rates while actual investment returns are believed to always exceed the growth of incomes over the long have been much higher. In other words, part of the run. Figure 4 below shows that this difference is effect being observed here is an implicit taxation of usually at least 200 basis points. And although the returns rather than poor investment performance. time periods considered here differ, public and private pension returns over long periods of time From the worker's perspective, the result of this can also be compared for a few countries. For implicit taxation is a low replacement rate. Assuming example, privately-managed funds in Sweden and that average wages and income per capita moved in Japan earned returns that were respectively, 300 tandem in both countries, a worker who began his and 500 basis points higher than their public sector career in 1960 would have seen his wage grow counterparts. eightfold in Singapore and almost tenfold in Malaysia by 1995. Meanwhile, the contribution he made in What explains poor performance? 1960 would only have doubled in Singapore and The direct causes for underperformance have tripled in Malaysia. Not surprisingly, criticism of already been mentioned: They include government these schemes has increased as workers about to interference in investment ranging from the retire after contributing their entire working lives find imposition of social or development objectives on that low balances are insufficient to maintain pre- the pension fund to forcing pension funds to retirement living standards. finance deficits or state enterprise losses often at interest rates lower than what is available on the market. The common prohibition on investment This note was prepared by Robert Palacios of the World Bank. It is based on a longer World Bank social protection discussion paper. Full details of the pension reform primer series are available from the social protection advisory service, World Bank, 1818 H St NW, Washington DC 20433 and on the internet at www.worldbank.org/pension. 6 Public management Private managers do better 4 shown that these decisions are largely determined by the mandates and restrictions imposed on public pension fund managers. Asset allocation decisions are largely political and have little to do Average private schemes with any application of portfolio theory. Switzerland (70-90) Japan (70-87) In short, the problem is that investment policy is United States (70-90) driven by political motives. This means that Canada (75-89) performance differentials observed across Denmark (70-88) countries are most likely correlated with lack of Hong Kong (83-96) transparency that allows the fund to be used for Netherlands (70-90) non-pension purposes. Other potential causes of Japan (84-93) underperformance would also reflect the quality of Switzerland (84-96) governance: Some governments may be less Denmark (84-96) dependent on cheap borrowing from pension Australia (87-94) reserves while others may be more resistant to United Kingdom (70-90) lobbies for social investments. Some may be even Spain (84-93) be more susceptible to corruption. Netherlands (84-96) Ireland (82-97) Recently, good government has been found to be Chile (81-96) associated with a wide array of positive Belgium (84-96) development outcomes across many countries. United States (84-96) Sweden (84-93) Conversely, the countries that rank poorly in terms United Kingdom (84-96) of bureaucratic efficiency and corruption are found to have worse living standards. -1% 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% Based on the evidence, one study tested the Gross returns minus income per capita growth following hypotheses: abroad posed another major challenge to public · the level of governance in a country affects pension fund managers trying to diversify their (often performance across countries significant) country-specific risk. While we noted that and that, thin and/or badly regulated capital markets might not · there is a general "public management effect" provide the investment opportunities required by that itself reduce returns large funds, this constraint did not seem to be binding in most countries. It certainly could not explain why These hypotheses were tested in a multivariate returns were lower than bank deposit rates. regression analysis using gross returns minus bank deposit rates as the dependent variable. An index We also noted that normalized returns vary widely of governance and private management were used across publicly managed pension funds. This suggests as explanatory variables. The governance measure that additional factors are either exacerbating or is an average of three indices covering surveyed mitigating the general deficiencies that affect public perceptions of (i) the efficiency of the judiciary management. The obvious question is why do some system, (ii) the amount of "red tape" and (iii) countries perform better than others do? corruption. The specific proposition here is that poor governance leads to conditions - corruption, Explaining private pension fund performance has politicized investments, the need for captive focused on strategic asset allocation. These decisions financing for deficits etc. that lead to poor returns. have been found to explain more than 90 percent of the variation in risk adjusted returns across private The sample used included 16 country observations pension funds. In public pension funds, we have of long term, gross returns minus bank deposit rates from Figure 2 above. The governance index World Bank Pension Reform Primer was not available for the other countries. The sample governance index. In other words, the "public also included observations of the same indicator from management effect" is strongly negative. 14 countries for which private pension returns were available. Observations of both public and private Figure 5 also shows that privately-managed management were available for Canada, the United schemes are likely to produce reasonable returns in States, Sweden and Japan. countries with medium or high governance ratings. In countries rated 7 or better, private returns What were the results? Better country governance exceeded bank deposit rates by more than 400 rankings were associated with higher returns while basis points. On the other hand, countries with public management lowered them. The extent of the worst governance ratings, say below 4, would these two effects can be seen in Figure 5 which plots not produce reasonable returns even with private the fitted lines for publicly and privately managed management. The difference between returns in a returns against different values of the governance privately-managed scheme in a country with good index. The effect of governance is strong: on a scale government and a publicly-managed scheme in a of 1-10 with 10 being the best governance, an poorly governed country would be on the order of increase from 4-5 to 8-10 increases returns relative to 10 percentage points. bank deposits by 5-7 percentage points. Most of the gains come at the lower part of this range and begin The policy implications are significant: First, unless to disappear after a country achieves a governance public management can be improved, it is not rating of around 8. likely to be effective in achieving the objectives of either a partially funded, DB or a funded DC Governance and pension returns 5 scheme. Second, poor performance of publicly- managed schemes is likely to signal misallocation of savings with important macroeconomic 10% implications if the fund itself is large. Third, good governance is important for private management. Finally, poorly governed countries should probably sti 5% avoid funding their mandatory pension systems osped altogether. 0% nk These findings are preliminary and subject to ba e -5% several caveats. In particular, the data set used ovba suffers from certain problems of quality and sn -10% comparability. Valuation methods vary and urteR administrative charges to the fund are ignored. Publicly-managed -15% Also, the sample is quite small despite the fact that Privately-managed the 30 country observations represent more than 400 country/year data points for each variable. -20% And unfortunately, time series data on governance 0 1 2 3 4 5 6 7 8 9 10 is not readily available. The analysis implicitly Governanceindex(Best=10) assumes that relative governance capacity has not changed over the time periods analyzed. The difference between the parallel lines in Figure 5 The results may also be skewed by the fact that the are due to the public management effect. Private privately-managed schemes in our sample are from management produces returns that are about 430 countries with better governance. Finally, it basis points higher than publicly-managed schemes should also be noted that the privately-managed after taking into account differences in the returns are sectoral averages hiding potentially significant variation across funds within a country. This note was prepared by Robert Palacios of the World Bank. It is based on a longer World Bank social protection discussion paper. Full details of the pension reform primer series are available from the social protection advisory service, World Bank, 1818 H St NW, Washington DC 20433 and on the internet at www.worldbank.org/pension. 8 Public management Nevertheless, the preliminary findings of the analysis The box on Cameroon is based on: are quite robust and are consistent with a large body Mounbaga, E. (1995), `The investment of social of anecdotal evidence. security reserves during periods of crisis: The experience of Cameroon', International Social The most visible consequence for the members of the Security Review, Vol. 48, 2/95. fund is the low rate of return. For partially funded DB schemes, low returns hasten the day when benefits must be cut or contributions raised. This has already happened in many developing countries where deficits have arisen or contribution rate Conclusions and recommendations increases have been forced after just one or two publicly-managed pension reserves are decades of operation, largely due to poor investment often used to finance non-pension policy returns. In other words, poor investment policy leads public pension fund managers tend to to a lower internal rate of return to participation in invest based on objectives unrelated to the scheme. pension provision these include social and economically The members of a publicly-run DC scheme feel the target investments such as housing effects much more directly. A worker that but often governments look to pension contributed 100 shillings to the Kenyan provident reserves as a convenient and cheap way fund in 1978 would have retired in 1990 with about to finance deficits 60 shillings in real terms. And even where the real one result is that public management returns are positive, the fact that they do not keep up produces poor returns relative to what with the growth of incomes means that their could potentially be earned consumption smoothing function is compromised. good governments perform better but public management produces inferior returns across all countries Further reading as a result members of the scheme have Asher, M. (1999), `The Pension System in Singapore', to pay higher contributions or receive Pension Reform Primer Series, Social Protection lower benefits Discussion Paper no. 9919, World Bank. the evidence suggests that public Angelis, T. (1998), `Investing Public Money in management of pension reserves should Private Markets: What are the Right Questions?', generally be avoided paper presented at the 1998 National Academy of any prefunding of long term pension Social Insurance Conference. obligations requires some minimal level Iglesias A. and Palacios (2000), `Managing public of good governance pension reserves: Evidence from the international experience', Pension Reform Primer Series, Social Protection Discussion Paper no. 0003 , World Bank. Munnell, A. and C.N. Ernsberger. (1989), `Public Pension Surpluses and National Saving: Foreign Experience', New England Economic Review, March/April 1989. Weaver, C. ed. (1990), "Social Security's Looming Surpluses: Prospects and Implications", American Enterprise Institute, Washington D.C..