Examination of Witnesses (Questions 20-39)|
WEDNESDAY 30 OCTOBER 2002
20. That would be perverse.
(Mr Cook) When it comes to people arriving in the
United Kingdom we allocate them according to what they tell us
on their international passenger card. When it comes to leaving
we use the information that we get from people who are leaving
to tell us where they have come from.
21. Yes, but if you are losing an additional
100,000 people per year that must have a knock-on consequence
for wherever in Britain it was that those people were living.
(Mr Cook) Yes.
22. How do you attribute that?
(Mr Cook) We attribute the flows from in and out to
the international passenger survey. We then can work out how people
are moving within the United Kingdom through changes in addresses
and National health Service registrations.
23. I am not really getting to what I want.
Let me put the question a different way. You began by saying that
the million people we appear to have lost were never there in
the first place.
(Mr Cook) Yes.
24. This question therefore sounds a bit odd
but you will understand why. Where in Britain were the million
people we did not have living?
(Mr Cook) We can give you an analysis of each of the
376 local authorities and tell you what was their population estimate
as at June 2001 and we can tell you what was the census base estimate
for that period. We can give you an analysis of each of the local
authorities if you wish. I can send that to you.
25. I am getting more and more confused here.
One the one hand you seem to be suggesting they did not exist
in the first place, and then you went on to say that there were
problems with measuring people leaving, and you went on to talk
about European countries. It seemed to me from your answer that
you could have been suggesting that there was an inter-European
balance going on with some people coming here and some of our
people going there, but because we were measuring people coming
here from wherever in Europe we knew about them but we do not
know about ours leaving, and that would be more significant because
they are much more likely to go through ordinary channels in Customs
and not be recognised by your survey. Is that what you are telling
us and, if that is what you are telling us, can both situations
(Mr Cook) No. What I was saying was that when it comes
to measuring people coming into the country we get an estimate
from the international passenger survey. There is other information
that we can get from our labour force survey, from registrations
into administrative services such as the NHS, that gives us confirmatory
information of those flows. The only confirmation we have of the
flows of people going out comes from looking at other countries'
migration statistics. One of the few that we have that can give
us good information, for example, is Australia where we discovered
that we had 50,000 more people from the United Kingdom go to Australia
than we had assumed in our own analysis from the international
26. But that suggests that the million did exist
at some point. If it is true that people have gone and we have
lost them and we did not know that they had left, then they must
have existed at some point or the 1991 census was wrong and we
over-estimated. There must be some balance between the two and
if you are suggesting that we have this problem particularly in
the European nation states what is your suggestion that we can
use within the whole of Europe to get this balance right, because
if it is affecting us it is affecting them?
(Mr Cook) Can I answer your last question first? We
have not completed the work that we are doing. We have got a small
team of people looking at the whole range of options that we have.
If we were to look to a solution similar to that used for foreign
trade then it would require work within Europe to get some agreement
on how we proceeded. A good many of the European countries have
the same problem as we do and therefore getting an agreement is
not starting from a completely fresh position because the issue
has been discussed quite a few times in the European statistical
27. I hope you are going to tell me what progress
you are making.
(Mr Cook) We have really only made significant progress
in the last four months in that our time beforehand has been much
more about investigating options. We have now got some very able
people working on the problem full time and we expect to have
a report in the new year.
28. Which new year are you talking about?
(Mr Cook) Perhaps I can get John Pullinger to be more
(Mr Pullinger) The census gives a new benchmark of
who is here so it is because we found in the census that growth
in the population was a million lower than we had thought that
we now need to do this investigation. We knew there was a problem
with migration statistics anyway for the reasons that Mr Cook
has identified, and we had already commissioned a national statistics
quality review of the sources of information we could use to monitor
migration and how we might improve them. It is that which we will
report on in the new year and that will give us a much better
fix, we believe, on how we can capture these flows both into the
country and out of the country. The key problem that we have identified
with the census is that we are confident that there is a bias
in our current method for estimating the people who are leaving
here. Those people were never here in the sense that the population
was never a million higher, so it was never 59.9 million. What
we are saying now is that growth in the population was 0.1 per
cent lower each year over the last ten or 15 years.
29. You are not going to count people who were
(Mr Pullinger) We are not going to count people who
were never here.
30. How do you know that this is the source
of the problem though, the immigration/emigration? How do you
know that it is not just that you under-counted by two million
or that two million escaped the census instead of one million?
(Mr Cook) When we look at the certainty that we can
have of the various sources of information that we have got, births,
deaths and the census, and the effectiveness of the census coverage
surveywe have in the United Kingdom a very large census
coverage survey which gives us the ability to measure for each
of the 376 local authorities separately and independently of the
census, an estimate within five per cent of what the population
waswe can be quite sure that the majority of the problem
lies within the migration statistics that we have got.
31. Jim Cousins asked you about how the additional
migration which you reckon to have discovered is attributed to
the local authorities basically. You very kindly said you could
provide detailed figures. As you probably are aware, there is
considerable debate in Parliament at the moment about the standard
spending assessments according to, among other things, local population
and I know my borough council is in dispute with the authorities
on how many people live there. How confident are you that the
attribution of the migration statistics to this correction has
been taken account of in the SSA debate? Are you aware whether
it has or not?
(Mr Pullinger) Obviously the precise calculation of
the standard spending assessment is a matter for the Deputy Prime
Minister's office but as far as the statistics are concerned the
numbers that we are now putting forward are the census numbers
which provide the new benchmark of the numbers of people who were
here. They do not rely on correctness or incorrectness of migration
32. That is the statistics you are now putting
forward but the current debate is based on the figures from last
year and those will not have the correction yet, will they?
(Mr Pullinger) The current debate is about the difference
between the figures that were produced, which are the mid-year
estimates for the year 2000, and what we are now finding from
the census. Inevitably in some local authorities the numbers are
lower, in some they are higher.
33. The current SSA debate, you are saying,
or the current debate on the census? I am sorry; we may be talking
of two different things.
(Mr Pullinger) The calculation of standard spending
assessments for next year is being informed by the census information
which has now been released. That was one of the things that was
driving our timetable for the release of the census information
so that it would be available in time for this debate.
34. How do you feel ideally that the definition
of the retail price index should be set? Who should set it? What
would command most public confidence?
(Mr Cook) In most countries it is set by the National
Statistician, often informed by a committee, in some cases set
up by the minister. In the case of the small country where I used
to be it was the responsibility of the Government Statistician
but the Government Statistician was advised by a committee set
up and appointed by the Minister for Statistics and the advice
that that committee gave was presented to Parliament and therefore
the Government Statistician had a very clear accountability of
explaining if he were not responding according to that committee's
35. How does that differ from what we do here?
(Mr Cook) In practice it has not differed at all to
the extent that in 1993-1994 there was a committee chaired by
the then Director of the Central Statistical Office, Bill McLennan.
That committee was appointed by the Chancellor and it contained
about 20 people; I have forgotten the exact number but it contained
a large number of people. It presented two alternative views as
to what the RPI in the United Kingdom should look like. The Director
of the Central Statistical Office provided a way ahead for that
and that was accepted by the Chancellor.
36. In commenting on the code of practice the
Statistics Commission called for the special arrangements for
the Chancellor's involvement in the RPI to be spelt out and for
a public register to be maintained of those occasions on which
ministers are consulted about proposals which impinge on government
policy. Have you been asked to comment on these proposals?
(Mr Cook) No.
37. May I invite you to comment now?
(Mr Cook) I would expect, if I were given a direction
on the RPI, that it would be naturally a matter in the public
domain. Certainly it would be very unlikely that the National
Statistician would be expected to act on a direction that was
not in the public domain. It would certainly give me incredible
difficulty if that were the case.
38. So what you are telling usI do not
want to put words in your mouthis that there have been
no secret communications from the Chancellor or Treasury of their
views on the RPI?
(Mr Cook) There have been no secret discussions on
the RPI of the United Kingdom that involved me or my Office. Certainly
during the time I have been here there have been odd discussions,
whether it is with the Bank of England or Treasury economists,
and there is the odd exchange of views, but it is certainly no
different than I would have experienced in my previous post or
would happen in any other country.
39. Do you feel that the current definition
of the RPI is reasonable up to date or is it time for a review?
(Mr Cook) I think the Bank of England Act has altered
the nature of the way in which we would treat individual major
statistics. For example, in the national accounts there is an
expectation that quarterly GDP should be able to be measured with
greater reliability than we might have expected before that because
of the very significance of the decisions, the indexation almost
of interest rates to GDP and RPI changes. I think there is a greater
likelihood that a future RPI would be focused much more on the
measurement of inflation rather than the mix with an outlays approach
to inflation measurement as it exists in the United Kingdom. That
is a trend which is happening in pretty much every other country
in the world. If we were not to follow that trend (which we might
not) we would obviously have some good reason for not doing so.