Debates in recent months about Cambridge stepping back from state school admission targets, including commentary by Lee Elliot Major, show how conversations about inequality in higher education often ricochet back to familiar terrain around schooling background and admissions policy.
These debates, of course, matter and contain value. But they also dominate the discourse so completely that they obscure a deeper issue, namely that the contours of regulation shape what is visible and therefore what can be acted upon.
It is here that structural racism becomes difficult for regulation to see.
Yes, disparities are widely measured but are not often interpreted as evidence of it, though it radiates through student experience and remains largely unnamed within the regulator’s core vocabulary as a systemic risk to quality.
By structural racism, I mean the ways in which everyday practices and the design of systems can, over time, produce inequitable patterns within higher education. This is not to suggest universities can unilaterally resolve societal inequalities beyond their reach, however, it recognises that institutional systems within higher education can nonetheless reproduce unequal conditions that shape student experience.
Inequality in higher education is not only about who gets in, it is also shaped by the structures students encounter once they turn up, the organisational conditions that influence whether they feel fully included and able to thrive.
Yet structural inequality, particularly structural racism, remains difficult for regulation to articulate directly, even when disparities are clearly visible in the data. Where issues are not explicitly articulated within the regulatory framework, they are harder to assess, challenge, or treat as systemic risks to student experience and quality, allowing patterns to become embedded and begin to calcify.
Design or disparity
Access and participation plans (APPs) have sharpened the focus on outcomes, with institutions expected to demonstrate progress in closing gaps.
The framework assumes institutions will analyse the causes of persistent disparities, and regulatory scrutiny largely concentrates on whether gaps narrow over time rather than on how deeply institutions interrogate the systems that produce them.
Higher education repeatedly returns to questions of access, while the organisational conditions shaping student experience remain less visible within the regulatory view.
Since the 2021 Sewell Report questioned the language of institutional racism, public bodies have been more cautious in how they frame structural explanations for inequality. Disparities associated with race are extensively measured, but they are more often described as outcome gaps or equality risks than as evidence of structural discrimination.
England measures inequality extensively but is twitchy about describing the systems that produce it.
This caution reflects a broader political climate in which discussions of heritage, migration, and national identity have intensified sensitivities around how racism is named. As Stuart Hall argued in It Ain’t Half Racist, Mum, the language used to describe racism shapes whether it is recognised as structural or dismissed as incidental. Where structural racism is not explicitly articulated within regulatory vocabulary, it is less likely to be treated as an organisational design issue and more likely to register as a statistical disparity.
Name it
Looking comparatively, South Africa’s higher education framework explicitly names racism and structural inequality, reflecting a post-apartheid nation building project that’s far from resolved. Movements such as “Rhodes Must Fall” remind us of that unfinished work. The system nevertheless begins from a premise that inequality is embedded.
Unlike countries that have had to confront colonial injustice within their own public institutions, England has not reconstituted its higher education system through an explicit coming to terms with colonialism.
In New Zealand and Canada, treaty and reconciliation based frameworks embed obligations to Māori and Indigenous peoples across public institutions, including universities. These systems differ in history and design, but they share one idea: system-level harms require system-level vocabularies. England’s regulatory framework has evolved without an equivalent domestic vocabulary that calls out structural racism or systemic inequality.
As a result, inequality is often treated as a measurable outcome rather than as a structural pattern. Staffing illustrates this. Chronic under-representation of racially minoritised staff, particularly in senior academic and professional roles, is well documented. While staffing itself sits outside the Office for Students’ (OfS) direct remit, institutional hierarchies shape many of the experiences and outcomes regulators are concerned with across the student journey.
Culture is not abstract
Research by Nicola Rollock, alongside work such as Gatby and Sanderson’s on professional services hierarchies, shows how racialised harm is reproduced not so much through isolated acts as through everyday institutional practice. For example, how credibility is assigned, whose concerns are taken seriously, and who is proximate to decision making power.
Within the regulatory framework, however, these dynamics tend to register simply as disparities or isolated incidents, rather than as the product of oppressive systems shaping institutional life.
This is where debate can become muddled.
Vice-chancellors are right to argue that culture cannot be regulated directly. But culture is not abstract. It is produced through institutional architecture that shapes who is hired and promoted, who sits on committees, whose complaints are believed, and whose knowledge is recognised. Seeing these systems as peripheral to quality narrows what regulation is able to see, and therefore what it is able to address.
There is also a regulatory asymmetry worth acknowledging. Racial harassment is explicitly addressed within the regulatory framework, yet structural racism is not articulated as a systemic risk in its own right.
The OfS’s approach to sexual misconduct under E6 shows what happens when an issue is treated as serious and deserving of sustained attention. Institutions were required to revisit staff–student relationship policies that may have gone untouched for years, scrutinise power dynamics within postgraduate research supervision, redesign reporting routes, expand training provision, and work more closely with specialist charities and support services. Public communications and institutional websites were updated to clarify expectations and processes.
Sexual misconduct came to be understood not as a series of isolated complaints, but as a systemic risk to student experience and quality. While racial harassment falls within E6, it has not been operationalised with comparable visibility or sustained regulatory focus. Addressing it with similar force would require recognising structural racism itself as systemically produced rather than primarily as an outcome disparity.
Expanding the language
Disparities can signal structural patterns, but only when they are interpreted and assessed as such.
Data alone does not name the institutional processes that produce persistent racial inequalities. Without explicit articulation, structural racism is more likely to register as a statistical outcome than as a systemic condition shaping the whole arc of student experience. Across the sector, substantial structural work has already been underway including sustained efforts to interrogate awarding gaps and engagement with frameworks such as the Race Equality Charter, reflecting ongoing institutional commitment.
Structural racism could instead be understood through risks that the regulator already knows how to address. In practice, this would include risks to fairness in assessment processes, where persistent awarding gaps that cannot be fully explained by prior attainment may reasonably raise questions about how performance is evaluated and moderated, and risks to the credibility of complaints processes where patterns of harm undermine trust.
Framed in this way, structural racism becomes visible as a quality concern embedded in institutional practice. Naming it as such does not require universities to repair societal fissures beyond their reach but brings into view harms that might otherwise remain hidden from scrutiny.
An E6-style approach would not require the OfS to adjudicate racism or regulate culture, both of which sit outside its statutory role. It would shift the emphasis from asking where a gap exists to examining what institutional arrangements plausibly produce persistent patterns, and whether providers have credible mechanisms to identify and address racialised harm as it emerges.
These questions sit at the intersection of quality and equality of opportunity and fall within the OfS’s existing concerns. Greater harmonisation between APPs and E6 would allow these dynamics to be assessed in a single regulatory frame rather than as separate tracks.
A period of leadership transition creates a natural moment to reflect on the scope and vocabulary of its regulatory framework. Structural racism is not absent from regulatory discourse, but it is not positioned as a systemic risk.
Expanding the language through which it is recognised would not stretch the regulator’s remit so much as clarify it, which ultimately shapes what regulation is able to recognise and respond to.
This is so spot on! Great article.
“It would shift the emphasis from asking where a gap exists to examining what institutional arrangements plausibly produce persistent patterns”
I agree – it is the persistence of the same stagnant data patterns for the last 20 years that should be the greatest cause of concern.
It wasn’t until the NHS enacted the WRES and linked it with CQC ratings that we started to see shifts in outcomes.
Thanks so much, Nathan. I’m really glad some of the ideas here struck a chord, and the line you picked out gets to the crux of what I was arguing. The WRES example is an excellent one too, thanks for sharing it. It’s left me wondering how well it has been received over time and whether there are similar regulatory approaches in other sectors or countries. I suspect there’s a worthwhile piece of comparative research in that!
This piece seems (‘structural racism becomes difficult for regulation to see’, ‘structural inequality, particularly structural racism, remains difficult for regulation to articulate directly’) to try to persuade us that something exists, without providing evidence for it, and that it is hard to see or articulate or identify this thing that definitely exists. But does it exist?
Disparities in data are not evidence of structural racism. There are many potential reasons for disparities that should be fully explored before jumping straight to ‘must be racism’. I have seen prejudice and I am aware of students who have experienced racial harassment on a micro, social level but (as someone who is not white British) I have seen zero ZERO evidence of structural racism in higher education. If anything, there is positive discrimination, with universities being desperately eager to have people of colour in the room, to the point of prioritising this over knowledge, experience and skills. And people of colour are given extra space to talk about being a person of colour, to value being of colour over not and protected from alternative views.
I think the pushing of this agenda does not help anyone, especially not students of colour. I have seen students get lost in feelings of victimhood, edging towards paranoia. If you tell people that we live in a racist world, they will believe it and start to experience it even where it doesn’t exist. We are making the world (and the world of higher education) a more racist place.
People who emphasize the idea of structural racism tend to downplay the signifcance of personal racism, such as racial harassment. They argue (or more typically assert) that group disparties are caused by institutional processes and therefore the institutional processes must be changed regardless of how well-founded they may otherwise be. Ironically, such arguments quite often rest on racial stereotyping and various genetic fallacies. It’s not uncommon for advocates of this view to causually talk about European or African ways of thinking or learning. I saw this recently in a Wonke article.
Yes, there are many examples of what from my perspective is straightfoward racism in the ‘anti-racism’ space – eg concepts of Blackness and whiteness (even down to one being capitalised and the other not) and those seen on a binary of good/bad, with many racist stereotypes mixed in. I also find it racist that some groups are no longer even included in conceptual groups of what used to be BAME and is now often ‘global majority’ or ‘minoritised’. Groups seen as ‘white’ eg Roma and Jews are now no longer included, yet they remain minority groups and certainly (I’d suggest, more than ever) experience discrimination. And the whole colour lens has been foisted on us, it is a social construct of division, there are many more meaningful ways to distinguish differences.
This thread reinforces the point made in the article. Yes, data might not always necessarily indicate structural racism, but refusing to ask questions relating to it does. As Sivalingham says, we need to shift ‘the emphasis from asking where a gap exists to examining what institutional arrangements plausibly produce persistent patterns’. If we don’t, we reinforce the idea that some groups of students are ‘naturally’ predisposed to inequality and exclusion. The comment “universities being desperately eager to have people of colour in the room, to the point of prioritising this over knowledge, experience and skills” is problematic. I see a lot of diversity in knowledge, experience & skills across the student body that I teach. Supporting students to flex & extend this while navigating the academy & its expectations is what HE pedagogy is about. HE institutions exist to educate students & can learn a lot from the community they serve. Saying that students from certain backgrounds are ‘lost in victimhood’ is your opinion (and oddly something you afford ethnic groups viewed as ‘white’). It’s mine that this kind of comment is reflective of & reinforces structural racism, undermining the experiences of some students & further silencing those who have legitimate concerns.
Racialisation and structural racism do affect Romani and Jewish peoples & members of these communities (affecting these ethnic/religious populations & groups within them differently) & inclusion/exclusion in HE & society. The comment “seen as ‘white'” reflects this. Romani people are diverse, within and across groups, yet have the same ethnic heritage & are culturally distinct from majority social groups. Racialisation is institutionally weaponised (IMO) to divide Romani peoples as an ethnic group and mark some as ‘included’ (they aren’t) and others as ‘outsiders’. This is structural racism and racialised distinctions should be made visible with questions on how they have occurred, their purpose & value. The article does not exclude any group from this broader conversation. Asking students from ethnic/religious groups that may span the human Dulux skin colour spectrum could be a helpful exercise. In my experience, anti-racism movements have not excluded or racialised Romani peoples; successive govts & society broadly have = structural racism/inequality.
The article is pushing to broaden how HE recognises racism & inequality (note this term is used too) & their operation within and through institutions. This is good for all in HE settings.
The colour lens has been foisted on us all via structural racism – with origins in European colonialism – & has been reinforced in and by the academy. It was not created by scholars who point out the existence of structural racism.
Great article, Shaalinie Sivalingham. Thank you!
Is structural racism against white British people possible in your view?
Thanks for the question, Bobby. Before I answer, could I ask what you mean by strutural racism? I ask because people use the term in different ways and I didn’t want to assume we were working from the same definition. My post wasn’t really intended to settle that debate (though that’s certainly a debate worth having). Rather, it asks whether the OfS has the conceptual language and regulatory tools to examine structural racism given that it already regulates systemic and cultural harms including institutional culture through condition E6 (its harrassment and sexual misconduct condition).
It is fine to use your definition from the article: “By structural racism, I mean the ways in which everyday practices and the design of systems can, over time, produce inequitable patterns within higher education.”
I was so pleased to read your comment. Thank you for the vote of confidence and the insights you thoughtfully injected into the discussion. Your knowledge, especially in relation to the Romani example expanded the conversation in a valuable way. I dont think I’ll look at a Dulux paint chart in quite the same way again!
One can certainly theorise about structural racism, but that is not a job for the regulator. Also, it’s a false dichotomy to say that one has to either accept the thesis that group disparites are caused by structural racism or are natural predispositions.
Why do you think Roma and Jews are excluded from the category (and therefore the support groups, funding etc) of ‘global majority?’ Do you see this as structural racism (I do)?
Thanks for taking the time to engage with the post. I agree with you that disparities alone are not evidence of structural racism or do I argue that every disparity should be attributed to it. The contribution I was making was to explore what the absence of structural racism from the OfS’s regulatory framework means for regulation.
There’s a heap of evidence examining institutional contributors to racial inequalities. I’ve cited a few examples here, by no means an exhaustive list: the HEFCE commissioned Causes of Difference in Student Outcomes (2015), Advance HE, the Race Equality Charter (2015), the LKMco report comissioned by King’s College London, The Underepresentation of Gypsy, Roma and Traveller pupils in HE (2017), UUK’s Closing the Gap (2019), EHRC’s Tackling Racial Harassment: Universities Challenged (2019), which exposed widespread racial harassment and shortcomings in how universities recognised and addressed it.
I couldn’t help but note your suggestion that discussing structural racism is itself harmful to students and risks making higher ed more racist. That’s a significant claim, and I’d be interested in seeing the evidence supporting it.
I have seen students become hyperfocused on race and this leading to significant impact on their wellbeing and/or racism towards other groups. I have observed and heard of increased tension and stress when interacting with other racial groups, feeling of victimhood to the point of paranoia and loss of sense of self-agency and locus of control. I am concerned by anti-whiteness becoming more casual and mainstream. Putting race at the center of social analysis understates the importance of other factors such as class, family background, culture, education, or personal choices. Social inequalities have multiple causes and should not be viewed primarily through the lens of race. Eg a middle class black student from a two parent household has many privileges in comparison to a white working class student from a single parent, deprived family background, with adverse childhood experiences. I am concerned by the lack of support for working class white students and I have heard their frustration at being excluded from funding/projects exclusively for non-white students. I am concerned that constant focus on race reinforces racial categories rather than diminishes their importance. I am concerned about political/ideological bias skewing the overall narrative and understanding of history, putting individual ‘lived experience’ and qualitative data with aims of political activism over objective assessment of quantitative data that considers all possibilities. I am concerned about the stigma attached to questioning the narrative
It’s the start of a very interesting conversation about how we begin to challenge these institutional disparities from a regulatory perspective – and one that I think is critically important!
I’m fascinated by all of this.
When people talk about structural racism, the conversation very quickly turns into an argument about whether the structure really caused the inequality. A gap appears in the data and someone immediately says – mebbie it’s prior attainment, family background, subject choice or something else. Before anyone can discuss changing the system, they have to prove that the system is responsible.
With disability, thankfully we often skip that argument. We already accept that systems can disable people.
A building with only stairs disables someone who uses a wheelchair. A lecture without captions disadvantages some deaf students. An exam that requires someone to write quickly for three hours may test their processing speed as much as their knowledge. An attendance rule can penalise someone whose condition fluctuates. Nobody finds that basic idea especially strange. The problem isn’t simply located inside the person. It is partly created by the way the university has organised things.
That feels very close to what people mean by structural discrimination in other contexts.
Of course, universities tend to accept a fairly limited version of structural disablism. They accept that the standard system creates barriers, but the usual answer is to give an individual student an adjustment.
So the building stays difficult to navigate, but one student gets a special route. The assessment stays badly designed, but one student gets extra time. The timetable stays exhausting, but one student is excused from parts of it. The student has to disclose a condition, produce evidence and ask for an exception.
In other words, we accept the barrier, then individualise the solution.
The stronger argument is a lot more uncomfortable. It says that the “normal” university student has been imagined as someone who can attend every week, concentrate in noisy rooms, process information quickly, communicate in conventional ways, cope with unpredictable workloads and repeatedly explain their medical history to strangers. Once you say that, the issue is no longer just whether a particular student needs help. It becomes a question about who the whole system was designed around.
That version of structural disablism is, sadly, much less comfortably accepted. “The real world” gets invoked a lot. And so on.
Maybe disability has picked up a practical and relatively non-threatening language – barriers, access, adjustments – which allows us to admit that systems exclude people without necessarily questioning the system as a whole.
But structural racism tends to reach that more politically difficult question much sooner. Who designed the rules? Whose behaviour is treated as normal or professional? Whose account is believed? Who benefits from apparently neutral arrangements?
With disability, we often allow unis to say: “Yes, the system disadvantages you, so we will make an exception.”
With race, even saying that the system might be producing the disadvantage can become the entire argument. Including on here.
Dear Jim,
I completely agree with you when you write “With race, even saying that the system might be producing the disadvantage can become the entire argument. Including on here.” In fact, it is sometimes even worse than that.
The latest UCAS admissions data does not provide any graphs or tables about ethnicity. It does provide these for age group, gender, intention to live at home, deprivation, UK region, provider category (low, medium, high tariff) and subject group. Ethnicity data is known, but UCAS “hides” it. The wonkhe write-up of the data also does not include ethnicity data. The government website has a page about participation in higher education by ethnicity. This was last updated 9 July 2025. One would therefore have expected it to have been updated by now (a year and two weeks later) with new data, but it hasn’t (yet) been. References to this data in a wonkhe comment tend to get deleted by the moderators. To an outside observer, this does seem to be “the system” trying to prevent people from noticing an inequality. The conversation isn’t allowed to even get to the stage of “saying that the system might be producing the inequality”.
Race and disability are not like for like. The disability is inherent – eg someone cannot see. We can make structural changes to alleviate the impact of that. What is the inherent issue with race?
Absolutely. There is wide agreement that reasonable adjustments should be provided for students with motor, visual, or hearing disabilities, e.g., accessible classrooms, appropriate learning materials, and interpreters.
The claim that the teaching materials, assessment methods, or curriculum of, say, a physics degree are structurally racist is different, and requires clear evidence. What specific features are alleged to be discriminatory, by what mechanism would they affect students from different racial groups, and what evidence shows that they contribute to disparities in outcomes? Without a careful analysis of those questions, proposed reforms become tokenistic, or can lead to stereotypical views about different racial groups.
The argument is not that a physics curriculum is inherently racist, or that every disparity proves structural racism. It is that apparently neutral institutional arrangements can produce patterned disadvantage, and regulators should be able to ask how.
With disability, we accept that a system can create barriers and then examine the design. With race, you appear to require a complete causal proof before the design can even be investigated. That is the asymmetry I was pointing to.
TASO (the OfS-funded What Works Centre) has mapped 16 institutional approaches to ethnicity awarding gaps, tested curriculum reform using quasi-experimental methods, and developed theories of change covering assessment, belonging and departmental practice.
Its conclusion is not that every disparity proves racism. It is that persistent disparities justify identifying and testing possible institutional mechanisms – sometimes finding that proposed reforms have little effect. “Where is the evidence?” can’t sensibly mean that the system must not be investigated until its causal role has already been conclusively proved!
In the example of a physics degree, a starting point would be to ask what are the plausible mechanisms by which curriculum choices could lead to awarding gaps among different ethnic groups when controlled for prior attainment. I am not asking for watertight proof, just something intuitive, at the level of the idea that a deaf student is at a disadvantage in an oral presentation. What would you suggest?
The work by TASO is once again a perfect illustration of my point (thanks for the hat tip Jim). Filtering the reports by topic “race and ethnicity” gives three reports which all deal with ethnicity degree awarding gaps. There are no reports dealing with participation gaps by ethnicity. TASO stands for Transforming Access and Student Outcomes in higher education and their stated mission is “Building the evidence base on what works to reduce equality gaps in higher education” and they write “Our work is funded by the Office for Students”. One would think that the ethnicity access gap is something they would look into, but apparently not.