Examination of Witness (Questions 80 -
WEDNESDAY 14 NOVEMBER 2001
80. Would it be fair to describe what you are
trying to do with the management as trying to effect a culture
(Mr Broadbent) Yes.
81. How are you assessing the impact of the
changes you are making in terms of the outcome you want in management?
(Mr Broadbent) There are certain practical measures
I can look at which are mostly to do with the flow of information.
We are spending quite a lot of time investing money and time into
management accounts and controls so that decisions get triggered,
moved up the chain, discussed and you can see this happening.
You can measure the number of times these things happen and you
can see how often discussions get triggered. I can ask myself
the questionI have the benefit of this report which others
did not haveif this happened now, would I know? That is
an interesting test to ask yourself. There is a second and much
more difficult part of the answer, which I am afraid is really
to do with the nature of change management. If you are seeking
to change attitudes and behaviours you have to know very clearly
where you are coming from and where you are getting to but make
a rather subjective judgement about what point in the road you
82. Can I just turn to one specific recommendation
from the report? Roques said that the Department has been reluctant
to publish estimates of losses from fraud and other leakage in
each tax segment. He has just stressed to us that he thought this
was one of the most important areas on which he made recommendations.
Are you now preparing to make such estimates in each of the tax
(Mr Broadbent) I said to the PAC the week before last
that the Government will be publishing in the forthcoming PBR
estimates for excise taxes. It will also be publishing an estimate
for part of VAT, a particular set of frauds which we are confident
we understand and can measure. It will not be publishing more
estimates yet because we need to understand more about VAT before
we can do that.
83. Why are the responses partial?
(Mr Broadbent) In the time available actually the
response is quite full, but that is obviously my opinion. The
partial response in relation to VAT stems from the fact that VAT
is a significantly bigger and more complex tax than the excise
duties. The issues with VAT go into understanding not just fraud
in the sense of criminals attacking the system, of which there
is an element and that is the element we understand best, but
also error, law. The custom and practice relating to VAT is extremely
complex and the way that is applied is very complex. Legitimate
avoidance is quite large legal avoidance. Unpacking that as an
analytical exercise is a very considerable undertaking which we
have started but I cannot at this stage say when it will be completed.
84. Do you accept in principle that full publication
of estimated losses in all those areas is desirable?
(Mr Broadbent) Provided one can get to estimates which
are meaningful, yes. This is quite an important proviso.
85. Do you think you can get there?
(Mr Broadbent) I do not know yet. What we are going
to do in a PBR is publish estimates in the area of VAT where we
have got there, which I hope is indicative of our approach to
86. Are estimates going to be done in terms
of tobacco and alcohol as well?
(Mr Broadbent) They will be for tobacco and alcohol
and we shall also publish a technical paper showing how we have
reached all our estimates so they are transparent.
87. Paragraphs 4.3 to 4.5 of Roques are really
a depressing catalogue of what can only be called incompetence.
How can your organisation explain the fact that although you had
warnings in 1994 and 1995 from NIS and in 1996 from your own internal
audit review, your organisation could not get to grips with outward
diversion until 1998?
(Mr Broadbent) Are you referring to sections 4.3 and
88. And 4.5.
(Mr Broadbent) Sections 4.3 and 4.4 refer to bootlegging;
4.5 refers to drawback fraud.
89. I am talking about September 1994.
(Mr Broadbent) Outward diversion is the 1994-95 figure
you gave me, which is section 4.6.
90. I do beg your pardon, I am referring to
pages 11 and 12, the executive summary. September 1994 there is
a note from an investigating officer to a manager in policy which
talks about "The fraud is extraordinarily easy . . . The
risks . . . are considerable". Then on 7 December 1995 we
go slightly higher up the food chain, do we not? A briefing paper
to the head of NIS, reports, ". . . the last twelve months
or so has seen a growing acceptance that the various transit systems
for the movement of excise goods provide inadequate controls in
the single market". In April 1996 an Internal Audit Division
review of the Movement of Goods Liable to Excise Duty "highlighted
significant control weaknesses in the Holding and Movements system".
Those are three specific points in the story where the report
concludesand I use the words of Mr Roques"I
find it surprising that this audit, which can only be described
as damning, was not acted on more quickly". So just to repeat
the question: in 1994, 1995 and 1996 you are on notice that there
is a major problem and in relation to outward diversion fraud
jack-all happens until 1998 of any significance. Why?
(Mr Broadbent) Two things are worth saying. The first
thing, at the risk of repeating myself, is that the critical question
here is at what point an organisation should know something is
happening which is different from what was going on before. This
is a big challenge for any organisation, including commercial
organisations. In 1994 and 1995, the years you are talking about,
the actual level of fraud was less than 0.1 per cent of alcohol
revenue. Yes, there were the first signs at operational investigation
level of the fraud beginning. But the interesting question we
have to ask ourselves is at what point it is fair to say the organisation
should have started piecing the bits together, an organisation
should have known at the top that something was going on? My own
judgement about thatand I do not think either the National
Audit Office or in particular John Roques disagreedis that
that is probably some time around about the time of the IAD report
to which you referred in 1996, at which point as a percentage
of alcohol revenues the fraud was still quite small, but there
was at that point a document at high level which said there was
a control failure here. That leads me into my second point, that
to me the most important issue is not in fact that this damning
report was not acted on, it is that it was acted on. The report
made 14 recommendations, they were all implemented and it did
not deal with the problem. Why did it not deal with the problem?
Because most of those recommendations are of a sort not untypical
from an organisation which tends to deal in process, which said
we have to promote this sort of practice, we have to remind people
of that thing they have to do. It did not say there is a problem
and you need to change the way you do things fundamentally to
tackle it. That is very much to do with the point I was making
earlier, that the challenge here is to bring an organisation to
move more quickly to something which is happening in the external
environment and take action. The IAD report was implemented; it
did not actually deal with the problem. What dealt with the problem
at the end of the day was a series of actions, quite practical
actions, which were started to be taken in around about April
1997, which is about 12 months after the IAD report and that is
what I regard as the real gap which this organisation should answer
91. How is the organisation going to "answer
for it", to use your words?
(Mr Broadbent) The best thing you can do in answering
is to understand what has happened, acknowledge your mistakes
and learn from them, which personally is what I do.
92. In responding to the earlier bits of evidence
in September 1994 and 7 December 1995 that the report sees fit
to draw attention to, your defence to this is that it was such
a small percentage of the total it was not a developed problem,
it was not something the report should have been drawing attention
to. You are saying that it was early days, it was unreasonable
to take action. But Mr Roques seems to take a contrary view, does
(Mr Broadbent) I do not think he takes a contrary
view. The bit you are quoting sets out a chronology of events
and he is quite right to point out that there were early signs,
as early as 1994 and 1995, that there was a problem.
93. That is 1994 and no serious or effective
action kicks in until 1998. I could be getting this wrong, but
four years is quite a long time for you to work out that there
is a bit of a problem and provide solutions to cracking that problem.
Twelve months, maybe. Eighteen months, fair enough. But four years?
Surely the job of your organisation is to think strategically
and think ahead. If there was an embryonic problem in 1994, are
you seriously inviting us to believe that it should take four
years to get it sorted?
(Mr Broadbent) Where I disagree with you is in your
assumption that the first warning sign is tantamount to full knowledge
and full understanding of the problem. That is just not how organisations
work. No organisation in the world, public sector or private sector,
works that way.
94. "The fraud is extraordinarily easy
. . . The risks . . . are considerable". That is in the note
in September 1994. I am not quite sure of your interpretation
of the English language, but "The fraud is extraordinarily
easy ... The risks ... are considerable". That is a problem.
(Mr Broadbent) It is clear that within the organisation
documents existed which at a working level identified the fraud.
The question you are asking me, is: what is it sensible to expect
the organisation to do? I have to go back to asking what the knowledge
of the organisation was. The existence of a piece of paper deep
down in the organisation is not the same as the organisation knowing.
The challenge the organisation faced was drawing together quite
a lot of disparate sources of information to a senior level where
strategic action could be taken. That did take too long, but it
did not take four years. The point at which it could be fairly
said that the organisation should have known something was seriously
astray here was not as early as I should have liked.
95. I do not wish to be unhelpful or badger
the witness but this really will not do. In September 1994, you
say it is a piece of paper buried down in the depths of the organisation,
"The fraud is extraordinarily easy . . . The risks . . .
are considerable". Then on 7 December, in a briefing paper
to the head of NIS, for heaven's sake, we have the quote, "...
the last twelve months or so has seen a growing acceptance that
the various transit systems for the movement of excise goods provide
inadequate controls in the single market". This is the head
of NIS and it is talking about what has been monitored in the
preceding 12 months. I really must invite you to reconsider your
answer. You are saying that the organisation did not know the
problem. Those two pieces of paper to which Roques actually draws
attention are there in September 1994, December 1995 and nothing
is done effectively to tackle it until 1998.
(Mr Broadbent) We just disagree what
it means to say to "the organisation". I wish it were
the case that the organisation
96. Hang on. This is to the head of NIS in December
1995. That presumably is suitably high enough up for you, is it?
(Mr Broadbent) No, not necessarily. Not everything
the head of NIS has will have crossed the Chairman's desk or the
Board's desk or the director's desk.
97. Absolutely extraordinary statements. So
when the head of NIS is advised that for 12 months there has been
a growing acceptance that there is a problem, inadequate controls,
you do not think that is serious enough for anyone in your organisation
to take seriously.
(Mr Broadbent) I did not say that. What I said was
that the question is: did the organisation know. The failure here
98. So what is the answer to that question.
(Mr Broadbent) The failure here was not that people
had information necessarily and did not act on it. What Roques
goes on to say a bit later on is that the first the Chairman knew
of this was an operation at the end of 1995. There was then the
IAD report in 1996 and I do agree that at some point in this process
it is fair to say the organisation, at the top, the senior people,
should have known. What I do not accept is that the fact there
was one reference by one person earlier than that is tantamount
to saying the organisation knows. That is the complexity of a
99. It is not one reference, it is two references;
I have given you the two references. Let me ask one question in
relation to these findings of the Internal Audit Division in April
1996. When were they brought to the attention of the Board? First
question. Secondly, when were they brought to the attention of
Her Majesty's Ministers?
(Mr Broadbent) They were not brought to the attention
of Her Majesty's Ministers at all. This is an audit report within
the organisation and I do not think many if any audit reports
go to Ministers.