funding unfunded pensions: governance and investments of ......reserve funds in japan, south korea,...

No. 1|2008

International Pension Papers

Funding Unfunded Pensions: Governance and Investments of Asian Reserve Funds

2

No. 1|2008Allianz Global Investors International Pension Papers

Imprint

Publisher: Allianz Global Investors AG, International Pensions, Nymphenburgerstr. 112-116, D-80636 Munich, [email protected],

http://www.allianzglobalinvestors.com | Author: Dr. Alexander Börsch, Senior Pensions Analyst, Allianz Global Investors AG, [email protected]

Layout: volk:art51 GmbH, Munich | Printing: repromüller, Übersee | Closing Date: March 5, 2008

The entire content of this publication is protected by copyright with all rights reserved to Allianz Global Investors AG. Any copying, modifying, distributing

or other use of the content for any purpose without the prior written consent of Allianz Global Investors AG is prohibited. The information contained in this

publication has been carefully verified by the time of release, however Allianz Global Investors AG does not warrant the accuracy, reliability or complete-

ness of any information contained in this publication. Neither Allianz Global Investors AG nor its employees and deputies will take legal responsibility for

any errors or omissions therein.

This publication is intended for general information purposes only. None of the information should be interpreted as a solicitation, offer or recommen-

dation of any kind. Certain of the statements contained herein may be statements of future expectations and involve known and unknown risks and un-

certainties which may cause actual results, performance or events to differ materially from those expressed or implied in such statements.

3

No. 1|2008Allianz Global Investors International Pension PapersAllianz Global Investors International Pension Papers

Critical IssuesMany countries around the world have introduced pension reserve funds to fund parts of future liabilities in order to cope with the adverse effects of population ageing on public pension systems.

The pension reserve funds in Asia have, and will continue to accumulatehuge amounts of assets. This paper investigates the four biggest reserve fundsin Asia, which together have aggregate assets of EUR 766 billion under manage-ment, putting them among the biggest pension funds worldwide.

While pension reserve funds traditionally invested conservatively and weresubject to political control, there is substantial change in the governance struc-tures and investment policies of the Asian funds. Three key trends can beobserved:

– Pronounced return orientation in the management of these funds has taken root, resulting in a retreat from public project financing

– Governance structures of Asian reserve funds are becoming more profes-sional through board composition reforms and increasing independence of the reserve funds from government

– Outsourcing to private asset managers has risen considerably with the main motivation and consequence being an ongoing diversification ofinvestments

The size of the Asian reserve funds means their investment strategy has substantial effects on the sustainability of national pension systems, the domestic economy and international financial markets.

Funding Unfunded Pensions: Governance and Investments of Asian Reserve Funds

4


ContentIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

The Rationale for Pension Reserve Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

The Financing of Pension Reserve Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6Traditional Management of Reserve Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

Trends in Asian Reserve Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

Country Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

Japan: the Government Pension Investment Fund . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13South Korea: the National Pension Fund . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14Australia: the Future Fund . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17China: the National Social Security Fund . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

5


1 Pension reserve funds

should not be confused

with sovereign-wealth

funds. The capital of

sovereign-wealth funds

comes from current

account surpluses and re-

sulting foreign exchange

reserves, which are then

reinvested.

There is an increasing trend in manycountries around the world to build an

additional element into their pension sys-tems, namely pension reserve funds (alsoknown as sovereign pension or demographyfunds). The main aim is to partially fund theliabilities of existing pay-as-you-go systemsto cushion the future impact ageing societieswill have on unfunded public pension sys-tems. In this way, these funds can contributeto a higher sustainability of public pensionsand avoid skyrocketing pension contribu-tions for future employees. Pension reservefunds can differ in many dimensions, butthey have an important influence on globalfinancial markets and national economies.Their importance is illustrated by the factthat three of the world’s four biggest pensionfunds are pension reserve funds.1

This paper focuses on the Asian pensionreserve funds in Japan, South Korea, Aus-tralia, and China, the four biggest reservefunds in the region. The Japanese and SouthKorean funds are among the biggest world-wide and are quite mature, while their Aus-tralian and Chinese counterparts have onlybeen recently established. Despite differ-

ences in maturity, all four funds show strik-ing similarities in terms of governancestructures and investment policies. The keytrends are a:

• Substantial increase in the return orienta-tion of these funds, coupled with a retreatfrom public project financing

• Reform of board structures and composi-tions with a view to professionalise man-agement, accompanied by an increasingindependence of reserve funds from gov-ernment

• Considerable rise of outsourcing to pri-vate asset managers, with the main moti-vation being to diversify investments

This paper first deals with the reasons whycountries introduced pension reserve funds,then classifies the pension reserve fundsand explains the traditional ways of manag-ing these reserves before analysing recenttrends in governance and investing. The lastpart describes the design, governance andinvestment policies of pension reserve fundsin Japan, South Korea, Australia and China.

Introduction

6


The impact of ageing on public pay-as-you-go systems is one of the main mo-

tivations for establishing pension reservefunds. In pay-as-you-go systems, the firstpension pillar in most countries of the world,pension contributions of current employeesare used to finance pension benefits of cur-rent retirees. Ageing populations imply thatthere will be fewer employees in the future,but more retirees. This means that, if the liv-ing standard of retirees is to be preserved,contributions to the pension system mustrise substantially. If contributions remainconstant, then pension payments to retireesmust decrease. It has to be kept in mind thatpension payments are already one of thebiggest items in national budgets and thatthey will grow substantially. The scope of theproblem can be illustrated by the concept ofimplicit pension debt, a measure for a coun-try’s future pension obligations or pensionpromises. This estimate aims to convert fu-ture streams of benefits into a current debtequivalent. For example, World Bank calcu-lations show that China’s implicit pension

debt amounted to 141% of GDP in 2001 for a projected period of 75 years (Sin 2005).

There are several reform options, such as providing greater weight to occupationaland private pensions in old-age provisionand cutting back public pension benefits. A measure that directly addresses the prob-lems of pay-as-you-go systems is the estab-lishment of pension reserve funds. Pensionreserve funds pre-fund a part of future pen-sion payments of the pay-as-you-go systems.This supports the working of the pay-as-you-go systems by smoothing the future costs ofan older generation by limiting otherwisenecessary future contribution hikes. Put dif-ferently, at the time when the ratio betweenemployees and pensioners becomes unfav-ourable, the pension system can draw onaccumulated capital to pay benefits insteadof being purely dependent on contributions.In this sense, pension reserve funds act as abuffer against the adverse impact of ageingsocieties on public pensions.

This buffer function is central to dedicatedpension reserve funds. Another type of re-serve funds are those used for precautionarypurposes, where some reserves are main-tained to bolster the impact of the businesscycle on the scheme balance. However, thesefunds only build up minor reserves. They arenot designed for long-term purposes and donot accumulate long-term capital to disbur-den the pay-as-you-go system.

The main difference between long-term-oriented pension reserve funds relates totheir funding. Pension reserve funds can beestablished intentionally as a means to copewith demographic change. This is the case

for the Chinese Social Security Fund and the Norwegian Government Pension Fund.These dedicated funds, which could becalled strategic reserve funds, can be fi-nanced from various sources external to the pension system, such as privatisationproceeds, general budget surpluses or oilrevenues. Hence, their main characteristic is that governments take countermeasuresagainst foreseeable financing problems ofpublic pension systems by earmarkingfunds from other sources to partially fundthe pension systems in place.

A second type of long-term reserve fundemerges as a by-product of the operation of

The Rationale for PensionReserve Funds

The Financing of Pension Reserve Funds

7


the pension system in place. When contribu-tions to the pay-as-you-go system are higherthan the payouts, the surplus can be used tosafeguard the pension system in the future.This is often the case in relatively young pay-as-you-go systems, like in South Korea, butalso in established systems like Japan. TheSouth Korean pay-as-you-go system was in-troduced in 1988, but the minimum member-ship period for pension benefits is 20 years,so that the system could accumulate all thecontributions over this period and investthem in the National Pension Fund.

Both types of reserve funds have the sameeffect, namely to fund a part of future liabili-

ties. The only difference is that capital accu-mulated from the operation of the pensionsystem itself can be considered a part of thesocial security system, while in the case offiscal transfers the managing institutions areseparated from the social security system.Another difference is that reserve funds, fedby fiscal transfers, are often not allowed tomake payouts in the near or medium future(OECD 2007a). However, the distinction basedon belonging to the social security systembecomes blurred as reserve funds funded bycontributions are increasingly establishedas independent entities, a topic analysedlater. (Table 1)

Pension reserve funds have often been criti-cised for their poor investment performance.This performance is perceived to stem frompolitical influence on investment policies(Vittas, Impavido, O’Connor 2008). There isone crucial difference between pension re-serve funds and other pension funds, namelythat the government or a government-affili-ated managing institution has control overthe capital (Yermo 2008). This naturallymakes it tempting for governments to utilise

the assets for political or other ends not di-rectly related to financial performance.

As a result, the key issue regarding invest-ments of pension reserve funds is the invest-ment philosophy and the question: Shouldpension reserve funds be focused solely onmaximising returns for the beneficiaries orshould they promote “public” purposes aswell? Traditionally, the pursuit of public pur-poses was stressed, often at the expense of

Surplus from contributionsover payouts

Short-term impact

Long-term impact Pension Reserve Fund (Japan, South Korea)

Strategic Pension Reserve Fund(Australia, China)

Precautionary FundPrecautionary Fund

Surplus from fiscal transfers

Table 1 Types of pension reserve funds

Traditional Management of Reserve Funds

8


returns. Many pension reserve funds, as wellas other public pension funds, have beenforced to invest in low-yielding governmentprojects. For example, as in Japan and SouthKorea, in housing loans or infrastructure, orgenerally in “economically-targeted invest-ments” in the interest of economic develop-ment. There is also a danger that pensionreserve funds loan to government at belowmarket rates or have to confine investmentsto the domestic markets, which generatesexcessive country-specific risk (Weaver 2003).Besides, pension reserves can be used as away of deficit financing for government, ifdomestic bond investments dominate. As aresult, the general suspicion is that publicreserve funds generate below-market returnsbecause of political influence, which preventsthem from maximising long-term returns.

However, financial market investments ofreserve funds are complex. The main reasonis that due to their sheer size, they can movethe markets. If public reserve funds directly in-vest assets to a substantial degree in domes-tic equities, they would significantly influenceshare prices. They would also become impor-tant company owners. In this way, the govern-ment would become deeply involved in thecorporate governance of those firms and con-flicts of interest would arise. The fact that re-serve funds are government or government-affiliated institutions also complicates directforeign investments. If the reserve funds in-vested directly in foreign firms, a politicalbacklash in the respective countries would belikely, as has been recently the case with theinvestments of sovereign-wealth funds fromemerging economies in Western companiesin the wake of the subprime crisis.

9


At the same time, demographic develop-ments require investments in higher-

yielding assets. Given the unfavourable pro-portion of contributors and beneficiaries inthe future, the capital of reserve funds is animportant factor to smooth contributions.The more efficient the risk-return profile, themore disburdened the public pension sys-tems. This implies diversification of assetsbetween asset classes and internationally.

A way to do so and to circumvent theabove-mentioned investment issues is tooblige the managing institution to adhere tothe goal of return maximisation by outlaw-ing other goals, or by outsourcing of assets orby introducing greater distance from govern-ment interference. These measures supporta de-politicisation of investment policies,which eases equity investments and assetdiversification. In addition, these measuresimply increased professionalisation of assetmanagement, which helps the goal of returnmaximisation to be realised. Asian pensionfunds have suffered from the constraints of

reserve fund investing because their assetsare huge. (Chart 1)

With EUR 560 billion in assets, the Japan-ese Government Pension Investment Fund is by far the biggest reserve fund in Asia andthe largest pension fund worldwide.2 Thiscapital originates from a long-standing sur-plus of contributions over pension benefits,which has persisted since its inception in the1940s, but has been getting smaller (Saka-moto 2005). South Korea’s National PensionFund is the fourth largest pension fund glob-ally and has assets of EUR 142 billion undermanagement. The Korean pay-as-you-go sys-tem was established in 1988 and the mini-mum period of membership is 20 years, socapital accumulation could proceed rapidlyover this period. Founded in 2006 and in-tended to fund public service pension liabil-ities, the Future Fund in Australia is theyoungest reserve fund in the sample. Fi-nanced by budget surpluses and privatisa-tion proceeds, it could accumulate EUR 35.9billion within its first year of existence.

Trends in Asian Reserve Funds

Chart 1 Size of Asian reserve funds 2006* [EUR bn] * Or latest year available

Source: OECD 2007b, national sources

600560

142

35,9 27,5 18,7

500

400

300

200

100

0Japan Korea Australia China Taiwan

2 Although the Social

Security Trust Fund in

the U.S. is larger in terms

of asset volume, all of its

investments are in non-

tradable securities issued

by the US Treasury, so

that it normally is not

considered.

10


China’s National Social Security Fund wasestablished in 2000 and is capitalised fromseveral sources. Its assets amounted to EUR27.5 billion in 2006.

These funds are still accumulating assets.It is estimated that the assets of Japan’s Gov-ernment Pension Investment Fund will peakin 2050; in South Korea, assets will probablyaccumulate until mid-2040. China’s NationalSocial Security Fund is unlikely to start decu-mulation before 2030, and Australia’s FutureFund has been recently established.

The governance structures and the invest-ment policies of Asia’s pension reserve fundsare in a process of substantial reform. Threekey trends can be detected.

• Reform of the board structures and in-creasing independence of reserve funds:In three of the four countries (Australia,Japan, South Korea), the reserve funds areor will become an independent publicagency. This will diminish government in-fluence on the funds and, consequently,on investment policies. China is the ex-ception. At the same time, increasedboard professionalisation is noticeablewith financial experts exerting greaterinfluence in the governance of the fundseither through an advisory role to theboards, or through direct representationas board structures change from repre-sentative to professional boards staffedwith financial experts.

• Increased return orientation and a retreatfrom public project financing: Given theimportance of pension reserve funds forthe operation of public pay-as-you-go sys-tems in the future, reserve funds haveadopted a heightened return orientation.The reserve funds in Japan and SouthKorea used to be considerably involved in

public and government projects, whichreturned low yields. This involvement wasstopped and assets invested in the finan-cial markets. The reserve funds of theother countries have not been investing in this type of projects.

• Outsourcing to private asset managersand diversification of investments: Con-gruent with increased return orientation,the outsourcing of assets to private assetmanagement companies has increasedsubstantially. The main motivation is toachieve higher returns through the profes-sionalisation of investment management.This goes hand in hand with a diversifica-tion of assets. There is a move away frominvesting primarily in bonds and anincreasing tendency to invest in the inter-national markets, mostly through out-sourced assets, and, in the case of SouthKorea, also in alternative asset classes.3

Among the four countries investigated, Aus-tralia and China run strategic reserve fundsfinanced by fiscal transfers and dedicated tocushion the impact of ageing on the publicpension system (although the Australianfund is confined to pension liabilities fromthe public service). In Japan and South Korea,the capital of reserve funds stems from asurplus of contributions over payouts. As the capital accumulated stems from publicsources, contributions or general budgetrevenues, the government is necessarilyinvolved in the governance of the funds. Inall countries, the boards or the chair of theboard, are appointed by the government.However, the degree of government influencediffers between full government control ofthe fund in China and the establishment ofthe fund as an independent agency, as is thecase in Australia, Japan and soon in SouthKorea. (Table 2)

3 These trends are in line

with recent World Bank

recommendations on the

governance of public

pension funds, namely

clear objectives, institu-

tional independence from

government, long-term

funding sources and a

small board of experts

among others (Vittas, Im-

pavido, O’Connor 2008).

11


In terms of internal governance, the maindifference is between a professional and arepresentative / government board. The for-mer consists of members appointed becauseof their financial expertise, while the latterincludes groups with a stake in the pensionsystem, such as employers and employees.Obviously, the professional board is moreconducive to a focus on returns in invest-ment management, while the latter type alsoallows non-financial concerns to be consid-ered. Reform plans in South Korea foresee

that a board of investment professionals willreplace the representative board structure,consisting of government officials and stake-holders. Investment professionals also makeup the management board of Australia’s fu-ture fund. In Japan’s reserve funds, expertshave been given a much greater role than in the past though the establishment of anadvisory body. China is an exception; herethe board is made up of current or formergovernment officials.

Australia China Japan South Korea

Type of fund Strategic reservefund

Reserve fund Reserve fundStrategic reservefund

Independence from government

No Yes Currently no, in thefuture yes

Yes

Type of board Government-controlled

Professional Representative (to be changed intoprofessional board)

Professional

Appointing body State Council Ministry of Health,Welfare and Labour

Ministry of Healthand Welfare

Treasury / Ministryof Finance and Administration

Outsourced assets[%]

37 60 7TBD / Fund not fully operational

Foreign investments[%]

N.A. 26 83.9

Bond /deposit share[%]

63 63 8774

Predefined targetreturn

No overall target return

Yes YesYes

Table 2

Source: OECD 2007b, national sources

12


Despite these differences in governance,the investment policies of all countries havechanged substantially over the last years.Outsourcing and diversification of assetshave grown considerably. Outsourced assetsof China’s NSSF increased from zero at thetime of establishment in 2000 to 37% in 2006(OECD 2007b). Australia’s fund will out-source active management; what sorts ofassets will be managed internally is not yetdecided. In South Korea, outsourced assetshave grown from 3.1% to 7.3% between 2003and 2005 and outsourcing will accelerate inthe years to come. In addition, in 2007, theAsian reserve funds granted several majoroutsourcing mandates. South Korea is also a good example for the increasing diversifi-cation of assets. Its National Pension Fund,which already invests in alternative assets,granted a major mandate for alternativeinvestments last year, and will substantiallyincrease the portion of domestic and inter-national equities in the near future.

Closely related to investment policy is apredefined target return for the reserve fund.It helps to clearly define goals of reserve fundinvestment management and to measurethem. In this sense, clear targets also sup-port transparency of the investment processand a focus on performance. Australia, Japanand South Korea use this instrument. Chinahas set target returns for fixed-income prod-ucts and strategic investments, but no targetreturn for the whole fund.

In all, these tendencies imply that invest-ment management of the Asian reserve fundshas become much more sophisticated andmore return-oriented. The remainder of thispaper provides a more detailed analysis ofthe design, the operation and the investmentpolicies of the respective reserve funds.

13


Of all the pension reserve funds consideredin this paper, the Japanese Government Pen-sion Investment Fund (GPIF) has the longesthistory. It is the largest pension fund in theworld and, for reasons considered later, itwill even increase its assets substantiallyover the next few years. The GPIF managesthe reserves of the two pillars of public pen-sions in Japan, the flat rate National Pensionand the earnings-related Employee PensionInsurance. The former was established in1959, the latter in 1944. The reserves stemfrom surpluses of contributions over pay-ments.

The GPIF itself is a relatively new institu-tion. Until the 2001 reform, the reserves wereentrusted to the Trust Fund Bureau of theMinistry of Finance. The pension reserves, to-gether with the savings of Postal savings andPostal life insurance, were used to financethe Fiscal Investment and Loan Schemes,investing primarily in public infrastructure(Usuki 2002). A portion of the reserves wasearmarked for loans for welfare facilities,such as hospitals. In the 1980s and 1990s,there were steps to invest the money moreprofitably by placing it with financial insti-tutions. In all, the surpluses of the pensionsystem were publicly managed with a focuson fostering public projects at the price ofvery low returns (Casey 2004).

This has changed with a substantial re-form in 2001, which was helped by the grow-ing awareness of the magnitude of the demo-graphic change Japan faces and the resultingcentrality of pension reserves.4 Responsibilityover the reserve fund was transferred to theMinistry of Health, Labour and Welfare andinvestment decisions should not be taken inthe “public” interest any longer, but in the in-terest of the beneficiaries, implying a higher

return orientation. The GPIF was founded tomanage the reserves. Simultaneously, disclo-sure and audit were strengthened. In 2006,the GPIF gained independence from the gov-ernment when it became an independentadministrative organisation.

The board consists of two members, thepresident and the executive managing direc-tor. The Minister of Health, Welfare and La-bour appoints the president, who then inturn appoints the executive director. TheMinister also sets the mid-term objectivesalthough he is not involved in operationalmatters, as used to be the case (Global Pen-sions 2006). The Investment Committeewithin the GPIF provides expert advice. TheMinister for Health, Labour and Welfare appoints its members, who must have ex-pertise in finance and economics or otheracademic qualifications. The committeesupervises GPIF investments and gives re-commendations to the president of the GPIF.

Investment policy takes place within pre-determined limits. The asset allocation to bereached by the end of fiscal 2008, when thetransfer of all assets will be completed, ispreset. This principle portfolio aims to meetthe targeted rate of return, which is 3.37%.(Chart 2) There is a permissible range ofdeviation of 5-8% for each asset class. At theend of fiscal 2006, domestic bonds account-ed for 52% of the portfolio, domestic stocksfor 22%, foreign bonds for 11% and foreignstocks for 15% of the assets.

There are guidelines regulating the scopeof internal asset management. The GPIF isonly allowed to manage bonds in-house(Sakamoto 2005). In reality, only passivelymanaged domestic bonds are administered

Country Developments

4 Japan has one of the

fastest ageing popula-

tions in the world. In fact,

many observers consider

the country to be the old-

est society in the world

even now. The current

old-age dependency ratio

stands at 30 and is pro-

jected to worsen to 74 in

2050. During the same

period, Japan’s population

will decrease from 128

million to 102 million

(Allianz Global Investors

2007).

Japan: the Government Pension Investment Fund

14


South Korea established a partially fundedpay-as-you-go system in 1988. As the mini-mum membership period for benefits is 20 years, benefit payments will only start in2008. Consequently, the National PensionSystem has accumulated a substantialamount of assets. The National PensionFund is the largest institutional investor in

South Korea and fourth largest pensionfunds in the world with assets under man-agement of EUR 142 billion (KRW 172 trillion)in 2006. According to official statements, itssize could increase ten-fold until 2025. How-ever, due to the rapid ageing of South Korea’spopulation, the fund will be depleted by 2062,according to IMF projections (IMF 2007).5

Allianz Global Investors International Pension Papers

internally. All other assets are outsourced(Global Pensions 2006). Generally, the GPIFrelies on passive management for most ofits assets. Only a fifth of assets are managedactively, because active management canonly be used when the GPIF has a strongbelief in outperformance. Besides, the sheersize of the GPIF complicates active manage-ment (Global Pensions 2007). Foreign in-vestments in the portfolio account for 25.5%of assets.

In 2005, the GPIF had assets of around EUR560 billion (JPY 88 trillion) under manage-ment. The magnitude of outsourced assets isestimated to have been EUR 337 billion (JPY

53 trillion) (Nomura Research Institute 2006).However, not all assets have yet been trans-ferred to the GPIF; the transfer is supposed tobe completed in 2008. Then, according to esti-mates, the GPIF will have assets of aroundEUR 1.1 trillion (JPY 166.5 trillion) and out-sourced assets of EUR 719 billion (JPY 113 tril-lion) (Nomura Research Institute 2006).

In all, the management of Japanese pen-sion reserves has undergone a shift towardsprofessionalisation of asset management.This is marked by a bold move away frominvesting primarily in government or wel-fare projects and by turning the CPIF into anindependent agency.

Chart 2 Target portfolio of the Government Pension Investment Fund 2008 [%]

Source: OECD 2007b

Short-term assets: 5

Foreign bonds: 8

Foreign stocks: 9

Domestic stocks: 11

Domestic bonds: 67

South Korea: the National Pension Fund

5 South Korea faces one

of the most severe demo-

graphic challenges in the

world. Today’s dependen-

cy ratio stands at 13. It is

expected to rise to 64 by

2050. The dependency

ratio in South Korea is

changing faster than in

any other OECD country.

While the proportion of

elderly people is currently

the second lowest in the

OECD, it will be among

the highest by 2050.

15


The governance structure of the NationalPension Fund is in transition. Currently, thefund is managed by the government, or morespecifically the Minister of Health and Wel-fare. The highest decision-making body with-in the National Pension Fund is the NationalPension Fund Management Committee. It isa representative board and decides on issueslike fund management guidelines and plans.The committee consists of twenty-one mem-bers; the chair is the Minister of Health andWelfare. The other members are deputy min-isters from related ministries (5), the chair ofthe national pension service, representativesof business and employer organisations (3),representatives of labour organisations (3),representatives of several other organisations,such as agriculture or consumer protection(6) and financial experts (2). The committee’splans have to be coordinated with the Min-istry of Planning and Budget and the Ministryof Health and Welfare, before they are sub-mitted to the Cabinet Council and the Nation-al Assembly, which has to agree.

In the second half of 2007, the govern-ment announced plans to fundamentallyrestructure the NPF’s governance system. Theplans foresee that seven professionals fromthe private sector will replace the current 21members of the management committee.Also discussed is the recruitment of non-Korean financial experts for the committee.The chair would be appointed by the Koreanpresident on recommendation of the primeminister. Equally important, the NationalPension Fund should be run as an independ-ent public enterprise, separated from otherpension organisations, in which it is current-ly included. The fund might also be dividedup (Korea Times 15. November 2007 and 11.September 2007).

The second governance layer is currentlythe National Pension Fund Evaluation Com-

mittee. It advises and assists the managementcommittee regarding the management of thefund in areas such as performance measure-ment, composition of fund assets and moni-toring. Its composition is similar to that ofthe management committee with 21 mem-bers, the Vice Minister of Health and Welfareas Chairman, seven officials and representa-tives of societal groups. Finally, there is theFund Management Centre for internal assetmanagement and the National Pension Re-search Institute, which has several researchtasks regarding the pension system and themanagement of the fund.

Over the last years, the investment policyof the National Pension Fund has seen sub-stantial changes. The fund traditionally in-vested in three sectors: the public, welfareand financial sectors. Public sector invest-ments refer to public projects, such as re-gional development or infrastructure. Wel-fare sector refers to the funding of loans andwelfare facilities. Examples include loans fortuition or medical expenses, while welfarefacilities include old-age care centres or hos-pitals. Allocation in these two sectors usedto demand a significant share of the Nation-al Pension Fund’s assets with a strong pre-ponderance of investments in the public sec-tor.

It is obvious from table 3 that investmentsin the public and welfare sectors were almostcompletely abandoned in favour of financialmarket investments. Part of the explanationis that until 2004 government bonds wereclassified as public sector investment, afterthat time they have been counted as invest-ments in the financial sector. Despite that,the main reason for the decrease to nil isthat, since the early 1990s, several funds hadto deposit capital in the Public Fund to beused for public projects. This obligation wasrescinded in 2000, mainly due to poor returns,

16


and the deposited funds given back by 2005.The returns on these investments were con-siderable lower than those for financial mar-ket investments. For example, in the period1995 to 2005, the returns on financial mar-ket investments were 2.3 percentage pointshigher than the public sector investmentsand 3.4 percentage points higher than thewelfare sector investments (National Pen-sion Research Institute 2005).

The main principles of the NPF invest-ment policy are the long-term stability of the fund, maximising returns at a given risklevel, and a contribution to the national

economy and the development of financialmarkets. The long-term target investment re-turn is higher than the expected GDP growthrate (National Pension Research Institute2005). The NPF invests in domestic and for-eign bonds, domestic and foreign stocksand, in a recent development, in alternativeslike private equity. Each asset class is expect-ed to perform at least as good as the bench-mark. Looking more closely at the currentasset allocation, it shows that most assetsare invested in bonds. (Chart 3)

Blue chip equity investments are usuallypassively and internally managed, while

Chart 3 Asset allocation of NPS investment in the financial sector 2005 [%]

Source: National Pension Research Institute 2007

Cash: 0.4

Alternatives: 0.5

Equities: 12.5

Bonds: 86,6

Public

1990

1995

2000

2005

46.3

65.4

56.9

0.0

0

4.0

1.2

0.2

53.7

30.6

41.9

99.8

Financial Welfare

Table 3 Asset allocation of NPF funds

Source: National

Pension Research

Institute 2007

17


outsourced mandates are actively managed.Bonds are normally managed in-house, butcan be outsourced according to limits statedin the annual fund management plan. Alter-natives are managed externally, except for in-frastructure investments and loans (Nation-al Pension Research Institute 2005).

There has been a tendency to increase outsourcing of assets over the last years,although at a low level. For example, in 2005,7.3% of assets were outsourced, in 2003 it was3.1%. Within the bond category, there is a shifttoward foreign bonds, which increased from0.6 to 7.3% of assets. Since then, outsourcinghas continued. In 2007, the NPF awarded twomandates of USD 950 million (EUR 687 mil-lion) combined to foreign external managers.These mandates are for overseas invest-ments.

As with the governance structure, the in-vestment policy of the NPF is also about tochange fundamentally. The fund plans toincrease its equity investments to 30% by

2012, while overseas equity investments areset to rise to 10% from 0.4% in 2005. At thesame time, bond investments are intendedto decrease to 50% (Korea Times 15. Novem-ber 2007). The motivation for this change inpolicy is the same as for the governance re-forms, namely to boost investment returns.

In all, the NPF is currently a governmentcontrolled institution, which over the lastyears has withdrawn from financing govern-ment or welfare projects and shifted itsinvestments almost completely towards thefinancial market. Simultaneously, there is anincreasing trend towards outsourcing toasset managers as a consequence of a morepronounced return orientation. This higherreturn orientation also induced plannedreforms in governance and investment poli-cy. Governance is set to be more profession-alised, while the investment policy will bemore geared to equity and internationalinvestments. In this sense, South Korea is apronounced example of the trends under-way in Asian reserve funds.

The Australian pension reserve fund wasonly established in 2006. The Future Fundaims at funding costs arising from unfund-ed public sector superannuation liabilities.These liabilities will become payable to pub-lic servants and defence personnel from 2020onward. Current unfunded liabilities, whichamount to EUR 2.7 billion (AUD 4.5 billion)per annum, are covered by the governmentbudget. Liabilities are expected to grow toaround EUR 88.5 billion (AUD 148 billion) by 2020 and to more than EUR 119.6 billion(AUD 200 billion) by 2046. The Future Fundis funded by budget surpluses and privatisa-tion revenues. Its balance in 2007 was approx-imately EUR 35.9 billion (AUD 60 billion). It

is expected to hit its planned asset target of EUR 88.5 billion (AUD 148 billion) in 2020.

The Future Fund Board of Guardians, anindependent body, supervises the fund. Theboard is responsible for strategic investmentdecisions and is accountable to the govern-ment for asset performance. The governmentappoints the members of the board for a peri-od of up to five years, the main selection cri-teria being expertise in asset management.Indeed, all members of the board do not havea background in government services, but inprivate, financial market-related companies.Administrative and operational functions

Australia: the Future Fund

18


are carried out by the Future Fund Manage-ment Agency, which supports the board andrecommends on investment strategies.

The targeted long-term return of the FutureFund is determined in the investment man-date issued by the Treasury and the Ministerfor Finance and Administration. Currently, thetarget return is 4.5-5.5% per annum above theConsumer Price Index over the long-term,while during the transition period on the wayto a long-term asset allocation, a lower returnis expected.

In 2007, the Australian government trans-ferred its remaining stake in the telecommu-

nications and media company Telstra to thefund, valued at EUR 5.3 billion (AUD $8.9 bil-lion). The fund is not yet fully operational, sothere is a strong overweight in cash. (Chart 4)As soon as the long-term investment plan isset-up, asset allocation will change substan-tially. The goal is to have a widely diversifiedportfolio in the long run, the foundations ofwhich will be built up over the next years.According to announcements, active manage-ment will be outsourced, although the splitbetween active and passive management aswell as the degree of outsourcing is not clearyet.

Chart 4 Asset allocation Future Fund 2007 [%]

Source: Future Fund

Australian equities: 3.5

International equities: 3.9

Telstra equities: 18.6

Cash: 74.0

The Chinese National Social Security Fund(NSSF), a strategic pension reserve fund, is fedby fiscal transfers in order to build up capitalfor coming public pension deficits resultingfrom upcoming demographic development.6

In its baseline scenario, the World Bank esti-mates that China’s implicit pension debtstands at 141% of its 2001 GDP (Sin 2005).7 The

NSSF was founded in 2000, and is in the accu-mulation phase. It is unclear when paymentswill start. Some sources think that decumula-tion will begin when the fund reaches assetsof at least EUR 97.2 billion (RMB 1 trillion).Others suggest that decumulation will com-mence from 2030 onwards, when the demo-graphic situation is projected to deteriorate.

China: the National Social Security Fund

6 Between 2005 and

2050, the old-age depend-

ency ratio in China is ex-

pected to worsen from 11

to 39; the median age will

rise from 32.5 to 48 years

in the same period. While

the absolute numbers ap-

pear relatively favourable

compared to industrialised

countries, the speed of

ageing in China is excep-

tional. Essentially, China

will age significantly with-

in one generation.

7 Implicit pension debt is

the present value of un-

funded pension liabilities

or promises.

19


Assets of the NSSF come from four sources:• Fiscal transfers from the central govern-

ment budget• Proceeds from the listing of state-owned

enterprises• Lottery proceeds• Investment income

The bulk of NSSF assets comes from fiscaltransfers, followed by the proceeds of pri-vatisation, investment income and lotteryincome. The proceeds from privatisationcome from the Initial Public Offerings (IPO)of state-owned enterprises, which mustgrant 10% of money raised from an IPO tothe NSSF, as well as from overseas listings.In 2006, the NSSF’s total assets amounted toEUR 27.5 billion (RMB 283 billion).

The NSSF is managed by the NationalCouncil of the Social Security Fund (NCSSF),which was established simultaneously to thefund. It is an organisation with the status ofa ministry and directly reports to the StateCouncil, the government’s cabinet. The NCSSFcomprises 17 executive board members andis led by a chair and three vice chairs, whichare nominated by the State Council. The re-

maining directors are also appointed by theState Council. Many NCSSF officials have abackground in other key government depart-ments (Leckie, Pan 2007). This means theNCSSF is not independent, but fully integrat-ed in the state administration. Tendencies to grant pension reserve funds a higher in-dependence from government, as in othercountries, are not discernible in China. How-ever, the independence of reserve fundstends to occur in countries with a highereconomic development.

The NCSSF is responsible for managingNSSF assets, developing its investment poli-cy, selecting external fund managers, and forfinance and controlling. The NSSF investmentpolicy is based on the priorities of asset se-curity and liquidity. As a result, regulationsspecify that deposits and government bondscombined must amount to at least 50% ofassets. At least 10% must be invested in de-posits alone. The maximum limit for corpo-rate bonds is 10%, and the combined limitfor shares and mutual funds is 40%.

The asset allocation of the NSSF was con-servative in the initial period of operation.

Chart 5 NSSF asset allocation 2006 [%]

Source: Leckie, Pan 2007

Cash: 8.3

Private equity holdings: 13.7

Listed shares: 23.5

Bonds: 54.5

20


In 2000, all assets were invested in bankdeposits and government bonds. However,this picture has begun to change. In 2006,the share of equities has increased signifi-cantly to 23.5% of assets managed. (Chart 5)A notable share of assets is devoted to directequity holdings, mainly pre-IPO investmentsin Chinese enterprises, especially in banks.NSSF management plans foresee that directequity investments should generate a returnof at least 6%, while fixed-income productsshould generate at least 3.5% (Sekine 2007).

Besides investments in higher-risk assets,outsourcing is a second major trend in theNSSF investment policy. The share of out-sourced assets has increased steadily, from24.1% in 2003 to 37.3% in 2006. In 2003, theNSSF began outsourcing its assets to domes-tic asset managers, which were allowed toinvest up to 40% of these assets in stocks. In2005, five national and/or joint ventureswere selected to manage NSSF assets. In2006, new regulations were issued that

allowed the NSSF to invest internationallythose assets raised through IPO proceedsabroad. Following this, the NSSF allotted fiveinvestment mandates to ten foreign assetmanagers and one joint venture to manageinternational investments of EUR 758 mil-lion. The mandates foresee investments inHong Kong stocks, global stocks, US stocks,global bonds and foreign currency.

The shift in investment policy towardshigher risk assets and outsourcing seems tohave paid off for the NSSF. The realised rateof return has jumped from 2.75% in 2002 to9.34% in 2006 (Sekine 2007), helped by theperformance of the Chinese stock markets.Generally, the NSSF fills an important gap in the pension system. While the country isonly in the process of setting up a compre-hensive pension system, from a publicfinance point of view, it is a good idea to pro-vide for the foreseeable liabilities of this newsystem to make it sustainable in the longrun.

21


The governance and investment policiesof Asian pension reserve funds are in

transition. Their strategic reorientation canbe described succinctly as putting returnsfirst. This strategy is supported by increasedoutsourcing of assets and their diversifica-tion, the retreat from financing functions forgovernment and public projects and theprofessionalisation of board governance.These developments benefit future pension-ers, whose pension payouts are partlydependent on the performance of the assetsmanaged by the reserve funds. In this way,professional management of the huge pen-sion reserves is critical for the sustainabilityof public pensions. As pension reserve fundsare necessarily investors with a very long-term horizon, sophisticated risk manage-

ment is decisive for long-term stability ofpension reserves.

The patterns observed also have widerramifications for financial markets. The rising share of international and equityinvestments in the portfolios of Asian re-serve funds imply that the importance ofthese funds as an actor on the global finan-cial markets will increase substantially. Asin the case of sovereign-wealth funds, theirsheer size and societal impact will bringissues like internal governance, investmentprocesses, political influence and trans-parency inevitably into the public spotlight.This will make clear goals, sophisticatedinvestment strategies and solid governancestructures all the more important.

Conclusions

22


Allianz Global Investors 2007. Asia-Pacific Pensions 2007. Systems and Markets.http://publications.allianzgi.com/en/PensionResearch/Pages/PensionStudies.aspx

Casey, Bernard 2004. The Japanese Retirement Income System: A Special Case?. Cass Business SchoolPensions Institute Discussion Paper 0407, London.

Global Pensions 2007. GPIF faces Challenges of Independence. August.

Global Pensions 2006. Reforms Fail to Deliver Independence. August.

International Monetary Fund 2007. Republic of Korea: Selected Issues. IMF Country Report No. 07/345,October 2007.

Korea Times. Foreigners will Help Manage Pension Fund. 15. November 2007.

Korea Times. Pension Fund to Manage Assets Independently. 11. September 2007.

Leckie, Stuart; Pan, Ning 2007. A Review of the National Social Security Fund in China. Pensions: AnInternational Journal, Vol. 12, No. 2, pp. 88-97.

National Pension Research Institute 2007. 2005 Annual Report on National Pension Fund Management.http://www.nps.or.kr

Nomura Research Institute 2006. Japan’s Asset Management Business 2006. http://www.nri.co.jp/english/opinion/jamb/JAMB2006.pdf

OECD 2007a. Pension Markets in Focus. No. 4, November 2007, OECD, Paris.

OECD 2007b. Asian Pension Systems. Unpublished Reports.

Sakamoto, Junichi 2005. Japan’s Pension Reform. World Bank Social Protection Discussion Paper No. 0541.

Sekine, Eiichi 2007. ‘Recent Moves By China's Social Security Fund, an Increasingly Important Partici-pant in Chinese and Global Capital Markets’. Nomura Capital Market Review Vol. 10, No. 3.

Sin, Yvonne 2005. China. Pension Liabilities and Reform Options for Old Age Insurance. Working PaperSeries Human Development Sector Unit, No. 2005-01. World Bank, Washington.

Usuki, Masaharu 2002. The New Investment Management Scheme for Japan’s Public Pension Fund. Dis-cussion Paper 02/09, Centre for Pensions and Superannuation, University of New South Wales, Sydney.

Vittas, Dimitri; Impavido, Gregorio; O’Connor, Ronan 2008. Upgrading the Investment Policy Framework ofPublic Pension Funds. Policy Research Working Paper. World Bank Financial Policy Division, Washington.

Weaver, Kent 2003. Whose Money is it anyhow? Governance and Social Investment in Collective Invest-ment Funds. Working Paper 2003-07, Center for Retirement Research at Boston College.

Yermo, Juan 2008. Governance and Investment of Public Pension Reserve Funds in Selected OECD Coun-tries. OECD Working Papers on Insurance and Private Pensions, No. 15, OECD Publishing, Paris.

References

Allianz Global Investors International Pension Papers No. 1|2008

23


www.allianzglobalinvestors.com

Allianz Global Investors AGInternational PensionsNymphenburgerstr. 112-116D-80636 Munich

funding unfunded pensions: governance and investments of ......reserve funds in japan, south korea,...

Documents