This year, along with over 10,000 other organisations across the country, English NHS organisations employing more than 250 people published headline details of their gender pay gap, which was a requirement of the 2017 amendment to the 2010 Equality Act. This data revealed that around nine out of 10 NHS organisations in England had a median hourly pay gap that favoured men.
Over the weekend, it was announced that Professor Jane Dacre, president of the Royal College of Physicians, will lead an independent review into the gender pay gap for doctors.
As part of our own investigation into the gender pay gap, we have analysed over 1.2 million NHS staff records in England to examine the issue in more detail.
In this first blog of a short series, we look at differences in median pay between men and women across the NHS in England, and the variation in pay gaps between staff groups.
Further blogs will decompose the reasons for the pay gaps that we identify here, as well as pay gaps based on ethnic identity.
Staff-level analysis of the gender pay gap
For directly employed NHS staff in the English health service – 77% of whom are women and 23% men – the estimated median basic full-time equivalent (FTE) pay gap between men and women for the month of December 2017 was 8.6%* in favour of men. This was equivalent to an earnings gap of £207 over that month.
* On announcing the inquiry into the gender pay gap in the NHS, the Department of Health and Social Care noted that there is a pay gap in favour of men of 23%. This is based on differences in average pay between men and women, not the median difference as we report here.
Although this is highly significant statistically, this does not mean the median gap is significant in policy terms. (Given a degree of aggregation in the data – see the appendix notes – the median calculated is an estimate.) While not completely comparable, the basic FTE pay gap of 8.6% is similar to that reported by English NHS trusts of 10.3% (based on the ‘median of the medians’) and also that reported by the ONS across all economic sectors: 9.1%. These latter measures are based on an hourly wage, whereas our measure is for a single month.
This chart below shows the distribution of pay for men and women (plus the median basic FTE pay for December 2017) for all 1.2 million staff in the NHS in England. The basic FTE earnings for the majority of staff ranged from £500 to £4,500 a month. But the distribution is very skewed and there is a noticeable group – mainly men – with earnings from £6,000 to £9,500 a month.
Staff and pay groups
The pay distribution in the above chart is an aggregate of all the various staff groups in the NHS, and includes nurses, consultants, managers and others. So how does the gender pay gap vary between staff groups?
This next chart shows the basic FTE pay gap for five staff groups that make up 38% of all staff – consultants, managers and senior managers, nurses and health visitors, midwives, and ambulance staff. All groups have a pay gap in favour of men, ranging from 3.8% for midwives and 4.3% for nurses and health visitors (neither of which are statistically different from 0%), to 2.8% for consultants, 5.9% for ambulance staff and 13.7% for managers and senior managers.
An obvious explanation – though not a justification – for these pay gaps is that men may be disproportionately represented in higher pay grades and/or in staff groups with generally high pay than women.
The pay of the vast majority of NHS staff – around 88% – are set within the Agenda for Change pay system, introduced in 2004 partly to address gender pay differences. As is clear from the next chart, for these 1.1 million NHS staff on Agenda for Change pay bands, men tend to be disproportionately represented in higher pay bands.
However, how this translates into median pay gaps is not necessarily straightforward. Across all Agenda for Change pay bands there is in fact a pay gap of 3.9%, but in favour of women (as shown in the next chart). Within each pay band, however, there is a link between the median pay gap and the proportion of men in each band, with higher pay bands (with a disproportionate proportion of men) showing a (small) pay gap in favour of men.
However, for the 12% (around 135,000) of NHS staff not covered by Agenda for Change – in the main, doctors and some managers – the overall pay gap is 47% in favour of men (as shown in the next chart). This gap is driven by the combination of two sets of staff groups (evident from the final chart below), which combines junior doctors on relatively low pay with consultants and others on relatively high pay.
For associate specialists, those on core training and first year foundation doctors, the gap is not statistically significant. For other non-Agenda for Change staff, the gender pay gaps ranges from fractionally over 0% (for foundation year 2 doctors) to 12.9%.
In terms of the overall gender pay gap across all NHS staff of 8.6%, clearly an important factor is the influence of a fairly small, non-Agenda for Change group of relatively highly paid staff that disproportionally consist of men.
52% of the staff in non-Agenda for Change staff groups are men, compared to 19% of those within Agenda for Change. It is also notable that, within the non-Agenda for Change group, one relatively highly paid group of staff – consultants – not only make up the largest proportion (around 50,000), but is also dominated by men.
Our analysis of over 1.2 million staff employed by the NHS in England has revealed a basic median pay gender gap of 8.6% in favour of men.
Analysis by staff group has also shown that, while the pay gap for the 88% of staff covered by the Agenda for Change pay system is 3.9% in favour of women, the overall pay gap for all staff is partly driven by consultant and other non-Agenda for Change staff groups, which are generally more highly paid and dominated by men.
While gender composition and relative pay between staff groups will explain part of the overall pay gap, other factors will be important too. Further statistical analysis from us will reveal more about the contribution of factors such as staff group, age and gender split (across and between staff groups) on the overall pay gap. We will also investigate pay differences between ethnic groups.
About the data
This analysis is based on a special extract from the NHS Electronic Staff Record (ESR) held by NHS Digital, and covering staff in all but two NHS trusts in England and for one specific month (December 2017). Pay data for that month separately included basic and non-basic pay, as well as total earnings. All pay figures are rounded to the nearest £100. The data is provisional and may change. The data is also anonymous, with 98.4% (1,186,823) of records aggregated on the basis of the seven non-pay characteristics included in the data extract (i.e. where staff group, grade, primary area of work, age group, gender, ethnicity and specialty group matched). Aggregated groups vary in size from two to over 6,500.
Median pay figures were calculated on a disaggregated data set, where the pay data was averaged across those staff in each aggregation. This will have affected the median figures estimated. It is not possible with the data supplied by NHS Digital to calculate the size of this difference from the actual median (based on actual staff-level data within each aggregated block).
However, given the method of aggregation (on seven non-pay characteristics, including pay grade) it is not thought the difference between the estimated median and the true median is substantial. Moreover, as one point of triangulation shows, NHS trusts’ published Equality Act data shows that the hourly ‘median of the medians’ basic pay gap is around 10.3% – compared to the December 2017 month pay gap figure we have calculated of 8.6%.
Appleby J and Schlepper L (2018) "The gender pay gap in the English NHS”, Nuffield Trust comment. https://www.nuffieldtrust.org.uk/news-item/the-gender-pay-gap-in-the-english-nhs