The data doesn’t include people working for:
- the NHS
- non-departmental public bodies
- the Civil Service abroad
- the Northern Ireland Civil Service, which supports the devolved government in Northern Ireland
- central government in security posts
It doesn’t include people not on the payroll or not being paid during the time period covered, for example contractors or staff taking unpaid leave.
There are 2 measures of the senior Civil Service available: the Senior Civil Service and ‘SCS Level’. ONS statistics measure ‘SCS Level’ employees, which include a number of health professionals, military personnel and senior diplomats, as well as those working in the Senior Civil Service. As a result, these statistics may be different to those used for monitoring diversity of the Senior Civil Service.
Ethnicity wasn’t known for 25% of civil servants and, as a consequence, it is possible that average annual pay is under- or over-estimated for staff in some ethnic groups in some grades or regions. The median pay of staff with unknown ethnicity is included in the tables for reference.
Median annual pay is presented in this measure, rather than mean annual pay. For each ethnic group, if every respondent was lined up in the order of their full-time equivalent (FTE) annual pay, the median would be the pay of the person in the middle. Using the median as a measure of average annual pay prevents the average being influenced by staff with very high or very low pay, which may not give an accurate representation for the majority of staff, so distorting the figures.
However, measures of average pay are less reliable for ethnic groups with small numbers of staff. It should be noted, for example, that median pay among Senior Civil Service staff is based on 10 staff in the Chinese group, 20 staff in the Other group, 50 staff in the Black group, and 60 staff in the Mixed group (headcounts are available in the download file).
Mean annual pay (worked out by adding all staff’s FTE annual pay and dividing by the total number of staff) is not presented here. Where some employees have much higher salaries than others in the same ethnic group, the mean annual pay for that ethnic group will be higher, and as a result may not accurately represent the annual pay of the less well-paid employees in that group.
Data on the Civil Service workforce shows that the percentage of staff in each region, grade, gender and age group varies by the different ethnic groups. An ethnic group with a higher percentage of staff in regions with higher pay (such as London, which has a regional allowance) or a higher percentage of staff in more senior grades will have a higher average annual pay than an ethnic group with more staff in lower paid regions or junior roles.
For example, this data shows that White staff are amongst the lowest paid in the Civil Service overall and within each grade. However, within many of the regions, White staff are the most, or among the most, highly paid of all ethnic groups. The Civil Service workforce data shows that White staff make up a higher percentage of staff at senior grades but a lower percentage of London-based staff (who receive a regional allowance) than those in other ethnic groups.
Note that in this measure, the median annual pay by ethnicity is shown by grade and then by region. While this enables salaries of staff living in the same region or at the same grade to be compared, partly overcoming the complexities mentioned, there is no comparison by ethnicity, grade and region simultaneously.
When comparing pay of staff at the same grade, please be aware that these grades are broad, and the grouping together of all Senior Civil Service grades in particular encompasses a very wide pay range. The balance of staff within these broad grades may vary by ethnicity and also by region, so affecting estimates of median pay.