The new demography of death

Click to find out more about the Oxford Institute
Click to find out more about the Oxford Institute of Population Ageing

Today’s post is by Dr. George W. Leeson, Co-Director and Senior Research Fellow in Demography, at the Oxford Institute of Population Ageing (OIPA), as well as co-editor of the Journal of Population Ageing and chair of the editorial board of Ageing Horizons.

In Europe, population ageing continues and brings with it increasing numbers of centenarians and supercentenarians as well as a new demography of death. In the mid-19th century, European populations were young and short-lived with high levels of infant mortality. More than half of the almost 370,000 deaths in England and Wales in 1850 would have occurred among people aged less than 60 years. By the early 21st century, these same populations have become old and long-lived, and almost 90 per cent of the deaths in England and Wales now occur among people aged over 60 years.

Of course, the absolute number of deaths as a demographic measure is not a helpful measure, but in this context we are considering mortality in simple absolute terms to reveal the changing composition of the demography of death behind the story of improving survivorship which is a part of population ageing. In the middle of the 19th century, the absolute number of deaths per year in England & Wales was increasing from around 350,000 to around 600,000 by the end of the 19th century. The number peaked in 1918 at just over 610,000, after which deaths in absolute terms declined to around 440,000.

An element of the dramatic changes in the structural development of death and the demography of death was the decline in infant mortality in England and Wales throughout the 20th century. At the turn of the 20th century, infant mortality had been as high as 154 deaths under 1year per 1000 live births. This rate has been halving every 25 years roughly except for the period from1975 to 1990, where the halving time decreased to around 15 years, returning thereafter to be on track for 25 years halving again, ending at just 4.4 deaths under 1 year per 1000 live births in 2011.

Declines in mortality among the extreme aged have been striking. One hundred and seventy years have seen late-life life expectancy increase by just over 7 years for males and almost 9 years for females—something of an achievement given the previous conviction that mortality at older ages was intractable.

Population ageing is often equated with an increasing number and proportion of frail, dependent older people who become an increasing burden on society and family. The numbers themselves are a warning that societies need to change with the changing demography. One consequence of the numbers change is the new demography of death. Mortality at advanced ages is being delayed and although the future remains difficult to predict, there does seem to be an increasing body of evidence that around the world lives will continue to be extended for some time to come. By the turn of the next century, life expectancies at birth are predicted to be 93 years for males and 95.6 years for females in England and Wales, while at age 65 years, life expectancies are expected to be 29.9 years for males and 31.1 years for females.

The number of people aged 100 years and over in England and Wales increased from less than 200 in 1922 to 570 in 1961. By 1981, this number had climbed to 2,418 and to 12,318 in 2012 and by the middle of the century the number is expected to be close to 300,000 and more than 1 million by 2100.

So more people are living longer and the longest lived are living longer too.

What then of the future and the new demography of death?

The development of this demography of death over the 200 year period from the mid-19th to the mid-21st century is striking. The total number of deaths in England and Wales increases from 342,760 in 1838, when 50 percent of a cohort was dead by age 45 years, to almost twice that number, 666,253 in 2050,when 50 percent of a cohort will be surviving to age 90 years.

It is the structure of this new demography of death that is interesting.

Since 1959, death has been dominated by deaths of people aged 60 years and over and this domination has increased and will continue to increase at least until the middle of this century. In 1959, 78 percent of deaths were people aged 60 years and over. This had increased to 88 percent by 2009 and is predicted to reach 94 percent by 2050. And in line with the ageing of the population of England and Wales, the proportion of the 60-plus deaths aged 80 years and over has also increased and continues to increase—from 34 percent in 1959 to 60 percent in 2009 and 78 percent in 2050.

While this is in all respects a natural consequence of the ongoing demographic development in England and Wales and similar developed economies, there remains the question: are we prepared for this new demography death, its scale and structure, as individuals, families, communities, and societies?

The ageing of European populations in the latter part of the 20th century was a demographic surprise brought about by a combination of demographic resistance to dismissing the idea of a limit to human longevity and the creeping decline in mid- and late-life mortality as the prevention and treatment of, for example, heart disease improved. Experience proved we had pushed old age into our 80s. The future could be an equal demographic surprise if we ignore the evidence of the new demography of death, which also would suggest that the lives of more and more people will continue to be extended and centenarians and supercentenarians would comprise an increasing number and proportion of our populations.

The new demography of death is also a 21st century challenge for the emerging economies of the world, where life expectancies continue to increase. However, these economies are challenged additionally by the speed of their fertility transitions, which in many instances are occurring in one or at most two generations.

How could/should we begin to prepare ourselves for this new demography of death?

It is clearly a challenge to longstanding concepts of old age and retirement – indeed one could ask whether retirement even at age 75 years is sustainable. Family dynamics will be challenged by the survival of extreme aged generations delaying intergenerational succession and depending on smaller families for support in frail and dependent old age.

The additional and confounding prospect of declining population size raises different issues as the workforce contracts. This would lead to policy discussions about controlled labour immigration, perhaps, to compensate for the declining local workforce.

Buried in this demography of death is, however, a success story of survival. Let us not be dismayed by that but let us begin to discuss what it means.

Useful links

Dr. Leeson is a leading member of The Complex Environmental Population Interactions Project which unites key demographers, economists, anthropologists, philosophers and environmentalists.

The video below, introducing population growth and global ageing, is part of the e-learning programme GOTO – Global Opportunities and Threats – by Said Business School. It includes contributions by George Leeson as well as Sarah Harper and Kenneth Howse, also from OIPA.

Solving the super-ageing challenge by Katsutoshi Saito, Chairman, Dai-ichi Life Insurance Company, Ltd at OECD Forum 2014

Looking for ghosts in China

A friend in Kunming laughed when we told her our travel plans: “Chenggong? But there’s nothing there!” More accurate, perhaps, if she’d said there’s no one there.


Under construction since 2003, Chenggong is a satellite city of Kunming, capital of mountainous Yunnan province in southwest China. Such projects are not rare in China, and they tend to follow a familiar pattern: Universities and government offices are relocated, students and officials move in and, eventually, so do other people.

Chenggong is different: Yes it has students and, as far as we know, bureaucrats but – so far at least – pretty much no one else. Its wide-open highways and empty estates have won it plenty of attention in the blogosphere and earned it a reputation as one of Asia’s biggest “ghost cities”.

In truth, the place is not as quiet as all that. Around the spacious university campuses there’s a noticeable buzz. A few small streets are filled with shops selling student necessities – cheap grub and trendy clothes. On one campus, an English-language student, Tina, showed us around and enthused about the quality of the facilities. When we asked her if she liked Chenggong, she nodded enthusiastically: “Oh yes, the air is so clean.” That, too, is often a novelty in China.


But, elsewhere, it’s hard to escape the feeling that Chenggong lacks people. Cars bowl along six-lane highways at speeds unthinkable in most of China’s congested cities, while passengers seem sparse on the light railway that will eventually link the city to the provincial capital. Driving back to Kunming in the evening, we passed housing estates where not a single light seemed to be shining. According to the World Bank’s Holly Krambeck, Chenggong has over 100,000 empty apartments.

Regardless of how many people Chenggong eventually manages to attract, the city is part of one of the most remarkable human transformations in history. In just a little over three decades, China has gone from being a country that was overwhelmingly rural to one where just over half the population now lives in urban areas.

That might sound high but, by international standards, and for a country at its level of development, China still has a relatively small urban population. That wasn’t always the case. As a fascinating recent OECD paper by Vincent Koen, Richard Herd, Xiao Wang and Thomas Chalaux notes, China was once one of the most urbanised places in the world, admittedly in an era when globally around 90% of people lived in the countryside. By around 700 C.E., the city of Chang’an – now known as Xi’an, home of the terracotta warriors – is believed to have had around a million inhabitants, levels that London and Paris would only reach in the 1800s.

By the dawn of the 20th century, however, the proportion of Chinese living in cities hadn’t changed much in 400 years and, throughout that troubled century, it grew only very slowly. By 1949, urban dwellers accounted for just 12% of the population, around a third of levels then typical in the world.

Today, China is playing catch-up. In the last decade, the urban population rose by about 20 million a year; by 2020 the government expects around three out of five people to be living in urban areas. And by 2040, it’s estimated that around one billion Chinese will be living in cities.

China’s government believes cities are essential to building a modern economy. For one thing, cities make it easier for companies to do business, making them hubs for growth. For another, city dwellers tend to spend more than their country cousins, which should help China meet its goal of relying less on exports and more on domestic demand. So, cities will be key to China’s economic and social future. But planning them will require careful thinking about their “hardware” and “software”, to borrow a metaphor.

For an example of the hardware, take transport. By international standards, Chinese cities have relatively low levels of public transportation. That’s evident even in a new town like Chenggong, where, as urban planner Luis Balula notes, “the wide streets and superblocks […] continue conveying the image of a car-oriented urban environment waiting to be populated by cars.”


According to the OECD paper, China would need to invest the equivalent of around 11% of its GDP just to bring transport provision in its ten biggest cities up to international standards.

As for the software, think of that as the people who will live in China’s cities. Many will be rural migrants and, so far at least, the legal situation in China’s cities hasn’t worked in their favour. And that’s a topic we’ll return to soon.

Useful links

Policies for Inclusive Urbanisation in China (2013, OECD Economics Dept.)

OECD work on China

网站 (中文) (The OECD’s Chinese-language site)

Of kissing, the less the better

1 in 7 billion. Now click to see how population growth affects you

Today’s post to mark the birth of the 7 billionth baby was written in collaboration with Lynn Robertson of the OECD’s  Directorate for Financial and Enterprise Affairs, and follows on discussions that led to this post we did in September on children’s happiness

The world’s 7 billionth baby was born today, or a couple of years ago, or maybe will be born in a couple of years from now. Demographers can’t say exactly when we reach the magic number, but Halloween 2011 is as good a guess as any. Many experts agree though that there’s never been a worse time to be a child. Extensive research in a number of bars has proved that when we were kids, life was much better, and so were children. And this result is robust over time: whatever period you look at, it turns out that you missed the golden age by a generation.

One change over the millennia is that now we talk about children being more selfish, lazy, etc rather than just people. That in itself is an improvement the golden agers probably haven’t thought about. Until the late 19th century, children enjoyed no special status or protection. They worked long hours often in dangerous conditions, could suffer barbaric punishments, and be abused with impunity by practically any adult. In fact, we’re only now learning of many cases of abuse that took place when today’s adults were children.

When things did start to change, progress was slow, with great faith in harsh treatment. L. Emmett Holt’s 1894 bestseller  Care and Feeding of Children explained that: “Babies under six months old should never be played with, and of kissing the less the better”.  In 1928, the most influential child psychologist in America, John B. Watson, echoed this advice in Psychological Care of Infant and Child when he warned against the dangers of “too much mother love”, telling women “Never hug and kiss them, never let them sit on your lap.” Otherwise, the result would be spoilt, self-centred, unproductive brats.

This drivel is sometimes called a “no-nonsense” approach to child rearing, and obviously influences people who want to return to a time when they had no responsibility and someone else always told them what to do. It’s impossible to argue against nostalgia, fond memories, and the feeling that things were better before despite the world wars, genocides, diseases and squalor of times past, but the fact that this view can be retraced back over the centuries suggests that it’s nothing more than a myth. The objective facts paint a different picture.

Take health for example. A child born fifty years ago when the OECD was created could expect to live to over 73 on average in only two countries of the world compared with 87 countries today. In OECD countries, average life expectancy reached 79.1 in 2007, more than ten years better than in 1960, and exceeded 80 years in almost half the OECD countries.

Elsewhere, improvement has been dramatic too even over short time spans.  According to UN figures on the Millennium Development Goals, the global mortality rate for children under five has declined by a third, from 89 deaths per 1000 live births in 1990 to 60 in 2009. Despite population growth, the number of deaths in children under five worldwide declined from 12.4 million in 1990 to 8.1 million in 2009, or nearly 12,000 fewer children dying each day. Access to education is also improving, and there is progress on the other MDG targets too.

Much of the evidence on more subjective aspects comes from small-scale studies with only a few participants, but last year Brent Donnellan of Michigan State University and Kali Trzesniewski of the University of Western Ontario published an analysis of nearly half a million high-school seniors spread over three decades. They argue that “more often than not, kids these days are about the same as they were back in the mid-1970s” regarding a whole range of topics, including individualism, happiness and antisocial behaviour. Their data also suggest that compared with previous generations, today’s youth are more cynical and less trusting of institutions;  less fearful of social problems such as race relations, hunger, poverty and energy shortages; and have higher educational expectations.

Still, the basic problem remains: they’re not us. The good news is that in a decade or two they will be, and they’ll be boring their own kids with stories about how they were happy with a simple iPad for their birthday. So, whoever you are little 7 billion, good luck. With parents like us, you’ll need it.

Useful links

The OECD is partnering 7 Billion Actions, established by the UN Population Fund to “inspire change that will make a difference by highlighting positive action by individuals and organisations around the world”. You can find out more about the OECD Better Life Initiative and make and share your own Better Life Index here

Doing Better for Children provides data, analysis and policy advice on child well-being in OECD countries

UNICEF IRC, the OECD, the Learning for Well-being Consortium and the European Commission are co-organising an expert consultation on cross-national research on child indicators and well-being on 2-3 November 2011 at the OECD. You can download the papers here.

Wikichild is part of the OECD’s Wikiprogress platform for sharing information on global societal progress

Chinese demography: One child, many consequences

Photo courtesy of Jacques Bron

Coinciding with the China Development Forum in Beijing, the Insights blog is focusing on China this week

How many wedding dresses does it take to change the world economy? That’s right, lots. And how? High saving rates in certain countries, notably China, contributed to housing price bubbles and the global financial crisis. Shang-Jin Wei of Columbia Business School and Xiaobo Zhang of IFPRI have an intriguing explanation as to why China’s saving rate is so high (around 50% of GDP in 2007 just before the crisis started): blame it on the brides.

Or rather, blame it on “competitive saving”, one of the unintended consequences of the one-child policy, introduced in 1978 to slow population growth. The government claims that thanks to the policy, the population is three to four hundred million fewer than it would have been, but since families prefer boys, many of the missing millions are girls, including victims of female infanticide and sex-selective abortions.

The normal ratio of boys to girls is around 106:100, an evolutionary corrective for the fact that male babies are more likely to die. In China, however, the ratio is much higher. The exact figure is not known, but estimates range from 119:100 to over 130.

What is known though, is that there are now tens of millions more young men of marriageable age than women. So to improve their son’s chances, a family saves to be able to buy a nicer, better furnished house than rivals and give the young newly-weds a good start in life. According to Wei and Zhang, even families who don’t have a son to marry off have save as much, since prices are driven up. Families with sons save more than those with daughters, and savings rates are higher in regions with higher gender imbalances.

Another unintended consequence is the rapid ageing of the population. In 1975, just before the one-child policy started, there were six children for every person aged over 60. By 2035, there will be two over-60s for every child. Population ageing is happening all over the world, but it is happening much earlier in China’s economic development than in OECD countries.

As Richard Jackson and his colleagues at the Center for Strategic and International Studies point out, when the elderly share of the population was the same in the US as it is China today, per capita income was four times what China’s is at present. Some Chinese express the fear that the country will grow old before it grows rich.

China’s growth has depended on what some media like to call limitless supplies of cheap labour. But the working-age population will peak in 2015, and will shrink by almost a quarter by 2050, with an even sharper decline for people in their 20s and 30s. Total population will still grow, because people will live longer, and China will have an older population than the US in 2050.

Given that this elderly population will not be able to count on large numbers of children to support them, how will they live? The government’s goal is universal coverage for the basic pension system by 2020 and it has also taken a number of steps to encourage schemes to complement this.

Yet public pension coverage remains far from universal, and has large unfunded liabilities for early retirees from the state-owned sector. Moreover, benefits are not fully portable, so workers often have to choose between job mobility and retirement security, and the rate of return on personal pension plans is too low to replace a salary.

At the same time, the decline in the working-age population may allow employees to negotiate better pensions as part of their pay packages. Labour shortages have already been reported in some regions, partly because rural migrants who went home during the worst of the recession don’t want to come back. Last week, the Guangzhou Daily reported that the local authorities had raised the minimum wage from 860 yuan to 1030, a higher rate than Beijing even.

All this is leading to calls to abandon the one-child policy. Especially since China is now suffering from a phenomenon that has cursed every country that ever existed at every epoch in world history – kids today are not as nice as us. Or, as the People’s Daily complained last month: the one-child policy is breeding brats.

Useful links

China in the 2010s: Rebalancing growth and strengtheneing social safety nets

2010 年代的中国:经济增长再平衡和强化社会安全网 (Chinese version)

OECD work on China

The OECD’s Chinese-language site – 网站(中文)

China at the OECD Development Centre

Photos by Jacques Bron