Examination of Witnesses (Questions 20-39)
MR MICHAEL
FALLON MP, MR
SIMON BRISCOE
AND MS
JILL LEYLAND
10 JULY 2008
Q20 Paul Flynn: Your recollection
is wrong. There were patent examiners and trademark examiners
that were hugely specialised, and you do not pick those up at
the local labour exchange. The numbers who moved down were greatly
in excess of the figure you quoted. The Lyons Report did do a
job on this and presented it as a model of relocation but there
were many people who did not relocate at that time. I think there
was a feeling, if you accept that this is a success now, that
there was strong opposition which could have put the relocation
in jeopardy, but it was very much a London-centric view. I do
not know if any of those who made submissions at the time did
take submissions. I do not know if the Treasury Committee took
submissions from Newport or from those who had been involved in
previous relocations.
Mr Fallon: Yes, we did. We had
evidence from the Staff Unions and so on. I do not think it is
so much London-centric. I hesitate as to how to put this. Statisticians,
I think, are perhaps naturally conservative people. They have
made their choices as to where they live and where they are commuting
in, and there were a number, obviously, who had bought their houses,
made their arrangements to commute into Pimlico, who were suddenly
told the real choice was to move to Newport. I think that would
be disruptive for anybody.
Mr Briscoe: I think you also have
to look at the career move of a government statistician. There
is a Government Statistical Service. People would have in the
past expected to move between the Home Office, the Health Department,
the ONS, other government departments around London. What will
happen in the future is that will become extremely difficult,
so I think there is going to be a long-term issue about cross-fertilisation
of expertise. In the very short term, I think problems are beginning
to appear. We have recently learned that the Blue Book, the annual
statement of national accounts, is going to be delayed—its
publication for the summer is going to be delayed until the end
of the summer or the early autumn—and that improvements
that were scheduled for it are not going to be appearing this
year; they are going to be delayed another year. So there are
already signs beginning to appear that the loss of staff is affecting
the output of the Office.
Q21 Paul Flynn: Are you advocating
the relocation of even more jobs to places like Newport so that
cross-fertilisation can take place?
Ms Leyland: I would just like
to finish. Obviously, all these problems are manageable but there
is a particular concern at the moment about the loss of expertise
on national accounts. This is a very complex area. There is only
one place you become an expert in it and that is in the Office
for National Statistics. It is not something that is taught at
university. Perhaps it could be the subject at one of the universities
near Newport in the future but at the moment it is not a university
discipline and it is an area where people need a number of years
of experience. It is very important that a critical mass of expertise
is retained. Obviously, there are ways you can do that by flexible
working. I know the ONS management has it in mind but obviously
we are concerned that that does actually happen. Longer term,
I think obviously it is important that Newport does not become
an ivory tower, with people with a lot of technical expertise
but who do not know enough about what is going on in the rest
of the country. I would suggest, without being too London-centric
about it, it is probably easier not to go that way in somewhere
like London, where first of all you have a lot of users on your
doorstep, and also, of course, there are a lot of people who come
to London in the course of their work. To me, it is very important
that the staff in Newport have the time and the budget to travel
round the rest of the country.
Q22 Paul Flynn: The 2011 census:
is it not madness to have a census of everyone in the country
when a sample can be taken that would produce results that were
virtually as valid as doing a mass census, and that this huge
census of every person in the country is something that should
be part of the past and we could have a far more sophisticated
way of getting the results that we get than these very expensive
censuses?
Mr Fallon: This was the subject
of our last report before we handed over the reins to your Committee.
This is clearly one of the big issues your Committee is going
to have to think about fairly quickly after the 2011 census, as
to whether the census has a long-term future and what form that
future should take. There are clearly issues. We found examples
in other countries where they have abandoned the ten-yearly censuses
because they are simply too far apart, and they have a census
every five years. That is true in Ireland, for example. Other
countries use samples or have much more accurate ways. We found
in Canada, for example, and in Sweden much more accurate ways
of measuring the population all the time, every six months, through
the local authorities and so on. We were fairly disturbed to find
the lack of confidence in the population figures in-year in between
the censuses, particularly the extent of migration, not simply
inward migration but we discovered that really nobody has any
idea about emigration, as to how many people actually leave on
a permanent basis every year. There is a great deal of work to
be done there in improving the way the population is counted between
each census but also the census itself, which is an extremely
expensive operation. If it is not to command confidence in the
future, if this one is not more successful than the last one,
I think your Committee will have to think quite seriously about
how this is done in future.
Q23 Paul Flynn: Should we take a
decision now on the next census, Mr Briscoe?
Mr Briscoe: I do not see any hurry
to make a decision right now. I think it can wait until after
the next census but I think that some parallel running now of
checking and spring-cleaning various administrative records, allowing
the statisticians to have access to them for statistical purposes,
so that we can actually see what the administrative records were
saying on census day, so that you can benchmark the two and see
whether administrative records are a way forward, I think, is
something that could be put in train now. I also think that one
of the very unfortunate aspects of the last census was the failure
to have an address list. This might sound an incredibly boring,
technical point but if you do not know where the people can live,
it is very difficult to then count the people in the houses. The
Government has essentially given up and has left the Ordnance
Survey, the Post Office and local authorities with their own separate,
competing address lists, and I think that is a terrible shame
because if we at least had an address list, we would know where
the people were.
Ms Leyland: There are doubtless
going to be a lot of challenges with the census but the key thing
is, almost certainly, it is generally agreed that the way forward
is to move more towards the use of administrative data and other
methods, but before we drop the census, you do have to remember
that it gives you a huge amount of information because it is all
collected at the one time. You can get a lot of information for
very small areas, for example. The way forward has to be examined
very carefully and it is not something that can be decided overnight.
Q24 Chairman: In general, just as
a footnote to that, is it the case that the massively more mobile
population that we have now, the amount of people movement in
and out, people spending parts of the year here, parts of the
year elsewhere, has just transformed the difficulties of collecting
reliable data?
Ms Leyland: Yes. That is certainly
one of the major elements, probably the major element, but it
is also other things. There is less respect for authority than
there was, so people are less willing to fill in forms. Obviously,
with migration you have a greater problem of people who feel that
their status may not be legally accepted and therefore are even
more unwilling to fill in forms or are unwilling to answer any
sensitive questions. Then you have problems like gaining access
to gated communities, to blocks of flats with entry phones. It
has become a lot more difficult.
Q25 Chairman: If I am right, it was
in your memorandum that you were raising questions about the Government's
desire to have additional migration questions on the census, were
you not? You were preferring to have income questions.
Ms Leyland: Yes. Basically Keith
Dugmore, who ideally would have been here to represent the Statistics
User Forum, could give you much more information on this but,
as I understand it, the Government held a lot of consultations
about the questions on the census. That was fine and agreement
was reached. Then some government departments, because of the
furore over migration statistics, at the last minute wanted to
drop some of the agreed questions and replace them with others.
First of all, we think this is not a solution to the problem because
you cannot just measure migration once every ten years; you have
to measure it all the time. Secondly, we think that some of the
questions, like asking people details about passports, are such
that they will help to dissuade people from filling in the form,
so they could be damaging. In other words, we think this is a
problem that needs to be solved by other means.
Q26 Julie Morgan: Just on that issue,
some people in Wales did not fill in the form because there was
not a census box to tick to say you were Welsh. You remember there
was quite a furore about that at one time.
Ms Leyland: I do not think they
will make that mistake again.
Q27 Julie Morgan: I think it has
been agreed now that there will be one. As a Welsh MP, there are
some slight statistics I find quite difficult to get hold of.
For example, the number of Welsh women in prison; I have to ask
parliamentary questions to get that sort of answer. Data from
the Home Office is not broken down automatically by gender for
different offences. Then there are issues like teenage pregnancies
for Wales. I know the reasons for this are partly to do with devolution
and the way statistics are gathered, and partly to do with the
methods used by departments. Could you suggest how we can have
a more consistent way of getting statistics that is easier to
access?
Ms Leyland: This is something
that one can never solve a hundred per cent, but we can certainly
do better. We should be able to do better than we are at the moment.
Access, as I said to start with, and outreach of statistics is
one of the areas in which the UK has not done that well in the
past. As a first step, obviously, there are a lot of issues. There
is a lot of work being done at the moment on the website, which
is one way forward, but I think one of the things that the UK
Statistics Authority should be looking at early on is the whole
question of dissemination, outreach, access, holistically. Of
course, it is not just for the UK statistics administration. It
has to have the co-operation of all the other government departments
involved in that. I agree with you it is a big problem. I would
hope it is an issue that your Committee will certainly follow
up and expect to see progress on and ask questions if there is
not progress.
Mr Fallon: If I may just add very
briefly to that, it is an important feature of the new arrangements
that it is now a UK Statistics Authority. It is not simply something
for England. A lot of thought was given during the passage of
the legislation to how we can strengthen the various protocols
that exists between the Welsh Assembly and the Scottish Executive
and the Office for National Statistics. I would hope that the
new Authority would be able to take much more of a UK-wide view
of statistics when it comes to some of these issues.
Q28 Julie Morgan: So you think the
new Authority could ensure that some of the UK-wide statistics
could be broken down?
Mr Fallon: Yes.
Mr Briscoe: One of the principal
hopes I would have of the new Authority and where I would hope
you would give them support, and Parliament would, is in ensuring
that there is a decent planning system for deciding what data
is going to be collected. Some of that will be a devolution issue
and some of it will be a male/female issue. Others will be whole
new areas. I am not sure what the ONS and the Government Statistical
Service thinks it does by way of planning at the moment but really,
I do not see there being any planning. Effectively, to my mind,
there is no planning. I jotted down before coming here some of
the data that people at the Financial Times had asked for
in recent weeks. If I can just give you a few of these as examples:
estimates of illegal immigration; the gap between men and women's
pay when they are doing the same job; effective literacy and numeracy
rates; robust estimates of the amount of activity done by the
voluntary sector in the economy; on health care, information about
the private sector and alternative medicines; the number of uninsured
and untaxed cars; the proportion of people below the Minimum Wage
who are in non-compliance with the legislation; more information
about pre-school childcare activity; harmonised measures of disability.
The list could go on. There is an awful lot of these sorts of
data which I do not think are particularly special requests and
they are quite reasonable things that you might want to ask for,
or a citizen might want to know about the country, but we do not
have it. In fact, what has happened in the last few years in the
ONS is that their spare capacity, their intellectual reserves,
has been sapped by one-off reports that have been asked for by
the Treasury. The two most high-profile ones were the Allsopp
Report on regional data and the Atkinson Report on measuring public
sector productivity—both important subjects but they were
both launched without warning, without consultation, the reports
were produced and they have essentially been left to collect dust.
In the last two or three years management time has been so focused
on relocation that we have lost any mechanism for users being
able to put forward requests for data like this.
Q29 Julie Morgan: It is quite alarming
if there is no planning and the responses have been ad hoc
rather than looking long-term to what the needs are. Some of these
issues are very important, such as, obviously, the teenage pregnancy
issue, which relates to what, for example, each country is doing
in terms of addressing that, and you cannot get the data easily.
Ms Leyland: Yes, indeed. I am
not sure I would agree with Simon that there has been no planning
but there certainly has been nothing like enough. I think it is
largely a consequence of the decentralised and devolved nature
of the UK statistical system. The ONS can plan for itself. It
has not really been able to plan for other departments because
it is other departments who hold the purse strings and who make
the decisions. It can influence and argue but it cannot command.
Obviously, even before devolution there was a huge problem about
consistency across the UK of different statistics, and with devolution,
obviously, there are powers there for the devolved administrations,
so there is a need for co-operation from the devolved administrations.
The UK Statistics Authority only has power over the ONS. It has
some indirect powers via the granting of National Statistics status
to the other departments but that is not enough. If there is one
thing that I would hope this Committee does, it is to support
the UKSA in its endeavours—I am sure it will have the endeavours—to
actually implement a much better planned and coherent system and
hold other departments or administrations to account and ask questions
of them if they do not appear to be taking their full share in
their work, because I would agree with you it is right. Just a
point on devolution. Obviously, with things like teenage pregnancy,
there is a need to be able to look at that in the different countries
of the UK so that if, for example, Wales is suddenly finding a
way of reducing teenage pregnancies, other countries can actually
see that yes, this is what succeeds and this does not. It is that
sort of thing that we need.
Mr Fallon: I could add simply
that the ONS last year was under extraordinary pressure, as Simon
has indicated. It had the reorganisation under the new legislation,
it had relocation, it had staff and efficiency savings imposed
on it, and then, of course, it was preparing and doing the pilots
for the 2011 census, so it was an organisation under pressure
and I would hope your Committee will take an interest in how it
is coming through that this year.
Q30 Chairman: Will the UK Statistics
Authority have a role in identifying the gap-filling work that
Simon has just talked about? Does it have a remit to chase that
issue, do you know?
Mr Fallon: It has a work programme.
It has already listed some of the things it wants to focus on.
Q31 Chairman: Yes, but that is validating
areas of statistical work. I think what Simon was identifying,
as I understand it, were areas where the data is not at the moment
being produced.
Ms Leyland: The remit, as I understand
it, is that it has to monitor and promote statistics that serve
the public good, and that covers everything.
Mr Briscoe: If it has an interaction
with users, which it has an obligation to do, and reports those
gaps to Parliament, we can hope that that will help to plug some
of the gaps.
Q32 Kelvin Hopkins: The census: one
does not want to be sentimental about it but it has been with
us for 200 years. Is it not possible that we might lose some valuable
consistent time series, that we can look at our history with some
interest and even some use for the future? Is it possible we might
lose those?
Ms Leyland: Yes, if we stopped
the census, there would be swings and roundabouts and that is
why we have to plan very carefully. I think the ideal way would
be if we put a new system of administrative and other statistics
in place and then dual-run that with the last census. It is too
late to do that for 2011 because we do not have the new systems
in place and I do not imagine they could be put in place by then
but if, for example, you were able to set up systems between now
and 2021 and dual- run that with the 2021 census as the final
test, that would be an ideal way of going about it, I think.
Q33 Kelvin Hopkins: When time series
are rebased, they often have figures for the same year on the
different bases so one can make a comparison over a longer period.
Ms Leyland: If you do things at
the same time, you can have some form of a link or a check but
you are absolutely right that the census is a very rich source
of information, despite the shortcomings we have seen in the last
one. It is not something that should be dropped without a lot
of care and attention being paid to what replaces it.
Mr Briscoe: I do not think we
should be hooked on a ten-year cycle though. That is how it has
evolved because 200 years ago that was all that could be done.
If we can have a head count done every year, supplemented by survey
information, there is no reason why we could not almost have a
real-time or at least an annual census. In that sense, we would
be losing the ten-yearly census but we would be having something
much better that was much more frequent.
Q34 Kelvin Hopkins: I have a different
question. I very much welcome the move away from the Treasury
but have we gone far enough in making the whole system arm's length
from government, given that we know the Government is compulsively
mischievous?
Mr Fallon: Perhaps I could help
here? No, the crucial issue of pre-release was retained by government
and it still seems to me quite extraordinary that you set up a
brand new authority, put it at arm's length from government and
then say the one big issue that everybody has concerns about,
Ministers fiddling around on the day of announcing important statistics,
should be something determined by Ministers alone and not by the
Authority. It seems to me quite bizarre. I think we have seen
the repercussions of that. We have still not had the arrangements
laid. They were supposed to come into effect in April and they
have still not been laid before Parliament. That is one issue
clearly where we felt the Government had not gone far enough and
indeed, Sir Michael Scholar has himself said publicly that the
Government should have given that whole matter to the Authority
to deal with.
Q35 Kelvin Hopkins: Each year the
Budget Red Book seems to paint a glowing picture of a wonderful
British economy, doing so well. Then I also read, as my favourite
bedtime reading, the Institute for Fiscal Studies Green Budget,
which gives a slightly different flavour. One would like a consistent,
objective report on the economy which is not just the Treasury's
glowing report on its own success.
Mr Fallon: Yes, and I think that
matters now not least because the Government has quite fairly
set itself fiscal rules and it is incredibly important for the
credibility of those rules here and amongst the opposition parties
but also in the City and more widely that the statistics that
are used to measure compliance with those rules do command confidence.
There has been a real difficulty there in the past, when the ONS
was virtually part of the Treasury, with the way in which certain
lines of public expenditure were suddenly reclassified between
capital and current as you came to the end of the financial year,
that kind of thing; the extent to which PFI schemes involve real
liabilities as well as contingent liabilities and how they should
be measured on the public balance sheet. These are not just political
issues; they are very important issues for understanding more
clearly whether or not the Government is coming close to breaking
its rules.
Mr Briscoe: As one measure of
independence, if I could just try and provoke imagination, this
would not work in the UK at the moment but for those of you who
are not aware of what happens in Norway, it is quite interesting.
When a general election is called, government departments close
up, as is the case here, with the exception of the statistics
office, which every day during the three or four weeks of the
campaign will produce statistical briefings on the key issues
affecting the electorate. That is an example of how independent
they are and how different it could be if we wanted. At a time
when politicians are tempted to try and pull a fast one over us,
at a time when the media is a bit confused and a bit biased, to
have an independent body, akin to the IFS, is but it being the
National Statistics Office, producing the core data, I think that
is quite something to have in mind as a target.
Q36 Kelvin Hopkins: As always, Scandinavia
is more progressive and enlightened. I keep pressing, unsuccessfully
within my own party, for the adoption of more Scandinavian approaches
to all sorts of things. Another concern is about numeracy. I am
at risk of offending my electorate now by saying this but the
Moser Roser report some years ago suggested that 50% of the population
is not functionally numerate, and illustrated this by saying that
more than 50% do not understand what 50% means, so the Government
has an easy job to fiddle figures. Is it partly that people do
not trust figures because they are not comfortable with numbers
themselves? Is that possible?
Ms Leyland: It could be, I guess.
We are not as numerate a country as we should be. It certainly
does not help but I think it probably makes it all the more important
that the statistics are put out with integrity and honesty, so
that the messages are clear and are not distorted.
Q37 Kelvin Hopkins: One area where
people are currently very distrustful of government is on inflation
because, depending on what your income is, you have a different
basket of purchases—fine wine and air fares for some people,
basic foodstuffs and fuel for old age pensioners. But the Government
constantly uses one preferred statistic for inflation. Would it
not be more honest if we had an objective statement of inflation
or a range of inflation figures for different communities, different
levels of income, which was produced not by government but by
this independent statistical service, so if one is a pensioner,
this is what it is; if you are into fine wine and air fares, then
you get a different picture. I know they do now have different
measures but the Government focuses on the one they prefer.
Ms Leyland: One thing for sure
is that inflation statistics are controversial in every country.
It is nothing unique to the UK here. The ONS already does some
work like this. I think you still have this on the website, but
you can actually go to their website and try and use it to calculate
your own personal inflation index according to what you purchase,
so there are some aspects of this already being done. Also, the
UKSA has announced that one of its very first inquiries will be
into the communication of inflation figures, so hopefully that
will be looked at properly. I agree with you that it is a very
important issue, particularly at the moment, when we have so many
concerns about inflation.
Q38 Kelvin Hopkins: I used to teach
statistics at a very modest level, and one of the things I used
to do was to show how statistics are used to lie, the misuse and
abuse of statistics, fiddling the base years, lopping off the
base of a column so one can exaggerate the ups and downs and so
on. Would it not be an idea suggest to government that mathematics
education ought to have within it a little section about how statistics
can be abused, so people would not be misled quite so much in
future?
Mr Fallon: I think that would
be very useful and would certainly help also with financial education,
where people do not properly understand the percentage interest
rate or whatever they are paying on certain things. Can I come
back your first question, because you have really touched on one
reason why statistics have become much more important in a political
sense, which is that this Government, possibly a previous Government,
but in a way that has not happened in the previous 20 or 30 years,
seeks to measure itself by performance. There is a whole range
of targets, and therefore issues like whether crime is falling
or what types of crime are falling, whether inflation is rising,
whether waiting times in the Health Service are being met, are
much more important statistically than they were perhaps ten or
20 years ago. What has happened with inflation, of course, is
not only has there been a target but the target was changed statistically
from the RPI index to the CPI index, and I think that has seriously
undermined, whatever you think of the change, public confidence
in it. If you look at the Bank of England survey, which they do
regularly now, of inflation expectations, not only is the median
expectation that inflation will be over four per cent in 12 months'
time but there is a median assessment that most people think inflation
was over four per cent over the preceding 12 months. So clearly,
expectations and confidence in the inflation figures are seriously
at odds with what you would like to see.
Mr Briscoe: If I could add a small
rider to that, under the regime of national statistics we have
had until this year, the Treasury had decided to keep some residual
control over the RPI numbers on the grounds that it was so important
that it could not be entrusted to independent statisticians, which,
to my mind, was exactly the wrong way round. If it is really important,
you should actually take it out of the political sphere in the
way that, say, monetary policy has been. I do not know where the
RPI sits in the new regime but I think it is something that, for
the reasons Michael has mentioned, and given the history of the
RPI in the last handful of years, is something that is worth looking
at.
Q39 Mr Liddell-Grainger: Can I go
over some of the duties of the Statistics Board, because there
are things in here which are quite interesting and we have to
try and understand them. Every year you lay a report before Parliament
and the devolved legislators, the Scottish Executive, as you said,
and the Northern Ireland Assembly, on its activities. How do they
manage to get a coherent report when in fact there are different
statistical departments in all of these other places to try and
lay before Parliament and the devolved parliaments that is going
to make sense?
Ms Leyland: We will have to wait
and see how they do it, I guess.
|
