Child and Working Tax Credits error and fraud statistics, tax year 2020 to 2021
Updated 11 April 2024
© Crown copyright 2024
This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: psi@nationalarchives.gov.uk.
Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.
This publication is available at https://www.gov.uk/government/statistics/child-and-working-tax-credits-error-and-fraud-statistics-2020-to-2021/child-and-working-tax-credits-error-and-fraud-statistics-tax-year-2020-to-2021
Original and revised estimates for tax year 2019 to 2020
The first assessment of the level of error and fraud for tax year 2019 to 2020 was published in July 2021, with a central estimate that the level of error and fraud favouring the claimant was around £880 million, or 5% of finalised tax credits entitlement.
The publication explained that as in all previous years, the estimates were based on incomplete data. In particular, some of the cases used in the estimation were still under investigation, and the compliance officer decisions that underpinned the error and fraud estimates were subject to appeal by households. Because of these factors HMRC revisits the estimates each year to take account of any new information received after the original publication and commits to republish the estimates when complete data on all sampled cases is available.
We have now revisited the tax year 2019 to 2020 estimates to take account of new information, and estimate that the level of error and fraud favouring the claimant now stands at £940 million. Estimated error and fraud favouring HMRC is £170 million.
- the final central estimate for claimant favour error and fraud in tax year 2019 to 2020 is 5.3%, up from the first estimate of 5%
- the final central estimate for HMRC favour error in tax year 2019 to 2020 is 0.9%, up from the first estimate of 0.8%
The first and final central estimates for tax year 2019 to 2020 can be found in tables 1.1 and 1.2, which also illustrate 95 per cent confidence intervals associated with these central estimates. When referring to previous year figures for comparison, the final central estimates for tax year 2019 to 2020 will be used.
Key figures
Figure 1: tax credits error and fraud at a glance in tax year 2020 to 2021
Figure 1 shows that:
- tax credits error and fraud in favour of the claimant is estimated to be £710 million
- this is 4.7% of total tax credits entitlement
Figure 2: tax credits error and fraud as a proportion of entitlement - central estimates, tax year 2011 to 2012 to tax year 2020 to 2021
Figure 2 shows that:
- the central estimate of tax credits error and fraud in favour of the claimant has decreased from 5.3% of entitlement in the tax year 2019 to 2020 to 4.7% in the tax year 2020 to 2021, although this change is not statistically significant
- the central estimate of the claimant favour error and fraud rate has been between 4.4% and 5.5% since tax year 2012 to 2013
Error and fraud rates since tax year 2006 to 2007 can be found in annex B.
Figure 3: tax credits error and fraud by risk category - central estimates, tax year 2016 to 2017 to tax year 2020 to 2021
Figure 3 shows that:
- over the past 5 years, the biggest reasons for claimant favour tax credits error and fraud have been Income, Work and Hours and Undeclared Partner and this has continued in tax year 2020 to 2021
- the Income risk group is the biggest reason for HMRC favour error in tax year 2020 to 2021 and previous years
- the overall value of error and fraud has reduced over time in line with reductions in the tax credits population, although the income risk saw a proportionately larger fall in tax year 2020 to 2021
A detailed breakdown and description of the different risk groups is provided in section 2.
Estimates of error and fraud in tax credits for tax year 2020 to 2021
Section 1: estimated levels of error and fraud
This report presents results from the Error and Fraud Analytical Programme (EFAP), which is designed to measure error and fraud (E&F) in finalised awards across the tax credits population.
This publication will be of particular interest to academics, policy makers and operationally within HMRC. Historical error and fraud estimates dating back to tax year 2006 to 2007 can be found in annex B.
The details presented in the following tables are based on a sample of cases and hence there are margins of error associated with these estimates. Therefore, tables 1 to 4 also illustrate the 95 per cent confidence intervals associated with these central estimates. More details about the sampling methodology can be found in annex A.
Estimates in the tables are rounded to the nearest £10 million or 10,000 in tables 2, 4, 5 and for all the overall totals in the other tables. The breakdowns in the other tables are rounded to the nearest £5 million or 5,000. The error and fraud rates are rounded to the nearest 0.1% in tables 1 and 3. Note that the numbers in these tables may not sum to the totals due to rounding.
Error and fraud favouring the claimant refers to cases where the claimant has been found to be non-compliant in a way that has led HMRC to pay them more tax credits than they were entitled to for the year, in other words there was a monetary gain for the claimant and a monetary loss for HMRC.
Error favouring HMRC refers to cases where the claimant has been found to be non-compliant in a way that has led HMRC to pay them less tax credits than they were entitled to for the year, in other words there was a monetary gain for HMRC and a monetary loss for the claimant.
For tax year 2020 to 2021, the central estimate for the error and fraud rate has decreased from 5.3% in the previous year to 4.7% in the current year, as shown in table 1, although the fall is not statistically significant.
Table 1.1: total error and fraud as a proportion of finalised entitlement (%), first estimate for tax year 2019 to 2020
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error and fraud favouring the claimant | 4.5% | 5% | 5.5% |
| Estimated error favouring HMRC | 0.7% | 0.8% | 1% |
Table 1.2: total error and fraud as a proportion of finalised entitlement (%), final estimate for tax year 2019 to 2020
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error and fraud favouring the claimant | 4.8% | 5.3% | 5.9% |
| Estimated error favouring HMRC | 0.8% | 0.9% | 1.1% |
Table 1.3: total error and fraud as a proportion of finalised entitlement (%), tax year 2020 to 2021
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error and fraud favouring the claimant | 4.1% | 4.7% | 5.4% |
| Estimated error favouring HMRC | 0.6% | 0.8% | 0.9% |
Table 2 shows that:
- there were an estimated 370,000 tax credits claims containing error and fraud in the claimant’s favour in tax year 2020 to 2021
- the total value of this error and fraud is estimated at £710 million
- there were an estimated 310,000 claims containing error favouring HMRC, with a total estimated value of £120 million
It is important to note that due to the introduction of Universal Credit, the total number and entitlement of tax credits awards is decreasing year-on-year, and comparisons of the absolute number of awards in error and fraud and absolute value of error and fraud to previous years should consider the decreasing size of the tax credits population.
Table 2.1: overall level of error and fraud, tax year 2020 to 2021 - number of instances
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error and fraud favouring the claimant | 340,000 | 370,000 | 400,000 |
| Estimated error favouring HMRC | 280,000 | 310,000 | 340,000 |
Table 2.2: overall level of error and fraud, tax year 2020 to 2021 - amount
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error and fraud favouring the claimant | £620m | £710m | £810m |
| Estimated error favouring HMRC | £90m | £120m | £140m |
When Claimant Compliance Officers find error and fraud in EFAP cases they assess whether they believe it was due to genuine error or fraud. To be classified as fraud, a caseworker needs to have found evidence that the claimant deliberately set out to misrepresent their circumstances to get money to which they are not entitled (for example claiming for a child that does not exist).
Error covers instances where there is no evidence of the claimant deliberately trying to deceive HMRC. It covers a range of situations, including cases where a claimant inadvertently over-claims because they simply provided HMRC with the wrong information. It could also cover a situation where the correct information has been provided but this information has been incorrectly processed by HMRC.
Tables 3 and 4 show that error makes up more than 95% of the total value of error and fraud in claimant favour, with less than 5% coming from fraudulent activity.
Table 3: error and fraud favouring the claimant as a proportion of finalised entitlement (%), tax year 2020 to 2021, split out into separate error and fraud components
| Lower bound | Central estimate | Upper bound | |
|---|---|---|---|
| Estimated error favouring the claimant | 3.9% | 4.6% | 5.2% |
| Estimated fraud favouring the claimant | 0% | 0.1% | 0.3% |
| Total | 4.1% | 4.7% | 5.4% |
Table 4.1: level of error and fraud favouring the claimant, tax year 2020 to 2021 split out into separate error and fraud components - numbers of instances
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error favouring the claimant | 330,000 | 360,000 | 390,000 |
| Estimated fraud favouring the claimant | 0 | 10,000 | 10,000 |
| Total | 340,000 | 370,000 | 400,000 |
Table 4.2: level of error and fraud favouring the claimant, tax year 2020 to 2021 split out into separate error and fraud components - amount
| Lower Bound | Central Estimate | Upper Bound | |
|---|---|---|---|
| Estimated error favouring the claimant | £590m | £690m | £780m |
| Estimated fraud favouring the claimant | £0m | £20m | £40m |
| Total | £620m | £710m | £810m |
Error can be made by both the claimant and HMRC, and table 5 provides a breakdown into claimant error and HMRC error. It shows that the value of the error made by claimants is substantially higher than the value of the errors made by HMRC. This is consistent with previous years.
Table 5: overall level of error split between claimant error and HMRC error - central estimates, tax year 2020 to 2021
| Claimant error - number | Claimant error - amount | HMRC error - number | HMRC error - amount | |
|---|---|---|---|---|
| Estimated error favouring the claimant | 360,000 | £690m | under 10,000 | under £10m |
| Estimated error favouring HMRC | 300,000 | £110m | 10,000 | £10m |
Figure 4: Distribution or error and fraud favouring the claimant by value of finalised award - central estimates, tax year 2020 to 2021
Figure 4 shows that:
- the majority of tax credits awards with claimant favour error and fraud have an award value of less than £10,000
- the £0 to £2,000 award band has the highest number of claimants in error and fraud, at 90,000
- the £0 to £2,000 band has the highest value of error and fraud at £230 million
- note that the value of the award shown in figures 4 and 5 is the value of the finalised award and includes the value of error and fraud
Figure 5: Distribution of error and fraud favouring HMRC by value of finalised award - central estimates, tax year 2020 to 2021
Figure 5 shows that:
- the majority of tax credits awards with HMRC favour error have an award value below £10,000
- the £2,000 to £4,000 band has the highest number of claimants in error at 65,000
Figure 6: Distribution of claimant favour error and fraud amounts - central estimates, tax year 2020 to 2021
Figure 6 shows that:
- the majority of awards with claimant favour error and fraud have an error and fraud value below £2,000
- the £500 to £1,000 error and fraud band has the highest number of awards at 90,000
- 25,000 awards have an error and fraud value over £6,000, and these account for £230 million in error and fraud
Figure 7: Distribution of HMRC favour error amounts - central estimates, tax year 2020 to 2021
Figure 7 shows that:
- the majority of HMRC favour error has a value of less than £500, with fewer than 5,000 awards having HMRC favour error of over £3,000
- the largest proportion of total HMRC favour error is in awards with an error value of £100 to £500 at 155,000 awards and £35 million error
Section 2: reasons for error and fraud
Error and fraud can enter the system due to a range of circumstances being incorrectly reported. At a high level there are 7 key risk categories. These are:
- Income - inaccurately reporting income
- Undeclared Partner - making a single claim instead of a joint claim
- Childcare Costs - incorrectly reporting childcare costs
- Children - incorrectly including or excluding children or young persons on a claim
- Work and Hours - overstating or understating hours worked
- Disability - incorrectly reporting disability status
- Other - risks that cannot be assigned to one of the other high level categories. This category includes residency and situations where a partner has been declared but is not present
The associated level of error and fraud for each risk category can be found in tables 6 and 7. Note that some claimants will have more than one risk identified in their claim so the numbers will not sum to the total number of awards presented in the other tables. The confidence intervals for the estimated value of error and fraud for each risk group are shown in the column after the value.
Table 6: reasons for claimant favour error and fraud - central estimates, tax year 2019 to 2020 and tax year 2020 to 2021
| Number of instances - 2019 to 2020 | Amount - 2019 to 2020 | Confidence intervals - 2019 to 2020 | Number of instances - 2020 to 2021 | Amount - 2020 to 2021 | Confidence intervals - 2020 to 2021 | |
|---|---|---|---|---|---|---|
| Income | 250,000 | £325m | +/-£75m | 175,000 | £180m | +/-£45m |
| Undeclared Partner | 50,000 | £175m | +/-£65m | 40,000 | £155m | +/-£65m |
| Childcare costs | 95,000 | £90m | +/-£30m | 50,000 | £55m | +/-£20m |
| Children | 55,000 | £60m | +/-£25m | 50,000 | £70m | +/-£25m |
| Work and Hours | 160,000 | £245m | +/-£55m | 80,000 | £165m | +/-£45m |
| Disability | 30,000 | £45m | +/-£25m | 25,000 | £50m | +/-£20m |
| Other | under 5,000 | £5m | +/-£10m | 45,000 | £40m | +/-£15m |
| Total | 640,000 | £940m | 460,000 | £710m |
Figure 8: value of claimant favour error and fraud (£ million) - central estimates by risk category, with confidence intervals, tax year 2019 to 2020 and tax year 2020 to 2021
Table 6 and figure 8 show that:
- most of the error and fraud favouring the claimant is due to the Income, Work and Hours and Undeclared Partner categories
- the Income risk has the highest number of awards in error and fraud
- there has been a significant reduction in the value of error and fraud from the income risk in tax year 2020 to 2021
Table 7: reasons for HMRC favour error - central estimates, tax year 2019 to 2020 and tax year 2020 to 2021
| Number of instances - 2019 to 2020 | Amount - 2019 to 2020 | Confidence intervals - 2019 to 2020 | Number of instances - 2020 to 2021 | Amount - 2020 to 2021 | Confidence intervals - 2020 to 2021 | |
|---|---|---|---|---|---|---|
| Income | 335,000 | £130m | +/-£30m | 275,000 | £90m | +/-£20m |
| Undeclared Partner | 0 | £0m | +/-£0m | 0 | £0m | +/-£0m |
| Childcare costs | 10,000 | under £5m | +/-£0m | 5,000 | under £5m | +/-£0m |
| Children | 15,000 | £15m | +/-£10m | 5,000 | £10m | +/-£5m |
| Work and Hours | 25,000 | £10m | +/-£10m | 15,000 | £10m | +/-£10m |
| Disability | 15,000 | £10m | +/-£10m | 5,000 | £5m | +/-£0m |
| Other | 0 | £0m | +/-£0m | 30,000 | £5m | +/-£5m |
| Total | 400,000 | £170m | 340,000 | £120m |
Figure 9: value of HMRC favour error (£ million) - central estimates by risk category, with confidence intervals, tax year 2019 to 2020 and tax year 2020 to 2021
Table 7 and figure 9 show that:
- error favouring HMRC is mainly due to the Income risk category, which accounts for around 80% of HMRC favour error
- the other risk groups contributing to HMRC favour error are Children, Work and Hours and Disability
- it is not possible to have Undeclared Partner error in HMRC favour
We are able to break down the larger risk categories further to see the underlying reasons for the error and fraud entering the system. There is not enough information on the smaller risk categories to provide a further breakdown.
Income risk
Income error and fraud occurs when a claimant under or overstates their actual income. This can come from a range of different sources as can be seen in table 8 and figure 10 below.
Table 8: Income error and fraud favouring the claimant broken down by different sources of income - central estimates, tax year 2020 to 2021
| Number of instances | Amount of error and fraud | |
|---|---|---|
| Self-Employed Income | 50,000 | £60m |
| Employed Income | 20,000 | £25m |
| Social Security Benefits | 45,000 | £35m |
| Benefits in kind | 10,000 | £10m |
| Rents | 10,000 | £5m |
| Dividends | 10,000 | £10m |
| Other | 30,000 | £35m |
| Total | 175,000 | £180m |
Figure 10: Income error and fraud favouring the claimant broken down by sources of income - central estimates, tax year 2020 to 2021
Table 8 and figure 10 show that:
- self-employed income has the highest error and fraud among the different sources of incorrect income at 50,000 awards and £60 million error and fraud
- employed income has the highest value of error and fraud per award at £1,250 per case
Error and fraud can occur because the claimant has not informed HMRC about any of their income from a certain source or because they have under or overstated the amount of income they receive. This varies depending on the type of income as can be seen in figure 11.
Figure 11: Proportion of error and fraud broken down by declaration of different sources of income - central estimates, tax year 2020 to 2021
Figure 11 shows that:
- overall there is approximately a 50-50 split between claimants understating their income and claimants not informing HMRC of their income. However, this varies substantially by type of income
- the main sources of undeclared income are benefits in kind and social security benefits
- claimants are more likely to understate self-employed income than not declare it at all
Work and Hours risk
Work and Hours error and fraud can occur when a claimant provides an incorrect start or end date for their qualifying employment, or provides an incorrect assessment of their weekly hours. We are able to break down Work and Hours into those categories and also distinguish between employed and self-employed work. This can be seen in table 9.
The Work and Hours risk group includes the commercial and with a view to profit (C&P) test, an assessment of whether a self-employed WTC claimant is engaged in qualifying remunerative work that is structured, regular and ongoing. This was introduced in tax year 2016 to 2017.
Table 9.1: Work and Hours error and fraud (employed) favouring the claimant broken down by reason - central estimates, tax year 2020 to 2021
| Number of instances | Amount of error and fraud | |
|---|---|---|
| Incorrect start and/or end dates | 25,000 | £45m |
| Overstated hours | 40,000 | £75m |
| Other (Employment ceased & fictitious employment) | under 5,000 | under £5m |
| Total (employed) | 65,000 | £120m |
Table 9.2: Work and Hours error and fraud (self-employed) favouring the claimant broken down by reason - central estimates, tax year 2020 to 2021
| Number of instances | Amount of error and fraud | |
|---|---|---|
| Overstated Hours | 10,000 | £20m |
| C&P | 5,000 | £25m |
| Total (self-employed) | 15,000 | £45m |
Figure 12: Work and Hours error and fraud favouring the claimant broken down by reason - central estimates, tax year 2020 to 2021
Table 9 and figure 12 show that:
- most Work and Hours error and fraud is from employed work when claimants overstate the number of hours worked (the claimants are working fewer hours than they originally claim which affects their eligibility for Working Tax Credits)
- C&P has a higher value of error and fraud per award than other reasons for Work and Hours error and fraud at £5,000 per case
Further information
Child Tax Credit (CTC) and Working Tax Credit (WTC) were introduced in April 2003. They are flexible systems of financial support designed to deliver support as and when a family needs it, tailored to their specific circumstances. They are part of wider government policy to provide support to parents returning to work, reduce child poverty and increase financial support for all families.
The flexible design of the system means that as families’ circumstances change, so does their (daily) entitlement to tax credits. This means tax credits can respond quickly to families’ changing circumstances, providing support to those that need it most.
Tax credits are based on household circumstances and can be claimed jointly by members of a couple, or by singles. Entitlement is based on the following factors: age, income, hours worked, number and age of children, childcare costs and disabilities.
The introduction of Universal Credit has meant that since 1 February 2019, new claims to tax credits are no longer accepted, except in a limited number of specific circumstances.
For further information on who can claim tax credits please refer to the GOV.UK website: https://www.gov.uk/topic/benefits-credits/tax-credits
For the tax year 2020 to 2021, this exercise took a stratified random sample of 3,000 cases which were selected to be representative of the tax credit population. These cases were taken up for examination by claimant compliance officers who worked the cases as they would for any other compliance enquiry. The sample is stratified because of the size and diversity of the claimant population and the possible variation in compliance risk. More details about the sampling methodology can be found in Annex A.