Does a shrinking population mean we need immigration?

New figures now anticipate a gradual fall beginning in the 2050s.

Updated population projections from the Office for National Statistics (ONS) released last week anticipate that the UK population could start to shrink from 2055 onwards. This was quite a stark change from the most recent projection before that, which showed the population continuing to grow overall until at least the year 2122.

By the time we get to that distant year nearly 100 years from now, the UK’s population is projected to be 17 million people smaller than it would have been under the previous estimate which was released just two years ago.

The reason for this big revision of what the future may look like is that the ONS has significantly reduced the expected level of future net migration, compared with when the last projections were made just over a year ago. The way net migration is projected is really simple - it has to be, since changes in immigration policy have a big impact and are impossible to predict - with future figures being an average of the ten years before that point.

When projections were last made in 2025, net migration had just been at an all time high from which it has since fallen sharply.

If not for immigration, the population would shrink

Population projections are not firm predictions. No one can say with confidence what’s going to affect the drivers of population change, which are births, deaths and migration, even five years from now let alone 50 or 100 years in the future. That said, they are used in policy planning at a very high level, including in decisions over how much will need to be allocated to welfare spending, education, healthcare and pensions in future years.

Births and deaths are the easiest part of the equation to predict, at least in the short to medium term, and on that score the UK is following virtually all other countries worldwide on a trend of declining fertility.

In the case of the UK, if we don’t expect a radical change to this trend, natural change in the population (births minus deaths) is going to be negative in every foreseeable year from 2026 onwards.

In Scotland and Wales, natural change has already been negative since 2015. Without in-migration from the rest of the UK and abroad, their populations would have shrunk. Their future projections anticipate natural change to keep reducing the population, while migration will balance this out at least in the medium-term.

So do we ‘need’ immigration?

The question really is whether it matters or not if the population is shrinking.

There’s no optimal level for the size of the UK population - it has been a lot smaller, and it could conceivably grow a lot bigger without negative consequences. Fearing population shrinkage in its own right is something readers of Sum of Us might recall from eighteenth century England, when the influential Reverend Richard Price tried to convince others that the nation’s size was on the wane.

Back then, the size of the population in terms of headcount was a lot more important than it is now. If Price had been right - and he wasn’t - an abruptly shrinking population would have caused some real problems. For one thing, national defence relied much more on the number of young men who could be mustered rather than, as it does now, on technology. For another, having a lot of children was the only economic safety net available to the majority of families, providing they were healthy enough to start working from a young age.

It goes without saying that we live in a very different world now. That said, the ONS has also calculated a ‘zero net migration’ scenario for its population estimates, in which it projects that without any addition to the population from migration at all, the population of the UK would halve by 2122. This would take us back to a population of 35,300,000 people - last observed in the UK in 1883.

This is an unlikely scenario but would definitely be one to worry about. It would be a catastrophe worthy of Reverend Price levels of concern as it would be very hard to avoid significant GDP shrinkage from that kind of decline.

Some level of positive net migration would seem to be a good way to avert that. But even so, it’s less about the absolute numbers than who exactly is migrating.

The main issue for us now is that our population is ageing and that, as was not the case in the late eighteenth century, we have committed to providing people with free healthcare and a pension later in life so that they don’t have to work until the day they drop. What is cause for concern is the declining ratio of working age to non-working age people, which looks set to inevitably rise.

Depending on what planning generations now do for future ones, there could come a time when the tax base gets severely stretched in trying to cover social security commitments.

As a side note, the UK’s dependency ratio is high by global standards but not by those of developed and high-income countries. Elsewhere in Europe, including in Poland, Italy and Spain, the ratio of children and pensioners to the working age population is projected to be over 90% by 2100, while in the UK it will ‘only’ be 80%. In parts of Asia, the situation is worse, including in China, South Korea and Japan.

Migrants are much younger on average than the general population and, overall, more likely to be in work. Over the span of a lifetime, an average-earning migrant contributes more in tax than they receive in benefits.

That said, there is great variation as to whether people are net contributors across different migrant groups. Dependants (children and partners) of workers on the Health and Care visa, for example, are expected to be significant net recipients of state spending over their lifetimes.

Whether migration solves the problem of an ageing population depends on its composition. Having lots of high-earning migrants come in and even settle for good in the UK should achieve both a short- and long-term boost for the public finances.

On the other hand, admitting lots of low-paid workers, children, and people who may never work and the gain in population numbers will not address the reason that an otherwise shrinking population is a problem in the first place. It may indeed make it worse.

There’s also evidence that a sudden increase in net migration can disrupt certain sectors and the economy as a whole. There are lots of positive and negative examples of this, including that a sudden increase in low-skilled labour can depress native wages (observed with Cuban refugees in the USA and EU migrants in the UK), and that immigration can stimulate further job creation and boost the economy.

Migration is certainly not the only solution for a rising dependency ratio. Raising the retirement age is a simple way to expand the tax base, although it relies on people being in good enough health to work into more advanced age. There’s also the option of getting people to contribute more during their working years to fund retirement. Automation is also a realistic way to reduce the manpower needed in many sectors.

These are all considerations for policy makers, and ones which require long-term thinking in the order of fifty or one hundred year stretches. That’s easier said than achieved, and let’s not forget that there are a great many other pressures deciding the level and type of migration to the UK.

Is there a right level of immigration?

As with the issue of population size, from an economic perspective there’s no optimal level of immigration, nor of net migration which is the level of population increase once you factor in emigration as well.

Trying to put a number on it - as the 2010 Coalition Government famously tried to do with its pledge to cut net migration to the tens of thousands - is like coming up with a solution without being clear what the problem is.

Net migration to the UK in recent years - and particularly since 2021 - has been far higher than would have been needed to achieve population replacement (and in England, the population would have continued to grow in any case from natural change).

Because of this, even with the population decline the ONS is now projecting it will take until the year 2108 for the UK population to fall to the size it was in 2020. By 2122, the furthest into the future that these projections go, it will still only be back at its 2015 level. This factors in some level of net migration (an average of 230,000 people per year, to be specific).

We’re also not facing the kind of population drop-off as some comparable economies. Germany, Spain and Italy, for instance, are projected to go into sharper and more immediate decline which will accelerate population ageing, compared with the UK.

The final policy-related point to make on this is that migration as a whole can’t be switched on and off like a tap. Yes, since Brexit the UK has far more control over labour and student migration than it has had for decades. But it has no control over the number of British people emigrating, nor the millions of non-citizens who have indefinite leave to remain or EU settled status. Even hard-line migration sceptics rarely propose an end to family migration for the partners and children of British citizens, nor a total end to humanitarian migration in times of crisis.

There will always be some level of in- and out- migration and, unfortunately for statisticians, it will always be nigh-on impossible to predict.

In other news…

The paperbacks of Sum of Us arrived and they are GORGEOUS 😍

We changed the subtitle from ‘A History of the UK in Data’ to ‘The Untold Story of Britain in Numbers’ which I think is a bit friendlier and more accessible. Yes, I went back and forth over whether ‘Britain’ is enough of a substitute for the UK but concluded that it’s widely understood as such and it’s so much less of a mouthful (sorry Northern Ireland).

We also went back and forth over whether to change the colours. For a while I was favouring a slightly faded red background and white title font but decided in the end that the original looked better.

Available now in all good bookshops and online!