Select Committee on Public Accounts Thirty-Fifth Report


The Committee of Public Accounts has agreed to the following Report:




1. All taxes are subject to the risk of fraud and evasion. Indirect taxes are particularly vulnerable because they are applied at the point of transaction and there is a limited timescale to identify and collect the taxes.[1]

2. The estimated overall revenue lost due to all types alcohol duty fraud in 1999-2000 was 800 million, 7 per cent of the total tax receipts collected from alcohol, mainly through imports into the UK without payment of duty.[2] The Comptroller and Auditor General's Report examined the known losses between 1993 and 2000, following the introduction of the Single European Market in 1993, of alcohol duty worth 884 million from outward diversion fraud (Figure 1).[3] He took into account the Report, in December 2000, of an independent investigation into Customs handling of these losses, by John Roques, a former senior partner of Deloitte and Touche.



Figure 1: Amounts of alcohol duty lost from outward diversion fraud between 1993 and 2000


Type of loss over the period 1993-2000


Cases where duty suspended goods destined for overseas excise warehouses were diverted onto the UK market without payment of duty.



Cases where duty suspended goods destined for overseas excise warehouses were diverted onto markets outside the UK, without payment of duty in those countries. Under European Rules, these duty losses are accounted for in the country where the warehouse was located.






3. At the time of our hearing in October 2001, Customs were in the process of producing estimates of and strategies for tackling fraud, including smuggling, against each of the main indirect taxes. They had completed this exercise for tobacco smuggling, which was their highest priority and where the estimated loss for 2000-01 was 2.9 billion.[4] Since then, in Tackling Indirect Tax Fraud published in November 2001, Customs set out their estimate of revenue loss in the alcohol and oils sectors and from missing trader fraud, what they were doing in these areas and possible new initiatives being considered. In total, the estimate of tax not collected as a result of fraud was between 6.4 billion and 7.3 billion (6.3 per cent-7.1 per cent of the tax collected in 2000-01), and this did not include a full assessment of fraud against VAT where work was still ongoing.[5]

4. On the basis of a Report by the Comptroller and Auditor General, we looked at how Customs were developing a better corporate approach to tackling fraud and the way they responded to the loss of revenue from alcohol diversion.[6]


5. In the light of this examination, the Committee draws four overall conclusions.

There are wide differences between rates of duty across the European Union, so fraudsters have strong incentives and Customs face an uphill struggle to collect all the revenues due. Nevertheless, fraud totalling 6.4 to 7.3 billion excluding most of VAT where further work has to be done, results in a serious loss to public finances. This sum could make a major contribution to improving the health service, education or other priority areas.

Assessing the scale of the problem is an important first step in developing anti-fraud strategies and operational measures, and in securing sufficient resources to stem and then reduce the loss of revenue. The estimate could rise substantially once Customs complete their assessment of VAT fraud, since this tax represents some 57 per cent (58 billion) of the duties and taxes they collect.

Customs have begun to develop strategies for tackling fraud, starting with tobacco. Since our hearing they have outlined their strategies for dealing with alcohol, hydrocarbon oils and VAT missing trader fraud. They also propose to develop outcome targets and measures to reduce levels of fraud on all duties and taxes, following their initiative on tobacco. These will not only help them to assess their own effectiveness, but also inform the allocation of resources overall and between tax streams. This information will provide an important insight into Customs' performance, and should be published annually.

Our examination of alcohol diversion frauds showed significant deficiencies in Customs' management information and intelligence systems for identifying emerging frauds early enough, and in their arrangements for getting the balance right between facilitating trade and protecting the revenue. There were weaknesses in controls over decisions on letting frauds continue, and on when to give indemnities to warehouse keepers, so as to obtain evidence and secure convictions. Customs are now well underway in implementing the recommendations of the Alcohol & Tobacco Fraud Review, the Roques Report and the Comptroller and Auditor General's Report, which should help them address these weaknesses for the future.

6. Our more specific conclusions and recommendations are as follows.

On developing a better corporate approach to fraud

  (i)  As Customs tighten up on fraud in one area, so determined fraudsters will move on to test other aspects of the system. So, even though Customs have addressed the problem caused by outward diversion, latest estimates show that the level of alcohol fraud is still some 750 million. The main threat now comes from inward diversion and smuggling. Customs have identified the need to improve their intelligence information systems, to help them respond quickly to emerging fraud before losses escalate.

  (ii)  The limitations of their revenue forecasting process were a key reason why the rising level of alcohol diversion was not identified sooner. Customs have started to improve their forecasting but the systems should be further developed, drawing on best practice in other sectors such as market research.

  (iii)  Alcohol diversion fraud, as with other frauds such as tobacco and hydrocarbon oils smuggling, occur across the European Union and across the borders of Member States, especially where there are significant differences in rates of duty. Customs should encourage the European Commission to implement the system for the exchange of information on transactions more quickly than the five years currently planned.

  (iv)  Underpinning the delayed response to rising levels of alcohol diversion was a lack of joined-up working. Law enforcement did not communicate well enough with other parts of Customs, leading to a delay of 9-12 months in tackling weak controls and significant loss of revenue. Customs have since re-organised their operations, and introduced new oversight arrangements at Board level. They should also review the scope for better internal management information systems.

On responding promptly and effectively to fraud

  (v)  Customs face difficult judgements in balancing the need to secure the revenue with the desirability of facilitating trade and reducing the burden they impose on business. In implementing the risk-based controls that followed implementation of the Single Market, Customs misjudged this balance, with the loss of significant sums of public money. The measures they have now taken to strengthen control over warehouses, owners of goods and hauliers, including wider application of guarantees, have tightened the regime against outward diversion fraud.

  (vi)  Partly because Customs did not focus soon enough on seeking to confiscate cash from fraudsters, the amount confiscated by the courts between 1994 and 1998 totalled only 23 million. They have now taken action to pursue money from the outset, and thereby improve their ability to confiscate assets from warehouse keepers, owners of goods and hauliers involved in fraud. The Proceeds of Crime Bill will help further. Customs need to develop a more active prosecutions policy and increase the level of asset confiscations, to send clear messages to potential fraudsters that rigorous action will be taken.


7. At the time of our hearing in October 2001, Customs were in the process of producing estimates of and strategies for tackling fraud, including smuggling, against each of the main duties and taxes they collect (Figure 2). They had produced an estimate of, and a strategy for, tackling tobacco smuggling, which was their highest priority and where the estimated loss for 2000-01 was 2.9 billion.[7] We looked at the level of losses due to alcohol fraud and the need for a strategy to reduce them, the adequacy of Customs' management information systems for spotting frauds early enough, and organisational barriers to effective anti-fraud activity.

Figure 2: Alcohol duty formed 6.6 per cent of total revenue collected by HM Customs & Excise in 2000-01

(a)  The scale of losses of revenue on alcohol goods and the need for a strategy to reduce them

8. The Comptroller and Auditor General's Report focused on the circumstances surrounding the known losses between 1993 and 2000, following the introduction of the Single European Market in 1993, of alcohol duty worth 884 million from outward diversion fraud (Figure 1).[8]

9. At the time of the Comptroller and Auditor General's Report, the estimate for the overall revenue lost due to fraud against alcohol duty in 1999-00 was 800 million. This amounted to 7 per cent of the total tax receipts collected from alcohol, comprising 500 million for spirits, 200 million for beer and 100 million for wine. The main element was alcohol imported into the UK without payment of duty. Customs published full estimates later in November 2001, in the Pre-Budget Report, which estimated the annual loss for 1999-00 at 750 million.[9]

10. Customs said that at 7 per cent of receipts the level of fraud was excessive and that the figure for spirits, at 18 per cent, was manifestly too high. As Customs had toughened up their counter fraud methods, determined fraudsters had move on to test other aspects of the system. As a result, the nature of the fraud had changed from outward diversion to inward diversion and smuggling.[10]

11. Since early in 2001 Customs had taken a series of measures in response to outward diversion frauds (Figure 3), but they accepted that they needed a strategy to reduce the scale of fraud. They were committed to publishing a strategy on alcohol later in 2001, and were working on oils and VAT as well. On VAT in particular there was a lot to do because they were dealing with error and avoidance as well. The emphasis would shift from outputs to outcomes, for example by setting targets for reductions in the level of fraud as they had done for tobacco.[11]



Figure 3: Key measures taken in response to the Roques Report

a more rigorous approach to the approval of warehouses;

tighter registration procedures for warehouse keepers and the owners of goods;

improved information on the holding and movement of excise goods where the duty has not been paid;

improved exchange of information with other European Union Member States;

increased checks on warehouse keepers' compliance with holding and movement regulations;

tighter controls on hauliers;

considering the use of fiscal marks on alcohol.



12. Following our hearing, in Tackling Indirect Tax Fraud the Government announced further action to reduce revenue losses in the next stage of the strategy (Figure 4).[12]



Figure 4: Key additional measures announced in November 2001, in Tackling Indirect Tax Fraud


increase the rate of checks carried out on spirits consignments when they enter the UK under duty suspension;

improve intelligence to help target additional checks on suspicious consignments, helping limit the impact on legitimate traders and hauliers;

strengthen controls on bonded warehouses within the UK;

wide consultation on the costs, benefits and practicalities of introducing a tax stamps system for spirits, designed to assist in the identification of smuggled products;

if it is decided to proceed with the tax stamps system, new sanctions for those found to be dealing illegally with un-stamped goods;

setting Customs an objective of achieving gradual reductions in the market share of illicit spirits over the next three years.


(b) The adequacy of Customs' management information system

13. The creation of the Single European Market in 1993 brought with it the removal of restrictions on the movement of goods between European Union Member States, with the aim of making trade easier.[13] In February 1994, Customs told the Committee of Public Accounts, in evidence on the Comptroller and Auditor General's Report on the Department's 1992-93 accounts,[14] that there was at that time no sign of increased fraud as a result of the Single Market arrangements.[15]

14. By 1995, Customs became aware that fraudsters were taking advantage of weak controls over goods movements to and from excise warehouses to divert alcohol on to UK and overseas markets without paying the required duty. Customs knew that the scale of alcohol diversion fraud was growing and by 1996 the losses were significant. In a Report to the Department in July 1996 the National Audit Office highlighted the risks to the revenue from Customs' approach to handling traders and excise warehouses. The Report was followed up in the National Audit Office's Reports on the Department's 1996-97 and 1997-98 Accounts. A Customs' Internal Audit Report of April 1996 also highlighted weaknesses in the movement system and the increased numbers of fraud cases. In response, Customs took action to tighten up the weaknesses in the system and the level of alcohol diversion fraud fell significantly after 1997-98.[16]

15. Customs produce forecasts twice a year of receipts from all the taxes and duties they collect, and comparisons with outturn, which provide another potential indicator of the loss of receipts through fraud. Over the period from 1991-92 to 1997-98, these comparisons showed a fall in consumption of alcohol and a widening gap between forecast and actual receipts. However, Customs lacked the information to monitor accurately their performance either in collecting the revenue or in tackling fraud. In his Report, John Roques saw the need for forecasting of Customs' revenue to be improved as an aid to timely detection of problems affecting revenue streams.[17]

16. Customs told us that the while the gap between forecast and actual receipts was large in relation to alcohol, it was very small when taken against receipts overall. The earlier estimating techniques for consumption started from revenues in the previous year, rather than estimating the pattern of consumption. At the time, Customs and the trade sector had interpreted the shortfall in revenue as a switch from spirits to wine, and had failed to consider that fraud might have been a factor. They accepted that they needed a clearer idea of the total amount of tax to be collected, and that one of the problems on alcohol had been the absence of a clear view on the total market. They had since been working on expenditure surveys and in collaboration with the industry.[18]

17. More widely, Customs had taken action to improve their information on the movement of goods where duty had not been paid. They planned improvements in their intelligence systems, including the use of electronic knowledge management. In addition, there was now commitment and funding from the European Commission to improve intra-Community exchange of information although this could take up to five years, because it was less of a priority in some Member States. In Tackling Indirect Tax Fraud, published after our hearing, the Government announced action to improve Customs' information and intelligence.[19]

(c) Organisational change

18. Eighty per cent of the fraud took place over a two year period in 1996-97 and 1997-98. Customs told us that by late 1995 and early 1996 they should have been aware of what was happening. However while one part of Customs was pursuing frauds vigorously, other parts were not making use of that knowledge. In terms of law enforcement they reacted quickly, but the information was compartmentalised. As a result, the corporate response had been slower than it should have been, by about 9-12 months. Had they acted more quickly, they would have saved considerable sums of money.[20] The lesson they had drawn was not just to rely on controls in a dynamic and fast changing area, but to look outside and respond rapidly. They needed to create a flexible organisation, so that when the external environment changed again and fraud rose they could deal with it. Customs were now in a better position to respond to changing circumstances and were much more conscious of the risks.[21]

19. Another factor leading to increasing loss of revenue was that over the period concerned, their manpower was falling year on year. In retrospect some parts of the excise regime had been under-resourced. This they were now trying to reverse, for example by redeploying additional assurance officers to strengthen the excise holding and movements system. Customs were however more concerned about skills than the volume of resources, and a major objective was to upskill their staff, particularly in management and IT.[22]


20. In response to the Single Market, Customs adopted a risk-based approach, removing many of their physical checks of goods and largely passing operational control over the system to the traders themselves.[23] However, the resulting control framework meant that the system for the movement of alcohol from excise warehouses quickly became a target for fraud. The weaknesses included:

the difficulty of detecting fraudulent export documents;

the reduced level of Customs' checks of excise warehouses;

incomplete records of the movement of goods;

non-compliance with regulations by traders; and

failure to impose appropriate sanctions.[24]

21. Against this background, we looked in more detail at three main aspects of Customs' approach to tackling alcohol diversion frauds:

Controls over warehouses, warehouse keepers, goods owners and hauliers;

Balancing surveillance of frauds to secure convictions against disruption of frauds to protect the revenue; and

Prosecution of fraudsters and confiscation of assets.

(a) Controls over warehouses, warehouse keepers, goods owners and hauliers

22. In response to management concerns and the reports from the National Audit Office and Internal Audit, Customs took action to strengthen controls, leading to a significant fall in the level of alcohol diversion fraud after 1997-98. These actions included requiring warehouse keepers and the owners of goods to be registered, and putting more obligations on warehouse keepers to control their loads more carefully. The Roques Report went further, and recommended:

a more rigorous approach to the approval of warehouses;

tightening the registration procedures for warehouse keepers and the owners of goods;

increasing the checks on warehouse keepers' compliance with holding and movement regulations; and tightening controls over hauliers.[25]

23. Excise goods were moved from the warehouse under guarantees, which could be provided by the warehouse keeper, the haulier or the owner of the goods. When the Single Market was created in 1993, guarantees were compulsory for all intra-European Union movements of duty suspended alcohol and tobacco. Customs could call on these guarantees where the principal failed to pay an assessment of duty owed and revenue was therefore lost. Guarantees covered the amount of duty that was likely to be suspended in a single consignment, and were intended to cover "occasional" or minor losses, such as the theft of a lorry load where the owner was unable or unwilling to pay the duty owed. They were not intended to underwrite revenue lost through organised diversion fraud or prevent the smuggling of excise goods. The Roques Report recommended that guarantees should be more readily invoked by Customs, who should treat irregularities as absolute offences which were dealt with immediately.[26]

24. The guarantee system had been loosened during the mid-1990s as part of a continuing debate within Customs to balance revenue security with staff costs. It had now been tightened up again. They had widened the system to ensure that no load moved without somebody giving a guarantee, whether it was the warehouse keeper, the owner or the haulier. They encouraged warehouse keepers to get others to give the guarantees wherever possible, and made it clear to the warehouse keeper that if they chose not to do so, then Customs would hold them responsible.[27]

25. Guarantees were largely set in percentage terms relative to the value of goods moving through, up to 1 million. They were not an absolute guarantee against fraud, because they were not big enough to cover certain organised fraud and because the person who gave the guarantee might not be the one at fault. However, Customs felt that 1 million was sufficient to deter the casual law breaker and significantly encourage rigorous controls by warehouse keepers.[28]


(b) Balancing surveillance to secure convictions against disruption of frauds

26. A key approach of Customs' National Investigation Service to tackling diversion frauds between 1995 and 1998 was to identify the principals behind the frauds and collect sufficient evidence to secure their conviction. The main method was to identify suspect consignments and allow fraudsters to move goods from excise warehouses under observation (known as "letting loads run"). Selected consignments would be followed and, when enough evidence was obtained, the perpetrators would be arrested.[29]

27. However, at the time the alcohol diversion frauds were occurring, Customs did not assess whether "letting loads run" was a more effective way of tackling the frauds than disruption. The intended objective was to protect the revenue by convicting fraudsters and recovering losses, but it inevitably led to the build up of arrears of duty. The main alternative was to disrupt fraudulent activity as it was discovered, which protected potential revenue loss but was less likely to lead to conviction of the perpetrators.[30]

28. In 1996 Customs became concerned at the conduct of the investigations because of the high level of revenue losses. But it was not until 1998 that they took action to bring outstanding investigations to a conclusion. In 1999 Customs introduced a system of authorisation levels within the National Investigation Service to ensure that decisions were taken by an appropriately senior person, before consignments of diverted or smuggled goods were allowed to proceed without interception. Thus the Chief Investigation Officer, who headed the National Investigation Service, was required to authorise all cases where the extent of fraud was likely to be over 1 million, with approval at Director-level (i.e. a Board Member) required where losses were likely to be over 3 million.[31]

29. The Roques Report concluded that "letting loads run" was necessary in individual cases to ensure that sufficient evidence could be assembled to mount a successful prosecution. However, the Report recommended that the level at which Director approval was required to put revenue at risk be reduced to the level at which their approval was required to write-off or remit debts (i.e. 1 million). This recommendation has been implemented and should reinforce the need to consider significant operational decisions at the strategic (i.e. Board) level.[32]

30. Customs accepted that the judgements on the balance between disruption and letting loads run were not properly made in every case. It was not possible, even with the benefit of hindsight, to say conclusively whether more revenue was lost as a result, but they thought that it had been. They had introduced a new set of controls, a new case management system for investigations, which from the earlier stages of the case identified what the issues were and triggered a series of control levels requiring more and more senior attention, depending on how much money was at stake. As a result, they were much more confident about the judgements.[33]

(c) Prosecution of fraudsters and confiscation of assets

31. The co-operation of warehouse keepers was an essential part of "letting loads run", both in alerting Customs to possible fraudulent consignments and in letting consignments leave the warehouse under surveillance. However, some warehouse keepers were given indemnities, removing their liability for duty evaded in those cases where the liability would properly have fallen to them. Warehouse keepers were indemnified for an estimated 53 per cent (355 million) of the 668 million revenue evaded through diversion on to the UK market, and this was a major factor in their failure to raise assessments on 59 per cent (397 million) of the total identified losses. In some cases, the indemnities were written, but most were verbal or implied. The indemnities meant that warehouse keepers had no liability for the duty evaded.[34]

32. At the time they were given, Customs did not have a clear policy on the use of indemnities. From 1999, however, any proposal to indemnify a warehouse keeper had to be approved by senior policy staff in Customs. And Customs issued guidance to staff in April 2000 on the circumstances in which indemnities could be provided to traders. Customs confirmed that indemnities were still available where needed, for example where they were faced with fraud and needed the collaboration of a warehouse keeper. However, the Chairman of Customs would now be consulted about any indemnity in excess of 1 million, and no indemnities had been given in 2001.[35]

33. Between 1995 and 1998, the National Investigation Service dealt with 72 cases of outward excise diversion fraud, and Customs staff in regional offices investigated a further 58 cases. Over 100 successful prosecutions were made, with a total of some 200 years in sentences passed on those involved in the frauds, and Customs recovered around 25 million of goods and 23 million in cash.[36]

34. Customs told us that within the 156 prosecutions, 13 individuals had received sentences of 7 years or more. The value of cash confiscated, at 23 million, reflected what Customs could identify in people's ownership at the time. Customs were now much clearer in pursuing money right at the beginning of an investigation, which was not done before. As a result, the value of cash confiscated during 2001 was more than half as much again as the sum raised in the whole of the period 1994-1998.[37] In addition, the Proceeds of Crime Bill would be an important piece of legislation in speeding up the procedures to confiscate the cash.


1   C&AG's Report, para 2; Q2 Back

2   C&AG's Report, para 3.6 Back

3   ibid, para 1 Back

4   ibid, para 2 and Figure 1 Back

5   ibid, para 4; HM Customs and Excise Tackling Indirect Tax Fraud (not printed here) Back

6   C&AG's Report, Losses to the Revenue from Frauds in Alcohol Duty (HC 178, Session 2001-02) Back

7   C&AG's Report, para 2 and Figure 1 Back

8   C&AG's Report, para 1 Back

9   ibid, para 3.6 and Figure 7; Qs 4, 14, 34-40, 117; HM Customs and Excise Measuring Indirect Tax Fraud, Tables 2.3, 2.5, 2.11 (not printed here) Back

10   C&AG's Report, para 3.10; Qs 19-20, 112-113 Back

11   C&AG's Report, para 16; Qs 4-6, 87, 93-94 Back

12   HM Customs and Excise Tackling Indirect Tax Fraud, paras 4.23, 4.25 (not printed here) Back

13   C&AG's Report, para 3 Back

14   HC 11-XII, 1993-94 (Appropriation Accounts, Volume 12: Classes XVIII to XX, 1992-93) Back

15   19th Report of the Committee of Public Accounts, HM Customs and Excise: Account Matters (HC 216, Session 1993-94), para 36 Back

16   C&AG's Report, paras 6, 11-12, 14, 1.18-1.21 and Figure 4 Back

17   ibid, paras 3.4-3.7 and Figure 6 Back

18   Qs 21-24, 45-49, 63-66, 80, 106, 109-11, 114-116, 146 Back

19   C&AG's Report, paras 16, 1.24, 1.28, 3.13-3.14; Qs 8-9, 150; HM Customs and Excise Tackling Indirect Tax Fraud, para 4.23 (not printed here) Back

20   Qs 15-18, 48, 80, 146, 157-160, 166 Back

21   Q48 Back

22   C&AG's Report, paras 1.28, 3.13; Qs 78-79, 131-133 Back

23   C&AG's Report, paras 3, 1.2-1.8 Back

24   ibid, para 11 Back

25   C&AG's Report, paras 12, 16; Qs 16, 53-54, 62 Back

26   C&AG's Report, paras 1.14-1.15 Back

27   Qs 25-27, 95-99 Back

28   Qs 25-27 Back

29   C&AG's Report, paras 8, 2.3 Back

30   ibid, paras 8, 25, 2.3-2.4 Back

31   ibid, para 2.7 Back

32   ibid, paras 2.3, 2.7 Back

33   Qs 10-11, 50-52, 84, 100-101, 121-122, 163-164, 172, 182-184 Back

34   C&AG's Report, paras 2.8-2.9 Back

35   Qs 28-29, 187-193 Back

36   C&AG's Report, paras 10, 2.2, 2.10, Figure 5 Back

37   Qs 83, 139-141, 143-145, 147; Ev 20, Appendix 1 Back

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 17 May 2002