Taxation logo taxation mission text

Since 1927 the leading authority on tax law, practice and administration

The right target?

CHRIS CHADBURN reviews the National Audit Office report on HMRC’s management of civil tax investigations


  • Targeted checks on small and medium-sized enterprises.
  • Local Compliance yield increased in 2009/10.
  • Effectiveness of referring cases to a central HMRC unit.
  • Learning from the Liechtenstein disclosure facility.
  • Average penalty paid is 21% of the tax due.

In 2009/10, HMRC collected £435 billion in tax, according to the National Audit Office’s report HMRC: Managing civil tax investigations. In the previous year, the tax gap was estimated to be £42 billion.

The department believes £15 billion of this emanated from fraud, evasion and criminal attack deriving from the non‑compliance of approximately 5% of individuals and 7% of businesses, the equivalent of around 1.5 million individuals and 300,000 businesses.

The report examines ‘whether the department is managing its civil investigations efficiently and effectively by making best use of resources to maximise levels of compliance and tax revenue’.

It received little coverage in the national press; publication on Friday 17 December could have been significant, as could the fact that, in many ways, the message was very positive from HMRC’s perspective, with large increases in yields achieved during difficult times.

The Guardian was one paper that did highlight the report. Its news item had the headline Tax cheats avoiding fines when caught and included the comment that ‘critics say the deals mean there is little deterrent for tax cheats who will have nothing to lose other than the tax they should have paid’.

The NAO report has no criticism of the long-standing fundamental idea that the penalty system should reward those who disclose their misdemeanors and co-operate with the Revenue.

Settlements in substantial cases will usually comprise not only the tax that should have been paid but also penalties and interest. Professional fees will need to be paid and there will often be considerable expense for the taxpayer in terms of time and energy.

The report is in four parts, each of which will be considered in the rest of this article.

Managing resources

The first part looks at how HMRC manages its civil investigation resources.

The NAO notes that the department is working on a new business plan for 2011 to 2015 aimed at bringing in an additional £7 billion a year by 2014/15, as announced in the October 2010 spending review. This is a tough ask representing an 82% increase on the 2009/10 results.

Although the report provides little detail on the plan, we are told that HMRC are ‘using the new customer-centric strategy to tailor its interventions to specific customer groups and focus efforts where they will have the greatest effect’. Despite the management-speak this does not represent any strategic change. The idea of matching your resource to the perceived risk is not new.

What will make the difference are initiatives such as the targeted check of the adequacy of the records of 50,000 small and medium-sized enterprises (SMEs) a year. HMRC believe that ‘poor record keeping is a problem in around 40% of all SME cases (five million)’.

I am sceptical of the accuracy of this claim but it is clear that these checks will be in an invaluable tool in identifying targets for investigation.

Directorates shows the various directorates within the enforcement and compliance business area.

HMRC use yield as the key measure of effectiveness in this area. See Yield. The Local Compliance yield shot up in 2009/10 largely as a result of assessing the validity of backdated VAT repayment claims. Excluding these reduces the annual increase from 59% to 14%.

Nevertheless the overall increase in yields and cost ratios is impressive, particularly bearing in mind reductions of 8% of expenditure and 16% in the workforce. The figures generally significantly exceed the targets.

The NAO makes two major criticisms. First, the focus on yield omits other key performance aspects, in particular:

  • the full range of activities, for example the impact of preventative work;
  • the effect on taxpayer behaviours or the overall level of compliance; and
  • the balancing of generating short-term revenues with longer term improvement in taxpayer compliance.

Second, the NAO considers that management information is not sufficiently robust to determine either the cost effectiveness of different types of enforcement activity or productivity.

For example:

  • the figures of local civil investigation of fraud teams revealed unexplained wide variations in performance;
  • Specialist Investigations did not collate similar information as their teams dealt with both avoidance and evasion.

According to the report, performance hubs lack a ‘focused end-to-end view of enforcement and compliance work; for example, by establishing key input and output metrics on the key stages of investigations.

Directorates should establish better metrics on the performance of their teams and ensure data can be aggregated consistently to inform board reporting’.

HMRC have been spending £1 billion a year on tackling non-compliance. Looking at the Specialist Investigations example above, it is astonishing that adequate measures of performance are not available because investigators cover different types of work.

Mixed case loads emerged following the development of the civil investigation of fraud teams from 2005. HMRC may have heard of a very simple ‘metric’ to solve these sorts of problems for relevant staff: time sheets.

Referring cases

The second part of the report is called ‘Referring cases for civil investigation’. In 2007, HMRC introduced a mandatory central referral system where Local Compliance officers send cases to a central unit where the tax at risk is likely to exceed £10,000.

The case may be adopted by Criminal Investigations or Specialist Investigations, Civil Investigation of Fraud or returned to the originating officer to pursue.

The main objective is to ‘provide a consistent and standardised mechanism for the referrals and escalation of cases’. See, for example, Enquiry Manual EM0303 and Employer Compliance Handbook ECH6065 for details of the system. The availability of the referral data also helps HMRC identify characteristics of known evaders and use feedback from completed investigations.

The report reveals that in 2008/09 caseworkers conducted enquires on 265,000 selected cases and that suspected fraud and evasion above £10,000 was identified in some 4,500 of these (1.7%). After filtering, 2,420 cases were referred on and 480 adopted (20%).

The NAO highlights the lack of systematic feedback in terms of the reasons for rejection and the quality of the referrals. Following a substantial reduction in the volume of referrals contributory factors were identified as:

  • caseworker disillusionment following rejections;
  • the ‘mandatory’ system was not being used across HMRC; and
  • yield targets can act as disincentives.

Managing civil investigations

The third part of the report focuses on the cases where HMRC are investigating under Code of Practice 9, i.e. where serious fraud is suspected and no criminal charges are being pursued for the tax offences under investigation.

Civil investigations shows key statistics for Specialist Investigations and Local Compliance in 2009/10. Specialist Investigations will typically be handling cases where a settlement of more than £500,000 is anticipated.

Local Complicance will typically be looking at cases from £75,000 to £500,000. Local Compliance shows variations of the time taken to conclude for Local Compliance civil investigation of fraud teams for 2009/10.

The NAO rightly highlights reducing the time being taken to deal with cases as key to improving performance and estimates that if investigations were concluded within the target time of 18 months, additional revenue of between £30 million and £60 million a year could be generated.

Clearly the present targets do not reflect the realities and the NAO refers to the following suggestions for improvement from focus groups of investigators:

  • two-tier disclosure reports tailored to the size and complexity of the investigation;
  • working more closely with agents to provide greater upfront advice; and
  • a more focused approach to settling cases.

It is also noted that HMRC’s approach to the Liechtenstein Disclosure Facility ‘suggests opportunities to reduce the time taken to conclude civil investigations of fraud’.

The main factor mentioned is the closer working with agents to agree technical and process issues before submission of the report that has resulted in an average of six months from registration to settlement in the early cases.

It is worth looking at the main features of the facility to establish how HMRC could be operating their civil investigation of fraud procedures in future.

First, incentives to disclose and co-operate are made. These were broadly that no tax etc. was sought pre-5 April 1999, and the penalty was fixed at 10%.

These are certainly generous and I suggest that at the very least the future general civil investigation of fraud offering should include a clear guaranteed maximum penalty where an acceptable report is delivered on time.

Second, there are fixed deadlines for the submission of the disclosure and settlement. The present one-size-fits-all report deadline of six months is unsatisfactory but difficult deadlines can always be met if sufficient resource is thrown at the report and HMRC are flexible and commercial.

For example, where overseas bank statements cannot be obtained in a reasonable time, HMRC take a view on the information that you do have. If there is a will to have clear, realistic and achievable deadlines then some simple measures of size and complexity can easily be developed.

Third, there is a clear and concise disclosure and payment package. The standard report regarding the Liechtenstein Disclosure Facility already exists. The investigator could add a handful of bolt-ons covering the appropriate major headings to provide a comprehensive framework to fit any case.

Finally, the operation of a light touch review. As the facility booklet says ‘we expect the majority of disclosures to be accepted without in-depth enquiry’. I believe HMRC’s Liechtenstein desk has been responsive and helpful.

HMRC are generally adopting a more collaborative approach with agents when tackling enforcement and this seems to be absent at least in the early stages of the civil investigation of fraud procedure. There needs to be a clear focus on the major risk areas as soon as is practical.

It seems to me there is also an issue in terms of what investigators consider at an early stage the agent can deliver. There is a substantial element of lack of co-operation.

The average time to review the disclosure report and conclude the investigation is approaching one year which indicates the general quality of reports may be well below HMRC expectations.

This kind of approach needs sticks to help the carrots work. So if the report is seriously incomplete there has to be a clear and fixed penalty uplift or prosecution.

The NAO also questions the basic effectiveness of the civil investigation of fraud procedure in deterring tax evasion and points out that the level and trends of unprompted voluntary disclosure and taxpayer co-operation would help demonstrate the position.

HMRC indicate a 20% level of non co-operation with their investigations which does not sound like a resounding confirmation that all is well. HMRC do not routinely collate data on voluntary disclosures.

HMRC reported on the effectiveness on the procedures in 2009 and highlighted that it was important for potential fraudsters to be clear about the penalty policy and for prosecutions to underpin the civil procedure.

In my experience, actual fraudsters have little idea of the potential penalties and even less of the potential for prosecution. I suspect potential fraudsters are usually in the same boat. The major deterrent is the likelihood of getting caught.

Maximising impact

Part four of the reports looks at maximising the impact of civil investigations. It refers in passing to prosecutions and sets out the basis of the old and new penalty regimes. Penalty levels shows the penalty levels for the civil investigation of fraud cases settled in 2009/10.

The NAO says that there has been no analysis of the factors leading to different penalty levels, for example by assessing the distribution of penalties by size and age of cases or the level of co-operation and disclosure.

Certainly previous NAO reports have been clear that penalty levels were generally too low and this was a factor in the introduction of the new regime from April 2008.

The 2009/10 figures match my experience in thepast few years where early disclosure and a good report resulted in a penalty of around 25%, even in very serious cases.

Enquiry Manual EM 6080 states that the abatement for seriousness in fraud cases should not normally exceed 15% and more serious cases dealt with by Specialist Investigations would often justify abatements well below this. So a typical Specialist Investigations case should merit a penalty of at least 25% even with full abatements for disclosure and co-operation.

Fifty per cent of cases result in penalties of less than 20% and, on the basis of what the NAO is saying, HMRC cannot easily explain this. Likely factors are:

  • no or low culpability taxes included;
  • inappropriate cases registered for investigation;
  • the department taking what it can in ‘can’t pay’ situations; and
  • deals struck to avoid litigation or to clear old cases.

HMRC do not routinely monitor the collection of tax etc. from investigations. Management information was limited but of the £58 million of 2008/09 settlements passed over for collection that could be traced, 16% was unpaid.

The final point raised is one that is regularly flagged up by the NAO: HMRC’s civil investigation successes receive little publicity. Time and again it has been pointed out that overseas tax authorities adopt approaches to improve the visibility of their work, driving home the message that evasion is unacceptable, that evaders are regularly brought to book and are heavily penalised.

Examples include:

  • publicising activity including the publication of compliance programmes;
  • tax fraud alerts; and
  • publicity of successful cases.

The increased use by HMRC of campaigns is welcomed without any detailed analysis of their effectiveness. Reference is made to ‘some success’ in the offshore investments campaign in an earlier report.

There has certainly been criticism from advisers of HMRC’s ineffective publicity in relation to the offshore disclosure facilities.


Although the report is generally positive, there are some strong criticisms and greater value should be obtained by HMRC through ensuring:

  • resources are deployed closer to risks;
  • cases are settled quicker; and
  • the deterrent effect is strengthened.

The department is taking action in most of the areas highlighted in the report, but the impression is that this is on the back of the NAO’s comments. If HMRC are going to get anywhere near their investigation targets leading up to 2014/15, they need to make changes along these lines.

Their autumn 2010 staff survey report indicates that the department’s employees do not have much confidence in its ability to lead and manage change. These are the positive responses (‘agree’ or ‘strongly agree’) to the following statements:

  • When changes are made in HMRC they are usually for the better: 9%
  • I have the opportunity to contribute my views before decisions are made that affect me: 16%
  • I feel that change is managed well in HMRC: 11%
  • Overall I have confidence in the decisions made by HMRC’s senior managers:11%
  • I feel that HMRC is managed well: 12%

Will the NAO now be getting stuck into some of these issues?

Chris Chadburn is a founder member of venntax, an independent association of tax advisers. He can be contacted on 01323 736375 or by email

back to top icon