Projections in world population statistics have revealed a forthcoming drop in global numbers. Dr Abul Rizvi examines the reasons why.
LATER THIS YEAR, in May or June, the U.N. will update its 2019 population projections.
Projections of the world’s population are produced by three major organisations. The United Nations Population Division; the Wittgenstein Centre in Austria; and the Institute for Health Metrics and Evaluation at Washington University.
These organisations use different approaches to forecasting future rates of fertility, life expectancy and migration — the underlying basis of all population projections.
From a global perspective, the key is the assumed fertility rate. For this, the U.N. uses statistical forecasting methods based on historical trends.
The U.N. divides the evolution of fertility into three phases — high fertility; transition to low fertility; and low fertility post-transition.
Crucially, based on past experience of countries in Europe and East Asia, the U.N. assumes the fertility rate in most low fertility countries will very gradually rebound to 1.75 births per woman by 2100 while the decline in the fertility rate in high fertility countries over the past 30 years is projected to gradually slow.
The Wittgenstein Centre uses a blend of expert opinion and statistical modelling. For low fertility countries, it also assumes, based on expert opinion, that the fertility rate in low fertility countries will converge to 1.75 births per woman, but that this will not be until 2200. This means a slower rate of recovery in the fertility rate of developed nations with low fertility rates than assumed by the U.N. For example, the fertility rate of Taiwan is assumed to gradually rise from 1.04 births per woman in 2017 to 1.75 births per woman by 2200 rather than by 2100 as assumed by the U.N.
In addition, the Wittgenstein Centre assumes fertility will continue to decline at a similar rate to recent years in sub-Saharan Africa while the U.N. has assumed this will decline at a slower rate. The combination of these two factors led the Wittgenstein Centre to conclude the global population will peak much earlier than the U.N.
The Washington University researchers use a more sophisticated causality model to forecast fertility and life expectancy. In terms of fertility, the researchers use a completed cohort approach to measuring fertility rather than the traditional total fertility rate measure. The latter is more volatile as it initially falls if women delay childbirth due to more time spent in education and career development and then recovers as they have babies later in life. The completed cohort approach is a much more stable basis for forecasting fertility.
The Washington University team model completed cohort fertility as a function of rising education levels (particularly of girls) and increasing access to contraception. These two factors drive over 80% of the variance in fertility across the world. They then examine the impact of faster and slower rates of change in education levels and access to contraception.
All three organisations assume life expectancy will grow, but very slowly — note that life expectancy in the U.S. and UK had in recent years actually fallen pre-COVID-19. They make net migration assumptions based on past trends noting these can change significantly subject to the relative economic performance of individual countries and are thus highly uncertain.
All three organisations project the world’s population growth rate to slow and age considerably over the next 80 years. Compared to the U.N. Population Division and the Wittgenstein Centre, the Washington University researchers conclude the world’s population will age faster, peak sooner and then decline faster.
Based on quite plausible assumptions, researchers at Washington University have projected the world’s fertility rate (births per woman over her lifetime) will be below the replacement level of 2.1 from 2035; its population will peak by the mid-2060s and then go into increasingly more rapid decline.
Well before that, the world’s population must get a lot older, accelerated by more baby boomers heading into retirement over the next 20 years.
Over the past 29 years, the global fertility rate has fallen from 3.2 births per woman in 1990 to less than 2.5 births per woman in 2019. If fertility continued to decline at that rate for the next 30 years, the decline in the global population growth rate would be much more rapid than currently projected by the U.N.
It is likely the U.N. will again revise down its projection of the world’s population in its next revision due in 2021. In its 2019 revision, the U.N. reduced its projected world population peak by over 300 million (a bit less than the current population of the USA).
In their reference or base case scenario, the Washington University researchers forecast the global fertility rate may fall to replacement level – that is 2.1 births per woman – by 2034 and then continue to decline to 1.66 births per woman by 2100. On that basis, the world population would peak much earlier at 9.733 billion by 2064 and then decline to 8.758 billion by 2100 (see Table 1).
Table 1: World Population Projections — Washington University
The Washington University researchers point to a major shift in world population away from China, whose population is projected to almost halve by 2100 and, to a lesser degree, away from South Asia, North America, Europe, Central Asia, East Asia and South America. This is offset by major population growth in sub-Saharan Africa, North Africa and the Middle East.
But population growth, even in these latter regions, is projected to peak by 2100 or soon afterwards.
The population of 23 mainly developed nations is projected to fall by 50% or more by 2100 including Japan, South Korea, Italy, Spain and Greece. Many of these nations are in the so-called “low-fertility” trap.
This has three dimensions, all of which drive down fertility and keep it low:
- population ageing associated with low fertility itself reduces the number of potential mothers and keeps fertility low;
- the ideal family size is linked to the smaller family size of previous cohorts — the children of the baby boomers have increasingly adopted smaller families and are likely to continue to do so; and
- fertility is the result of aspirations and expected income — as aspirations rise while the expected income of younger cohorts falls, low fertility is further reinforced. Inadequate government policies supporting families in which both partners work puts further downward pressure on fertility.
As a result of these changes, the population of China, the world’s most populous country, is projected to halve by the end of the century. In perhaps the biggest backflip in global social policy history, the Chinese Government is now encouraging couples to have more children.
It is also trying to attract back the young and highly qualified amongst its diaspora in an attempt to convert China from a nation of net emigration to one of net immigration.
Due to their higher rates of net migration, the populations of “Anglo-Celtic” countries such as the USA, Canada, the UK and Australia are comparatively younger and are ageing more slowly. Unlike most other developed nations, their populations are not projected to peak until well into the second half of the 21st Century — once again, primarily due to immigration, including its effect on the number of births.
While populations of less-developed nations will also continue to grow, these, too, are seeing rapidly declining fertility rates, particularly amongst teenage girls who are increasingly more educated and demand equal career opportunities and now have significantly easier access to affordable contraception.
Indeed, there are now less-developed nations that are also ageing rapidly and facing major population decline. These include Thailand as well as most countries in Eastern Europe and Central Asia — low fertility is not just a feature of developed nations.
Before any nation’s population goes into long-term decline, it must first age.
The median age of developed nations as a whole has increased from 28.5 in 1950 to around 42 today — an increase of 13.5 in 70 years. This is projected by the U.N. to rise to 46 by 2040 — a further increase of four in 20 years. That means by 2040, more than half the women of developed nations, plus China and Russia, will have completed child-bearing.
The median age in developed nations will increase even more quickly if the Washington University researchers are correct since the projected median age of the totality of developed nations is based on the U.N.’s assumption that the decline in fertility around the world will slow markedly and for many developed nations begin to rise.
If fertility rates of developed nations as a whole do not rise as assumed by the U.N., or indeed fall even faster due to the COVID-19 recession, their populations will age more rapidly and their populations will, as a whole, go into decline earlier than currently projected by the U.N. — that is before 2035. The combined populations of Europe (not including the UK), China, Russia, Japan, South Korea and Taiwan are already projected to go into decline well before 2030.
The world’s five major economic/military powers – that is the USA, China, the EU, Russia and Japan – will all age rapidly over the coming decades even though the U.N. has assumed their fertility rates will steadily rise for the rest of the century towards 1.75 births per woman.
But even if their fertility rates increase as assumed, the populations of China, the EU, Russia and Japan will decline sooner and more rapidly than the USA.
Due to its higher rates of net migration, the population of the USA is projected to continue to grow for at least another 50 years. But it, too, has a below-replacement fertility rate. In 2019, the U.S. fertility rate was 1.78 births per woman and its life expectancy was falling. COVID-19 is likely to exacerbate these two trends.
It would be surprising if China, the EU, Russia and Japan just allowed the projected decline in their populations to be realised without response; either to increase fertility rates or increase net migration. But neither of these are simple public policy choices — both require a major change in societal attitudes to achieve even a very minor shift in fertility or net migration.
Australia has slowed its rate of population ageing through a targeted immigration policy, particularly from 2000 onwards. This has been driven by the younger age structure of migrants and temporary entrants at time of arrival.
Skill stream migrants are particularly concentrated in the age range 25 to 44 at time of arrival or permanent resident visa grant while overseas students are concentrated in the age range 20 to 34. By comparison, almost 34% of Australian-born adults were aged 55 or over at the time of the 2016 census. This differential will rise as Australia’s population ages further.
Australia’s achievement is based on 70 years of “managed migration” that few other countries can boast. Nevertheless, there is no realistic level of immigration that can stop Australia’s population ageing further — it can only slow the rate of ageing which can be valuable in assisting the adjustment to an older population.
It is likely that sometime in the second half of this century, Australia’s population will also cease growing and go into decline. The specific timing will depend on the future direction of Australia’s fertility rate, the level of net migration and growth in life expectancy which appears to be slowing across Anglo-Celtic nations.
- Population growth and the economy: Challenging the dogma
- From boom to bust: The Treasurer’s population forecasts
- George Monbiot must bring population back to the table
- Treasury stops pretending Australia’s fertility rate will rise
- Australia’s COVID-19 recovery must question our breakneck population growth
Support independent journalism Subscribe to IA.