4          

Personal tax credits error and fraud

4.1    

Since the tax credits error and fraud statistics started to be published, the CIOT’s Low Incomes Tax Reform Group (LITRG) have raised concerns about the accuracy of the statistics both in terms of overpayments and underpayments[10].

4.2    

Our concern is that the underlying practice on the part of HMRC in assessing the cases used in the sample to compile the statistics likely fails to provide an accurate picture of the true extent or causes of error arising in the tax credit system. This means the chances of developing solutions are reduced because the true cause has not been identified.

4.3    

According to the latest published statistics, the sample size for compiling the 2020/21 statistics was 3000. Our understanding is that compliance officers work the cases as normal tax credit enquiry cases and they are required to indicate whether they believe any non-compliance is due to genuine error or to fraud. On average around a quarter of claimants in the sample do not respond to HMRC’s investigations.

4.4    

Our concerns with the process are that:

• If people do not respond, then it is very difficult for the compliance officer to know whether it was error or fraud. If an error, they will not know the source of that error.
• In cases of non-response, HMRC rely on data which may or may not be accurate. For example, HMRC may use real time earnings data to determine a claimant’s income but this will not show all the allowable deductions that the claimant may be able to legitimately make.
• There is no attempt to distinguish between genuine claimant error and ‘contributory error’ – which is error that HMRC has contributed to, for example by not providing guidance or providing misleading guidance.
• Fraud and error are still conflated in headline figures.
• Our understanding is that even where a claimant engages with the enquiry process, the process is no different to normal enquiry cases and there is no in-depth conversation with the claimant to understand their whole award. HMRC appear to focus on the risk area for that specific case when we believe the proper way to build the sample is to look at every aspect of the claim to ensure it is correct. That requires a detailed discussion with the claimant.
• Underpayments may be under-reported in the sample due to the methodology, especially in non-response cases. Claimants often don’t know they can deduct expenses or other items when calculating their income for tax credits. Some claimants do not know that they get additional elements if they have a disability or their child has a disability and receives certain disability payments. These items would not be picked up in a normal compliance investigation and so any findings of underpayments are likely to be accidental or based on data.

4.5    

Tax credits debt still remains a significant issue. £3.4 billion has been transferred to DWP and there is still £2.3billion to transfer[11]. We are concerned about the level of tax credit debt and the transfer of that debt to DWP:

• It is still the case that many tax credit claimants are not aware they have historic tax credit debts. Award notices are notoriously difficult to understand and it is not always obvious. In addition, HMRC have not always followed up with recovery. We are often contacted by people with very old tax credits debts – some dating back to the very early years of the system – which come as a shock to them. In most cases, they are outside of time limits to challenge the overpayments.
• The number of people needing support with these overpayments will continue to increase as more debts are transferred to DWP. Currently old tax credit debts resurface when a universal credit (UC) claim is made and DWP will start recovery of the debt from the UC award. In addition, HMRC are sending old tax credit debts to DWP for recovery even where someone has not claimed UC. This is because DWP can use attachment of earnings powers to recover the money (which HMRC cannot do).
• The process of debt transfer is confusing. People who are transitioning to UC may receive multiple notifications about tax credit overpayment debts. There are complexities around old joint debts that have to be split, around debts that have been partially recovered and debts that are challenged. Some people have reported recovery starting in UC without warning. Claimants with issues relating to these debts are often passed back and forth between DWP and HMRC.
• With the current cost of living crisis, people are likely to find deductions for historic tax credit debts even more difficult to manage.

4.6    

As managed migration starts in earnest, it is important that HMRC and DWP work together to ensure their guidance and communications are robust to help people understand what is happening and their options. For example, the most common questions we get asked when someone finds out they have an old tax credit overpayment are:

• Why have I got this overpayment?
• I haven’t heard from HMRC in 10 years – can they ask me for the money now?
• I don’t think I owe this money, how can I challenge it?

4.7    

It would make sense for letters to direct people to GOV.UK and FAQs which answer these common questions and make it clear which options are available and which are not. In England, the Limitation Act 1980 applies to tax credit debts, so claimants often assume HMRC cannot take any action after a number of years have gone by with no contact. However, that legislation effectively just stops HMRC taking someone to County Court (where the conditions are met); it doesn’t stop them recovering in other ways (such as by passing the debt to DWP and making deductions from UC). Yet the current guidance focuses on how to pay back any overpayment and what to do if you are struggling with the deductions that are being made. Whilst this is necessary information, claimants could be far better supported with fuller guidance around historic debts.

6          

Management of tax reliefs

6.1    

The PAC was right to say, in its 2020 report, ‘Management of tax reliefs’,[13] that the Government knows too little about the tax reliefs it provides. We are pleased that, since the publication of that report, HMRC has significantly improved how it reports on tax reliefs. Their most recent tax relief statistics[14] include headline statistics on the largest reliefs, a detailed analysis for the most ‘significant’ reliefs, and a comparison of the forecast cost of reliefs with the latest estimates of their actual cost.

6.2    

However, the way HMRC presents this information makes it difficult to analyse. There is no total figure for the cost of structural and non-structural reliefs. The statistics provided are at an individual tax relief level, per tax. The estimated costs of non-structural tax reliefs[15] appear on a spreadsheet with eighteen different tabs representing different taxes, each containing numerous individual reliefs. None of these spreadsheets are added up, nor is there a summary page showing totals of each. This can be contrasted with earlier years’ publications, which provided a list of non-structural (‘Tax Expenditures’) and structural reliefs, and their total costs over a five-year period, in the same document.[16] HMRC should be encouraged to report at a ‘macro’ as well as a ‘micro’ level in future.

6.3    

Of the 339 non-structural tax reliefs, less than one-quarter remain uncosted.[17] However, out of around 250 structural reliefs, nearly two-thirds are uncosted.[18] The main reason for this appears to be the availability and accuracy of relevant data.

6.4    

When a relief is introduced, there should be a mechanism to obtain sufficient data to monitor its cost. In many instances, taxpayers will be willing to provide a reasonable level of additional data if the quid-pro-quo is entitlement to a relief. It is important that HMRC’s systems can easily capture and interpret that data.

6.5    

For proposed tax reliefs, we suggest that determining the data provision requirements could represent a mandatory part of the tax consultation process. This would ensure that the means of evaluating the relief are also explored, alongside the pros and cons of the relief itself, as well as the potential administrative burdens and costs of the taxpayer providing the additional information.

6.6    

In our ‘Better Budgets’ report,[19] the CIOT, the Institute for Fiscal Studies (IFS) and the Institute for Government, jointly called for effective and routine post-legislative review of whether tax measures (including reliefs) are achieving their objectives at an acceptable cost.

6.7    

We are pleased that HMRC has recently set out in broad terms its approach to evaluating reliefs,[20] which forms part of its wider evaluation framework,[21] as well as a commitment to publishing its findings. However we would like to see:

• A more systematic approach to evaluation activity, highlighting which reliefs will be reviewed and by when.
• Initial focus on the more ‘expensive’ reliefs, and those costing more than forecast.
• A commitment to publishing the results of the reviews (rather than ‘subject to Ministerial approval’ as currently stated), and on a timely basis.
• The inclusion of structural reliefs in the review process; not just non-structural reliefs.

6.8    

We note that there is a Government Social Research Publication Protocol on publishing research and analysis in government.[22] This seems to include the type of review and evaluation we are considering ie ‘research and analysis to clarify or quantify a policy problem or to evaluate a policy and/or its delivery at pilot or full roll out stage’. However, historically, many research reports are published years after the research has taken place, rendering them almost useless. We would expect to see publication within the 12-week time period specified in the protocol.

6.9    

On the specific matter of pension tax reliefs, in its 2020 report, the PAC recommended that ‘HMRC should within 12 months, have evaluated the impact of pension tax reliefs’. However, the Government disagreed with the Committee’s recommendation, pointing to previous consultations, in particular the responses to the 2015 wide-ranging consultation on pensions tax relief[23] which indicated there was no clear consensus for reform at that time. But much has happened since 2015: a pandemic and a cost of living crisis, and in our view it would be valuable to ascertain that it is the tax relief (rather than simple financial planning) that is incentivising pension saving to an extent sufficient to justify its price tag.

7          

Revenue collection and debt

7.1    

As noted in HMRC’s annual report, the tax debt balance fell from 2020-21 to 2021-22. However the Q1 2022-23 figures show that the debt balance increased again in that quarter. It would be interesting to understand why HMRC think this happened given it was not a big income tax payment deadline time.

7.2    

At 31 March 2022 only 18% of the tax debt was in a managed position (time to pay arrangements (TTPAs) mostly). Whilst this is a 7% increase on pre-pandemic levels, this still seems to be a low percentage. Given that over 90% of debt subject to a TTPA is subsequently paid, surely HMRC should be encouraging the use of TTPAs and getting a much bigger percentage of the overall debt into TTPAs.

7.3    

The anecdotal experience of our members suggests that HMRC’s debt management teams are now considering TTPAs (outside the online application process) far more strictly than in the past (more so than anticipated under ‘Collecting tax debts as we emerge from coronavirus’).[24] In other words they are acting as they did pre-pandemic and as though there was not a cost-of-living crisis with its consequential impacts on businesses. If a TTPA cannot be agreed then often the business has no option but to go into some kind of insolvency/liquidation process. We would be interested to see statistics on the percentage of TTPAs agreed each quarter split between TTPAs arranged following an online application and those arranged after calling or writing to HMRC (either direct or via an agent). Our members report that it seems quite difficult to get a TTPA now unless one is able to apply via the online route.

7.4    

We are also concerned about a lack of consistency in decision-making following a TTPA request (again outside the online system). The answer seems to depend on which person or team at HMRC takes the call.

7.5    

HMRC expects more insolvencies over the coming months, partly due to a catch up from the insolvency hiatus over the pandemic and also from the current tough trading conditions. Whilst clearly some businesses do fail for a multitude of reasons, this represents loss of future tax revenue, jobs etc. If other lenders/creditors are prepared to help by agreeing to a restructure of debts, can HMRC please give full consideration to a TTPA? It would save the business, people’s jobs (so they don’t need to claim benefits) and hopefully lead to future tax receipts.

7.6    

It would be helpful to see up-to-date statistics on HMRC’s usage of their direct recovery of debt (Section 51 and Schedule 8 Finance (No.2) Act 2015) powers and on use of HMRC field force officers and bailiffs. It would be helpful to know whether HMRC consider their approach in this area is working and whether it is cost effective?

7.7    

As at January 2020 HMRC assessed the amount of international tax debt (that owed by taxpayers based abroad) at around £1 billion - 5% of the total tax debt owed to HMRC at the time.[25] International tax debt is harder to collect than domestic tax debt. It would be useful to know what HMRC’s current estimate of international tax debt is and what proportion of total tax debt it constitutes.

8          

The tax gap

8.1    

As noted by the NAO in its report, the ‘tax gap’ in 2020-21 remained steady as a share of the tax that should be collected. The figures suggest HMRC are still collecting about 95 per cent of tax due, which compares well internationally.

8.2    

On the face of it the pandemic has not had a significant effect on the tax gap, though as HMRC and the NAO note, the estimates for 2020 to 2021 are subject to even more uncertainty than usual due to COVID-19. Fraud in covid-support schemes such as the Job Retention Scheme anyway falls outside the tax gap.

8.3    

Error and carelessness

HMRC’s analysis puts tax lost due to taxpayers’ failure to take reasonable care in 2020-21 at £6.1 billion (1.0% of total theoretical liability) and tax lost due to taxpayer error at £3.0 billion (0.5%). That more than £9 billion of the tax gap relates to taxpayers not getting things right through error or carelessness is indicative of the complexity of the tax system.

8.4    

We would like to see a stronger focus from ministers on tax simplification – a simple tax system, with clear rules and easy to navigate guidance will lead to fewer mistakes by both taxpayers and tax authorities. The new government should undertake a more ambitious tax simplification programme and resist the temptation to make major structural changes to the tax system until this is done.

8.5    

The latest analysis means we now have figures for the first two years of ‘Making Tax Digital’ (MTD) – the introduction of compulsory digital record keeping and quarterly digital reporting for VAT by HMRC. HMRC said that this would ‘reduce the amount of tax lost to avoidable errors’.[26] It is hard to judge whether this is actually happening. In the first year of MTD (2019-20) the amount of tax estimated as being lost to taxpayer error rose from £3.1 billion to £3.5 billion. The following year it fell back to £3.0 billion. Failure to take reasonable care increased from £5.9bn to £6.1bn in 2019-20, and remained at that level in 2020-21. But given the extent to which tax gap figures get revised in succeeding years, as well as the particular circumstances in 2020-21, we will need further years of figures before we can judge whether this fall is a lasting trend.

8.6    

The theory behind digitalisation of the tax system is sound – and experience during the pandemic has shown its merits. But our view is that HMRC must thoroughly assess the effectiveness of MTD, as well as whether the administrative burden it is imposing on business is reasonable, before they expand it further.

8.7    

We believe the complexity of the tax system is hindering efforts to digitalise the tax system. The pace of change is such that software solutions are lagging behind. In 2015 the Government published its vision that by 2020 every individual and small business would be able to see and manage their tax affairs through their digital tax account, removing the need for annual tax returns. However, seven years later this vision appears to be a long way from reality, and we note that the project is rated red in the latest annual report on Government infrastructure projects[27].

8.8    

Priority should be given to the development and roll-out of the single customer account based on the principles which were published in 2015.

8.9    

Crime and tax evasion

HMRC’s analysis puts tax lost due to evasion in 2020-21 at £4.8 billion (0.8% of total theoretical liability), tax lost due to criminal attacks on the tax system at £5.2 billion (0.8%) and tax lost due to the hidden economy at £3.2 billion (0.5%).

8.10            

After substantial falls in evasion and criminal attacks since 2013, progress on evasion seems to have stalled and in absolute terms the amounts being stolen from the Exchequer appear to be going up. This suggests to us that while the government’s actions in the early 2010s to put extra resources into identifying and tackling tax evasion and other illegal activity were successful they now need to think more imaginatively and make effective use of the latest techniques for spotting and tackling this activity.

8.11            

We would suggest that the main focus for HMRC should be on continued investment in data analytics to analyse the large amount of data they now have access to, and ensure that they have adequate numbers of compliance officers with the right kind of training to identify and target illegal activity.

8.12            

Non payment

HMRC’s estimates put tax lost due to ‘non-payment’ (largely taxes written off as a result of insolvency) in 2020-21 at £4.9 billion (0.8% of total theoretical liability). This is slightly up on 2019-20 (£4.6 billion; 0.7%) and in absolute terms the highest loss in this area since the tax gap series started in 2005.

8.13            

Non-payment reflects in large part the state of the economy, rather than anything HMRC can control. It increased sharply between 2006-7 and 2009-10, then dipped, but now appears to be on another upward trend.

8.14            

While the 2020-21 figures cover a period when the pandemic closed many businesses the year also saw economic support measures including not just furlough, the self-employed scheme and bounceback loans, but also temporary tax cuts and taxpayers being given more time to pay. The economic support measures undoubtedly helped many business survive the pandemic but the long-term situation is less clear. We will probably have to wait a number of years for the full impact of the pandemic to become clear, as we find out how much of the tax deferred in 2020-21 will ultimately go unpaid due to business failure.

8.15            

Also worth noting is that legislative changes to give HMRC a higher priority in recovering debt that businesses owe when they become insolvent took effect in December 2020. The government predicted these would cut the tax gap by £150-200 million a year in a typical year. Next year’s tax gap figures will be the first opportunity to see whether this is having a measurable impact.

8.16            

Legal interpretation

HMRC’s tax gap report puts tax lost due to ‘legal interpretation’ in 2020-21 at £3.7 billion (0.6% of total theoretical liability). This is down from £5.6 billion (0.8%) in 2019-20. This looks a substantial fall in the legal interpretation part of the tax gap, though it follows a year when it rose significantly, so it is hard to know at this stage if this is a trend or a blip.

8.17            

Legal interpretation is about tax which may or may not legally be due but the system is so complex that neither taxpayer nor HMRC yet know! 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 supply. The £3.7 billion figure makes the spurious assumption that HMRC’s view of the law is right in 100 per cent of cases. We query why this category is part of the tax gap at all.

8.18            

This is another argument for a simpler tax system. If even HMRC are uncertain of the law to the tune of £3.7 billion what hope is there for the ordinary taxpayer faced with the same complexity in the tax system?

8.19            

We note that, from April 2022, this element of the tax gap has been targeted with a new requirement for large businesses to notify HMRC if they have adopted a tax treatment where the business believes that HMRC, or a tribunal or court, may not agree with their interpretation of the legislation, case law, or guidance.4 We will watch with interest to see if this has a measurable impact.

8.20            

Avoidance

HMRC puts tax lost due to avoidance in 2020-21 at £1.2 billion (0.2% of total theoretical liability). These numbers are both unchanged on 2019-20. The first tax gap figures (2005-6) put the ‘avoidance gap’ at £4.8 billion (1.1% of total theoretical liability).

8.21            

Tackling avoidance is the big success story of the government’s efforts to tackle the tax gap. The avoidance gap is just a quarter of what it was 15 years ago, less than a fifth as a share of the total tax that should be collected. It is noteworthy that more than twice as much is lost to errors as to avoidance.

8.22            

Offshore tax gap

We note that HMRC plans to publish a new stand-alone offshore tax gap, estimating the amount of offshore tax not being correctly reported by UK taxpayers, for the ‘Measuring tax gaps 2023 edition’.[28]

8.23            

This is welcome. The Common Reporting Standard (CRS) has given HMRC huge amounts of data on assets and income outside the UK, but until now it has been unclear whether HMRC already knew about most of these cases, or whether they had the appetite or capability to check. It now appears that we will soon have a better idea of the scale of offshore evasion and avoidance than we have had before.

9          

HMRC customer service performance

9.1    

We continue to be concerned about the difficulties both advisers and taxpayers face getting timely responses and action from HMRC. HMRC’s performance standards need to be improved if the tax authority is to play its essential role in supporting taxpayers and businesses.

9.2    

In a letter sent in June to HMRC’s Director General for Customer Services, sent jointly with the ATT, ICAEW and ICAS, we stated: ‘We all continue to be contacted by our members about poor HMRC service levels, and asking what we are doing about it… Issues seem to exist across a range HMRC services, and can change over time. While not a comprehensive list, the following issues get raised with us most frequently: self-assessment registrations and refunds, correspondence about VAT grouping and the option to tax, section 455 refunds, responses to technical queries, corporation tax post and returns not processed. Feedback also indicates that performance on the agent dedicated line appears to be erratic; there are sometimes very long waiting times, and calls can be cut off unexpectedly and promised call backs do not happen. Members are also concerned that helpline hours have not been restored to pre-pandemic levels.’[29]

9.3    

Most of the problems identified above continue, and new ones have emerged as HMRC’s resources are redeployed onto other priorities.

9.4    

It is not simply that targets are being missed. Some targets are not fit for purpose even if met. For example the target for section 690 determinations (under which HMRC give an employer permission to exclude a portion of the pay of an employee who is not UK-resident from the operation of UK PAYE) is four months. That in practice means at least five months, and more likely six months, that a taxpayer will suffer double taxation, as their pay will usually be subject to withholding in the country that they are resident in during this period.

9.5    

We recognise that HMRC, like everyone else, can always be challenged to become more efficient. As a critical friend of HMRC, we support policies to improve its effectiveness and efficiency. However, in the move to digitalise the tax system we are concerned that staff numbers within HMRC are being cut in anticipation of securing savings from digitalisation when these savings have not yet been realised, with the result that delivery of existing performance standards has been compromised.

9.6    

The government should undertake to maintain HMRC’s existing resources and capabilities, coupled with a more ambitious mandate to improve standards of basic performance across the full range of HMRC activities including answering telephone queries, dealing promptly with correspondence, investigation and compliance activity and timely processing of new tax registrations and agent authorisations, as well as ensuring that these improvements are sustained for the remainder of the life of this Parliament. The investment required to sustain these improvements is urgent and essential, not just for taxpayers but for the sustainability of future government revenues, so should not be put at risk by any new wider public spending cuts or constraints.

9.7    

We have a particular concern about HMRC making it difficult and time-consuming for digitally excluded taxpayers to comply with their tax affairs. In addition to those who are digitally excluded by necessity, our Low Incomes Tax Reform Group (LITRG) have had a number of people contact them over recent times saying they cannot pass HMRC security checks to be able to deal with their tax affairs online, so they are forced into doing things on paper in that case even though they are willing to comply online. These ‘enforced excluded’ then find themselves in a frustrating loop of trying to get hold of a paper form, with HMRC telling them this could take up to 15 working days to receive. (And while they are on hold to HMRC to speak to someone, they keep getting told how easy it is to do things online and that this is the best way to deal with your taxes!)

10      

Repayments to taxpayers

10.1            

A perhaps surprising omission from HMRC’s annual report is the problem of unacceptable practices by some repayment agents.

10.2            

Large numbers of angry and unhappy taxpayers have contacted LITRG raising serious concerns including (but not limited to) –

• Taxpayers believing that they have simply filled in an enquiry form online – only to later realise later that an ‘electronic’ deed of assignment has been generated and been lodged with HMRC
• People paying a fee for assistance to claim a specific refund, but in the process inadvertently signing a wider deed of assignment, meaning unrelated refunds are also then diverted to the tax refund company
• Some tax refund companies charging people anything up to 50 per cent of the amount to release refunds to the taxpayers (or sometimes not releasing them at all)
• Deeds of assignment lodged with HMRC where the taxpayer says the signature on the form is not theirs
• Taxpayers interacting with tax refund companies when they thought they were dealing with HMRC direct  
• Tax refund companies being unresponsive when taxpayers try to contact them  
• Concerned about protection of data where the repayment agent requests significant personal information in order to process payment of a refund to them 

10.3            

We welcome HMRC’s acknowledgement that there is a serious problem here, and the launch of a consultation in June ‘to consider ways to better protect taxpayers from Repayment Agents’, with proposals including ensuring repayment agents display material information to prospective clients in line with existing rules for financial services; considering whether HMRC should restrict the use of assignments and whether HMRC should require repayment agents to formally register with HMRC to allow credentials to be checked and swifter action taken where instances of poor practice are found.

10.4            

However we are concerned that any action could take time to materialise, and taxpayers could lose out in the meantime. We urge HMRC to consider what more they can do to protect people in the short-term.

10.5            

For example, there is a serious question over whether some of the deeds of assignment that HMRC are currently accepting are valid in accordance with HMRC’s own guidance because the process under which signatures have been collected and attached to assignment documents is not always transparent. This means taxpayers may not have seen, understood or approved the deed. Greater scrutiny is urgently needed on HMRC’s part before accepting such deeds.

10.6            

Especially disturbingly, we have heard of a number of cases where claims have been made fraudulently in the name of a taxpayer by a refund company where the taxpayer had no knowledge the specific types of claim were being made. The taxpayers believed the company were helping them claim relief for work expenses. Instead, these have typically related to Enterprise Investment Scheme (EIS) claims. The refunds have been paid to the refund companies but HMRC, realising the claims are fraudulent, have pursued the taxpayer. This surely cannot be appropriate.

11      

Acknowledgement of submission

11.1            

We would be grateful if you could acknowledge safe receipt of this submission, and ensure that the Chartered Institute of Taxation is included in the List of Respondents when any outcome of the consultation is published.