Equality before the law?

19th February 2021

HMRC’s use of criminal prosecutions for tax fraud and other revenue crimes. A comparison with benefits fraud.

Download PDF


In this report we look at criminal investigations and prosecutions into tax fraud and other revenue crimes initiated by HMRC, and contrast this with the Department for Work and Pensions investigations and prosecutions for benefits crime.

We find that on the basis of the government’s own figures, tax fraud costs the Treasury nine times the amount lost to benefits fraud.

Despite this, the government allocates significantly more resources (relative to the tax and benefits gaps) to compliance in the benefits system than in the tax system. This inequality of arms, combined with a stricter policy on when to refer cases for prosecution practised by the DWP, leads to a vast disparity between the amount of benefits crime prosecuted as a criminal offence when compared to tax offences.

This disparity demonstrates that the law is not being applied equally with regards to crimes committed against the public purse.

In numbers

  • Tax fraud cost the Treasury an estimated £20bn in 2018/19 – 9x more than benefits fraud (£2.2bn)

  • DWP employs 3.5x more staff in compliance than HMRC (adjusted for size of tax and benefits gaps)

  • Over the past 11 years there have been 85,745 criminal prosecutions for benefits crimes, 23x more prosecutions than for tax crime of all types (3,665)

  • 8.5x more suspended or immediate custodial sentences have been handed down for benefits crime vs tax crime over the last 11 years

  • The number of criminal prosecutions relating to tax crime of all kinds have decreased by 39% since 2015

  • The number of live criminal investigations into serious and complex tax crime has increased from 50 to 400 between 2015 and 2020

Prosecutions for tax fraud and other tax crimes

HMRC has faced calls to do more to tackle tax fraud through criminal prosecution for some time.

In 2015, Parliament’s Public Accounts Committee described the number of prosecutions for offshore tax evasion as “woefully inadequate”.1 In 2016 the Committee called on HMRC to:

“do more to tackle tax fraud and counter the belief that people are getting away with tax evasion. It needs to increase the number of investigations and prosecutions, including wealthy tax evaders, and publicise this work to deter others from evading tax and to send out a message that those who try will not get away with it.”2

The Ministry of Justice publishes data on the number of criminal cases prosecuted under several broad categories defined by the Home Office. This data is a National Statistics publication. The categories relevant to tax law are: “Acting with Intent to Defraud and to the Prejudice of Her Majesty the Queen and the Public Revenue”; “Revenue Law Offences” and “Other Offences (against Revenue Laws)”.3

When looking at this data there is a marked decline in the numbers of cases being proceeded against for all three categories of revenue crimes, from 430 in 2015 to 261 in 2019.

The Home Office categorisations of crimes used by the MoJ are broad and cover a wide variety of tax crimes including offences relating to customs and excise.

In order to find out more about the prosecutions for tax fraud TaxWatch sent an FOI to the Crown Prosecution Service (CPS) seeking to discover how many offences they had charged relating to specific revenue offences related to tax fraud. The figures we received are reproduced below.

The data are not consistent with the statistics from the Ministry of Justice as the CPS numbers look at all charges. It is possible to charge an individual with several offences and the 2,278 offences charged in the table below relate to 887 individual case files. The Ministry of Justice data relates to individual defendants.

Overall, the trend was the same: a clear decline in prosecutions for almost all forms of tax fraud.4

2015-2016 2016-2017 2017-2018 2018-2019 2019-2020
Common Law – Cheating the Public Revenue 154 144 113 127 79
Criminal Law Act 1977 { s.1(1) } – Conspiracy to Cheat the Public Revenue 3 9 13 49 27
Customs and Excise Management Act 1979 { s.167 } – Providing false documents or information to HMRC 1 0 0 0 0
Taxes Management Act 1970 { s.106A } – Fraudulent evasion of income tax 244 173 127 62 69
Value Added Tax Act 1994 { s.72(1) } – Fraudulent evasion of VAT 182 228 184 161 129
Totals 584 554 437 399 304

Table 1: Offences Charged And Reaching A First Hearing At Magistrates’ Courts5

This downward trend in prosecutions is likely to continue as HMRC’s latest accounts state that due to Covid-19 there has been a significant reduction in the amount of criminal investigations they have undertaken.6

Prosecutions for serious tax fraud and complex tax crimes

One area where there has been particular pressure for HMRC to do more is in the area of serious and complex tax fraud.

In HMRC’s 2015-2020 Business Plan a commitment was made to increase the number of criminal prosecutions of serious and complex tax crime, with a particular focus on wealthy individuals and corporates, to 100 per year by the end of Parliament. At the time, Parliament was scheduled to run until 2020.7

In 2016, HMRC brought together various elements of its criminal investigations activities under the Fraud Investigations Service (FIS). One of the ideas behind this was to tackle more complex areas of tax crime through a more co-ordinated approach.

The department also launched the J5 in 2018, the Joint Chiefs of Global Tax Enforcement, which brings together directors of criminal investigations from the revenue services of the UK, United States, Canada, Australia and the Netherlands. This group is aimed at creating a more effective approach to transnational tax crime.

HMRC’s increased focus on complex tax crime does appear to be paying some dividends. The only charge that has seen an increase in prosecutions in the CPS data is “conspiracy to cheat the revenue”. Conspiracy charges are generally more amenable to more complex cases (although that is not always the case).

HMRC told TaxWatch that since 2016 it has increased the number of investigations into the wealthiest and most sophisticated offenders by a factor of eight, from 50 to 400.

With regard to the 100 prosecutions target, HMRC provided us with the following data on charging decisions related to serious and complex tax crime over the last three years.


Positive Charging Decisions







Table 2: Positive Charging Decisions against the target of 100 prosecutions for serious and complex tax crime. Data provided by HMRC.

HMRC told us that they had expected to meet their 100 prosecutions target by 2021, but that this is now unlikely to happen due to the Covid-19 crisis.

There has not been one prosecution of a company from 2015–2020, despite the Corporate Criminal Offence coming into force in September 2017.8 HMRC told us CCO investigations take a significant amount of time and it is too soon for prosecutions to start coming through the system. They told us that they currently have 13 live investigations under the Corporate Criminal Offence with a further 17 cases under review.

Overall, despite the clear decline in the number of prosecutions for all types of tax crime, the evidence shows that HMRC has had some success in focusing its limited resources on increasing the investigation and prosecution of some incidences of more serious and complex crime.

But it must be stressed that this all starts from a very low base. Before 2015, the UK government launched several amnesty programmes designed to encourage people to declare offshore income. The UK Government’s Lichtenstein Disclosure Facility even went as far as to give assurances against criminal investigation, and offered a favourable 10% penalty on undeclared income that was declared through the scheme.

By 2015, the year in which HMRC told us they opened 50 criminal investigations into serious tax fraud, 5,644 people had made disclosures under the Lichtenstein amnesty programme.9 If HMRC took the same approach as the DWP, all of these individuals would have been referred to the Crown Prosecution Service for criminal prosecution.

HM Government’s policy on tax fraud and benefits fraud

Criminal investigation leading to prosecution is just one way that the government can seek to recoup money from tax underpayment or benefits overpayment. There are also civil recovery processes, administrative penalties, and the tax courts.

Both HMRC and the DWP have the choice whether to pursue fraud as a criminal or civil matter. Both have a different policy and approach with regards how they make that decision.

Like the former Inland Revenue, HMRC has maintained a selective prosecutions policy. This has a strong preference against using criminal investigation and prosecution to deal with tax fraud.

Instead, HMRC’s stated policy is to deal with tax fraud by civil investigation and settlement. The department’s criminal investigations policy will “consider” opening a criminal investigation in certain circumstances.

HMRC’s policy reserves criminal investigation “for cases where HMRC needs to send a strong deterrent message or where the conduct involved is such that only a criminal sanction is appropriate.”10

The Department for Work and Pensions’ policy is much stricter. The Department states that they “will normally” refer a case to the CPS or Crown Office and Procurator Fiscal Service under a number of proscribed circumstances, including cases where the recoverable amount is more than £5,000.11

Fraud investigations

Before a decision is made on whether or not to prosecute, an investigation has to be carried out.

The DWP annual report reveals that the organisation completed 46,000 fraud investigations and 332,000 “compliance cases” (which is a review of a case  where there may be an overpayment but fraud is not suspected) in 2019-2020. Of these fraud investigations over 2,000 were referred to prosecuting authorities.

HMRC deal with tax fraud through a variety of procedures: criminal investigations under the Police and Criminal Evidence Act 1984, and civil fraud investigations under Code of Practice (COP) 8 or 9.

A Freedom of Information request from an accountancy firm revealed the number of investigations carried out by HMRC under COP 8 and 9.








Cases opened






Cases closed






Yield recorded







Cases opened






Cases closed






Yield recorded






Total cases opened






Total cases closed






Total yield recorded






Table 3: HMRC COP8 and COP9 investigations 2016-202012

HMRC told us that in 2019-20 they concluded 864 criminal investigations. This means that between COP 8, COP 9 and criminal investigation, HMRC’s Fraud Investigations Service completed 1,720 investigations for tax fraud in 2019-20.

HMRC also told us that they concluded 337,000 civil compliance checks in 2019-20 and that since its launch in 2016 the FIS had launched over 76,000 “civil cases”.

These cases involve the use of a wide range of civil tools used by the FIS, including account freezing orders, the seizure of alcohol from illegal distribution sites as well as HMRC’s work with supply chains, including regular monitoring and engagement with legitimate traders at risk of involvement with illicit trade, to VAT assessments and penalties for companies knowingly involved in fraudulent trading.

If we limit our analysis to fraud investigations completed by the DWP and tax fraud investigations carried out under COP 8, 9 and criminal procedure, we find that the DWP completes 27 times more civil and criminal fraud investigations than HMRC.

Even if we were to include the “civil cases” launched by the FIS, these 76,000 cases were spread over four years, and so do not account for the difference in fraud investigations completed by DWP on a yearly basis. In any event, the information provided to us about what constitutes a “civil case” suggests that they are not analogous to fraud investigations carried out by the DWP.

Perhaps the better measure is in criminal investigations. The number of cases referred to prosecuting authorities by the DWP is more than double the total number of criminal investigations carried out by the FIS in the latest year we have figures for.

Criminal prosecutions for tax crimes vs benefits crimes

The obvious outcome of the DWP completing more fraud investigations and having a policy of referring more cases of fraud to prosecutors is that vastly more people face criminal charges leading to a criminal record, fines or a prison sentence for benefits related offences compared to tax related offences. Our analysis of government criminal justice system statistics reveal that in the 11 years to 2019:

  • 23 times as many people are prosecuted for benefits related offences as they are for tax related offences.
  • Tax crime is generally considered a more serious offence, with 44% of those prosecuted for tax related offences handed a suspended or immediate custodial sentence, compared with 16% for benefits related offences.
  • Despite prison sentences being more common for tax crimes, the difference in the volume of prosecutions for benefits crime means that many more people end up going to prison for benefits crime than tax crime. Over the 11 year period 13,540 people were handed suspended or immediate custodial sentences for benefits related offences, compared with 1,601 people for tax related offences.

The table below sets out the total numbers of people prosecuted for tax and benefits crime between 2009 and 2019.




Offence categories/ outcomes

Acting with intent to defraud and to the prejudice of Her Majesty the Queen and the Public Revenue

Revenue law offences

Other offences against revenue laws

Total tax related offences

Benefit fraud offences summary

Benefit fraud offences triable either way

Other social security offences

Total benefits related offences

Proceeded against









Found Guilty



























Suspended sentence / immediate custody









Per cent proceeded against resulting in a suspended sentence or immediate custody









Average Fine




Average £780




Average £234

Average custodial sentence (months)




Average 33.3 months




Average 4.1 months

Table 4: Criminal justice system statistics 2009-201913

It should be noted that the statistics from the Ministry of Justice on criminal prosecutions bundle a variety of tax crimes together, and include a number of customs and excise offences. There is no easy way of breaking out the data to look at fraud offences specifically. It also should be noted that as set out in more detail below, HMRC’s remit goes substantially beyond the policing of revenue offences only.

If we look at more serious revenue and benefits crimes by excluding summary only offences (the “Other offences against revenue laws”, “Summary benefit fraud offences” and “Other social security offences” categories) we see that in the 11 year period there were 1,973 prosecutions of individuals for revenue law offences and 46,151 offences for more serious benefits fraud, a ratio of 23:1, the same as for all benefits and revenue crimes.

The tax fraud gap and the benefits fraud gap

It may not be surprising that the DWP completes more fraud investigations and more prosecutions than HMRC. Benefits crime tends to involve lower amounts of money and tax fraud can be very complex. This means that tax cases can require more resources to be dedicated to each case. It may be that looking at case numbers alone does not give us the whole story.

To delve more into this issue, our research attempted to look into whether the relative amount of resources dedicated to pursuing tax fraud and benefits fraud were proportionate to the size of the problem being faced.

The DWP categorises losses in the benefits system under three categories: fraud, official error and claimant error. These figures reveal that the total benefit overpayments due to fraud and errors were worth £4.1bn in 2018-2019, 2.2% of the total benefits bill.14 Of this, DWP categorise £2.2bn as arising from fraud, accounting for 1.2% of the total benefits bill.15 These figures are gross figures, before any collections arising from compliance work.

This is no small sum, but it is a relatively small amount compared to the amount of tax underpayments from fraud, error or negligence. The gross tax gap for the same period was £44bn and the net tax gap after cash collected from compliance work was £31bn.16 This means that the loss to the Treasury due to the underpayment of tax accounts for 4.9% of tax revenue on a net basis or 7% on a gross basis.17

HMRC’s Measuring Tax Gaps publication does not categorise tax losses in the same way as the DWP. Instead of simply using fraud and error, HMRC define tax losses using a number of different behaviours. The tax gap figures are also reported on a net basis, whereas DWP figures for losses to fraud are reported on a gross basis.

The following behaviours in the tax gap are very clearly analogous to fraud as categorised by the DWP: evasion, criminal attacks, hidden economy and avoidance (which is defined as the use of avoidance schemes and therefore could not be considered to be the same as either claimant error or official error). These total £13.4bn in the latest tax gap figures – 2.1% of the total tax take.

This is equivalent to 43% of the total tax gap.

In order to create a gross fraud figure to compare with the DWP statistics, we added back in 43% of the cash received due to compliance activities (£13.2bn). This produces a total gross tax gap attributable to fraud as £20bn.

This will not include all tax losses to fraud, as in reality there will be fraud in the behavioural categories of “legal interpretation” and “non-payment”, which account for £4.9bn and £4.1bn of the tax gap respectively.

If we limit ourselves to just categories which are clearly analogous to the DWP categorisation of fraud, we find that the potential value of tax fraud is approximately nine times the potential losses to benefit fraud.

Compliance resourcing

One way of looking at the resourcing of counter fraud activity is to compare the staffing levels at HMRC to the DWP. This is not a straightforward comparison.

The Counter Fraud, Compliance and Debt (CFCD) Department of the DWP had approximately 8,000 staff in 2019-2020.18 We asked the DWP whether it would be possible to give us the number of staff directly employed in fraud investigations, rather than broader compliance work. They were not able to provide this data.

According to an FOI request made by TaxWatch, HMRC has 24,191 people working in its Customer Compliance Group, a reduction of 800 staff members over the past four years.

HMRC’s Customer Compliance Group is a large group that cuts across several directorates and includes a number of functions that would not be carried out by the DWP’s CFCD Department.

The number of people working in HMRC’s Fraud Investigations Service, which focuses on more serious incidences of fraud, is 5,000. The number of staff employed in the Wealthy and Mid-Sized Business Compliance directorate unit reduced from 1,007 in 2016-2017 to 866 in 2019-2020.19,20

Because we are not able to break down the DWP’s CFCD department by function, we felt the fairest comparison would be to look at staffing in compliance as a whole, looking at all compliance staff in HMRC and the DWP’s CFCD department and compare this to the gross tax and benefits gaps.

This comparison demonstrates that for every £1m of tax underpaid HMRC employs 0.55 compliance officers, whereas the DWP employs 1.95 compliance officers for each £1m in benefits overpaid due to fraud or error. On this basis the government dedicates three and a half times the amount of resource to benefits compliance as it does to tax compliance.

If the government were to dedicate the same amount of resources, on a per pound at stake basis, to closing the tax gap as they deployed to recouping overpayment in the benefits system, it would mean that HMRC would need to employ an additional 62,000 staff in their compliance department. This is more than ten times the amount of staff currently employed in the Fraud Investigation Service, or more than twice the amount of staff currently employed in compliance.

It should be noted that the figures we have produced for HMRC’s resourcing compared to the gross tax gap will be a significant underestimate, because HMRC’s tax gap does not include significant areas of avoidance such as profit shifting by multinational companies which means that the real tax gap is significantly higher.

It also needs to be noted that HMRC’s remit includes a broad range of compliance activities that go well beyond tax crime, and even includes the enforcement of benefits crime.

In correspondence, HMRC were able to provide us with figures for all prosecutions initiated by the department over a ten year period. This demonstrated that the department secured 6,017 convictions over the last ten years, for all offences that come under its remit. The list of offences provided by HMRC included an impressive array of offences ranging from a failure to pay the minimum wage under National Minimum Wage legislation, entering the country illegally contrary to the Immigration Act, and offences relating to Tax Credits and Child Benefits, which are forms of welfare benefits administered by HMRC and which would likely be counted as “benefits fraud” under the categorisation of crime by the Home Office.

FY Convictions Acquittals Prosecutions % Success
2011/12 413 36 449 92%
2012/13 540 36 576 94%
2013/14 716 45 761 94%
2014/15 642 67 709 91%
2015/16 808 72 880 92%
2016/17 807 79 886 91%
2017/18 835 82 917 91%
2018/19 648 101 749 87%
2019/20 608 83 691 88%
Total 6,017 601 6,618 91%

Table 5: HMRC 10 year prosecutions, all offences, data provided by HMRC

This data shows that the number of convictions HMRC secures goes far beyond those categorised as tax crimes under the national statistics database. It demonstrates that HMRC’s resources in tackling tax crime are significantly more stretched than our analysis would suggest. HMRC have fewer resources than the DWP relative to the size of the problem they face, and in addition to tax crime, they police a variety of other crimes including some benefits crime.


“Don’t tell me what you value, show me your budget, and I’ll tell you what you value.”

– Joseph Biden Jr

Both revenue fraud and benefits fraud are similar in that in both cases the victim is the Treasury. Although comparisons between the DWP and HMRC are not straightforward given the much broader remit of HMRC and the differences in the way data is collected, 86,000 prosecutions over 11 years for benefits offences against a few thousand for tax offences tells a pretty clear story.

As a society we treat benefits fraud much more severely than revenue fraud – and are prepared to dedicate significant resources in pursuit of that policy.

Our research has found that the DWP employs more than three times (3.5x) the number of staff in compliance compared to HMRC when the amount of cash at stake from benefits overpayments and tax underpayments is taken into account.

If HMRC applied the same policy as the DWP and referred every fraud case worth more than £5,000 to the Crown Prosecution Service, the criminal justice system would topple over. Prosecutors would not be able to cope with the caseload and the criminal courts would become flooded with tax cases.

HMRC’s highly selective use of criminal prosecution is not new. It is a policy that was previously practiced by the Inland Revenue and has been around for at least 70 years. HMRC contends that the preference it shows for the use of civil procedure is a cost-effective way of making sure that taxes due are paid and historically the Inland Revenue did not see its role as being the punishment of offenders.21

However, if the result of the differences in policy and resourcing is that someone committing benefits fraud is more likely to face criminal prosecution and go to jail than someone engaged in tax fraud, this raises serious questions about whether the rule of law is being applied fairly across society.

This disparity cannot be explained away by any difference in severity of the crimes committed. When people are convicted of tax fraud, on average they receive longer sentences, reflecting the more serious nature of the crime. Our research demonstrates that even on HMRC’s own figures, tax fraud costs the Treasury nine times the amount lost to benefits fraud.

The issues raised by this report go far beyond the administration of the tax and benefits system, and speak to the way in which government and society as a whole treat crimes committed by benefits claimants and those committed by tax cheats.

The reasons for the differences in treatment are beyond the scope of this paper. However, whatever the reason, the unequal treatment between similar types of offence seems difficult to justify.

Picture credit James Cridland, used under the Creative Commons Licence

Note on data and sources

As has been set out in this report, we have used a number of data sets to look at the rate of criminal investigations and prosecutions for tax and benefit fraud.

The data sets are not directly comparable and do not always compare like for like, but together we believe that they present strong evidence for our findings.

The sources of our data include national statistics publications, freedom of information requests and data provided directly to us by HMRC. We are grateful to HMRC for engaging with us in this project and providing us with additional data for our report.

Here we set out a bit more detail on the various data we collected.

Table 1 was derived from a Freedom of Information Request made to the Crown Prosecution Service. The CPS collects data on offences which are charged and proceed to a first hearing at a magistrates court. This will include indictable only offences which are then referred to the Crown Court.

The data in this table relates to offences and not prosecutions. As such it is not comparable to other data sets. A single person can be charged with multiple offences, in the data given to us, there are 2,278 offences charged which relate to 887 individual case files.

It should be noted that this CPS data is for England and Wales only.

Table 2 presents figures from HMRC on the number of cases referred to prosecutors resulting in a charging decision in relation to its own categorisation of serious and complex tax crime. This categorisation exists to keep track of the target set to reach 100 prosecutions a year in this area. HMRC provided us with this data directly.

Table 4 is derived from Ministry of Justice’s Criminal Statistics Quarterly Publication. This is a national statistics publication and is based on a Home Office categorisation of crimes.

The offence categories are taken directly from the criminal justice system statistics database. In the data there are no categories for tax fraud, whereas there are for benefits fraud. Some crimes included under the “other revenue offences” category will relate to customs offences and other offences perhaps not directly related to tax.

Because of this we thought the fairest comparison was to compare all types of benefits crime to revenue crimes, As such, the “total tax related offences” will likely overstate the amount of people who are prosecuted for tax related offences.

The data is derived from court records in both the crown and magistrates courts. It covers all countries of the United Kingdom and Northern Ireland. In this dataset, data for “Acting with intent to defraud and to the prejudice of Her Majesty the Queen and the Public Revenue” is only available from 2011 onwards.

Table 5 is a list of all prosecutions coming out of HMRC. This data was provided to us by HMRC directly. HMRC has a broad remit which goes beyond tax crime. The full list of offences included in this table is included below:

We asked HMRC specifically how many cases it had prosecuted under the Tax Credits Act and under the Immigration Act. They told that there were fewer than 20 prosecutions under Tax Credits Act in the dataset. The Immigration Act prosecution figures relate to a single case where individuals were arrested for both PAYE and immigration fraud.

  • ALDA 1979
  • CAA 1981 – s 1(1)
  • CJL Act 2010 – s49
  • CLA 1977 – s1(1)
  • Common Law – Fraud and Fraudulent Scheme
  • CRCA 2005 – s30
  • CRCA 2005 – s32
  • Forgery and Counterfeit  Act 1981 – s1
  • Fraud Act 2006 – s1
  • Fraud Act 2006 – s2
  • Fraud Act 2006 – s3
  • Fraud Act 2006 – s4
  • Fraud Act 2006 – s6(1)
  • Fraud Act 2006 – s7(1)
  • HODA 1979 – s13(4)
  • Immigration Act 1971 – s24a(1)
  • MLR 2007 – s45
  • MLR 2007 – s7
  • NMW 1998 – s31(1)
  • NMW 1998 – s31(5)(a)
  • Perverting the Course of Justice
  • POCA 2002 – s327(1)
  • POCA 2002 – s327(1)
  • POCA 2002 – s328(1)
  • POCA 2002 – s329(1)
  • POCA 2002 – s330
  • POCA 2002 – s334
  • POCA  2002 – s327(1)
  • Section 684(4A) of the Income Tax (Earnings and Pensions) Act 2003
  • Serious Crime Act 2007 – s25
  • Social Security Admin Act 1992
  • TCA 2002 – s35
  • TDA1979 – s8H
  • Theft Act 1968 – s17
  • TMA 1970 – s106(B)
  • TPA 1979 – s8(g)
  • TPDA 1979 – s8(H)
  • TPDA 1979 s8(G)
  • VAT Act 1994 – s73 (3), s72 (8) s72(10), s72(11)

1Committee of Public Accounts, HM Revenue and Customs performance in 2014-15, Sixth Report of Session 2015-16, HC393, para 9: http://www.publications.parliament.uk/pa/cm201516/cmselect/cmpubacc/393/393.pdf

3Statistics available from Criminal Justice Quarterly https://www.gov.uk/government/collections/criminal-justice-statistics-quarterly

4We have provided a note on our data at the end of this document

5FOI Response from the CPS to TaxWatch, 16 December 2020

6See HMRC, 2019-2020 Annual Report and Accounts, R10, available from: https://www.gov.uk/government/publications/hmrc-annual-report-and-accounts-2019-to-2020

8HMRC fails to deliver on pledge to increase criminal prosecutions by end of 2020, FOI request reveals, Kingsley Napley, 21 December 2020, https://www.kingsleynapley.co.uk/insights/blogs/criminal-law-blog/hmrc-fails-to-deliver-on-pledge-to-increase-criminal-prosecutions-by-end-of-2020-foi-request-reveals

9FT Advisor, Crown Dependency tax disclosure failure, August 3 2015, available from: https://www.ftadviser.com/2015/08/03/regulation/crown-dependency-tax-disclosure-failure-p49cHwoUDuXo9nAukPCRyI/article.html

10HMRC’s criminal investigation policy, Gov.uk, Updated 13 May 2019, https://www.gov.uk/government/publications/criminal-investigation/hmrc-criminal-investigation-policy

11DWP Policy Paper, “Penalties policy: in respect of social security fraud and error”, August 14th 2017 https://www.gov.uk/government/publications/penalties-for-social-security-fraud-and-error/penalties-policy-in-respect-of-social-security-fraud-and-error

12HMRC Serious Tax Investigations – latest statistics, Lancaster Knox, 15 July 2020, https://lancasterknox.com/hmrc-serious-tax-investigations-latest-statistics/

13All data sourced from Criminal justice system statistics quarterly: December 2019, Gov.uk, 21 May 2020, https://www.gov.uk/government/statistics/criminal-justice-system-statistics-quarterly-december-2019.

15DWP do have figures for 2019-20, but the latest HMRC data relates to 2018/19 and so we are using the previous year’s figures

16To calculate a ‘gross tax gap’, we took the 2019-2020 tax gap of £31bn from ‘Measuring tax gaps 2020 edition’, and added to it the £9bn cash HMRC expected to recoup due to its compliance work from HMRC’s November 2020 corporate report on compliance yield.

17The measuring tax gaps publication by HMRC 2018-19 edition quotes a figure of 4.7%. We have measured the potential tax losses against the 2018/19 total tax revenues figure in the HMRC annual report from that year which gives a slightly different figure.

19FOI Response from HMRC to TaxWatch, 21 December 2020

20TaxWatch correspondence with HMRC

21The former Inland Revenue’s selective prosecution policy is set out in R v Inland Revenue Commissioners, ex parte Mead and Cook [1992] STC 482