Examination of Witnesses (Questions 1-19)|
WEDNESDAY 30 OCTOBER 2002
1. Mr Cook, welcome back to the Committee. Perhaps
you could identify yourself and your colleagues for the benefit
of the shorthand writer.
(Mr Cook) I am Len Cook, Registrar General of England
and Wales; John Pullinger, who is the Executive Director of Social
Statistics in the Office for National Statistics, and Robin Lynch
who is the Head of our National Accounts Division.
2. The Treasury told the Committee during our
inquiry into general parliamentary accountability that departments'
spring reports would give a full picture of each department's
organisation, aims and objectives, performance and use of resources.
Are you satisfied that your spring 2002 departmental report can
meet those criteria?
(Mr Cook) I think it provides quite a
comprehensive view much more of where we are going and I would
expect it to have more detail in future years. We have put a lot
of effort into basically changing the strategy of the Office and
improving very significantly the Government's documents, for example,
the National Statistics Plan, the second one this year, the business
plan for the Office, and I suspect with the new service delivery
agreement that we are negotiating with the spending review for
2002 there will be a much more comprehensive framework than I
believe we have at the moment. This reflects the quality of the
SAR 2000 service delivery agreement which I would not think embraces
as much as I want the 2002 one to embrace. It is also important,
just to conclude, to recognise that a huge part of our performance
analysis of course comes through delivering on the basic statistics
programme. We publish well over 500 reports a year and the actual
delivery of those on time is a hugely significant share of the
accountability we have to deliver the performance of the Office.
3. Let us just stick with this report if we
can for the moment, not the 500. Could you turn to page 33 where
you list the various performance indicators in annex C? There
you simply put alongside most of the performance indicators the
comment "Monitored and reported quarterly". What does
that mean? It does not tell us what happened.
(Mr Cook) No. This could have more detail and I have
got much more detail, for example, in the background material
that we used to prepare this.
4. What is the point of listing performance
indicators with a commentary if it does not tell us what happened?
What is the point of the commentary?
(Mr Cook) When you come to our annual report, of course,
5. Let us stick with this report. What is the
point of the commentary if it does not tell us what happened?
(Mr Cook) The body of the report has provided quite
a comprehensive statement within it of the huge number of things
that are going on in the Office but I certainly accept the criticism.
6. In future years we would like to see some
outturn properly reported here. Turning to your annual report,
one of the key activities of course was the 2001 census and you
devote some four pages to this. You do not refer to our report
on the census. Why is that? This is your report, Mr Cook, pages
7-11. It does not make any reference to our own report or recommendations.
Why is that?
(Mr Cook) I was not aware that we had not, Mr Chairman.
7. You are aware presumably of the Government's
response to it.
(Mr Cook) Very much so.
8. You do not refer to that either. Are you
not taking this Committee's report seriously?
(Mr Cook) On the contrary, we do take it very seriously
and we have regarded our response to you as a very serious piece
of the direction of the work of the Office.
9. But you do not refer to it at all.
(Mr Cook) The contents of the report have played a
very significant role in the business plan. I believe that we
have properly taken account of your report much more in the prospective
work of the Office than the reporting of the year 2001-2002. What
I can promise is that the annual report for 2002-2003 will provide
a clear distillation of what we have done in this year for that,
which is in fact the year where most of the action on your recommendations
will have taken place.
10. Turning to the census results itself, your
preliminary results show that the population was about a million
less than you had previously forecast. Was the methodology wrong
with the interim census? What went wrong here? Were you surprised
about this? How did we lose a million people?
(Mr Cook) We never had them in the first place because
the population estimates that we produce in the intervening years,
in the 10 years between the census, are based first on the previous
census, in which we now believe we made an error in assuming was
less correct than it was. We believe that we over-adjusted for
missing people in the 1991 census, and I can come on to that,
for the same reason that we believe we over-counted in the estimate
for 2001, which is simply that whereas we can measure births and
deaths in the United Kingdom with a high level of accuracy, because
we believe we have a highly effective birth recording system and
a highly effective death recording system, we can measure migrants
into the UK with a level of precision that we believe we can improve
somewhat on but which we believe is of a reasonable degree of
accuracy but we have real difficulty in measuring the outflow
from the United Kingdom. We believe that we are able to use a
lot of administrative information to confirm the estimates that
we are able to make of people coming into the United Kingdom from
administrative sources and surveys such as our labour force survey.
When people have left the United Kingdom we do not have information
that can confirm whether they have gone or not and we have no
way of validating estimates of the outflow of people from the
United Kingdom. We have the disadvantage of being on the one hand
an island state but we do not have the border documentation that
smaller island states have and we do not have the registers that
landlocked countries have to measure population flow in the United
Kingdom, so it is very difficult to measure, with the level of
precision that we need for these estimates, the migration flows
in the UK. We estimated 250,000 departures a year between 1991
and 2001 and we now believe that there were just on 300,000 departures
11. You can count them all in but you cannot
count them all out; is that right?
(Mr Cook) We do not have the ability to do so. We
do not have the methods for doing that and as migration has been
more significant our ability to measure population change is less
than it was, say, 20 years ago.
12. Getting the forecast wrong by a million
is not very good, is it?
(Mr Cook) It is 0.1 per cent a year.
13. It is a million people you were wrong by.
(Mr Cook) When you accumulate 0.1 per cent a year
for ten years it comes to just on a million, yes.
14. Why do you think migration has been significantly
(Mr Cook) The process by which we measure migration
is a survey of international passenger arrivals and departures
at Heathrow and other airports and it is a sample which is designed
and has been traditionally designed to measure with sufficient
reliability the balance of payments flows of United Kingdom citizens
and of people coming into the United Kingdom. We also use it for
visitor arrivals and it is extending its effectiveness to use
it as we do for migration flows. We currently have an investigation
under way to look at alternatives.
15. Are you going to get this better for the
next census? Are you going to remedy these deficiencies in terms
(Mr Cook) It is probably one of the most important
things that we need to do in official statistics at the moment.
It is a significant activity in my office at the moment, to look
at alternative means. We had started to do that about a year ago
but the significance of it obviously has changed.
16. How are you going to get better if we do
not record people coming into the country and whether they leave
or not? Short of doing the American system where, when you come
in, they take your name and they tie it up when you go back out
so that they know who has come and who has gone, how are you going
to do it? I do not see how it is possible to do it so the inaccuracies
(Mr Cook) It is most likely that we will have to use
some modelling methods and a series of approaches
17. So you will guess?
(Mr Cook)some of which mean that we look at
other countries. For example, the problem that we face is not
much different than exists in other European countries. One of
the things that we have been discussing in the European Union
Statistical Offices is, is there a common solution to this, just
as you will be aware there is with foreign trade, where each of
the countries in the European Union follows the Intrastat trade
process where we can measure from either country the trade flows
between countries, and so with an attempt to find a European-wide
solution one of the things we know is that we can measure inflows
into countries better than we can measure outflows. If we were
able to measure the inflows of United Kingdom citizens into other
countries with a degree of accuracy we could use that to compute
part of the movement, for example, on outflows. That is not saying
that that is the solution but there is a set of second order indirect
approaches to measuring that we would look at using.
18. Can I ask how these problems with the international
passenger survey statistics affect our internal population figures?
For example, if we have been losing 100,000 more people than we
thought, and by that I do not mean that we have been losing 100,000
in that sense but they have been choosing to do whatever it is
they want to do with their lives, where have they been coming
from inside Britain? What knock-on effect does that have for population
figures for various parts of Britain?
(Mr Cook) We do have more means of measuring internal
19. I am not talking about internal migration
flows. I am talking about the impact on urban regional populations
of losing an additional 100,000 people a year. You are not attributing
that loss, for example, of 100,000 a year to wherever in Britain
Heathrow Airport happens to be recorded?
(Mr Cook) No.