Official Report 172KB pdf
The final item on the agenda is the commencement of evidence taking for the committee's inquiry into the Scottish Executive's fresh talent initiative, which the committee decided upon a number of weeks ago. We have a paper from Duncan Macniven, the registrar general for Scotland, and Celia Macintyre, the head of the demography branch of the General Register Office for Scotland. It is my pleasure to welcome them both to the committee today. We look forward to hearing your evidence at what is the start of this inquiry. I invite Duncan Macniven to begin with some words of introduction and any opening remarks that he wishes to make.
Thank you for your words of welcome. I am grateful that you started with the background facts. In our paper I have done my best to set out in a general way what is happening to Scotland's interesting demography, and I have focused on some of the information that we have on migration. Behind the paper, which I have kept brief, we have a huge amount of extra information. As your inquiry proceeds, if you want to focus in on any aspect, please feel free to ask us to expand on matters or open up another issue or come back here and give formal evidence again. Would it be helpful if I gave a resumé of my submission, or do you take it as read?
Are members comfortable with the contents of the submission?
Okay. We will pore over the details in the paper, but if you would like to make any general comments please make them now.
What I would say by way of introduction is that interesting and unique things are happening to Scotland's demography. We do not know precisely the reasons why they are happening, but we can speculate on some of them. The issue is important. You would expect us to say that, as we are the experts who are focused on the matter, but there is public understanding—fostered by me and my predecessor—that something needs to be done about the issue.
Would Celia Macintyre like to make any opening remarks?
No.
Thank you for those comments and for a very informative submission about the population of Scotland.
You are right to say that the picture is a changing one. Since my predecessor first gave prominence to the issue there has been an unexpected upturn in the birth rate. That may or may not be temporary. The long-term trend in the birth rate is still downwards. The number of births is below the number of deaths, so there is still a gap between those two numbers. If births and deaths were the only factors the population would be tending to decrease.
Your view is, therefore, that migration has effectively flattened out.
As the graph shows, the position is nothing like as negative as it was in the 1960s or even in the 1970s or 1980s. However, it is not really flattening out. There is still an oscillation of perhaps plus or minus 10,000 around the point of zero net migration. Some years we lose, some years we gain.
Your submission also states that it is scientifically difficult to provide robust data on migration. What more can you tell us about the difficulties of tabulating migration patterns? How confident should we be in the estimated net migration figures contained in the submission?
The difficulty is that which I referred to, namely, that people do not clock in or out as they cross the border. To overcome that difficulty, we use two main sources: general practitioner registrations, which assumes that people register with a GP when they come to Scotland and, perhaps more crucially, deregister when they move away; and a survey known as the international passenger survey, which samples passengers travelling to and from Britain by the main air, land and sea routes.
Paragraph 2 of the submission states that you expect the population to reduce to 4.88 million by 2028. Notwithstanding what you said about migration patterns, does it remain your robust prediction that there is a problem that requires to be addressed?
That is correct. Tomorrow we will issue fresh estimates of the population in the middle of last year. That will give us a basis for revisiting these forward projections. The estimated figures will change because we are taking a different starting figure. However, the long-term trend is undoubtedly downwards, as will be seen from the information that we have provided. The submission makes a further point that we have not yet touched on, namely, that we are dealing with an aging as well as a falling population.
I will follow up on those points. I am trying to grapple with how we measure population and to what degree of accuracy. I presume that the census provides a fairly accurate snapshot at a given point in time. Is that correct?
Correct. It is obligatory to put in one's census return. In 2001, 96 per cent of people did so. By conducting a follow-up survey to detect the people who did not, we were able to apply to the raw census data—to our minds, with a great degree of statistical accuracy—a correction that accounted for the remaining 4 per cent. That correction is not perfect; in particular, it tends to miss out young men, but it is the gold standard.
You can account for the people who did not take part in the census. Social scientists would say that if they took an opinion poll today, there would be a margin of error of plus or minus 3 per cent. Can you build in a margin of error to your census data?
That is exactly what we did when we published the census data. I can produce the figure for the committee, although it is not on the tip of my tongue. Celia Macintyre might know what it is.
I am afraid not.
Are we talking about a figure of plus or minus 3 per cent?
We are talking about a very small margin of error.
It seems that there are two other kinds of measure of population—projections that look into the future and mid-year estimates. How often do you produce mid-year estimates?
We produce them every year in the spring; they relate to the previous summer. We will publish such an estimate tomorrow.
I do not suppose that you will be able to tell us about that estimate today.
Your supposition is correct.
The mid-year estimates are based on updates to the census, so the mid-year estimates and the census are inextricably linked. The estimates use the census data as their starting point and make the relevant adjustments on births, deaths and migration.
Births, deaths and migration are the only three components of the adjustments. We have very good information on births and deaths; migration is more difficult to measure, as I have explained.
The census gives us a very clear picture of the population and the mid-year estimates give us a reasonably clear picture of it. It is when we come to population projections that we get into the territory of crystal ball gazing—even when we look ahead only to 2028. Is that correct?
Absolutely. That is more true the further into the future we move. I was perhaps doing us an injustice by using the phrase "crystal ball gazing"—unless you have great faith in soothsayers with crystal balls—because the work that we do is rational. As I have said, it is based on assumptions about fertility, mortality and migration, which we make explicit. For example, the assumption on migration that underlies the projected population figure of 4.88 million for 2028 is that we will lose 1,500 people per year, but we did not lose 1,500 people in 2002-03. Perhaps the migration picture is changing. I have given what we call the principal assumption—the assumption with which we are happiest—but we commission the Government Actuary's Department to work on a range of variants that make different assumptions about births, deaths and migration.
I think that I read somewhere in your submission or in the paper from the Scottish Parliament information centre that the population figure had been reviewed. At one point, we estimated that the population would go below 5 million by 2009, but following the review the revised date for that figure was 2017. Was that revised timescale calculated on the basis of mid-year estimates?
Yes; we reviewed last autumn our forward projections in the light of the mid-year estimate announcement that we made a year ago tomorrow.
So, even if we feel that the information is reasonably accurate, in 18 years there will be quite a small population change.
Yes, because what we are seeing is quite a gentle change around the zero net-migration figure. If we were losing lots of people year after year, as in the 1960s, the calculation would be simpler, in a way. The oscillation around zero is a difficult environment for us to predict from.
Did the revision relate to migration?
It related to migration, but also to births.
The revised projections were different because the starting point was higher. The long-term assumptions were the same: the assumption of the loss of 1,500 migrants and the birth and death assumptions were the same because the assumptions are revisited normally only every two years. However, the Government produced new projections for England and Wales, which had recently revised their population estimates quite a lot, so we also revised ours, in order that a UK figure could be produced. We kept everything basically the same, except that we had to take into account the new mid-year estimate because we had one. It would have been pointless for us not to have done that.
You said that death and birth rates are crucial—they are clearly measurable. Have you built into your population projections the impact of policies that are being put in place now—for example, the potential anti-smoking legislation, the healthier lifestyles agenda and the campaign to get people to take more exercise? Are such things being built into the death-rate and birth-rate predictions?
Yes—we do our best to take account of such things. Until the Smoking, Health and Social Care (Scotland) Bill becomes law, we will not make any assumptions about smoking. However, we will examine carefully past trends in the death rate and work out how those will play out in the future, which is difficult to do. What impact will the national health service make on the long-standing problems of life expectancy in the west of Scotland? It will certainly make an impact, but it is difficult to say how great and over how long that impact will be.
Thank you. I look forward to tomorrow's announcement.
Yes. It will be an interesting announcement. I am very sorry that I cannot say more.
Your paper states that Scotland's population is projected to decrease to fewer than 5 million people by 2017, hence the fresh talent initiative that has been announced by the First Minister. What are the Executive's targets for immigration to deal with the problem of depopulation, and how were those targets calculated?
They are obviously the Executive's targets; they are not targets that I set. Therefore, in answer to your question I shall explain the Executive's announcement. The target about which the First Minister spoke was that we will prevent the population from falling below the important—psychologically, at least—figure of 5 million. When the First Minister made that announcement, we were forecasting that the population would fall below five million in 2009. All the Scottish Executive did was work out how many extra people would be needed to fill that gap between 2004—when it made the announcement—and 2009. It came up with the figure of 8,000, but I do not know whether it dignified that figure with the term "target". It was not a sophisticated calculation.
Who will monitor progress towards achievement of the Executive's target and how will they do that, bearing in mind the difficulties that you have mentioned in measuring migration in and out of Scotland? People can move from here to other parts of the UK and the Republic of Ireland without a passport and have freedom of movement within the European Union. Not everybody goes to the bother of deregistering with their general practitioner when they leave, or of registering with a GP when they come in. The international passengers survey is based on samples. How accurate will the monitoring process be and will your departments be involved in it?
Yes. The Scottish Executive will do the monitoring because the targets are its targets. It will be conscious of the uncertainty in the figures that we have been discussing and which it will use, but I do not want to exaggerate that uncertainty. The figure of 8,000 is beyond the margin of error that we would expect; we would expect to be able to pick up a migration change of 8,000. I am not saying that our figures are perfect, but they should be adequate for the purpose, especially if they are taken not just one year at a time but over a longer period. The population changes that I have been talking about are slow and happen over a long period. My impression of the fresh talent initiative is that it is a long-term policy that needs to be monitored over the long term. That will help us to overcome the inadequacies of our figures, which are to a certain extent about determining whether change happened this year or last year, rather than whether it happened at all.
What is your margin of error? You said that you would be able to pick up a change of 8,000, but would you be able to pick up a change of 1,000?
No, I do not think that we could. The yardstick for margin of error that I would use is the adjustment that we made when we got the cross-bearing of the 2001 census, which was in the low thousands—fewer than 5,000.
So that change was somewhere between 1,000 and 5,000.
It was something of that order. I am sorry that my answer is imprecise, but you will understand why.
Can you give us a more precise figure?
No—I do not think that I can, although Celia Macintyre might be able to be more helpful.
The context that we can provide is the report that was produced on the census and in which evaluation was made of the estimates that we had rolled forward from 1991. Comparison was made between those estimates and the figures in the census, and with other data sources for age groups that might be considered to be reliable. A report was published on the quality assurance of the census results, which fed through into our decision to make an adjustment each year from now until such time as we can improve the migration situation.
The impression that I have left you with of a margin of error of between 1,000 and 5,000 is of the right order. I would not hang my hat on a particular figure, but that is the order of magnitude.
If it takes your department, with all of its expert statisticians, almost a year to estimate Scotland's population at the mid-point of the previous year, how long will it take the Scottish Executive to estimate net migration or emigration?
The Scottish Executive does not need to calculate that figure; we provide it nine months on, as you say. The Executive does not need to make a fresh calculation. The urgency that underlies your question does not quite fit the long-term nature of population changes; we are not in a "population crisis"—to use a term that is beloved of sub-editors across the land—but we have a substantial problem that needs to be played out over the longer term.
Earlier, the convener said that having read the introduction to your excellent paper, it seemed to him that the problem had virtually been solved before we started. We can perhaps take some relief from that.
Scotland is in the negative part of the graph in figure 1 because, like many parts of Europe, we are suffering a natural decrease in population because deaths exceed births, which is not unusual. What is unusual is that, although we are an attractive region within the UK, we are not attracting anything like the number of people that London, the most attractive region, attracts. I suspect that there is a geographical reason for that. What is also unusual is that the UK is not experiencing the levels of immigration that other parts of Europe experience. The populations of other parts of Europe, such as the Netherlands, Belgium, Sweden and Austria, are rising because of that larger volume of immigration.
On deaths exceeding births, I point to the fact that life expectancy seems to be rising in Scotland and across the UK.
I would not want to give the impression that Scotland is a very fertile nation; I am sorry if my paper does that. Our birth rate is almost exactly at the EU average, and there are plenty of European countries, such as Ireland, that have a much higher birth rate than we do. Indeed, other parts of the UK have significantly higher birth rates than we do. I do not want to leave you with the impression that Scotland is a very fertile country.
You say that it is not in your remit to consider that matter, but would not it be a factor in any future population growth? Would you be prepared to produce some figures on it?
I am sure that the clerk could get the figures from the health service without difficulty. The number of abortions underlies our projections of the birth rate in future.
On a different tack, you have suggested that Scotland is attracting more migrants, and you said that most of those migrants are fairly youthful. Given that our principal concern, which is about the working population in Scotland supporting the aging population, is connected to the issue of youth, will the fact that we are pulling in those youthful migrants assist us in the future? On another aspect, are we in a position yet to assess the increase in the number of migrants as a result of the accession of the new EU countries?
Will youthful migrants help? Yes, they will. I draw members' attention to our paper, which mentions the fact that the majority of migrants, both from the UK and from overseas, are in the age group—16 to 34—from which we are losing people. They are filling the trough in the population very nicely. Of course, it is not Peter Pan who is migrating here, which is a fact that we need to build in for the future. However, as you say, in the short to medium term the pattern is helpful. We have figures for the number of people who come here from the accession countries, although those figures are not on the tip of my tongue. Mr Gallie probably knows this better than I do, but Britain chose the option to get people from the new accession states—except Malta and Cyprus, I think—to register when they come here. We have figures from the Home Office for Scotland's share of that. If Celia Macintyre does not have those figures at her fingertips, we can easily provide them. Would that be helpful?
I would be very much obliged if you would do that.
You referred to figures that show that Scotland has relatively high population decline and relatively low fertility compared to the rest of the UK. Have you analysed those figures? Are there any explanations for that?
One can speculate on a whole lot of things, but last weekend we announced research on precisely that issue. The research, which is funded by the Economic and Social Research Council and the Scottish Executive but promoted by us, aims to try to explain two things on which two separate research projects that were announced last week will bear. One project is to consider why we are different from the rest of Britain, and the other is to consider why there are differences within Scotland. I allude to those internal differences—for example the relatively high fertility in rural areas—in the paper. We do not really have an explanation for them, although we have hypotheses.
How long will it take to get answers?
We expect interim results about a year from now.
That will not help our inquiry.
That is probably too far away for your inquiry.
What you show in the paper and in the figures is that we are dealing with long-term demographic trends. Can you think of any examples where such long-term demographic trends have been changed in a country like Scotland?
I am sure that such trends will have been changed. It is not impossible to do that.
That is what we want to hear.
I could give the committee a paper on the international comparison, if that would be helpful. Over the past couple of years in Scotland, we have seen two things change very interestingly. One is the birth rate. It is perfectly normal in developed countries for the birth rate to fall, but it is unusual to see an upturn over two or three years—we do not know whether it is a swallow or a summer. The other matter is the changing migration pattern, which seems to be flowing in Scotland's direction. Granted, the demographic decline is quite slow, and there is good scope for changing that.
You deal with facts and figures in your profession. Do you give credence to politicians or anyone else who might try to change such trends by public policy measures or is that impossible?
I would not like to comment at this gathering on whether I give credence to politicians.
Feel free.
I certainly think that levers can be pulled by Government and that, if they are pulled, it is likely that they will make a difference. I wonder whether even just having the debate that we have had this afternoon, and in the media over the past few years, changes things. Perhaps that is naive; I do not have evidence that that has happened, but it would be one way of explaining why the curve of births has turned upwards unexpectedly.
It would be helpful if you could give us evidence of successful initiatives to deal with such situations in comparable countries around the globe.
The two examples that we have examined were in Sweden—where I think steps were taken that were effective for a while, but which did not change the long term—and in Singapore, which is so demographically different from Scotland that comparison is not useful.
It is worth pointing out that there is in the 2002 annual report an article on policy levers by one of the researchers who worked on patterns of fertility. That is not unconnected with the work that is being done at the moment, and it might be useful to look at it. I will give the committee a copy.
The report is on the web and is easy to refer to.
I am not sure that just talking will do much for the birth rate.
That might have been my being a starry-eyed idealist.
I can see from graphs in the submission that there are clear natural patterns; there is a boom, then 20-odd years later, there is another. However, the overall picture shows a decline in the birth rate. I also noticed the fairly stark period of net migration from Scotland throughout the post-war period until the 1990s, with the heaviest emigration occurring in the 1960s. I presume that that consisted of emigration of mainly youngish, economically active people of an age at which they were most likely to have children. That had an impact then, but would have had another one 20 or 30 years later, when the children of those people were having their children in England.
Yes. You are absolutely right. In the births graph—figure 5—two obvious spikes immediately follow the two world wars and there is a lumpier rise in the early 1960s, which echoes the post-second world war baby boom. There is scope for other echoes.
That can be considered with figure 9, which shows net migration by age. The average age for having children has significantly increased, which suggests that there might be hope for us. There seems to be net migration into Scotland by people in their late 20s and early 30s. Is that likely to have an impact on the birth rate? I hope that I am reading my chart—which is in black and white rather than colour—properly.
We must consider figure 12, which is a snapshot of the year before the census. You are dealing with a much longer term. Figure 12 shows that we are losing people in the 16-to-34 age group, which includes the most fertile ages. Therefore, there is a deficit to be made up.
I appreciate that. However, the charts seem to suggest that there is a deficit at one stage but that people return as they get slightly older.
I am sure that that happens to some extent. The fact that people come back to work is interesting. Figure 12 shows that Scotland is gaining people in the 35-to-64 age group, many of whom will still be working. Many in-migrants are returning Scots. I agree with your interpretation.
You have spoken about the difficulties of measuring in-migration, but figure 14 shows incredible variations among local authority areas. My area—Moray—has the fourth highest number of in-migrants. How do you measure the number of in-migrants?
Again, that is a census snapshot. The census gives us a huge wealth of information. One question that it asks is where people lived a year ago—figure 14 refers to in-migrants from outwith the United Kingdom. Moray is favoured partly because of its climate and partly because of the friendliness of the people, but there are also Royal Air Force bases there.
Moray is certainly an attractive area in which to live, but it has just lost more than 1,200 jobs in the space of a month, so finding work for people there will be complicated.
One graph shows age distributions. Many young people are migrating for educational reasons—they want to become students—so there are high rates of migration in areas with universities. If we considered different age groups, we might be able to disentangle the student factor from other factors and find out where older migrants were moving to.
Is there any indication that some of the in-migrants are following their families, who are already here? Is there an indication that that plays a role in the trend?
I do not know that we have the data to answer that question. My gut feeling is that that is not a big factor.
So, many of them are young people who do not already have relatives, roots or contacts here.
In the census, people are asked about their country of birth. It is possible that some people who were born here happened to live abroad a year ago, while other people have no apparent connection. The information is based only on the questions that were asked at the time about people's current situation.
We could delve further into that, if it would be helpful.
I would certainly find it helpful, although I cannot speak for the committee as a whole. It seems that you have picked up a lot of issues on which you will have to do additional work.
We are not saying that you should not ask the question. We would much rather that the committee's inquiry was properly informed and that members were not deterred from pursuing a line of inquiry for want of data that we could make available.
In-migration is easier to measure, because someone who takes part in the census is obviously here. People are also likely to register with a doctor when they arrive, whereas there is no incentive to deregister. We can also identify, from the census, the country from which the migrants have come. As well as asking what country people were born in, we can look at where they have come from.
There is a sort of twilight area of inward migration involving seasonal workers who may not be known to the authorities. Do you have a handle on that? Looking at the figures for some local authorities, I wonder whether that is what is going on.
No, I do not think that that is what is going on, partly because, by definition, we do not have a way of catching up with illegal immigrants. The census also picks up people who are ordinarily resident at an address, but a seasonal worker will not necessarily be here in the spring, when the census is taken.
Figure 13 tells us—if I understand it correctly—that the net migration to Scotland that was recorded in the 2001 census is accounted for by the recruitment into Scotland of full-time students, people who were unemployed and people who were economically inactive, whereas the migration out of Scotland was of people in a variety of skilled, professional activities. Is that a fair assessment of that figure?
That is correct.
What ability do you have to update those data from year to year? That strikes me as fundamental to the issues with which we are wrestling.
That is a difficult question. It takes us back to the beautiful census snapshot, of which this is one. We can do a little. For instance, in issuing work permits, the Home Office has available to it data on the sort of people who are coming in. However, that deals with in-migration from outside the EU and provides only a small part of the picture. I do not think that we can produce those data except in the year before a census. Am I being too pessimistic, Celia?
No, I think that that is fair.
That profile strikes me as being fundamental to the debate. It is undeniable that we are successful in attracting what is the largest group of migrants to Scotland—students. Of course, we have world-class universities, so that is not a surprise. However, it is alarming that we cannot satisfy the need for economic opportunities for people who are in professional occupations, who are senior officials, who are in associate professional and technical occupations or who are in skilled trades—we cannot command such people. That goes to the nub of why we bother with the fresh talent initiative, which must be about economic development, if nothing else.
I very much agree. That is the pool that fresh talent fishes in. The previous figure—one page back—gives another example of the pool, in the bars for 16 to 24-year-olds and 25 to 34-year-olds.
I understand that. However, I do not understand the correlation between figures 12 and 13 in relation to students. If we are performing effectively in attracting people to Scotland as students—as figure 13 shows—we should not have the net out-migration in the 16-to-24 age bracket that figure 12 shows.
Figure 12 does not show the migration just of students, although students are part of the picture.
On that logic, the gains that we make by attracting students to Scotland because of our world-class universities in no way compensate for the population loss of other economically active people.
The same people may be involved, but they have graduated, so they are no longer full-time students.
However, the figures show snapshots of people at the same time.
Exactly. The figures show to-ings and fro-ings with the rest of the UK. On the outflow side of figure 13, we pick up from the English, Welsh and Northern Irish censuses the people who said that they were in Scotland a year ago and what they were doing. Some of those people came here to study then returned to England or, as many people do, went back over the Irish channel to Northern Ireland to take up professional or technical employment there. Those are the students whom we attracted—as shown in the left-hand bar of figure 13—but who have returned home after graduating.
On the basis of the 2001 census data, can you tell us what has happened to the indigenous Scottish population across similar criteria? Can you strip out the indigenous English, Welsh and Northern Irish population to give us an analysis of the indigenous Scottish component of figure 13?
We could do that. Do you define the indigenous Scots component as people who are Scotland born or Scotland domiciled? We would need to play around with those elements.
I think that you understand the point that I am trying to get at. If all that is involved is students coming in and going out, we could sit here until the cows come home debating the initiative's influence. As you said, we are interested in affecting the long-term patterns in the Scottish population. We must understand whether the initiative is all about numbers of students in or out—which it could be, as it is intended to encourage students who have been here for a while to stay a bit longer—or whether it is about using the policy levers to which you referred to influence people's ability to remain, to be economically active in Scotland and to contribute to our society's health. We might need to discuss offline the data that we would like, but I think that that information could help us.
The proposal is feasible.
Will retaining Scottish graduates be part of that? Do you have any information on what has happened to the retention of Scottish graduates in Scotland since devolution? I presume that that is a factor in the analysis.
Yes. That information is in figure 10. The people about whom you are talking are those in the first line, whose domicile and location of higher education institution were Scotland. I am afraid that the table does not go back before devolution, but in 1999-2000—the first year of devolution—79 per cent stayed on in Scotland for their first job and that figure had risen three years later to 89 per cent.
So there have been year-on-year increases in that figure since devolution.
Yes. That is right.
Equally, there have been year-on-year increases in the percentage of Scottish students who leave Scotland to study and stay out of Scotland.
No, there have not.
Yes, there have. Figure 10 shows that, for those who were domiciled in Scotland but whose higher education institution was elsewhere in the United Kingdom, 25 per cent had employment in Scotland in 1999-2000 and 30 per cent had employment in Scotland in 2002-03.
Yes. That line shows that the number of such people who have come back to Scotland for their first jobs has increased.
Oh, yes; I am sorry.
Am I right in thinking that the figures about students in the tables from the census relate to the academic year 1999-2000, because the census was taken before the end of the academic year 2000-01?
It is probably not an exact correlation but, broadly speaking, yes.
So, in 1999-2000, a lower percentage of students found employment in Scotland than has been the pattern since. However, we might be looking at a rogue year. That is the problem with a snapshot.
It is possible, but I do not have data on that, I am afraid, and figure 10 is based on a sample. As note 1 to that figure explains, it is based on a voluntary survey of individual students of all EU nationalities. The data are not as good as those in the census; we are a wee bit at the mercy of whoever fills in the survey.
How much weight do you attach to the reliability of that information?
The figures are not mine—I am retailing figures from the higher education world—so I do not feel that I can give an authoritative answer, but I would not have put the figures in my paper if I had not been reasonably confident about them.
I thank Mr Macniven and Ms Macintyre for coming along. Our evidence-taking session on those data has been interesting and helpful. We will correspond about the further information that we want once we see the Official Report of our discussion. As I indicated to you earlier, we might invite you back later to correct or put into context some of the things that we hear during our inquiry.
That would be a delight. Thank you very much.
I have no announcements to make other than that the committee meets again on 10 May and I look forward to seeing committee members then.
Meeting closed at 16:13.
Previous
Sift