Sources of household income
Last updated 8 August 2023 - see all updates
1. Main facts and figures
In the 3 years from April 2018 to March 2021:
on average, households received 73% of their total income from wages, salaries and self-employment in the 3 years to March 2020 – this is based on ‘gross’ income before tax and other deductions
households in the white other ethnic group received the highest percentage (87%) of their income from employment out of all ethnic groups
households in the Bangladeshi and Chinese ethnic groups received the lowest percentage (68%) of their income from employment
households in the Pakistani and Other ethnic groups had the largest increases in the percentage of income from employment since the 3 years to March 2012
on average, the percentage of income that White British households received from private pensions (10%) was at least 3 times higher than any other ethnic group in the 3 years to March 2021
households in the Bangladeshi ethnic group received the highest percentage (20%) of their income from benefits out of all ethnic groups; this has decreased by 7 percentage points since the 3 years to March 2012
households from the Pakistani ethnic group had the largest decrease in the percentage of income from benefits since the 3 years to March 2012
2. Things you need to know
What the data measures
The data shows the percentage of income from different sources for households in the UK.
Income is measured before tax and other deductions. It includes all household income contributed by everyone who lives there, including children.
The data only covers private households. In this data, a household is either one person or a group of people sharing cooking facilities and a living room or dining area.
Percentages are rounded to the nearest whole number. This means some figures may not add up to 100%. Figures in the commentary may be different to the figures in the charts and tables due to rounding.
Not included in the data
The data does not include:
- people who live in communal accommodation (for example, care homes)
- people with no fixed address (for example, homeless people)
- households where not everyone living there completed the survey
Percentages based on fewer than 100 responses are not shown. This is because estimates based on smaller numbers of people are not reliable. Percentages of less than 0.5% before rounding have been suppressed.
The ethnic groups used in the data
The data uses the 18 ethnic groups from the 2011 Census. Data is aggregated for the Black, Mixed and Other ethnic groups, which means estimates are shown for these groups as a whole.
Some households contain people from different ethnic backgrounds. The ethnicity assigned to the household is that of the head of the household (usually the person with the highest income).
The data does not account for people of different ethnic backgrounds who live in the same house.
Read the detailed methodology document for the data on this page.
Because the data varies from year to year, the Department for Work and Pensions (DWP) takes 3 years’ worth of data and works out the average for that period. This is to make sure there are enough survey respondents to be able to make reliable generalisations. You can find out more about combining multiple years of data to create reliable estimates.
Responses are weighted on the basis of regional population totals by age and sex to give estimates for the entire UK household population. You can read more about how weighting is used to make survey data more representative of the group it is about.
The figures on this page are based on survey data. Find out more about interpreting survey data, including how reliability is affected by the number of people surveyed.
3. Sources of income by ethnicity
|Ethnicity||Wages/salary||Self-employment||Investments||Tax credits||State pension||Private pension||Disability benefit||Other benefits||Other|
|Bangladeshi||61||7||withheld because a small sample size makes it unreliable||7||3||1||2||11||6|
|Chinese||63||5||1||1||2||1||withheld because a small sample size makes it unreliable||3||23|
|Pakistani||64||10||withheld because a small sample size makes it unreliable||6||3||1||2||8||6|
|Black||70||7||withheld because a small sample size makes it unreliable||3||3||3||1||7||5|
|White other||75||12||2||1||2||2||withheld because a small sample size makes it unreliable||2||4|
4. Data sources
Type of data
Type of statistic
Department for Work and Pensions
Note on corrections or updates
The figures for the share of income in the download file before the year ending March 2015 may not match the DWP’s published figures. This is because the way of adjusting for inflation changed from the Retail Price Index to the Consumer Price Index from the year ending March 2015 onwards.
Before 2019 to 2020, respondents from the Irish Travellers (NI) ethnic group in Northern Ireland were included in the White Other ethnic group. From 2019 to 2020, the reporting of this ethnic group was changed for all years presented to harmonise with published figures for the Family Resources Survey and Households Below Average Income, and the ONS harmonised output guidance. Irish Travellers (NI) ethnic group responses in Northern Ireland are now included in Other ethnic group.
Purpose of data source
The main purpose of the Family Resources Survey (FRS) is to give the Department for Work and Pensions (DWP) data to develop, monitor and evaluate social welfare policy.
The survey is also used by other government departments, including for tax and benefit policy modelling by HM Treasury and HM Revenue and Customs. The FRS is also used extensively by academics and research institutes for social and economic research.
5. Download the data
This file contains the following: measure, source, time, ethnicity of household reference person, ethnicity type, geography, value, value type and denominator