Examination of Witnesses (Questions 266
- 279)
TUESDAY 20 NOVEMBER 2007
Professor David Coleman
Q266 Chairman:
Good afternoon. Thank you very much for coming along to talk to
us. You are very welcome. You have submitted some written evidence
to us which we appreciate. Do you want to say anything to start
off with or are you happy to go straight into the questions?
Professor Coleman: Let us go straight in, if
you would.
Q267 Chairman:
How likely is it, in your view, that the Government's current
net immigration projections will be accurate? What factors could
cause the level of net immigration to diverge significantly from
those projections?
Professor Coleman: All population projections
are always wrong. What matters, of course, is how wrong they are.
With respect to migration, it is the most volatile and fragile
element of any projection and therefore the most likely to be
incorrect. In a way, that does not matter because one of the merits
of population projections is that they show the necessary logical
consequences of a particular set of assumptions projected into
the future. If those assumptions hold, then the projection is
what absolutely must happen. So it gives you information, a warning,
whatever you like to call it, of what will happen if things continue
as they are trending at the present time. Having said that, the
position recently is that there have been, I think, seven successive
upward revisions of international migration by the Government
Actuary's Department over the last decade or so. There have not
been any downward revisions. Migration can of course, and does,
decline as well as increase, to misuse the financial services
health warning. This must be kept in mind. But the general trend
over some time now has been for an increase, although perhaps
at a slightly decreasing rate of increase. My expectation would
be that unless something happens which is not yet expected, migration
will continue to go up to a moderating extent and maybe even higher
than 190,000. I do not think this will continue for long and certainly
by 2050 I would be astonished if it was still 190,000. I did not
answer your question about the factors that might alter it. Migration
is very tiresome and complicated. First of all, because it is
not one process but several. There are so many different streams
of migrants entering for different purposes and with different
consequences, as I am sure other people have made clear. Also,
it is influenced by not only what goes on in this country to some
extent, its economy and its migration policies, but also what
goes on in a dozen or two dozen other countries in terms of their
political change and their economies as well. That makes it uncommonly
difficult to forecast. I think it would be helpful if we could,
at least in theory, split up migration into its different streams.
One important one which is not considered very much in the economic
context, because it is not an economic migration, is of course
the migration of spouses for family reunion and, increasingly
importantly, for family formation. The latter especially is reasonably
strongly correlated with the growth of the Asian and ethnic minority
populations in the UK and is becoming important both here, in
Holland, Germany and other countries which have such populations.
Until that culture changes or the policy changes, one might expect
that element of migration to continue in an upward direction.
Economic migration, though, is much more related to economic trends
and also to government policy affecting those economic trends
as to how far the needs of employers are going to be met or not
met. Those are just two of several components. A third one is
education. At the moment, universities can only stave off bankruptcy
by recruiting as many as possible of non-EU graduate students
because they are the only ones who pay the economic costs of what
universities actually provide. This has an unintended effect upon
migration naturally. Were university finances to be changed, there
would possibly be a downward trend in that recruitment and it
would become less frantic and less essential to salvation from
the economic point of view of the universities.
Q268 Lord Layard:
You have made some remarks about the Office for National Statistics,
but are you thinking that it is now doing enough to improve its
statistics? I am talking about its actual measurement of migration
stocks and flows. If not, what else should we be doing? How accurate
can we ever hope to be?
Professor Coleman: The ONS is doing a highly
professional job to the best of its ability with what seem to
be rather limited resources. The fundamental problem that it has
is that it is trying to measure migration with instruments which
are really quite unsuited for the purpose. The international passenger
survey, as you know, was invented back in the 1960s as an instrument
for the Board of Trade for balance of trade, tourism and things
of that kind. It is only incidentally used to measure migration.
It does so by using a small fraction of its interviews, as I am
sure you know very well, to take a voluntary sample of those coming
in, about 1,800 a year, and those going out, about 800 a year
or thereabouts. Those are then grossed up to make migration assumptions
from different categories of people by age, marital status, country
of origin, nationality and all the rest of it. That is intrinsically
unsatisfactory because it is voluntary and it is quite a small
sample survey. They are also trying to use the labour force survey,
which is also a voluntary survey, although a much bigger one,
60,000 a year, which has a question on migration over the last
12 months: whether someone was living abroad 12 months ago. The
hope would be that all these things can reinforce and correct
each other. The trouble is, as they are all defective, it is difficult
to know which one to trust. The 2001 Census for example was deemed
to be infallible to begin with, as you know, and turned out not
to be infallible. As a consequence of the belief in its infallibility,
the migration estimates were toned down quite a lot because it
was assumed that the only thing that could be wrong were the emigration
estimates of white males. You may recall the particular excitement
in the papers, the supposedly 800,000 white Britons cavorting
on Bondi Beach unknown to the authorities in Australia and in
the UK. This deficit has been squeezed down to quite a small number.
We do not really have a complete set of data which tell us how
these things really add up. There is one other source, the Home
Office source, of data on admissions, and admissions for settlement
in particular. The trouble is that in all except a few categories
these are not compatible with the International Passenger Survey
data, not least because they are only gross inflow data and do
not have any kind of departure element to them. Matching those
is difficult. It seems to me and some of my colleagues that the
instruments which the ONS has at its disposal are really at the
end of their useful life. The ONS is doing its best in a number
of ways to improve them. It has increased the sample size on emigration,
as I am sure you know. It has increased the cover of entry in
the less used ports of entry to try and make the coverage more
complete. It has been involved in the long term e-borders initiative.
That will not come on stream until 2013 and the extent to which
that will record information of a useful kind in both directions
is still quite unclear. It is almost impossible, for me anyway,
to find anything in the public domain about the details of the
e-borders initiative. That is partly because it is still being
put out to tender and there is still much which is unclear. It
seems to me that the ideal way forward, if one could stump up
the cost and if the political objections could be overcome in
some suitable way, would be to abandon these approaches and to
have instead a population register of the kind which a number
of foreign countries have, which would at the same time probably
remove the need for a census. It would mean that the e-borders
initiative, which would log everyone coming in and everyone going
out, would be linked to a population register which would show
where they were, how long they were staying and when they left.
That would also, of course, give us a reliable total, were it
to be done properly, of the total number of people in the country
and their condition and where they are living and so on. The Swedes
do this and a number of other countries do this without any noticeable
erosion of their civil liberties, and I think we should do as
well. It was toyed with by ONS for a while but I think it was
dropped.
Q269 Lord Best:
Do you think this debate as to whether or not we should be looking
in some detail at foreign-born persons as opposed to foreign nationals
is a productive debate for us to worry about? If it is, do we
have accurate figures of these two groups in terms of arrival
since 1997, foreign-born versus the foreign nationals? We did
actually ask your predecessors in that position the same question
and I was not entirely clear of the answer. From your perspective,
is this an important ingredient for us to get our heads around?
Professor Coleman: It is particularly important
when so much emphasis has been placed in recent Government policy
on the specific advantages of immigrant movement into the UK from
the point of view of its effects on the economy, so whether people
were born abroad or not, usually whether they are of foreign origin
or not in addition to that, is clearly very important in the context
of the present debate. As regards the accuracy of the data, the
data on the foreign-born coming into the country as measured by
the International Passenger Survey is just as accurate as any
other element of data from the International Passenger Survey.
That is to say that it is strongly constrained by the sampling
error which such a small sample necessarily imposes, and it means
that you cannot easily split up that data into different countries
of origin unless those countries send very large streams. You
have probably noticed that in the tabulations which the ONS produces,
the country of origin categories, whether they are birthplace
or nationality or where they physically come from before immigration,
are really quite coarse India, Sri Lanka and Bangladesh are all
lumped together; Pakistan[12]
used to be separate, I am not sure it still is separate; the West
Indies used to be separate but has now been abolished as a separate
category; "Other foreign" covers 150,000 or 200,000
people. With one year's data it is almost impossible to divide
those up in any kind of statistically respectable fashion; to
do that you have to amalgamate several years data which may lose
interesting information. Of course, if you could really substantially
increase the IPS sample or have a complete count of those coming
in which collected such data, you would to be on a very much firmer
basis: knowing exactly where people are coming from and what the
immigrant population is. Elsewhere, if I may just develop that
for a second, in the national statistics system as a whole we
tend to lack data on birth places of individuals. We have perhaps
put too much emphasis on collecting data by ethnic minority status,
perhaps for some very understandable reasons, and also perhaps
because the notion of migrant or immigrant has become slightly
pejorative. This means that when it comes to answering questions
of the kind which are raised by the Government's recent policy,
we are rather deficient in data on where people were born, not
so much from the IPS but from other kinds of statistical sources
and from samples of various kinds. Above all, we are deficient
on nationality; nationality has not been asked in the census since,
I think, 1971[13]
and it is a grievous error not to have kept it in. It is going
in, I gather, in 2011. That is quite an important dimension, especially
on migration.
Q270 Lord Macdonald of Tradeston:
Your written evidence concludes that "overall the effects
of the new [immigration] policy are counterproductive and damaging
to the national interest". How do you define the "national
interest"? How should the impacts on employers be balanced
against those on workers, and what weight should be given in policy
design to the interests of the migrants and their countries of
origin?
Professor Coleman: That could detain me for
a long time. To turn first to your first question about defining
the national interest, I suppose the answer is that I do that
with some difficulty But one can do it on a multi-level basis.
It perhaps could be regarded as the sum of the effects on the
population as a whole from the point of view of their economic
development, their wages, their incomes and the security of those
things at the economic level; also, at the individual level on
the prospects which they have for housing their families, for
advancing the interests of their children and advancing their
own ambitions. At the community level one might think of the extent
to which a population is experiencing continued cohesion and a
sense of identity which is, I would have thought, good in a general
way for morale, or on the other hand is finding itself rather
divided up in a way which may erode civil society, which may erode
support for the welfare state, which may make people feel that
they are belonging to a rather inward-looking community rather
than to a broader society based on a common citizenship. At the
national level, what effect it is having upon the macroeconomic
characteristics of the country, on its security from the point
of view of internal threats and also external ones, and the extent
to which it can conduct an independent foreign policy—in
so far as anyone can these days—which is based upon the
interests of the country as a whole as opposed to being driven
or altered in some way by powerful minority groups inside, which
may have a different set of priorities from that of the national
average. That is the first question. The second one was?
Q271 Lord Macdonald of Tradeston:
How should the impacts on employers be balanced against those
of workers?
Professor Coleman: I hope I can extend that
slightly because it is not just workers versus employers in this
case, in a slightly old-fashioned class war kind of way, it is
more the effect of employers on everybody else in the country.
When employers bring in immigrant labour, then that of course
is a benefit to them, it is a benefit to the immigrants, it is
a benefit to those who consume the services and buy the goods
the immigrants provide. That is not in dispute. It does, of course,
have much wider consequences for the wider society, because those
immigrants, quite naturally, will want to have families, they
will want their children educated, they may or may not want their
children to be part of the country in which they have settled
or they may have some reservations about that, as we know is the
case with some groups. The migration which the employers promote
will increase population size, which has all sorts of consequences
for housing, for land, for the countryside and the environment,
so it affects the whole country. It is much more, therefore, than
the interests of workers versus employers, it is the employers
versus the interests of a very wide spectrum of everyone living
in the country who is affected to a greater or lesser degree by
what the employers do and what happens to their workers once the
employers have finished with them. The third one?
Q272 Lord Macdonald of Tradeston:
What weight should be given in policy design to the interests
of immigrants and their countries of origin?
Professor Coleman: The interests of immigrants
themselves are something which of course have to be, and are,
protected by law and by any normal standards of decency. The interests
of the countries from which they come are a different matter.
A basic principle of democracy, which I suppose might seem a rather
selfish one, but nevertheless I think is true and is something
on which national democracies rest, is that the interests of the
local population, of the electorate, are paramount and their consent
would be required if their interests were to be set aside for
some possibly good moral reason for the advantage of anybody else.
It certainly is not possible to solve the problems of countries
overseas by promoting large-scale migration into countries like
the UK; apart from anything else, the scale is out of the question.
The populations of the countries from which most poor migrants
come; the countries of origin of those migrants who are culturally
different from Britain are hugely greater than that of the UK
and for almost all practical purposes infinite. There is no way
in which, say, population pressure, could be in any way relieved
by promoting migration to Britain. We do do so, of course, in
a more specific way by agreeing to international asylum conventions,
which is one way in which we do benefit people who are in trouble
and who show they are in trouble in foreign countries. That is
rather a special category.
Q273 Lord Kingsdown:
How do the costs and benefits of immigration differ from those
of population growth without immigration? Is there an optimal
population for the UK?
Professor Coleman: First of all, it seems to
me—if it is not too heretical to say so—that population
growth is not particularly desirable and may in the long run not
even be possible. We are running out of planet in rather a big
way and countries like Britain, as we all know, are consuming
ten or twenty times as much resources per capita as those in the
rest of the world and are producing a correspondingly large amount
of contaminants and greenhouse gases and all the rest. From the
point of view of world interest, if one can conjure up such a
concept, it would be very helpful if the developed countries would
stop increasing their population size, as well as improve their
performance in terms of more efficient use of resources and a
moderation of greenhouse gas emission and all that. There is no
evidence that I am aware of that a growing population size or
even a large population size, in the developed world countries,
has any positive effect upon per capita income; when you do the
statistics the line is flat. Small countries are just as rich,
some indeed are clearly richer than large ones, and within the
western world countries which are growing somewhat are no richer
than those which are declining somewhat. This may not last into
the future of course, it depends how long the decline goes. We
ought to be thinking about population stabilisation and reconciling
ourselves to some small degree of decline in the future, so long
as that decline is very slow and so long as that decline comes
to an end, which of course is not something which can be guaranteed,
although there are some reasons why it might indeed do so. Population
decline is not on offer for the UK at the moment on present projections,
as you will well know. A stable population, can be maintained
if the birth rate is about the level of replacement. Then the
support ratio, the ratio of people of working age to the people
of pensionable age, is more or less as good as you can get under
normal circumstances and certainly would be better than the future
level expected at the moment. With a replacement level of fertility
the potential support ratio would more or less stabilise, subject
to the death rate, at about 2.7, 2.8, something of that kind,
compared with the 4.2 which we have now. That is rather better
than would be the case if the birth rate were lower. If you try
and preserve the support ratio by migration, then you can do so.
But you can only do so at the cost of really quite considerable
population growth. The gearing between immigration and the support
ratio is not very favourable, it is rather inefficient. It would
mean putting up with quite a lot of population growth, which is
not a good thing at the moment. As far as the optimum is concerned,
there have been lots of attempts to find the optimum population.
I do not see that any of them are entirely convincing. Like the
Loch Ness monster, I would like to believe in it but I do not
yet. I do feel that there is no case for a further increase in
population and that a small decline might be better, but that
is not to say that I pretend to know what any kind of optimum
might be.
Q274 Lord Kingsdown:
If we had much more immigration, would that be making a contribution
to this balance at all, or is it something that one wants to avoid
as a matter of national interest?
Professor Coleman: The balance of what, excuse
me?
Q275 Lord Kingsdown:
At any given moment we have so many hundred thousand immigrants
coming in and so many hundred thousand emigrants going out.
Professor Coleman: Yes.
Q276 Lord Kingsdown:
Supposing the immigrants stopped coming in but the emigrants continued
to go out, would that be healthy for the economy?
Professor Coleman: It would entirely depend
upon the numbers. It is impossible to imagine that migration in,
or emigration out, would ever stop. In recent years, in the 1980s,
both were quite modest and the net level of migration was sometimes
negative, sometimes positive, but it was quite close to zero,
certainly in terms of the total population and in terms of the
main drivers of population change, which are the numbers of birth
and deaths. That has now changed quite a lot because immigration
has increased so much. If there were to be no people coming into
the country and very large numbers leaving, as in the case of
Bulgaria for example, and one or two of the other Eastern European
countries, that would normally be symptomatic of something seriously
wrong with the country, with its economy, its politics, its standard
of life, its health service and so on. It would be an indicator
of malaise, as in the case of Bulgaria it clearly is. I would
have thought that things would have to be very bad indeed for
migration into Britain to stop or the pressure for it to stop,
and emigration to be very high.
Q277 Lord Lamont of Lerwick:
You have said that the Government has exaggerated the economic
benefits of immigration. Could you say why and give some examples?
Professor Coleman: There are two ways in which
it has been exaggerated. One is in the evidence which has been
presented of a statistical kind, which is, I think, responsible
for the exaggeration in two ways. First of all, because it concentrates
upon gross domestic product without reference to population growth.
GDP is not a terribly important thing to emphasise because if
you increase the population of a country by any means, unless
none of those people who comprise the increment are actually working,
then the gross domestic product goes up. That seems to me a rather
irrelevant conclusion. The small numbers of per cent by which
the gross domestic product is clearly increasing as a result of
immigration, as it must do, have no bearing really upon the average
level of living of the people in the country. It is the average
standard of living of people in the country which matters, both
from the point of view of logic and also from the point of view
of what governments always say: they want to improve the standards
of living. Immigration does not improve overall standards of living
for each individual by anything more than a tiny degree, and some
of the evidence suggests that it may be negative, at least under
certain circumstances. That is one exaggeration, this persistent
emphasis upon gross domestic product and not per capita gross
domestic product where the claims do affect a very small number
indeed. The second is when making this calculation the assumption
has been that all migration is in some way economic migration.
But until the A8 influx, economic migration, that is to say people
coming ostensibly and specifically for the purposes of work, has
always been the minor part of migration. In some European countries—in
fact most European countries—over the last 30 years there
has been quite a small minority of "economic" migrants.
Most immigrants to Europe in the last few decades have been new
spouses, old spouses, child dependants, older dependants, students,
people coming for no particular known purpose, asylum seekers,
people entering who have been given an amnesty and so forth. The
A8 inflows have changed that in Britain recently, but that may
not last. The calculations and assumptions which the Treasury
keeps making assume that the workforce participation rate and
levels of income of migrants are the same as the rest of the population,
and it just is not true. There is variation, some are much higher,
some are lower, but unemployment is higher and the participation
rate is somewhat lower overall than the rest of the population,
as is the case in almost every other European country, and this
tends to be ignored. The second way in which things are exaggerated,
it seems to me, is just in the use of language. In Mr Blair's
speech to the CBI—I think it was in April 2003—and
there were various speeches by Mr Blunkett and Mrs Roche and others,
we heard that immigration is essential to the success of the British
economy. It would be disastrous if this migration had not happened.
Furthermore, attempts were made to persuade us that this has been
always the case, notably in Mr Blair's speech. That is just ahistorical;
it is not the case and it can be clearly shown not to be the case,
not least because migration into Britain from overseas had been
really quite modest until the 1960s. That rhetoric has been unfortunate,
to say the least.
Q278 Lord Lamont of Lerwick:
I do not know if you were here when Ms Riley from NIESR argued
that actually the contribution of immigration had been to boost
GDP per capita; did you hear that?
Professor Coleman: I did, yes, and I acknowledge
that there are some calculations which show this to be the case.
But there are others which show that it is not the case, that
either it is too trivial to be made a fuss of or actually is negative.
One of the sadnesses about the Government official response to
this Committee, which she also commented on, was that almost all
that evidence was, it seemed to me, ruthlessly ignored. There
are a number of papers which, for example, Professor Rowthorn
mentioned in the course of his submissions to this Committee which
list quite clearly a number of studies both in Britain but particularly
abroad, both on the Continent and in the US, where the per capita
effect is negative. It is normally small, most of these effects
are small when computed on GDP or on a fiscal basis, a small positive
or a small negative; there are hardly any studies which have shown
more than 1% plus or 1% minus on that calculation.
Q279 Lord Lamont of Lerwick:
I do not know if you heard her point about the contribution of
capital towards GDP growth and suggesting that some of these estimates
that showed negative growth were ignoring the actual contribution
of capital towards the increase in GDP.
Professor Coleman: I did not hear that, either
because I had not arrived or because I did not quite catch it.
I am surprised to hear that because one of the criticisms sometimes
raised about the effects of migration is that they bring labour
but do not bring capital, except in the very few cases of entrepreneur
migration, and therefore that capital is being diluted. Part of
the immigration shock which I heard her talk about is deemed to
be the effect of a large influx of labour at one particular point
in time where capital formation then took some time to catch up—if
I can be ungrammatical—and for a time therefore the effect
is negative although in due course, perhaps in five years or so,
capital formation should have caught up and the labour can be
more effectively employed. There was something quite specific
to that effect in a report by the OECD in the last year or two,
if I recall correctly.
12 Note by witness: Now, all lumped together
as Indian sub-Continent! Back
13
Note by witness: Last asked in the 1961 census. Back
|
