Examination of Witnesses (Questions 120-139)|
WEDNESDAY 30 OCTOBER 2002
120. The £105 billion figure, what was
the base figure?
(Mr Cook) Sorry?
121. The revision was £105 billion, what
was the total it was a revision of?
(Mr Cook) It was £760 billion.
122. Of the order of 15 per cent.
(Mr Cook) It was 100 billion on 760.
123. So it was of the order of 15 per cent of
(Mr Cook) Yes.
124. What sort of decisions might that have
(Mr Cook) It did not influence the national accounts
because our national accountants took the data and validated it
against other sources which were not part of the survey. There
has been no change to national accounts. What it would have influenced,
I have no doubt, is anyone looking at the performance of the pensions
125. They would have had too rosy a picture?
(Mr Cook) They would have had a very different picture
from what we presented, yes.
126. There was another change. The Times
on 22 October reported that following an investigation by the
ONS the estimated value of pension contributions for last year
has been reduced from £86.4 billion to £43.7 billion,
almost half. You had an investigation I believe which showed the
original Government figure for new investment included billions
of pounds of pension transfers. Can you explain this to us?
(Mr Cook) That error was a result of labelling which
was not precise in a statistical table which had been prepared
accurately itself, and the statistics were accurate, but the labelling
did not make it clear that it was dealing with gross flows into
pension schemes and out of them rather than netting off the flows
which go from one pension scheme to another.
Chairman: Double counting.
127. Which should it have been? Should it have
been the net flow?
(Mr Cook) No, it was reporting on the information
available from those inquiries so it was reporting on the gross
flows. Net flows were not estimated in that survey. It should
have been more effectively labelled. It had been labelled for
well over 20 years with a footnote at the bottom of the table
of what it was actually rather than being very clear in the actual
128. I am getting confused. What was the table
meant to be: two gross flows?
(Mr Cook) Two gross flows, yes.
129. Is that what these figures I quoted were?
(Mr Cook) The £85 million was a gross flow, yes.
130. The £84 million was gross?
(Mr Cook) Yes. In fact, the net flows are obtained
from other information that we gather so when we produce again
the national accounts side of the office, another source of information
is obtained which estimates the netting out that goes on in the
insurance industry. This particular survey did not collect that
131. This figure of £86.4 billion was a
gross flow and that gross flow then reduced to £43.7 billion,
that was the nature of the gross flow after revisions?
(Mr Cook) When you took the flows between pension
funds, yes. When you derive net flows from the gross flows figure
by using other information about the transfers between pension
schemes, the net inflows into pension schemes becomes £46
132. How do you end up with an error of this
magnitude? Does anybody have a reality check on this? Do people
calculate figures to three decimal places and find out by a factor
(Mr Cook) What will rectify this failure more significantly
is the production, which we will produce now, of a table which
takes this information and works right through to national accounts
and brings in each of the alternative sources in pensions so you
can see the complete flow of individual sources of information
right through to the final net flow into pension funds which is
the aggregate that we publish. The problem with this was that
we were publishing an entirely separate survey and yet its effective
use required the bringing together of other information and that
was not clear.
133. Lightening seems to have struck the same
part of your organisation twice, does it not, one error in 1999
and the other in 2002? Is there something wrong with the pensions
area in the statistics they cover?
(Mr Cook) I think it is an area in which we believe
we have not responded quickly enough to the significance of the
statistics which are used in terms of the materiality of the debate
on the numbers rather than reflecting the effort needed to produce
the statistics. The systems which have been used have been quite
old. We have inherited this work from other organisations.
134. Pensions have moved to the centre of the
political forum now. There is a lot of attention put on them.
How reliable are national statistics on different aspects of pensions?
Will they bear the weight of so much attention?
(Mr Cook) One of the most important elements of pensions
policy, of course, is population estimates and projections. I
think there we do a huge amount of work with Government actuaries
and with other organisations and we have very high quality and
very analytically rich pensions statistics. We have quite a comprehensive
set of income measures which tell you about the income position
of people both in retirement and near retirement. We have very
effective savings information. We have proposals for a survey
of wealth which will give us information about the wealth holdings
of people before and after retirement. This area of pension funds
is one part of the whole picture of the information that is needed
in the pensions policy but what I can assure you is that the broader
mix of statistics that we have is an area of considerable concern
to my office and one that I think we are able to perform very
effectively in. The United Kingdom has an incredibly rich amount
of work and understanding about the income and position, for example,
of its retired populations which provides a very informed amount
of public policy in a debate on its pensions. The area I think
where we have more vulnerability is understanding the wealth position
of people who are retired and coming to retirement age and there
we have got quite a comprehensive proposal for a wealth survey
that we are very keen to proceed with.
135. It seems to me that first of all we had
a million missing people and now we seem to have billions of pounds
worth of money which never existed either. From what you have
just told us the money was never there. I would be interested
to see what your wealth survey tells us but perhaps meanwhile,
when we put similar questions to Sir John Kingman, he said to
us ". . . we are trying to get to the bottom of the series
of mistakes that occurred there which seem to be somewhere around
the interface between the Department of Work and Pensions and
the Office for National Statistics. What we are going to do is
to challenge the National Statistician to take hold of the statistics
in the Department of Work and Pensions and to make sure that department
operates to the highest standards. . ." Are there weaknesses
there and what are they?
(Mr Cook) I am not aware of the weaknesses that Sir
John was talking about. In terms of our work with Work and Pensions
we rely hugely on benefits data, for example, which is a significant
driving force of neighbourhood statistics. What I can say there
in terms of information management is they are probably one of
the leading departments in the public sector in the effectiveness
of the use of administrative data and statistics.
136. He put to us that he thought there was
a cultural problem in the Department. If there is a problem in
the culture of developing and using statistics in the Department
of Work and Pensions that would have a significant effect on policy.
Are you aware of such a culture?
(Mr Cook) No, I am not. My Department has enacted
occasionally a vigorous relationship with the Department of Work
and Pensions but it is one where on a professional basis it is
a very productive relationship.
137. You are satisfied entirely that the interface
between you and them is of the highest professional standard and
we are getting absolutely spot-on information and we can all go
away and sleep easily in our beds tonight knowing that our pension
funds are safe in their hands?
(Mr Cook) I ceased recently the production of vacancy
statistics, for example, some nine months ago because with the
change to Employment Direct the registered vacancy statistics
had a somewhat different coverage than they have had traditionally.
I regarded it as inappropriate to regard that as a time series
that meant the same thing after Employment Direct came in as it
did before so I ceased producing that. That is a decision I have
made but that is not a criticism of the professionalism of the
Department of Work and Pensions, it is saying that one of their
administrative sources has changed its quality. We have a considerable
variety of debates and discussions of that sort but certainly
we have an effective professional relationship. We have had, as
you might have noticed in the media, occasional differences in
other ways but on a professional basis I would have no concerns,
it is similar very much to the other relationships we have.
138. These differences you have just alluded
to, would they affect these outputs?
(Mr Cook) No.
139. So the next time a minister makes an announcement
on pensions we can be confident that the figures he is using are
(Mr Cook) If not you should be coming back at me.