Data from across civil service departments and agencies has consistently shown that employees from BAME backgrounds are less likely to receive the top performance rating and more likely to receive the lowest rating compared to their colleagues from non-BAME backgrounds.

The same is true for employees across other protected characteristics including disability and sex.

For example, among the Ministry of Defence civilian workforce, the differences between staff who declared as White and BAME were statistically significant for both ‘exceeded’ and ‘partially/not met’ appraisal markings – 11.5 percentage points and 2.8 percentage points respectively in 2017-18.

In the Ministry of Justice, a higher proportion of White staff (14%) were awarded an ‘Outstanding’ appraisal rating than BAME staff (8%). BAME staff (3%) were more likely than White staff (1%) to be awarded a ‘Improvement Required’ rating in 2017-18.

Civil Service Employee Policy (CSEP), the behavioural insights team (BIT) and seven government departments ran a project that aimed to improve outcomes for black, Asian and minority ethnic employees.

From September 2016 to February 2018, BIT designed and implemented a randomised control trial to test the effect of a small ‘nudge’ on manager behaviour.

BIT emailed a best practice management checklist to line managers which asked them to carry out a number of actions throughout the year.

The hope was that this would:

• encourage more effective performance management
• lead to greater consistency between BAME and non-BAME employees’ experience of performance management, and
• result in improved line management for BAME employees.

BIT compared the group who had received the email with a control group who received a placebo checklist unrelated to performance management.

It did this by sending a survey to employees before and after the manager had received the checklist. Based on initial analysis of the survey results, CSEP and the BIT concluded that:

• delivering the best practice checklist to managers via a link in an email had no impact on its primary outcome measures for BAME and non-BAME employees
• overall manager behaviour matters to employees
• the trial generated three key insights for further exploration and testing; these were:

1. Imbalances in performance ratings could be result of unconscious bias; the actual quality of line-manager/employee interactions; or the need for changes in wider processes and practices in the civil service.
2. The checklist, and other management interventions, may be more successful at improving line manager behaviour and employee sentiment if delivered by more direct and timely channels. But BIT did caution that this “may not close the gap in performance ratings between BAME and non-BAME employees.
3. There is confusion among managers about the importance of implementing best practice management. Many line managers said they did not need to meet employees formally on a regular basis because ‘they sit next to them and talk to them every day’.

The report on the trial, published in October 2018, said that although the trial was not successful in achieving the desired outcome, “we gathered useful new evidence on how to reduce diversity imbalances and improve performance management.”

Prospect deputy general secretary Garry Graham said: “For any performance management system to operate effectively, it must have the confidence of staff.

“We still have a long way to go in ensuring that performance management systems operate fairly and do not result in discriminatory outcomes and this is a key issue for Prospect.

“Prospect has consistently argued for the decoupling of pay and performance management outcomes against the backdrop of these types of figures.”

BIT is owned by the UK Cabinet Office, innovation charity Nesta and its employees.