23
Examination of witnesses
Dr Ben Brindle and Mary Gregory.
Q20 The Chair: Welcome to the second evidence session of our Economic Affairs Committee inquiry into preparing for an ageing society. We particularly welcome Dr Ben Brindle, researcher at the Migration Observatory, and Mary Gregory, director of population statistics at the Office for National Statistics. We are very grateful for your time. Thank you very much. I should say that this session, as you know, is being broadcast on Parliament Live TV. We will give you a full transcript that is being taken shortly after the session. We will send it to you to make any corrections to the record. Thank you again.
Let me start by asking you a scene-setting question. We would be grateful if you could briefly detail the current demographic trends with respect to ageing, and the headline issues, as it were, that that survey reveals to us.
Dr Ben Brindle: I cannot speak too much about the picture more generally, but with regard to migrants compared to the UK-born, if we look at census data from 2021, people born outside of the UK and Ireland were marginally younger than the UK-born. In England and Wales, their median age was 38. That compares to 40 for the UK-born. In Northern Ireland, their median age was 36, compared to 38 for the UK-born.
That, of course, does not tell us much about how the structure, in terms of age and the age profile of migrants, has changed post-Brexit, because that is 2021. On that front, we know that, according to ONS net migration statistics, around 80% of arrivals from outside the EU post-Brexit have been between the ages of 16 and 64, and the majority of that remaining 20% or so have been under the age of 16.
We can get a better idea of whether non-UK nationals are more towards the start or the end of their working lives with HMRC data. They produce employee counts by different age bands. That shows that the average non‑UK national was marginally older in December 2024 than in December 2019. There are a couple of different patterns that underpin that. In terms of EU-origin employees, they have become slightly older on average, which is a combination of the fact that somebody who was in the UK in 2019 is older now than they were then and emigration among younger age groups, within the 18-to-34 age bracket. Non-EU-origin employees are younger on average, and that reflects the high levels of migration of skilled workers, and students as well, many of whom work while they are in the UK.
The question is how that will change moving forward, because we could see emigration of certain groups being more likely than others, such as international students, and that might change the age profile. There are a couple of caveats, though, on that HMRC data. It does not tell us about the non-working population or the self-employed population—it is employee counts—and there is no year of arrival in that data. Those conclusions are based on what we see with compositions, not based on direct data. That would be possible to get, but with a freedom of information request rather than from public data.
Mary Gregory: I will step back a bit. I guess the main headline is that the population is getting older, in terms of numbers of people over 65 or over 85 but also in proportion of the population. You probably all know that the population is growing, but even within that, the percentage of people over 65 is increasing. In mid-2023, 18% of the population were over 65, and that compares with 13% in 1973. If you look at the over-85s, you have 2.5% over 85 in mid-2023, compared with less than 1% in 1973.
We expect that to continue. Our projections do not take account of policy or changes in the world, but our expectation, if things stay the same as they have in the past, is that by mid-2047 you would have 23% of the population over 65 and 4.3% over 85. That is 3.3 million people over 85. Is it helpful to say now some of the factors that are leading to that?
The Chair: Yes, that would be good, especially the relative contribution of longer life expectancy.
Mary Gregory: There are three main things leading to that increase in the older population. There is improved life expectancy. Up until the 1950s, there were improvements in infant mortality and child mortality that improved life expectancy. From around the second half of the 20th century, there have been improvements in lifestyle, working conditions and health in general, and that has led to improvements in life expectancy.
I can go into more, but, in general, we have seen improvements throughout and it has been relatively stable for the last few years. It is hard to know at this point how much the pandemic has led to that. You might see another increase. In other countries, we have seen a continued increase, so that might carry on, but at the moment it is a bit stable. So, life expectancy is one area.
Another is around the high numbers of people who were born in particular years. Just after World War I and just after World War II, we had very high birth rates. Across the UK, just after World War I is about the only time we have seen more than a million births in a year, and of course, that led to high birth rates in the 1960s as well. If you think about that, of those people born after World War I, those who are still alive are the ones who have now just passed 100, so that is increasing the numbers over 100. Those who were born in the 1960s are the people who are coming into retirement age around now, so they will be over 60. That is leading to more people in those age groups, because they were big birth cohorts.
The third big factor is that there have been lower births in terms of numbers since the 1970s, but also in terms of the birth rate. We look at the fertility rate as a number of people born per woman. A replacement rate, to put it in context, is about 2.1. If you assume no migration, you would need a total fertility rate of around 2.1 for the population to stay stable. Our latest data for 2023 is 1.44 for England and Wales, so you are quite a lot lower than that replacement rate at the moment. I can expand on any of that, but that gives a summary.
The Chair: We will come back to a lot of those things, yes. Thank you very much. That is really helpful.
Q21 Lord Turnbull: Last week, I was trying to get an idea of the growth of the ex ante population, or the native population, and we had two numbers. The professor sprung it on us and did not have the numbers, but I have looked them up. For the first time, deaths have exceeded births, but only by about 15,000. Let us say 690,000 deaths and 15,000 fewer births. The population is determined by two sets of figures. One of them was determined nine months earlier, so these are the births. The other is determined by events 80 years earlier.
My supposition is that we are going through a phase where the number of deaths is actually quite low, because these are the people who were born in 1941, 1942 and 1943. I think the number of deaths will start to rise. The population may still rise, but the number of deaths will rise. With this low population number, we will see, for the first time, quite a significant fall in the ex ante population. Do you think that is the correct deduction?
Mary Gregory: I do not know for sure. You are right that, because we have the higher numbers ageing, we would expect the number of deaths to increase, as our population projections show. At the moment, we have more births than deaths in some parts of the UK, but more deaths than births in other parts, which leads to the overall figure. It is getting very close to being even numbers right now. In future, the expectation is absolutely that there will be more deaths than births across all parts of the UK. Sorry, within all countries of the UK—within England, there will definitely be a variety.
Lord Turnbull: The ex ante population could start to decline quite sharply, being dominated then by the 1.4.
Mary Gregory: The best way to look at that is that, if you assume zero migration, we would have a falling population, definitely.
Q22 Baroness Wolf of Dulwich: I would like to dive into fertility. I have some questions about differential fertility, but first, people put various arguments forward for the decline in the fertility rate. My memory is that, on the whole, we make up these arguments after the changes have happened, rather than being very good at predicting them. My first question is whether anybody saw this very sharp decline coming everywhere, including Scandinavia, France and here. Particularly in the developed countries, there has been a very sharp fall, and to extremely low levels in east Asia. Did anybody see it coming, or are we trying to create explanations because it has happened?
Mary Gregory: I am sure that there are some people. Of course, there are always so many people with a view that someone will, in hindsight, say that they saw it coming, but I do not think that it was obvious. There is a really important thing to note. Even with the fertility rate—the fertility rate, as you say, has gone down a lot—if you look at completed family size, so how many children a woman has had by age 45, that has actually stayed relatively stable. We do not know right now whether the people who are making up that number now within the UK are going to be having children later, and so they will still have a similar completed family size, or whether some of them will not have children, or have fewer children, and therefore we will see a reduction in future. I do not think that anybody can predict that at this point.
Baroness Wolf of Dulwich: It is too early to say whether we are actually seeing a further shift in average age, or whether there is a decline? That is very interesting.
Can I ask a couple of questions about specific cohorts and comparisons? You said, for example, that, in some regions, births are exceeding deaths. Is that because of people moving in and out, or is it because different regions seem to have different fertility rates? For example, you would expect births to be higher in London with a high, relatively young migrant population. Is there anything more to it than that? In terms of family completion and so on, is the country fairly standard, or are there real differences?
Mary Gregory: It is definitely not standard. It is a bit hard to unpick exactly what is causing what. For example, we know that there are far fewer older people in London and other big cities, so that will contribute to places like that having a different make-up.
Baroness Wolf of Dulwich: My question was whether, if you take comparable people in different regions, they are basically the same, or are there actually real differences?
Mary Gregory: I do not know for sure if we take comparable people. I would need to check that. We could look at the fertility rate. I do not have it with me, but we have the fertility rate by region. It varies, definitely. To give an example, in an area with high numbers of students, you quite often find a relatively lower fertility rate, because you have a lot of people who might be at an age that you think might have children, but they typically might be at university rather than thinking about having a family at that point, so you get some slightly strange numbers. Cambridge has one of the lowest fertility rates in the country, but that is perhaps related to having a lot of students. I do not know for sure. It is hard to say exactly.
Then there is the migration bit. I do not know if Ben wants to add, but that can change things as well. For example, where there are high numbers of migrants, we do not know exactly why, but they might have a lower fertility rate in some ways. It might be because they are not coming here and having a child straightaway: they have either already had it or are going to wait a few years. If you see a high level of migration, it might get a bit lower. I do not know exactly, but there are definitely things we can pick out that give an indication.
Baroness Wolf of Dulwich: That leads me to my last question, which is related to the behaviour of migrants. We know that, at least in the recent past, immigrant groups tended to have a higher birth rate, but also that it tended to verge towards—I do not know what you call it—the ex ante means, I suppose. Can you say a little more about that, whether it is true of all migrant groups or whether it is an average, and what we know about how quickly, if at all, fertility rates shift to that of the wider United Kingdom?
Dr Ben Brindle: There is agreement in the academic literature and analysis on this topic that the fertility rates of migrants’ descendants converge with that of the UK-born, but the question of how fast this convergence takes place is disputed. There are a couple of important things at play here. The first is where migrants are from initially. Migrants and their descendants from European countries have lower fertility rates than those, for example, from South Asia. That is an important part of it.
As well, the speed of convergence with regard to fertility rates is affected by two things that can work in opposing directions. There is this concept of assimilation, where, as migrants’ descendants become more integrated into the labour market, education, the housing market and UK culture more broadly, their fertility rates will look more like the UK-born. On the other side, there is the concept of socialisation, so the descendants’ fertility rates will resemble more the country of origin, due to upbringing, attitude and behaviours that they are exposed to during childhood.
With regard to what the literature finds there, to give you a couple of examples, there was a paper by academic Sylvie Dubuc, who found a fall in the fertility rates of the descendants of Pakistani, Bangladeshi and Indian migrants. She put that down to increased educational attainment among these descendants, so acting much in the same way it does for the native population. Hill Kulu at St Andrews has done quite a lot of research on this and found, as I say, that origin plays a big part here, but residence plays a bigger role in the timing of childhood and the number of children that migrants and descendants have, which taps into what Mary was saying. We have this convergence, but as for how quick that is, it is a little bit of an unknown and is disputed.
Mary Gregory: One thing I looked at, because I was interested, is whether the age that migrants are having children is different, even just in the current data. There is a bit of a difference, but it is not as big as I had anticipated it would be.
Lord Londesborough: Coming back to the factors behind the decline in fertility rates, I am interested to get your views on the economic factors. It seems that there is very strong evidence, internationally and within the UK, if you look over the last 50 years, around those countries or regions that have had the strongest level of economic growth and development. The one that keeps being mentioned, of course, is South Korea, whose economy has grown phenomenally, and yet its birth rate has fallen to, I think, 0.7, which is the lowest in the world.
I also read last week an interesting piece on Bristol, which has been one of the fastest-growing cities outside London, in economic terms. That has coincided with the steepest fall in birth rates, down to about 1.1. What are your views on the correlation between a growing economy and falling birth rates?
Mary Gregory: It is really hard to unpick what is causing each aspect. We do not have specifics within the ONS data at this point, but there are all sorts of things we can speculate, based on the data. Is it that education is taking a higher priority initially and that is why people are having children later, or is it because getting on the housing ladder is getting harder, so people are trying to do that first and have children later? A lot of the things you might imagine all lead to the conclusion that you would perhaps end up having children later. We are seeing that people are having children later, so it is quite likely that it is economic factors that are supporting that, but it may be other things as well.
Q23 Lord Liddle: Is there a correlation between level of women’s education and fertility?
Mary Gregory: I am pretty sure that there is, but I would have to check the numbers. That is not something that we have published in the ONS, but I think the academic literature would support that.
Lord Liddle: Has anyone tried to correlate levels of fertility with housing stress? Where there are the greatest problems in setting up a home, does that affect fertility?
Mary Gregory: I have not seen anything. I have seen a lot of articles that make assumptions, but it is very hard to know for sure.
Lord Liddle: There is no hard data.
Mary Gregory: Not that I am aware of. It might well exist and we can have a look.
Q24 Lord Verjee: Ben, if we were to base population projections on the native fertility rate alone, what rates of immigration would be required to keep the current population age structure unchanged?
Dr Ben Brindle: In the UK, the level of net migration that we would need would be large and potentially verging on unrealistic. In 2000, the UN Population Division estimated that the UK would need net migration of over 1 million annually between 2000 and 2050. To put that into context, that is three to four times the level of net migration we saw in the 2010s and is higher than the historically high levels that we have seen over the past few years. The peak was 906,000.
There are more recent projections by the ONS. They suggest that net migration of 245,000 each year, which is a more plausible figure, would still lead to an increase in the dependency ratio—the number of people that the working-age population would be supporting.
There is similar evidence in other countries. This is not a unique challenge for the UK. However, I do not think that the most important thing here is simply the numbers of migrants coming to the UK. The most important thing is the composition in regard to their fiscal contribution, given that an ageing population is primarily a financial problem, in that we have increasing age spending, particularly pensions and healthcare, but a smaller working-age population to pay for that. Migrants will help address the public finance challenges only if they are paying more than they are taking out. The evidence on this from the past decade or so suggests that migrants, as a whole, pay in about as much as they take out.
There are some studies that look over a longer timeframe and make assumptions about the future, and they are a bit more positive in the medium term. One of the largest assumptions there is that the children of migrants will grow up, enter the labour force and offset some of the costs from when they were children. In the long run, migration is less effective as far as improving the fiscal outlook is concerned, because, although some may return home when they reach retirement age, many migrants will stay in the UK and add to age-related spending pressures. That does not do much to change the long-run fundamentals.
In 2023, the OBR projected what might happen to the level of debt relative to GDP if there were high levels of migration. It found that, although there would be less debt relative to GDP by 2072-73, debt would still be 310% of GDP, which it calls an unsustainable path for debt. What really matters here is not, as I say, how many people are coming to the UK, but whether they are contributing more to the public finances and helping to address that financial equation.
Mary Gregory: The other bit I would expand on there—it has already been mentioned by Ben—is whether they stay or not. We know that things have changed a lot about migrants, not least because we now have a lot more from outside the EU. We know that some nationalities are far more likely to stay in the UK longer than others. That is changing a lot over time and we do not know how long they will stay. It is really hard to predict what impact that will have on the profile of the population in the future.
Lord Verjee: Mary, what effect do life expectancy and birth rate projections have on productivity growth and GDP per capita projections?
Mary Gregory: Do you mean because productivity per head is particularly critical, or in terms of how much they are adding to the workforce and what they produce? We know that population changes have an impact on productivity, because, as mentioned, if you take an example like migration, it depends on who those people are and what they are doing.
To take an example, we have had a lot of Ukrainians come to the UK over the previous years. They are not adding a lot to productivity, even though they may well have really valued skills, because they are not necessarily getting jobs that correspond to the skills that they had in Ukraine, perhaps because they do not have good enough English. There are different aspects. To take that aspect, it really depends on the routes and the policy levers that are used. You could have a migration regime that really adds to productivity or one that is less helpful for productivity.
Dr Ben Brindle: The relationship between migration and productivity is more under-researched than I would have thought, given how much of a big political topic it has been. Generally, the findings are that migration increases productivity and this seems to be more concentrated among high-skill individuals.
Quite what the mechanisms are is not entirely clear. Perhaps it is about knowledge transfer, but then you would think this would potentially be in rather narrowly defined jobs where that is more important, rather than more widespread. It could be related to innovation. There is a link between that and high‑skill migration. Research has also found a relationship between the share of high-skill labour in the workforce and the level of automation that firms undertake. They are some of the pathways, but this is relatively under-researched for the moment.
Lord Burns: Looking at this from a public policy point of view, there are four magnitudes that immediately come to my mind. One is what the prediction is. What is going to happen to the working-age population? Secondly, what is going to happen to the number of people over the age of 65, from a health and social point of view? How many households will there be, from the point of view of housing? Then, of course, what about the number of people in education from the point of view of schools? Do we have projections for each of these categories going forward that would help us to look at some of the policy implications of what is happening? They are obviously very different for the different groups.
Mary Gregory: Yes, we do for most of those categories. In terms of the projection of the population as a whole, we published that in January this year, and that breaks it down by age. Soon it will be broken down by local area as well, so you can look by local area by age to understand that, right up until 2021-22. That is something that is used by the Department for Education to look at schools and by the DWP to understand pensions. That is absolutely available. You can look at the old age dependency ratio, if that is something you are familiar with—so, how many over-65s there are relative to working-age population, through that. We also publish household projections. The most recent ones are a little out of date now, but there will be some more in the near future.
Q25 Lord Davies of Brixton: We are talking about population projections. What is the null case in terms of public policy? You could assume all sorts of changes in public policy that would have an impact on future levels of population. What is your central assumption? Is it that things do not change?
Mary Gregory: Yes, exactly that. That is where they have their limitations, but we publish variants as well and I can talk a bit more about that. It is based on what has happened. At the moment, we are using the last 10 years for population projections. We assume that, in the medium term, what happened over the last 10 years will happen again. We look at the short term a little bit more closely, saying, “How do we get to that point?”
If you take something like migration, we have a long-term assumption of 340,000 per year. You may or may not know that that is a lot lower than it is right now, but that is still higher than the historic numbers from 10 or 15 years ago. It assumes no change to policy, because we, as an independent office, would not want to make any assumptions about policy, but someone like the OBR will then make assumptions. In the Budget, it may use one of the other variants, and often does. We do all sorts of different variants, but it could be something like a high migration or a low migration. We have a tool on our website that lets people put in almost any level of migration and see what impact that has on the figures.
Lord Davies of Brixton: What assumptions are you making about changes in mortality rates?
Mary Gregory: For all our assumptions, we gather a group of experts to input into those assumptions. For mortality, we make some assumptions about improvements based on expert judgement. We do not assume it stays the same as it is right now, but they are worked through quite carefully. They are not thinking about what policy might change but what we expect from the future.
Q26 Lord Blackwell: Dr Brindle, I want to crystallise my understanding of the implications of what you said about the immigration rate that would be necessary to keep the population structure stable. If we define the economic problem as an increasing dependency ratio that will put pressure on GDP per head, and you are saying that the migration level required to keep that stable would be a million or so, is the conclusion from that that migration is not going to be a feasible or practical solution to the economic challenge?
Dr Ben Brindle: In short, it is unlikely that it will be, in part because of the numbers needed and in part because migrants also age. There are various ways that immigration regimes could be set up to help that along, such as a selective migration policy, I guess not too dissimilar from what we have at the moment, where the selection criteria is based on those things that we think are more beneficial for migrants’ fiscal and economic contributions, such as education, occupation, age and language proficiency. That might help that balance somewhat, but probably not to the extent where migration is the solution to this problem.
Regardless of the migration system that there is, there are limited opportunities to increase migrants’ average fiscal impact. If we look at migration today, most migration is non-work migration, and this is something that is also the case in other high-income countries. This non-work migration is likely to have a less positive impact on the public finances, but is also less amenable to policy.
A good example of that is asylum seekers. We have seen over the past few years how policymakers have tried to reduce inflows of asylum seekers, but it has proved very difficult. Among those people who come on work visas, there are not too many opportunities to increase their average fiscal contribution, particularly following the changes made by the previous Conservative Government, such as increasing salary thresholds for skilled workers and, related to that, preventing care workers from bringing their partners, and most students too.
Perhaps the biggest opportunity in that regard and the biggest exception to this would be care workers, who tend to make a less positive fiscal contribution. Of course, there is then potentially a challenge where, if you have an ageing population, somebody needs to care for those people. So care workers are arguably one of the exceptions to admitting migrants that make a more positive fiscal contribution being the best way to address the challenges that come from an ageing population.
Q27 Lord Agnew of Oulton: I want to pick up on your last point about raising the financial threshold for earnings. Is £38,000, give or take, the level at which an immigrant starts to make a net financial contribution to the economy?
Dr Ben Brindle: That will very much depend on who one is thinking about. In terms of how that £38,700 threshold was set, my understanding is that that was the median income among those roles that are eligible for a skilled work visa.
As for at what point a migrant makes a positive fiscal contribution, that will depend on a number of assumptions. Household is important here and whether that person has children. It will also depend on their health outcomes and the time period we are considering, because somebody might have a positive impact today but then a lesser one, or the reverse could be true. It very much depends on some of these differences. More generally, those working in higher-paid roles will make a more positive fiscal contribution. Most of those on a skilled-work visa, particularly now with these new thresholds, will be making a positive contribution.
Lord Agnew of Oulton: Is there any data to show us this? You sound a bit vague.
Dr Ben Brindle: Yes. The Migration Advisory Committee in its annual report—I am more than happy to provide this. It looked at different household structures and at what point people would break even in terms of their fiscal contribution.
Lord Agnew of Oulton: That is where we should look.
Dr Ben Brindle: Yes. I am happy to provide that to you.
Lord Razzall: I think that you touched on this earlier, but it is important, as we are on the record, for you to indicate what effect Brexit has on the age and economic activity of immigrants. I think that you touched on it earlier, but perhaps you could be specific.
Dr Ben Brindle: We have certainly seen a change in the reasons people are coming to the UK. There has been a very sharp rise in the number of non-EU people who are employed in the UK. That is not just people coming to the UK on work visas, but lots of groups that come on non-work visas too. That would include the partners of people on work visas, students and their partners, and Ukrainians. Before Brexit, there was perhaps an expectation that lower-wage industries would be largely cut out from being able to recruit overseas workers. I do not think that that has actually been the case in reality, because of these dependants and non-work groups coming to the UK.
The overwhelming story, in terms of what has happened in the world of work, is with the health and care sector. That has overwhelmingly been the reason that work visas have been granted. There were points in 2023 when care workers made up 50% of work visa grants, so it has very much been the key driver. That is underpinned by poor pay and poor working conditions in the sector, which have meant that UK-based workers have not considered those roles attractive to join.
Lord Razzall: Has this had an effect on the age profile?
Dr Ben Brindle: I am not entirely sure, beyond those figures that I cited earlier. This is something that could be requested from HMRC in terms of the age of migrants at year of arrival, but the best that we can work on is that HMRC data with these changes in the compositions.
Mary Gregory: We have some data on the age when they arrive. I cannot remember what it is offhand, but absolutely we are seeing differences. Some of that is around nationality, but different nationalities have different typical patterns, if that makes sense. It is hard to tell how much is the visa type and how much is a change in the nationalities.
To take one example, if you look at students, we have a lot more people who have been coming from Nigeria recently. They are more likely to be slightly older students who are likely to stay longer. They bring more dependants, when they were allowed, or for those small numbers who are still allowed. China has always had quite a high amount of student migration to the UK, but they tend to stay much less time. They will quite often come, do a one-year degree or course and then leave again, and they will not necessarily bring dependants. That change in the make-up of the nationalities is also feeding through to a change in the make-up in the age. We have some numbers that we can share.
Lord Razzall: That would be interesting, yes. I have a follow-up question. Will changing demographics in other countries affect immigration to the UK?
Dr Ben Brindle: Demographic changes in countries of origin might be more important than in peer countries for the UK. We are seeing fertility rates falling around the world, particularly in middle and high-income countries where those citizens are more likely to have a positive fiscal impact if they come to the UK. However, I do not think that it will be the case that the supply of migrants, at least in the foreseeable future, will stop as a result of declining fertility rates, because the UK is still an attractive country by international standards - we admit fewer people than want to come here each year. But we might see a change in the origin in terms of where migrants are coming from as a result of that.
With regard to the UK’s peers and “competitor countries”, we would not see too much of an impact, broadly, from different countries stepping up attempts to bring migrants to them, because migrants do not tend to shop around when deciding where to migrate to, at least as far as immigration policy is concerned. In the academic literature on this, high‑skill workers tend to think about job opportunities or the economic environment more than about immigration policy. The primary thing that students tend to think about is the prestige of the university that they are looking at going to, or the suitability of the course on offer.
Immigration policy tends to be a factor later down the line according to rather limited research, I must admit. In some cases it is considered right at the end, when almost everything else has been decided. Broadly, I do not think that would have much of an impact. The exception to that might be in specific areas, such as for health and care workers, where there is a more finite supply because it is a more limited group. Perhaps if other countries—for instance, the USA—began trying to bring those workers to their countries, the demand might outstrip the supply.
Lord Razzall: Is there not a pressure, for example, from sub-Saharan Africa, which seems to be the only area with an increased fertility rate? Will that not cause pressures on us and, indeed, all European countries?
Mary Gregory: There might be some impact on the economic environment as well. At the moment, we know that we have had quite a lot of students from Nigeria, as I mentioned, but the economy might mean that they cannot afford to come here in future. I think that there will be a bit of an impact from the demographic challenges, but the economic ones are more likely to make the difference.
Q28 Lord Lamont of Lerwick: Part of my question was already asked and answered when the Chair asked his opening question about past trends and whether it was reasonable to expect life expectancy to continue to rise. You dealt with that quite comprehensively. More specifically, what can be said about the expected health of the nation with respect to trends in life expectancy? Are healthy life years—that is to say, the number of years that someone might expect to live without health problems—rising pari passu with life expectancy, or is there a difference between the two?
Mary Gregory: That is a really excellent and important question because, unfortunately, healthy life expectancy is not rising as fast as life expectancy, or has not been in the recent past. That has a lot of implications for helping older people, care support and that kind of thing. We have not done a lot on this in the ONS, although we would like to do more, but the UN has done quite a lot of research. We are seeing, not just in the UK but across the world, that there is an impact on smaller improvements in healthy life expectancy.
Lord Lamont of Lerwick: You are saying that we are keeping people alive, existing, but not very satisfactorily, and the burden is getting greater?
Mary Gregory: Yes, and in some cases healthy life expectancy is actually reducing rather than increasing.
Lord Lamont of Lerwick: Do you have anything to add?
Dr Ben Brindle: I have very little to add on this particular question. In general, migrants tend to be healthier than the UK-born. Part of that is just down to differences in age, but even within age bands the non-UK-born are more likely to say that they do not have their day-to-day activities limited by illness or disability. Those health differences tend to converge though. Over time, the longer a migrant has lived in the UK, their health becomes more like the UK-born. These gaps also tend to narrow as people move into older age, which suggests that age eventually catches up with us all.
Mary Gregory: I have found a couple of numbers that might illustrate the point. If you look at healthy life expectancy now compared with pre‑pandemic, so the 2017-to-2019 numbers, for males there is quite a large decrease in the north-west. Healthy life expectancy has dropped by 31 months, whereas in London it has improved by 2.2 months. For females, the south-west has seen the biggest drop, a drop of 32.4 months, and in London it has increased by just 0.7 months.
Q29 Lord Londesborough: Can we turn to the question of economic activity for the 50-plus age demographic? The first question I have is on overall trends. Looking at what I think is ONS statistics, I see that there has been a very big increase in the number of 65 year-olds in employment, which back in 2014 was 27%, and last year was 40%. That is a pretty major shift. Is that the case for most developed countries, or is that a UK-specific phenomenon? What does it tell us going forward, given what you have told us about declining birth rates? That rate, I presume, is going to have to continue to go up pretty steeply if we are going to see economic growth.
Mary Gregory: I am afraid that I am not as knowledgeable about the economic side, but we can definitely come back to you with more. One thing that is contributing to the increase is the rise in the pension age for females, so of course there are more people working longer who are in the older bracket there. We have also seen some really interesting things in the workplace since the pandemic, with things changing. We saw for a while that the number of people over 50 was reducing, but some of them are coming back into the workplace. There are a lot of things going on, but I will get some colleagues to provide some more written briefing for you, if that is helpful.
Dr Ben Brindle: I have relatively little to add on this particular question. Again drawing on census data, if we look at the activity rate of UK versus non-UK-born males, that is similar at the ages of 50 to 64, but after 65, non-UK males are more likely to be economically active, although that gap narrows as we move from the ages of 65 to 69 and to the 70 to 74 age group. Among females from age 50 to 64, the non-UK-born are less likely to be active in the labour market: they are much more likely to be looking after the family or home. From age 65 onwards, they have higher activity rates, although there is a smaller gap than what we see for men.
What is going on there? I am not entirely sure. It is not simply the case that less well-off migrants are continuing to work once they reach retirement age, because we also see this difference in professional occupations. It could be that some people are returning home when they reach retirement age, so you are then left with those who are still in employment or active in the labour force, but I am not entirely sure what is driving some of those differences.
Lord Londesborough: Following up on Lord Lamont’s question in relation to HLE—healthy life expectancy—is it right that it has stopped growing since the pandemic? Would it be fair to say that that is going to present an employment challenge to the Government? Is there not a correlation between HLE and the ability to continue being in the workforce? While we have seen it improve over the last 50 years quite dramatically, if that has now plateaued, is it not going to present a problem for getting either the 50 to 64 year-olds or the 65-plus cohort back into work?
Mary Gregory: It is a policy question. As you say, the numbers suggest that it is going to be a challenge. The challenge is how you deal with it.
Lord Londesborough: Are there any studies on the relationship between healthy life expectancy and being in employment?
Mary Gregory: I am sure there are, yes.
Lord Londesborough: It sounds like an obvious correlation, but I have not actually seen data that backs that up.
Mary Gregory: I have not.
Dr Ben Brindle: I am not familiar with any, sorry.
Baroness Liddell of Coatdyke: Where I am getting a bit confused is the correlation between ageing and the nature of the jobs that are done. Where I come from, 50 years ago it would have been coal, iron and steel and shipbuilding. Now it is bioscience and computers. How does that fit in, where people do not have the intense physical jobs that we used to in shipyards, where it used to be very rare to see people over 50. Now, in bioscience and in computing, you can have people who are 60, 65 or 70. How can you take that into account?
Mary Gregory: It is a real challenge and there is not any clear evidence about whether, for example, productivity is greater for older or younger workers. There is some suggestion that older workers have more experience and therefore there is greater productivity. In the examples you give where they are less physical, you could imagine that that would be the case. Equally, younger workers are considered more dynamic, so they are more able to change or move area if they need to. At the moment, the evidence is not clear. I hope that will become clearer over time.
At the ONS, we are in the early stages of trying a project called linked employer employee datasets, linking data to try to look at what happens as people move employers. We can start to understand a bit more about the wages and productivity and get that greater insight.
Lord Blackwell: There was a question earlier about what level of immigration would be required to keep the age structure and the old age dependency ratio constant. Has anyone done the calculation, or can you do the calculation, on how much you would need to raise the retirement age in order to offset the increasing dependency ratio—in other words, to keep the proportion of employed versus retired people constant?
Dr Ben Brindle: I am not aware of such a study.
Mary Gregory: I do not know whether anyone has done it, but I do not think it would be very hard to do, because we have the projected numbers by year.
Lord Blackwell: Are you volunteering?
Mary Gregory: I will take it away and see what we can do.
Lord Blackwell: It would be an interesting calculation.
Mary Gregory: The data is all published, so there may well be somebody listening who wants to give it a go.
Lord Davies of Brixton: I have seen the figures. They are available. I cannot remember where though.
Baroness Wolf of Dulwich: It would be very good to have that.
Mary Gregory: To give a sense of the scale, at the moment the old age dependency ratio is 30 for 2023, so that is 30 people over 65 for every working-age person. It is expected in 2047 that it will be 37, so there is quite a big change over time. It would be interesting.
Lord Blackwell: It would be useful to know whether that means everyone has worked five more years or 10 more years.
Mary Gregory: I would imagine that DWP will have looked at that when thinking about its pension plans, but we can check.
Q30 Lord Turnbull: I will come on to care in a minute. Following up on Lord Blackwell’s question, on the numbers you gave on the amount of migration that is needed to make good the shortfall in the growth of the native ex ante population, we have only just passed the breakeven point. At the moment it is something like 15,000. There is an idea that it is going to go up to 250,000 or 500,000. Something has to change. It cannot be based on a projection of what is happening now. It must all be based on projections of how these two variables will develop in the future.
Dr Ben Brindle: My understanding is that that is the case.
Mary Gregory: I am not sure. I do not know the million figure but, in terms of what is happening now, that figure is 16,300 more deaths than births. The important thing to remember is that bringing in a migrant is not exactly equal. To eradicate the impact of that, you would need more people to come in as migrants than the difference between the deaths and births, because they are coming in at an older age, so their impact on the structure of the population is a bit different.
It is a lot more complex than picking a number, because it depends on what age they come in and how long they stay. You can imagine, if it is students, they would be coming in, maybe staying for a few years and going again, and they would help bring down that dependency ratio. If you have people coming in at more like 40 or 50 and staying for a long time, they will then add to the older age groups and that will have a much bigger impact.
Lord Turnbull: Can I come back to this figure, which I think Lord Agnew raised, of £38,000 as a threshold? A lot of people, I think, would draw the conclusion from that that, if you are allowing in healthcare workers earning, say, £20,000, there is somehow a net loss for the country. I will give you a scenario where I do not think that that is the case.
Let us take women. Over the last 30 years, their education and access to high-paid jobs has grown enormously. They get to the age of about 50 and then there is a crisis in the family. There is an elderly relative who needs to be looked after and then a choice that that family has to take: “Can I find care at a price and availability that I am satisfied with, or do I do the job myself?” Part of the problem in the crisis in the care sector is that more people are being forced to do the caring themselves when they really prefer to work.
If the effect of bringing in a £20,000 care worker from the Philippines enables someone with possibly a PhD or a professional qualification to carry on working, this must be quite a big benefit to us. With this £38,000, you have to look at what is displacing what. If it is enabling well-trained people to extend their careers, it must be beneficial rather than the opposite.
Dr Ben Brindle: Logically, that makes sense. The studies I have referred to tend to think about the direct impact of migrants, so that a person earning, say, £20,000 might have a negative fiscal impact. That is what will be reported in those studies. As you say, it is very true that, if that avoids the displacement of somebody who would make a positive fiscal contribution, the net effect could be more positive between them both.
Lord Turnbull: If we look at how this is recorded in the employment statistics, if that person makes a choice to withdraw from work and care for granny, the labour supply goes down. If it is the opposite and that person stays in work and employs someone else to look after the elderly relative, the activity rate goes up. We need to disentangle this. Is this care problem sufficiently statistically significant to affect the aggregate composition of the workforce?
Mary Gregory: I do not know the answer to that, but it is one of many questions that we would really like to have greater insight on. As we mentioned, the Migration Advisory Committee does some really good research to understand more, but there is a lot more that can be done.
One of the things that we are trying to do in the ONS is a lot more of linking with the administrative data so that we can understand a lot more about the journey of someone who comes in as a migrant. You could also start to try to work out—this is very long term—what those kinds of impacts are. As I say, it is a lot to unpick. It is very hard to work out whether there is a single thing that has an impact.
Lord Turnbull: The journey of the person whose job that person takes has a big impact.
Mary Gregory: Yes, exactly.
Baroness Wolf of Dulwich: I have a question that follows on from that. Andrew, you are coming back to this issue later. It does seem to me that, as you say, it is going to be very complicated. It is going to depend on what job they do, when they drop out, how many extra years they spend in the workforce, whether the care worker comes in with a family and all the rest of it.
I have a very straightforward question, which is about the availability of the data. What data do we have on exactly what sorts of jobs care workers are doing and the population of people who they are looking after? Do we have very detailed data on the demographics of the cared for?
Mary Gregory: It is a bit of a mix. At the moment, what we use for our migration statistics is mainly the visa that someone comes in on. We know whether they came in on a care worker visa. We can see when they change visa as well, which is really interesting. We will publish more on that in future, because that tells us quite a lot about what is going on. Some people might come in as a dependent and then get a work visa, for example.
In the future, we are looking to do some linking with such things as HMRC data. We might get a much better understanding not just of whether someone has come in, but what job they got, how often they changed jobs, how stable their job was and what their earnings were after being in the country for however many years. There is huge potential, but at the moment it is a potential that we are working towards.
Dr Ben Brindle: Just to add to that, the Home Office will have a record of who has sponsored care workers on the health and care visa. That data will exist in some form. I am not sure it is publicly available. It has been reported on by groups such as the ICIBI, the Independent Chief Inspector of Borders and Immigration. That data certainly does exist, if we want to see which sponsors were doing it and to discover more about who those people are, who is being cared for and things like that.
Lord Blackwell: Baroness Wolf asked a question earlier about whether anyone had forecast the decline in the dependency ratio. Listening to the conversation about care for the elderly, if people act rationally, if it is a rational response to economic and social pressures, could we be looking at a situation where it becomes a rational thing to have more children in order to ensure care in your old age and this problem goes away?
Mary Gregory: That is not for us to comment, but I guess the big question is whether people act rationally.
Lord Blackwell: Nobody has predicted this at the moment?
Mary Gregory: Not that I have seen, no.
Q31 Lord Davies of Brixton: On the birth rate, we have heard evidence that nobody knows anything about how to make women have more children. I had thought that 10 years ago Norway, or Scandinavia more generally, had provided the answer. They provided very good child support, nurseries and long paternal leave. The general reaction was, “They have got it. That is it. You just support women”. They have now let us down by coming into line with everyone else. Is that correct? That is my understanding.
Mary Gregory: Internationally there is very little evidence of what policies make a difference. As you say, there are some examples where people think they have got it, but I do not think anyone is fully confident. It is also because there are so many different things interlinked that you cannot identify one thing that is going to make all the difference. It is probably a combination of factors.
Lord Lamont of Lerwick: Can I ask Mary about falling populations? Some European countries have falling populations. Some Asian countries have as well, notably Japan. Is everything you have been saying the opposite when you have a falling population? Could a falling population, in some circumstances, modify some of these effects?
Mary Gregory: That is a really interesting question. Typically, a falling population is because you have an ageing population. Japan is a prime example of that. I would imagine it would be very hard to turn that around, but I have not given it much thought.
Lord Lamont of Lerwick: Could you give it some thought?
Mary Gregory: I will try.
The Chair: We have given you lots of extra work to do, I am afraid.
Lord Davies of Brixton: This is probably an unfair question, so you do not have to answer for the ONS. It has been a bit of a torrid time for the ONS recently with some of the statistics, although they have broadly been sample statistics rather than your end of the thing. Could you let us know how good your population projections have been for the last 20 years? You see these fan charts of where it was predicted to go year by year. It would be really good to know how you score yourself.
Mary Gregory: Yes, the easiest way to see this is to look at the projection each time we publish it and how that chart changes. It is quite hard to describe. If you could see that picture, you would very quickly get the impression, so I will send that over afterwards.
The one thing that you can be sure of is that they are always going to be wrong, because they are talking about the future and we do not know what the future is going to look like. The really important thing—this is something Tim Harford said a few months ago—is not whether they are right but is whether they are useful. We are not trying to model the future. We actively say we are not trying to do that. We are trying to provide something that is useful for policymakers, so they can understand, “If the world looks like this in future, what should we be doing?” We produce variants as well, so they can make a judgment if they need to: “Do we think we will have a high-migration scenario? Do we think it will be a low-migration scenario?”
The really important question is whether they are useful. I very much hope they are, but I will leave others to judge. There are lots of things built in to make clear they are not a prediction of the future.
Lord Davies of Brixton: It is about whether they are useful in terms of making decisions now.
Mary Gregory: Yes, exactly.
Lord Davies of Brixton: It is not about making sure we know what is going to happen.
Mary Gregory: Yes, exactly. Do they inform policy decisions or business decisions? How many houses need to be built? How many schools need to be in place?
Q32 Lord Burns: You spoke earlier about this question of measuring fertility rates. There are issues about what the current rate is and what the rate is when you get to the end of women’s fertility life. What are the latest figures, if you take the second of those measures?
Mary Gregory: I cannot find it here, but I think the completed family size is 1.94; it has stayed pretty similar for the last 15 years or so. It is really interesting. The narrative in the media does not ever give that sense of the completed family size. It is quite a surprising fact for a lot of people.
What we do not know, as I said earlier, is whether the world has changed for people right now. It will look different by the time they all reach 45. The last cohort of people that have reached 45 are people who were born in 1977. Tomorrow we will be publishing the 1978 numbers, but that is still quite a big difference.
We can see things about more recent cohorts. We can look at people born in 1993 and how many children they have had at this point, by the time they are 30. That is lower. People are definitely having children older, but we do not know whether they will go on to have them—
Lord Burns: We do not know whether they will make up for not having them at a younger age.
Mary Gregory: Yes, exactly.
Lord Burns: I was looking at some numbers at some point in one of the pieces of reading. It seemed to me that you got quite a different answer if you looked at different measures. It is possible that maybe what has been happening is not quite as stark as simply measuring the current rate.
Mary Gregory: Absolutely, yes. As you say, if you look at the birth rate just as a number, it changes massively, but that is because it depends on how many people there are in the population already. The total fertility rate is definitely a better measure than just doing that. It is useful for short‑term policy decisions to know how many children there are at a given time, but it is not helpful for understanding whether that trend will continue.
Lord Burns: When you come not so much to forecasting but to predicting present trends, the second of those measures is more relevant to what is going to happen to the population going forward, is it not? It is going to be about completed family size.
Mary Gregory: You are right, but it is not so easy to put that into the projections. We have to use the fertility rate in the projections.
Lord Davies of Brixton: Just trying to work this through, it is about the interaction between completed family size and the period over which that size is reached. If women are having their children older, it will not make any difference to the ultimate size of the population, assuming it is a steady state. It will just affect how long it takes to get to that steady state. Is that right?
Mary Gregory: It depends on whether it continues. It cannot continue indefinitely.
Lord Davies of Brixton: Yes, but there is a step change now.
Mary Gregory: Yes, exactly. There is almost a delayed impact.
Lord Turnbull: In response to Lord Davies’ earlier question, you offered to compare outturns with projections that had been made, say, 10 years earlier of where we would be in 2025. There is also the problem of being right for the wrong reason. You may get that figure spot on, but the division between the native ex ante population and migration might be completely different. Could you tell us, if a projection was made for 2025, what the balance is in terms of where these people came from?
Mary Gregory: Yes, we could look at that. I love the chart that shows the different lines. They are obviously not all in the same place, but you get a sense of that variation. We can also provide those for fertility, mortality and migration.
Q33 Lord Turnbull: I have a question about care and the care problem. A lot depends on how the care is being provided and what the alternative is. If we call upon informal care to a much greater extent, it reduces the average skill level, because experienced people are taken out of the labour force to look after elderly people. If lower-paid people come in to do it, it releases experienced workers in the labour force to carry on. I do not know whether we can know the answer to that without examining that.
I also want to follow up on one thing. If it is simply the case that the fertility rate goes down until the point at which women are delaying births to the maximum extent that is biologically possible, might the trend revert? In an extreme case such as Korea, is this simply a spreading out of families, or are people having fewer children? Completed family size has gone down there as well.
Mary Gregory: That is the big question, is it not? We do not know at the moment. At the moment, the delay is getting longer and longer so the fertility rate is staying lower. At some point you would imagine that that has to stop, but when and by how much?
Lord Turnbull: It has not stopped yet.
Mary Gregory: That is the big question. To go back to the question about carers, we have a little information on unpaid care. We do not know what the future looks like, but the census does include some information. A lower percentage of people were providing unpaid care in the 2021 census compared with 2011. That is different depending on the number of hours. That was mostly around those providing 19 hours or less.
Lord Turnbull: That is what you would expect with the rising participation of well-educated women in the labour force.
Mary Gregory: Yes. What we have not done is attempt to predict the future.
Q34 Baroness Liddell of Coatdyke: I am quite interested in how productivity changes as workers get older. To some extent, this is associated with what I asked you last time around. How might the productivity of the workforce in aggregate change as its composition shifts towards older employees?
Mary Gregory: At the moment, we do not have the evidence to know. As I said before, there is some evidence that suggests older employees will be more productive. There are some studies, not by the ONS, that look at start-ups. There are a lot more younger people starting start-ups, but the ones started by older people tend to be more likely to be successful and to continue. That is one of the only small chinks of evidence, but we do not know for sure.
Baroness Liddell of Coatdyke: That is interesting. That is a very interesting thing to look at.
Lord Razzall: That is very complimentary to this committee.
Lord Londesborough: It is something to do with our experience and wisdom.
With physical, manual labour, it would be fairly extraordinary if there were not a correlation between declining productivity and age, but for office‑based work, it will be very interesting, because it is going to become more and more relevant to this economy. The population is clearly ageing and the workforce is ageing with it.
It would be interesting to know whether 50 to 64 year-olds are more productive and perhaps get less distracted by certain things that I will not mention. In the private sector, there are lots of privately held views about employing certain cohorts, and their productivity. It would be really interesting data if it could be produced, if it does not exist at the moment.
Mary Gregory: I will commit my colleagues in the ONS, when they have done this work on the linked data, to come back and tell you more.
Lord Londesborough: It is another one to add to the list.
Q35 Lord Agnew of Oulton: You made a very interesting comment a few questions back about this widening gap between longevity and healthy longevity. Parts of England have deteriorated by about three years. When you do the bits of homework that Lord Blackwell and others have asked you to do, can you just expand on this a bit to see whether there is a trend here? I am very worried about it. We are not going to get to grips with the NHS any time soon. I am not sure that personal responsibility for health in old age is particularly good either. It then folds back into this point about a later retirement date. Is that really feasible if people are getting sicker earlier?
Mary Gregory: Yes, I will.
The Chair: I have a follow-up question specifically on migration. We have now had a number of decades of significant migration. Do we know the percentage of the 65-and-over population that are historically migrants? How has that changed over time? In other words, are migrants leaving? At what rate are they leaving or staying? Do we have any sense of that over their lifetimes and whether those life cycles have changed with different cohorts? Does that make any sense?
Mary Gregory: I do not know the answer, I am afraid, but I know that we could get some information on it. For example, the census gives us country of birth. We could look at that. More recently, we have much better information because we do a lot more of linking of records. We see when the same people come and go. Previously, we asked someone if they had arrived and we asked someone if they had left, but we did not link the two. Since the more recent past, we can get much better insights, over time, about who is coming and going.
The Chair: It is not a perfect match, because foreign-born is not the same as being a migrant.
Mary Gregory: No, exactly.
The Chair: Is it easy to find out what the foreign-born percentage of over‑65s has been over time?
Mary Gregory: Yes.
The Chair: That is publicly available. Ben and Mary, thank you so much. We are really grateful for your time. We have learned a lot. I am afraid the bad news is that we have given you lots of homework to do.
Mary Gregory: It is all so exciting.
The Chair: If you could send it in by the end of the day, that will be fine. We look forward to getting more data from you, but we are really grateful. Thank you very much. The meeting is now concluded.
