Pro-natalism (the idea that people, or rather, women, should have more babies than they choose to do at present) has become an established orthodoxy,[1]. The central claim is that, unless something changes soon, human populations both global and national, are going to decline rapidly, with a lot of negative consequences. This is simply not true, on any plausible assumptions about fertility[2]
There’s no need for me to do any calculations here. For many decades he Population Division of the UN Department of Economic and Social Affairs has been producing population projections for the world, and individual countries, under a variety of scenarios. One finding is unambiguous. Short of a drastic decline in fertility, far beyond what we are now seeing there will be more people on Earth at the end of this century than there were at the beginning
The range of projections considered plausible is the shaded area. All the projections in that range show population increasing for several decades to come, and remaining higher than at present at the end of the century. The reason is simple. Global fertility is close to the replacement level (one surviving daughter per woman) at present, but past growth means that a large proportion of the population is in, or approaching, child-bearing years. It’s only when this group ages out that the effects of declining fertility, assumed in the lower projections will start to dominate
What about the blue dotted lines? These assume drastic reductions in fertility. On the low side, that involves the entire world becoming like South Korea, where the combination of high employment rates for women and pre-modern male attitudes on gender role has produced reproduction rates below 0.5.
But even in this extreme case, world population in 2100 only falls to 6 billion, the same as in 2000. I was around at the time, and did not feel as if there were too few people about.
One reason these predictions have only a limited range of variation is that most of the growth in population is already baked in. There are 2 billion or so children under 14 at present in the world, and most of them will be around in 2100 as will their soon-to-be-born siblings.
What about the need for workers? One unsatisfactory feature of long-running projections like this is the use of outdated statistical concepts such as the “dependency ratio”, that is, the ratio of people aged 15-64 to everyone else. That made sense 50 years ago, when this range represented the period between leaving school and retiring in most industrial societies. But these days (and it will be even more so in 2100) education continues well past 20 and retirement is often deferred to 70 or more. A look at the age group 25-69 shows that it is going to remain more or less stable in absolute numbers declinging only marginally relative to the growing population
Population projections for individual countries depend largely on what happens to migration. In the absence of stringent restrictions, the flow of migrants from poorer to richer countries will largely offset differences in fertility, meaning that the trajectory for individual countries will look similar to that for the world.
Of course, if you combine low fertility and an already-old population with hostility to immigrants, and you can’t stop your own young people from seeking a better life abroad, you end up with a sharply declining population, as in South Korea and Hungary. But it’s much easier to let more migrants in (there are plenty of young adults, many of them well-educated, knocking at the door) than to persuade people to have more babies.
There is no difficulty in gaining access to these projections, and anyone with a spreadsheet and a bit of time can reproduce them. Yet I’ve read dozens of pro-natalist articles in both traditional and new media and the evidence is never mentioned. Maybe I’m living on the wrong planet.
fn1. Some this is driven by racists worrying specifically about the lack of white babies. But the belief that declining fertility is a crisis is also dominant among centrists, like those pushing the “abundance agenda”, who also support high levels of immigration. Archetypal example is Matt Yglesias who advocates One billion Americans
fn2 There are plenty of ways in which we are risking massively increased mortality (nuclear war, climate catastrophe, pandemics, AI apocalypse etc), but having babies won’t help in those cases.
John Quiggin 
Test.
The global average fertility rate is set to dip below replacement level (2.1 births per woman) by the mid-2040s. As the OP explains, due to demographic momentum the global population will continue to increase until, according to the most likely scenario shown in the graph in the OP, it reaches a peak at 10.3 billion in 2084 before declining to 10.2 billion by 2100.
It is the nearest thing to an Iron Law of Demography that a country’s fertility rate will fall below replacement level once it reaches a high level of human development as measured by the UN Human Development Index. Given a choice, we would prefer to live in a country with below replacement fertility than one with high replacement fertility.
The fertility rate is not like the rate of GST (which can be dialed up or down by Parliament amending the Taxation Act) or the base interest rate (which can be dialed up or down by a meeting of the Reserve Bank board). It is the expression in aggregate of the personal decisions of millions of individuals, and those personal decisions are in turn path-dependent on multiple previous decisions by those millions of people (cf. Julia Gillard’s “decision-creep).
The failure of attempts by governments to engineer reversals in fertility decline, and the range of unintended consequences of such attempts (which are often undesirable, not least from the perspective of “family values” conservatives) shows the radical incompleteness of governments’ understandings of the relationships between their policies and the life choices of those they govern (particularly younger women). The most promising and constructive application of human ingenuity in coming decades in relation to this issue will be devising economic models that can work with stable or slowly declining populations.
The global average fertility rate is set to dip below replacement level (2.1 births per woman) by the mid-2040s. As the OP explains, due to demographic momentum the global population will continue to increase after that until it reaches a peak that, according to the median scenario shown in the graph in the OP, will be 10.3 billion in 2084, followed by a fall to 10.2 billion by 2100.
It is the nearest thing to an Iron Law of Demography that a country’s fertility rate will fall below replacement level once it reaches a high level of human development as measured by the UN Human Development Index. Given a choice, we would prefer to live in a country with below replacement fertility than one with high replacement fertility.
The fertility rate is not like the rate of GST (which can be dialed up or down by Parliament amending the Taxation Act) or the base interest rate (which can be dialed up or down by a meeting of the Reserve Bank board). It is the expression in aggregate of the personal decisions of millions of individuals, and those personal decisions are in turn path-dependent on multiple previous decisions by those millions of people (cf. Julia Gillard’s “decision-creep).
The failure of attempts by governments to engineer reversals in fertility decline, and the range of unintended consequences of such attempts (which are often undesirable, not least from the perspective of “family values” conservatives) show the radical incompleteness of governments’ understandings of the relationships between their policies and the life choices of those they govern (particularly younger women). The most promising and constructive application of human ingenuity in coming decades in relation to this issue will be devising economic models that can work with stable or slowly declining populations.
JQ wrote:
“fn2 There are plenty of ways in which we are risking massively increased mortality (nuclear war, climate catastrophe, pandemics, AI apocalypse etc), but having babies won’t help in those cases.”
This is a case of burying the lead in the footnote. The dangers J.Q. mentions here vary in probability from non-trivial lows to near certainties. The near certainties at this stage are climate catastrophe and pandemics. Why avoid grappling with these issues?
I have my own theories on why we avoid grappling with these issues but I am interested in the theories of others too. Why are we not grappling with the big existential issues of imminent climate catastrophe and sweeping pandemics?
Geoff Miele, recently on this blog, has alluded to the clear evidence that on current trends we are highly likely to breach civilization-destroying temperature limits of +3.0 C by about 2054 to 2065.
Eric Feigl-Ding has posted recently about the imminent full destruction of the USA’s pandemic fighting capability.
https://threadreaderapp.com/thread/1934700967892525061.html
I write “full destruction” because much of the USA’s ability to fight pandemics with public health measures and non-pharmaceutical interventions has already been largely destroyed by the victory of GBD (Great Barrington Declaration) ideology.
JQ – “Population projections for individual countries depend largely on what happens to migration. In the absence of stringent restrictions, the flow of migrants from poorer to richer countries will largely offset differences in fertility, meaning that the trajectory for individual countries will look similar to that for the world.”
Trying times are ahead for the 99% in that scenario. Without restrictions on migration socioeconomic and governance standards in third world countries won’t rise to a first world level; just as first world countries standards will fall into a second world trap. In such a future most countries shall certainly be similar in not being of the first world.
Adding to those global difficulties is perhaps a newly discovered hidden third world rural population that may prove to be a far larger problem for the global 99% than has been reckoned.
As they have the most to gain have the manipulating 1% known about the possible miscount all along? There’s easy money to be extracted from bigger populations and bigger migrations.
https://www.newscientist.com/article/2472604-have-we-vastly-underestimated-the-total-number-of-people-on-earth/
New Scientist*
Humans
Have we vastly underestimated the total number of people on Earth?
A new way of estimating rural populations has found that we may be undercounting people who live in these areas, potentially inflating the global population beyond the official count of 8.2 billion – but not everyone agrees
By Chris Stokel-Walker 18 March 2025
Japan, in spite of the population stratagems of their governing 1% has it right. Few migrants, fewer babies, and good times ahead:
https://thewest.com.au/opinion/james-pethokoukis-japan-baby-slump-might-not-be-an-economic-disaster-c-19160516
The West Australian*
Japan baby slump might not be an economic disaster
James Pethokoukis Fri, 27 June 2025
“A Goldman Sachs analysis this year found that wage growth has risen from 0.3 per cent in the 2010s to 1.2 per cent in the 2020s, while core inflation has climbed from 0.5 per cent to a healthier 1.5 per cent, an encouraging development in a country that was long stuck in a deflationary trap…
As Goldman sees things, the demographic decline that once drained vitality is now creating a “virtuous cycle” of tightening labor markets, increased worker bargaining power and more investment in productivity-enhancing tech. These trends are helping prop up the economy even as it weathers a shock from the US-led trade war…
But rather than just grumbling about the lack of workers, businesses are finding ways to use fewer of them. From software to machines — here’s where AI and robotics can really lend a hand — productivity has gone up in Japan’s most labour-starved sectors, with corporate profits hitting record highs in fiscal 2024…”
*The links above are from the SPA newsletter this week.