Select Committee on Public Accounts Minutes of Evidence

Examination of Witnesses (Questions 20-39)



  20. Yes, but the fact of the matter is that the sight test is unreliable. One assumes that large numbers of people have evaded being demonstrated to be culpable simply by the fact that you have been using a sight test rather than a more objective chemical test.
  (Mr Broadbent) No, I do not think that is the case. A sight test is not unreliable. The chemical test is the only test which can produce a definitive answer in the case of doubt but in many cases a sight test is quite definitive: it is red diesel. That is why we put in the dye so we can inspect it with a sight test.

  21. Is it not the case that your officers are expected to use the chemical test rather than the sight test, or have been?
  (Mr Broadbent) What I expect of my officers is to use judgement. This seems to me absolutely at the heart of it. If you look at the breakdown of the way testing is run, in the case of intelligence-led tests 87 per cent of the tests are full chemical tests. In the case of roadside tests, 29 per cent are full chemical tests. That reflects guidance which says that the full chemical test is the best case where you want to produce absolutely objective evidence but clearly if you are wanting to do large numbers of tests in a motorway layby, you want to do a volume of tests. We invite our officers to use their judgement. We say to them that the full chemical test is the preferred test, particularly where you want to get complete certainty, but I do expect my officers to use judgement, that is what they are paid for.

  22. I assume that these are tests at the side of the road, 20,000 tests in 2000-2001. Eleven per cent were found to be misusing the fuels. I assume this is based largely on sight tests which are unreliable according to paragraph 3.9. I deduce and it seems to me that you do not, that we are understating the amount of misuse and underestimating the extent to which this is going on as a result of using sight tests.
  (Mr Broadbent) I do not quite accept that. Thirty eight per cent of those tests were intelligence led and in the case of the intelligence-led tests 87 per cent were full chemical tests. The hit rate of those chemical tests was very high. In the case of the random testing, which was the balance of those tests, 29 per cent of the tests were chemical tests because you are doing a random test and the hit rate was four per cent, which is about the level of fraud. It shows a pattern of activity which is broadly consistent with people judging the right test in the circumstances. What I do agree with you is this. It is very clear that the pattern of fraud is changing. It is becoming more sophisticated and as it becomes more sophisticated we are saying to our officers that they should use the chemical test more. The circumstances where it is appropriate will increase in number and there you are absolutely right. That is correct.

  23. May I ask about the so-called developing strategy described in paragraph 3.13? It is a year since this was used, January 2001, 14 months now, in the North East of England so people in the County of Durham no doubt know something about it. Why has so much time elapsed since then? As far as I can seem there are some startling figures: 44 per cent of traders showed some evidence of duty evasion. This is a staggering amount more than one would expect from the other testing you have been doing. The question I really want to ask you is: why are we now almost in March 2002 and this has not been rolled out?
  (Mr Broadbent) That was not an isolated test. We are doing those exercises now fairly regularly and as I left the office I brought out of my press cuttings one from Friday about vehicles seized in diesel tax blitz, 87 vehicles during a six-day blitz on diesel tax dodgers in the North East. That is an exercise which has just finished running. We are running exercises fairly frequently. The reason we do not focus all our activity into this effort is very simple, which is that the ability to build the intelligence profile which drives them is quite difficult. Intelligence is a very scarce commodity and it takes considerable time to build up the profile which allows you to run one of these exercises with these very high hit rates. A lot of effort goes in behind the scenes to make them work.

  24. I am not quite sure how difficult it is. It seems to me that a lot of information must be available from VAT returns of traders up and down the country as to how they are trading, how much they are buying, how much they are selling. A lot of fraudulent invoices must be being issued. I am quite sure that large amounts of intelligence are available with the use of tachographs and other information. What I was asking you was why we are now almost in March 2002 when the pilot—I presume it was a pilot—was done well over a year ago? When are you going to roll this out to the rest of the country. It seems to me that in West Yorkshire or the South West or anywhere else the risk of detection is substantially lower, is it not?
  (Mr Broadbent) That was not an isolated exercise. We have run these exercises on a regular basis since then and indeed ran one as recently as last week.

  25. So there has been a series of pilots. Why have you not rolled it out nationwide?
  (Mr Broadbent) No, sorry, it is not a series of pilots, it is a series of exercises and the limiting factor on these exercises is not because we see them as pilots, the limiting factor is our ability to create and define a sufficiently high quality intelligence profile to run an exercise, which is not a simple task.

  26. You are still not answering my question. I suppose it is easy to alight on a particular word such as "pilot". Let me ask you again. When do you intend to roll this out nationwide?
  (Mr Broadbent) I am sorry, I am not trying to avoid answering your question, I can only say what I said.

  27. I deduce you are not intending to do it nationwide. What you intend to do is to hotspot areas around the country as intelligence emerges.
  (Mr Broadbent) We will always tend to hotspot areas, not because they are focusing on one particular part of the country, but because the nature of an intelligence profile is that it is much more efficient to build it up in part of a certain region and then blitz that region using the whole intelligence package over a four, five or six-week period and then when you have done a region you move to another region. That is a function of the way the intelligence is built up. It is not saying we are never going to do the South East we are going to do the North West. It is simply a more effective way of building an intelligence picture. I am sorry, I did misunderstand.

  28. I am not sure that we have communicated very adequately in relation to that particular matter but there is something else I want to move onto, which is the action taken against fraudsters once they have been detected. Once again this appears to lack a degree of coherence or at least strategy since a range of different actions is available but it is not clear to me that they are being operated in a strategic way from reading this report. I want to ask you for example about paragraph 3.18, "Prosecution is a costly option and Customs only take such action if the likelihood of securing a custodial sentence is high". Why a custodial sentence? You only prosecute if a custodial sentence is likely.
  (Mr Broadbent) We have a number of criteria which guide prosecutions and one of them is the expectation of a severe penalty, custodial or very significant financial penalty.

  29. This report says custodial.
  (Mr Broadbent) Certainly we would prefer custodial. If we are going to prosecute we would prefer to believe that there is going to be a custodial sentence.

  30. It says, "Customs only take such action if the likelihood of securing a custodial sentence is high". Are you saying that is not exactly accurate?
  (Mr Broadbent) What our guidance actually says is that there has to be a good prospect of a custodial sentence. I am afraid I do not recall the exact wording.

  31. In terms of taking action once fraud has been detected, what is your strategic objective then? What objective do you set yourself? It seems to me that you are necessarily recovering the total amount of duty which has been lost.
  (Mr Broadbent) Our strategic objective is to reduce the amount of fraud and prosecution is one tool in that. We try to judge whether to prosecute or not by reference to our aim and our aim is to reduce the amount of fraud.

  32. If we take the largest category of action which is taken, which is seizure and restoration, what exactly does that involve?
  (Mr Broadbent) That would involve effectively that the vehicle is sealed and restored on payment of the duty and normally a civil penalty and usually a restoration charge as well. You might finish up paying 600 or 700.

  33. Do you feel that mechanism which is used substantially more than any other actually is recovering for the public purse the amount of fraud which is being undertaken by the perpetrators?
  (Mr Broadbent) It recovers the duty in that case. What I would say about these figures—and it is quite important I make this point—is that they relate to a period nearly two years ago. They relate to a period when the pattern of fraud was somewhat different. It tended to consist of large numbers of small-scale misusers. There is no doubt that the pattern of fraud has now changed and I entirely accept that as the pattern of fraud changes we should expect to see the pattern of sanctions change. The pattern of sanctions here is probably appropriate for the activity at the time. It does not mean to say that I necessarily think it is the right pattern today.

  34. What do you think is the right pattern today then?
  (Mr Broadbent) I think you would expect to see first of all more prosecutions because there is more commercial fraud, more large-scale organised fraud. You would expect to see more assessments because that is the way you hit the cashflow of fraudsters and the way you really attract large sums of money back into the public purse. We would need to look at tougher penalties and there is reference to this in the PBR. We would need to look at having a more rigorous vehicle seizure policy, for example, which has proved to be very effective in the case of tobacco.

  35. There is a distinction between criminal action and civil action, is that right?
  (Mr Broadbent) Yes.

  36. I am not sure I entirely understand it. It seems to me that you can take a fixed civil penalty, which presumably is negotiated between yourselves and the alleged perpetrators.
  (Mr Broadbent) The civil penalties are fixed penalties, a 250 fine; it is not negotiated, it is a fixed penalty for all excise offences. You are right that there is a distinction between civil penalties and criminal action.

  37. Two questions arise from that. Let me just focus on one for a second. How do you feel about the levels of penalty which are currently available to you? I notice that there is quite a variation elsewhere in the European Union.
  (Mr Broadbent) It is fair to say that we are giving thought to this. I do not necessarily believe that the right answer is simply to increase the fixed penalty. I do think there are some areas where we may want to apply higher penalties than we currently apply. I instanced vehicle seizure a moment ago. In some cases we have these powers, for example we do have powers to seize vehicles.

  38. But then you restore them to the owners, do you not?
  (Mr Broadbent) No, we seize them and keep them in some cases. The cases in which we can seize and keep them are somewhat bound about by tribunals and the Human Rights Act. We must keep our punishments proportionate. This is a problem we have recently seen in immigration. If a punishment is not proportionate, then you start to lose cases in the courts.

  39. Would common sense not indicate that the right thing to do is to criminalise as many people as possible who are perpetrating this kind of fraud and that that would argue for prosecutions whether or not custody was a likely outcome?
  (Mr Broadbent) These are judgements which have a wide public interest. I am not certain that criminalising large numbers of small-scale users, as opposed to putting perpetrators behind bars, is an approach which would be effective in reducing fraud, which ultimately is my job. It would certainly be prohibitively expensive. It takes 165 man days to prosecute one person; that is half a man year. If you start talking about 4,000, 5,000, 6,000 offences, that is a very large number of staff. That is not the main point. Prosecution is a very expensive tool, but the acid test has to be its effectiveness in reducing fraud and I am not sure that criminalising large numbers of people who are small-time fraudsters would necessarily achieve that.

  Jon Trickett: I do not know about that. That is to some extent a political judgement. Certainly taking criminal action where custody might not be the outcome might well be appropriate in significant numbers of cases and yet that does appear to be ruled out. We have had an interesting exchange and I shall listen to the rest of the debate with interest as well.


previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 18 July 2002