Official Statistics

1. Tax gaps: Summary 

Published 18 December 2012

 Introduction  

This annex provides further detail on the data, methods and quality considerations underpinning the tax gap estimates published in ‘Measuring tax gaps 2026 edition’. It explains the different approaches used to estimate individual components of the tax gap and the main sources of uncertainty affecting those estimates. 

Methods used to estimate the tax gap 

Methodological approaches 

There are several methodological approaches used to estimate the tax gap. HMRC uses the most appropriate approach for each component, depending on the nature of the tax, the available data and the level of detail required. 

Top-down methods use external independent data sources to estimate total consumption of taxable products to calculate the total theoretical liabilities. An example of this is the VAT gap. 

Bottom-up methods include several techniques. 

• random enquiry programmes — these involve undertaking compliance checks for a randomly selected sample of customers and scaling up the findings from the sampled cases to the relevant whole population
• statistical methods — unlike random enquiry programmes, these use risk-based compliance checks that are not representative of the whole population and require statistical methods to scale up the results to the whole population
• population surveys — we use results from a bespoke research survey to estimate part of the hidden economy tax gap
• management information — these methods use management information such as:
  • risk registers — a list of identified tax risks, together with information such as estimated value, nature and status
  • data extracted from accounting systems
  • other databases or systems used to manage HMRC’s business

The total tax gap is estimated using established statistical and illustrative methods. Illustrative methodologies, formerly called experimental methodologies, are used to produce estimates where there is no direct measurement data. For these tax gap components, we use the best available data and simple models to build an estimate of the tax gap. 

We employ the most appropriate methodology for each tax gap component, based on the factors listed below: 

• availability of quality HMRC data 

• availability of quality independent data 

• structure of the tax regime 

• cost and impact for both HMRC and taxpayers 

• level of granularity required 

Generally, following good international practice, we use ‘top-down’ methodologies for indirect taxes and ‘bottom-up’ methodologies for direct taxes. The tax gap estimates may, however, also be produced by compiling the results from a combination of 2 or more methods. 

Methods used for different tax gap components 

Table A1.1 below shows the general methodological approach used to estimate each tax gap component. 

Table A1.1: Tax gap methodologies 

Top-down	Bottom-up (management information)	Bottom-up (statistical and survey)	Bottom-up (random enquiries)	Illustrative
Alcohol duties	Alcohol duties	PAYE mid-sized businesses	PAYE small businesses	PAYE large businesses
VAT	Hidden economy	Hidden economy	SA business and non-business	SA large partnerships
Tobacco duties	—	CT mid-sized businesses	CT small businesses	Stamp taxes
—	—	CT large businesses	Diesel duties	Other excise
—	—	Inheritance Tax	—	Other remaining taxes
—	—	Avoidance	—	—

Notes for Table A1.1 

1. Alcohol duty gaps are produced using both top-down and bottom-up methodologies. 

2. Hidden economy tax gaps for ghosts and moonlighters are produced using bottom-up management information and bottom-up statistical and survey methodologies. 

Tax gap by methodology 

Figure A1.1 below shows a summary of the tax gap by methodology. A degree of assumption and judgement has been applied to attribute some elements of the tax gap to methodology types, especially where a combination of methods is used. 

Figure A1.1: Tax gap by methodology for 2024 to 2025 (£ billion) 

Methodology	Total
Bottom-up (management information)	1.7
Bottom-up (random enquiries)	30.8
Bottom-up (statistical and survey)	5.0
Illustrative	7.3
Top-down	14.2

We have increasingly used established top-down or bottom-up methods to estimate the tax gap, reducing reliance on illustrative approaches. The shares presented are based on the absolute value of the tax gap, rather than the number of estimates. This means a small number of estimates can account for a large share of the total if they relate to large components of the tax gap. Comparing figures over time is challenging due to ongoing revisions from both methodological improvements and new data, which can affect the share of the tax gap across methodologies. 

Table A1.2 shows that the proportion of the tax gap estimated using established methods rose from 76% in 2019 to 88% in the 2026 edition.

Table A1.2 : Share of total tax gap by established and illustrative methodologies in ‘Measuring tax gaps’ editions 

Measuring tax gaps edition	Illustrative	Established
MTG19	24%	76%
MTG20	15%	85%
MTG21	14%	86%
MTG22	21%	79%
MTG23	16%	84%
MTG24	14%	86%
MTG25	13%	87%
MTG26	12%	88%

Accuracy and reliability 

Our tax gap estimates are official statistics produced to the highest levels of quality and adhere to the UK Statistics Authority’s Code of Practice for Statistics framework. This framework ensures statistics are trustworthy, good quality, valuable, and provides producers of official statistics with the detailed practices they must commit to when producing and releasing official statistics. 

A Measuring tax gaps quality report accompanies this statistical release, providing information about the quality of outputs as set out by the Code of Practice for Statistics. 

The figures presented in the ‘Measuring tax gaps 2026 edition’ are our best estimates based on the information available, but there are sources of uncertainty and potential error. For this reason, it is best to focus on the trend in the results rather than the absolute numbers when interpreting findings. 

Accuracy 

Accuracy refers to the closeness of estimates to the true values they are intended to measure. Due to the methodologies used, uncertainty is an inherent aspect of all tax gap estimates. Uncertainty relates to a range of possible factors that can affect the accuracy of a statistic, including the impact of measurement or sampling error (related to sample surveys) and all other sources of bias and variance that exist in a data source. 

Reliability 

Reliability refers to the closeness of estimated values with subsequent estimates. The methodologies used to estimate tax gaps are subject to regular review and can change from year to year due to improvements in methodologies and data updates. These can result in revisions to any of the previously published estimates. Estimates are made on a like-for-like basis each year to enable users to interpret trends. Where data sources change over time, every effort has been made to ensure consistency in the time series, but this is another potential source of uncertainty. 

Uncertainty 

Statistical uncertainty is caused by 2 factors: 

• sampling error — errors that arise because the estimates rely on information collected from a sample, rather than from the whole population; sampling error can lead to year-on-year fluctuations in the tax gap estimates that do not reflect true changes in the size of the tax gap

• bias or non-sampling error — systematic errors where the modelling assumptions or errors in the data lead to estimates that are consistently either too low or too high. 

Where possible, HMRC has estimated the likely impact of sampling errors by calculating statistical confidence intervals and presented these where possible in the online tables. These give margins of error within which we would expect the true value lies 95% of the time, if there were no systematic errors. They provide an indication of the extent to which changes in the estimates between years can be confidently interpreted as true changes. They do not take account of systematic errors that might lead the central estimate to be too low or too high over the whole series. 

Systematic error is less straightforward to resolve as it is not defined by statistical assessments that allow for easy interpretation. To give an indication of the effect of these biases, upper and lower bounds are produced where possible for specific tax heads. The central estimate is reported in the main chapters of ‘Measuring tax gaps 2026 edition’. The upper and lower bounds are available in the online tables. 

Tax gap uncertainty assessment 

To show the uncertainty of tax gap estimates in a systematic and transparent way, we assign an overall uncertainty rating for each tax gap component ranging from ‘very low’ to ‘very high’. Table A1.3 provides a definitionof the overall uncertainty ratings. 

Table A1.3: Overall uncertainty rating guide

Overall uncertainty rating	Definition
Very low	Very high confidence that the estimate is close to the actual
Low	High confidence that the estimate is close to the actual
Medium	The estimate is likely to be close to the actual but there is a possibility that it is different
High	Low confidence that the estimate is close to the actual
Very high	Very low confidence in the estimate — the actual is likely to be markedly different

To determine the uncertainty ratings of each tax gap component, we assess the uncertainty arising from each of 3 sources: the model scope, the methodology used, and the data underpinning the estimate. 

In assessing model scope, we evaluate each estimate’s methodology against relevant criteria including: 

• coverage of the appropriate tax base and taxpayer population
• whether the model accounts for all potential forms of non-compliance
• any overlap between any 2 components of the tax regime

Table A1.4 provides a guide on the interpretation of the uncertainty rating for model scope. 

Table A1.4: Model scope uncertainty rating guide 

Rating	Model scope
Very low	Accounts for whole potential tax base and population. Accounts for all potential forms of non-compliance. No overlap with other estimates.
Low	Accounts for nearly all of the tax base and population. Accounts for nearly all potential forms of non-compliance. No overlap with other estimates.
Medium	Accounts for most of the tax base and population. Accounts for most potential forms of non-compliance. No overlap with other estimates.
High	Missing some of the tax base and population. Some forms of non-compliance are risks not being accounted for. Some potential for overlap with other estimates.
Very high	Almost all the tax base and population are missing and almost no risks are being accounted for. Likely overlap with other estimates.

In assessing the methodology used, we evaluate each estimate’s methodology against relevant criteria including: 

• complexity and challenges of the model, including the quality and impact of its assumptions
• potential sources of bias, such as methodological bias, sampling errors, or reliability concerns 
• model uncertainty factors, including volatility, margin of error, confidence intervals, and unaccounted external risks

Table A1.5: Model methodology uncertainty rating guide 

Rating	Model methodology
Very low	Few or no sensitive assumptions. The model is logical and straight forward with no complex analytical challenges. Model risks are robustly mitigated.
Low	Some sensitive assumptions and the model is logical and straightforward with few complex analytical challenges. Model risks are mitigated.
Medium	Some sensitive assumptions and challenges. The model is analytically complex with multiple stages. Some external risks with most unlikely and with good risk mitigation.
High	Multiple sensitive and unverifiable assumptions. The model may be too analytically complex or simplistic. Many risks, some of which have weak mitigation in place.
Very high	Assumption based sensitive model, most of which are unverifiable. Many risks with no strong mitigation.

In assessing both HMRC and third-party data underpinning an estimate, we evaluate each estimate’s methodology against relevant criteria including: 

• data suitability for purpose
• understanding of data
• sensitivity analysis

Table A1.6: Model data uncertainty rating guide 

Rating	Model data
Very low	High quality and assured data used throughout. Completely understood data and highly suitable for use in tax gaps.
Low	High quality data (small amount of unsensitive projecting/non-detection multiplier uplifts). Mostly understood data and suitable for use in tax gaps.
Medium	Not complete data (a lot of assumptions but all logical and verifiable). Some data not fully understood but still acceptable for use in tax gaps with some caveats.
High	Little suitable data and of poor quality. Most of the data is not properly understood, many caveats but no alternative.
Very high	No suitable data, what is available is not well understood and is of low quality.

Table A1.7 below shows the uncertainty rating for 2024 to 2025 for each tax gap component; by model scope, methodology used, and data underpinning the estimate, and the overall uncertainty rating.  The overall uncertainty rating considers the relative importance of each uncertainty source for each tax gap estimate component. 

Table A1.7: Tax gap model uncertainty ratings, 2024 to 2025

Tax gap model	Scope	Methodology	Data	Overall uncertainty rating
VAT	Very low	Medium	High	Medium
Tobacco duties — cigarette duty	Low	High	Very high	High
Tobacco duties — hand-rolling tobacco duty	Low	High	Very high	High
Alcohol duties — beer duty	Medium	Very high	Very high	Very high
Alcohol duties — spirits duties	Low	Very high	Very high	Very high
Hydrocarbon oils duty	High	Medium	Medium	Medium
Other excise duties	Very high	High	High	Very high
Self Assessment — non-business taxpayers	Low	Medium	Medium	Medium
Self Assessment — wealthy	High	High	Very high	High
Self Assessment — large partnerships	Very high	Very high	High	Very high
Self Assessment — business taxpayers	Low	Medium	Medium	Medium
PAYE — small business	High	Medium	Low	Medium
PAYE — mid-sized business	High	Medium	Medium	Medium
PAYE — large business	High	Very high	Very high	Very high
Income Tax, NICs, CGT hidden economy —moonlighters	High	High	High	High
Income Tax, NICs, CGT hidden economy — ghosts	Very high	High	Very high	Very high
Income Tax, NICs, CGT — avoidance	High	High	High	High
Corporation Tax — small businesses	Medium	Medium	Very high	High
Corporation Tax — mid-sized businesses	Low	Medium	Medium	Medium
Corporation Tax — large businesses	Low	Medium	Medium	Medium
Inheritance Tax	High	High	High	High
Landfill Tax	High	Very high	High	High
Other taxes, levies and duties	Very high	High	Very high	Very high
Stamp Duty Reserve Tax	High	Very high	Very high	Very high
Stamp Duty Land Tax	Medium	High	High	High

Notes for Table A1.7 

1. ‘Other excise duties’ includes betting and gaming duties, cider and perry duties, spirit-based ready-to-drink duties and wine duties. 

2. Ghosts are individuals whose entire income is unknown to HMRC. 

3. Moonlighters are individuals who are known to HMRC in relation to part of their income but have other sources of income that HMRC does not know about. 

4. ‘Other taxes, levies and duties’ includes Aggregates Levy, Air Passenger Duty, Customs Duty, Climate Change Levy, Digital Services Tax, Insurance Premium Tax, Plastic Packaging Tax and Soft Drinks Industry Levy. 

Figure A1.3 shows the share of the tax gap by uncertainty ratings since ‘Measuring tax gaps 2021 edition’. 

Figure A1.3: Share of tax gap by overall uncertainty rating compared to previous editions 

Measuring tax gaps edition	Very low	Low	Medium	High	Very high
MTG21	0%	64%	19%	6%	12%
MTG22	0%	31%	40%	12%	18%
MTG23	0%	21%	51%	13%	15%
MTG24	0%	20%	56%	9%	15%
MTG25	0%	0%	79%	8%	13%
MTG26	0%	0%	51%	35%	14%

Notes for Figure A1.3

1. MTG stands for ‘Measuring tax gaps’.
2. Figures may not appear to sum due to rounding.
3. ‘%’ refers to percentage of the total tax gap.

Projections in the tax gap estimates 

For the most recent tax years, complete data is not always available. We therefore use projections to produce estimates for these years so that a consistent time series can be presented across all taxes. 

In this edition, projected estimates are shown using lighter shading for bars and a dashed line to distinguish them from estimates based on more complete data. This has been added to Figure 1.1 in 1. Tax gaps: Summary to highlight that a large proportion of the most recent years are based on projected estimates. 

Projections use the latest available data and assumptions about how key drivers may change to estimate the tax gap for later years. The specific approach varies by tax and is described in the relevant chapters and tables. 

Projected estimates are less certain than those based on complete data and are more likely to be revised in future publications as outturn information becomes available. 

Tax gap by customer group  

Tax gap estimates by customer group are aligned with HMRC’s customer group classifications used for measurement purposes. Table A1.8 summarises the customer groups used in this publication. 

Table A1.8: Customer group definitions 

Customer group	Definition
Individuals	Incomes below £200,000 and assets below £2m in each of the last 3 years
Wealthy	Incomes of £200,000 or more, or assets equal to or above £2 million in any of the last 3 years
Small businesses	Turnover below £10 million and typically fewer than 20 employees
Mid-sized businesses	Turnover between £10 million and £200 million, or typically 20 or more employees
Large businesses	Usually turnover exceeding £200 million but other factors such as complexity, level of risk and global mobility are considered

A separate tax gap estimate for criminals is also published and is not included within these customer group definitions.