Examination of Witnesses (Questions 240
THURSDAY 16 NOVEMBER 2000
JOHNSON, MP AND
240. On this question of scope, Minister, the
aims and objectives that are set out in this Framework document
say that the purpose is to inform Parliament and assemblies and
the citizen about the state of the nation and provide a window
on the work and performance of government, allowing people to
access government policies and actually to be assessed. I have
looked through this list of topics and initiatives for the ONS
with two policies in mind. One is the Government's urge to improve
national productivity which, amongst other things, would involve
particulars on investment in research and development both by
Government and by the private sector, and indeed the Department
of Trade and Industry has in the past published a research and
development scoreboard, but also the health policy which is to
try and ensure that the treatment available in different parts
of the country is similar, which involves a certain amount of
epidemiological statistics. I can see neither of those topics
mentioned in here at all. The question really is, how are people
to know what is covered by the National Statistics, what are the
National Statistics, and will there be a full index published
at some point saying what at any one moment are considered to
be the National Statistics that people can refer to and therefore
pick out what they want to look at?
(Miss Johnson) Perhaps I could make an initial response
and Mr Cook might want to come in. The research and development
statistics are collected by ONS and therefore they are in the
scope. Perhaps a problem that you are highlighting with the initial
document is that there is a one-line entry which relates to ONS
statistics which obviously does not elaborate the content of those
statistics being included in the scope. We will certainly do our
best to produce a comprehensive list of what is included in the
scope by doing it on a line by line basis, but when the document
was being published it was just not feasible to put an entire
list under that ONS heading which is I think the first line on
the document that you are referring to.
(Mr Cook) Can I add that studies such as epidemiological
studies, where they come from, the demographic data that is collected
in ONS (and the population census of course is the key foundation
for that) will themselves be part of National Statistics where
they are regularly produced. There is a whole range of measures
that come from statistical frameworks. The Minister referred to
PAT 18. National accounts is another, balance of payments statistics,
demographic statistics, where, even if all the components are
not part of National Statistics, the resulting statistics will
be of such importance that the components will be expected to
have the qualities of National Statistics in terms of their basic
statistical integrity. I think that is one of the ways that we
will see National Statistics being defined in terms of the significance
of those very valuable statistical outputs.
241. This is a question for both Miss Johnson
and Mr Cook. By what criteria do you develop the list of statistics
that shall initially be included within National Statistics?
(Miss Johnson) It has been a question partly for departments
what they would like to see included in the scope of National
Statistics. It is obviously also a question about the professional
integrity and the quality of those statistics. They have to meet
up with standards which the National Statistician is confident
meet up with the requirements to be of a standard such as to be
put into the National Statistics scope. I think that is a very
important proviso. People cannot just volunteer things they would
like to put in there. They have to meet up with the quality required
in order to be included. I have to say that because some things
are not included it does not necessarily reflect in any way on
the quality of some of the things that are not presently included.
(Mr Cook) I am not aware of any country in the world
that has come up with a perfect definition of what is an official
statistic and what is not. Of course National Statistics is our
label for that. In essence it is the visibility, the significance
and the expectation of the accountability arrangement for the
quality of these statistics, because in fact National Statistics
in essence are statistics that people have to trust because of
the significance of the decisions that they make about them, and
therefore the National Statistician has an accountability for
ensuring that that trust is justified. I think that there will
always be a measure of pragmatism in the boundary as to what is
a National Statistic or not.
242. Did I understand you, Minister, to say
a moment ago that there is likely to be published a progressive
list of the statistics available?
(Miss Johnson) I think the line in particular in here
says "all ONS statistical outputs and public access databases"
which is not very explicit and I appreciate that that will cause
people problems. Much of the rest of this list is pretty specific
about what is included, but I think it would probably be very
helpful to expand with a list that single line that was used for
ONS's own contribution to the scoping exercise.
243. Would you not agree that it would be useful
to have a list of National Statistics, what is included and what
is excluded, in the way, say, that the Scottish Executive have
produced in Scotland?
(Miss Johnson) We are looking at what is included
in the GSS statistics, but to say all the statistics that exist
that might be outside both of those would be a very difficult
task for anyone to perform. For one thing many of the things that
are outside are done for various purposes across government. We
may not even be conceivably aware of them in some cases. There
may be small things going on in departments for their own internal
uses that have no major bearing on anything. It would be very
difficult to produce a total list across the board of "these
are in and those are out". What we can do is provide a fuller
list of the GSS statistics.
(Mr Cook) We are preparing a comprehensive list of
National Statistics. We will not of course be able to produce
a comprehensive list of statistics which are not National Statistics
because, as the Minister has pointed out, in essence what is not
a National Statistic but should be is a measure of judgement.
What we will identify are significant statistics that are not
National Statistics, but that, as I say, will always be a little
244. Do you think that non-government statistics
like house prices ought to be included in the National Statistics?
(Mr Cook) I think we need to think of the category
of statistics which are measures which we expect to have the quality
of National Statistics in terms of their statistical properties
because of course if they are produced for commercial reasons
they cannot be expected to have the same integrity of release
as we define it in terms of available to all at the same time.
There are aspects simply by being commercial statistics that make
it impossible for them to be National Statistics in the full sense.
But certainly in terms of quality, yes, we should seek to do that.
We may not of course have any influence or choice over that unless
245. Do you as the National Statistician believe
that the dividing line between those government statistics that
are included in the National Statistics and those that are excluded
is clear and capable of rational derivation?
(Mr Cook) We have a very strong mix, a very broad,
very comprehensive mix of statistics in National Statistics. If
you went to any country in the world it would certainly be a very
comprehensive and extensive mix. Where the exact boundary line
is placed in some areas from something at the margin being one
side or the other is debatable, but I think the value of the arrangement
is that it makes it very obvious.
246. If you were somebody in a company with
a particular exercise to do and someone was interested in a particular
topic, how would they know whether to look in the National Statistics
for statistics of relevance or not? What is the dividing line?
How do people outside the Commission and the ONS know what is
in and what is out?
(Mr Cook) Our aspiration is that the National Statistics
web site will become almost the first port of call for anyone
seeking good information about the United Kingdom, and of course
the power of the internet, the broad accessibility of it virtually
for anyone in the country who has access to the internet, all
the key gatekeepers, public libraries, commercial information
sources, that vehicle, as we become better at providing a web
site that anyone can use will actually provide most of that fairly
automatically for a user. There are also, as you know, people
in the value added information industry who take official statistics,
National Statistics, with commercial measures and tie them together
often for particular subjects and one might want to know what
is happening to the motor vehicle industry, for example. They
will use the official statistical sources and qualitative information
and quantitative information from other organisations, market
research for example.
247. Can you explain why the retail price index,
which is fundamental to an awful lot of calculations during 12
months, is not included in the National Statistics?
(Mr Cook) It is a National Statistic. It is one of
the most important measures in the United Kingdom. Its methodology
has been objectively developed by national statisticians. The
scrutiny of it by the Statistics Commission in terms of an overall
assessment of how objective any decisions about the RPI have been
made on professional matters is very clear. The National Statistician
has a very clear and unambiguous role on the methodology of the
RPI. Where the Chancellor has an important role in articulating
the particular importance of the RPI for government, then of course
that will be very clearly transparent as well.
248. So it will be included in the National
(Mr Cook) Yes.
249. As the Government's chief professional
adviser on statistical matters do you consider it will be within
your responsibility to advise Ministers on the scope of National
(Mr Cook) I will certainly have a view and I think
the arrangements make it important to define a process for me
presenting that view.
250. Minister, the National Statistics Framework
document does not make any mention of the role of the National
Statistician in relation to the scope of National Statistics either
as an adviser or in any other way. Would you like to comment on
(Miss Johnson) The document makes extensive references
to the role of the National Statistician and I think it is clear
that he is the top professional in the country as it were. He
is responsible for the quality and the integrity of National Statistics
and he is obviously able to make, as he has just said, comments
about these things. We envisage primarily the role of the Statistics
Commission to comment, if they wish to do so, on questions of
scope and obviously we are bound to make public and transparent
responses to those comments.
251. So that you do envisage it being within
the role of the National Statistician to be proactive in producing
advice on scope as well the role of the Commission?
(Miss Johnson) The Framework sets out clearly what
the position is in relation to scoping matters and it says the
National Statistician will take the lead in advising on methodological
questions concerning the RPI, for example. But obviously the scope
of that is a matter, as he has just been remarking, which rests
with the Chancellor of the Exchequer.
252. But methodological questions are not scope,
(Miss Johnson) The scope is a different issue. We
have said that Ministers will take those decisions about the coverage
of National Statistics in the light of the costs and benefits
that are involved. For the reasons that I was explaining to Sir
Michael earlier on, there are very often (in fact almost always)
significant resource implications to decisions about whether things
will go into the scope of National Statistics or not.
(Mr Cook) Could I make the point that the National
Statistician's role on scope is as an adviser to Ministers and
subject to the same obligations that an adviser to a Minister
has on any other matter. On professional matters it is the National
Statistician who is the lead authority and that is a very important
distinction. Certainly I do not think you should assume that any
national statistician will be inactive.
(Miss Johnson) The National Statistician will provide
key professional input as to whether something meets up with the
quality to go into National Statistics or not.
253. I understand that. The important question
is quite clear and it is quite clear in the document. The nub
of the point I am raising is whether Mr Cook sees it as part of
his responsibilities to advise you where he feels that the scope
is not adequate because it is not mentioned in the document in
those terms. I think the outcome of this, just for the record,
is to say yes, it is one of Mr Cook's responsibilities as an adviser.
Is that correct?
(Miss Johnson) I think his advice is on the professional
side of things. It is quite difficult to distinguish between the
sort of dialogue that often goes on about things behind the scenes
as it were and the public dialogue that sometimes takes place
as well, but in terms of the public dialogue it is the Statistics
Commission which has a very clear public role in commenting, and
I am sure the Statistics Commission as well as Ministers, obviously,
is working closely with the National Statistician on occasions
on these matters. Although they are providing an independent comment
they will be no doubt asking questions of the National Statistician
just as they may be of Ministers. You will certainly be free to
ask questions of them and their comments and our responses will
be a public matter.
254. This Framework document again does not
mention the role of the Commission in relation to the scope of
(Miss Johnson) The Commission's role is obviously
that it is independent of both Ministers and of the producers
of statistics, so it has got a key role as an independent body
in all these matters. It is free to comment on all the matters
that pertain to its responsibilities publicly, as is set out in
the Framework document. It will obviously have a role in helping
Ministers to take decisions about the scope and we have always
envisaged that and I believe the Statistics Commission are well
aware of the fact that they do have that role.
255. One thing was slightly unclear to me about
your answer to Mr Beard on the question of the list of statistics.
I think you said that you were thinking of publishing a comprehensive
list, an index. I was not quite sure when you answered the question
about the criteria on which that list would be based whether you
would publish that as well.
(Miss Johnson) There are two fundamental criteria.
Obviously Ministers are content for any particular statistic to
go into the scope. Secondly, and a very important secondly, is
that it meets up with the criteria which the National Statistician
has for the professional integrity and quality of the statistical
set that is being added. There is not a large set of criteria
for this. Obviously, the second criterion on professional integrity
and quality of the statistics is something which in a sense will
be unpacked in a number of ways, not least through the Code of
Practice. In terms of a detailed set of criteria, it is fundamentally
those two things.
256. So it is costs and quality?
(Miss Johnson) It is really, yes.
257. Minister, you have described the key professional
input that Mr Cook has into this system. Presumably Mr Holt, his
predecessor, also had a key professional input.
(Miss Johnson) Indeed.
258. What is the actual difference?
(Miss Johnson) There are a lot of differences I think.
First of all, we have got a Framework for National Statistics
which we never had before. We have got a set of statistics which
are a long set of statistics which are included in the scope of
National Statistics, which we did not have before. We have got
a Statistics Commission, independent of government, independent
of producers of statistics, able to comment publicly on what both
Ministers and the producers of statistics are doing. We have got
much greater engagement with users than we had before. I believe
that it is not only a question of there being greater quality
but also of there being seen to be that quality and that integrity.
It has not only been a question of whether there has been interference
in the past or changes to the statistics that are used for political
or other purposes, but also questions about whether there is anybody
there who can produce an independent view on these matters and
whether the system encapsulates the kinds of qualities of transparency
and openness and accountability which we believe it should. I
think that the changes we put in place provide a framework within
which those things can now be guaranteed and they can be seen
to be guaranteed by the public, by yourselves as part of those
who are scrutinising our new arrangements, both now and in the
259. Sure, but I was asking you really about
Mr Cook, and you replied, quite reasonably, about the Framework
document and so on and the Commission. Can Mr Cook provide a more
independent view than Mr Holt did?
(Miss Johnson) I believe that he can but I would not
want in any way to suggestand I am sure you are not suggestingthat
Mr Holt was not supplying an independent view at all times.