Examination of Witnesses (Questions 60-79)
MR MICHAEL
FALLON MP, MR
SIMON BRISCOE
AND MS
JILL LEYLAND
10 JULY 2008
Q60 Mr Prentice: You can ask questions
about ethnicity, but ethnicity does not cover Muslim women in
employment or issues like that, does it? What I am saying is,
is that another gap that we should be aware of?
Mr Briscoe: There was a fairly
shambolic performance a few months ago. There was a week-long
debate in the media when Gordon Brown mentioned the phrase about
"British jobs for British people". There were then different
numbers coming from the DWP and the ONS. One of them was using
nationality as recorded on surveys; the other was using country
of birth. They gave subtly different trends and that just allowed
the whole debate to rumble on for a week. It was front page news
for a week which is quite extraordinary on a statistical issue,
so we ought to have more clarity than that.
Q61 Mr Prentice: We have a Government
that likes to outsource everything that is not nailed down. Is
there a problem about outsourcing statistics or the collection
of statistics? I am looking at you, Jill.
Ms Leyland: I think it is an area
which needs clear guidelines. There have been one or two cases.
Obviously the Dr Foster issue is one that came into the public
arena where there are some question marks over what happened.
There may well be occasions, as with anything, where it is appropriate
to outsource and there may not. I would say it is an area where
guidelines need to be developed particularly because, as we have
had this in the Health Service, the whole issue gets very tangled
up with the debate over public versus private in the NHS so we
need guidelines that are based on statistical need.
Q62 Mr Prentice: I asked the question
because we had a memorandum from the Statistics User Forum and
it tells us that the ONS has been asked to make substantial efficiency
savings of £25 million and lose 700 jobs and then they go
on to say that these pressures are placing an impossible burden
on the ONS. That is damning, is it not?
Ms Leyland: This was the point
Michael was making earlier, that there have been a lot of pressures
on the Office for National Statistics for the combination of efficiency
cuts, IT modernisation, the new legislation and communication
altogether, so it is a somewhat different area.
Mr Fallon: On outsourcing, may
I make one addition to what Jill has said. There is one important
issue coming up, Mr Prentice, which is the possibility that one
of the contractors bidding for the 2011 census may be an American
company and what its obligations would be under the Patriot Act
to hand over, if required, some information it collected to the
US authorities if it was to win the contract. That is something
we sought assurances on from the Minister and I am sure it is
something you will want to continue to take an interest in as
we get up to the period when the contracts are awarded.
Q63 Mr Prentice: There are two outside
bids, I know that, but there is an in-house bid, yes. 2[2]
Could I finish on this one, Chairman. In the written memorandum
that we got from the Royal Statistical Society, Jill, you raise
a series of areas of concern and there is a fascinating one here
right at the end about a possible referendum on Scottish independence,
it is D1. You say that statistics will be needed to assist voters
to come to a sensible conclusion on whether Scottish independence
is a good thing or a bad thing. The Scottish nationalists up in
Holyrood would tell you it is Scotland's oil and how do you handle
that one?
Ms Leyland: I am not saying that
the statistics will answer the question for everybody, of course
they will not. Just to pick up on what Simon said earlier about
the Norwegian experience, where every day during an election period
the Norwegian statistics office produces data on issues of the
moment, I think if there were a Scottish referendum it would be
very important that there were some good facts and figures out
there about the relationship of Scotland to the rest of the UK,
for example good information on public spending in Scotland and
Scotland's share of that which is always a big issue. I think
we will be saying to the UK Statistics Authority, "You need
to bear in mind that if there is a referendum on Scottish independence,
you need to be prepared to produce appropriate information".
Q64 Mr Prentice: This is an impossible
question to answer, but would the UK Statistics Authority be prepared
to face down the Scottish Government in Holyrood on something
as critically important as statistics?
Ms Leyland: I do not see why it
would need to face it down because it would only be putting out
the facts and the figures and there may be some debate over that.
Q65 Mr Walker: Mr Briscoe, you said
that an increasing number of surveys were going unanswered when
they are sent to people's homes, perhaps even their workplaces.
Why do you feel that is?
Mr Briscoe: I suspect it is a
whole host of reasons. People just do not feel compelled to fill
them in, I suspect. There is not that sense of duty that perhaps
existed in the past. Although I am not an expert on surveys, it
is just the evidence.
Q66 Mr Walker: Do you think it could
be that they believe the forms they fill in are not used to benefit
them but are used perhaps to disadvantage them? We read in the
papers every day about people being snooped on and spied on and
so on and so forth, and there may be a concern that it is another
attempt by Government to get a look into what they regard as their
private lives and private business. Do you think there is a sort
of cynicism now creeping in?
Mr Briscoe: I am sure there is
a part of that, but there are ways of addressing that. One of
your colleagues earlier mentioned very specifically the declining
response rate to business surveys. What some other countries'
statistical offices are doing is interesting: they get the response
from the firm and when all the figures have been processed they
return to the firm their figure but set against the average for
that region, that county, that country, whatever, so the firm
is able to see their figure in the context of a broader picture.
I think there is some evidence that where there has been feedback
like that you can help to restore response rates because, of course,
companies spend a small fortune on that sort of market research
and market positioning information. The analogy does not necessarily
work with individuals but certainly personally—and I might
be a bit of an extreme here—I would love to have figures
returned to me, if I had filled in a form, saying what my community,
the borough and the city was like. It might be a way of trying
to engage people and show them the importance.
Q67 Mr Walker: My concern here is
that my constituents view it, rightly or wrongly, as sort of another
method for extracting more cash out of them at a later date, for
example with the amount of rubbish they throw away or increased
charges for parking. Is that a concern, Mr Fallon, that you think
is something we need to address?
Mr Fallon: Yes. I think it does
underpin the growing reluctance of people to fill in forms just
because they are told to fill in forms, but there is a trade-off
here. If you want a more accurate census and more accurate information
on which to plan public policy or public spending, then you need
your statistics to be as accurate as possible. Every time we look
at that, people come up with ludicrous examples of where data
is simply not being shared, it is publicly collected but is not
even being shared, never mind problems with its collection. It
has been collected but is not being shared between different departments.
I think there is a lot more progress to be made there before you
have to go out and ask people for more and more information.
Q68 Mr Walker: If you were a betting
man, would you have thought that the forthcoming census, I think
you said 2011, would get a higher percentage rate of response
than the one conducted a decade earlier?
Mr Fallon: I think this time local
authorities will have much more interest and incentive in making
sure that people do co-operate with the census because they can
see more clearly the public expenditure implications if they get
a poor response rate in their own particular area. My hunch would
be that local authorities would work much more closely with the
census arrangers and look at ways of encouraging turnout.
Q69 Mr Walker: Mr Briscoe, you mentioned
that it might be sensible to go to a rolling annual census. Some
of the concerns we have had over, say, the last five years are
we have had high levels of inward migration into certain parts
of the world and public services have not been able to keep pace
with the additional demands on them because nobody has been anticipating
or measuring these demands. Would you see this sort of annual
census as a way of smoothing out these peaks and troughs in demand
to make sure additional demand can be recognised extremely quickly
and then additional resources found to meet that demand?
Mr Briscoe: The ONS does produce
mid-year estimates for population for every year and it benchmarks
them against the census. Enough local authorities, I think, have
kicked up a stink suggesting that the modelling the ONS does between
censuses is not fit for purpose and, therefore, an alternative
method of calculating mid-year estimates is required. Whether
you call that an annual census or something else, let us not get
hung up on terminology. I think that populations do move a lot
faster now than they did 200 years ago and, therefore, the systems
have to move on.
Q70 Mr Walker: Mr Fallon, are you
confident that lessons are being learned?
Mr Fallon: Yes. We reported on
this, the pilots have been done and I would hope that next time
there will be a higher response rate and the data collected will
be more accurate. We were quite struck as a Committee by how successfully
this was done in other countries, these mid-year estimates seemed
to be more accurate. I think it was true when we went to Sweden
that we asked how up-to-date the information was on the actual
population count and we were told it was up-to-date from the previous
Friday because everybody in Sweden has a personal ID number there,
which, again, is an another issue. Of course it is a smaller country,
but Canada is the same, they seemed to have much more sophisticated
ways of estimating the population change mid-year and every year.
We have to get better at it and I hope with the next census we
will achieve that. Censuses are very expensive operations. I think
it will cost something like half a billion pounds. It is a formidable
amount of money and if it is to be justified, it really has to
produce the results, and I hope your Committee will have a look
at that.
Mr Briscoe: If I could mention
one technical point about the response rate to the census. I have
never had an answer to this, but I think the response rate was
said to be 96% or something like that. What we do know is the
number of forms that came back and the number of individuals who
were on those forms. What we do not know is what the denominator
was from which they have calculated the response rate. I do not
know how they got that figure, it was some sort of estimate that
someone came up with about how many people they thought were in
the country. It would be a delight if you would be able to find
out how they did calculate that. My fear is that after the last
census where an awful lot of men in their 20s and 30s suddenly
disappeared, a million of them just vanished, if somehow Mr Fallon
is right and the census does get a better response rate this time,
would the ONS feel obliged somehow to hide these million men that
they have just re-found so that the 2001 census did not look too
awful. I do think there is a little bit of thinking to be done
about whether the response rate is the best way of measuring success.
Q71 Mr Prentice: They claim the 2001
response rate was 96%.
Mr Briscoe: Whatever the figure
was.
Q72 Mr Prentice: You would imagine
in real terms it was probably a little lower than 96%.
Mr Briscoe: We were told at the
time that these million young men had all gone off for gap years
to Australia or something, which was strange given areas of London
like Lambeth had lost a sixth or a quarter of their young men.
I do not know whether there was a sudden influx of young Lambeth
men in Australia, but if you added back the million people, I
think the response rate would come down to 93-94%, I am not sure.
Chairman: I hope you keep us supplied
with a steady stream of these questions that we will ask.
Q73 Kelvin Hopkins: On this very
point, our local authority has done some research in Luton and
they estimate that the population of Luton is 20,000 larger than
the official estimate, which is ten per cent of the population.
I was astonished, but if that is repeated across the country,
clearly the statistics are way out of line and something serious
must be done.
Mr Fallon: We took evidence from
a number of authorities showing exactly the same kind of scale:
Slough, Westminster, Manchester and so on, and there are probably
others.
Mr Briscoe: It is worth mentioning,
it is not just the financing of local authorities and the provision
of services, the census is a very important set of figures because
it acts as a denominator for all sorts of other statistics. If
you think about a target for getting young people to university,
because the number of young people fell in the census, the proportion
of young people going to university rose from 41% to 43%, which
helped edge towards that 50% target. You get all of these sorts
of side effects from the mis-measurement, which, again, influence
perceptions about success or otherwise of public policy.
Q74 Paul Flynn: I am totally alarmed
at of Mr Briscoe's suggestion that there could be finessing of
the figures. Do you agree with the hopes for the Act which went
through that it would be almost as valuable as the independence
of the Bank of England and we really are entering into a brave
new world of independent statistics. Should not the fiddling of
figures that you fear of hiding one million men be impossible
under the Statistics Authority?
Mr Briscoe: I hope so and that
there will be a new air of transparency and honesty and that an
increasing number of Members of Parliament and the press will
take interest. I take enormous heart from what I said at the beginning,
that there was such widespread hope and support for the new legislation
from all parties, from both Houses and from the media and elsewhere,
so I think there is a lot of goodwill to get this sorted out.
Q75 Mr Prentice: We would not be
able to mislay one million people once the national identity scheme
is up and running, would we?
Mr Briscoe: You are straying into
other territory.
Q76 Chairman: Can I ask one thing
as we end because it picks up what you said, Simon, about Norway
and, Michael, what you said about the new era of performance measurement
that we are in and, Jill, you said about the need for more outreach
work. I have always had a sort of whacky idea that we ought to
have a statistical ombudsman so that when all this political argument
rages about whether figures have gone up or down and so on, there
is some place that someone can go to have some sort of reputable
answer to these questions. What I am really asking is should we
see the role of the statistics service as providing regular kinds
of briefings aimed at public consumption which would cut through
all the rival political claims about things and tell the public
in a reasonably straightforward way whether crime has gone up
or whether it has gone down, whether MRSA has gone up or whether
it has gone down, some of these sorts of things.
Ms Leyland: I think that is right.
The statistics should be published in a way which allows that
to be said. I also think there is no reason why the UK Statistics
Authority cannot correct misperceptions when they happen. It does
occasionally happen if you have a minister who misleads badly
that they are corrected and put right.
Q77 Chairman: A kind of umpire.
Ms Leyland: Yes. The UK Statistics
Authority should be a body to which people turn in case of doubt.
Mr Briscoe: I do not know whether
the Statistics Authority will feel it can fulfil that role or
whether it can help others, but there are one or two groups around
trying to do this. There is a group run by Channel 4 which is
called Factcheck, which was borne out of arguments in election
campaigns and they do occasional releases of this sort. There
is a Member of the Lords who is currently establishing a group
called the Campaign for Real Statistics, so there are people out
there who are edging in that direction. There is a growing awareness
that something like this is needed.
Q78 Chairman: I cannot quite see
why it cannot be done officially. I did try this on Sir Michael
Scholar the other day and he was not terribly keen on it and,
Michael, you do not look terribly keen on it either.
Mr Fallon: I am not keen on it,
no, because I do not think you can subcontract everything. Parliament
has now voted for this Statistics Authority and I think it is
interesting that Sir Michael has chosen the title "Authority".
I think the "Authority" ought to get into the position
of being able to rule on these things definitively. Where there
are arguments leading up to the point of that ruling or afterwards,
then I think your Committee has a big role to play in supporting
the Authority in its work in making rulings like that.
Q79 Chairman: I think we have covered
a lot of ground. Are there things that ought to be on our radar
which we have not mentioned?
Ms Leyland: Just one thing which
is in our briefing, but I would like to emphasise it. In fact,
we have touched on it, it is the importance of evaluating, consulting
and engaging with everybody in the planning of statistics for
the future. At the moment, national Government departments have
a lot of influence in what goes on. Clearly Government departments
are important users so they need to have some influence, but they
are not the only people. It is very crucial that the UK SA sets
up a structure where it engages with people and where it evaluates
what is needed.
Chairman: Thank you very much indeed.
I think we are very conscious of our responsibilities in this
area and want to do a good job. I think we are conscious that
Michael's Committee has set a rather high bar in the past and
we want to try and live up to that. We wanted to cast our eye
around the territory today, which you have enabled us to do, so
we can begin to have an active presence in this area. Thank you
very, very much indeed to all of you.
2 2 For correction see Ev 26 Back
| 
