Workforce Race Equality Standard 2025: Did we read the data too quickly?
There is a particular kind of discomfort that comes with data like this. The latest Workforce Race Equality Standard figures suggest that white applicants are now 1.77 times more likely to be appointed from an NHS shortlist than applicants from minority ethnic backgrounds, the widest gap in a decade. It is difficult to look at that statistic and not feel that something has gone backwards, particularly given the sustained policy attention, reporting infrastructure, and organisational effort that has been directed at this issue since the WRES was first introduced in 2016 (NHS England, 2025).
And yet, I can't help but wonder whether we are asking it the right question.
The dominant interpretation is clear and, in many ways, understandable. If the gap is widening, then discrimination must be worsening or at least not improving at the rate we would expect. That interpretation is reinforced by long-standing evidence of unequal labour market outcomes. For example, UK field experiments have consistently shown that applicants with minority ethnic names are less likely to be invited to interview than otherwise identical white applicants (Wood et al., 2009; Heath and Di Stasio, 2019; Department for Work and Pensions, 2019).
But that is not the point at which the WRES recruitment metric is measured.
The WRES indicator does not capture who gets onto the shortlist. It captures what happens after shortlisting. And that distinction matters more than we sometimes acknowledge.
What the metric actually tells us
At its simplest, the indicator is a ratio: the likelihood of appointment from shortlist for white candidates compared to minority ethnic candidates. Implicit in its use is the assumption that the shortlist itself represents a pool of broadly comparable candidates, such that differences in appointment outcomes point towards bias or discrimination within the selection process (NHS England, 2025).
That assumption is not necessarily wrong, but nor is it automatically true.
There is a growing body of research which suggests that diversity initiatives focused on shortlisting can, in some circumstances, change the composition of candidate pools without necessarily changing underlying pipelines of experience, preparedness, or perceived readiness for senior roles. Dobbin and Kalev’s longitudinal analysis of diversity programmes found that compliance-driven interventions, including formalised recruitment processes, often have limited impact on outcomes unless paired with deeper organisational changes around development and progression (Kalev et al., 2006; Dobbin and Kalev, 2016).
Their argument is not that diversity efforts are misguided, but that interventions which focus on procedural fairness at the point of selection can become decoupled from the more difficult work of building capability and opportunity over time. In other words, organisations can become better at demonstrating fairness in process without materially shifting the underlying distribution of who is ready to succeed in that process.
If we apply that lens to the NHS data, an alternative interpretation begins to emerge.
The risk of procedural compliance without pipeline development
It is entirely plausible that NHS organisations, under sustained scrutiny and with the best of intentions, have become more diligent in ensuring that shortlists are diverse. That is, after all, one of the explicit aims of many equality and inclusion frameworks.
But if the pipeline of candidates has not been strengthened to the same degree, then shortlists may increasingly include candidates whose career trajectories have not yet provided the same depth of experience, exposure, or sponsorship as their peers. This is not a statement about inherent capability. It is a reflection of structural differences in opportunity.
There is strong evidence that such structural differences persist. The NHS Race and Health Observatory has highlighted disparities in access to career development, leadership pathways, and informal networks, particularly for internationally educated staff and those entering the NHS from different health systems (NHS Race and Health Observatory, 2022). Similarly, The King’s Fund has pointed to uneven access to acting-up opportunities, mentoring, and stretch assignments as key barriers to progression (The King’s Fund, 2020).
More broadly, labour market research reinforces the importance of these factors. Work by the OECD finds that differences in recognition of prior experience, access to networks, and familiarity with local professional norms can significantly affect progression, even when formal qualifications are comparable (OECD, 2020). In the UK context, the Institute for Fiscal Studies has shown that ethnic pay gaps are driven in large part by differences in occupational progression rather than entry-level disparities (IFS, 2021).
Taken together, this suggests that the issue may not simply be what happens at the point of recruitment, but what happens in the years leading up to it.
When good intentions produce difficult signals
If organisations are improving the diversity of their shortlists without a commensurate investment in preparing candidates to succeed at that level, then the WRES metric could deteriorate even as intentions improve. That is a difficult proposition, because it runs counter to the narrative of straightforward progress. It also risks being misinterpreted as a defence of unequal outcomes, which it is not.
Rather, it is an argument that the metric may be picking up a different problem than the one it is most often used to diagnose. There is some support for this interpretation in the broader diversity literature. Research on “diversity without inclusion” has highlighted how organisations can meet visible diversity targets while underlying power structures and decision-making norms remain unchanged (Shore et al., 2011; Ferdman, 2014). In such contexts, individuals from underrepresented groups may enter processes but still face disadvantages in how their experience is evaluated or translated into appointment decisions.
At the same time, there is also evidence that when organisations invest in structured development pathways, the outcomes do change. Evaluations of targeted leadership programmes and sponsorship initiatives have shown measurable improvements in progression rates for underrepresented groups (McKinsey, 2020; NHS Leadership Academy, 2018).
The distinction, then, is not between fairness and unfairness, but between surface-level compliance and deeper capability building.
Reframing the question
If we take this perspective seriously, then the WRES recruitment metric becomes less a definitive statement about discrimination at the point of appointment and more a signal that something upstream is not working as well as it might.
It invites a different set of questions.
Are organisations systematically identifying potential among minority ethnic staff early in their careers, or are they relying on individuals to navigate complex systems without structured support?
Are internationally trained staff being given the contextual grounding, mentorship, and sponsorship needed to translate their experience into the UK context?
Are leadership pathways transparent and accessible, or do they still depend heavily on informal networks and sponsorship that are unevenly distributed?
These are more difficult questions than simply asking whether recruitment panels are fair. They require sustained investment, organisational introspection, and a willingness to move beyond metrics that are relatively easy to measure.
A more demanding form of equity
None of this is to suggest that discrimination does not exist, or that recruitment processes should not be scrutinised. The evidence on bias in labour markets is too strong, and too consistent, to ignore. But there is a risk in allowing a single metric to carry too much interpretive weight.
If we focus only on the end point of recruitment outcomes, we may miss the more fundamental issue, which is whether we are enabling individuals to become genuinely competitive candidates for the roles they aspire to. That requires time, investment, and a level of organisational commitment that cannot be reduced to compliance with reporting frameworks.
In that sense, the uncomfortable possibility is that the worsening WRES indicator is not simply a story of failure, but a reflection of partial progress. Organisations may have become better at opening the door, but not yet at ensuring that those who walk through it are equipped to succeed.
And that, perhaps, is the harder work that now lies ahead.
If you would like to talk more about building sustainable career support for minority groups then please get in touch.
References
Department for Work and Pensions (2019) A test for racial discrimination in recruitment practice in Great Britain 2018. London: DWP.
Dobbin, F. and Kalev, A. (2016) ‘Why diversity programs fail’, Harvard Business Review, 94(7–8), pp. 52–60.
Ferdman, B.M. (2014) ‘The practice of inclusion in diverse organizations’, in Ferdman, B.M. and Deane, B.R. (eds.) Diversity at Work: The Practice of Inclusion. San Francisco: Jossey-Bass.
Heath, A. and Di Stasio, V. (2019) ‘Racial discrimination in Britain, 1969–2017: a meta-analysis of field experiments’, British Journal of Sociology, 70(5), pp. 1774–1798.
Institute for Fiscal Studies (IFS) (2021) Ethnicity pay gaps in the UK. London: IFS.
Kalev, A., Dobbin, F. and Kelly, E. (2006) ‘Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies’, American Sociological Review, 71(4), pp. 589–617.
McKinsey & Company (2020) Diversity wins: How inclusion matters. London: McKinsey.
NHS England (2025) Workforce Race Equality Standard (WRES) 2024 report. London: NHS England.
NHS Leadership Academy (2018) Stepping Up Programme Evaluation Report. London: NHS Leadership Academy.
NHS Race and Health Observatory (2022) Ethnic inequalities in healthcare workforce and leadership. London: NHS RHO.
OECD (2020) International Migration Outlook 2020. Paris: OECD Publishing.
Shore, L.M. et al. (2011) ‘Inclusion and diversity in work groups: A review and model for future research’, Journal of Management, 37(4), pp. 1262–1289.
The King’s Fund (2020) Workforce race inequalities and inclusion in NHS leadership. London: The King’s Fund.
Wood, M. et al. (2009) A test for racial discrimination in recruitment practice in British cities. London: Department for Work and Pensions.
Comments