Policy Review Banner

Growing Old the Hard Way: China, Russia, India

Saturday, April 1, 2006

Over the past decade and a half, the phenomenon of population aging in the “traditional” or already affluent oecd societies has become a topic of sustained policy analysis and concern.1 The reasons for this growing attention — and apprehension — are clear enough.

By such metrics as median age or proportion of total population above the age of 65, virtually every developed society today is more elderly than practically any human society ever surveyed before the year 1950 — and every single currently developed society is slated to experience considerable further population aging in the decades immediately ahead. In all of the affluent oecd societies, the proportion of what is customarily called the “retirement age population” (65 years of age or older) will steadily swell, with the most rapid expansion occurring among those aged 80 or more. Simultaneously, the ratio of people of “working age” (the cohort, by arbitrary though not entirely unreasonable custom, designated at 15–64 years) to those of retirement age will relentlessly shrink — and within the working age grouping, more youthful adults will account for a steadily dwindling share of overall manpower.

Whether these impending revolutionary transformations of national population structure actually constitute a crisis for the economies and societies in the industrialized world is — let us emphasize — still a matter of dispute. To be sure: This literal upending of the familiar population pyramid (characteristic of all humanity until just yesterday) will surely have direct consequences for economic institutions and structures in the world’s more affluent societies — and could have major reverberations on their macroeconomic performance. Left unaddressed, the mounting pressures that population aging would pose on pension outlays, health care expenditures, fiscal discipline, savings levels, manpower availability, and workforce attainment could only have adverse domestic implications for productivity and economic growth in today’s affluent oecd societies (to say nothing of their impact on the global outlook for innovation, entrepreneurship, and competitiveness).

But a host of possible policy interventions and orderly changes in existing institutional arrangements offer the now-rich countries the possibility (to borrow a phrase from the oecd) of “maintaining prosperity in an aging society” — and in fact of steadily enhancing prosperity for graying populations. Today’s rich countries may succeed in meeting the coming challenges (and grasping the potential opportunities) inherent in population aging — or then again, they may fail to do so. The point is that an aging crisis is theirs to avert — and they have considerable scope for so doing.

In contrast to the intense interest currently accorded the issue of population aging in developed countries, the topic of aging in low-income regions has as yet attracted relatively little examination. This neglect is somewhat surprising, for over the coming decades a parallel, dramatic “graying” of much of the Third World also lies in store. The burdens of pronounced population aging, furthermore, are unlikely to be born as easily by countries still poor as by countries already rich. Simply stated, societies and governments have fewer options for dealing with the problems imposed by population aging when income levels are low — and the options available are distinctly less attractive than they would be if income levels were higher.

Over the next generation, it seems entirely likely — indeed, all but inevitable — that a large fraction of humanity, peopling countries within the grouping often termed emerging market economies, will find themselves coping with the phenomenon of population aging on income levels far lower than those yet witnessed in any society with comparable degrees of graying. For such countries, the social and economic consequences of aging could be harsh — and the options for mitigating the adverse effects of population aging may be fairly limited. In some of these countries, population aging could potentially emerge as a factor appreciably constraining long-term growth and development.

As we will detail in the next few pages, rapid and pronounced population aging represents a highly uneven, largely unappreciated, and as yet almost entirely undiscounted long-term risk for the world’s emerging markets.

 

Dynamics and dimensions

Venturing predictions about the world outlook 20 years hence is a hazardous enterprise. Nevertheless, we can state quite confidently that a tremendous wave of population aging is virtually certain to sweep through the developing regions between now and 2025. How can we talk so boldly — and so categorically — about events that have not yet unfolded? Over the course of two decades, a country’s economic or political circumstances can change tremendously. By contrast, given the stubborn realities of population change — demographic evolution tends to be gradual, contingent on previous developments, and tightly bound by existing social tendencies — there is inherently less leeway among plausible alternative demographic scenarios 20 years hence.

Barring a cataclysm of Biblical proportions, about five-sixths of the roughly 5.3 billion people alive in the less developed regions of the world today are likely still to be around in 2025 — and something like two-thirds of the approximately 6.7 billion population projected for those areas as of 2025 will comprise people alive now, already inhabiting those regions today.2

Between 2005 and 2025, current projections anticipate the population of the less developed regions to increase by about one-fourth. The most rapidly growing cohort within that population, however, will be the 65-plus group: Once again barring only catastrophe, the Third World’s senior citizens will roughly double in numbers over those years, to about 570 million person, or about 8.5 percent of the total for 2025.

While the less developed regions in 2025 would meet the United Nations’ definition of an “aging population” (i.e., with persons 65 and older accounting for over 7 percent of total population), their overall population profile would nevertheless be roughly as youthful as that of Europe from the late 1950s or Japan in the mid-1970s. But the extent and tempo of population aging will vary tremendously within the less developed regions: Some territories are slated to experience practically no population aging at all over the coming two decades, while others will become positively aged.

The demography of aging explains this differentiation. As a purely arithmetic matter (and perhaps somewhat counterintuitively), the scope and scale of population aging in any regularly convened human society is determined primarily by changes in its fertility levels, not by changes in mortality.3 It is low birth rates, rather than long life expectancies, that drive population aging — and in much of the Third World, fertility levels are already very low. Today, in fact, the majority of humanity probably already lives in countries with “sub-replacement fertility” regimes (that is to say, childbearing patterns which, if continued, would result in indefinite population decline, absent immigration).4

Since the world’s more developed regions account for less than a fifth (19 percent) of the current global population, this means that the great majority of the planet’s sub-replacement populations today are found in low-income regions. And since fertility levels in low-income areas continue to drop — even in spots where sub-replacement is already the norm — the momentum for rapid population aging continues to build.

Not in sub-Saharan Africa, to be sure: There, median age is likely to remain just barely over 20 years some two decades from now. And certainly not in those parts of the Arab/Islamic expanse where total fertility rate levels still apparently exceed five births per woman per lifetime (for example, Yemen, Oman, Afghanistan). But in much of East Asia, South Asia, the Middle East, Eastern Europe,5 and Latin America, sub-replacement fertility is now the norm: which is to say that the areas of the developing world presently set on a trajectory for rapid population aging are precisely the areas that encompass today’s most promising emerging market economies.

 

Emerging markets

To place the low-income regions’ population aging phenomenon in international economic perspective, we can begin by comparing the current correspondences between graying and income levels in today’s major emerging market economies with the old-age-to-income trajectories charted out by the major Western economies over the course of their postwar development (see Figure 1). As may be seen, the picture for the “emerging markets” is highly varied. On the one hand, some developing economies appear to be much more “youthful” today than Western Europe, the United States, or Japan at comparable levels of per capita income: Proportionately, for example, Turkey, Brazil and Mexico presently support barely half as many senior citizens as Western Europe did at those levels of per capita output. (The proximate explanation for this contradistinction is that Brazil, Mexico and Turkey all had much higher birth rates than the Western countries when approaching the $5,000 gdp per capita threshold.)

 

figure 1
Percentage of the Population Aged 65+ vs. GDP per Capita:
Developed Countries 1950–2000 vs. Emerging Economies 2000

Sources: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects: The 2004 Revision and World Urbanization Prospects: The 2003 Revision, http: //esa.un.org/unpp (April 19, 2005); Angus Maddison, The World Economy: A Millenial Perspective (Development Centre Studies, Organization for Economic Cooperation and Development: Paris 2001), 276, 304, 337, 144, 101, 184, and 194.

Note: gdp per capita in 1990 International Geary-Kamis Dollars. Europe data for 12 countries: Austria, Belgium, Denmark, Finland,.France, Germany, Greece, Italy, Netherlands, Norway, Sweden, Switzerland, U.K.


 

Some other emerging market economies appear to be at age–income coordinates almost identical to those traced by Western countries during their economic ascent: India and Thailand, for example, are each separately at points earlier delineated by Japan on its postwar aging/growth path.

But some other emerging market economies already have distinctly higher old age burdens than Western countries bore at similar stages of development. As of the year 2000, China’s 65+ cohort accounted for about the same share of population as Japan’s in 1970 — but per capita output was three times higher for Japan in 1970 than for China in 2000. By the same token, at the point when Western Europe and the United States had the same ratios of elderly to total population as registered by Russia in 2000, their per capita income levels were, respectively, three and six times higher than Russia’s.

Turning from the relatively recent past to the immediate future, Table 1 indicates the prospective dimensions of the coming wave of population aging for nine of the largest emerging market economies.

 

table 1
Population Aging in Developed Economies Today vs. Emerging Market Economies Tomorrow

Sources: Global Competitiveness Report 2004/2005, Annex Tables 1.01–1.03; World Development Indicators 2004; U.S. Census Bureau International Database.


 

By un criteria, all nine will count as aging societies by 2025 — but not surprisingly, the degree of population aging differs greatly from one country to the next. India, Indonesia, and Mexico, for example, will still likely have more youthful age profiles in 2025 than the United States of 2003. By 2025, on the other hand, Turkey and Brazil will be roughly as gray as the United States today, but not yet so gray as contemporary Japan or today’s Western Europe. By 2025, China and Thailand will likely have population profiles almost as elderly as today’s eu–15. Russia and Poland, for their parts, will likely have populations more aged in 2025 than Japan’s today: That is to say, they will be grayer than any population yet seen in human history.

Although most of these emerging market economies will have age profiles similar to — or in some cases, even more extreme than — today’s developed economies, their income levels are far lower than those of the affluent oecd societies today — and in almost every case will almost certainly remain below today’s oecd levels two decades from now.

Over a 22-year interim (i.e., 2003 to 2025), Poland could reach today’s eu–15 Purchasing Power Parity (ppp)-adjusted per capita output levels if it grew at a steady 4 percent per capita per annum — an ambitious though hardly impossible hope. To hit current Japanese ppp-adjusted income levels, however, per capita Russian growth would have to be maintained at 5 percent a year for nearly a quarter-century. Attaining current eu ppp-adjusted income levels by 2025 would require annualized per capita growth rates of 8 percent for China and over 10 percent for India. One need not be a Russo-pessimist, a Sino-pessimist, or an Indo-pessimist to suggest such tempos may be out of reach.6

A closer examination of the Chinese, Russian, and Indian aging problems can lay out the constraints over the coming decades for each of these economies in greater detail.

 

China: A tightening “triple bind”

China’s impending aging wave is illustrated by Figure 2, which compares the country’s actual population structure in 2000 with its projected profile for 2025. China has evidently experienced subreplacement fertility levels since the early 1990s.7 Consequently, as may be seen from these projections, every age group under 35 years of age is anticipated to be smaller in China in 2025 than it was in 2000 — while all of the older groups are expected to be larger. China’s coercive population control program may have succeeded in limiting numbers for the country’s rising contingents of young people, but its elderly population will be exploding in the years ahead — waxing at an annual pace of something like 3.5 percent per year. Between 2005 and 2025, about two-thirds of China’s aggregate population growth will occur in the 65+ grouping — and that cohort will likely double in size, to roughly 200 million people. By 2025, under current un and Census Bureau projections, China would account for less than a fifth of the world’s population but almost a fourth of the world’s senior citizens.

 

figure 2
Estimated and Projected Population Structure of China: 2000 vs. 2025

Source: United Nations Population Division, World Population Prospects 2004 Revision. Available online at http://esa.un.org/unpp/index.asp?panel=2.


 

China, of course, is a vast land with plenty of regional variation: As regards aging and economic development, the country’s provinces are characterized both by notable differences in provincial demographic profiles and truly tremendous disparities in levels of per capita output. (In 2001, local per capita gdp was reportedly 13 times as high in bustling Shanghai as in isolated Guizhou.) While there is a broad overall correlation between graying and income levels in China’s provinces today, the correspondence is not a tight one — and over the next two decades, some of China’s most dramatic aging trends are set to unfold in some of the areas that are poorest today.

We can see this when we compare the U.S. Census Bureau projections of Chinese provincial age structures for 2025 with official Chinese data on per capita levels of gdp by province as of 2001. Currently (2005), Japan reports the world’s highest proportion of persons 65 and older, at about 19.5 percent of total population. By 2025, however, the 65-plus cohort is projected to account for 21 percent of total population in one part of China — and that place is not relatively prosperous Beijing or booming Shanghai. Rather, it is Heilongjiang, in China’s Manchurian rustbelt — where per capita provincial output (at current exchange rates) was just over $1,100 in 2001.

Currently, Japan also reports the world’s highest median age, at about 42.5 years. Yet as of 2025, nine of China’s 31 provinces and major municipalities are projected to have higher median ages than contemporary Japan: among them, Liaoning (where the exchange-rate-based gdp per capita was around $1,450 in 2001); Jilin (around $925); and Chongqing (under $690). (Heilongjiang’s projected median age in 2025, we might note, would be over 51 years.) No purchasing-power-parity adjustments of these exchange-rate-based figures can alter the basic message they impart: In just 20 years, large parts of China will have to support very aged populations on very low average income levels.

How will China’s elderly population be sustained financially two decades hence? All uncertainties about the future notwithstanding, we can fairly confidently surmise that these prospective pensioners will not be supported through the country’s existing state pension arrangements. For all the justifiable anxiety about the current actuarial health of national pension systems in various oecd countries, the financial disarray of China’s official pension apparatus is in a league of its own. For this patchwork, covering perhaps a sixth of the total Chinese workforce, the net present value of unfunded liabilities is estimated to exceed current gdp — perhaps substantially. China’s pension system is clearly unsustainable — but despite the better part of a decade of high-level policy deliberations in Beijing on the national pension problem (the issue has been addressed by directives from the highest levels of government since 1997), no alternative formula has been officially offered.8 Nor is promulgation of legislation for a new and unified social security system obviously on the policy horizon for China today.9

Thus China’s national pension system as of 2025 promises today to be more or less the same system that has always provided for the country’s elderly and infirm: namely, the family unit. But herein lies a problem: The “success” of the Chinese government’s continuing antinatal population drive will necessarily translate in coming decades into a plummeting ratio of working-age children to elderly parents. Whereas the average Chinese woman who celebrated her sixtieth birthday in the early 1990s had borne five children during her lifetime, her counterpart in 2025 will have had fewer than two.

No less important: Over the next 20 years, China’s rising cohorts of prospective retirees face a growing “son deficit.” In Chinese cultural tradition, it is sons, rather than daughters, upon whom first duty for support of aged parents customarily falls (a daughter is obliged to care for her husband’s parents as well as her own). In the early 1990s, about 7 percent of China’s 60-year-old women had never borne a son. Today that proportion is about 10 percent. By 2025, the figure will shoot up to roughly 30 percent. Some of the male children in question, furthermore, will not live to adulthood to be able to help support their parents. Taking both fertility and survival trends thus into account, it seems likely that a third or more of Chinese women approaching retirement age two decades from now will have no living sons. For tens of millions of aged Chinese just 20 years from now, seeking financial and material help from one’s children will amount to competing for resources with one’s son-in-law’s parents (given the presumed continuation of China’s long-standing cultural norm of near-universal marriage for women). Suffice it to say that such a “niche” is not promising from the perspective of social ecology.

With official government pension guarantees a distinctly more limited and problematic set of options than official policy might wish, and with the traditional social security system known as “the Chinese family” a rather more fragile construct than at any time in the recent past, the grim reality may be that a great many elderly Chinese men and women in the coming decades will have to come to the conclusion that they must sustain themselves by continuing to work. Paradoxically, despite China’s tremendous material progress over the half-century beginning in 1975, the nation’s elderly will face a continuing need — quite possibly a growing need — to support themselves in old age through their own labor.

But China’s elderly population is not ideally placed as competitors in their country’s labor markets — either today or tomorrow. And here we enter into the second constraint of China’s tightening “triple bind.” China’s elderly workers occupy a decidedly unfavorable position in the country’s labor force today. At the start of the new century, in comparison with China’s overall workforce, workers 65 or older are six times as likely to be illiterate or semiliterate, almost 50 percent more likely to have only primary education, and only a tenth as likely to have a high school or college diploma. They are also much more likely to toil in the agricultural sector: In 2000, 87 percent of China’s elderly workers were in farming, as opposed to 66 percent for the workforce as a whole.10 Thus consigned to the low-income sector of the economy, to labor there with low levels of human capital, China’s older labor force provides almost a textbook definition for the working poor.

Despite China’s educational advances, its older population will still be disadvantaged in 2025. We know this because we have data today on the educational attainment of the people who will make up China’s elderly cohorts 20 years from now As of 2025, something like two-fifths of China’s senior citizens will have a primary school education — or less. That circumstance contrasts starkly with prospects for today’s developed countries: In both Japan and the United States, for example, nearly five-sixths of the 65-plus population will have at least a high school diploma in 2025.

In 2025, moreover, the fate of China’s low-skill older workers will still primarily be to toil in the field. Despite rapid structural transformation in the Chinese economy, agriculture promises to remain a major source of employment two decades from now — and older workers are likely to remain over-represented in China’s primary sector.

The point to bear in mind about farming in China is that the occupation entails regular strenuous activity: It is not only low-paying but physically demanding. This point can be drawn more broadly, for even in occupations Western readers do not commonly associate with physical exertion, stamina and muscle-power are routinely required for job performance in China. This is so, quite simply, because China still lacks the capital investment per worker to provide “labor saving” alternatives to human strength in the production process. With mechanization so much more limited in China than in Western economies, the machines powering much of Chinese economic activity today are human bodies — and this circumstance is unlikely to change appreciably over the next 20 years.

Perversely, because China’s older workers suffer from lower levels of education and training than the general labor force, they are precisely the cohort most directly obliged to rely upon physical effort to earn a living. This will hold true tomorrow as well as today. And that brings us to the third strand of the “triple bind” that defines China’s looming aging problem: In the years ahead, China’s senior citizens are not only likely to face real and perhaps mounting pressures to support themselves through paid labor, and not only likely to find that their employment opportunities are principally in low-paying, physically demanding jobs; they are also likely to be less healthy and more fragile than counterparts in other countries where the physical demands of employment are much less forbidding for the elderly and nonelderly alike.

Upon reflection, the proposition that the health of senior citizens in China is more tenuous than that of older populations in Western countries should not surprise: Over their life histories, after all, older people in China have on average been exposed to more in the way of health and nutritional risk, and have less scope for protecting and/or recovering from illness and injury, than older people in affluent societies. In all, China’s availability and quality of food, housing, education and health services (and many other factors influencing health status) do not compare favorably with oecd levels. Consequently, China’s senior citizens live shorter lives than old people in oecd countries — and their remaining years are more heavily compromised by serious health problems.

Estimates of “disability-free life expectancy” in China and Japan at age 65 make the point.11 (These particular data come from circa 1990, but the patterns they revealed still obtain.) For men and women alike, overall life expectancy at age 65 was over 50 percent longer in Japan than in China — but whereas Chinese men could reportedly expect to be encumbered by disability for nearly a third of their remaining years, the fraction was just half as high among Japanese men. Chinese women, likewise, could reportedly expect to spend a third of their final years affected by disabilities — whereas the corresponding fraction for Japanese women was one-fifth. These data, incidentally, may considerably understate the extent of health impairment in later life for China’s elderly. For one thing, they only refer to disability — not to the broader matter of potentially serious health problems. Secondly, the data on disability come from self-assessments — and as is well known, subjective expectations regarding health tend to be lower for populations with lower income and educational levels.

China’s outlook for population aging, in sum, can in some respects be likened to a slow-motion humanitarian tragedy already underway. On the current trajectory, the graying of China threatens many tens of millions of future senior citizens with a penurious and uncertain livelihood in an increasingly successful emerging market. The incidence of individual misfortune implied by current trends looks to be sufficiently broad to suggest that impoverished aging may emerge as a major social problem for China in the years ahead. The impending fault lines for impoverished aging, furthermore, promise to magnify yet further the social inequalities with which China is already struggling.

How the country will deal with the social and political tensions generated by impoverished aging remains to be seen. Since 1997, Beijing’s policymakers have recognized the question but have been deferring the response to it. Quite predictably, feasible policy options have narrowed over these years of indecision — and the remaining alternatives are steadily becoming more expensive.

 

Russia: Graying with “unhealthy aging”

In some respects, Russia’s outlook for population aging can be regarded as an ordinarily “European” tale. Between 2000 and 2025, total population is expected to decline, while the number of Russians 65 and above is slated to grow substantially. Consequently, Russia’s “population support ratio” or psr — a metric that relates the working age population (15–64 years) to the retirement age population (65+) — falls by Census Bureau projections over this quarter-century from 5.5: 1 to 3.3:1.

The implied burden of potential pensioners on the potential Russian workforce of 2025 is high by contemporary standards, to be sure — but such ratios look unexceptional for the Europe that awaits in 2025. By Census Bureau projections, the psr for all of Europe would be about 3.0: 1 in 2025, indicating the Russia would have a slightly higher working age population to retirement age population than the rest of the continent. For 2025, in fact, Russia’s proportion of population 65 and older promises to be a bit higher than America’s and a bit lower than Western Europe’s.

It would be tempting to describe Russia’s prospective aging profile as that of a typical developed society — and thus to expect Russia’s aging problem to be similar to those of oecd Europe. The weight of the aging burden that Russian society will have to bear in coming years, however, cannot be measured adequately by population pyramids or “population support ratios” alone. Russia’s particular vulnerabilities in population aging pivot not so much on the size of the nation’s prospective elderly population as on the exceptional fragility of the workforce that will be expected to support it.

Pronounced long-term deterioration of public health for an industrialized society during peacetime is a highly anomalous, indeed counterintuitive proposition for the modern sensibility: Nevertheless, over the four decades between 1961–62 and 2003, life expectancy at birth in Russia fell by nearly five years for males. It also declined for females, although just slightly, making for an overall drop in life expectancy of nearly three years over the past four decades. Between the mid-1960s and the start of the twenty-first century, the country’s age-standardized death rates climbed by over 15 percent for women and by a shocking 40 percent for men. This upswing in mortality was especially concentrated among the group of “working age,” where the upsurges in death rates were breathtaking. (Between 1970–71 and 2003, for example, every female cohort between the ages of 25 and 59 suffered at least a 40 percent increase in death rates; for men between the ages of 30 and 64, the corresponding figures uniformly exceeded 50 percent, and some cases exceeded 80 percent.12)

To get a sense of just how bad health and mortality conditions are for Russian adults nowadays, we can compare survival rates for Swiss men with their Russian counterparts for 1999, thanks to information compiled by the Human Mortality Database13 (see Figure 3). In Switzerland, a 20-year-old man had a five-out-of-six chance of making it to 65 years of age; in Russia, he stood less than even odds. (In 1999, by the way, Russian adult survival schedules were somewhat better than they are today.)

 

figure 3
Male Survival Schedule, Ages 20–65: Russia vs. Switzerland, 1999

Source: www.mortality.org, accessed January 14, 2005.


 

If Russia’s adult mortality levels are so “unnaturally” high today, won’t more “natural” levels more or less automatically reassert themselves in the coming decades? The dismaying answer is: not necessarily, for in Russia’s demography the “abnormal” seems to have become the new norm. Unlike other parts of the industrialized world, Russia’s health trends are characterized by a heavy measure of what we might call negative momentum: that is to say, an unfavorable accumulation of immunological insults and health risks in today’s adult population by comparison with their parents’ generation. To the extent that death rates provide evidence about general health conditions, modern Russia’s mortality data strongly suggest that each new cohort is more fragile than its predecessor.

In other industrialized Western societies in the postwar era, younger generations have routinely come to enjoy better survival rates than their predecessors: In contemporary Japan, for example, men born in 1950 have, over their adult life course thus far, experienced age-specific death rates 30 percent to 80 percent lower than those recorded for the cohort born 20 years before them.

By contrast: In Russia today (to invert the popular “boomer” mantra) “30 is the new 40.” That is to say, there has been no improvement in survival schedules among the two generations of Russian men born between the late 1920s and 1970. Quite the contrary: Over the life course, each rising cohort of Russian men seems to be charting out a slightly more dismal mortality trajectory than the one traced by its immediate predecessors. (The same sorts of patterns, incidentally, are evident among Russian females.)

This negative momentum makes the objective of major, sustained improvements in public health especially problematic. Partly for this reason, demographers generally have low expectations for health progress in Russia over the years immediately ahead. The U.S. Census Bureau, for example, imagines the male life expectancy in Russia will steadily lag below India’s, Pakistan’s, and even Bangladesh’s through 2025.14

Russia’s lingering health and mortality crisis promises to be an anchor against rapid economic development, frustrating the effort to move Russia onto a path of swift and sustained material advance. It is difficult to see how Russia can expect, in some imagined future, to achieve an Irish standard of living if its labor force still faces an Indian schedule of survival — or worse. Widespread debilitation and premature mortality among working age cohorts depresses economic potential directly and immediately — but also has adverse and far-reaching effects on longer-term productivity. The expectation of a seriously foreshortened working life alters the cost-benefit calculus for higher education and technical training, lowering investment in human capital. And since Russian working-age adults “present” as far older than Western counterparts of the same calendar birth-year,15 the scope for “economically active aging” — for enhancing the labor force participation rates and economic contributions of persons in middle age and beyond — will be far more constrained for Russia than for oecd Europe.

Population aging in the context of unhealthy aging poses additional special economic and social challenges to Russia. Given the country’s steep and forbidding age-specific health gradient today and the limited prospects for health improvements over the coming two decades, the prospective aging of Russia’s working age population — the median age of the population within the 15–64 grouping will be about 42 years in 2025, three-and-a-half years higher than today — means that the health and mortality outlook for Russian manpower could actually be less favorable than today — perhaps even appreciably so. Second, the specter of a swelling population of elderly pensioners16 dependent for support on an unhealthy and diminishing population of low-income workers suggests some particularly unattractive trade-offs between welfare and growth. Should Russian resources be allocated to capital accumulation or to consumption for the unproductive elderly? Given Russia’s population structure, the question cannot be finessed. As of the year 2002, Russia had only 1.7 workers for every pensioner17 — and that ratio will only fall in the years ahead. Though the government officially embarked upon pension reform in 2002, that process looks to be a long and complicated one — and what the eventual arrangements will presage on the one hand for the availability of investable funds, and on the other for living conditions of Russia’s steadily growing ranks of the elderly and the infirm, is as yet an open question.

 

India: A tale of “two countries”

India’s population profile will age over the coming 20 years, but the country will nevertheless remain relatively youthful. Although projections indicate that India’s 65+ cohort is slated to double in size between 2005 and 2025, those elders account for less than 8 percent of overall population 20 years hence; the country’s median age then is only just over 30; and the psr is almost 9:1 — a level last witnessed in today’s more developed countries before the Second World War.

But in many ways vast India is a sort of arithmetic expression that averages the sum of its many diverse components: so, too, with population aging. Closer examination reveals that with population aging there are in reality two Indias, with very different aging prospects and challenges: one that stays remarkably youthful over the next 20 years; the other already embarked on a very rapid graying.

As already noted, the pace and scale of aging tomorrow is always largely determined by local levels of fertility today. India’s total fertility rate has dropped by more than two-fifths over the past three decades — from about 5.4 births per woman per lifetime in the early 1970s to approximately 3.1 today — but the pace of change has varied strikingly from one region and setting to the next.

It will surprise some readers to learn that sub-replacement fertility today prevails in many of India’s huge urban centers — New Delhi, Mumbai (Bombay), Kolkata (Calcutta), and Chennai (Madras) among them.18 But even more surprising, throughout much of rural India — especially rural South India — fertility levels today are also near or already below replacement.

Dr. P.N. Mari Bhat of Delhi University’s Institute for Economic Growth has laid out the implications of these discrepant patterns for future aging in the supra-statal regions he labels “north India” and “south India.”19 Currently (2005), fertility levels for the roughly half-billion population of this “north” are almost twice as high as for the nearly quarter-billion people of this “south.” (India’s remaining 350 million people live in states and union territories not included in Mari Bhat’s “north/south” analysis.)

By 2025, “north” India’s population would still be very young. Its projected median age would be just 26 — and the 65+ group would account for less than 6 percent of total population. On the other hand, “south” India’s population structure in 2025 would bear unmistakable signs of population aging. There, median age would be about 34 (a level comparable to Europe’s in the late 1980s), and 9 percent of the population would be 65 or older (about the same share as Japan’s in 1980).

Mari Bhat does not break down his 2025 projections to the state level, but Professor Tim Dyson of the London School of Economics has done so for India for 2026.20 His projections differ from Mari Bhat’s in some particulars (most importantly, in positing a slightly faster pace of fertility decline over the next generation), but his calculations similarly depict a growing “aging gap” between north India and south India — with aging by 2026 already having progressed considerably in a number of southern states. In each of India’s four southernmost states, median age would be over 35 — and over 37 in both Kerala and Tamil Nadu. (For reference, think of Western Europe around 1990.) In these projections, Kerala and Tamil Nadu’s proportion of persons 65 and older both exceeded 10 percent by 2025. (Think here of Japan in the mid-1980s.)

A generation before Western Europe’s median age reached 35, however, the region’s gdp per capita was in the range of $6,000–$8,000 (at 2000 prices and exchange rates); a generation before Japan’s share of 65+ to total population hit the 10 percent mark, the country’s per capita gdp (at constant 2000 exchange rates and prices) was likewise around $7,000.21 In 2001 — that is to say, 25 years before the two states in question are projected to comport to those same demographic specifics — exchange-rate-based gdp per capita was less than $450 a year in both Kerala and Tamil Nadu. By any international or historical benchmark, indeed, many areas of India are facing the onset of rapid population aging on current levels of per capita output that are astonishingly low.

At the moment, India has nothing like society-wide old-age pension coverage: To the contrary, only about 11 percent of India’s workforce participates in any sort of guaranteed retirement income systems. (An “emergency” needs-based monthly stipend is publicly available for persons over 60 in India, but this mechanism offers less than $2 a month to its beneficiaries and is not guaranteed to be available to all who apply for it and meet its hardship qualifications.) Although Indian policymakers and academics have been discussing alternative potential paths to universal old age income protection with a reasonable measure of intellectual seriousness in recent years, no plans for national coverage are even remotely on the national policy agenda.22 Lacking a tangible comprehensive national retirement pension policy, India’s implicit strategy for meeting its coming aging challenge is, at least for now, to “grow its way through it.” Like many unspoken and thus unexamined game plans, this one looks highly problematic. Even if India, like the Japan of an earlier day, were poised to grow at a 5.5 percent per capita rate per annum over the coming generation, significant parts of India would be approaching the advent of an “aged society” on income levels almost an order of magnitude lower than those of Japan in the mid-1980s.

A sustained 5.5 percent per capita growth rate for India over the next generation, furthermore, is hardly a matter that can be taken for granted at the moment. In the period since her 1991 economic crisis, India has averaged a highly respectable 4.0 percent annual rate of per capita growth and has become a presence in the global it economy through its enclaves in places like Bangalore. But Bangalore — like the rest of the Indian south — is part of what may soon be known as “old India”: While its labor force is relatively skilled, it is also older, and absolute supplies of available manpower will soon peak and begin to shrink. Other parts of India, by contrast, will have abundant and growing supplies of labor — but a disproportionate share of that manpower will be either entirely unschooled or only barely literate.

Figures 4 and 5 highlight this paradox. They compare projected age/sex/education pyramids in 2026 for the Indian states of Kerala and Bihar (work by Anne Goujon of the Vienna Institute for Demography and Kirsty McNay of Oxford23). The extreme contrast is intentional — Kerala is India’s most educated state, while Bihar is its least schooled — but the story from the graphics is clear enough. In the Kerala of 2026, almost everyone of working age (15–64) will have some schooling, and the majority of the economically active manpower will have a high school diploma or better — but the largest population cohort would have been people in their 40s, with every successive cohort a little bit smaller. In the Bihar of 2026, on the other hand, each new birth cohort entering the labor pool would be larger than the one before — but fewer than one-third of the economically active population (15 to 65 years) would have even completed grade school, and well over two-fifths of the economically active age population would be illiterate, with no schooling whatsoever. (These projections, one may feel compelled to note, posit a continuation on into the future of the progress achieved over the 1990s in expanding access to education in India.)

 

figure 4
Age and Education Pyramid for Kerala, India: 2001 and 2026

Source: Anne Goujon and Kirsty McNay, “Projecting the educational composition of the population of India: selected state-level perspectives.” Applied Population and Policy 2003:1(1) 25–35, Figure 1.


 

figure 5
Age and Education Pyramid for Bihar, India: 2001 and 2026

Source: Anne Goujon, and Kirsty McNay, “Projecting the educational composition of the population of India: selected state-level perspectives,” Applied Population and Policy 1:1 (2003), Figure 1.


 

To appreciate how very educationally disadvantaged India stands to be a generation hence, we may compare the Goujon–McNay projections with earlier work by Robert J. Barro and Jong-Wha Lee on historical patterns of worldwide educational attainment.24 By the Goujon–McNay projections, in 2026 nearly one third (32 percent) of Indians 25 years of age or older would be illiterate, with no formal schooling. By contrast, as of 1960 — that is to say, nearly three generations earlier — the illiteracy rate for the 25+ group in 23 oecd countries would have been about 6 percent. Even more telling, perhaps, are comparisons with educational levels from other developing regions. To go by the Goujon–McNay projections and the Barro–Lee estimates, India’s rate of adult 25+ illiteracy in 2026 would be roughly comparable to the levels in Latin America and the Caribbean or East Asia around 1970 (35 percent and 31 percent, respectively) — that is to say, two generations earlier. Indeed, striking as this may sound, India’s future adult illiteracy rates would not be that much lower than the current (2000) levels prevailing in sub-Saharan Africa (43 percent).

Educated and aging, or untutored and fertile — this looks to be the contradiction for India’s development in the years immediately ahead. Human resources are the foundation for economic growth in the coming century — but if poor health is the Achilles Heel of Russia’s human resources base, seriously inadequate educational opportunity for all too much of the population looks to be the Achilles Heel for India.

 

What is to be done? Can anything be?

In any plausible future, population aging in low-income Eurasia is unlikely to be pretty. Absent only unthinkable catastrophe, the trajectory for a rapid graying is already essentially set for China, Russia, and much of India. Even with rapid and uninterrupted economic growth, by 2025 these aging or aged societies will be far poorer than was the West when Western societies had to face comparable degrees of population aging. There is no “fix” to this basic dilemma — at least, not in the next two decades.

That being said, it is also clear that informed and deliberate public action can still mitigate some of the adverse consequences that rapid graying would otherwise entail.

At first glance, the obvious policy avenue for coping with rapid population aging in Russia, China, and India might appear to be strengthening — and broadening the coverage of — the national pension arrangements in the countries in question. Certainly there is plenty of room for improvement here: Of the three countries, only Russia has a national pension system with anything approaching universal worker coverage (and the soundness of the evolving Russian system still remains to be seen). Moreover, given the enormous economic disparities within all three of these countries, even a relatively small measure of redistribution within the workings of a nationwide pension system could provide an important measure of protection for the most vulnerable of the elderly (a group that will be disproportionately rural, agrarian, and female).

In actuality, however, the potentialities of pension reform may offer less opportunity in each of these countries than would first seem to meet the eye. Post-communist Russia’s polity, for example, betrays scant interest (on the part of either policymakers or voters) in protecting the health and well-being of economically productive citizens, much less social dependents. By the same token, the idea of extending pension coverage to the countryside and the slums is not unpopular in India: It simply remains beyond the realm of serious policy discussion. As for China — where political decision making is, shall we say, more “streamlined” — policy interest in extending the nation’s pension coverage to poorer rural regions is indeed evident, but action has been precluded by the immense obstacle of the vast existing unfunded pension liabilities for comparatively well-off urban and state employees. However separate those two questions may appear intellectually, in practical terms they are inseparably linked, and the government’s inability to deal with the latter means it cannot make progress on the former either.

Perhaps counterintuitively, then, the best hope for controlling the toll of the aging tsunami in low-income Eurasia may lie outside pension policy — more specifically, in embracing politically feasible strategies for poverty-reducing growth. And here it is not hard to identify directions worthy of vigorous pursuit.

In China, for example, the country’s banking and financial system fairly begs for overhaul — and a transition to a more transparent, efficient, and sustainable system for financial intermediation will be essential if China is to maintain economic growth in the lower savings rate era that surely lies ahead.

In Russia, public health interventions to reduce premature mortality not only could reduce the country’s tragic and unnecessary loss of life, but also could augment the size and the productivity of the working-age population that will be sustaining pensioners (and preparing for self-financed retirement).

As for India, with outward economic orientation and sound macroeconomic policies, high rates of return can be expected from investment in human resources — and the highest returns of all may be expected from extending primary education to those currently excluded from school. (Reaching the young is critical here: A century of efforts around the world has repeatedly demonstrated the limited efficacy of such programs.) India is already committed rhetorically to “Education for all” — but achieving universal primary education will require much more than words from the country’s political leadership — and there is no good reason that India should be the sole great modern democracy incapable of meeting this objective.

For China, the case for financial sector reform wins on its own merits — as does the case for public health policy in Russia and universal primary education in India. Yet such policies would have the subsidiary, but hardly trivial, effect of making the scale of the aging problem a little less unmanageable in each of these societies.

 

Waiting for the “demographic dividend”?

Not all emerging markets, of course, are poised to settle their aging populations upon income levels so low as to be without historical precedent. Such middle income economies as Mexico, Brazil, and Turkey — among others — promise to have relatively youthful populations in 2025. Given current productivity levels and reasonable future economic prospects, moreover, a number of these places promise to fall well within the Western age-income experience over their presumed ascent in the coming decades — or perhaps even to remain more youthful than the Western economies were at similar levels of per capita income.

Is there a special economic benefit to be had from relatively youthful development? By some thinking in the economics literature — a hardy strand reaching back nearly half a century — the answer would seem to depend not just upon graying, but upon the ratio of all dependents, young and old, to economically productive workers.

To this way of thinking — oversimplifying the argument, some might say brutally — the lower the “dependency ratio” of children and oldsters to working age people, the higher the rates of capital accumulation and growth. In the argumentation of some economists today, the special moment in the “demographic transition” when birth rates are low and elderly populations are not yet burgeoning provides a sort of “sweet spot” for very rapid economic growth. According to this viewpoint, societies that have recently transited from high to low birthrates — but have yet to experience significant graying — are positioned to reap a “demographic dividend.” In this telling, indeed, crucial shifts in demographic structure are integral to the modern era’s great development success stories, from East Asia’s remarkable economic ascent over the past generation-and-a-half to Ireland’s more recent transformation into Europe’s “Celtic tiger.”25

The tale of the demographic dividend is certainly compelling to hear, and it may be heartening for those low-income societies whose dependency ratios are now about to begin a long-term decline — but is the story right? Are dependency ratios critical to a population’s prospects for sustained and rapid material advance?

It is surely incontestable that per capita output will be higher, all else being equal, in the society where people of working age account for the greater share of total population. And theoretically, variations in dependency ratios may also plausibly be associated with changes in savings ratios — or more strictly speaking, with changes in a population’s disposition to save and the facility with which it can attempt to accumulate savings. But in an era of truly global capital markets, relatively low savings rates in any given locale would not necessarily look like a binding constraint on that area’s development. Postwar economic history would seem to indicate that dependency ratios are not a decisive influence on long-run economic performance in our increasingly globalized times. Consider the developmental records of two major global regions — East Asia on the one hand, and Latin America and the Caribbean on the other.

Over the 40 years between 1960 and 2000, dependency ratios for the low-income countries in East Asia and Latin America/Caribbean traced quite similar trajectories (although dependency ratios were always somewhat lower in East Asia). Despite the close correspondence between the indicators for the two regions over the course of the past four decades, economic results in the two areas have been anything but similar: Per capita output septupled for East Asia, whereas it rose by about 85 percent for Latin America and the Caribbean. In both regions, moreover, the dependency ratio commenced a steady decline around 1975. But while the pace of growth accelerated in East Asia in the quarter-century following this drop, Latin America suffered a growth slowdown. In the high dependency ratio era (1960–75), Latin America’s per capita output growth averaged almost 3 percent a year; in the years of steadily falling dependency ratios (1975–2003), annual per capita growth was under 1 percent.26 Clearly other factors — including government policy — are more important in determining prospects for material progress.

P.N. Mari Bhat’s cautious assessment of the possible influence of the demographic dividend on economic growth for India could be extended to other low-income areas as well:

[D]uring the next 10–20 years demographic conditions would be favorable for growth. . . . However, as Bloom and Williamson note, their effect is by no means inevitable. To realize the effect, it is necessary to support it with appropriate economic, social and political institutions and policies. Otherwise it would only lead to higher levels of unemployment.27

Countries with declining dependency ratios, in short, may enjoy some potential for economic advantage — under the right cultural, institutional, and policy conditions. But demographic dividends are nothing to bank on.

 

Global implications?

All other things being equal (which is to say, without proactive policy and institutional adjustments), the pronounced population aging that awaits the affluent oecd countries over the generation ahead would be expected to have a depressing effect on both local savings rates and local rates of economic growth. Much the same may also be said of the aging trends that are due to affect some of the major “emerging markets” in the next few decades. The conjunction of population aging in the world’s developed economies and important parts of the developing world naturally raises the question of potential global impact. Global capital markets may be efficient in allocating investment to promising countries, corporations, and projects — but the availability of capital will affect the cost of capital and thus the profitability or attractiveness of undertakings worldwide. By the same token economic slowdowns in one major region would be expected to have spillover impacts on growth in other regions in an environment of liberalized global trade. Will the aging of the Third World have unanticipated spillover effects for the world economy? The answer is not yet clear — but it is none too early to begin asking the question.

1A select representation might include: Alan J. Auerbach, Laurence J. Kotlikoff and Robert Hageman, “The Dynamics of an Aging Population: The Case of Four oecd Countries,” NBER Working Papers 2797 (1989); David Cutler et al., “An Aging Society: Opportunity or Challenge?” Brookings Papers on Economic Activity 1 (1990), 1–56; Richard Disney, Can We Afford To Grow Older? A Perspective on the Economics of Aging (mit Press, 1996); oecd, Maintaining Prosperity in an Ageing Society (Paris: oecd, 1998); United Nations Population Division, Replacement Migration: Is It a Solution to Declining and Ageing Populations? (un Department of Economic and Social Affairs, March 2000); Peter S. Heller, Who Will Pay?: Coping with Aging Societies, Climate Change and Other Long-Term Fiscal Challenges, (imf, 2003); Landis MacKellar et al., The Economic Impacts of Population Ageing in Japan, (Elgar, 2004); James Proterba, “The Impact of Population Aging on Financial Markets,” NBER Working Papers 10851 (2004); International Monetary Fund, World Economic Outlook: The Global Demographic Transition (imf, September 2004).
   More popularized presentations include Peter G. Peterson, Gray Dawn: How the Coming Age Wave Will Transform America — and the World (Times Books, 1999); Robert Stowe England, Global Aging and Financial Markets: Hard Landings Ahead? (csis Press, 2002); Philip Longman, The Empty Cradle: How Falling Birth Rates Threaten World Prosperity and What To Do About It, (Basic Books, 2004); and Laurence J. Kotlikoff and Scott Burns, The Coming Generational Storm: What You Need To Know About America’s Economic Future (mit Press, 2004).

2Demographers today generally use global population projections by the United Nations Population Division and the United States Census Bureau as their reference standards: unpd’s “World Population Prospects” are available electronically at http://www.unpopulation.org; uscb’s “International Data Base” can be accessed at http://www.census.gov/ipc/www. Either set of projections (unpd “medium variant” or uscb sole-variant) can be used to support the calculations above.

3An illustrative example will emphasize the point. In a high-mortality society where female life expectancy stayed at 50 years and births per lifetime forever averaged six per woman, median age would eventually stabilize at about 16 years. On the other hand, if female life expectancy were 50 but births per lifetime averaged two instead of six, median age would stabilize close to 40! With an average of two children and a female life expectancy of only 50, in fact, the 65-plus group would ultimately account for over 15 percent of total population — a higher fraction than that estimated for the “more developed regions” at the dawn of the twenty-first century. (Estimates derived for “stable population” structures with female life expectancy of 50 under either West, North, East, or South “model” life tables — cf. Ansley J. Coale and Paul Demeny with Barbara Vaughan, Regional Model Life Tables and Stable Populations (Academic Press, 1983) and uscb International Data Base.)

4Chris Wilson and Gilles Pison, “La majorité de l’humanité vit dans un pays où la fécondité est basse,” Population & Sociétés 405 (October 2004).

5Technically, Eastern Europe is defined as “more developed” rather than “less developed” in the un’s global taxonomy. Because these “more developed” countries are nonetheless relatively low-income societies, we will include them for consideration here.

6The task of reaching today’s oecd per capita income levels, incidentally, may be even more daunting than those numbers suggest, for China, Russia and India all enjoy extremely generous ppp adjustments in this table, each of which scales up the country’s actual exchange-rate-based per capita output level by a factor of four or more. By actual exchange-rate-based per capita gdp estimates, Russia’s is one-fifteenth of Western Europe’s, and China’s is less than a twentieth of Japan’s.

7The unpd currently suggests that China’s total fertility rate was about 1.92 in 1990–95, a level roughly 16 percent below that required for long-term population replacement, and that China’s tfr has subsequently declined to a bit over 1.7 today. The U.S. Census Bureau’s reading is quite similar: For 2005, it projects a tfr of 1.72 for China. These estimates, of course, are prepared under uncertainties, the most important of these being the lack of complete annual vital registration data for China, and the unknown degree of under-reporting of infants and children in the country’s censuses and demographic surveys. For background, see Daniel Goodkind, “China’s missing children: The 2000 census underreporting surprise,” Population Studies 58:3 (2004), 281–295, and Guangyu Zhang and Zhonwei Zhao, “China’s Fertility Puzzle: Data Collection and Data Use in the Last Two Decades,” paper presented at Population Association of America 2005 Annual Meeting, Philadelphia, Pennsylvania (April 1, 2005).

8For background, see World Bank, Old Age Security: Pension Reform in China (World Bank, 1997); Song Xiaowu, ed., Perfect the Pension System (Enterprise Management Publishing House, 2001) [in Chinese]; Jinxing Huang, “Economic Restructuring, Social Safety-Net and Old-Age Pension Reform in China,” American Asian Review 21:2 (2003), and Xin Wang, “China’s Pension Reform and Capital Market Development,” China & World Economy 12:3 (2004).

9Professor Mark W. Frazier of Lawrence University may have pinpointed a critical factor in this continuing delay: “Until the fundamental question of what the state owes its citizens in terms of guaranteed basic pension benefits is resolved, the debate over the necessity of pension legislation to supplant the current patchwork of national and local regulatory controls is largely academic.” “What’s in a Law?: China’s Pension Reform and its Discontents,” in Neil J. Diamant, Stanley J. Lubman, and Kevin J. O’Brien, Engaging the Law in China: State, Society and Possibilities for Justice (Stanford University Press, 2005), 108–130, cite at 125.

10Data drawn from China Ministry of Labor and Social Security, China Labour Statistical Yearbook 2003, Tables 1–43, 1–51, and China National Bureau of Statistics, Tabulation on the 2000 Population Census of the People’s Republic of China, Volume 2, Tables 4–4, 4–4c.

11Yasuhiko Saito, Xiaochun Qiao and Sutthichai Jitapunkul, “Health Expectancy in Asian Countries,” in J.M. Robine et al., Determining Health Expectancies (Chichester, England: John Wiley & Sons, 2003), 289–317.

12How are we to explain modern Russia’s awful health tragedy? The fact is, demographers and public health specialists do not fully understand the reasons for these gruesome results. Diet, smoking, sedentary lifestyles, and health care (or the lack of it) all play their part. Russia’s romance with the vodka bottle is also deeply implicated here. Part of the mystery of the ongoing Russian health disaster, however, is that the problem looks to be worse than the sum of its parts: that is to say, death rates are significantly higher than one would predict on the basis of observed risk factors alone.

13Available at http://www.mortality.org.

14Those Census Bureau projections, furthermore, do not formally take into account the possibility that additional and perhaps severe new health setbacks may lie in store for the Russian Federation. Yet precisely such problems are, quite plainly, on the horizon today: Think of the country’s still-gathering hiv/aids and drug-resistant tuberculosis epidemics.

15A single example: In 1999 — by no means the worst year for health in post-Communist Russia — the same death rates experienced by 40-year-old Russians were matched in Italy by women at age 55 and by men at age 60, respectively. (Estimates from the Human Mortality Database, www.mortality.org.)

16Given Russia’s grim health problems, it may seem paradoxical that the population should be aging so rapidly. Remember, however, our earlier discussion: Population aging is driven much more by fertility patterns than mortality — and Russia’s fertility levels today are far below replacement.

17For a penetrating overview and analysis of the Russian pension situation, see Leon Aron, “Privatizing Pensions,” aei Russian Outlook (Summer 2004).

18See, for example, Christophe Z. Guilmoto and S. Irudaya Rajan, “District Level Estimates of Fertility from India’s 2001 Census,” Economic and Political Weekly (February 16, 2002), data from Table a–1. For a more comprehensive treatment, see Christophe Z. Guilmoto and S. Irudaya Rajan, eds., Fertility Transition in South India (Sage Publications, 2005).

19P.N. Mari Bhat, “Demographic Scenario, 2025,” Study #S-15, Research Projects on India — 2025 conducted by Centre for Policy Research (New Delhi, July, 2003).

20Tim Dyson, “India’s Population — The Future,” in Tim Dyson, Robert Cassen and Leela Visaria, eds., Twenty-First Century India: Population, Economy, Human Development, and the Environment (Oxford University Press, 2004), 74–107. Professor Dyson also generously shared with me some of the additional unpublished details from this same series of projections.

21Derived from World Bank, World Development Indicators 2004 cd-rom. Estimates for Western Europe are for the 12 current countries of the European Monetary Union for 1960–65; estimates for Japan are for 1960.

22For background, see First Report of Project oasis (Old Age Social and Income Security), India Ministry of Social Justice and Empowerment (February 1, 1999); World Bank, India: “The Challenge of Old Age Insurance Security,” Report 22034-in (April 5, 2001); B.C. Purohil, “Policymaking for diversity among the aged in India,” Journal of Aging and Social Policy 15:4 (2003); and Robert Palacios, “The Challenge of Pension Reform in India,” in Edgaro M. Favaro and Ashok K. Lahiri, Fiscal Policies and Sustainable Growth In India (Oxford University Press, 2004), 282–300.

23Anne Goujon and Kirsty McNay, “Projecting the educational composition of the population of India: selected state-level perspectives,” Applied Population and Policy 1:1 (2003). The discussion below benefits from additional unpublished projections from that effort, kindly transmitted to me by Dr. Goujon. The discussion refers to the Goujon-McNay “Scenario One” projections.

24Robert J. Barro and Jong-Wha Lee, International Data on Educational Attainment: Updates and Implications, Harvard University Center for International Development Working Papers 42 (April 2000).

25For example: Ansley J. Coale and E.M. Hoover, Population Growth and Economic Development in Low-Income Countries (Princeton University Press, 1958); Stephen Enke, “The economic aspects of slowing population growth,” Economic Journal 76 (1966); David Bloom and Jeffrey Williamson, “Demographic Transitions and Economic Miracles in Emerging Asia,” World Bank Economic Review 12 (1998), 419–456; David Bloom, David Canning, and Jose Sevilla, The Demographic Dividend: A New Perspective on the Economic Consequences of Population Change (Rand, 2003).

26Comparisons derived from World Bank, World Development Indicators (2004) cd-rom.

27Mari Bhat, “Demographic Scenario, 2025,” 9.