GP Earnings and Expenses Estimates, 2020/21

GP Earnings and Expenses Estimates provides a detailed study of the NHS and private earnings and expenses of GPs working in the NHS under a General Medical Services (GMS) or Personal Medical Services (PMS) contract in the UK in 2020/21.

Earnings and expenses estimates are based on a sample as at March 2022 from HM Revenue and Customs’ (HMRCs') Self Assessment tax database. All analyses are carried out by HMRC statisticians on an anonymised data set, which is adjusted to take into consideration individuals’ pension contributions.  Only aggregate non-person-identifiable information is supplied to NHS Digital for publication, in line with data protection legislation.

All results presented are in cash terms. For real terms income before tax estimates, see the Excel Time Series (available to download under 'Resources' on the Overview page), which supplements this report. The time series has been produced to collate results from the GP Earnings and Expenses Estimates reports from 2002/03 onwards.

This report has been agreed by the Technical Steering Committee (TSC), which is chaired by NHS Digital and has representation from the Department of Health and Social Care, UK Health Departments, NHS England, and the British Medical Association (BMA) which represents the interests of GPs.

NHS Digital welcomes feedback on the methodology, commentary, tables and charts within this publication. Please contact NHS Digital with your comments and suggestions by telephone on 0300 303 5678 or by email to [email protected] stating ‘GP Earnings and Expenses Estimates’ in the subject heading.

Only GPs working for the NHS in the UK are included in the report. The figures may include both NHS and private earnings.

GP type

The report contains results for contractor and salaried GPs only. “Combined GP” results are those of contractor and salaried GPs together, and consist of a weighted average of earnings and expenses data from self-employment for contractor GPs along with a combination of employment and medical self-employment figures for salaried GPs.

GPs that also work as Primary Care Organisation (PCO) Medical Directors are included. Fixed share partners bound to a practice via the partnership agreement are included in the contractor group.

PCO-employed GPs that are salaried are treated as salaried GPs employed by the practice.

It is possible that some salaried GPs may not need to complete a Self Assessment tax return. Such individuals do not appear in HMRC’s tax Self Assessment database and therefore cannot be included in the sample.

GPs who work solely as locums or freelancers are not included.

Contract type

Only GPs working under a GMS or PMS contract are included. GPMS results are those of GMS and PMS GPs together. No PMS contracts exist in Wales or Northern Ireland, so results shown for these countries relate to GMS GPs only. PMS GPs are referred to as Section 17c in Scotland. Primary Care Organisation (PCO) employed GPs that are salaried are given a GMS/PMS flag depending upon whether the contractor GP(s) for the practice hold a GMS or PMS contract.

GPs with multiple GMS or PMS contracts are removed, as are those GPs who are solely Alternative Providers of Medical Services (APMS).

Accounting year

Only GPs whose information HMRC could trace and who had an accounting year ending in the final quarter of the fiscal year (i.e. between 1 January and 5 April 2021) are included.

Important definitions

The report uses the following terms:

Gross earnings – self-employment and/or employment income from NHS and private work before the deduction of total expenses.

Total expenses – business expenses from NHS and private work allowable for tax purposes.

Income before tax – taxable income before pension contributions are deducted, made up of gross earnings less total expenses, also known as net income.

Expenses to earnings ratio (EER) – the proportion of gross earnings taken up by total expenses.

Earnings

Gross earnings for contractor GPs are based on all medical income from self-employment sources. Income from employment sources is not included in averages for contractor GPs.

The majority of salaried GPs have a combination of self-employment and employment income. Therefore, their average total income before tax is based on medical income from self-employment sources and all employment income. The total expenses figure for salaried GPs will also be based on an average total of self-employment and employment data.

Expenses

Expenses are split into the following categories:

• Office and General Business
• Premises
• Employee
• Car and Travel
• Interest
• Other, which includes:
  • ‘Advertising and business entertainment costs’ (which were classified as business expenses up to 2006/07)
  • Interest for businesses where turnover is <£85,000 and interest is not reported separately
  • Expenses for businesses (where turnover is low) and detailed expense breakdown not available
  • Cost of drugs for dispensing GPs
• Net Capital Allowance.

Expenses figures exclude disallowable expenses.

Populations and samples used in the report

Report population - the identifiable sample of GPs from the NHS workforce census and workforce Minimum Data Set (wMDS) who qualify for inclusion in the report, used by HMRC to match against self-assessment tax returns.

The report population is not the same as the totals of populations published in the GP censuses in Wales, Scotland and Northern Ireland and the General Practice Workforce publication in England. The totals in these publications are based on a count of unique General Medical Council (GMC) numbers, whereas for this report, certain types of GPs are excluded (see above). Additionally, incomplete coverage within the data source also has the effect of lowering the report population when compared to the published totals of GPs in the four countries (see ‘Change in data source’). Some other GPs were subsequently excluded from HMRC’s sample for data quality reasons. For more information, please see the Methodology section.

HMRC sample - the number of GP records that HMRC traced and was able to match to a Self Assessment tax return. These records must pass validation checks and have an end-date in quarter four. In addition, contractor GPs must have self-employment income in order to be included in the sample. Salaried GPs must not have solely self-employment income but must also have employment income for all quarters of the financial year to be included. 

The results presented in this report are estimates based on samples. To give a more accurate result, HMRC weights its sample to the report population when calculating averages.

Figures are subject to sampling error. Differences between groups and subgroups of GPs and differences in results when compared to previous years may not be statistically significant. Standard errors to support the statistical significance of the income before tax results are available in the Excel Time Series (available to download under 'Resources' on the Overview page).

In addition, small populations for some subgroups mean that extreme values can have noticeable effects upon the averages. In such cases, results may be subject to additional uncertainty. Footnotes identify results based on a sample size of fewer than 125 GPs, which should be treated with caution.

Tables A1 to 4 and B1 and B2 show comparative figures for population counts.  The Known Issues section below provides more information on the report population and sample counts in this report.

Table A1: Workforce Minimum Data Set population, report population and HMRC sample, GPMS, 2019/20 and 2020/21, England

^{Notes on Table A1:}

^{1. In England, prior to 2015 figures were sourced from NHAIS GP Payments (Exeter) System. From 2015 England GP information has been sourced from the workforce Minimum Data Set (wMDS) which collects information on the workforce directly from general practices. This workforce data has been collected on a quarterly basis since September 2016 and its completeness and coverage has improved since the collection began meaning the report population and consequently, the size of the matched HMRC sample, has also increased each year. See the Change in Data Source section below for more information.}

^{2. Information from the workforce Minimum Data Set collection is published in the General Practice Workforce publication. The 2019/20 and 2020/21 Population column totals (as at 30 Sep) do not sum the total of contractor and salaried GPs rows because there are a small number of GPs (around 120 in 2020/21) who work as both a contractor and salaried GP.}

Table A2: Census population, report population and HMRC sample, GPMS, 2019/20 and 2020/21, Scotland

^{Notes on Table A2:}  

^{1. Census population figures are the totals of populations published in the GP census in Scotland at 30 Sep. The total row includes only contractor and salaried GPs, and not GPs outside the scope of this publication, such as locums.}

Table A3: Census population¹, report population and HMRC sample, GMS, 2019/20 and 2020/21, Wales

^{Notes on Table A3:}

^{1. Census population figures are the totals of populations published in the GP census in Wales at 31 March. The total row includes only contractor and salaried GPs, and not GPs outside the scope of this publication, such as locums.}

Table A4: Census population, report population and HMRC sample, GMS, 2019/20 and 2020/21, Northern Ireland

^{Notes on Table A4:}

^{1. Census population figures are the totals of populations published in the GP census in Northern Ireland at 31 March. The total row includes only contractor and salaried GPs, and not GPs outside the scope of this publication, such as locums.}

^{2. Census populations taken at 31 March 2020 and 31 March 2021}

^{3. c - figures suppressed due to small sample sizes.}

Table B1: Report population by contract type, 2020/21, England

Table B2: Report population by contract type, 2020/21, Scotland

^{No PMS contracts exist in Wales or Northern Ireland, so all results shown relate to GMS GPs only. See Tables A3 and A4 for Wales and Northern Ireland GMS GP report population counts.}

Averages, rounding and suppression

All averages are means unless otherwise stated. In addition to providing the mean taxable income from NHS and private work, the median is also provided in places. Standard errors and medians for all taxable income results are provided in the open data text file (csv) published alongside this report.

Earnings and expenses estimates have been rounded to the nearest £100. This means that in some tables and figures the rounded numbers presented for taxable income may not equal gross earnings minus total expenses, and the sum of numbers in a table may not equal the total. As the population estimates provided in this report are not intended to be definitive, these are rounded to the nearest 50 unless otherwise stated. In expenses tables, the percentage of total expenses for each category is displayed in parentheses.

Calculations such as percentage change and estimates of the expenses to earnings ratio are performed on data prior to rounding unless otherwise stated.

Symbols used in tables

• c = confidential
  Results have been suppressed by HMRC due to low sample numbers
• z = not applicable
  Used when an observation is not applicable
• : = not available
  Data unavailable for reasons other than those already specified

Superannuation contributions

The results for contractor GPs exclude an estimate of employers’ superannuation contributions but are before the deduction of employees’ contributions.

The results for salaried GPs include an estimate of employees’ superannuation contributions and Additional Voluntary Contributions (AVCs). This puts the salaried results on a comparable basis with the contractor results and allows results for combined GPs to be produced for the report.

For more information, please see the Methodology section.

COVID-19

The first cases of COVID-19 in the UK were confirmed late January 2020 and the first UK-wide lockdown was announced in March 2020. Differing but similar arrangements were put in place across England, Scotland, Wales and Northern Ireland to support general practice during this period with the focus being managing the pandemic. Practices also played a role to varying degrees across England, Scotland, Wales and Northern Ireland in their respective vaccination programmes which may have impacted practice income. The pandemic is also likely to have impacted on expenses incurred during this period. Details can be found in the results chapters for each country.

Self-employment Income Support Scheme (SEISS)

2020/21 saw the introduction of Self-Employment Income Support Scheme (SEISS) payments that were made by the government to eligible businesses that had been adversely affected by the coronavirus pandemic. These are subject to income tax and self-employed National Insurance contributions. As this is classed as taxable income these payments have been included in the GP earnings data for 2020/21. A table showing the number of SEISS claimants and the average amount claimed for each country can be found in table C1.

Table C1: Number of SEISS claimants and average amount of SEISS claimed by country, 2020/21

_Notes:  

_{1. These figures are unweighted and as such do not follow the methodology referred to in the stratification section of the Methodology page.}

_{2. Figures are based on valid sample in both NHS digital and HMRC database.}

_{3. Counts are rounded to the nearest 50, amounts are rounded to the nearest 100.}

_{4. c - figures suppressed due to small sample size.}

Change in data source

In England, the primary source for GP data changed in September 2015 from the National Health Authority Information System (NHAIS) to the workforce Minimum Data Set (wMDS), now collected via the National Workforce Reporting Service (NWRS), but before 1 April 2019 via the Primary Care Web Tool (PCWT)  Workforce Census module and the workforce Minimum Data Set Collection Vehicle (wMDSCV).

Data completeness for the wMDS collection has improved each year since September 2015 meaning that the report population - the identifiable sample of GPs which HMRC used to match against Self Assessment tax returns - has also increased each year.

Although data for England are now extracted on a monthly basis, this report uses data that were extracted shortly after 30 September, which is in line with the previous NHAIS extracts and is consistent with the timing of the annual GP census data still used for Scotland, Northern Ireland and Wales.

NHS and private earnings

Earnings and expenses results relate to both NHS and private work. It is not possible to provide an NHS/private split using HMRC as a data source as most GPs submit a Self Assessment tax return which contains information on all of their self-employment earnings while practising as a GP, but which cannot differentiate between NHS and private earnings. GPs can perform both NHS and private work inside or outside the practice, including the NHS Out of Hours service.

Please note there is also the possibility that income from work outside of general practice, but elsewhere in the NHS, such as A&E clinical streaming or working as Primary Care Network clinical directors, is also included that cannot be reported separately.

Working hours (England)

For three years we published experimental analysis of earnings and expenses estimates by working hours bandings for GPs in England, covering 2016/17 to 2018/19. However, as it is not possible for HMRC to distinguish between general practice earnings, income from other NHS work and private earnings, all earnings were included in the calculations whilst the working hours used reflected only those worked in general practice. This in turn could have inflated the income figures when considering earnings by working hours bands.  

Because of the growing numbers of GPs working in additional roles as well as in general practice, we believe that analysis by working hours bands would become increasingly unreliable and it has therefore been removed from the report.

Small samples

As noted in the Populations and samples used in the report section, the results from the sample count are weighted up to the report population and are subject to sampling error. Differences between groups and subgroups of GPs or comparisons between years may not be statistically significant. In addition, small samples for some subgroups mean that extreme values can have noticeable effects on the averages. In such cases results may be subject to additional uncertainty. Footnotes identify results based on a sample size of fewer than 125, which should be treated with caution.

Multiple contracts in England

Some GPs in England hold contracts with several practices, leading to more complex business arrangements. GPs with multiple contracts are excluded from the data set. In addition, GPs who are solely APMS are not included in the report. However, if a GP holds both a GMS/PMS contract and an APMS contract, their earnings and expenses from that contract may be included and reported within their GMS/PMS figures.

Dispensing-only income

Income received solely from dispensing cannot be separately identified as this is not recorded separately on the Self Assessment tax return.

Clawback

Earnings and expenses reflect any Primary Care Organisation (PCO) clawback or reimbursement for previous years. This may occur if the GP or PCO had over or under-estimated predicted earnings for the year.