The tax fraud gap – 2021 edition

16th September 2021

  • The Tax Gap Attributable to Fraud was at least £15.2bn in 2019/20


  • At least 43% of tax losses arise from fraudulent behaviour

Executive Summary

HMRC’s annual estimate of non-compliance, “the Tax Gap” is regarded by the department as an important indicator of their long term performance and is used as a tool in developing HMRC’s strategy towards compliance. It is listed as a key performance indicator in HMRC’s annual report under their primary objective, “collecting revenues due and bearing down on avoidance and evasion”.

It is a broad measure of non-compliance defined by HMRC as “the difference between the amount of tax that should, in theory, be paid to HMRC, and what is actually paid”. What should be collected is the total amount of tax due under the law.

There are a number of reasons why any taxpayer may be non-compliant. This can range from the taxpayer making a mistake on their tax return, to not knowing about a liability to pay tax, through to criminal attempts to defraud the Treasury through filing false claims or hiding income.

HMRC break down the Tax Gap by eight “taxpayer behaviours”, which appear to be related to the department’s internal arrangements, namely: (1) criminal attacks; (2) evasion; (3) hidden economy; (4) avoidance (which does not include Base Erosion and Profits Shifting (BEPS) or tax planning); (5) legal interpretation; (6) non-payment; (7) failure to take reasonable care; and (8) error.

In this paper, we propose an alternative categorisation of non-compliance based on the three behavioural categories defined in law: Fraud, Negligence, and Honesty.

Fraudulent non-compliance is a deficiency of tax where the underlying behaviour is dishonest. Dishonest tax behaviour can lead to criminal charges, but can also be addressed through civil and administrative penalties.

Negligent non-compliance arises from carelessness or a failure to pay due care and attention on the part of the taxpayer with regard to a tax liability. Negligence, where detected, results in a tax liability and civil penalties.

Honest non-compliance can arise if a taxpayer makes an honest mistake in the filing of a tax return which is not caused by negligence or dishonesty.

If a taxpayer has ended up as non-compliant through an honest mistake and that mistake is discovered, they will have to pay any taxes due, but may not suffer any penalties, although there are some strict liability cases where a penalty may be levied.

We believe that defining the Tax Gap in these terms would provide a number of advantages. Firstly, as a matter of principle, it is right that a measure of non-compliance with the law should be defined in terms that are recognised by law.

Secondly, these three categories of behaviour are easily understood by the public. If HMRC were to present their Tax Gap in these terms, we believe that the public’s understanding of the nature of non-compliance would be significantly improved.

Finally, using these categories would provide clarity in developing HMRC’s strategy, as there is a risk that what is not recognised as fraud will not be treated as fraud.

For the purposes of this paper, we have assessed HMRC’s behavioural categories as found in the Tax Gap and find that many easily fall under the legal definitions of fraud, negligence or honesty.

We find that when categorised in this way, fraudulent behaviour accounts for at least £15.2bn of the Tax Gap – 43% of the total Tax Gap and 2.25% of the total amount of tax due according to HMRC.

In order to reach this figure, we added the sum total of tax lost to what HMRC term, “Criminal Attacks – £5.2bn”, “Evasion – £5.5bn”, “Hidden Economy – £3bn” and “Avoidance – £1.5bn”.

The categorisation of Avoidance as fraud arises because, unlike HMRC that consider only “taxpayer behaviours”, we consider the behaviour of tax professionals and conclude that this approach places tax avoidance in the fraud category.

Although tax avoidance is generally thought of as “legal” activity, it is clear that avoidance as defined by HMRC, which is an incidence of non-compliance arising from the use of a scheme, developed by tax professionals, which seeks to “exploit” the tax system through “contrived or artificial” transactions that have no commercial purpose, should properly be defined as arising from dishonest behaviour on the part of the professionals who design and market the schemes.

Our interpretation is supported by the new definition of tax fraud adopted by HMRC in the latest edition of Measuring tax gaps:

“Any deliberate omission, concealment or misinterpretation of information, or the false or deceptive presentation of information or circumstances in order to gain a tax advantage.”

More detail on our approach to this issue is provided in the main body of this paper.

However, some of the behaviours used by HMRC do not easily fall into one of the proposed categories. For example one of the largest components of the Tax Gap is “legal interpretation” which comprises £5.8bn of the Tax Gap. This is where a taxpayer disputes HMRC’s interpretation of the law. These disputes could easily arise from either dishonest or honest behaviour.

Furthermore, HMRC does not count most international tax avoidance, characterised as “BEPS”, in their Tax Gap calculations. It is clear, from HMRC’s publications in this area, that much of what HMRC categorise as BEPS arises from fraudulent conduct. If all of this is taken into account, it would not be unreasonable to assume that the Tax Fraud Gap is at least £20bn.

Even at the lower estimate, which only includes categories clearly falling under the definition of fraud, tax fraud is a far larger problem than fraud impacting other areas of public finance. For example, the latest estimates of fraud in the benefits system show that fraud accounts for £6.3bn of potential losses – before any recoveries (HMRC’s figures are after compliance efforts).1

Given that the behaviours we identify are grounded in law, it should be relatively easy for HMRC to publish a more detailed estimate of the amount of tax lost to fraud, having assessed the amount of fraud included in categories such as Legal Interpretation, BEPS and Non-Payment. We recommend that HMRC do this in their next update to the Tax Gap.

HMRC’s Tax Gap

The Commissioners for Her Majesty’s Revenue and Customs (HMRC) was established through the merger of the Commissioners of Inland Revenue (IR) and Her Majesty’s Customs and Excise (HMCE) by the Commissioners for Revenue and Customs Act 2005.

Section 5 provided that the Commissioners shall be responsible for “the collection and management of revenue” for which the Commissioners of Inland Revenue and the Commissioners of Customs and Excise were previously responsible.

The management of revenue includes policing the tax system and conducting criminal investigations with a view to prosecution. While the former Customs and Excise prioritised both the collection of revenue and the punishment of offenders, the former Inland Revenue considered their primary objective to be the collection of revenue and not the punishment of offenders.

Against this background, HMCE published estimates of the Tax Gap in HMCE-administered taxes known as the indirect tax gap from 2001 in technical papers published alongside each Pre-Budget Report (PBR): Measuring Indirect Tax Fraud (Nov 2001), Measuring indirect tax losses (Nov 2002), Measuring and Tackling Indirect Tax Losses (Dec 2003, Dec 2004), published with the 2001, 2002, 2003, and 2004 Pre-Budget Reports.

Following the 2005 merger, HMRC continued to publish the indirect tax gap. At the same time, a broader estimate of tax losses of all taxes administered by HMRC was developed for internal use. After some resistance, HMRC made this information public after the journalist Richard Brooks sought the estimates under the Freedom of Information Act.

HMRC began regularly publishing estimates of the Tax Gap in all HMRC-administered taxes (including direct taxes) alongside the 2009 PBR. It estimated the tax gap to be around £40 billion in 2007-08 and identified the eight underlying taxpayer behaviours: (1) criminal attacks; (2) evasion; (3) hidden economy; (4) avoidance (which does not include Base Erosion and Profits Shifting (BEPS) or tax planning); (5) legal interpretation; (6) non-payment; (7) failure to take reasonable care; and (8) error. These eight behaviours remain the foundation for HMRC’s analysis of non-compliance today.

Protecting Tax Revenues detailed HMRC’s approach in using analysis of the tax gap to tackle the drivers of the Tax Gap and described the range of measures that HMRC were taking to reduce the Tax Gap. The behaviours are critical to this approach. According to the report:

“Analysis of the underlying behaviours that drive the tax gap is useful as by identifying these behaviours HMRC can most effectively develop a targeted approach, prioritising operational resources and identifying where policy solutions are required.”2

HMRC’s stated reasons for measuring the Tax Gap are as follows:

“The tax gap provides a useful tool for understanding the relative size and nature of non-compliance. This understanding can be applied in many different ways:


It provides a foundation for HMRC’s strategy — thinking about the tax gap helps us understand how non-compliance occurs and how we can address the causes and improve the overall health of the tax administration system


our tax gap analysis provides insight into which strategies are most effective at reducing the tax gap


although the tax gap isn’t sufficiently timely or precise enough to set annual targets or manage detailed operational performance, it provides important information which helps us understand our long-term performance.


The tax gap also provides important information to the public on tax compliance, creating greater transparency in the tax system.”3

Fraud, Negligence and Honesty

As far as the law is concerned, there are only three types of behaviours that lead to non-compliance: Fraud, Negligence and Honesty. Every incidence of non-compliance can be said to arise from one of these three behaviours depending on the knowledge, abilities and circumstances of the taxpayer or tax professional involved.

Fraud, cheating and dishonesty

In tax and indeed other areas of law, the terms ‘fraud’, ‘cheating’ and ‘dishonesty’ mean essentially the same thing and can be used interchangeably.

According to Justice Hardy’s widely-accepted definition of the common law offence of Cheating the Public Revenue in R v Less:

“The common law offence of cheating the Public Revenue does not necessarily require a false representation either by words or conduct. Cheating can include any form of fraudulent [or] dishonest conduct by the defendant to prejudice, or take the risk of prejudicing, the Revenue’s right to the tax in question knowing that he has no right to do so.”4

Dishonesty is also the essence (or essential requirement) of the relatively new criminal offence of fraud under the Fraud Act 2006, which applies to tax and other areas of law.

UK tax legislation also contains criminal offences that criminalise dishonesty. These include: fraudulent evasion of income tax (section 106 of the Taxes Management Act 1970); fraudulent evasion of VAT (section 72(1) of the Valued Added Tax Act 1994); and fraudulent evasion of excise duty (section 170(2) of the Customs and Excise Management Act 1979).

Each Act also provides for civil penalties for fraud, where the essential requirement is dishonesty.5

In 2021 HMRC adopted the following definition of fraud in their “Measuring tax gaps” report.6 The definition is as follows:

Any deliberate omission, concealment or misinterpretation of information, or the false or deceptive presentation of information or circumstances in order to gain a tax advantage. Tax evasion is fraud.

This was a significant departure from previous editions of Measuring tax gaps and other publications, which always stated that tax fraud is tax evasion. For example, in Measuring tax gaps 2020 edition fraud is defined as simply, “Deliberate, dishonest evasion of tax.”7

The broader definition now used by HMRC begs the question, what else, other than “evasion” should be considered fraudulent behaviour?

As we demonstrate in this report in relation to the Tax Gap behaviours, what HMRC consider to be “Tax Avoidance” and some “Legal Interpretation” could easily be described as behaviour arising from “any deliberate omission, concealment or misinterpretation of information, or the false or deceptive presentation of information or circumstances in order to gain a tax advantage.”


The classic common law definition of negligence was set out by Baron Alderson in Blyth v Birmingham Waterworks as follows:

“Negligence is the omission to do something which a reasonable man, guided upon those considerations which ordinarily regulate the conduct of human affairs, would do, or doing something which a prudent and reasonable man would not do.”8

In the context of tax non-compliance negligence is defined as carelessness in section 95 of the Taxes Management Act 1970 or failure to take reasonable care under Schedule 24 of the Finance Act 2007.

In tax disputes, the courts have broadly followed the test set out in Blyth when considering penalties under these provisions of these Acts. For example in Anderson v HMRC Judge Berner stated:

“The test to be applied … is to consider what a reasonable taxpayer, exercising reasonable diligence in the completion and submission of the return, would have done.”9


Honesty is simply the opposite of dishonesty or fraud or cheating. It is a subjective assessment based on an individual’s knowledge at the time. According to Lord Nicholls in Royal Brunei Airlines v Tan:

“Honesty has a connotation of subjectivity, as distinct from the objectivity of negligence. Honesty, indeed, does have a strong subjective element in that it is a description of a type of conduct assessed in the light of what a person actually knew at the time, as distinct from what a reasonable person would have known or appreciated.”10

The role of professional advisers

The role played by professional advisers is crucial to the understanding of the nature of non-compliant behaviour on the part of a taxpayer.

Paragraph 18 of Schedule 24 to Finance Act 2007, which deals with the liability of a taxpayer to penalties for negligence or fraud where professional advisers are acting on his behalf, was considered in Hanson v HMRC. Judge Cannan confirmed the well-established law and practice thus:

“What is reasonable care in any particular case will depend on all the circumstances. In my view this will include the nature of the matters being dealt with in the return, the identity and experience of the agent, the experience of the taxpayer and the nature of the professional relationship between the taxpayer and the agent. In my view, if a taxpayer reasonably relies on a reputable accountant for advice in relation to the content of his tax return then he will not be liable to a penalty under Schedule 24.”11

The corollary of the highlighted principle is that a taxpayer using a tax avoidance scheme, which is invariably devised and implemented by the professional enablers, to misrepresent or conceal his tax liability in a tax return submitted to the Revenue is more likely to do so honestly than negligently or fraudulently.

Tax evasion, tax avoidance and tax mitigation

The terms tax evasion, avoidance and mitigation are commonly used to describe a range of behaviours associated with both compliant and non-compliant tax behaviour.

The terms have no universally accepted definition, and are used in different and sometimes opposing ways in different contexts. HMRC have particular definitions they use, which will be set out later on in this report. In this section we look at meaning of these terms in law.

Tax evasion

Tax evasion is when a taxpayer dishonestly fails to make a tax return when they had a legal requirement to do so, or when a taxpayer makes a false tax return by failing to declare all of their income. It is tax fraud and punishable under the common law offence of cheating the public revenue. In either case, the key issue is the behaviour and knowledge of the taxpayer in their dealings with the tax authority.

As set out in R v Mavji:

“This appellant was in circumstances in which he had a statutory duty to make value added tax returns and to pay over to the Crown the value added tax due. He dishonestly failed to do either. Accordingly, he was guilty of cheating HM The Queen and the public revenue.”12

In R v Hudson the taxpayer was convicted of cheating the public revenue by sending in false accounts relating to their farming business which deliberately understated the profits of the business. At the court of appeal, Goddard CJ stated:

“We think that the offence here consisted of sending in documents to the inspector of taxes which were false and fraudulent to the appellant’s knowledge … for the purpose of avoiding the payment of tax. That is defrauding the Crown and defrauding the public.”13

Avoidance and tax mitigation

On a proper analysis of the law, Tax Avoidance could be defined as a form of tax fraud by professional advisers that design, market, implement and otherwise facilitate the use of tax avoidance schemes in which the taxpayer using an individual scheme may or may not be complicit.

Tax avoidance is distinguished from tax evasion by the use of a tax avoidance scheme created and marketed by professional tax advisers.

In a tax avoidance scheme, a taxpayer reduces their tax liability by entering into an arrangement, or a series of transactions which has the effect of making a taxpayer appear to suffer a reduction in their taxable income when in fact no real reduction has taken place.

As stated above, it is well established in tax law that a taxpayer should be able to reasonably rely on professional advice in the field of tax, and if they do so then they should not be considered to be negligent (let alone dishonest!) even if the tax return they make on the basis of that advice turns out to be wrong.

It follows from this that a tax payer that submits an incorrect tax return based on the use of a tax avoidance scheme is more likely to be behaving honestly rather than dishonestly, or negligently (depending on whether their reliance on the professional advice in question was “reasonable”).

However, just because the taxpayer may be acting honestly by entering into a tax avoidance scheme, does not mean that all participants in a scheme are acting honestly.

R v Charlton, Cunningham, Kitchen and Wheeler14, was a case which started as a standard enquiry into a taxpayer’s return, but became a criminal investigation after the Inland Revenue raided the premises of the accountants that had devised the tax avoidance scheme used by the taxpayer. It became the longest running prosecution by the Inland Revenue and ended with the conviction of a number of tax professionals for cheating the public revenue for their roles in devising, marketing, implementing and otherwise facilitating the use of tax avoidance schemes. According to Lord Justice Farquharson:

“The case for the prosecution was that Charlton had devised a dishonest, tax-avoidance scheme for the benefit of some of the firm’s clients and that the Appellants were involved with the implementation of the schemes or the concealment from the Revenue of the existence of the fraud.”15

The schemes in Charlton were designed to reduce taxable income in the UK by shifting profits using artificial transactions to intermediaries in Jersey. The type of scheme would readily be described as Base Erosion and Profit Shifting (BEPS) schemes. As set out by Farquharson LJ:

“It was the case for the Crown that the accounts presented to the Revenue by the United Kingdom companies were false in that by using Charlton’s scheme to transfer part of their profits to the Jersey companies they were not disclosing the full extent of the profits they had made. It was this lack of disclosure which formed the basis of the false representations alleged in the indictment. Each of the Appellants was charged in the relevant counts with cheating the Revenue by ‘… falsely representing that the apparent purchases (by the United Kingdom company) from (the Jersey company) were bona fide commercial transactions’.”16

Tax avoidance is usually not considered to be fraudulent behaviour by HMRC because the tax system as it currently operates is designed for the relationship between the Revenue and the taxpayer (who is usually an honest participant in the scheme). Where HMRC discover tax avoidance their usual approach is to amend the taxpayer’s return and deny them the benefit of using the scheme.

Where a tax assessment is appealed by the taxpayer, the fraudulent nature of tax avoidance scheme is obscured, because the dispute is between the participating taxpayer and the Revenue to which their tax advisors or the scheme operators are not parties.

This, combined with the fact that HMRC vary rarely prosecute dishonest tax advisors, has led to some confusion over the true nature of tax avoidance, a point neatly summarised by a leading criminal barrister Robert Rhodes in his commentary on the Charlton case:

“Amongst professional tax advisers, alarm and concern have been expressed at the approach of the Revenue and the conduct of the case. It has been argued that there is a general move to ‘blur’ the ‘very clear’ distinction between legal tax avoidance and illegal evasion. However, it might well be suggested that the distinction is not and has never been as clear as many professional advisers (and their clients) would like to believe. Where avoidance arrangements are wholly artificial and have no substance then clearly it is and always has been open to the Revenue and the courts to consider whether they are in fact ‘devices to cheat the public revenue’.

Moreover, the terms ‘tax avoidance’ and ‘tax evasion’ have been created by the legal and accountancy professions as convenient generic terms to distinguish what is legal from what is illegal, and the fact that they have also been adopted by the courts should not blind us to what they actually are.”17

It is important to understand that tax avoidance is separate from tax planning or tax mitigation, which is often confused with avoidance in common usage.

Tax planning is properly defined as when a taxpayer reduces their taxable income by making a real expense that takes advantage of a real tax benefit provided for by Parliament. This could be for example investing in plant an machinery that attracts capital allowances or putting money into an ISA. As set out by Lord Templeman in CIR vs Challenge, where the concept of tax mitigation was first developed:

“Income tax is mitigated by a taxpayer who reduces his income or incurs expenditure in circumstances which reduce his assessable income or entitle him to reduction in his tax liability. In tax mitigation … the taxpayer’s tax advantage is not derived from an ‘arrangement’ but from the reduction of income which he accepts or the expenditure which he incurs….18

Analysing HMRC’s Tax Gap behaviours using the legal concepts of fraud, negligence and honesty

HMRC’s Tax Gap is defined as “the difference between the amount of tax that should, in theory, be paid to HMRC, and what is actually paid”. It is therefore a measure of non-compliance and all non-compliance can fall under the three behavioural categories found in law and set out above – Fraud, Negligence or Honest non-compliance. Indeed, these three categories of behaviour are the bedrock of how the courts approach tax law, both civil and criminal.

HMRC presents the Tax gap as arising from eight different “taxpayer behaviours”. The use of the term “taxpayer behaviours” is significant because it underscores HMRC’s focus on the taxpayer and the failure to consider the behaviour of professional advisers, which is critical to tax avoidance. The eight behaviours are: error; failure to take reasonable care; evasion; hidden economy; criminal attacks; avoidance; legal interpretation; and non-payment.

Some of these behaviours are clearly dishonest, negligent or honest, whereas others can cover more than one category. In this section we go through each of the behaviours contained in the Tax Gap analysis used by HMRC to see into which legal category the behaviour should fall.

Tax Evasion, Hidden Economy and Criminal Attacks – £13.7bn

HMRC define “evasion”, “hidden economy” and “criminal attacks” as three separate behaviours. According to the latest Tax Gap publication, HMRC considers “evasion” to be “where registered individuals or businesses deliberately omit, conceal or misrepresent information in order to reduce their tax liabilities.”

This focus on “registered individuals or businesses” distinguishes “tax evasion” from “hidden economy” which is defined as where “whole sources of income have not been declared to HMRC for tax purposes”.

HMRC define “criminal attacks” as “co-ordinated and systematic attacks on the tax system” including “smuggling goods such as alcohol or tobacco, VAT repayment fraud and VAT Missing Trader Intra-Community (MTIC) fraud.”

In reality, all three behaviours would be considered tax evasion by the general public with the category “criminal attacks” specifically covering tax evasion which relates to the work of the former Customs and Excise.

Indeed all three are considered tax evasion by HMRC in all other publications apart from the Tax Gap. For example, the joint HMRC & HMT document Tackling tax avoidance, evasion, and other forms of non-compliance, states:

“Tax evasion is always illegal. It is when people or businesses deliberately do not declare and account for the taxes that they owe. It includes the hidden economy, where people conceal their presence or taxable sources of income.”19

As our discussion of tax evasion above demonstrates, all these forms of evasion found in the Tax Gap are from a legal perspective cheating the public revenue by a taxpayer, and so can easily be categorised as fraudulent or dishonest behaviour.

Avoidance – £1.5bn

HMRC’s definition of avoidance defines tax avoidance in terms of “schemes” which often involve “contrived” or “artificial” transactions designed to “exploit” the tax system. The definition provided in Measuring tax gaps 2021 is as follows:

“Avoidance involves bending the tax rules to try to gain a tax advantage that Parliament never intended. It often involves contrived, artificial transactions that serve little or no purpose other than to produce a tax advantage. It involves operating within the letter but not the spirit of the law.”

This is a from the previous edition of the Tax Gap which described avoidance as “exploiting the tax rules” rather than “bending them”.20

Tax planning is clearly differentiated from avoidance in the HMRC’s tax gap analysis. As set out in Measuring tax gaps 2021:

“Tax avoidance is not the same as tax planning. Tax planning involves using tax reliefs for the purpose for which they were intended. For example, claiming tax relief on capital investment, saving in a tax-exempt ISA or saving for retirement by making contributions to a pension scheme are all forms of tax planning.”

Tax professionals involved in fraudulent tax schemes could be pursued by HMRC under the criminal law, as they were in the Charlton case, but these kinds of prosecutions are exceptionally rare.

Where they do occur, the approach taken by HMRC confirms the proposition that tax avoidance can involve both dishonesty on the part of tax professionals and honest behaviour on the part of the taxpayer.

To give an example of a more recent case, in 2019 three men pleaded guilty to various counts of cheating the revenue for their role in promoting and enabling what HMRC termed “a fraudulent tax avoidance scheme”. Anthony Blakey, John Banyard and Professor Ian Swingland were convicted on indictment after enticing wealthy people into investing an a scheme which purported to invest in carbon credits and research into a cure for HIV. However, HMRC found little evidence of the investments having actually been made.

As set out in the press release issued by HMRC:

“Investors were able to claim tax rebates on the losses that the businesses apparently generated, or lower their tax bills, by offsetting losses against £160 million of income, attempting to avoid £60 million in tax. The majority of repayments claimed were withheld by HMRC….

There is no suggestion that the investors knew the scheme was a sham, or knew that their money was not being spent on research and development and carbon trading business activity.”

The press release went onto say:

“HMRC is working with the tax profession to tackle those who promote tax avoidance schemes. Promoters of Tax Avoidance Schemes legislation, introduced in Parliament in 2014, is aimed at tackling those who push the boundaries of the rules, and carries consequences for those who fail to change their behaviour.”21

HMRC clearly categorises the Banyard scheme as “Avoidance”, yet the successful prosecution of the tax professionals behind the scheme confirms that the tax losses arose from fraudulent or dishonest conduct, even in the circumstances where the participating taxpayers were were unaware of the fraudulent nature of the scheme.

Given that HMRC’s definition of “avoidance” expressly excludes planning and is limited to schemes that are exploitative, contrived, and artificial, then the tax loss under HMRC’s “avoidance” category should be considered to be part of the fraud gap.

Error – £3.7bn

HMRC’s definition of “error” is clearly limited to error arising from honest mistake because it only includes errors that arise “despite customers taking reasonable care”. It therefore excludes errors arising from negligence.

Under the description of behaviours, “Error” is described as “Errors result from mistakes made in preparing tax calculations, completing returns or in supplying other relevant information, despite the customer taking reasonable care” in the Measuring tax gaps 2021 edition.

Failure to take reasonable care – £6.7bn

The tax behaviour which HMRC describes as “Failure to take reasonable care” clearly corresponds to negligence as explained above. According to the Measuring tax gaps 2021 edition:

“Failure to take reasonable care results from a customer’s carelessness and/or negligence in adequately recording their transactions and/or in preparing their tax returns. Judgments of ‘reasonable care’ should consider and reflect a customer’s knowledge, abilities and circumstances.”

Non-Payment – £4bn

The Non-payment component of the Tax Gap reflects the impossibility of collecting every penny of tax that is owed because HMRC cannot collect outstanding tax from individuals and businesses that become bankrupt or insolvent.

According to Measuring tax gaps 2021:

“For direct taxes, non-payment refers to tax debts that are written off by HMRC and result in a permanent loss of tax — mainly as a result of insolvency. It does not include debts that are eventually paid.

VAT non-payment differs as it is based on the difference between new debts arising and debt payments.”

Non-payment can, therefore, be honest (genuine inability to pay) or fraudulent (deliberate failure to pay such as the use of phoenix companies).

Legal interpretation – £5.8bn

The behaviour which HMRC define as “legal interpretation” is the second largest component of the Tax Gap. It is also one of the most difficult to interpret as the wording is wide enough to encompass honest and dishonest behaviours. There is no mention of the term in the methodological annex of the Tax Gap.

Under the heading ‘Resolving issues of legal interpretation’, ‘Protecting Tax Revenues 2009’ stated:

“Legal interpretation relates to the potential tax loss from cases where HMRC and customers have different views of how, or whether, the law applies to specific and often complex transactions. Examples include the correct categorisation of an asset for allowances, the allocation of profits within a group of companies, or VAT liability of a particular item. In these situations the customer will have an alternative view of the law and of how it applies to the facts in their case to that held by HMRC.22

It follows from this that legal interpretation could cover issues arising from both honest and dishonest behaviour, even tax compliance. For example, it is possible that HMRC’s interpretation of the law in any particular case turns out to be wrong, in which case the incidence of non-compliance defined by HMRC as “legal interpretation” will be compliant, assuming that the issue under dispute does not fall foul of any other law.

If, after the case has been resolved in favour of the taxpayer HMRC continues to maintain that the disputed amount should not have been claimed, they can seek to change the law but losses that arise from a deficiency in the law should not be counted in the Tax Gap as defined by HMRC.

If HMRC end up prevailing in their view, the claiming of an allowance the taxpayer is not in fact actually entitled to claim could have been an honest mistake or a dishonest interpretation of the law.

Many avoidance schemes are characterised as honest disputes over legal interpretation, usually by the people that design and operate them, when in fact the legal interpretation claimed by the creators of the scheme is dishonest.

In Charlton, which involved a transfer pricing scheme using a Jersey registered company, the behaviour of the barrister involved, Cunningham, was described in the following terms:

“Charlton used Cunningham to reassure any doubting participants. The Crown’s case against Cunningham had been that he advised Wheeler that the scheme was effective although to his knowledge it was not.”

The fact that many tax avoidance schemes will involve the provision of legal advice to scheme users (taxpayers or customers in HMRC’s terminology) that testifies to the legality of the scheme means that by definition avoidance includes “cases where HMRC and customers have different views of how, or whether, the law applies to specific and often complex transactions.”

The examples given by HMRC of what constitutes “legal interpretation” and in particular “the correct categorisation of an asset for allowances” and “the allocation of profits within a group of companies” indicate that legal interpretation was intended to cover tax non-compliance, or avoidance by companies, particularly multinational companies as opposed to tax avoidance by individual taxpayers which is defined under the “tax avoidance” behaviour.

This proposition is fortified by the following passage contained in ‘Protecting Tax Revenues 2009’:

“HMRC’s approach to issues of legal interpretation is strategic and risk based. This has been developed to deal with the tax affairs of large businesses”23

Whether or not an incidence of legal interpretation arises from honest or dishonest behaviour will be a matter of subjective judgment. The case of GE vs HMRC provides a good example of where HMRC have significantly changed their position over time.

In this case, which involves a dispute over whether anti-avoidance legislation should have applied to a number of transactions carried out by GE, HMRC allege that GE failed to disclose relevant information regarding the scheme. At first, HMRC alleged that this failure to disclose was due to an honest mistake on the part of GE, which resulted in HMRC being mislead as to the true nature of the scheme. More recently HMRC applied to the High Court to amend their case to allege that the non-disclosure was fraudulent.24 GE and HMRC have now settled the case with no blame to either party.

In their 2019 edition of Measuring tax gaps, HMRC attempted to draw a distinction between avoidance and legal interpretation for the first time:

“Legal interpretation losses arise where the customer’s and HMRC’s interpretation of the law and how it applies to the facts in a particular case result in a different tax outcome, and there is no avoidance. Specifically, this includes the interpretation of legislation, case-law, or guidelines relating to the application of legislation or case-law.

Examples include categorisation such as an asset for allowances or VAT liability of a supply, the accounting treatment of a transaction, or the methodology used to calculate the amount of tax due as in transfer pricing, or VAT partial exemption.

The definition adopted in 2019 remains the same in the 2021 edition of Measuring gax gaps.

The reference to transfer pricing is interesting, as transfer pricing disputes frequently arise from tax avoidance by multinational companies, which demonstrates the difficulties of seeking to distinguish between “avoidance” and “legal interpretation”.

This, and the proposition that a significant amount of “legal interpretation” is fraudulent is further demonstrated by HMRC’s guidance on their profit diversion compliance facility. The facility is a form of amnesty for multinationals that have moved their profits out of the UK in a non-compliant manner. Under the heading: behaviours and conclusions on penalties, the facility states the following:

“If we find that additional tax is due, we will always consider the behaviours that have given rise to the error and whether penalties should be charged. The Facility does not offer special terms and the normal penalty provisions and HMRC practice apply.

Our investigations into Profit Diversion to date have established that in a large number of cases the factual pattern outlined to HMRC at the start of an enquiry does not stand up to scrutiny once tested. That may be a result of a careless error (for example individuals within a group being unaware of what the actual facts are) but it may also be a result of a deliberate behaviour, that is a group knowingly submitting a TP [Transfer Pricing] methodology in a Corporation Tax Return based on a false set of facts….

Where HMRC suspects there has been an attempt by a group to deliberately mislead, then we will refer the issue to Fraud Investigation Service for consideration of a criminal investigation or civil investigation into fraud.”25

It should be noted that disputes over transfer pricing methodology are the most significant cases that HMRC take on. The latest figures for tax under consideration show that “Transfer Pricing and Thin Capitalisation” make up 1/3rd of the total tax receipts being disputed between HMRC and large businesses – £10bn.26

Drawing all of this together, the term “legal interpretation” appears broad enough to cover a wide range of tax behaviour.

However, the subjective nature of how HMRC assesses tax behaviour, and the fact that questions of legal interpretation will often arise before any assessment of whether or not a company or individual’s interpretation is honest or not, will mean that a significant part of the “legal interpretation” category will arise from fraudulent behaviour.

A good exercise would be for HMRC to conduct an analysis of cases that fell into the legal interpretation category five years ago, and publish what the outcomes of those cases were with an assessment of any underlying behaviour that led to non-compliance.

Base erosion and profit shifting (BEPS)

HMRC appear to recognise a separate type of tax behaviour – BEPS – which falls outside the Tax Gap. BEPS is a term which emerged from the OECD’s 2013 study commissioned by the G-20 entitled ‘Addressing Base Erosion and Profit Shifting’ and has been used to describe tax avoidance by multinational enterprises.

BEPS appeared for the first time in HMRC’s ‘Measuring tax gaps 2014 edition (Tax gap estimates for 2012-13)’ following tax avoidance and incorporating the OECD definition in these terms (emphasis added):

“It [tax avoidance] does not include international tax arrangements such as base erosion and profit shifting (BEPS). Measures for tackling this are overseen by the Organisation for Economic Co-operation and Development (OECD). The OECD defines BEPS as tax planning strategies that exploit gaps and mismatches in tax rules to make profits disappear for tax purposes or to shift profits to locations where there is little or no real activity, but the taxes are low resulting in little or no overall corporate tax being paid.”27

In ‘Measuring tax gaps 2015 edition (Tax gap estimates for 2013-14)’ 2015, HMRC added the following paragraph to the existing description (emphasis added):

“Where we can challenge cross-border tax avoidance or aggressive tax planning under UK law, it is reflected in the tax gaps for avoidance and legal interpretation, but where the effect of such activity is the result not of frustrating UK law but of exploiting the international tax framework, we do not include it in the avoidance tax gap.28

Measuring tax gaps 2021 contains a more nuanced but essentially similar description:

“Some forms of base erosion and profit shifting (BEPS) are included in the tax gap where they represent tax loss that we can address under UK law.

As new measures introduced in accordance with recommendations made in the BEPS project by the G20 group of world-leading economic nations and the Organisation for Economic Co-operation and Development (OECD) take effect, our ability to address BEPS under our domestic law will be greatly strengthened.

The tax gap does not include BEPS arrangements that cannot be addressed under UK law and that will be tackled multilaterally through the OECD.

HMRC have never provided a breakdown of how much BEPS activity they categorise as avoidance and legal interpretation, and how much they don’t count at all.

The descriptions given by HMRC suggests that the tax authority believes that a substantial amount of tax avoidance by BEPS arises from the honest use of international tax system, albeit with outcomes that the UK government may not like. This is the natural conclusion of the proposition that BEPS cannot be dealt with under UK law and requires changes in the law agreed internationally to be addressed.

That confusion is not aided by the OECD’s own characterisation of BEPS (which is repeated in HMRC’s Measuring gax gaps document), which describes BEPS as follows:

What is BEPS?

Base erosion and profit shifting (BEPS) refers to tax planning strategies that exploit gaps and mismatches in tax rules to make profits ‘disappear’ for tax purposes or to shift profits to locations where there is little or no real activity but the taxes are low, resulting in little or no overall corporate tax being paid.”29

The use of the term “planning” is clearly a misnomer. The definition, with references to attempts to exploit the tax rules by what are clearly artificial transactions closely aligns with HMRC’s own definition of avoidance.

The schemes in Charlton would be described today as Base Erosion and Profit Shifting (BEPS) schemes because they were devised to erode the UK’s tax base by shifting the taxable profits of UK companies to Jersey intermediaries. According to Farquharson LJ:

“It was the case for the Crown that the accounts presented to the Revenue by the United Kingdom companies were false in that by using Charlton’s scheme to transfer part of their profits to the Jersey companies they were not disclosing the full extent of the profits they had made. It was this lack of disclosure which formed the basis of the false representations alleged in the indictment. Each of the Appellants was charged in the relevant counts with cheating the Revenue by ‘… falsely representing that the apparent purchases (by the United Kingdom company) from (the Jersey company) were bona fide commercial transactions’.”30

The avoidance scheme used by Google, which would clearly be described as BEPS by both the OECD and the UK government, was subject to criminal procedures for “aggravated tax fraud” in France31 – an OECD member.

This suggests that a significant amount of tax behaviour described as BEPS and not counted in the Tax Gap should be classified as dishonest or fraudulent tax behaviour.

Measuring the tax fraud gap

From the above analysis, we can see that even on its own terms the behavioural categorisation by HMRC of non-compliance into eight categories (with a ninth BEPS, which falls outside the tax gap) is highly problematic.

The level of tax evasion and avoidance are grossly misstated by separating evasion into several different behavioural categories and via the inclusion of some avoidance in the “legal interpretation” category and the exclusion of BEPS. Through this approach, the overall level of dishonest behaviour is obscured.

The artificial nature of HMRC’s categorisations of tax behaviours in the Tax Gap may well be explained by the historical division between the work of the Inland Revenue and Customs and Excise and the fact that historically the Tax Gap was used as an internal performance measure and strategic tool. This has meant that Tax Gap the categorisation has followed how HMRC internally treat different types of tax non-compliance.

However, the dilution of categories such as avoidance and evasion as well as the exclusion of categories like BEPS also has the effect of diverting criticism that “HMRC has not been sufficiently challenging of multinationals’ manifestly artificial tax structures”32 in the words of the Public Accounts Committee.

Given that the Tax Gap has now developed into a measure by which HMRC presents their performance to the public and parliament, it would be better to move away from system based on technical definitions derived from HMRC’s internal arrangements, to categories based on clear legal concepts that everyone can understand. Fraud, Negligence and Honest non-compliance.

Categorising the Tax Gap in this way would provide a clearer way of presenting Tax Gap data to the public, and provide a better understanding of the scale of unlawful non-compliance.

The Department for Work and Pensions (DWP) already follows a similar approach in their equivalent of the Tax Gap, “Fraud and Error in the Benefit System”. This categorises non-compliance into just three categories, fraud, claimant error and official error (which does not differentiate between errors arising from honest mistakes or negligence).

As our analysis demonstrates, calculating the fraud, negligence, and honest non-compliance tax gaps should be relatively easy to do. There are already several categories which are clearly analogous to fraud, negligence and honesty.

If we take the HMRC behaviours that clearly arise from fraudulent conduct, we find that in 2019/20 fraud accounted for at least £15.2bn or 43% of the Tax Gap. To this would need to be added any fraudulent conduct that can be found in the categories of non-payment and legal interpretation, and of course BEPS.

HMRC have never published an estimate of how much tax is lost to BEPS and there are a number of studies from both academics and NGOs. Research from the University of Oxford calculated that profit shifting by multinationals results in foreign owned multinationals shifting 50% of their taxable profit outside of the UK, leading to tax losses of £25bn in 2014.33 A study by a number of academics produced for the European Parliament found that profit shifting could have cost £20bn in 2013.34

Taking BEPS into account, and assuming that a proportion of tax losses to Non-Payment and Legal Interpretation result from fraudulent tax behaviour, it would not be unreasonable to assume that Tax Fraud Gap is at least £20bn in the UK. However, more work would be needed to come to a reliable estimate. We recommend that HMRC complete this analysis as part of next year’s Tax Gap estimates.

TaxWatch, September 2021

The full report can be found as a PDF here.


4 The Times, March 30, 1993.

5 The standard of proof differs. Fraud is proved beyond reasonable doubt in criminal proceedings and on a balance of probability in civil cases.

8Blyth v Birmingham Waterworks (1889) 14 App. Cas. 337, 374 c

9 Anderson vs HMRC [2009] UKFTT 206 at [22].

10 Royal Brunei Airlines v Tan [1995] 2 AC 378, 389. Emphasis supplied.

11 Hanson vs HMRC [2012] UKFTT 314 at [21]. Emphasis supplied.

12 R vs Mavji, [1987] 84 Cr App R 34

13 R v Hudson, [1956] 2 QB 252, 261-262.

14 R vs Charlton and others, [1996] STC 1418.

15 Ibid

16 Ibid

17 Robert Rhodes et al, ‘Regina v Charlton, Cunningham, Kitchen and Wheeler’ (1999) Journal of Money Laundering Control, 197 page 206.

18 Commissioner of Inland Revenue (New Zealand) v Challenge Corporation Ltd. [1986] BTC 442

21 HMRC Press Office, Two jailed for £60m fraudulent HIV cure tax fraud, 25 February 2019,

23 Ibid, pages 16-17, paragraphs 5.15-5.16.

24 For more information on the GE case see TaxWatch, Around the world with $5bn,

25 Profit Diversion Compliance Facility Guidance, para. 4.4.1.

26 HMRC, Customer compliance: how HMRC’s compliance yield is split by business area and our approach to tax compliance and large businesses.

30 R vs Charlton and others, [1996] STC 1418.

31 Following criminal investigations and with criminal proceedings looming, Google agreed to a EUR 1 billion settlement under a non-prosecution agreement. See: Reuters, Google to pay $1bn in France to settle fiscal fraud probe, September 12 2019,

32 Public Accounts Committee, Ninth Report, Tax Avoidance – Google, 10 June 2013

33 Bilicka, Comparing UK tax returns of foreign multinationals to matched domestic firms, American Economic Review, 2019, 109(8), 2921-53,

34 European Parliamentary Research Service, “Bringing transparency, coordination and convergence to corporate tax policies in the European Union I – Assessment of the magnitude of aggressive corporate tax planning”, September 2015