Examination of Witnesses (Questions 170
THURSDAY 16 NOVEMBER 2000
170. Good morning, Sir John.
(Sir John Kingman) Good morning.
171. We are going to move at quite a pace today
because we have the Minister coming in three-quarters of an hour.
Could I ask you first of all whether you have noticed any specific
changes and improvements, if I can put it that way, since the
new framework has been established?
(Sir John Kingman) Of course this is
a time of very great change in National Statistics with the arrival
of the new National Statistician as well as the changes in the
structure, and I think things are moving really quite fast, but
the long-term effect of the changes will take some time to emerge
and I think the jury is still out.
172. You say things are moving fast; in what
direction? Can you be specific about that, Sir John?
(Sir John Kingman) Yes, I can. Clearly the structure
that has been laid down in the White Paper and the Framework Document
gives a central position to the office of the National Statistician.
Len Cook has picked up the managerial challenges of the Office
for National Statistics itself but he also has a role in relation
to the quality of National Statistics which are produced outside
ONS and a further role in terms of the whole Government's statistical
service as the professional head of that service. Inevitably,
I think, the most pressing problems are in the first area, in
the ONS itself, and I do not want to comment in detail because
the Commission has not yet investigated in detail what is going
on in ONS, but my impression is that his arrival is resulting
in some quite profound changes being initiated but not of course
173. Have you yourself identified any elements
which need further improvement?
(Sir John Kingman) We have already said to the National
Statistician that we would like to discuss with him and his colleagues
ways in which the statistical output from ONS and National Statistics
at large can be accompanied by indications of reliability and
accuracy; that is good statistical practice that we would like
to see widely adopted within National Statistics. We have also
had discussions with him on more detailed matters which I think
it would be premature to go into but that is our first substantive
request of National Statistics, that we should all have indications
of just how accurate the figures are that come out, because all
statistical figures are subject to error and it is good practice
to quantify that error.
174. One of the fundamental worries of this
Committee has been the question of whether there is too much political
control over National Statistics. That has not changed very much,
has it, in terms of the scope of the statistician or indeed the
priorities of the ONS which are still laid down by Ministers?
(Sir John Kingman) I think it will be much more difficult
for any Ministers who wanted to try to distort the output of National
Statistics to do so, because the role of the National Statistician
in terms of the quality of the statistics has been much more clearly
spelt out than at any time in the past, so he has a much stronger
position than his predecessors had, and the Statistics Commission
is there as a watch dog and will report if it sees any sign of
any political interference with statistics. That is one of its
main functions and it will do that as well as it possibly can.
175. Sir John, how did you devise your initial
programme of work?
(Sir John Kingman) We are still considering our initial
programme of work. We have identified two particular areas so
far but we are trying to be sensitive to the needs of users and
to what users of all sorts of National Statistics think is important.
So, for instance, I attended the recent conference of the Statistics
Users' Council, which is a very widely representative body, and
I listened very carefully to the concerns which were being expressed
there, and we shall feed that into our discussions about our work
programme. We have identified the two areas so far, one which
I have already mentioned to Sir Michael, and the other that we
are going to have a look at the situation with cancer statistics,
because that exemplifies some of the tensions and dilemmas which
exist in National Statistics generally.
176. You have said you have spoken to the Statistics
Users' Council; would you listen to what Parliament says about
(Sir John Kingman) Of course.
177. How would you go about amending your programme
of work given the need to ensure this political impartiality and
independence but also the need to respond to key users of your
(Sir John Kingman) I think it is important to understand
just how wide and how heterogeneous the community of users is.
The Statistics Users' Council represents a lot of the groups of
what you might call "conscious statistical users", people
who get up in the morning thinking, "I am going to use some
official statistics", but there is a whole range from Parliament
at one end through to the Clapham Omnibus at the other, with the
Governor of the Bank of England and other interesting people in
the spectrum. They all have different expectations, different
needs, some of them know what they need and can express that very
clearly, others really do not know what they need and it is up
to us to try and interpret that. Parliament, I think, is very
important and we shall listen very carefully to what is said in
this Committee, what is said in the House and so on, and I understand
there is to be an Adjournment Debate which we will be very interested
in. We are really in the market for concerns about statistics
which we will then decide whether we should pursue or not.
178. Since it is my Adjournment Debate you are
referring to tomorrow, could I raise one of the points I will
be raising, because it relates to one of your tasks, which is
the extent to which National Statistics successfully communicates
information about the reliability of its figures? This relates
to a recent ONS study into the way the ICT industry is measured
and the effect that has on industrial production. You will be
aware that recent survey suggests if industrial production is
measured using the hedonic pricing methods of the USA, production
would be 6 per cent higher since 1995 than the 3 per cent measured.
Does your programme allow you to deal with those sorts of real
concerns because those sorts of errors could have a major impact
on the whole set of economic statistics?
(Sir John Kingman) Of course. That is certainly something
that we should consider very seriously for inclusion in our work
programme. I agree.
179. Well, "consider it", but I would
argue it is a very high priority. Are you actually looking actively
to include it as a top priority?
(Sir John Kingman) We are now, as of three minutes