Household wealth by ethnicity, Great Britain: April 2016 to March 2018

• Median total wealth for all households in Great Britain was £286,600 between April 2016 and March 2018, with medians ranging from £34,000 among those with a household head from the Black African group to £314,000 for the White British group.
• There were statistically significant differences in wealth by ethnicity of the household head, even after adjustment for a range of household characteristics including age and household composition, with largest differences estimated between the White British reference group and the Pakistani and Black African groups.
• Households with a White British head were approximately nine times as likely to be in the top quintile of total wealth (wealth above £865,400) as those of Black African ethnicity and 18 times as likely as those of Bangladeshi ethnicity.
• The percentage of households with financial debts that exceed their financial assets was highest for the Black African and Other Asian groups (both 44%) and was twice as likely for these households compared with the White British group.
• Households with an Indian, Pakistani or White British head had the highest net property wealth (medians of £176,000, £115,000 and £115,000 respectively) and were the most likely of all ethnic groups to hold net property wealth, with 80% (Indian head), 73% (Pakistani) and 69% (White British) of households having net property wealth.
• There is evidence of differing degrees of private pension participation by ethnicity with lowest participation in Bangladeshi (48%), Chinese (57%), Any other ethnic groups (58%) and Black African (59%) headed households (compared with 83% and 82% for Indian and White British ethnic groups respectively).

The ethnicity pay gap has been widely reported, highlighting inequality in earnings and income by ethnic group. The ethnicity wealth gap is less well defined, primarily because of limited data.

There are two main reasons for this; there are few data sources on household wealth in the UK and social surveys can be limited in terms of sample size to be sufficiently robust when considering minorities within a population. The Wealth and Assets Survey (WAS) has a relatively large sample size (around 18,000 households) compared with many social surveys and has a breadth of coverage on wealth, assets and debt, which make it well suited to this type of analysis.

Analysis of household wealth by ethnicity is further complicated by differing demographics of the population (and survey sample) by ethnicity. Some ethnic groups are younger on average than others (for example, Black African versus White). This is reflected in the survey sample, but is also true for the population as a whole, according to the census. Age has a particularly strong influence on wealth, much more than on income or earnings. Median wealth in Great Britain increases by more than a factor of 10 from ages 25 to 34 years to 55 to 64 years, while average income and earnings by comparison vary little with age.

There are also differences in factors such as household composition, housing tenure (whether a homeowner or renter) and employment type between ethnic groups, which ultimately affect wealth. For example, there are relatively high levels of employment among those aged 16 to 64 years for White (77%) and Indian (76%) ethnic groups compared with Pakistani (57%) and Black ethnic groups (67%). Indian and White individuals are also more likely to be employed in higher-paid professional jobs, whereas Black individuals are more likely to work in lower-paid “elementary” jobs. Similarly with home ownership rates, households in the Black African (20%) and Arab (17%) ethnic groups have the lowest rates of home ownership, compared with 68% of White British and 74% of Indian households. Lower employment, lower earnings and lower home ownership rates will all limit the ability to accumulate wealth.

We will show in later sections that there are large differences in household wealth by ethnicity, which will be a result of these and other factors. In this section we use a modelling approach to isolate the impact of ethnicity from other household characteristics, to estimate differences in wealth associated with ethnicity alone. We used linear regression to model the difference in wealth between ethnic groups, controlling for age, sex, housing tenure, household composition, socio-economic group and educational attainment.

Figure 1 and Figure 2 show that even when controlling for these factors (in effect taking away differences in age, home ownership and so on) there remain statistically significant differences in the total wealth of a household depending on the ethnicity of the household head (household reference person). This suggests that the inequality in wealth by ethnicity is not explained by differences in the factors that we have adjusted for.

In some cases, the ethnicity of the household head is not statistically significant in explaining wealth differences. This does not mean that household wealth is the same when comparing groups, but suggests that differences in total wealth may be explained by differences in the characteristics controlled for in the analysis (for example, levels of home ownership or the age distribution).

Figure 1 shows the estimated effect of ethnicity on total wealth compared with a White British reference group, when the listed characteristics are held constant. Large confidence intervals represent the uncertainty around these estimated marginal differences in total wealth.

Nevertheless, comparisons for all except the Black Caribbean, Chinese and Other White groups were statistically significant. Total household wealth was estimated to be up to £271,000 higher for the White British group than the Pakistani ethnic group, after adjusting for characteristic differences such as age and household composition.

Differences in total household wealth were also estimated between ethnic minority groups (Figure 2). For example, total household wealth in the Mixed group was estimated to be lower than the White British group, but higher than the Pakistani and Black African or Other Black groups. No statistically significant differences were estimated between the Black Caribbean, Other White, and Chinese ethnic groups, or between the Black African or Other Black, Bangladeshi, Pakistani, and Indian ethnic groups.

Figure 2: Statistically significant differences in total household wealth were estimated between ethnic groups, after factors such as age and housing tenure have been accounted for

Statistically significant differences in total household wealth by ethnicity of the household head when controlling for other factors;Great Britain, April 2016 to March 2018

Notes:

1. Factors controlled for are age, sex, educational level, and socio-economic classification (NSSEC) of the household head, housing tenure and household composition.

Download the data

This section describes in more detail the average levels of wealth by ethnic group (with weighting applied to bring estimates up to the population of Great Britain). Variation in wealth seen across ethnic groups may be influenced by factors such as those controlled for in Section 2.

There are clear differences in average total wealth depending on the ethnicity of the household head (Figure 3). The median total wealth for a household in Great Britain is estimated to be £286,600. Only the White British and Indian ethnic groups have median wealth higher than the average for Great Britain. The Black African group has the lowest median total wealth (£34,300).

Mean total wealth is also highest for the White British group, while the pattern across other ethnic groups differs. However, the median is a preferred measure of central tendency for wealth, because it is less affected by a small number of very wealthy households.

From our data we can estimate the full distribution of wealth across households in Great Britain, and from this we can assess where households from different ethnic groups are most likely to sit within this distribution. All households were ranked from lowest to highest total wealth and split into five equal groups, with the same number of households in each (quintiles). Figure 4 shows the percentage of households within each quintile. If total wealth were equally distributed across ethnic groups, then each bar would be equal in size. However, when looking at households by ethnicity this is not the case, highlighting inequality between ethnic groups in total wealth. Apart from the White British and Indian ethnic groups, households from all other groups are more likely to sit in a lower wealth quintile than average.

Households from the Black African group were the most likely to be in the lowest quintile of total wealth. More than half (54%) were in the lowest quintile (total wealth below £42,200), while more than three-quarters (79%) were in the lowest two quintiles (total wealth below £185,600). Only 2% of households from the Black African group were in the highest quintile (total wealth above £865,400). In contrast, households from the White British group were approximately nine times as likely to be in the highest wealth quintile and 18 times as likely as those from the Bangladeshi ethnic group.

Factors such as age and tenure (described in Section 2) are likely to influence this variation in wealth across ethnic groups. Since age is an important factor in the accrual of wealth for example, we might expect ethnic groups that are typically younger (for example, Black African) to have less wealth. However, our modelling in Section 2 suggests this is unlikely to explain all the differences. Households from the Black Caribbean group are also considerably less wealthy on average than the White British group, despite having a similar age profile.

The total wealth of a household is made up of net wealth from four components: property, financial, physical and private pensions. As for total wealth, when looking at each component of wealth separately (Figure 5), households from the White British or Indian ethnic groups were consistently among those with the highest median wealth. The Black African, Bangladeshi or Any other ethnic group were typically among those with lowest median wealth.

Figure 5: Similar patterns of median wealth are seen across ethnic groups for each component of total wealth

Median components of total household wealth (£) by ethnicity of the household head; Great Britain, April 2016 to March 2018

Notes:

1. Analysis includes all households, whether or not they hold each type of wealth.
2. Results for Great Britain include households where the ethnicity of the household head is unknown.
3. Results for ‘Black – Other’ are not presented because the surveyed sample size is less than 30 households. Results for the Bangladeshi and Chinese ethnic groups are based on less than 50 unweighted cases and such data should be treated with some caution.

Download the data

Overall, the data show considerable differences in the composition of total wealth by ethnic group, both in the amounts held (Figure 5) and the mix of components (Figure 6). On average, net financial wealth accounts for a relatively small percentage of a household’s total wealth (12%) and the remainder is roughly evenly split between physical wealth (27%), net property wealth (29%) and private pension wealth (32%). However, Figure 6 shows that this composition differs by ethnic group. Property and pension wealth make up a larger proportion of total wealth in ethnic groups with higher average total wealth, while financial wealth makes up a relatively small component of total wealth across all ethnic groups.

Net property wealth

Net property wealth is the value of any property or land owned, less any outstanding loans or mortgages held against them. There is large variation in median net property wealth between ethnic groups (Figure 5). Households from the Indian ethnic group have the highest median property wealth (£176,000) and are the most likely to hold property wealth (80%, Table 1). In contrast, Black African, Black Caribbean and Any other ethnic groups have median net property wealth of £0; a consequence of (among other factors) being less likely to hold this type of wealth (Table 1).

Low or zero net property wealth can reflect low rates of home ownership, or households with larger values of loans or mortgages relative to the value of property owned. For younger ethnic groups, this may be partly linked to their age profile. For households purchasing property with a mortgage, net property wealth will grow over time as they pay more towards their mortgage and the amount outstanding is reduced. Households headed by one of the Black ethnic groups are also more likely to live in London than any other region, where high property prices coupled with relatively low earnings may hinder these groups from accruing property wealth. As a proportion of total wealth, households from the Pakistani or Indian ethnic groups have relatively high property wealth (44% and 46% respectively) compared with other households. Compared with Black ethnic groups, the Asian population in Great Britain is distributed more evenly across the regions, potentially allowing routes to home ownership through lower house prices outside of London.

Net financial wealth

Net financial wealth is the value of any financial assets (for example, savings and investments) less financial liabilities (such as outstanding credit card balances and loans). Of the four components, median net financial wealth is most similar across ethnic groups with only a £9,000 difference between the highest and lowest median values.

Some households have negative net financial wealth, where the value of their debts exceeds their assets. Negative net financial wealth has a proportionally larger impact on total wealth in the Black African group than other households (making up an average 7% of total wealth compared with 3% for all households). The highest percentage of households with negative net financial wealth was estimated for the Black African and Other Asian groups (44%). These households were approximately twice as likely to have negative net financial wealth as those from the White British group (Table 1).

Physical wealth

Physical wealth is the value of household contents, vehicles, possessions and valuables. When considering this form of wealth there is relatively little difference in median wealth between ethnic groups (Figure 5). However, as a proportion of total wealth, the contribution of physical wealth is highly variable and tends to decrease for groups with higher average total wealth. For example, physical wealth makes up 51% of total wealth in households with a Black African head compared with 20% for those with an Indian head.

Private pension wealth

Private pension wealth is the value of any pension pots accrued that are not related to the State Pension. There is evidence of differing degrees of private pension participation by ethnicity (Table 1) with participation (those with private pension wealth) lowest in Bangladeshi, Chinese, Black African and Any other ethnic groups. Median private pension wealth ranges from less than £5,000 (Bangladeshi, Black African, Chinese and Any other ethnic groups) to £80,000 (White British) and this trend across ethnic groups correlates with the percentage of households that hold this type of wealth (Table 1).

As pension wealth is built up over time until retirement, differences in age and household composition between ethnic groups are particularly important when considering this type of wealth. Proportionally, private pension wealth is highest for the Black Caribbean (35%) and White British (33%) groups, whose household heads are older on average than other ethnic groups. However, differences in age profile are unlikely to fully explain differences in private pension wealth.

Disparities in employment rates and earnings will contribute to the ethnicity pensions gap (both participation and level of amounts held), together with variation in rates of self-employment (PDF, 285KB). Knowledge of pensions and the likelihood of participating in available pension schemes are also known to be lower among some ethnic minority groups (PDF, 526KB).

Financial wealth (net)

The values of any financial assets held, both formal investments such as bank or building society current or saving accounts, investment vehicles such as Individual Savings Accounts (ISAs), endowments, stocks and shares, and informal savings (money under the bed or loaned to family or friends) and children’s assets; less any financial liabilities such as outstanding balances on credit cards, arrears on household bills, loans (including student loans) from formal or informal sources.

Household reference person

The household reference person (HRP) is the person that is the sole or joint householder or is responsible for household affairs. Where there are joint householders, the HRP will be the person with the highest income. In cases where income is the same for a joint householder, the eldest person is assigned as the HRP.

Physical wealth

The (self-evaluated) value of household contents, possessions and valuables owned such as antiques, artworks, collections and any vehicles owned by individuals (including the value of any personalised number plates).

Private pension wealth

The value of any pension pots already accrued that are not state basic retirement or state earning-related. This includes occupational pensions, personal pensions, retained rights in previous pensions and pensions in payment. Estimating the value of some private pension pots is straightforward. For example, if a pension is a defined contribution type scheme (not in payment) the valuation is obtained from the respondents’ latest statement from their pension administrator. This is an accurate estimation of each individual pension pot taking into account any relevant market influences (for example, investment returns).

Property wealth (net)

Respondents’ self-valuation of any property owned, both their main residence plus any other land or property owned in the UK or abroad, less the outstanding value of any loans or mortgages secured on these properties. Self-valuation tends to yield higher estimates of worth than most other property indicators may suggest – however, when assessing individuals’ opinions or behaviours, it is this perceived worth that will drive the individuals concerned.

Statistical significance

More information on uncertainty and how we measure it is available.

Data sources and collection

This article reports analysis of household wealth by ethnic group based on 18,034 households in Great Britain who responded to the Wealth and Assets (WAS) Survey between April 2016 and March 2018 (inclusive). The associated datasets contain further results from data collected in earlier surveys.

The Wealth and Assets Survey (WAS) launched in 2006 and is a biennial longitudinal survey conducted by the Office for National Statistics (ONS). This survey measures the well-being of households and individuals in terms of their assets, savings and debt and planning for retirement. The survey also examines attitudes and attributes related to these. Classificatory variables (age, sex, employment status) are also collected.

The survey is currently sponsored by a funding consortium, including the ONS, Department for Work and Pensions (DWP), HM Revenue and Customs (HMRC) and the Scottish Government (SG). Interviewers working on the survey refer to it as the Household Assets Survey (HAS).

The first wave of the survey commenced with interviews carried out over two years from July 2006 to June 2008 with approximately 30,000 households interviewed. A second wave took place two years on from initial interviews, covering the period July 2008 to June 2010, with approximately 20,000 households obtained. Wave 3 obtained approximately 21,000 households, Wave 4 obtained 20,000 and Wave 5 obtained 18,000.

This periodicity was maintained until Wave 5, which covered the period July 2014 to June 2016. The survey has now moved to a two-year, financial year-based periodicity (April to March) with this periodicity being referred to as a “round”. Therefore, Round 6 covers the period April 2016 to March 2018. This move to a two-year, financial basis allows WAS to be integrated with other household financial surveys that are based on financial years, and allows it to be analysed on a consistent basis alongside other components included within other household financial surveys (income and expenditure). Round 6 achieved approximately 18,000 household interviews.

More quality and methodology information on strengths, limitations, appropriate uses, and how the data were created is available in the Wealth and Assets Survey (WAS) QMI.

Ethnic groups used in the data

The data uses the ethnic categories from the 2011 Census. The ethnicity assigned to each household is that of the head of the household (referred to as the household representative person). The analysis does not account for people of different ethnic backgrounds who live in the same household.

The breakdown of ethnic groups used in this article was guided by the number of households available for analysis. Where sample sizes are large enough, estimates are shown for individual ethnic groups. However, we have combined data for some ethnic groups to ensure we can provide robust estimates. Data are aggregated for Mixed or Multiple ethnic groups, which means that estimates are shown for this group as a whole.

The Ethnic minority (Great Britain) group includes all ethnic groups other than White British.

Estimating average total wealth and contributions from different types of wealth

When estimating average wealth, we use both the mean and median as measures of the average. Mean wealth is the total wealth of all households divided by the number of households. Median wealth is the wealth of what would be the middle household, if all households were sorted from poorest to richest. The median is generally a better measure of central tendency for wealth, because it is less effected by very wealthy outliers included in the data.

We use linear regression modelling to investigate the effect of ethnicity on total wealth, while holding other household characteristics constant. Note that because of the method of estimation used, data in this section are reported in mean and not median terms. This must be considered when comparisons between the two are made, as the mean is more susceptible to the influence of outliers than the median.

The regression estimates relate to a hypothetical population in which several household characteristics are controlled for: age, sex, educational level and socio-economic classification of the household head, household composition and housing tenure. Geographical region was not controlled for, as while there is an association between region and wealth, it is unclear whether region determines wealth or wealth determines region. In the case of the latter, controlling for region would distort the model estimates.

The total wealth of a household is the sum of net wealth held in property, private pensions, financial and physical items. When estimating the relative contributions of each type of wealth to a household’s total wealth we consider both positive and negative values of wealth. This is because debt is an important factor for total household wealth. For example, if a household had physical wealth worth £9,000, net financial wealth of negative £1,000 and no net property or private pension wealth, it would be estimated that 90% of the household’s total wealth came from physical wealth and 10% from net financial wealth.