Official Statistics

Quality report: statistics on non-domiciled taxpayers in the UK

Updated 29 July 2021

1. Contact

• Organisation unit - Knowledge, Analysis and Intelligence (KAI)
• Name – A Hatton
• Function – Specialist Personal Tax Statistics
• Email - personaltax.statistics@hmrc.gov.uk

2. Statistical presentation

2.1 Data description

This methodology and quality report relate to the Official Statistic publication, Non-domiciled taxpayers in the UK, and the purpose is to provide users with information about the quality of the outputs as set out by the Code of Practice for Official Statistics.

This publication is released annually and sets out the number of people who are claiming non-domiciled taxpayer status in the United Kingdom (UK) and estimates their liability for UK Income Tax, Capital Gains Tax and National Insurance contributions. These statistics are based on data submitted by taxpayers on their Self Assessment Tax return.

2.2 Classification system

Breakdowns of the number of non-domiciled taxpayers, tax liabilities and reliefs presented in the statistics reflect relevant tax-related concepts as defined in section 2.4.

2.3 Sector coverage

This publication covers all taxpayers that HM Revenue and Customs (HMRC) regards as non-domiciled for tax purposes, irrespective of economic sector. In April 2017 a new status of ‘deemed domicile’ was introduced. This means that some UK resident taxpayers that have a common law domicile outside the UK are not included in these statistics.

2.4 Statistical concepts and definitions

Taxpayer

For the purposes of these statistics, a taxpayer is any individual who is liable to pay tax in the UK. A taxpayer is either UK resident or non-UK resident according to the terms of the statutory residence test (SRT).

Tax year

The statistics are aggregated into tax years. A tax year stretches from 6th April until 5th April of the following calendar year.

Residence status

An individual’s residence status in the UK is usually dependent on how many days are spent in the UK within a particular tax year. From tax year ending 2014, the SRT sets out rules to determine tax residence (i.e. it sets out what makes an individual a UK resident for tax purposes). Before tax year ending 2014 the rules were different.

Non-domiciled status

Non-domiciled status is declared by individuals on their Self Assessment tax return form. A non-domiciled taxpayer has their permanent home (‘domicile’) outside the UK for tax purposes. An individual’s domicile is usually the country that individual’s father considered his permanent home when that individual was born. It may have changed if an individual moves abroad and doesn’t intend to return. For some individuals, domicile has no bearing on their tax position, so they might not declare non-domiciled status. Such taxpayers are not included in these statistics.

Arising basis

Individuals who are UK resident are normally taxed on the arising basis of taxation. The arising basis of taxation means that all an individual’s worldwide income and gains are taxable in the UK.

Remittance basis

The remittance basis is an alternative tax treatment that is available to individuals who are resident but not domiciled in the UK and have foreign income and gains. If they have been UK resident for long enough, they may be liable to pay a charge to access the remittance basis. If taxed on the remittance basis, individuals are liable to UK tax in the normal way on their UK source income and gains. However, they are only liable to UK tax on any remittances of foreign income and gains that they bring (or ‘remit’) to the UK. All other non-domiciled individuals are taxed on the arising basis (i.e. their worldwide earnings are subject to UK tax) if they do not make a claim for the remittance basis.

Remittance basis charge

The remittance basis charge (RBC) is due once a non-domiciled individual claiming the remittance basis has been UK resident for at least 7 out of the previous 9 tax years. The charge is £30,000 per annum if an individual is claiming the remittance basis and has been UK resident for at least 7 out of the previous 9 tax years.

In April 2012, the charge was increased to £50,000 per annum for those remittance basis users who had been UK resident for at least 12 out of the previous 14 tax years. From April 2015, this £50,000 annual charge was increased to £60,000 and a new charge of £90,000 was introduced for those individuals who have been UK resident for at least 17 out of the last 20 tax years. Therefore, an individual can pay a different charge level depending on their length of UK residency.

Deemed domicile status

Following a policy change in April 2017, non-domiciled individuals are treated as UK residents (deemed domicile) for tax purposes if they meet one of 2 conditions, A or B. These individuals are unable to claim the remittance basis of taxation and will be assessed on their worldwide income and gains on the arising basis.

Condition A – Being born in the UK, with the UK as their domicile of origin, and being a UK resident for tax year ending 2018, or later years Condition B – Being a UK resident for at least 15 of the 20 tax years immediately before the relevant tax year.

Income tax

Income Tax is a tax on an individual’s income over the course of a tax year. Income Tax is the UK Government’s largest single source of tax revenue.

The amount of Income Tax an individual should pay, their tax liability, is determined by a number of factors including: their level of income, the type of income and the level of allowances to which they are entitled, and where in the UK they are located (as some aspects of Income Tax have been devolved).

Capital gains tax

Capital Gains Tax (CGT) is charged on gains realised on the disposal of assets. Types of disposal include sales, gifts, exchanges and gaining compensation for assets. Typical assets include most personal possessions above a certain value, property, shares and business assets. The capital gain is broadly the difference between the disposal proceeds and the cost of acquiring the asset. There are various reliefs and exemptions which may affect the amount of CGT to be paid.

National insurance contributions

Contributions paid by employees and self-employed individuals to qualify for certain benefits and the State Pension.

Business investment relief

Business Investment Relief (BIR) was first introduced in April 2012 and allows, current (or previous where the remittance occurred when claiming the remittance basis) remittance basis claimants to claim relief from UK tax charges on remitted foreign income providing it is invested in a qualifying UK company.

2.5 Statistical unit

The unit in the statistics is individuals.

2.6 Statistical population

All individuals claiming non-domiciled taxpayer status in the UK on their Self Assessment tax return forms.

2.7 Reference area

The geographic region covered by the data is the United Kingdom (UK).

2.8 Time coverage

The statistics cover the time period from tax year ending 2008 until the latest financial year for which Self Assessment returns are available. Summary statistics relating to non-domiciled taxpayer numbers and tax contributions can be provided for the latest year, however, further breakdowns cannot be provided such as regional statistics, BIR claims and taxation on the remittance and arising basis.

3. Statistical processing

3.1 Source data

All data used for this publication comes from information provided by taxpayers in their Self Assessment Tax returns. Self Assessment (SA), introduced in tax year ending 1997, is a business system operated by HMRC where individuals, trusts and unincorporated businesses (for example partnerships) return information on Income and Capital Gains.

The SA system calculates tax liabilities using tax return data, accounts for payments and repayments, and administers penalties and interest for late payment and late filing of returns. All non-domiciled taxpayers must complete an SA return if they meet the criteria.

HMRC’s primary purpose for collecting data through these administrative sources is to ensure that taxpayers are paying the correct amount of tax or receiving the correct reliefs and repayments. HMRC’s policy on the use of administrative data for producing statistics is set out in the HMRC Statement of Administrative Sources.

3.2 Frequency of data collection

The data from all SA forms submitted is extracted from HMRC’s Computerised Environment for Self Assessment (CESA) into HMRC’s Corporate Data Warehouse (CDW) on a daily basis. Data is sent from CDW to KAI at the end of each month if it is new or has been updated since the previous monthly extract. The data is updated on a monthly basis and the monthly cycle is scheduled for the Tuesday after the first Friday of the month.

These statistics are compiled on an annual basis from an extract of the live Self-Assessment system, this extract is typically taken in mid-June.

3.3 Data collection

Tax returns should be filled in after the end of the tax year to which they apply. Depending on how the tax returns are submitted, paper or online, there are different deadlines. The latest deadline for submission of an online return is midnight on 31st January following the end of the tax year, and midnight on 31st October for submission of a paper return, and 30th December for those who wish HMRC to calculate the liability and automatically collect it from their income.

The SA return comprises the main tax form plus supplementary pages (schedules) for different types of income. There are different main returns for Individual taxpayers, for Trusts and for Partnerships. SA customers in any of these groups need to submit a main return/tax form and supplementary pages/schedules as required.

In particular those who are non-domiciled will complete the SA109 form entitled “Residence, remittance basis etc”. The majority of information in this publication comes from data that has been provided by taxpayers on this form. We use additional information from the SA return to determine estimated liabilities for income tax, capital gains tax and National Insurance Contributions.

3.4 Data validation

For SA data, quality assurance processes are included in the operational systems that collect the data. Further quality checks are added by analysts using the data for analytical purposes. These checks include manual and automated checking processes such as checking components of sums agree with totals. In addition, further checking is applied during the process of calculating the statistics published and reasonableness checks are carried out when new statistics are produced (e.g. are they in line with previous figures, are they what one would expect given what has happened since, are there plausible explanations for changes?).

3.5 Data compilation

We aggregate data using taxpayer statuses derived from information provided on the SA109 form as described in statistical concepts and definitions (section 2.4). We use statistical software to extract, sort, filter and aggregate the data in order to produce the tables in the publication.

The SA deadline is the January following the end of the tax year. Statistical analysis suggests that around 5% of returns are received too late to be included for publication the following July. By the second July after the SA deadline, analysis suggests that the statistics are typically more than 99% complete. Therefore, we only adjust for late returns in the most recent tax year. We do not adjust data for previous years given it is likely to be around least 99% complete. Further details on the late filing adjustment can be found in section 6.3.3.

The latest years in all tables are subject to revisions in future publications, so any arising trends should be treated with caution.

4. Quality management

4.1 Quality assurance

All official statistics produced by KAI, must meet the standards in the Code of Practice for Statistics produced by the UK Statistics Authority.

Analytical Quality Assurance describes the arrangements and procedures put in place to ensure analytical outputs are error free and fit-for-purpose.

Every piece of analysis is unique, and as a result there is no single quality assurance (QA) checklist that contains all the QA tasks needed for every project. Nonetheless, analysts in KAI use a checklist that summarises the key QA tasks, and is used as a starting point for teams when they are considering what QA actions to undertake. Teams amend and adapt it as they see fit, to take account of the level of risk associated with their analysis, and the different QA tasks that are relevant to the work. The quality assurance review checklist used for the production of these statistics was last reviewed in July 2021.

The QA for this project adhered to the framework described in ‘Quality assurance’ and the specific procedures undertaken were as follows:

Stage 1 – Specifying the question

Up to date documentation was agreed with stakeholders setting out outputs needed and by when; how the outputs will be used; and all the parameters required for the analysis.

Stage 2 – Developing the methodology

Methodology was agreed and developed in collaboration with stakeholders and others with relevant expertise, ensuring it was fit for purpose and would deliver the required outputs.

Stage 3 – Building and populating a model/piece of code

• Analysis was produced using the most appropriate software and in line with good practice guidance.
• Data inputs were checked to ensure they were fit-for-purpose by reviewing available documentation and, where possible, through direct contact with data suppliers.
• QA of the input data was carried out.
• The analysis was audited by someone other than the lead analyst – checking code and methodology.

Stage 4 – Running and testing the model/code

• Results were compared with those produced in previous years and differences understood and determined to be genuine.
• Results were compared with comparable independent estimates, and differences understood.
• Results were determined to be explainable and in line with expectations.

Stage 5 – Drafting the final output

• Checks were completed to ensure internal consistency (e.g. totals equal the sum of the components).
• The final outputs were independently proof-read and checked.

At the start of a project, during the planning stage, analysts and managers make a risk-based decision on what level of QA is required. Analysts and managers construct a plan for all the QA tasks that will need to be completed, along with documentation on how each of those tasks are to be carried out, and turn this list into a QA checklist specific to the project. Analysts carry out the QA tasks, update the checklist, and pass onto the Senior Responsible Officer for review and eventual sign off. These measures enhance the level of assurance and provide a clear audit trail.

4.2 Quality assessment

This background quality report assesses the quality of the statistics produced. In addition, we have conducted internal reviews on specific areas affecting these statistics such as the late filing adjustment and BIR claims, which are reflected in this report.

We reviewed the BIR statistics because the relief is available to UK residents who were classed as non-domiciled individuals in the past. Our review showed this also included individuals who are now deemed domiciled following the policy change in 2017. As these individuals are no longer classed as non-domiciled, their BIR claims were excluded from the statistics from 2017 onwards, causing a substantial decrease in the previously reported value and number of BIR claims. Following an internal review, we have included in the statistics the BIR claims of deemed domiciled individuals.

5. Relevance

5.1 User needs

We know that non-domiciled tax policy is of interest to policy makers in government, academics, researchers and journalists that have an interest in taxation. These statistics might also be of interest to individuals or organizations working in fields related closely to non-domiciled taxpayers, for example accountants and other financial organisations.

Officials in HMRC and HMT use HMRC statistics to assess the effects of policies. Accountants and tax lawyers also use HMRC statistics to assess trends in the internationally mobile. The media have used them as background for commenting on policy changes initiated by the Government.

5.2 User satisfaction

HMRC are committed to providing impartial quality statistics that meet our customers’ needs. Contact details for the team are available in the bulletin in case users wish to provide feedback, comments or queries on our statistics.

Customers can also provide feedback directly to the contact detailed in the bulletin, via the contacts section on the HMRC statistics page or the HMRC blog.

5.3 Completeness

It is a legal requirement that all qualifying taxpayers submit an SA return, by the required deadlines. Penalties exist for non-compliance. The statistics contained in this report can therefore be considered as complete.

6. Accuracy and reliability

6.1 Overall accuracy

This analysis is based on administrative data, and accuracy is addressed by eliminating non-sampling errors as much as possible through adherence to the quality assurance framework.

The potential sources of error include:

• Taxpayers entering incorrect information onto their SA tax return form.
• Human or software error when entering the SA data into the CESA system.
• Individuals not completing their SA return form by the required date.
• Mistakes in the programming code used to analyse the data and produce the statistics.

6.2 Sampling error

Samples are not used to compile the analysis, which is instead based on administrative data from HMRC’s CESA system. Sampling error is therefore not relevant.

6.3 Non-sampling error

Coverage error

Non-domiciled individuals must complete an SA tax return if they report foreign income or gains of £2,000 or more in the relevant tax year, or if they bring any money into the UK. As the majority of non-domiciled individuals are likely to meet at least one of these criteria for completing SA returns, coverage error is likely to be small.

Measurement error

The main sources of measurement error can be categorised as respondent errors and includes individuals making errors when entering their information onto their SA tax return form, whether this is done on paper or electronically.

There is a risk that errors involving large incomes or tax amounts may distort the overall statistics. To mitigate against this, we conduct checks on the live SA data each month to ensure there are records for all cases, count the schedules, quality assuring a selection of records against the tax calculation and checking the records before loading them to the HMRC database.

Nonresponse error

When analysing tax liabilities for the latest available year, figures are not necessarily available for all individuals, as some may not have completed their SA return by the required date.

For individuals whose data is not available for the latest year, tax liabilities are derived by making a late filing adjustment based on trends observed in previous years. The adjustment is calculated by using data for previous years to compare the number of taxpayers and values of IT and CGT declared on SA returns before the deadline, with totals submitted before and after the deadline. These proportions are calculated for each tax year for the 5-year period between tax year ending 2015 to tax year ending 2019 and a simple average is taken to produce the adjustments for the latest year of tax year ending 2020. The adjustments for the number of taxpayers, IT and CGT are calculated separately for non-residents and residents. We also apply the late filing adjustments calculated for IT to NICs.

After conducting checks to identify anomalistic cases of late returns, we identified one late return data item that was much larger than any other single item in the last eight years. This single data point was dominating our late filing adjustment dataset. So we replaced this outlier with the next largest value from within the same subgroup.

Processing error

It is possible that errors exist in the programming code used to analyse the data and produce the statistics. This risk is reduced through reviewing and testing the programs that are used. We used a substantially different production process in 2021 including new programming code. We compared the outputs of the new and old processes to ensure consistency of results.

6.4 Data revision

The latest years in all tables will be subject to revisions in future publications, so any arising trends should be treated with caution.

Data revision - policy

The United Kingdom Statistics Authority (UKSA) Code of Practice for Official Statistics requires all producers of Official Statistics to publish transparent guidance on the policy for revisions.

Data revision - practice

We publish estimates of liabilities for the latest financial year, along with revisions for previous years. The revisions for previous years are normally small but reflect the latest position for taxpayers and any new data received since the previous publication.

We revise statistics in this analysis for the three to four years prior to the latest published year. We estimate that around 95% of returns for a tax year are received in time for publication the following July. By the second July after the SA deadline, our analysis suggests that the statistics are typically around 99% complete.

6.5 Seasonal adjustment

We do not use seasonal adjustment.

7. Timeliness and punctuality

7.1 Timeliness

We publish these statistics on an annual basis, usually in July. This timeline allows sufficient time for later filing of SA returns, processing the data, analysis and to complete quality assurance checks of these statistics when the data becomes available at the end of the tax year. For example, statistics for the 2018-19 tax year were published in July 2020, six months after the January 2020 SA filing deadline for the tax year.

7.2 Punctuality

In accordance with the Code of Practice for official statistics, the exact date of publication will be given not less than one calendar month before publication on both the Publication plan for HMRC national statistics and official statistics and the Research and statistics calendar of GOV.UK. Any delays to the publication date will be announced on the HMRC statistics announcements website. This release was first published in 2017. The 2019 publication was delayed by approximately one month, which was announced at the time.

8. Coherence and comparability

8.1 Geographical comparability

We produce this analysis for the UK.

8.2 Comparability over time

As explained under ‘Deemed domicile status’ in section 2.4 above, there was a policy change in April 2017 which led to many individuals who were previously classed as non-domiciled being re-classified as having deemed domicile status. Therefore, the number of non-domicile individuals are substantially lower from 2017 onwards when compared to previous years. This indicates that statistics relating to the number of non-domiciled individuals are not strictly comparable pre and post-April 2017.

For Capital Gains tax contributions, amounts of gains in tax year ending 2009 (when taper relief was abolished and the 18% CGT rate and Entrepreneur’s Relief introduced), tax year ending 2010 and tax year ending 2011 (when a higher rate of 28% was introduced after 23 June 2010) are not comparable with earlier years. Amounts of gains since tax year ending 2012 are comparable only with each other as they are the only full years when a higher rate of 28% applied to high enough gains.

8.3 Coherence - cross domain

The only data source used for this publication are SA returns. Therefore, there are no coherence issues arising due to differences in data sources. The data is always presented on a tax year basis, which eliminates coherence issues due to differences between sub-annual and annual statistics.

8.4 Coherence - internal

Rounding of numbers may cause some minor internal coherence issues as the figures within a table may not sum to the total displayed.

The NICs figures are annual estimates of NICs derived from SA annual return data. The NIC estimates shown in table 1 of the publication are the sum of Class 1 primary, Class 1 secondary, Class 2 and Class 4 contributions all modelled annually. This may differ from actual NICs due to differences between modelling methods of NICs and the way some classes of National Insurance contributions are charged. For more information on National Insurance see HMRC guidance.

9. Accessibility and clarity

These statistics can be found in the Statistics on non-domiciled taxpayers in the UK section on GOV.UK. The publication consists of figures and commentary presented as pages on GOV.UK and Open Document Spreadsheet format. There is no other access to the statistics and data, e.g. via micro-data. This publication is produced in line with accessibility requirements.

Any enquiries in accessing HMRC data should be directed to HMRC data lab - read more information on availability of HMRC data and how HMRC data can be accessed.

Email: HMRC Datalab

9.1 Useful links

Guidance Note: Residence, Domicile and the Remittance Basis

Deemed domicile rules

Guidance Note: Changes to the Remittance Basis

Government research and statistics publications.

Older versions of HMRC statistical publications are also available on The National Archives.

10. Cost and burden

We compile these statistics from SA returns which are primarily designed to ensure that taxpayers pay the correct amount of tax. Analysis of this data is a necessary end of its own to understand this segment of the tax base and develop more efficient taxation policy. Therefore there is minimal additional cost to producing these statistics.

Publishing this data also achieves a public good to understand the contribution that this segment of taxpayers make to public funds.

We continually look for ways to improve and streamline the production of our statistics. We have moved towards the use of ‘reproducible analytical pipelines’ (RAPs), which are automated statistical and analytical processes. They incorporate elements of software engineering best practice to ensure that the pipelines are reproducible, auditable, efficient, and high quality.

11. Confidentiality

11.1 Confidentiality - policy

HMRC has a legal duty to maintain the confidentiality of taxpayer information. Section 18(1) of the Commissioners for Revenue and Customs Act 2005 (CRCA) sets out our duty of confidentiality.

This analysis complies with this requirement.

11.2 Confidentiality - data treatment

The statistics in this release are presented using aggregated data.

Due to possible disclosure of individual taxpayers, we have excluded from this publication a geographical breakdown of those non-domiciled taxpayers on the remittance basis who are liable to pay the remittance basis charge.

The increased accessibility of data due to the internet and modern information technology translates into higher risk of identifying individuals from published statistics. Therefore, HMRC has polices in place to ensure that there is sufficient protection of the privacy of individuals contributing to official statistics yet still ensuring the statistics are adequate, robust and accurate. Statistical disclosure control (SDC) concerns safeguarding the confidentiality of the information that HMRC holds about people and businesses. It ensures that units and their attributes in a data release are not identifiable. These techniques can be applied to a variety of outputs.

SDC involves modifying the data so that it becomes sufficiently difficult to identify individuals. At a basic level, this involves removing information by not including it in a table, increasing the scope of the variables (e.g. measuring the range of repayments as being between (1 and 100 not between 1 and 10). Disclosure control methods attempt to find an optimal balance between the improvement in confidentiality protection and the reduction in data quality. We have conducted disclosure checks on the regional breakdown statistics.

As the data presented is at an aggregate level that in itself provides a level of disclosure control, as it would make it impossible to identify any data belonging to an individual. Statistics on the number of taxpayers are rounded to the nearest 100. Remittance basis tax and NICs liability statistics for regional breakdowns have been rounded to the nearest £100m, or displayed as a [disclosive] where the value is between £0.01 and £49.99m.

Dominance occurs if any single contributor to a cell can be estimated within a certain percentage of its original value.

Our confidentiality and access policy and pre-release policy are both available on GOV.UK, along with a list of those who have pre-release access to this publication.