Examination of Witnesses (Questions 20-39)
DEPARTMENT FOR
WORK AND
PENSIONS & JOBCENTRE
PLUS
MONDAY 4 FEBRUARY
2008
Q20 Mr Touhig: 60% of your reduction
of £1.2 billion is attributable to a definitional change.
Sir Leigh Lewis: If you take our
starting point which was £2 billion and where we now are,
£800 million, and remember that is against a background of
a rising total of benefit expenditure overall, then it is the
case that £550 million of that reduction, very openly and
very clearly, has been because of a re-categorisation of what
had been previously regarded as fraud as no longer being fraudulent.
Q21 Mr Touhig: The Report here makes
it £700 million. It says that the Department conclude that
these overpayments of £700 million should no longer be classified
as either fraud or error, so a 60% reduction at a stroke of a
pen.
Sir Leigh Lewis: No. Just to clarify
on the £550 million versus £700 million, we estimate
that, of the £700 million change, £150 million had previously
been classed as error and £550 million had previously been
classed as fraud. Another way of looking at this—and it
has been completely open—is that we were overestimating
fraud for a considerable period before, so it is true to say—the
Report is very clear and there is no attempt to pretend otherwise—that
a part of the reduction has been through definitional change.
Q22 Mr Touhig: The Report makes that
clear. Coming back to the point you explained to me a moment ago,
in 2004-05 you reviewed your measures for fraud and error and
found that health conditions of some customers had improved gradually
over time and although they could not be expected to self-diagnose
that their level of disability had changed it affected their entitlement
as a result and they were still being overpaid. Your Department
concluded that these overpayments, £700 million, were no
longer error or fraud. They were still overpaid.
Sir Leigh Lewis: We concluded
with our auditors—this was very thoroughly discussed and
very clearly stated in our Resource Accounts at the time—that
until the point where we had cause or our customers had cause
to talk to us it was neither fraud nor error because the customer
could not reasonably have been expected at that point to have
realised themselves that their condition had improved to the point
where they might no longer qualify for DLA at the rate or at all
that they were receiving it; and that it was therefore right and
appropriate no longer to regard that sum as either fraud or error.
Q23 Mr Touhig: You live in a different
world if you think that they may not realise their conditions
are improving while they still continue to claim benefit. That
is not my experience of the cases that I have. Your Department
had a PSA target to reduce its fraud and error. You met the target
but you only met the target because of the definitional change.
It is cosmetic. You have a target. The Chairman and I were discussing
targets the other day. We perhaps have different views on targets
but you have yours. You are ticking the box because you have changed
the rules. You have changed the guidelines. You just said, "This
is no longer fraud. This is no longer error. This is definitional
change" and, boom, there is £700 million and therefore
we have met our target that we have agreed with the Treasury.
Sir Leigh Lewis: I do not think
I am doing well in explaining this sufficiently clearly. Let me
just have one more go. The SRO2 target which was to reduce fraud
and error in income support and JSA was not affected by this definitional
change because it affected a different benefit, disability living
allowance. That target was met and was not affected by the definitional
change. The overall reduction in fraud from £2 billion to
£0.8 billion has been affected by that definitional change
to the extent of £550 million, but remember two things. One,
that probably means that we were always overestimating fraud in
those earlier figures and, secondly, benefit expenditure as a
global figure has continued to increase over that period.
Q24 Mr Touhig: It is still my contention
that definitional change has greatly helped you reach your target
but perhaps we will not agree on that. The Report also tells us
in paragraph 1.13 that the Department does not exactly know how
much it has spent in implementing its anti-fraud strategy. Why
is that?
Sir Leigh Lewis: I think we do
know what we have spent.
Q25 Mr Touhig: The Report says you
do not. Have you disputed the Report?
Sir Leigh Lewis: No. The Report
is of course an agreed Report. What is in a sense at issue between
us is that we know very clearly what we spend on each element
of our counter fraud strategy. As the Report very clearly says,
that is value for money and the Report says that all of those
activities should continue.
Q26 Mr Touhig: I would just point
you to paragraph 1.13: "The Department does not know the
exact costs of implementing the strategy ...". It is like
being pregnant. Either you are or you are not. Either you know
or you do not know and the Report is saying you do not know.
Sir Leigh Lewis: I do not dispute
the Report, because I would not have signed it off as Accounting
Officer. I was seeking only to put that comment into context.
Q27 Mr Touhig: Your Department for
the 18th year has had its accounts qualified. Are you having a
coming of age party?
Sir Leigh Lewis: I would quite
like to say something on this because I think it is becoming increasingly
indefensible on almost every test that we have a Department whose
accounts have been qualified for nearly 20 years. There are two
issues going on there, just to be absolutely clear with the Committee.
One is that both levels of fraud and error remain too high and
we need to bring them down. This Report says that fraud may be
close to being at an irreducible minimum. Error most certainly
is not. That is our first responsibility. The second is—I
have been very clear with this Committee before—if we continue
to have the current threshold of 1% of expenditure being the benchmark,
and the only benchmark, by which qualification is determined,
this Department's accounts will go on being qualified for all
time because I believe it is inconceivable that we will ever reach
a point where the level of incorrectness will ever be less than
1% of benefit expenditure, in a system which is hugely complex.
Sir John Bourn has now retired but we have been having a very
good dialogue with the National Audit Office because I want to
achieve a different basis on which we can look at the qualification
of the Department's accounts which gives this Department a serious
opportunity to remove that qualification. I said on my very first
day in this Department as its Permanent Secretary that I wanted
to be the Permanent Secretary who was present when that qualification
was removed. There is not I think a single member of my senior
management team who does not know that that is my absolute determination
and ambition, but we will not do that—just to be clear,
if we are having a serious conversation—while the present
test for qualification remains. There is not an organisation of
any standing dealing with mass financial transactions in the private
or public sector that is able to reduce its overall losses for
fraud and error to below 1%.
Mr Touhig: You give a very clear explanation.
I am sure the NAO will take note of that as well.
Q28 Chairman: I think the NAO should
have an opportunity to comment. I think it is very important.
Apparently discussions have taken place. Should we recommend changes
to the 1%? What is your view?
Mr Burr: I have not myself had
discussions with the Permanent Secretary. Yes, we have had discussions.
It is only fair to say that the Accounting Officer has expressed
a very clear view of the matter. We do not think we want to be
in a situation where we are removing the qualification simply
by moving the goalposts. How far we can progress in terms of giving
a rather clearer message on progress made and the materiality
of the error for Parliament's attention is the area in which the
debate is progressing.
Q29 Chairman: I will not ask you
to comment now because it is not fair but I think it would be
useful for the Committee to know what sorts of figures we are
talking about. 1% has been impressive. Are we now talking about
2%, 1.5%? Sir Leigh, what is your view? Equally we cannot make
it such an easy target or there is no point.
Sir Leigh Lewis: I am pleased
to have this as a really serious conversation. If we are to have
a different target, first of all, I want it to be one that is
really stretching my Department but, secondly, I want it to be
one that is capable of being achieved. I have run large organisations
in the Civil Service as you know, Chairman, for most of my recent
working life. What I know about targets is that if you have a
target which people believe is achievable then they will go to
incredible lengths to achieve it. If you have a target which people
believe can never be achieved then you cannot get that degree
of buy in. The conversation which we have been having—it
has been a good conversation with Sir John before he retired and
colleagues—is whether there is another way of approaching
that which is looking at the separate benefit streams and asking
a question about the level of materiality which we might attach
to each. I simply believe that there are two responsibilities
here. There is a responsibility on me and my Department to be
able to demonstrate to this Committee and to Parliament that we
are doing everything that could reasonably be expected of us to
bring down levels of fraud and error. Equally, I think there is
a responsibility perhaps, if I dare say it, on this Committee
and on Parliament to give us a threshold which is capable of being
achieved.
Q30 Chairman: Do you want to do us
a note on this?
Sir Leigh Lewis: I would welcome
doing you a note on that.[1]
Mr Burr: It is ultimately a professional
audit judgment as to what level of error should be brought to
the attention of Parliament. We have not put into circulation
any other figures than the ones mentioned but there is a question
as to whether one can devise a more illuminating way of displaying
this for Parliament.
Chairman: You will advise us. Thank you.
Q31 Dr Pugh: It is good to know it
is not only the EU that does not get its accounts signed off year
after year. £106 million is the total figure, is it not,
for what overpayments stand at at the moment? What we were moving
towards in the last few minutes was some sort of sense from you
of how low that could go, because there will never be no fraud;
there will always be some fraud. Can you, by using international
comparisons, work out what level fraud can be got down to if you
implement the most efficient practices and carry out all the advice
from the Committee of Public Accounts to the letter? Is there
a level at which you cannot get any lower at all because nobody
internationally so far has been able to do it?
Sir Leigh Lewis: Will you bear
with me if I just give you a little bit of factual background
and then absolutely address the heart of your question? One of
the quotes in the Report is that the UK has levels of social security
fraud which are similar to those in comparable countries but,
as we reported in 2006, the DWP has a better understanding than
other nations of the problems and is doing more to tackle them.
The NAO carried out a benchmarking study that was undertaken for
them by an organisation called Rand Europe in 2006 which looked
at eight countries similar to the UK in terms of wealth and diversity
of population. I do not want to take up more of your time than
you would wish but I could tell you the countries. The interesting
thing is that a number then at least—I cannot speak for
now—did not seek to measure levels of fraud and error either
at all or globally. Of those that did, none appeared to have a
level which was lower than ours for fraud and error as a totality.
If you take fraud, the Report itself says that we may now be approaching
a position where it will be difficult to reduce the level of fraud—I
draw a sharp distinction between that and error—substantially
further. We are at a point where only a little over half a penny
in every pound that is paid out is paid out fraudulently. As the
Report says, any welfare system will always be susceptible to
people who are seeking to defraud it. No system will be capable
of ever preventing that.
Q32 Dr Pugh: Do international comparisons
show there is any connection between the complexity of the system
and the level of fraud found in it, or is that not a viable correlation?
Sir Leigh Lewis: I certainly think
it is a very viable correlation when it comes to error. We have
had this conversation in the Committee before. When it comes to
error committed both by customers and by our staff, there absolutely
is a correlation between the complexity of the benefit system
and levels of error. I think fraud is different because the defining
characteristic of fraud, as I was saying to Mr Touhig, is that
the person is deliberately doing something that they know to be
wrong. Therefore I do not think they can cite the complexity argument
in defence of acting fraudulently. Absolutely it can be cited
in defence of unwittingly failing to interpret the rules correctly.
Q33 Dr Pugh: Coming to sanctions
now, I have some experience of these. I have had a court case,
not as a person who was a defendant but as a witness, a very high
profile one which involved a former Mayor of the borough in which
I reside. I was very impressed by the strength of the case presented
by the DWP and the rigorousness of it but also by the enormous
cost of it. I was witness number 52 and I was not expected to
be called but I was. It seems you are quite successful with court
cases. How many people plead guilty when threatened with prosecution
by you? What percentage?
Sir Leigh Lewis: I will just check
this as we speak but from memory the figure for those who plead
not guilty is about one in five of the cases that are prosecuted
or go to our prosecution division that are brought forward for
court proceedings are contested.
Q34 Dr Pugh: I presume most of them
are defended cases, are they not? What is the average cost of
a court case?
Sir Leigh Lewis: We could do the
maths. The total cost of our prosecution division—that is
the last piece in the chain—was about £9.5 million
in 2006-07.
Q35 Dr Pugh: The average cost of
a case?
Sir Leigh Lewis: No. That is the
total cost of running our prosecution division as a whole in 2006-07.
In that year we had 6,756 successful convictions and brought court
action in 7,483 cases. You are right to say that some cases are
very expensive to prosecute and bring to court. We had just one
reported in this morning's papers of an individual sentenced to
three and a half years' imprisonment as a result of a major fraud,
but that case and cases of a similar nature can inevitably be
very expensive.
Q36 Dr Pugh: Your success rate is
so high I ponder why you do not bring more cases to court. I can
only assume that the sheer cost of doing so deters you even when
you think you have a better than average chance of succeeding.
Sir Leigh Lewis: No, it does not.
We are not cash constrained in that way. Our prosecution division
accepts every referral to it that comes from the Fraud Investigation
Service which reports to my colleague, Neil Couling. It considers
whether a prosecution should go ahead and it applies two tests
in that respect. It applies the evidential test that all prosecuting
authorities apply: is there sufficient evidence to make it more
likely than not that this case will succeed if prosecuted? It
applies a public interest test because there can sometimes be
other reasons why a prosecution may or may not be in the public
interest, but we are not cash constrained in that sense. Clearly,
there are some resource implications in all of this but we are
not in a position where our prosecution division is saying, "If
we had more resources, we would prosecute more cases".
Q37 Dr Pugh: Could you give us a
note on the breakdown? Obviously there are two sorts of resources,
resources for the trial and resources for processing the investigation.
There were in the case I mentioned earlier considerable resources
being allocated to the investigation as opposed to the prosecution.
Can you give us a breakdown of that?
Sir Leigh Lewis: I will try and
do my best to give you a breakdown of both of those things.[2]
Q38 Dr Pugh: A point made is that, on
living together fraud—this is absolutely outrageous, when
people see somebody claiming full benefits when they are supported
by a person they are living with or a partner who is fully in
work—you seem somewhat reluctant to bring these to court
compared with other sorts of cases, or your success rate is rather
lower. Is that deterring you from following these sorts of cases
up?
Sir Leigh Lewis: I think it is
something that you might want to hear a little bit from my colleague,
Neil Couling, on. It is described as perhaps one of the Department's
enduring urban myths. Actually, we are just as determined to bring
living together cases to court as any other sorts of cases. Nearly
2,000 were referred to our prosecution division in 2006-07 and
the numbers have been increasing sharply.
Mr Couling: The Report itself
is more questioning of following up living together.
Q39 Dr Pugh: It is the NAO, not you?
Mr Couling: In 2005-06 we achieved
2,300 sanctions on living together cases. Last year that increased
to 3,782 and to December we had already exceeded that at 3,800.
Because it is an area of loss, I am directing my officers at it.
It is very tricky. I was in an appeal tribunal on Thursday and
we lost the case because it is an area where it is quite difficult
to prove in some cases, but we are going after it quite hard.
1 Ev 23-25 Back
2
Ev 21 Back
| 
