The Graying of theMiddle Kingdom Revisited

China’s Long Marchto Retirement Reform

by Richard JacksonKeisuke Nakashima& Neil Howe

with contributions byJiangong Zhou

TABLE OF CONTENTS

Foreword 1Introduction 2

Chapter 1: The Dimensions of China’s Aging Challenge 7China’s Demographic Transformation 8China’s Looming Retirement Crisis 10The Broader Economic Outlook 15

Chapter 2: The Evolution of China’s Retirement System 19From Iron Rice Bowl to Empty Accounts 19Liaoning and Beyond 22The Unfinished Agenda 25

Chapter 3: A New Direction for Retirement Reform 27Building the Floor of Protection 27Increasing Funded Retirement Savings 28Designing a National Personal Accounts System 30The Advantages of Funding 35Reform and China’s Capital Markets 37

Conclusion: China at a Crossroads 42

A Note on Data and Sources 44A Key to Chart Source Citations 46Acknowledgments 47About the Authors 48About CSIS andThe Prudential Foundation 49

China’s Long Marchto Retirement Reform

Foreword 1

Foreword

The year 2008 began in China amid celebratory preparation for the Beijing Olympics. Theworld watched in awe as the Chinese coordinated and choreographed a breathtaking spectaclethat captured the hopes and aspirations of this remarkable country. As it turns out, Chinacould not celebrate its achievement for long. Before year’s end, China found itself in aneconomic crisis precipitated by the global recession.

No one who knows China doubts that it will successfully weather today’s global economicstorm. Yet having said that, there is indeed cause for concern about China’s economic future.This concern has nothing to do with the current crisis. It is instead tied to a longer-term andin many ways more serious challenge: the dramatic aging of China’s population.

China’s Long March to Retirement Reform warns that the aging of China’s population couldusher in an era of slower economic growth and mounting social stress. The sheer magnitudeof China’s age wave—by 2030 China will be an older country than the United States—wouldalone pose a serious challenge. What makes the challenge even more daunting is that the agewave will arrive while China is still developing and modernizing. The public pension systemcovers only a fraction of the population and could collapse under the weight of unfundedbenefit promises. The traditional family support networks on which most Chinese elderlydepend are weakening. Without reform, tens of millions of Chinese could arrive at old ageover the next few decades without pensions and with inadequate family support.

The report outlines a plan for meeting the challenge. The plan provides for a universal povertybackstop that would protect all Chinese against an uncertain old age. It would also create anational and fully portable system of funded retirement accounts. This system would allowChina to care for a much larger number of older people without overburdening its smallerworking generation. As the report explains, it would also help China to maintain rates ofsavings, investment, and living standard growth as its population ages.

China is a lynchpin in the global economy. How it confronts its age wave will have profoundimplications for its own future and for the future of countries around the world. China’s LongMarch to Retirement Reform offers a compelling analysis of the risks and opportunities posedby China’s coming demographic transformation.

John J. HamrePresident and Chief Executive OfficerCenter for Strategic and International Studies

Edward P. BairdExecutive Vice President andChief Operating Officer, International BusinessesPrudential Financial, Inc.

1

China is still a demographically young nationpreoccupied with a young nation’s challenges—educating the young, creating jobs for a growingworking-age population, and building a moderneconomy. China, however, is being overtaken by astunning demographic transformation. In 1975, therewere six Chinese children for every one elder. By2035, there will be two Chinese elders for everyone child. (See figure 1.) During the single decadebetween 1995 and 2005, China added 107 millionworking-age adults to its population. During thesingle decade between 2025 and 2035, it willsubtract 79 million.1

China will have to pay foran age wave of developed-world size with just a fractionof the developed world’sincome and wealth.

What makes China’s age wave so daunting is notjust its magnitude or the speed with which it isapproaching, but the fact that it is arriving in asociety that is still in the midst of modernization.While today’s developed countries were all affluent

societies with mature welfare states by the time theybecame aging societies, China is aging at a muchearlier stage of economic and social development.No matter how fast its economy grows over the nextfew decades, it will have to pay for an age wave ofdeveloped-world size with just a fraction of thedeveloped world’s per capita income and wealth.

When China embarked on its “Open Door” programof economic reform in 1978, it was an impoverishedcountry with few ties to the global marketplace.Today it is a rising global economic power. China hasbecome the factory floor of the world. In 2007,exports of goods and services reached 41 percent ofGDP, ranking China (in dollars) as the world’s thirdlargest exporter, just behind the United States andGermany. Thanks to its massive current accountsurpluses, which reached 11 percent of GDP in 2007,it is also becoming a major global creditor. As ofmid-2008, China’s foreign exchange reserves totaled$1.8 trillion, or 50 percent of GDP. Along the way,the typical Chinese has seen a spectacular increase inliving standards. In the three decades since Chinabegan its transition to a market economy, real percapita income has risen elevenfold. Its economicsuccess is manifest in the breathtaking skyline ofShanghai’s Pudong business district and was ondisplay to the entire world at the Beijing Olympics.

A young nation is about to grow old.

Chinese Children(Aged 0-14)and Elderly(Aged 60 & Over),in Millions, 1975-2050

Figure 1

Mill

ions

ofPe

ople

Source: UN (2007)

0

100

200

300

400

500

2035 205020051975

366

64

284

144

208

393

185

438

Aged 0-14

Aged 60 & Over

1 All population projections cited in this report refer to the UN’s 2006 Revision “constant fertility” scenario. “Children” refers tothe 0 to 14 age bracket, “working-age adults” to the 15 to 59 age bracket, and “elderly” to the 60 and over age bracket. The UNprojections, as well as other major data sources, are discussed in “A Note on Data and Sources” at the end of the report.

2 China’s Long March to Retirement Reform

Introduction

Introduction 3

Yet despite its blistering economic growth, China isstill a low-income country. Per capita income mayhave risen elevenfold since the beginning of thereform era, but even taking into account differencesin purchasing power, it is still just one-fifth the levelin South Korea and one-ninth the level in the UnitedStates. (See figure 2.) In exchange rate dollars, Chinaproduced 6 percent of global GDP in 2007, while theUnited States, with one-fourth of China’s population,produced 25 percent. China has lifted more peopleout of poverty in a shorter time than any country inthe history of the world. But according to the WorldBank, at least 170 million Chinese still lived on lessthan $1 a day in 2004.2 According to the McKinseyGlobal Institute, less than 1 percent of Chinesehouseholds have incomes equal to or greater thanthe average income of U.S. households.3

China is also a country whose social fabric is beingstrained by rapid modernization. While the industrialrevolution in Europe and the United States unfoldedover a century or more, China’s is being telescoped intoa single generation. When the government launched itsprogram of economic reform, urban workers lost thecradle-to-grave social protection that many had enjoyedin the days of the planned economy. Tens of millions ofpeasants are moving from traditional agriculturalvillages to bustling manufacturing hubs, where they join

a rootless “floating population” now estimated tonumber 150 million.4 Industrialization and urbanizationare weakening the extended family and degrading theenvironment. Worker mobility and turnover are risingand the income gap between the rich and poor iswidening. China’s leaders know that they must deliveron the promise of mass affluence or face a majorsocial—and perhaps even political—backlash. But theyalso understand that the social costs of breakneckdevelopment threaten their goal of building a“harmonious society.”

The rapid aging of China’s population could act asa multiplier on the social and economic stresses ofrapid modernization. Less than one-third of China’sworkforce is now earning a public pension benefit ofany kind. Despite China’s lofty national savings rate,only a small minority of workers are accumulatingsufficient financial assets to support themselves inretirement. The great majority will have to fall backon the most traditional form of old-age insurance:children. But as the extended family weakens andbirthrates decline this informal safety net is beginningto unravel. Unless China prepares for the challenge, aretirement crisis of immense proportions looms justover the horizon. Imagine tens of millions ofmembers of China’s low-wage floating populationmaturing by the year 2020 or 2030 into tens of

2 “China Quarterly Update,” The World Bank Office, Beijing, February 2008, http://www.worldbank.org.3 Diana Farrell, Ulrich A. Gersch, and Elizabeth Stephenson, “The Value of China’s Emerging Middle Class,” The McKinsey Quarterly,Special Edition (2006), 63.

4 “China Sees Soaring Migrant Population,” Xinhua News Agency, October 29, 2006.

Despite its rapid economic growth, China remains a low-income country.GDP Per Capita (in 2005 PPP Dollars), by Country in 2007

$0

$10,000

$20,000

$30,000

$40,000

$50,000

U.S.Japan SingaporeEU15South KoreaChina

GD

PPe

rC

apita

Source: World Bank (2008)

$5,046

$23,363

$31,310 $31,607

$43,227$47,488

Figure 2

millions of indigent urban elders who lack pensionsor nearby families. Or imagine, in China’s westernrural regions, entire towns of demographicallystranded elders.

Unless China prepares for thechallenge, a retirement crisis ofimmense proportions loomsjust over the horizon.

A rapidly aging and rapidly developing China willneed a retirement system whose coverage is broadand whose benefits are adequate and affordable. Thefoundation of the system must be a universal floor ofprotection against poverty in old age that covers allChinese elders, whether or not they have participatedin the contributory public pension system. Above thisfloor, we believe that the best solution is for China torely more heavily on funded retirement savings. Werecommend that China transform its current “basicpension system” for urban workers, which consistsof a first tier of pay-as-you-go benefits and a secondtier of largely “notional” personal retirement accounts,into a national system of genuinely funded personalaccounts that would be publicly regulated but privatelymanaged and invested. We also stress the importanceof expanding supplementary retirement coverageunder China’s new private enterprise annuity system.

A genuinely funded retirementsystem would have enormouslong-term benefits.

The reform we propose would have enormous long-term benefits. The floor of protection would sparetens of millions of elders from a destitute old age—and might even avert widespread social and politicalunrest in an aging China. The national system offunded personal accounts would, over time,allow the majority of tomorrow’s elders to enjoya comfortable retirement without overburdeningtomorrow’s workers. It would also help an agingChina to meet its long-term development goals bybroadening and deepening capital markets and bymaintaining adequate rates of savings and investment.Along the way, it would turn most workers into property-owning stakeholders in the future of China’s economy.

The current global financial crisis and the drasticdecline in world stock markets—between its peak inDecember 2007 and the end of October 2008 China’sstock market fell by two-thirds—are causing

policymakers in some countries to question thewisdom of basing retirement security on fundedsavings. One country has gone beyond questioning.Argentina, which introduced a funded personalaccounts system in the mid-1990s, nationalized itspension funds in November 2008 and is now shiftingback to a purely pay-as-you-go system.

Although it is understandable, this concern is misplaced.An effective retirement policy must focus on the longterm—and over the course of a contributor’s workinglife there is little question that a funded system willdeliver higher returns and larger benefits than a pay-as-you-go system can. To be sure, funded systemssubject retirement benefits to the risks of ups anddowns in financial markets. Economists largelyagree, however, that workers nearing retirement canbe protected from sudden financial downdrafts byprudent regulations that require them to move intofixed-income assets at older ages. The financial risk of afunded system, of course, can never be entirely eliminated.But neither can the “political risk” of a conventionalpay-as-you-go public pension system—that is, the riskthat future politicians will change its benefits. Thatrisk will become overwhelming as China ages andthe ratio of beneficiaries to workers soars.

4 China’s Long March to Retirement Reform

Introduction 5

Government reformshave begun to push China’sretirement system in theright direction.

Over the past few years, a series of government reforminitiatives have begun to push China’s retirement systemin the right direction. The government has stepped upefforts to broaden participation in the basic pensionsystem beyond its original base among workers atstate- and collectively owned enterprises—and it hasset a goal of achieving universal coverage by 2020. Ithas also begun to increase funded retirement savingsby creating a national pension reserve fund; by pumpingcentral government subsidies into the basic pensionsystem that allow a growing share of personalaccount contributions to be saved; and by layingthe groundwork for a new supplementary systemof employer-sponsored “enterprise annuities.”

Yet despite the progress, the government’s recentinitiatives fall short of a complete solution. Publicpension coverage remains far from universal, even inthe cities, and the government has yet to take serioussteps toward extending it to the countryside. The basicpension system labors under large unfunded liabilitiesfor legions of early retirees from China’s downsizedstate-owned sector, which means that contributionrates, and consequently evasion, are high. Since itsbenefits are not fully portable, workers often face astark trade-off between job mobility and retirementsecurity. Its personal accounts, though now partiallyfunded, earn such a low rate of return that they cannotpossibly generate promised replacement rates.

China’s Long March to Retirement Reform turns thespotlight on China’s aging challenge. In chapter 1,we describe the coming demographic transformationand the economic and social challenges it poses.In chapter 2, we zero in on and evaluate thegovernment’s efforts to build a national pensionsystem. In chapter 3, we turn to the CSIS reformplan. We describe how the floor of protection wouldbe structured; how the transition from today’s largelyunfunded basic pension system to a new nationalpersonal accounts system could be managed; andwhat the architecture of the personal accounts systemmight look like. We also discuss the advantages of ourreform and the obstacles that would have to beovercome to implement it. The current report buildson our analysis in The Graying of the Middle Kingdom,our 2004 study on China’s aging challenge.

China’s remarkable economic success has becomean inspiration to countries around the world.That success is due, above all, to the hard workand sacrifice of the Chinese people and to soundgovernment policies that have leveraged theirinitiative. Yet it has also been underpinned byfavorable demographic trends that have leanedstrongly with economic growth. Over the past 30years, the demographic dependency burden in Chinahas plummeted and the share of the population in theworking years has soared, helping to push up labor-force participation, savings, investment, and livingstandard growth. Beginning around 2015, however,the demographic climate will change abruptly.

As this report goes to press, China finds itself in themidst of an economic crisis triggered by the deepeningglobal recession. Its export sector is feeling the impactof contracting consumer demand in the United Statesand other developed-world economies, unemploymentis rising, and economic growth is projected to slowdramatically in 2009 from the double-digit rates ofrecent years. In the midst of the near-term economicemergency, China’s leaders may be tempted to deferaction on the long-term aging challenge. That wouldbe a mistake. China’s age wave is approaching sofast—and its potential economic and social costs are solarge—that delay is not an option. Concerted action onthe long-term challenge is needed now to ensure thatChina’s economic fundamentals remain strong whenit emerges from the current crisis. It may even helpmitigate the crisis by bolstering investor confidencein the government’s economic stewardship.

How China navigates itscoming demographictransformation will go a longway toward determiningwhether it becomes aprosperous and stabledeveloped country.

How China navigates its coming demographictransformation will go a long way toward determiningwhether it achieves its aspiration of becoming aprosperous and stable developed country with anexpanding role in the global economic and politicalorder. The time to prepare is fast running out and theoutcome still hangs in the balance. We are confident,however, that China will rise to the challenge.

The magnitude of China’s coming age wave isstunning by any measure. In 2005, there were just16 elderly Chinese for every 100 working-age adults.This aged dependency ratio is due to double to32 by 2025, then double again to 61 by 2050. Bythe mid-2020s, China will be adding 10 millionelders to its population each year, even as it loses7 million working-age adults. (See figure 3.) By 2050,there will be 438 million Chinese aged 60 and over,103 million of whom will be aged 80 or older.

By 2050, there will be438 million Chinese elderly,103 million of whom willbe aged 80 or older.

Populations are now aging throughout most ofthe world. China’s age wave, however, poses anespecially daunting challenge. China is not onlytraversing the entire distance from young andgrowing to old and stagnant or declining at abreathtaking pace—far more rapidly than mostof the developed countries once did—but it isalso aging at a much earlier stage of economic and

social development. When the elderly share of thepopulation in the United States was the same as theshare in China today, the U.S. per capita incomewas four times what China’s now is. Even Japan andSouth Korea, which like China are experiencingextremely rapid aging, were much wealthier at thesame stage in their demographic transformation.(See figure 4.) While today’s developed countriesbecame affluent societies before they became agingsocieties, China may be the first major country togrow old before it grows rich.

China is unprepared for the coming age wave. Itspublic pension system leaves large gaps in coverageand its private pension system is still in its infancy.Most elders depend heavily on informal familysupport networks, but these are coming under intensepressure even before the age wave rolls in. The“premature aging” of China’s population, moreover,poses broader challenges that reach far beyondretirement policy. It threatens to impose a risingburden on the young, slow economic and livingstandard growth, and become a socially destabilizingforce in a country where the stresses of rapidmodernization are already straining the civic fabric.

The Dimensions ofChina’s Aging Challenge

CHAPTER 1

200

150

100

50

0

-50

-100

China’s working-age population will shrink even as its elderly population explodes.Growth in Chinese Working-Age (Aged 15-59) and Elderly (Aged 60 & Over) Population, in Millions, by Decade, 1970-2040

1970-1980 1980-1990 1990-2000 2000-2010 2010-2020 2020-2030 2030-2040

Figure 3

Mill

ions

ofPe

ople

Source: UN (2007)

127

17 23 31

163

92 9274

105

59

1

41

-65 -58Aged 15-59 Aged 60 & Over

Chapter 1 7

8 China’s Long March to Retirement Reform

China’s DemographicTransformationChina may be the world’s oldest living civilization,but for most of its history it has been ademographically young society. As recently as themid-1960s, when Mao Zedong called on the nation’syouth to launch the Cultural Revolution, China’smedian age was 20, meaning that half of thepopulation were children or teenagers. The elderlymade up just 7 percent of the population, about whatthey had since time immemorial.

Over the past few decades, however, a far-reachingdemographic transformation has been gatheringmomentum. The transformation, which will soonlead to a dramatic aging of China’s population, is theresult of two fundamental forces: falling fertility andrising longevity. The first force is decreasing therelative number of young in the population, while thesecond is increasing the relative number of old. Untilthe 1970s, China’s fertility rate—that is, the averagenumber of lifetime births per woman—still toweredaround 6.0. By the early 1990s, however, it hadplunged to 1.8, where it remains today. (See figure 5.)Meanwhile, improved nutrition, sanitation, andhealth care have brought about an equally stunningincrease in life expectancy. Since the founding of thePeople’s Republic in 1949, life expectancy at birth hasrisen by 32 years. Nationwide, it now stands at 73.(See figure 6.) In Beijing it has reached 80 and inShanghai 81—a higher life expectancy than in mostdeveloped countries.5

China is not the only developing country wherefertility has fallen and life expectancy has risen inrecent decades. Like the developed world before it,much of the developing world is now in the midstof what demographers call the “demographictransition”—the shift from high fertility and highmortality to low fertility and low mortality thatinevitably accompanies economic and socialdevelopment.

In China, however, the demographic transition hasbeen accelerated by the government’s strict family

8.0

7.0

6.0

5.0

4.0

3.0

2.0

1.0

0.0

Behind China’s age wave: A dramatic decline in fertility rates.Chinese Total Fertility Rate, 1949-2006

Figure 5

Tota

lFer

tility

Rat

e

Source: NPFPC(various years)

Chinese TotalFertility Rate

1949 6.1

1970 5.8

1980 2.2

1990 2.2

2000 1.8

2006 1.8

1949 1954 1959 1964 1969 1974 1979 1984 1989 1994 1999 2004 2006

$0 $4,000

U.S.

Japan

SouthKorea

China

$8,000 $12,000 $16,000 $20,000

$15,538 in 1943

$14,900 in 1972

$16,256 in 1999

$4,088 in 2005

Figure 4

Source: UN (2007); U.S. Census Bureau (2002);World Bank (2008); Maddison (2007);and CSIS calculations

China’s “premature aging”: A largerelderly population at a lower income level.GDP Per Capita (in 2005 PPP Dollars) in SelectedCountries in the Year They Had the Same ElderlyPopulation Share as China in 2005

5 “Shanghai Residents Live Longer, See Medical Bills Drop,” Shanghai Daily, February 26, 2007; and “Life Expectancy of BeijingResidents Reaches 80 Years,” Xinhua News Agency, March 10, 2007.

planning policies, which pulled down the fertilityrate much sooner and faster than would otherwisehave occurred. Beginning in the early 1970s, thegovernment began encouraging couples to limit theirfamily size to just two children. Then, in the early1980s, it implemented the one-child policy, togetherwith a system of targets, birth permits, and penaltiesto enforce it. In the countryside, where localauthorities typically allow families whose first child isa girl to “try for a son,” fertility remains well abovethe official one-child target—though even herefamilies are now barely replacing themselves. In thecities, fertility has fallen much further—and in Beijingand Shanghai, it has now dipped beneath 1.0.6

When China first put in place its strict birth policies,the nation faced a looming Malthusian crisis. Withmortality rates falling rapidly, continued high fertilitythreatened to lead to runaway population growth andleave the country mired in poverty. China’s leadersdetermined that they would have to lower birthratesdrastically in order to lay the foundations for futureprosperity. At the time, no one gave much thought tothe inevitable sequel to a sudden and steep decline infertility—the eventual dramatic aging of China’spopulation and the social and economic challengesthat it would bring. Those challenges lay over thehorizon in a distant future.

That once distant future is now fast approaching.Over the next few years, China’s last large

generation, born in the 1950s and 1960s, will beginto reach old age. As it does, the number of elderlywill surge. In 2005, 11 percent of China’s populationwas aged 60 or older, up significantly from the daysof the Red Guards, but still barely half of thedeveloped-country average. That share is due to riseto 15 percent in 2015, then leap to 24 percent in2030 and 33 percent in 2050. By then, China’smedian age will be 47, more than twice what it waswhen the demographic transition began. An elderlyshare of 33 percent and a median age of 47 will notmake China the oldest country in the world. But itwill make it an older country than the United States.(See figure 7.)

China will not only havean aging population, butalso a contracting one.

China will not only have an aging population, butalso a contracting one. China’s child populationalready peaked in the mid-1970s. Its working-agepopulation is on track to peak by 2015, less than adecade from now. Between 2015 and 2050, assumingcurrent demographic trends continue, China’s overallworking-age population will contract by 23 percent,with even larger declines among younger adults intheir 20s and 30s. Its total population will continueto grow for a while longer, but only because the

Chapter 1 9

80

70

60

50

40

30

20

10

0

Behind China’s age wave:An equally dramatic rise in life expectancy.Chinese Life Expectancy at Birth, 1950-2005*

1950-55 1955-60 1960-65 1965-70 1970-75 1982 1990 2000 2005

Life

Expe

ctan

cy

*UN five-year average data for 1950-1975 and Chinese census data for 1982-2005.Source: MOH (2008); and UN (2007)

4145

50

6063

68 69 71 73

Figure 6

6 Tabulation of the 2000 Population Census of the People’s Republic of China (Beijing: Population Census Office and National Bureauof Statistics of China, 2002), 1696.

10 China’s Long March to Retirement Reform

number of elderly Chinese will be increasing fasterthan the number of younger Chinese will be declining.By 2030, however, China’s total population will alsopeak and enter a gathering decline. A decade beforethen, in 2020, India will overtake China as theworld’s most populous country—a position thatChina has held for most of human history.

Nondemographers may suppose that populationprojections so far into the future must be highlyspeculative. But in fact, the rapid aging of China’spopulation is about as close as social science comesto a certain prediction about the future. Absent acatastrophic pandemic, longevity will continue to rise.The future trend in fertility is admittedly less certain,though a major surge is unlikely even if thegovernment were to repeal the one-child policy.7After several decades of low fertility, small familieshave become the new social norm in China.According to a recent survey of Chinese women,35 percent now consider one child to be ideal—andthough 57 percent would prefer to have twochildren, only 6 percent want more than two.8 In anycase, higher fertility would have no impact on the sizeof the working-age population or the ratio of workersto retirees over the next 15 years and only a modestone over the next 25. Demography is like an oceanliner. Once it is steaming full speed ahead it takes along time to turn around.

China’s LoomingRetirement CrisisIts massive age wave is overtaking China at anawkward moment in its economic and socialdevelopment. While the United States, Japan,and the countries of Western Europe were all fullydeveloped economies with mature welfare states bythe time they became aging societies, China’s agewave is arriving in a society which, despite its rapideconomic growth, is still in the midst of modernization.In the days of the planned economy, workers atChina’s state-owned enterprises, or SOEs, enjoyedextensive social benefits, including generouspensions. But these workers were always a privilegedminority—and the system of cradle-to-grave socialprotection they once enjoyed has been largelydismantled. Meanwhile, industrialization andurbanization are weakening the traditional means ofincome support on which the vast majority of Chineseelderly have always relied: continued work and theextended family. Yet China has not yet had time tofashion adequate government and market substitutes.

China’s “premature aging”threatens to subject agrowing share of tomorrow’selders to economic hardship.

7 Comments by a high-ranking official at the National Population and Family Planning Commission recently fueled speculation that the policywould be repealed. The government, however, denies that any near-term change is under consideration. “China Denies Plan to Scrap One-Child Policy,” Reuters, March 2, 2008; and Jim Yardley, “China Sticking with One-Child Policy,” The New York Times, March 11, 2008.

8 National Population and Family Planning Commission of China, Major Statistics of the 2006 Sampling Survey on NationalPopulation and Family Planning [in Chinese], Statistical Bulletin of Population and Family Planning, no. 2, 2007.

35%

30%

25%

20%

15%

10%

5%

0%

Within a generation, China will have an older population than the United States.

Elderly (Aged 60 & Over),as a Percent of thePopulation in China and theUnited States, 1970-2050

Source: UN (2007)1970 1980 1990 2000 2010 2020 2030 2040 2050

Perc

ent

of

the

Popu

lati

on

Elderly ShareChina U.S.

7% 14%11% 17%17% 22%24% 24%29% 25%33% 26%

197020052020203020402050

Figure 7

China

United States

China’s premature aging thus threatens to subject agrowing share of tomorrow’s elders to economichardship. The threat is greatest in the countryside,where living standards lag and the social safety netis rudimentary. But even in China’s relatively moreaffluent cities, the contours of a looming retirementcrisis are beginning to take shape.

Coverage under China’sbasic pension system islargely limited to workersat state- and collectivelyowned enterprises.

To be sure, China now has a mandatory “basicpension system” for urban workers, which thegovernment created to replace the old SOE system.In principle, the system, which consists of a pay-as-you-go benefit and a personal retirement account,covers the entire urban workforce. In practice,however, coverage remains far from universal. Asof 2007, just 65 percent of the urban workforcewas contributing to and earning a benefit under thebasic pension system or a separate system for civilservants. (See figure 8.) Coverage, moreover, ishighly concentrated among workers at state- andcollectively owned enterprises, who account for adwindling share of total employment. Most of thefast-growing private-sector workforce, includingChina’s vast floating population of rural migrants,remains uncovered.

Beyond its low coverage, the basic pension systemsuffers from serious structural problems that makeit a shaky foundation on which to build futureretirement security. As we will see in the next chapter,it inherited large unfunded liabilities from the SOEs,which means that contribution rates are high and imposea heavy burden on workers. Although benefits fortoday’s retirees are generous, the system’s relativelysmall coverage base, low rate of return on contributions,and lack of portability mean that it cannot possiblyprovide adequate benefits to tomorrow’s.

As for the countryside, formal retirement protection isvirtually nonexistent. Rural workers, including thoseemployed in manufacturing at town and villageenterprises (TVEs), are categorically excluded from thebasic pension system. There is a special rural pensionsystem that consists solely of personal accounts, butparticipation is voluntary and the benefits are tiny.Only 11 percent of rural workers now contribute tothe system, and the share that does has been falling,not rising. Beneficiaries on average receive a pensionof 85 yuan, or 12 dollars per month.

When we add up the public pension coveragenumbers, the bottom line is stark. All told, just 31percent of China’s total workforce, urban and ruralcombined, is now earning a public pension benefitof any kind. (See figure 9.)

0%

20%

40%

60%

10%

30%

50%

70%

Perc

ent

ofW

orkf

orce

Civil ServiceSystem 13%

Basic PensionSystem 52%

Rural System

65%

11%

UrbanWorkforce

RuralWorkforce

Chapter 1 11

Even in the cities, the public pensionsystem leaves large gaps in coverage.Percent of Chinese Urban and Rural WorkforceCovered by Public Pension Systems in 2007*

*Civil Service System data are for 2006.Source: MOHRSS (2008); and Sin and Mao (2007)

Figure 8

Overall, just three in ten Chinese workersare earning any public pension benefit.

Percent of Total ChineseWorkforce Coveredby Public PensionSystems in 2007*

*Civil Service System data are for 2006.Source: MOHRSS (2008); and Sin and Mao (2007)

Not CoveredBasic Pension SystemRural SystemCivil Service System

Figure 9

20%

69%

7% 5%

12 China’s Long March to Retirement Reform

Meanwhile, China’s private pension system is onlybeginning to take shape. Although employer-sponsored “enterprise annuities,” as China’s privatepensions are known, may eventually become asignificant source of retirement income, they nowcover just a tiny minority of workers, all of whomalready participate in the basic pension system. In2007, 9.3 million employees contributed to anenterprise annuity, or 3.2 percent of China’s urbanworkforce and 1.2 percent of its total workforce.(See figure 10.)

Social assistance provides some help to the most needyelders. Like all Chinese, the elderly are covered by theminimum living standard guarantee, a means-testedbenefit which tops up the income of qualifyinghouseholds to the poverty line. In addition, there haslong been a special program called the “three no’s” forrural elders who have no relatives, no income, and noability to work. More recently, the government hasadded a new benefit payable to poor rural elders whohave no sons. The poverty line in China, however, is setat such a low level—just 20 percent of per capita ruralincome and 16 percent of per capita urban income—that relatively few elders qualify for assistance.

Older workers have littleplace in China’s neweconomic order.

Continuing to work is an obvious way to maintainadequate income when pensions are lacking, but olderworkers seem to have little place in China’s neweconomic order. Elderly labor-force participation

remains relatively high in the countryside—76 percentamong men aged 60 to 64 and 38 percent past age 65.In the cities, however, it is very low by developing-country standards. Rates of labor-force participationfall precipitously beginning in the early 50s. Amongurban men aged 60 to 64, just 34 percent areemployed—and past age 65, a mere 13 percent are.

Part of the explanation for low levels of elderemployment lies in the history of SOE restructuring.Over the past decade, China has witnessed a masswave of early retirement, with SOEs offloading menas young as age 50 and women as young as age 40—a development known as the “40-50 phenomenon.”But longer-term economic and social trends are alsoundercutting the earnings potential of older workers.On the one hand, the low-wage agricultural andservice sectors where most elders are employed areshrinking. On the other, as Chinese industry movesup the global value-added scale, the speed with whichthe skills of the workforce become obsolete isaccelerating—rendering older workers unemployableeven as they are becoming more numerous.

One might suppose that a substantial share of elderswill be able to finance their retirement out ofpersonal savings. After all, Chinese savings rates arefamously high, in part because the social safety net isinadequate and workers understand that they mayhave to provide themselves for much of their ownsupport in old age.

The typical urbanhousehold possessesfinancial assets equal tojust one year’s income.

Yet here too, the outlook is worrisome. Mosthousehold savings is invested in the family home orthe family business, and may be difficult to cash outin retirement. Despite the recent development ofChina’s stock markets, the vast majority of financialsavings—over two-thirds—still languishes in bankdeposits that pay low nominal interest rates which,in recent years, have not even exceeded the inflationrate. As of 2002, the typical urban household inthe middle of the income distribution possessedfinancial assets equivalent to about one year’sincome—nowhere near what would be needed tofinance a retirement that may last 20 years or more.The circumstances of the typical rural household,with financial assets equivalent to less than sixmonths of income, are even more precarious.(See figure 11.)

0.0%1.0%2.0%3.0%4.0%5.0%6.0%

Perc

ent

ofW

orkf

orce 7.0% 6.9%

3.2%

EligibleWorkforce*

UrbanWorkforce

TotalWorkforce

1.2%

China’s private “enterprise annuity”system is still in its infancy.

Enterprise Annuity Coverage in China in 2007

*“Enterprise workers” contributing to the basic pension system.Source: MOHRSS (2008); and CSIS calculations

Figure 10

Chapter 1 13

To be sure, the wealth of younger generations is nowgrowing rapidly in China. In contrast to mostdeveloped countries, where household wealth peaksin the 50s or 60s, in China it now peaks in the late30s, suggesting that a powerful cohort effect mayimprove the economic circumstances of tomorrow’selderly.9 It will take decades, however, for this effectto translate into widespread retirement security—ifindeed it ever does. The social safety net for theyoung in China is just as inadequate as that for theold. Savings for retirement will thus have to competewith many other needs—especially education andhealth care, whose huge out-of-pocket cost routinelybankrupts families.

As things stand, tens of millions of elders retiring overthe next few decades will have little to fall back on butthe family. As in most East Asian societies, China’sConfucian ethic of filial piety requires children to carefor their aged parents. The majority of elders—59 percent in 2006—live with their grown childrenor other relatives.10 Even those who do not oftendepend on the extended family for financial support.

Yet as China modernizes, this pillar of old-agesupport is weakening. Multigenerational families,though still the norm, are less common than theywere a generation ago. Mass rural-urban migration isseparating children from their parents, disrupting thelifelong care networks that elders have traditionally

relied on for assistance. China’s floating populationalready numbers 150 million, and, according togovernment estimates, could grow by another 300million over the next two decades.11Meanwhile, asindividualistic or “Western” values gain currency,the ethic of filial piety is itself waning. In 1996, theNational People’s Congress passed a law obligatingchildren to support their aged parents—a sure signthat the family is in trouble. The government nowencourages rural families to draw up family supportagreements that specify how elderly members are tobe cared for. As of the end of 2005, more than13 million of these agreements had been signed.12Increasingly, parents are turning to the courts toenforce their rights. In 2005, rural elders filed morethan 2,000 suits in Beijing’s courts against migrantchildren who had defaulted on their filial obligations.13

China’s informal familysupport networks areweakening even beforethe age wave rolls in.

The rapid aging of China’s population threatens toplace a heavy new burden on China’s alreadyweakening families. As the number of children perfamily declines, it will become ever less likely that an

1.9

0.40.4

1.5

1.21.00.9

0.0

0.5

1.0

1.5

2.0

2.5

5Q4Q3Q2Q1Q

Ass

et/In

com

eR

atio

0.20.3

Rural Households

Urban Households

0.7

Relatively few elders will be able to count on personal savings to financetheir retirement years.Ratio of Financial Assets to Average Annual Income, by Quintile in 2002

Figure 11Source: CDRF and UNDP (2005); NBS (2007); and CSIS calculations

9 Urban Household Asset Survey [in Chinese], National Bureau of Statistics of China, 2002,http://www.stats.gov.cn/tjfx/ztfx/csjtccdc/index.htm.

10 China National Committee on Aging, Research Report on the Follow-Up Study of the State of the Elderly Population in Urban andRural China [in Chinese], December 17, 2007.

11 “300 Million Chinese Farmers to Move into Cities in Next 20 Years,” Xinhua News Agency, January 12, 2007.12 The State Council of China, The Development of China’s Undertakings for the Aged, December 2006.13 “Elderly Deserted by Own Families,” AFP, December 4, 2006.

14 China’s Long March to Retirement Reform

adult child or in-law will be available to care for anelder. According to demographer Xiaochun Qiao,rural women now turning 65 on average have 3.7surviving children, while urban women have 3.0.By 2025, rural women turning 65 will on averagehave 2.2 surviving children, while urban women willhave just 1.3.14 The Chinese call this the “4-2-1problem”—the prospect that one child will have tosupport two parents and four grandparents.

The ability of families to care for the elderly,especially the burgeoning number of “old old” aged80 and over, may be further strained by China’s largegender imbalance. Chinese families have alwaysexhibited a strong preference for sons, in partbecause, in Confucian culture, it is the son who isobligated to care for his parents in old age. In anormal population, there are roughly 105 baby boysborn for every 100 girls. In China in 2005, therewere 119.15 Alarmed by the trend, the governmenthas launched a major educational campaign topersuade parents that daughters and sons are equallyvaluable. As yet, however, the gender imbalanceshows no signs of narrowing.

China’s large genderimbalance will exacerbatethe challenge of caringfor the elderly.

This imbalance has ominous social implications.Today’s “missing girls,” as they are called in China,are already giving rise to large bride shortages andbachelor surpluses. By 2020, the governmentestimates that there will be 30 million surplus menof marriageable age.16 It is not hard to imagine thatthese growing legions of undomesticated males couldbecome a socially disruptive force, especially in asociety like China’s where the expectation ofmarriage is nearly universal. Even more worrisomethan the shortage of brides, however, is the loomingshortage of daughters-in-law. Ironically, the missingbrides of today will become the missing caregiversof tomorrow. After all, while it is the son who bearsresponsibility for caring for his aged parents inConfucian culture, it is the daughter-in-law whoactually does the caring.

A young China with a predominantly rural economyand strong extended families could get by with aretirement system that leaves large gaps in coverage.A rapidly developing and rapidly aging China withweakening families cannot. The Chinese peopleunderstand this. According to a recent survey ofurban residents, the overwhelming majority—87 percent—believe that society as a whole shouldassume the main responsibility for supporting theelderly.17 Yet 37 percent of urban elders and54 percent of rural elders report that the familyremains their primary source of support. Amongelders aged 80 and over, these shares rise to 60 and

0%

20%

40%

Aged 80 & OverAged 65-79Aged 60-64Aged 60 & Over

Perc

ent

ofEl

derl

y

37%

54%

27% 29%38%

61% 60%

88%

Urban

Rural

60%

80%

100%

A large share of Chinese still depend on the extended family for support in old age.Percent of Chinese Elderly Whose Primary Income Source Is the Extended Family, by Age andResidence in 2005

Figure 12Source: NBS (2006)

14 Xiaochun Qiao, “From Decline of Fertility to Transition of Age Structure: Aging and Its Policy Implications in China,”Genus 57, no. 1 (2001).15 National Population and Family Planning Commission of China, Research Report on the National Demographic Development

Strategy [in Chinese], January 11, 2007, http://www.chinapop.gov.cn/fzgh/zlyj/200806/t20080626_154129.htm.16 “China: 30 Million Men Face Bleak Future as Singles,” Xinhua News Agency, January 10, 2007.17 Raymond Zhang, “Views on Filial Piety See Change,” China Daily, December 7, 2004.

Chapter 1 15

88 percent, respectively.18 (See figure 12.) Thishuge disjuncture between the retirement systemthat China has and the system that it needs not onlythreatens the material security of the elderly, butalso increases the risks of social—and perhaps evenpolitical—crisis in an aging China.

The BroaderEconomic OutlookFor the past several decades, China’s demographicshave been leaning strongly with economic growth.When fertility first falls, societies enjoy a temporary“demographic dividend,” a period of time in whichthe number of children declines faster than thenumber of elderly grows and the share of the populationin the working years rises. A larger share of thepopulation in the working years, all other thingsbeing equal, means a higher per capita livingstandard. The falling dependency burden, togetherwith rising longevity, also tends to boost livingstandard growth by raising savings rates, encouraginginvestment in human capital, and freeing up adulttime, especially the time of women, for marketemployment. The demographic transition thusopens up a window of opportunity for countriesto boost economic growth.

China’s “demographicdividend” has helped tounderpin its stunningeconomic rise.

Economists who have studied the demographictransition agree that China’s unusually largedemographic dividend has underpinned its stunningeconomic rise. Between 1975 and 2005, the totaldependency ratio of children and elderly to working-age adults in China plummeted from 87 to 48,among the largest drops of any country in the world.Meanwhile, the share of the population in theworking years surged from 54 percent to 67 percent.(See figure 13.) Many studies have concluded thatbetween one-quarter and one-third of the totalgrowth in China’s per capita GDP since the mid-1970s can be attributed to the shift in the agestructure of its population.19

Beginning around2015, China’s favorabledemographics will bethrown into reverse.

18 2005 National 1% Sampling Survey, National Bureau of Statistics of China, March 16, 2006,http://www.stats.gov.cn/tjsj/ndsj/renkou/2005/renkou.htm.

19 See, for instance, Fang Cai, “Demographic Transition, Demographic Dividend, and the Sustainability of Economic Growth”[in Chinese], Population Research 128, no. 12 (2004); and Feng Wang and Andrew Mason, “Demographic Dividend and Prospectsfor Economic Development in China” (paper presented at the UN Expert Group Meeting on Social and Economic Implications ofChanging Population Age Structures, Mexico City, August 31–September 2, 2005).

102030405060708090

10087

75

6257 56 54

48 47 5054 59

6774 76 80

87

0

China’s period of “demographic dividend” is now coming to an end.Total Dependency Ratio of Children (Aged 0-14) and Elderly (Aged 60 & Over)per 100 Working-Age Adults (Aged 15-59), 1975-2050

1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Figure 13

Tota

lDep

ende

ncy

Rat

io

Source: UN (2007)

Working-AgePopulation Share1975200520302050

53.6%67.4%59.8%53.4%

16 China’s Long March to Retirement Reform

Beginning around 2015, however, China’s favorabledemographics will be thrown into reverse. As therelative growth in the number of elderly finallyovertakes the relative decline in the number ofchildren, the total dependency ratio will bottom outand once again begin to rise. At the same time, theworking-age population will peak and begin todecline. While China’s demographic transformationhas so far been leaning with economic growth, itwill soon be leaning against it.

A rising share ofworkers’ incomes willhave to be transferredto nonworking elders.

As China ages, a rising share of workers’ incomes willhave to be transferred to nonworking elders. In 2005,there were 6.1 working-age adults in China availableto support each elder. (See figure 14.) That ratio isdue to fall to 2.5 by 2030 and to just 1.6 by 2050—which means that the average burden that must beshouldered by each worker will nearly quadruple.As we have seen, much of the burden of supportingthe elderly is now handled informally throughfamilies, which explains why China spends just3 percent of GDP on public pensions. As China agesand develops, however, a larger share will inevitablyshow up in public budgets. If financed on a pay-as-you-go basis, a pension system that offered all workersthe same average benefit as today’s basic pension

system would cost at least 10 percent of GDP by 2030and at least 15 percent by 2050.20The rising cost ofhealth care for the elderly would come on top of this.

Along with its growing old-age dependency burden,the contraction of China’s working-age populationcould also become a drag on economic growth. Overthe three decades of China’s economic reform era,the expansion of its working-age population has onaverage added 1.8 percentage points per year to itsGDP growth rate. By the 2030s, the contraction ofits working-age population will be subtracting0.7 percentage points per year—a dramatic swingof 2.5 percentage points.

An aging China may face afuture of capital shortages.

An aging China may also face a future of capitalshortages. With China now awash in excess savings,running huge current account surpluses, andaccumulating massive foreign exchange reserves, thisprospect may seem remote. Yet numerous studieshave confirmed that the classic lifecycle motivationfor savings—accumulating assets during the workingyears to be drawn down during the retirementyears—functions far more powerfully in China thanin the developed countries, in part because the socialinsurance system is not yet well developed and in partbecause the Confucian ethic that traditionally allowedelders to rely on their children for support in old age is

20 This illustrative projection assumes, first, that the average pension benefit will remain unchanged (at its 2007 level) relative to GDP per worker;and second, that two-thirds of the future elderly will qualify for a pension benefit based on their employment and contribution history.

9.0

8.0

7.0

6.0

5.0

4.0

3.0

2.0

1.0

0.0

There will be far fewer working-age adults to support each elder.Chinese Aged Support Ratio of Working-Age Adults (Aged15-59) to Elderly (Aged 60 & Over),1975-2050

1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Figure 14

Age

dSu

ppor

tR

atio

Source: UN (2007)

7.7 7.7 7.8 7.67.0

6.56.1

5.4

4.43.8

3.12.5

2.1 2.0 1.8 1.6

weakening. This means that savings rates, which haverisen dramatically in recent decades as youth dependencyhas declined and the share of the population in theworking years has surged, could fall just as dramaticallyonce elder dependency begins to climb.

The decline in domestic savings could beaccompanied by a decline in the availability ofaffordable foreign capital.21Most of today’s richcountries face towering age waves that are peakingeven sooner than China’s. Most also have expensivepay-as-you-go pension and health benefit programswhose costs will climb steeply in coming decades.They too are therefore likely to experience a sharpdrop in savings rates, both private and public. ForChina, the result could be to curtail the massiveinflows of foreign direct investment that have playedsuch a central role in fueling its economic growth.

All of this means that China’s continued economicsuccess will hinge on its ability to efficiently allocatesavings to investment. Yet this is China’s greatesteconomic weakness. Although China’s labor marketsand product markets are now more liberal than thosein many capitalist economies, reform of its capitalmarkets has lagged. In effect, most of the capital inChina is still allocated by the government through thestate-owned banks. History has proven, time andagain, that no administrative process can approachthe efficiency of decentralized markets in allowinginvestment resources to seek out the highest return.Today in China many large favored enterprises areawash in investment funds despite poor expectedreturns and prospects, while many small unnoticedenterprises are starved for funding despite excellentexpected returns and prospects.

Over the past three decades, the underdevelopmentof China’s capital markets has not done much to hinderits economic performance—in part because of thelarge inflows of foreign direct investment, and in partbecause so much of its stellar gains in real GDP havebeen generated by the mass internal migration ofworkers from the country to the city. Underemployedworkers from nonmarket rural sectors who oncehardly participated in the national economy havebeen taking full-time, low-skilled manufacturing jobsthat are integrated with the global economy. With themigration comes a huge leap in the productivity ofthe workers, even when the firms offering the jobsare not accountable to financial markets (for example,SOEs engaged in primary production) or rely mainly onforeign capital (the so-called manufacturing platforms).

China’s developmentagenda will increasinglydepend on the vitalityof its capital markets.

Reliance on foreign direct investment, however,is a risky long-term strategy—and migrationcannot continue indefinitely as a major source ofproductivity growth. China is losing its competitiveadvantage in low-skilled manufacturing. As itsindustries move up the global value-added scale,a mismatch is emerging between the skills of itsremaining surplus rural labor and the demandsof the jobs being created in the growth sectorsof its economy. In the decades to come, gainsin productivity and living standards will dependincreasingly on workers with good jobs finding betterjobs—and on firms with established markets findinginnovative ways to win new markets. China’sdevelopment agenda will, in other words, dependincreasingly on the vitality of its capital markets.

The social stresses ofbreakneck developmentmay become less tolerablein an aging China.

The rapid pace of China’s economic developmentand the sweeping social changes that accompany ithave sometimes been likened to a speeding bicyclethat has to keep going just to keep from falling over.China’s aging increases the pressure. On the onehand, it makes rapid economic growth even moreessential, since workers will have to transfer a risingshare of their incomes to nonworking elders. On theother hand, it makes rapid growth more difficult toachieve. The social stresses of breakneck development,from widening income gaps to weakening families,have been bearable in a youthful China in whichincomes have been rising rapidly. They may becomeless tolerable in an aging China in which economicgrowth is slowing.

Throughout China’s long history, periods of strongcentral authority and growing economic prosperityhave alternated with periods of social and politicalchaos—or what the Chinese call luan. WhetherChina’s aging ushers in the next turn of the cyclemay well depend on how successful China is inmeeting its aging challenge.

Chapter 1 17

21 For a summary of the literature on demographics, savings, and global capital flows, see Richard Jackson and Neil Howe, The Graying of the GreatPowers: Demography and Geopolitics in the 21st Century (Washington, DC: Center for Strategic and International Studies, 2008), 97-108.

18 China’s Long March to Retirement Reform

Chapter 2 19

China’s retirement system, like many of its socialand economic institutions, is still evolving. Thegovernment began to lay the foundations of the“basic pension system” for urban workers in theearly 1990s. In 1997, it initiated a transition fromthe original pay-as-you-go plan to a two-tieredsystem consisting of a scaled-back pay-as-you-gobenefit and a personal retirement account. Sincethen, it has repeatedly adjusted the system’s design—testing reforms in provincial-level pilot projectsbefore introducing them nationally.

The government understands that a rapidlydeveloping and rapidly aging China needs a publicpension system capable of providing a decent levelof support to the old without imposing a crushingburden on the young. It understands that the systemmust be national in scope, and in November 2007 itset a goal of extending coverage to all Chinese, bothurban and rural, by 2020.22 It also understands thepotential advantages that a fully funded retirementsystem can confer in achieving these goals.

China’s basic pensionsystem suffers from seriousstructural problems.

Today’s basic pension system is supposed to be thefoundation on which tomorrow’s national systemwill ultimately be built. The basic pension system,however, suffers from serious structural problems thatmake it difficult for it to fulfill its mission. Althoughthe system is relatively new, it took over theobligations of preexisting SOE pension plans, andtherefore has large unfunded liabilities. Contributionrates are prohibitively high—generally 28 percent ofworkers’ taxable payroll—which encourages evasionand makes expanding coverage difficult. The systemoperates within central government guidelines, butcontributions are collected and benefits are paid bylocal social security bureaus at the municipal, district,or (in some cases) provincial level, which means thatpensions are not fully portable. Until recently,contributions to the personal account tier of the

system were routinely diverted to cover deficits inthe pay-as-you-go tier, leaving the accounts largelyunfunded. Even when contributions are saved,moreover, they earn far less than a market return,which means that the accounts cannot generateadequate replacement rates.

Over the past few years, the government has steppedup its efforts to build an adequate and sustainableretirement system. It has enacted a series of reformsthat begin to address the shortcomings of the basicpension system; it is building up a national pensionreserve fund; and it has launched a new system ofsupplementary private pensions. Although theseinitiatives have pushed China’s retirement systemin the right direction, more fundamental reform isrequired. In this chapter, we take a close look at theprogress that China has made in its long march toretirement reform. In the next chapter, we outline aplan that would allow the government to achieve itslong-term goals.

From Iron Rice Bowlto Empty AccountsTo understand the challenges facing China’s retirementsystem, we must begin by taking a look back at thehistory of market reform and SOE restructuring. Inthe days of China’s planned economy, most SOEworkers enjoyed cradle-to-grave social protection.Wages were low, but the state-owned enterprisesguaranteed lifetime employment along with socialbenefits, including pensions, health care, and housing.China’s “iron rice bowl” was based on the work unit,but backed up by the government, which subsidizedthe SOEs—and thus effectively guaranteed pensions.

The Evolution ofChina’s Retirement System

CHAPTER 2

22 Hu Jintao’s Report at the 17th National Congress of the Communist Party of China [in Chinese], October 15, 2007.

When China launchedits market reform in 1978,it broke the old SOE“iron rice bowl.”

When China launched its market reform in 1978,it broke the iron rice bowl. As each SOE assumedresponsibility for its own bottom line, the securityof its pension benefits came to depend on its financialstatus. Unprofitable enterprises delayed or canceledpension payments, throwing millions of retireesinto poverty.

Beginning in the early 1990s, the central governmentattempted to alleviate the crisis by establishing localsocial security bureaus to collect and “pool” SOEpension contributions. These pools, which ultimatelyformed the basis of today’s basic pension system,served two purposes. They facilitated SOErestructuring, since workers could now leaveunprofitable enterprises for profitable ones withoutlosing their accrued pension benefits—at least so longas they remained within the state-owned sector andthe local pool. They also improved retirement security,since pension payments no longer depended directlyon the solvency of a retiree’s former employer.

Pooling alone, however, was not enough to shoreup the pension system. As the state-owned sectordownsized, millions of younger workers left for theprivate sector, while millions of older workers wereforced into early retirement. In less than a decade

between 1989 and 1997, the basic pension system’ssupport ratio of contributing workers to retiredbeneficiaries plummeted from 5.4 to 3.4, far beneathChina’s demographic aged support ratio of working-age adults to elderly. (See figure 15.) In some heavilyindustrialized provinces, the system support ratiodropped even further. In Liaoning province inChina’s rust belt, it sank beneath 3.0, and inShanghai it sank beneath 2.5. Without a bailout,the system faced near-term financial collapse.

In 1997, the State Councilsettled on a blueprint for anew pension system forurban workers.

Meanwhile, China’s leaders were beginning to focuson the long-term aging challenge. During the mid-1990s, the central government consulted extensivelywith foreign pension experts, including teams fromthe Asian Development Bank and the World Bank.It also encouraged the provinces to stage pensionreform pilot projects. In 1997, the State Council,China’s highest executive body, settled on a blueprintfor a new pension system that seemed to addressboth the near-term and long-term challenges.

The 1997 reform for the first time extended coverageunder the basic pension system to private-sectorworkers. At the same time, it replaced the existingpay-as-you-go pension system with a new two-tiered

The demographics of China’s basic pension system are much older thanthose of the nation as a whole.Aged Support Ratio of Working-Age Adults (Aged 15-59) to Elderly (Aged 60 & Over) versusSystem Support Ratio of Covered Workers to Retirees in the Basic Pension System, 1989-2007

Source: MOHRSS (2008); and NBS (2007) Figure 15

Supp

ort

Rat

ios

Aged Support Ratio System Support Ratio

20 China’s Long March to Retirement Reform

7.7 7.5 7.2 7.0 6.7 6.5 6.4 6.3 6.1 5.95.4 5.2

4.43.9

3.4 3.2 3.0 3.0 3.1

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

1989 1991 1993 1995 1997 1999 2001 2003 2005

3.2

2007

Chapter 2 21

system consisting of a scaled-back pay-as-you-go(or social pool) benefit and a personal retirementaccount. Under the reform, the “old men,” those whowere already retired in 1997, continued to receivepre-reform benefits that typically replaced about 80percent of wages. The “new men,” those who joinedafter 1997, were promised a considerably lessgenerous 60 percent replacement rate, consistingof a social pool benefit that replaced 20 percentof wages and a personal account benefit that thegovernment estimated would replace another40 percent. The “middle men,” those who werecontributing workers in 1997, were promised prorata benefits based on years spent participating in theold and new systems. Although the generosity of thepension system was reduced for future retirees,retirement ages remained low: just 60 for menand 55 for women.

Because the new system took over the unfundedliabilities of the financially troubled SOE system, itscontribution rate had to be set at a high level. Thecentral government’s guidelines called for a totalcontribution rate of 24 percent of taxable payroll.Of this, 13 percent of payroll (contributed byemployers) was intended for the first-tier and11 percent (3 percent contributed by employersand 8 percent by employees) for the personalaccounts. In many provinces, the actual contributionrate was even higher.

Nearly half of contributionsto the basic pension systemsimply go to defray the legacycost of pensions under theold SOE system.

The reform was clearly an effort to help the SOEs bybringing new contributors into the pension system.The newly covered private-sector workers, after all,were disproportionately young men and women withyears of contributions to make before they couldclaim a benefit. Even the personal accounts, withtheir promise of personally owned savings, wereintended to attract new contributors. Yet the reformalso represented a major step toward addressingChina’s looming old-age dependency challenge. Theprivate sector, which was nonexistent in 1978, nowaccounts for over half of urban employment and aneven larger share of urban output. Excluding it fromthe pension system made no long-term policy sense.

At the same time, the personal accounts wereexpected to take pressure off government budgetsas China aged.

It soon became apparent, however, that the reformwas not working out as expected. Coverage underthe basic pension system failed to rise, which meantthat its support ratio of contributors to beneficiariescontinued to deteriorate. Private enterprises andworkers had little incentive to join the system, sincea large share of their contributions—according to theWorld Bank, nearly half 23—simply go to defray thelegacy cost of prior pension promises to SOE retirees.In effect, the government was asking the new menfrom the private sector to finance two retirements:their own and those of the participants in thepre-reform system, with its more generous benefits.The new men, for their part, were refusing to join.

The basic pensionsystem’s high contributionrate has made it difficultto expand coverage.

The failure to increase coverage meant that it wasimpossible to fund the personal accounts. Sincesocial pool contributions in most provinces wereinsufficient to cover current benefit payments, thelocal social security bureaus diverted personalaccount contributions to cover the shortfall. Therewas no legal or even procedural obstacle to doing so,

23 China: Evaluation of the Liaoning Social Security Pilot, Report no. 38183-CN (Washington, DC: The World Bank, December 2006).

since contributions to both tiers of the pension systemwere deposited in the same social security bureaubank accounts. In many provinces, the social pooldeficits exceeded the combined contributions to bothtiers of the pension system, which meant that allpersonal account contributions were diverted—whilestill leaving a cash shortfall that had to be plugged bycentral government subsidies.

The personal retirementaccounts set up by the1997 reform turned outto be entirely unfunded.

To be sure, personal account contributions werestill credited to workers’ accounts. The accounts,however, were purely “notional”—or, as the Chinesemedia came to describe them, “empty.” Like the U.S.Social Security trust fund, they represented a claim onfuture tax revenues, not real assets that could financefuture retirement benefits. The administrativelydetermined rate of return on the accounts, moreover,was far too low to generate the promised 40 percentreplacement rate. As a rule, the social security bureausset the return equal to the bank interest rate on one-year term deposits, which in real terms averaged 1.7percent between 1997 and 2007, far beneath the rateof real wage growth. Since 2000, the bank interestrate has averaged just 0.5 percent. Even if we generouslyassume a 2.0 percent real rate of return, the personalaccounts would in reality generate replacement ratesof no more than 15 to20 percent for full-careerworkers. (See figure 16.) Rather than the promised 60percent, replacement rates for both tiers of the systemcombined were unlikely to be more than 35 to 40percent. Understandably, workers came to viewtheir personal account contributions as a tax, whichfurther exacerbated the evasion problem.

Along with its high contribution rates and low returns,the system’s limited portability also discouragedpotential new joiners. Workers must contribute to thebasic pension system for 15 years before they qualifyfor a first-tier benefit, and even then the benefit isforfeit if they leave their social pool. Although personalaccount balances vest immediately, there is no provisionfor transferring them between social security bureausin different social pools. If workers move, their onlyoption is to request a lump-sum payout from theirsocial security bureau, which, to the extent the accountis unfunded, must draw on the social pool to comeup with the cash.

Liaoning and BeyondIn the decade since its landmark 1997 reform, thegovernment has launched a wide range of reforminitiatives designed to strengthen China’s retirementsystem. Although the government has made significantprogress, fundamental problems remain.

Although recent governmentreforms have helped tostrengthen China’s retirementsystem, fundamentalproblems remain.

The new round of reforms began with the 2001–2004 Liaoning province pilot project, which testeda plan to bolster the basic pension system’s financesand allow its empty personal accounts to be saved.The plan involved two steps, the first being toeliminate the social pool deficits. The governmentbegan by raising the basic pension system’s first-tiercontribution rate and downsizing its personalaccounts. Specifically, the first-tier contribution ratewas raised from 13 to 20 percent (still paid entirelyby employers), while the personal accountcontribution rate was lowered from 11 to 8 percent(now paid entirely by employees). The overallcontribution rate for both tiers was thus increasedfrom 24 to 28 percent. At the same time, the

22 China’s Long March to Retirement Reform

China’s current personal accounts systemcannot possibly deliver on its promises.Basic Pension System Personal Accounts:Government Target Replacement Rate versusPotential Replacement Rates under Different RealWage Growth Assumptions*

Figure 16

*Contribution rates are 11 percent for original 1997 system and8 percent for post-Liaoning system. Projections assume a 2.0 percentreal rate of return, a 40-year career, and a retirement age of 60.Source: CSIS calculations

40%

0%

5%

10%

15%

20%

25%

30%

35%

45%

30%

21%17% 15%

12%

GovernmentTarget

Rep

lace

men

tR

ate

3.0% Real WageGrowth Rate

5.0% Real WageGrowth Rate

Original 1997 System

Post-Liaoning System

Chapter 2 23

government provided for increased subsidies toclose any remaining deficit in the social pools. Thesecond step was to create a “firewall” between thepersonal accounts and the social pools. Theresponsibility for collecting contributions wastransferred from the local social security bureaus tothe local tax bureaus, which deposit contributionsearmarked for each tier in separate accounts atdifferent banks. The social security bureaus thendraw on the accounts to cover first-tier benefitpayments and personal account withdrawals.

Between 2004 and 2006, the new framework testedin the Liaoning pilot project, including the revisedcontribution rates and personal accounts firewall,was in principle extended to the rest of the country.However, in some highly industrialized provinceswith low system support ratios, as well as some poorprovinces, the State Council concluded that fullyfunding even the downsized personal accounts mightnot be a feasible near-term goal. These provincesinclude Heilongjiang, Henan, Hubei, Hunan, Jilin,Shandong, Shanghai, Shanxi, Tianjin, and Xinjiang.Here the central government and provincialauthorities signed explicit subsidy agreementsdesigned to ensure that at least 5 percent of the8 percent personal account contribution is saved.

While the personalaccounts are now partiallyfunded, they earn too low areturn to generate targetedreplacement rates.

The “reform of the 1997 reform” has succeeded inraising the basic pension system’s overall balance andreducing the share of personal account contributionsthat is diverted to pay first-tier benefits. The reform,however, did nothing to improve the dismal rate ofreturn on the accounts. Although the personalaccounts are now partially funded, centralgovernment policy requires that the local tax andsocial security bureaus invest saved contributionsalmost exclusively in bank deposits. The interest onthese deposits may be somewhat higher than the one-year term deposit rate, but the personal accounts aregenerally credited with the latter just as they alwayshave been.24Moreover, because the personal account

contribution rate was reduced from 11 to 8 percent,future personal account replacement rates (again, seefigure 16) will be even lower than before.

It is true that some workers will now earn a largerfirst-tier benefit. Under the original 1997 reform,the minimum and maximum first-tier replacementrates were the same. Workers earned a 20 percentreplacement rate after 15 years of contributions, butreceived no further credit for additional contributionyears. Under the new post-Liaoning benefit rules,workers will need to contribute to the system for 20years to earn a 20 percent replacement rate, but theirreplacement rate will continue to rise by 1 percent foreach additional year of contributions. The potentiallyhigher first-tier replacement rate, however, comes atthe cost of a higher first-tier contribution rate.

The National Pension Fundwill be able to defray onlya small fraction of thebasic pension system’sunfunded liabilities.

As part of its efforts to strengthen the basic pensionsystem’s finances, the government in 2000 establisheda reserve fund known as the National Pension Fund,or NPF. The NPF, which is supposed to serve as a“fund of last resort” when China’s age wave beginsto push up pension costs, is financed through acombination of central government subsidies,revenue raised from the sale of shares in state-ownedenterprises, and the profits of China’s national lottery.Initially, NPF assets were invested almost entirely ingovernment bonds and bank deposits. However,the fund’s investment guidelines have been graduallyliberalized as the government has come to view it as apotentially valuable policy tool for bolstering strategicindustries and developing capital markets. As of theend of 2007, 47 percent of NPF assets wereoutsourced to external fund managers, who handlethe fund’s growing investments in domestic andinternational securities. Although the NPF isincreasing in size, its total assets still amount to just2.1 percent of GDP—a tiny fraction of the basicpension system’s estimated unfunded liabilities ofsome 120 to 140 percent of GDP.25

24 See Bingwen Zheng, “The Origin of China’s Partially Funded Social Security Scheme and Its Future Direction,” The ChineseEconomy 40, no. 4 (July/August 2007); and Bingwen Zheng, “Assessment for Social Security Pilot in Jilin and Heilongjiang ofChina,” China and World Economy 14, no. 5 (2006).

25 Yvonne Sin, “China: Pension Liabilities and Reform Options for Old-Age Insurance,” World Bank Working Paper Series no. 2005-1(Washington, DC: The World Bank, May 2005), 33.

24 China’s Long March to Retirement Reform

Since 2006, the central government has encouragedprovincial governments to entrust the managementof personal account assets to the National PensionFund. The policy was in part a reaction to thatyear’s pension fund scandal in Shanghai, where thediversion of personal account assets into speculativereal estate investments that benefited the party chief ’sfamily members and close associates shook confidencein the ability of local governments to manage pensionfunds. The NPF now manages a portion of thepersonal account assets of nine provinces—or moreprecisely, the central government subsidies that ineffect fund the personal accounts. As is typicallythe case with locally managed personal accounts,the accounts managed by the NPF are credited withless than actual investment returns—in this case,far less. In the agreement between the NPF and theprovinces, which runs from 2006 through 2010, theaccounts are to be credited with interest calculatedat a contractually fixed 3.5 percent nominal rate—which means that, thus far, the real rate of returnearned by the accounts has actually been negative.Meanwhile, in 2006 and 2007, which were boomyears for China’s stock market, the NPF earned areal return of 20 percent on its investment portfolio.

The government has begunto offer migrants and otherinformal-sector workersspecial incentives to jointhe basic pension system.

The central government is also trying to increasecoverage under the basic pension system byencouraging local authorities to offer informal-sectorworkers special incentives to join. In some provincesand municipalities, migrants, self-employed workers,and even workers at small and medium enterprisesare being allowed to contribute to the system at alower 18 percent rate (10 percent to the social pooland 8 percent to a personal account), while earningthe same benefits as those who contribute at the 28percent rate.26 The contribution discounts appear tobe having some impact. After hovering between 45and 46 percent of the urban workforce between 2000and 2004, coverage under the basic pension systemrose to 52 percent in 2007.

China’s new 2008 labor law, which requires allemployers and employees to sign labor contracts,may further raise coverage by improving thegovernment’s enforcement capabilities. It is doubtful,however, whether a large increase beyond today’slevel is feasible without more fundamental reform ofthe pension system itself. Even with the special newdeals for informal-sector workers, the system stilloffers a poor payback on contributions, which meansthat many enterprises and workers will continue tofind ways to evade participation. This is especiallytrue for migrants, only a tiny fraction of whom willever qualify for a first-tier benefit. Most are alsolikely to cash in their personal accounts long beforethey reach old age, which means that their savings“leaks” from the system and contributes little ornothing to retirement security. The Chinese mediarecently coined the term “cash-in phenomenon”(tuibao) to describe the long lines of migrantswaiting in front of local social security bureaus tocollect their account balances before returning homeat the end of the year.27 In view of the obstacles tointegrating migrants into the two-tiered basicpension system, the government is studying thefeasibility of creating a separate stand-alone personalaccounts system for migrants whose benefits, at leastin principle, would be portable.28

To be sure, the government has made someprogress in improving portability for all workersby encouraging local governments to consolidatemunicipal and district social pools at the provinciallevel. The progress, however, has been slow. Outof China’s 31 provinces, just 13 (including Beijing,Shanghai, and two other cities that are administrativelytreated as provinces) had achieved provincial-levelpooling as of the end of 2007. National-level pooling,which is essential to ensure both worker mobility andretirement security in a rapidly modernizing China,remains a distant goal.

National-level poolingof pension contributions,and hence full portability,remains a distant goal.

26 The World Bank, Liaoning Social Security Pilot, 27.27 “Social Insurance Policy Has Become Worthless: A Wave of Migrant Workers Resign and Cash-in Their Pension” [in Chinese],

People’s Daily, January 8, 2008; and “Another Wave of Pension Cash-ins Emerges: Non-Transferable Pensions Have Become ‘LocalFood Coupons’?” [in Chinese], South Yangtze Times, February 27, 2008.

28 “China to Cover 200m Migrant Workers with Portable Pension System,” ChinaStakes.com, August 11, 2008; and Shan He,“A Large Social Safety Net Spun for 1.3 Billion,” China.org.cn, October 5, 2008.

Chapter 2 25

In addition to reforming the public pension system,the government launched a new voluntary system ofprivate employer-sponsored enterprise annuities, orEAs, in 2004—or more precisely, it restructured andrebaptized the small existing system. Any firm thatcontributes regularly to the basic pension system canset up a plan, and any of the firm’s employees canchoose to participate. The scheme, however, is primarilyintended to provide supplementary retirement incometo higher earners with wages over the basic pensionsystem’s contributable wage ceiling. Since this ceiling isrelatively low (300 percent of the local average wage),higher earners’ basic pension benefits may replace nomore than a tiny fraction of their total wages.

Enterprise annuities are defined-contribution planssimilar to U.S. 401(k)s. Annual contributions, whichare split evenly between employers and employees,are capped at one-sixth of the previous year’s wagebill for all employees. Benefits can be paid eitheras a lump sum or in monthly installments when anemployee reaches the minimum retirement age of 60for men and 55 for women, the same eligibility agesfor basic pension benefits. Although so-called legacyEAs set up before 2004 were often managed by thelocal social security bureaus, following the Shanghaipension scandal the government mandated that allassets be transferred to licensed independent fundmanagers—a process that is still underway.

China’s new system ofprivate enterprise annuitiesfaces considerable obstaclesto growth.

The new enterprise annuity system meets an importantneed, but also faces considerable obstacles to growth.The setup costs are high, since trustee, custodian,administrator, and fund manager functions must beperformed by separate financial institutions. Each EAservice provider must obtain approval from its ownsupervisory agency—a bank from the China BankingRegulatory Commission, an insurer from the ChinaInsurance Regulatory Commission, and an investmentfirm from the China Securities Regulatory Commission.Each must then obtain a license from the Ministryof Human Resources and Social Security. The taxtreatment of EAs is also inadequate, haphazard, andill-defined. Only a fraction of employer contributionsare tax deductible, and the amount varies fromprovince to province. There are as yet no cleargovernment guidelines on the tax treatment ofemployee contributions, investment earnings, and

withdrawals. Investment rules are highly restrictive,forcing fund managers to overweight portfolios withlow-return fixed-income assets. There is no provisionfor master trust arrangements, which allow small andmedium enterprises to pool contributions in the sameplan, thus economizing on setup and maintenancecosts. All of this helps to explain the low coveragerates noted in the previous chapter.

The Unfinished AgendaChina is clearly better prepared for the aging ofits population than it was just a few years ago.The Liaoning pilot project and its subsequentextension to the rest of China, the establishmentof the National Pension Fund, and the launch ofthe new enterprise annuity system are all helpingto increase funded retirement savings. At the sametime, the government has begun to increase subsidiesto the basic pension system, thus assuming theburden for at least a part of its unfunded liabilities.Coverage under the system has also begun to rise,and its benefits are gradually becoming more portableas pooling is consolidated at the provincial level.

The recent reforms, however, fall well short ofa complete solution. The basic pension system’spersonal accounts, though now partially funded,still earn far less than a market rate of return. Theincrease in central government subsidies, though astep in the right direction, has been too small to allowthe government to lower the system’s prohibitivecontribution rate. Indeed, the contribution rateis higher today than it was in 1997. Despite theprogress toward provincial-level pooling, tens ofmillions of workers still face a stark trade-offbetween job mobility and retirement security.Pension coverage remains far from universal,even in the cities. Meanwhile, the governmenthas yet to take serious steps toward realizing itspromised extension of coverage to the countryside.

Successfully confronting theaging challenge will require amore radical restructuring ofChina’s retirement system.

In short, though the outlook for retirement securityhas begun to improve, the government’s incrementalapproach to reform appears to be approaching itslimits. To successfully confront the aging challenge,it will have to undertake a more radical restructuringof the retirement system.

Chapter 3 27

The challenge of building a national pension systemin China could hardly be more daunting. China facesthe same obstacles to ensuring universal coverage thatother developing countries do: a large informalsector, high levels of self-employment, and limitedgovernment enforcement capabilities. But in addition,China faces unique hurdles that greatly complicatethe challenge. It is aging much faster than mostdeveloping countries are, which means that it has lesstime to prepare. And, as a legacy of the old iron ricebowl, its basic pension system has large unfundedliabilities that render it unsustainable.

The CSIS reform plan wouldrequire a significant up-frontinvestment, but would bringvast benefits as China ages.

To meet the challenge, CSIS proposes a four-stepreform plan. The first step is to create a universalfloor of protection against poverty in old age thatwould cover all Chinese, whether or not they havecontributed to the basic pension system. Thesecond step is to lower the basic pension system’scontribution rate by socializing the cost of itsunfunded liabilities—that is, by having the centralgovernment directly assume the burden. The legacycost of the SOEs is a collective problem, and shouldbe paid out of general government tax revenues, notworkers’ payroll contributions. An aging China canno longer afford to hold retirement reform hostageto SOE restructuring. The third step is to graduallytransform today’s two-tiered basic pension systeminto a national system of publicly regulated butprivately managed and invested personal accounts.Unlike today’s personal accounts, the reformedaccounts would be fully funded, fully portable, andoffer participants a market rate of return. The fourthstep is to expand supplementary retirement coverageunder China’s new enterprise annuity system.

The reform we propose would require a significantup-front investment, but would cost far less in thelong term than today’s system and would bringvast benefits as China ages. The floor of protectionwould spare tens of millions of elders from a destituteold age—and might even avert widespread social and

political unrest in an aging China. The nationalsystem of fully funded personal accounts would, overtime, allow the majority of tomorrow’s growingelderly population to enjoy a comfortable retirementwithout overburdening tomorrow’s workers. It wouldalso help an aging China to meet its long-termdevelopment goals by broadening and deepeningcapital markets and maintaining adequate rates ofsavings and investment. Along the way, it would turnmost workers into property-owning stakeholders inChina’s continued economic rise.

The government would have to overcome considerableobstacles to build the floor of protection and create agenuinely funded personal accounts system, from thelack of administrative infrastructure to the immaturityof China’s capital markets. But the reform is feasible,affordable, and offers great advantages over thealternatives.

Building the Floorof ProtectionThe foundation of the retirement system mustbe a universal floor of protection against povertyin old age. In developed countries, this floor is oftenprovided by redistributive benefit rules built into theregular contributory public pension system. In China,however, it is nearly impossible for a contributorysystem to achieve universal coverage. China is by nomeans unique in this regard. Developing countriesaround the world must grapple with the sameproblem: Workers in the informal sector often fail tocontribute to the public pension system—and evenwhen they do contribute, they do so irregularly,which means that the ultimate benefits they earn maybe inadequate.

The foundation of theretirement system must be auniversal floor of protectionagainst poverty in old age.

What China needs is a general-revenue financed floorof old-age income support—or what is sometimescalled a “social pension” or “zero tier.” This floor ofprotection, which would be jointly financed by the

A New Direction forRetirement Reform

CHAPTER 3

28 China’s Long March to Retirement Reform

central and provincial governments, would guaranteea minimum level of income to all Chinese elders,regardless of their employment history or contributionrecord. The eligibility age would initially be set at 60,but might be raised over time. Given the hugeregional disparities in living standards across China,benefits would have to vary depending on residence.A reasonable solution might be to peg the floor to20 percent of the average local wage, which untilrecently was the fixed replacement rate for the firsttier of China’s basic pension system. Twenty percentof the average local wage is somewhat higher thanthe government’s minimum living standard guaranteefor urban workers and almost precisely equal to theWorld Bank’s dollar-a-day threshold of abject povertyfor rural workers.

The poverty floor wouldprovide larger benefits andhelp far more elders than thecurrent minimum livingstandard guarantee.

The floor of protection would be means-tested tokeep its cost manageable. Benefits, however, wouldbe phased out gradually as incomes rise in order tominimize disincentives to participate in thecontributory pension system. Elders aged 60 and overwho have no other source of income would beeligible for a full zero-tier benefit equal to 20 percentof the average local wage. The zero-tier benefit wouldbe reduced by 1 yuan for every 5 yuan in otherhousehold income, so that it falls to zero for elderswith incomes equal to the average local wage. Thisgradual phase-out means that our floor of protectionwould provide larger benefits and help far moreelders than the current minimum living standardguarantee, which merely tops up income to thepoverty threshold.

In principle, the zero tier could be a universal flatbenefit—that is, the same fixed yuan amount for allelders. Some pension experts argue that flat-benefitpoverty backstops are superior to means-tested ones,since they avoid the incentive problems inherent inany means-tested system. But in fact, even a flatbenefit might create disincentives to participate in thecontributory pension system, especially for workerswith wages near the poverty line. In any case, a flatbenefit would be much more expensive. We estimatethat a universal flat benefit for all elders aged 60 andover equal to 20 percent of the local average wagewould cost roughly 3 percent of GDP today and at

least double that by 2030. The means-tested benefitthat we propose could provide the same level ofpoverty protection at about one-third of the near-term cost—1 percent of GDP—and an even smallerfraction of the long-term cost if our restructuring ofthe basic pension system encourages higherparticipation rates. Even this more economicalapproach, of course, represents an extra burden onthe budget. The floor of protection would thereforebe phased in gradually between 2010 and 2020.

Increasing FundedRetirement SavingsBeyond a broad floor of old-age poverty protection,China needs a contributory pension system that, overtime, will allow most elders to enjoy somethingapproaching their preretirement standard of living.The current basic pension system is supposed toperform this function, but as currently structured itcannot fulfill its mission. Its coverage remains uneven,its benefits are not fully portable, and its personalaccounts earn far less than a market rate of return.Yet despite the system’s inadequate benefits, it placesa heavy burden on workers that can only grow asChina ages.

Above the floor ofprotection, China will haveto rely more heavily onfunded pensions.

Financing a more adequate national pension systemon a pay-as-you-go basis is not a viable long-termoption. An aging China will have to rely more heavilyon fully funded pensions that allow elders to financea larger share of their own retirement incomethrough savings set aside during the working years.The choice is simple: Does China want to pay for itsage wave by imposing a rising tax burden on low-wage workers? Or does China want to pay for it outof the new wealth that a funded retirement systemwill create?

We believe that the best solution is to transform thecurrent basic pension system into a national system ofpublicly regulated but privately managed and investedpersonal accounts. The first pay-as-you-go tier of thebasic pension system would be phased out, since itsprimary purpose—providing a secure base of old-ageincome support—can be accomplished moreefficiently through our plan’s new means-tested floorof protection. Meanwhile, the second personalaccount tier of the system would be enlarged.

Chapter 3 29

When the transition is complete, the totalcontribution rate for the new pension system wouldbe 18 percent of covered payroll, roughly one-thirdless than the current 28 percent rate. Of this, 16percent would flow to the personal accounts tofinance retiree and aged survivors benefits. Theremaining 2 percent would be earmarked forpurchasing insurance to cover disability and youngsurvivors benefits. Unlike benefits to retirees and agedsurvivors, these benefits necessarily require risk-pooling and cannot be financed through asavings-based personal accounts system.

The best solution maybe to transform the basicpension system into anational system of fullyfunded personal accounts.

The transition to the new system would be financedas follows. Central government subsidies to the localsocial security bureaus, which now cover about 15percent of current basic pension benefits, would begradually increased between 2010 and 2020 untilthey cover 100 percent of current benefits, except forfunded payouts from personal accounts. Workerswould continue to contribute 8 percent of wages totheir personal accounts, just as they do today. As thegovernment subsidies grow, the current 20 percentemployer contribution rate would be graduallyreduced to 10 percent, of which 8 percent wouldflow to the personal accounts, leaving 2 percent tofinance disability and young survivors benefits.Current retirees would continue to receive benefitsthrough their local social security bureaus. Followingthe example of Chile, Mexico, and a number of othercountries, the accrued benefits of workers who havenot yet retired—or perhaps just of workers youngerthan a certain age—would be credited to their personalaccounts in the form of interest-earning government“recognition bonds.” The advantage of this approachis that it avoids a lengthy transition lasting 75 yearsor more for a residual pay-as-you-go system thateventually would be paying out modest benefits.

The transition cost wouldbe a manageable burden.

While the transition cost would be a burden, theburden would be manageable. We estimate that the

extra central government subsidies to the basicpension system would amount to between 2.0 and2.5 percent of GDP a decade from now when theyare fully phased in. Adding in the new poverty floor,the total cost of the reform would come to between3.0 and 3.5 percent of GDP at its peak. Although thissum is not trivial, it is not an impossible burden foran economy that is growing at 10 percent per year.By way of comparison, the $586 billion fiscal stimuluspackage that the Chinese government announced inNovember 2008 is equivalent to 14 percent of China’sGDP. The transition cost, moreover, would be rapidlydeclining by the 2020s as current beneficiaries andworkers who have accrued substantial benefits underthe old system begin to die off.

The CSIS plan wouldrealize the government’svision of achieving universalcoverage by 2020.

Our plan would not only transform the basic pensionsystem into a new system of privately managed andinvested personal accounts, it would extend thesystem to the countryside. Coverage would be madeimmediately mandatory for wage and salary workersat town and village enterprises, or TVEs. Thecombined employer-employee contribution ratewould begin at 8 percent and be increased by 1percent per year until it reaches the full 18 percentrate. To minimize what would be a substantial newburden on TVE employers and employees, the centralgovernment might consider requiring local governmentsto subsidize part of their contributions.29 Non-TVErural workers, most of whom are farmers, wouldinitially be encouraged to participate in the system ona voluntary basis—just as some now do in the existingrural pension system, which our personal accountssystem would replace. By 2020, however, coveragewould be made mandatory for these workers as well,perhaps at a lower or subsidized contribution rate.With their incorporation into the new nationalpension system, the government’s vision of achievinguniversal coverage by 2020 would be realized.

The new national system of mandatory personalretirement accounts would, over time, ensure anadequate retirement income to the majority ofChinese workers. Higher-earning workers, however,will need to accumulate supplementary retirementsavings. It is thus crucial that China also strengthen

29 The central government might also consider requiring local urban governments to similarly subsidize a part of the contributions ofmigrants and other low-income informal-sector workers.

30 China’s Long March to Retirement Reform

incentives for employers and employees to participatein the new private enterprise annuity system.

The enterprise annuitysystem could become animportant source ofsupplementary retirementincome for China’smiddle class.

The place to start is to streamline the burdensomelicensing process and to simplify the system’sstructure by allowing trustee, custodian, andadministrator functions to be combined, as theyfrequently are in developed-country pension systems.The government will also need to improve, clarify,and standardize the system’s tax treatment. The usualpractice in developed countries is to tax pensionsavings when it is withdrawn upon retirement, but tomake contributions (both employer and employee)fully tax-deductible and to allow investment earningsto accumulate tax-free. The government will alsoneed to liberalize investment rules and authorizemaster trust arrangements that ease the burden onsmall and medium employers. Industry experts agreethat the enterprise annuity system has considerablepotential for growth and that, with the rightregulatory and tax rules, it could become animportant source of supplementary retirementincome for China’s emerging middle class.30

Designing a NationalPersonalAccounts SystemLet us be clear: Our plan to restructure the basicpension system is not a plan to privatize it. Assets inthe personal accounts would be personally ownedand privately managed and invested. The system,however, would remain a public social insuranceprogram regulated and supervised by government.

Funded personal accounts systems can be structuredin a variety of ways. One option is the “providentfund,” or Singapore model, in which personalaccount assets are centrally managed and invested bygovernment. Following this model, China couldestablish a personal account for each worker withinthe National Pension Fund. The NPF would both

administer the accounts and manage their investment,either directly or by contracting with certified fundmanagers. The advantage of this approach is that itbuilds directly on recent pension pilot projects which,as we have seen, have already shifted the managementof a portion of personal account assets to the NPF.It would thus be relatively easy to implement. Thedisadvantage is that a centrally managed systeminvites political interference, rarely generates marketreturns for contributors, and creates considerableambiguity about who really owns the assets—workers or government.

Independent managementof personal account assetsis essential for thesuccess of reform.

We believe that independent management of personalaccount assets is essential for the long-term success ofreform. Here there are two basic models that Chinacould follow: the Chilean or the Swedish. In aChilean-model reform, workers would choose amongcompeting personal accounts (or PA) managementcompanies set up by participating financial servicesfirms. Contributions would be routed directly tothe management companies, which would bothadminister and invest workers’ accounts. In aSwedish-model reform, workers would choose amonga menu of investment funds. Contributions would berouted to the funds through a central government“clearinghouse,” which would handle accountadministration. Each model has its pros and cons thatthe government would need to carefully weigh. Thecentralized administration of the Swedish modeloffers important cost efficiencies, but also requiresmore sophisticated data management. Thedecentralized Chilean model sacrifices theseefficiencies, but would be easier to set up—which iswhy it has been adopted by many developingcountries in Latin America and Eastern Europe.31

Beyond the choice of administrative model for thepersonal accounts system, the government will haveto address many other difficult design issues. In thefollowing bullets, we describe what the basicarchitecture of a workable system might look like.

30 China’s Long March to Retirement Reform

30 Yvonne Sin and Leslie Mao, “Hidden Pot of Gold: Responding to China’s Pension Burden,” CLSA-U Speaker Series (Shanghai: CLSA,October 2007); and Donald R. Forest, “The Promise of Enterprise Annuities,” China Business Review (November/December 2007).

31 Hong Kong’s Mandatory Provident Fund scheme, or MPF, resembles the Chilean model in many respects. Although the term“provident fund” typically refers to a publicly managed scheme, the MPF personal retirement accounts, like the Chilean accounts,are administered and invested by private pension management companies, known as “trustees” in Hong Kong. The most importantdifference is that in Hong Kong employers select the management company, whereas in the Chilean model workers do.

32 China’s Long March to Retirement Reform

• Pension regulator. China will need a stronggovernment pension regulator to supervise theoperation of the personal accounts system. Ideally,the regulator would be constituted as an independentagency. The agency would have its own professionalstaff, but would draw on the expertise of the Ministryof Finance, the Ministry of Human Resources andSocial Security, the People’s Bank of China, and theChina Securities Regulatory Commission, InsuranceRegulatory Commission, and Banking RegulatoryCommission. The pension regulator would havebroad functions. It would be responsible for certifyingfund managers; for establishing guidelines forportfolio allocation, fee structure, and reportingrules; and for policing the system. Under a Swedish-model reform, it would also operate a contributionsclearinghouse and administer the personal accounts.

China will need a stronggovernment regulator tosupervise the personalaccounts system.

• Fund managers. To receive certification as fundmanagers, financial services firms would have to meetminimum standards for business experience, assetsunder management, and past investment performance.Once participating in the system, they would besubject to strict capital requirements, fiduciaryobligations, and reporting rules. A wide range offinancial services firms, from “trust companies” toinsurance companies, would be eligible to set up a PAmanagement company or investment fund—providedthat they meet the certification standards. To ensurethat personal accounts are fully portable, fundmanagers would be required to operate nationwide.The type of system we propose would requirerelaxing some current barriers to market entry. Thisis necessary to promote efficient competition andensure that the system is national in scope.

Participating fund managerswould be subject to strictcapital requirements, fiduciaryobligations, and reporting rules.

• Fund selection.Workers would register theirchoice of fund manager with their local tax bureau,either directly or through their employer. Splittingcontributions among multiple fund managers wouldnot be permitted, though in a Swedish-model reform

this restriction could be relaxed as the capabilities ofthe clearinghouse grow. Workers who fail to choosea fund manager would be assigned to a default fund.This might be the PA management company orinvestment fund with the lowest administrative feesor the highest net rate of return. Alternatively, itcould be a special government default fund, whichmight be managed by the National Pension Fund.The regulator would establish rules for howfrequently workers can change fund managers.Initially, changes would be limited to once per yearto minimize the administrative burden.

• Routing system. The personal accounts routingsystem would make use of existing channels.Contributions would be withheld by employers anddeposited to special local tax bureau holding accountsat a commercial bank. The self-employed would beresponsible for making their own deposits. The localtax bureau would match deposits against employmentand wage records to ensure compliance. It wouldthen transfer contributions to the PA managementcompany of the worker’s choice or to the centralclearinghouse for allocation to the investment fundof the worker’s choice. In principle, collection androuting of contributions could also be handled bythe local social security bureaus. There are severaladvantages, however, to using the tax bureaus. Itwould maintain a firewall between the personalaccounts and the rest of the basic pension system,which will still be receiving government subsidiesand paying benefits over the next few decades. Thetax bureaus also have greater enforcement capabilitiesand can make use of an existing national bank wiretransfer system.

The new personalaccounts system wouldbe fully portable.

• Account administration. Local tax bureaus wouldmaintain a database with a record for each registeredworker employed within their jurisdiction. Therecord would include the worker’s social securitynumber, residence, employer, and wage andcontribution history. These records would be sharedelectronically with the worker’s PA managementcompany or the central clearinghouse. The managementcompanies or clearinghouse would maintain theirown databases, which would include completeinformation on account balances and transactions.This system would be fully portable. When workersmove between tax jurisdictions, they would simplyregister with their new tax bureau, which would

reconnect them with their PA management company orthe central clearinghouse. The new tax bureau in turncould access the worker’s prior tax bureau recordsthrough his or her account administrator.Alternatively, these records could be transferreddirectly to the new tax bureau using the “thirdgeneration” IC social security card now being testedby the Shenzhen city government.32 In a Chilean-model reform, fund balances and personal accountrecords could, at the worker’s request, be transferreddirectly between PA management companies. In aSwedish-model reform, fund balances would betransferred through the clearinghouse; there wouldbe no need to transfer account records.

• Administrative fees. The personal accounts systemwould be subject to regulations designed to limitexcessive administrative fees. Fees could be cappeddirectly by the regulator, either as a percent ofcontribution flows or account balances. Alternatively,the regulator could institute measures that promoteefficiency-enhancing competition among fundmanagers. This could be done in a variety of ways—for instance, by keeping barriers to entry low; byallowing fund managers to offer loyalty discounts; byassigning workers who fail to choose a fund managerto the fund with the lowest fees; or by allowingworkers to switch fund managers more frequently solong as they move to the fund with the lowest fees.Either way, the regulator would need to establish astandardized fee structure that allows workers toeasily compare net investment returns.

• Reporting requirements. The personal accountssystem would also be subject to strict reportingrequirements designed to ensure that participantshave adequate information to make informeddecisions. The regulator would compile comparative

data on the performance of fund managers andpublish it regularly in user-friendly form in print andon the internet. Workers would also receive periodicaccount statements in a standardized format, eitherfrom their PA management company or, in aSwedish-model reform, from the pension regulator.

• Disability and survivors insurance. The regulatorwould establish guidelines for disability and youngsurvivors insurance policies. Under a Chilean-modelreform, PA management companies would berequired to purchase a standardized policy froma commercial insurer for each enrolled worker.Under a Swedish-model reform, the regulator wouldpurchase the policies for workers. As an alternativeto purchasing private insurance, it would also bepossible under this model to finance disability andyoung survivors benefits on a pay-as-you-go basis.

All contributions would beinvested in lifecycle funds.

• Portfolio guidelines.While personal account assetsthat are invested in capital markets are inevitablysubject to market risk, financial experts agree thatthis risk can be minimized by regulations that requireworkers to maintain an adequate spread in theirportfolios and to move into fixed-income assets atolder ages. The CSIS plan would require each fundmanager to offer three to five lifecycle investmentfunds with varying degrees of risk that workers wouldbe assigned to based on their age. Initially, theregulator would establish broad guidelines forallowable investments, with minima and maximafor different asset classes. Over time, however, therestrictions would be relaxed as China’s capital marketsdevelop and the experience of fund managers grows.Ultimately, the reform’s success will require movingtoward a “prudent man” investment rule that allowscontributions to flow to the investments with thehighest risk-adjusted returns. This is another wayof saying that the Chinese government would haveto allow the personal accounts to hold a globallydiversified portfolio.

• Withdrawal rules. Personal account balances wouldbe subject to strict rules governing their use. Until theminimum retirement age, no worker will be allowedto withdraw personal account funds. After the minimumage, all funds withdrawn must be used to purchase aninflation-indexed annuity until the following conditionis met: The annuity, together with any zero-tier benefit,

Chapter 3 33

32 Tingting Zhang, “Nationwide Applicable Social Security System Urged,” China.org.cn, March 4, 2008; and “New Social SecurityScheme on Cards in Shenzhen,” China Daily,March 26, 2008.

Chapter 3 35

ensures that the worker possesses an income forlife equal to some minimum threshold—perhaps50 percent of the local average wage, or two and one-half times the plan’s poverty backstop. Couples wouldhave to purchase a joint and survivor annuity. Theannuities would be purchased from insurance companiesor, if it is determined that companies cannot pricethem fairly, from the regulator. Fund balances inexcess of those required to purchase the minimumannuity will be subject to no use restrictions and may beconsumed or passed on to heirs. As an alternative tothe annuity rule, workers might be allowed to makeannual withdrawals, with the allowable amount of thewithdrawal determined by remaining life expectancy.

The plan would requiremandatory annuitization ofaccount balances to protectagainst longevity risk.

• Retirement age. The minimum retirement agewould initially be set at 60 for men and 55for women, just as it is in the current basic pensionsystem. These low retirement ages are necessarybecause today’s older workers often do not havethe skills to compete in China’s rapidly modernizingeconomy. But as these workers are replaced byyounger and higher-skilled cohorts and as China’spopulation ages, longer work lives will not onlybecome feasible, but essential. Our plan thereforeprovides for gradually raising the minimumretirement age for both sexes to age 65 by 2030,after which it would be indexed to longevity.

• Benefit guarantees.Many personal accountssystems have minimum benefit guarantees thatare pegged to the minimum wage or some otherpoverty threshold. Under our plan, this safetynet protection would instead be provided by thenoncontributory poverty floor. Many systems alsohave relative rate of return guarantees backed up byfund managers’ capital reserves. The advantage ofthese guarantees is that they protect workers whentheir fund manager performs much worse than theaverage fund manager. The disadvantage is that theyencourage overly conservative investment strategiesby all fund managers which may result in smalleraccount balances and retirement benefits for allworkers. The government will have to carefullyweigh the costs and benefits of this design feature.

TheAdvantages of FundingThe kind of personal accounts system we proposewould have decisive advantages over a pay-as-you-gosystem as China ages. To begin with, it could deliveradequate benefits at a lower contribution rate. Whilethe rate of return to contributions to a pay-as-you-gosystem is limited to the rate of economic growth—or more precisely, the rate of growth in taxablepayroll—the rate of return in a genuinely fundedpersonal accounts system would be equal to the rateof return to capital, which, as China ages and itseconomy develops, will almost certainly be higher.It is true that this might not be true in the nearterm. After all, real wages—and along with themthe payroll tax base on which pay-as-you-gofinancing depends—are now growing at double-digit rates as China’s economy modernizes andemployment shifts en masse from nonproductive toproductive sectors. At the same time, the returnto capital is relatively low because China is savingprodigiously and investing inefficiently.

A genuinely funded pensionsystem can deliverhigher benefits at a lowercontribution rate than apay-as-you-go system can.

In any success story for the Chinese economy,however, both of these circumstances will change. AsChina’s economy matures and the potential for hugeleaps in productivity is exhausted, real wage growthwill inevitably slow to something closer to developed-world levels. At the same time, China will also see aslowdown in employment growth as its working-agepopulation peaks and begins to contract. Together,these trends will dramatically lower the potential rateof return on a pay-as-you-go retirement system.Meanwhile, the rate of return to capital is bound torise as China’s capital markets develop—and wouldbe given a big extra push by the type of fundedpersonal accounts system we propose.

Indeed, it is difficult to imagine a plausible long-termscenario in which a genuinely funded personalaccounts system would not outperform a pay-as-you-go system.33 Consider a worker who joins theworkforce in 2010, contributes 16 percent of coveredwages to the retirement component of a personal

33 The following projections compare the performance of our personal accounts system with a hypothetical reformed pay-as-you-gosystem which, like our personal accounts system, has been unburdened of the cost of the basic pension system’s unfunded liabilities.They also assume that coverage under the pay-as-you-go system will be universal. Under China’s existing basic pension system,replacement rates would be much lower—or contribution rates much higher—than those projected here.

36 China’s Long March to Retirement Reform

*PAYGO calculations assume universal coverage and a retirement age of 65; personal account calculations assume a 40-year career,a retirement age of 65, and administrative charges equal to 0.5 percent of assets.

Source: CSIS calculations

13.5%

17.6%

20.2%

23.3%

27.0%

31.4%

15.4%

0%

5%

10%

15%

20%

25%

30%

35%

5.5% 5.0% 4.5% 4.0% 3.5% 3.0% 2.5%

Perc

ent

ofPa

yrol

l

Real Wage Growth Rate

Personal Account Contribution Rate: 16%

Pay-As-You-Go Contribution Rates Required in 2050 to Match Personal Account ReplacementRates under Different Real Wage Growth Assumptions: 5.0 Percent Real Rate of Return Scenario

Figure 17b

As China ages and develops, a funded pension system will almost certainlyoutperform a pay-as-you-go system.

Out

perf

orm

sPA

YG

O

5.5% 5.0% 4.5% 4.0% 3.5% 3.0% 2.5%

Real Wage Growth Rate

Rea

lRat

eof

Ret

urn

PAYGO

3.0%

3.5%

4.0%

4.5%

5.0%

5.5%

6.0%

26%

30%

33%

38%

43%

50%

57%

28%

32%

36%

41%

47%

54%

62%

30%

35%

40%

45%

52%

60%

69%

33%

38%

43%

50%

57%

66%

76%

36%

41%

47%

55%

63%

73%

84%

39%

45%

52%

60%

70%

81%

94%

43%

50%

58%

67%

78%

90%

105%

52% 49% 47% 45% 43% 41% 39%

Personal Account Replacement Rates in 2050 under Different Real Wage Growth and Rateof Return Assumptions versus PAYGO Replacement Rates at the Same 16 PercentContribution Rate*

Figure 17a

Chapter 3 37

account throughout a 40-year career, and earns anaverage real rate of return of 5.0 percent, areasonable long-term assumption for a prudentlydiversified global portfolio of stocks and bonds. For apay-as-you-go system with the same contribution rateto beat the personal accounts system’s replacementrate, real wage growth would have to indefinitelyaverage 5.0 percent or higher, quadruple the averagedeveloped-country performance of the past quarter-century. Even if the worker were to earn a real rateof return of just 3.0 percent, what many economistsconsider the long-term, risk-free rate of return, realwage growth would have to average 3.0 percent orhigher for the pay-as-you-go system to beat the personalaccounts system’s replacement rate. (See figure 17a.)

This is of course equivalent to saying thatcontribution rates would, in almost any plausiblelong-term scenario, have to be higher under a pay-as-you-go system to match the replacement rates thatcould be delivered by a funded system. Consideragain the worker who contributes 16 percent ofwages to a personal account and earns a 5.0 percentreal rate of return. If real wage growth averages 5.0percent or higher over the next 40 years, the pay-as-you-go system would enjoy a modest cost advantage.But if real wage growth averages less than 5.0percent, the pay-as-you-go contribution rate wouldhave to rise steeply to deliver the same benefits as thefunded system. At 3.0 percent real wage growth, thecontribution rate would eventually have to rise to 27percent—and at 2.5 percent real wage growth, itwould have to rise to 31 percent. (See figure 17b.)

A system of fundedpersonal accounts wouldmake most workersstakeholders in China’slong-term economic success.

Beyond its potential to deliver adequate benefitsat a lower contribution rate, there are otherimportant advantages to the system we propose.Because contributions would be personally owned,workers would be less likely to view them as a tax.Together with the lower contribution rate, this wouldimprove incentives to participate, potentiallybroadening coverage and reducing the cost of thefloor of protection as well. In the near term, thesystem would help to broaden and deepen China’scapital markets, the crucial next step in its developmentagenda. In the long term, it would help to maintain

rates of savings, investment, and living standardgrowth as China ages. Over time, the system wouldmake most workers direct stakeholders in China’seconomic success, foster middle-class values of thriftand stewardship, and perhaps even turn China intoa more democratic society.

Reform and China’sCapital MarketsTo set up the personal accounts system, China willhave to overcome considerable obstacles. To beginwith, there are the administrative challenges. As wehave seen, China will need to establish a new centralregulatory agency that is both highly professionaland enjoys a reasonable degree of independence.It will also have to put in place reliable routing,recordkeeping, and oversight systems. Although thesechallenges are manageable, building the infrastructureof the new retirement system will nonetheless requirea significant investment of time and money.

China is laying theinfrastructure of amodern financial system.

Then there are the even greater challenges posedby China’s immature capital markets, which sufferfrom serious structural problems, from poorcorporate governance and weak property rights tolack of liquidity, rampant speculation, and pervasivegovernment control. Until a few years ago, manyof the fundamental requirements for a funded personalaccounts system were lacking. Since the release ofthe State Council’s guidelines on capital marketdevelopment in January 2004, however, the pace ofreform has accelerated and the basic infrastructure ofa modern financial system has begun to take shape.

Accountability and transparency, though stillinadequate, are improving. China’s new companylaw, which became effective in January 2006, moreclearly defines executives’ fiduciary responsibilities,bolsters the independence of auditors, and protectssmall shareholders’ rights. Its new securities law,which became effective at the same time, strengthenssafeguards against insider trading and establishesstricter disclosure requirements for publicly tradedcompanies. Its new accounting and auditingstandards, introduced in January 2007, are beinghailed by international investors as an importantfirst step toward aligning Chinese norms withglobal best practice.

38 China’s Long March to Retirement Reform

The government is also strengthening property rights.In 2004, China took the epochal step of officiallyrecognizing a general right to private propertyownership in its constitution. Its 2007 property lawspells out that private individuals and companies canown, buy, sell, bequeath, and inherit all types ofproperty except land—and that they can sue in courtto enforce disputed claims. To be sure, there remainsconsiderable ambiguity about the practical divisionsbetween state, collective, and privately ownedproperty. Given the weakness and widespreadcorruption of the judicial system, moreover, enforcingproperty claims can still be difficult. As TheEconomist nicely puts it, “the passage of laws is notthe rule of law.”34 Still, China has made importantprogress in laying the most basic legal foundation ofa developed market economy.

At the same time, government reforms are pavingthe way for a significant broadening of China’scapital markets. Beginning in 2005, the governmentlaunched a process that will triple the effective size ofChina’s stock markets by converting all nontradableshares (after a “lock-up period”) into tradable shares.In 2007, it liberalized the approval process forcorporate bond issues and opened up the market toprivate firms, which had previously been excluded.Nor are the reforms restricted to broadening capitalmarkets. The government is also encouraging deeper

markets by authorizing new financial products, fromcurrency swaps to stock index futures.

There has also been tentative movement towardeasing capital controls. In 2002, the governmentestablished the Qualified Foreign InstitutionalInvestors (QFII) program, which opened the doorto foreign investment in Chinese capital markets.In 2006, faced with a growing mountain of foreignreserves, it established the counterpart QualifiedDomestic Institutional Investors (QDII) program.By depositing funds with participating financialinstitutions authorized to purchase foreign currencyfrom the government, Chinese households can now,for the first time, invest a part of their savings abroad.

Despite impressiveprogress, China’s capitalmarkets are still relativelysmall and illiquid.

Despite impressive progress, all financial expertsacknowledge that major problems remain. AlthoughChina’s stock markets are playing a growing role inbusiness finance, they are still relatively small andilliquid. As of mid-2008, bank deposits in Chinaoutweighed the value of tradable stocks by nearly8 to 1. In the United States, the ratio was 2 to 1 theother way around. (See figure 18.) Most small andmedium-sized firms find it impossible to meet thecapital and regulatory requirements for listing.Although private stock offerings have recently beenauthorized, there is still no market infrastructurein place for private placement. China’s bond marketis if anything less developed. The market consistsalmost entirely of central government bonds withshort-term maturities. Corporate bonds, whose issuanceis subject to central government approval, account forjust 10 percent of the market. Local governmentbonds, which are an important source of capitalfinancing in many developed countries, do not exist.

China’s investment cultureremains highly speculative.

China’s investment culture also remains short-termand highly speculative. The surging number of smallindividual investors—the number of Chineseinvestment accounts grew from 60 to 138 million in2007 alone—helps to set the overall market tone.

34 “China’s Next Revolution,” The Economist,March 8, 2007.

Chapter 3 39

As of the end of 2007, individual investors ownedroughly half (51 percent) of Chinese stocks by value,nearly double the individual investor share in theUnited States (26 percent).35 Although institutionalinvestors play a significant role in China’s capitalmarkets, pension funds and life insurance companies,the types most likely to foster a long-term investmentculture, are still tiny players. In the United States,pension funds and life insurance companies togetherown 30 percent of all stocks. In China, thecorresponding figure is just 3 percent, even includingthe National Pension Fund.36

Perhaps the greatest obstacle, however, is thegovernment’s continued control of China’s capitalmarkets.37 This control reaches far beyond thegovernment’s well-known interventions in exchangemarkets to keep the RMB from rising. Despite recentreforms, most savings in China is still allocated toinvestment by state-owned banks, often through apolitically determined process. The government triesto guarantee the banks a large supply of low-costcapital by setting nominal interest rates for depositsfar beneath the market rate, while regulating the

availability of alternative domestic investments andstrictly controlling the flow of savings abroad. Overthe past few years, with inflation rising, the real rateof return on bank deposits has actually been negative.Not surprisingly, Chinese investors are beginning tofind ways around the government roadblocks—byshifting bank deposits into the stock market,purchasing higher-interest insurance products, andeven investing their savings in black market loansto businesses denied access to cheap capital throughthe state-owned banks.

Although reform wouldhave to be phased ingradually, China cannot affordto wait before launching afully funded pension system.

All of this means that a fully funded personalaccounts system would have to be phased ingradually. It might even make sense, as some

166%

66%

50%65%

135%

216%

0%

50%

100%

150%

200%

250%

China United States

TradableStocks:22%

Perc

ent

of

GD

P

Bank Deposits Stock Market Capitalization Bond Market

CorporateBonds: 5%

In China, bank deposits still dwarf the stock and bond markets.Assets by Type in China and the United States, as a Percent of GDP, at the End of June 2008*

*U.S. bond market data are for the end of 2007.Source: ADB (2008); CSRC (2008); FRB (2008); PBC (2008); IMF (2008); and CSIS calculations Figure 18

35 China Securities Regulatory Commission, China Capital Markets Development Report (Beijing: China Financial Publishing House,2008), 275; and Flow of Funds Accounts of the United States (Washington, DC: Board of Governors of the Federal Reserve System,September 18, 2008), 90.

36 Ibid.37 For a discussion, see Nicholas R. Lardy, “Financial Repression in China,” Policy Brief no. 08-8 (Washington, DC: Peterson Institutefor International Economics, September 2008).

40 China’s Long March to Retirement Reform

experts have suggested, for China to retain a hybridsystem of notional and fully funded accounts for awhile.38 Rather than the bank interest rate, however,the notional accounts would be credited with aninterest rate equal to real wage growth or GDPgrowth, the standard practice in notional accountssystems in developed countries. An even betteroption, proposed by the World Bank, would be forthe government to issue special GDP-indexed bondsto the accounts that would eventually be madetradable when the system becomes fully funded.39

Funded pensions couldplay a crucial role inbroadening and deepeningChina’s capital markets.

Yet China cannot afford to wait for the fulldevelopment of its capital markets before launchinga privately managed and invested pension system.In fact, such a system may be the key to ensuringthat very development. In many countries, fundedpensions have played a crucial role in broadening anddeepening capital markets. This is true in the UnitedStates, the UK, and other developed countries withlarge employer pension systems. It is also true inLatin American and Eastern European countries thathave established mandatory personal accountssystems similar to the one we propose. As China’sfunded personal accounts grow, so will the size andliquidity of its capital markets. Along with professionalfund management will come greater accountability,transparency, and ultimately the long-term investmentculture that China now lacks.

Over time, the new personal accounts systemwould also help to liberalize China’s capital markets.Although the government is hesitant to relinquishcontrol, it understands that China’s long-termeconomic success demands broad and deep capitalmarkets capable of allocating domestic savingsefficiently to the most productive investments. Italso understands that long-term economic successdemands globally integrated capital markets. Chineseinvestors must be able to compare the return on anydomestic industry or company with similar industriesor companies abroad. Chinese firms must be able to“read” global markets to seek out new customers,diversify their exports, and make their own strategic

investments abroad. Several hundred million personalaccount owners demanding a true market rate ofreturn—which by definition is the global rate ofreturn to capital—could become a powerful forcepushing government policy in the direction that thegovernment already knows it needs to move.

A mandatory personalaccounts system could helpthe government manage thetransition to a “float.”

Capital market integration would of course bea gradual process lasting many years. It would,eventually, require floating China’s currency againstmost other nations. Transitioning to a float in turnraises concerns about the possibility of a large initialdrop in the RMB as Chinese investors start movinginto foreign securities. A mandatory personalaccounts system could help manage this transition bygiving the government a powerful tool for regulatingthe flow of Chinese savings abroad. For example, acap on the foreign share of assets in each accountcould be gradually raised (or frozen, if necessary).Funded personal accounts thus furnish an ideal meansto regulate a transition that most policy experts inany case believe is sooner or later unavoidable.

Funded pensions are theground zero where China’sdevelopment and agingchallenges meet.

China needs broader and deeper capital marketsto build the funded pension system on which asuccessful response to the aging of its populationrests. But it also needs a funded pension system tobuild the broader and deeper capital markets onwhich its economic development agenda ultimatelydepends. Funded pensions are thus the groundzero where China’s two greatest challenges meet.

38 See, for instance, the “8-8-8 solution” proposed by Bingwen Zheng of the Chinese Academy of Social Sciences. Bingwen Zheng,“The Origin of China’s Partially Funded Social Security Scheme and Its Future Direction,” The Chinese Economy 40, no. 4(July/August 2007).

39 China: Evaluation of the Liaoning Social Security Pilot, Report no. 38183-CN (Washington, DC: The World Bank, December 2006).

42 China’s Long March to Retirement Reform

How China confronts its aging challenge will domuch to determine the future shape of its economy,society, and role in the world order. If China succeedsin forging an effective retirement policy, the potentialbenefits of rapid economic development will beenlarged. If it fails, the costs will increase. Indeed,without an effective retirement policy, it is difficultto envision a prosperous long-term future for China.

China has been “peacefully rising” while itsdemographics have leaned with economic growth.But by the 2020s, demographic trends may beweakening the two principal pillars of thegovernment’s political legitimacy—rapidly risingliving standards and social stability. Not just aretirement crisis, but also a more general socialand economic crisis may loom in China’s future.While it is difficult to gauge the risks, the Chinesegovernment, with its new emphasis on “balanceddevelopment” and building a “harmonious society,”is clearly taking them seriously. Like many developingcountries, China is experiencing the stresses ofmodernization, from rapid industrialization andurbanization to weakening families and wideningincome gaps. These stresses, which have been bearablein a youthful China in which incomes have beenrising rapidly, may become less tolerable in an agingChina in which economic growth is slowing.

With an effective retirementpolicy, the trajectory ofChina’s living standardwill no longer be at themercy of demography.

The CSIS plan can help to avert crisis and ensurefuture prosperity. With a universal floor of old-agepoverty protection in place, elders will no longerhave to fear that China’s overburdened family supportnetworks will fail them. With a system of genuinelyfunded personal accounts in place, a growing shareof the elderly will look forward to a comfortableretirement. Workers would not have to transfer anever-larger share of their wages to nonworking elders.The trajectory of China’s living standard wouldno longer be at the mercy of demography.

Successful retirement reform would have many otherbenefits as well. If elders have a guaranteed minimumpension, young workers will be able to move away totake a better job without having to worry about theiraging parents. If workers have a fully portable personalaccount, they will be able to take greater risks instarting a new job, career, or business without havingto worry about their own future old-age security.The long-term economic result will be to increaseeconomic mobility and dynamism. The political resultmay be to bolster popular support for the publicplanners and administrators who wisely prepared forChina’s unprecedented demographic makeover.

Successful retirement reform would play an especiallyprofound and positive role in fostering the growth ofChina’s capital markets. Nearly all economists agreethat sooner or later a developing economy needs tobuild broad and deep capital markets. China is nodifferent. It needs them to raise capital efficiently fromthe savings of the hundreds of millions of workingfamilies who today invest their money in unproductivehousing or low-return bank deposits. China also needsbroad and deep capital markets to allocate capitalefficiently to the emerging industries and businessesthat can move its economy up the global value-addedscale in decades to come.

Clearly, many institutional reforms may be needed toprepare the way, including court-enforced due processthat protects ownership rights and an expert regulatoryadministration with enforcement authority that codifies“fair” trading in capital markets and protects investorsagainst deception and fraud. Once in operation,however, a national personal accounts system willbecome a vital source of strength. Beyond the sheervolume of funds that would be injected into China’scapital markets, the funded component of a reformedpension system could prove to be a powerful forcebehind the global integration of its capital markets.In effect, personal accounts would be the vehicle bywhich ordinary Chinese households help to diversifyChina’s portfolio. Personal accounts will also educateworkers about how financial markets function. In thelong run, workers will begin to see themselves asstakeholders in the existing economic system and—to the extent it protects and acts on behalf of thatsystem—as property-owning supporters of the existingpolitical regime.

China at a Crossroads

CONCLUSION

Conclusion 43

Reform could help steerChina toward a moredominant global role—andthis would redound to thebenefit of all nations.

The future of its capital markets will do much todetermine the future of China’s role in the worldeconomy. Strong and globally integrated marketswill ensure balanced economic growth and a moredominant world role. Weak and isolated marketsmay lead to unbalanced growth, overdependence onforeign direct investment, and a less important worldrole. Successful retirement reform could help steerChina toward the more dominant role—and thiswould redound not just to its own benefit, but tothe benefit of all nations.

If China is successful in meeting its aging challenge,the future will not only be one in which elders retirein greater comfort and families live with less worry,but it will also be a future in which capital formationis stronger, living standards are higher, and publictrust in government is firmer. Most of all, it will bea future in which China’s maturing role in the globaleconomy undergoes a basic shift—from a vast reservoirof unskilled labor to a dynamic, high-saving, high-investing, high-value-added economy. The wholeworld thus has a stake in the outcome.

There are many reasonsto be hopeful that Chinawill successfully meet itsaging challenge.

There are many reasons to be hopeful that Chinawill steer a successful course. The government iscommitted to achieving universal pension coverage.It also recognizes the decisive advantages of fundedretirement savings, which is why it is has begunto fill the basic pension system’s empty accounts,established a national pension reserve fund, andlaid the groundwork for a supplementary system ofemployer-sponsored enterprise annuities. There iseven a growing consensus, reinforced by the recentShanghai pension scandal, that independentmanagement of pension fund assets is preferable to

state management. At the same time, economic andinstitutional developments have made the type ofreform we propose more feasible than it was justa few years ago. On the one hand, the centralgovernment’s fiscal capacity to finance a povertyfloor and socialize the cost of the basic pensionsystem’s unfunded liabilities is much greater. Onthe other hand, recent government reforms thathave strengthened property rights and capitalmarket oversight have removed many of thepractical obstacles to a funded pension system.

China stands at a crossroads. Its age wave is fastapproaching, and along with it the steep upwardspiral in dependency costs that threatens to slow itseconomic rise. Now is the time—while there still istime—for China to take the next steps and put inplace a retirement system that will allow it toconfront the demographic gauntlet ahead.

44 China’s Long March to Retirement Reform

A Note on Data and SourcesIn researching and writing China’s Long March to Retirement Reform, CSIS consulteddozens of studies on China’s demography, economy, and, of course, retirement system. Thisnote makes no attempt to review this rich secondary literature. Its purpose is morelimited—to orient the reader to the basic data sources used in preparing the report.

Most demographic data, historical and projected, come from the UN Population Divisionand are published in World Population Prospects.1For some specific data series, however,the report relies on Chinese census data or other government surveys. These includepopulation by urban or rural residence, which come from China’s National Bureau of Statistics(NBS) and are published in the China Statistical Yearbook;2 historical estimates of Chinese totalfertility rates;3 historical estimates of Chinese life expectancy (for 1982–2005);4 and Chinesesex ratios at birth.5 All population projections refer to the UN’s 2006 Revision “constantfertility” scenario. CSIS believes that this scenario constitutes a better baseline than the morecommonly cited “medium variant” scenario, which arbitrarily assumes that fertility in allcountries will converge at a rate of 1.85. The two scenarios differ substantially for somecountries. In the case of China, however, the scenario choice is not critical, since China’scurrent fertility rate is close to the assumed convergence rate. By 2050, the elderly share ofChina’s population diverges by just 2 percent under the two scenarios—31.1 percent underthe medium variant versus 32.8 percent under the constant variant.

Data on GDP (except in international comparisons), inflation, and household income comefrom NBS and are published in its China Statistical Yearbook and its Statistical Bulletin.6Mostdata on the Chinese labor force, employment, and wages (total and by residence registration)come from the Ministry of Human Resources and Social Security (MOHRSS) and arepublished in its China Labor Statistical Yearbook7 and its Statistical Bulletin.8 Data on labor-force participation by age, however, come from an NBS survey.9 Data on Chinese financialmarkets come from the People’s Bank of China (PBC), the China Securities RegulatoryCommission (CSRC), and the Asian Development Bank (ADB). PBC data include total bankdeposits, interest rates for one-year term deposits, and foreign exchange reserves. (The data areavailable at http://www.pbc.gov.cn.) CSRC data include stock market capitalization (bothtradable and nontradable shares) and number of investor accounts. (The data are available athttp://www.csrc.gov.cn.) ADB data include bond market capitalization (both private andpublic).10

1 World Population Prospects: The 2006 Revision (New York: UN Population Division, 2007).2 National Bureau of Statistics of China, China Statistical Yearbook 2007 (Beijing: China Statistics Press,2007).

3 National Population and Family Planning Commission of China, Statistical Bulletin of Family Planning [inChinese], various years, http://www.npfpc.gov.cn.

4Ministry of Health of China, China Health Statistical Yearbook 2008 [in Chinese] (Beijing: China UnionMedical College Press, 2008).

5 National Population and Family Planning Commission of China, Research Report on the NationalDemographic Development Strategy [in Chinese], January 11, 2007,http://www.chinapop.gov.cn/fzgh/zlyj/200806/t20080626_154129.htm.

6 National Bureau of Statistics of China, 2007 Statistical Bulletin of National Economic and SocialDevelopment [in Chinese], February 28, 2008, http://www.stats.gov.cn.

7 National Bureau of Statistics of China and Ministry of Human Resources and Social Security of China,China Labor Statistical Yearbook 2007 (Beijing: China Statistics Press, 2007).

8Ministry of Human Resources and Social Security of China, 2007 Statistical Bulletin of Labor and SocialSecurity [in Chinese], May 21, 2008, http://w1.mohrss.gov.cn/gb/zwxx/2008-06/05/content_240415.htm.

9 2005 National 1% Sampling Survey, National Bureau of Statistics of China, March 16, 2006,http://www.stats.gov.cn/tjsj/ndsj/renkou/2005/renkou.htm.

10 Asian Bonds Online, Asian Development Bank, http://www.asianbondsonline.adb.org.

A Note on Data and Sources 45

Data used in cross-country comparisons of GDP and GDP per capita (in exchange rate andpurchasing power parity dollars) are from the World Bank’s World Development Indicatorsdatabase and are available at http://www.worldbank.org. Data on international trade,finance, and capital flows all come from standard international sources.

Data on China’s basic pension system, including coverage and finances, are compiled byMOHRSS and NBS and are published in the China Statistical Yearbook and the MOHRSSStatistical Bulletin. Most of the data are available at http://www.mohrss.gov.cn. Data onChina’s rural pension system are from the same MOHRSS sources. Data on civil servicepensions are from a specialized study.11 Data on the minimum living standard guarantee andother welfare programs are from the Ministry of Civil Affairs and are available athttp://www.mca.gov.cn. Data on the National Pension Fund are from the National Councilfor the Social Security Fund of China and are published in the Annual Report of theNational Social Security Fund.12Most of the data are also available at http://www.ssf.gov.cn.

11 Yvonne Sin and Leslie Mao, “Hidden Pot of Gold: Responding to China’s Pension Burden,” CLSA-USpeaker Series (Shanghai: CLSA, October 2007).

12 National Council for the Social Security Fund of China, 2007 Annual Report of the National Social SecurityFund [in Chinese], June 30, 2008.

46 China’s Long March to Retirement Reform

A Key to Chart Source CitationsADB (2008) = Asian Bonds Online, Asian Development Bank,http://www.asianbondsonline.adb.org.

CDRF and UNDP (2005) = China Development Research Foundation and United NationsDevelopment Program, China Human Development Report 2005 (Beijing: ChinaTranslation and Publishing Company, 2005).

CSRC (2008) = Monthly Statistical Report, China Securities Regulatory Commission,August 19, 2008, http://www.csrc.gov.cn.

FRB (2008) = Flow of Funds Accounts of the United States (Washington, DC: Boardof Governors of the Federal Reserve System, September 18, 2008).

IMF (2008) = Global Financial Stability Report (Washington, DC: International MonetaryFund, October 2008).

Maddison (2007) = Angus Maddison, World Population, GDP and Per Capita GDP,1–2003 AD, August 2007, http://www.ggdc.net/maddison.

MOH (2008) = Ministry of Health of China, China Health Statistical Yearbook 2008[in Chinese] (Beijing: China Union Medical College Press, 2008).

MOHRSS (2008) = Ministry of Human Resources and Social Security of China,2007 Statistical Bulletin of Labor and Social Security [in Chinese], May 21, 2008,http://w1.mohrss.gov.cn/gb/zwxx/2008-06/05/content_240415.htm.

NBS (2006) = 2005 National 1% Sampling Survey, National Bureau of Statistics of China,March 16, 2006, http://www.stats.gov.cn/tjsj/ndsj/renkou/2005/renkou.htm.

NBS (2007) = National Bureau of Statistics of China, China Statistical Yearbook 2007(Beijing: China Statistics Press, 2007).

NPFPC (various years) = National Population and Family Planning Commission of China,Statistical Bulletin of Family Planning [in Chinese], various years, http://www.npfpc.gov.cn.

PBC (2008) = 2008 Statistical Data, The People’s Bank of China, http://www.pbc.gov.cn.

Sin and Mao (2007) = Yvonne Sin and Leslie Mao, “Hidden Pot of Gold: Responding toChina’s Pension Burden,” CLSA-U Speaker Series (Shanghai: CLSA, October 2007).

UN (2007) = World Population Prospects: The 2006 Revision (New York: UN PopulationDivision, 2007).

U.S. Census Bureau (2002) = U.S. Census Bureau, Demographic Trends in the 20th Century(Washington, DC: U.S. Government Printing Office, November 2002).

World Bank (2008) = World Development Indicators 2008, The World Bank,April 2008, http://www.worldbank.org.

Acknowledgments 47

AcknowledgmentsThe authors would like to thank a number of individuals and organizations for theircontributions to China’s Long March to Retirement Reform.

First mention must go to Jiangong Zhou, a Chinese financial market expert and consultantto the CSIS Global Aging Initiative (GAI), who made valuable contributions at all stages ofthe research and translated the report into Chinese. Without his assistance, the report wouldbe much poorer. The authors are also grateful to Xiaoqing Lu, a research associate with theCSIS China Chair, who assisted with the translation; and to Xi Yang, an intern with GAI,whose dedicated efforts helped clarify many difficult research questions.

The authors wish to thank The Prudential Foundation for its generous financial support.They are grateful to Gregory M. Loder (Vice President, Issues Management, PrudentialFinancial) for believing in the project and shepherding the report through the productionprocess; and to Linda Brnicevic (Project Manager, Advertising and Creative Services,Prudential Financial) and Robert Gary (Graphic Designer, Advertising and Creative Services,Prudential Financial) for the attractive design and graphics. They also appreciate the helpfulinput of Wei Jin (Vice President, New Markets, Prudential International Insurance) andTeh-Hsiu Fu (Director of the Board and Chief Executive Officer, Everbright Pramerica).

Along the way, the authors sought the advice of many experts. They owe a special debt ofgratitude to Xiao Caiwei (Director, International Department, China National Committeeon Aging); Yvonne Sin (Head of Investment Consulting, China and Director of ChinaMarket Development, Employee Benefits, Watson Wyatt Worldwide); and Bingwen Zheng(Senior Research Fellow, Institute of Latin American Studies, Chinese Academy of SocialSciences). All three gave generously of their time in answering our many questions. Theywould also like to thank Fang Cai (Director, Institute of Population and Labor Economics,Chinese Academy of Social Sciences); Jacky Chau-kiu Cheung (Associate Professor,Department of Applied Social Studies, City University of Hong Kong); Takeshi Jingu (ChiefRepresentative in Beijing, Nomura Institute of Capital Markets Research); Alex Yui-huenKwan (Head and Professor, Department of Applied Social Studies, City University of HongKong); Keping Li (Vice Chairman, National Council for the Social Security Fund); andBin Qi (Director of the Research Center, China Securities Regulatory Commission).

While the authors gratefully acknowledge the assistance they received in researching thereport, they are solely responsible for its content.

48 T h e

Graying of the Middle Kingdom

About the Authors

Richard Jackson writes on public policy issues arising from the aging of America’s andthe world’s population. He is currently a senior fellow at CSIS, where he directs the GlobalAging Initiative, an adjunct fellow at the Hudson Institute, and a senior adviser to theConcord Coalition. Jackson is the author or co-author of numerous studies on theimplications of global aging, including The Global Retirement Crisis (2002), The AgingVulnerability Index (2003), The Graying of the Middle Kingdom (2004), Building HumanCapital in an Aging Mexico (2005), The Aging of Korea (2007), and The Graying of theGreat Powers (2008). Jackson regularly speaks on long-term demographic and economicissues and is often quoted in the press. He holds a B.A. in classics from SUNY at Albany anda Ph.D. in economic history from Yale University. He lives in Alexandria, Virginia, with hiswife Perrine and three children, Benjamin, Brian, and Penelope.

Keisuke Nakashima is a research associate at the CSIS Global Aging Initiative and theco-director of the Global Policy Initiative, a virtual public policy think tank. His researchinterests include the economics of population aging and social security reform, and he is theauthor of numerous articles on aging trends in East Asia. Nakashima holds an M.A. ininternational relations from the Maxwell School of Citizenship and Public Affairs atSyracuse University and a B.A. in Anglo-American studies from Kobe City University ofForeign Studies in Kobe, Japan. He has also studied at the Paul Nitze School of AdvancedInternational Studies at Johns Hopkins University and at Newbury College.

Neil Howe is an economist, demographer, and historian who writes and speaksfrequently on the aging of the population, long-term fiscal policy, and generations in history.He is a senior associate at CSIS, where he works with the Global Aging Initiative, a seniorpolicy adviser to the Blackstone Group, and a senior adviser to the Concord Coalition. Heis also founding partner and president of LifeCourse Associates, a marketing, HR, andstrategic planning consultancy serving corporate, government, and nonprofit clients. Howeis the author or co-author of numerous policy studies and books, including On BorrowedTime (1988), Generations (1991), 13th-Gen (1993), The Fourth Turning (1997), andMillennials Rising (2000). He holds graduate degrees in history and economics from YaleUniversity. He lives in Great Falls, Virginia, with his wife Simona and two children, Giorgiaand Nathaniel.

48 China’s Long March to Retirement Reform

Notes 49

About CSISThe Center for Strategic and International Studies (CSIS) provides strategic insightsand policy solutions to decision makers in government, international institutions, the privatesector, and civil society. A bipartisan, nonprofit organization headquartered in Washington,DC, CSIS conducts research and analysis and develops policy initiatives that look into thefuture and anticipate change. Founded in 1962 by David M. Abshire and Admiral ArleighBurke, at the height of the Cold War, CSIS was dedicated to finding ways for America tosustain its prominence and prosperity as a force for good in the world. Since 1962, CSIShas grown to become one of the world’s preeminent international policy institutions, withmore than 220 full-time staff and a large network of affiliated scholars focused on defenseand security, regional stability, and transnational challenges ranging from energy and climateto global development and economic integration. Former U.S. senator Sam Nunn becamechairman of the CSIS Board of Trustees in 1999, and John J. Hamre has led CSIS as itspresident and chief executive officer since April 2000.

The CSIS Global Aging Initiative (GAI) explores the fiscal, economic, social, andgeopolitical implications of population aging. CSIS established GAI in 1999 to raiseawareness of the challenge and to encourage timely reform. Over the past nine years, GAIhas pursued an ambitious educational agenda—undertaking cutting-edge research projects,publishing high-profile reports, and organizing international conferences in Beijing, Berlin,Brussels, Paris, Seoul, Tokyo, Washington, and Zurich that have brought together worldleaders to discuss common problems and explore common solutions. To learn more aboutGAI, visit its website at www.csis.org/gai.

CSIS, 1800 K Street, NW, Suite 400, Washington, DC 20006202-887-0200 www.csis.org

AboutThe Prudential FoundationFounded in 1977, The Prudential Foundation is the nonprofit grant-making organizationof Prudential Financial, Inc. It is part of Prudential’s Community Resources Department,a strategic combination of four units: the Foundation, which strives to build children andfamilies’ self-sufficiency; the Social Investment Program, which originates and managessocially beneficial investments; Local Initiatives, which coordinates employee volunteerismand fosters community outreach; and Business Diversity Outreach, which facilitates diversemarket development and outreach efforts for Prudential businesses. The mission of theCommunity Resources Department is to strengthen Prudential by investing financialresources, business expertise and associate volunteer skills in programs that increasehuman potential and individual self-sufficiency through education, job skills, economicrevitalization and basic community health.

Prudential Financial, Inc. (NYSE: PRU), a financial services leader with approximately$602 billion of assets under management as of September 30, 2008, has operations in theUnited States, Asia, Europe, and Latin America. Leveraging its heritage of life insuranceand asset management expertise, Prudential is focused on helping approximately 50 millionindividual and institutional customers grow and protect their wealth. The company’s wellknown Rock symbol is an icon of strength, stability, expertise and innovation that has stoodthe test of time. Prudential’s businesses offer a variety of products and services, includinglife insurance, annuities, retirement-related services, mutual funds, investment management,and real estate services. For more information, please visit http://www.news.prudential.com/.

© 2009 by the Center for Strategic and International Studies. All rights reserved.