University of Sussex v OfS - practical lessons for regulatory investigations

The High Court's reasoning provides valuable commentary on investigative conduct, the dangers of bias and predetermination and the risk of pursuing broader policy objectives through individual enforcement cases. That risk is not confined to the higher education sector and the judgment merits close attention from those who come into contact with regulatory investigations.

Bias and predetermination

The concepts of bias and predetermination are related but distinct.

There are three main categories of bias: actual bias, apparent bias and automatic disqualification for interest (sometimes referred to as presumed bias). Apparent bias was at issue in Sussex.

As explained by the Court of Appeal in Bubbles and Wine v Lusha [2018], the test for apparent bias is whether all the circumstances would lead a fair-minded and informed observer to conclude there was a real possibility of bias (Porter v Magill [2001]). Bias means a prejudice against one party or its case for reasons unconnected with the legal or factual merits of the case (Flaherty v National Greyhound Racing Club Ltd [2005]).

Predetermination is conceptually distinct from bias, although related. As Richards J explained in Georgiou v London Borough of Enfield [2004], predetermination involves the decision-maker "approaching the decision with a closed mind and without impartial consideration" of all relevant matters. However, Richards J notes that the test for predetermination should be treated with caution so as not to render decision-making impossible or unduly difficult.

On this issue, the Divisional Court's recent decision in R (Nobel Oil) v Oil and Gas Authority [2025] is instructive. The Court recognised that a regulator is entitled (and indeed required) to have a strategy and prioritise the allocation of its resources. These considerations will naturally inform its approach and a regulator may well have a predisposition towards a particular outcome. What it must not lose, however, is a genuine openness to evaluating the relevant material. Nor should it disregard the strategic context of its role or the practical constraints on its resources. In short, a decision-maker must have an “open mind”, but need not have an “empty mind”.

The vice of predetermination, as Ouseley J set out in R (Bovis Homes) v New Forest DC [2002], is that "the decision-making process will not then have proceeded from reasoning to decision, but in the reverse order. In those circumstances, the reasons given would not be true reasons but a sham." 

A conflict of interest?

A key issue in Sussex was the role played in the regulatory investigation by the OfS Director of Free Speech, Dr Arif Ahmed, who joined the OfS in August 2023 as Director for Freedom of Speech and Academic Freedom. 

Concerns arose due to his being known to and having expressed support for Professor Kathleen Stock during his previous career as an academic. For example, in 2022 Dr Ahmed published an article in which he wrote: "Kathleen Stock suffered victimization for holding and legally stating the beliefs of a gender-critical feminist. Although these beliefs, like race, constitute a legally protected characteristic, her employer [the University of Sussex] seems to have permitted the harassment to continue, and to escalate, for three years."

The court considered whether the OfS' decision to fine the university was invalidated either by apparent bias (because of Dr Ahmed's involvement) or predetermination. Ultimately, it held that: "The Final Decision was vitiated by bias because the OfS approached the decision with a closed mind and had therefore unlawfully predetermined the decision." 

The complexities of bias are beyond the scope of this article. It is, however, worth noting that the judgment brings together the concepts of bias and predetermination. As explained above, these are analytically distinct doctrines: bias is assessed by reference to the perspective of the fair-minded and informed observer, whereas predetermination concerns whether the decision-maker had a closed mind. Although the court’s reasoning engages with each concept, the conclusion draws them together, reflecting the close relationship between a closed mind and the appearance of partiality on the particular facts.

On whether there was apparent bias in respect of Dr Ahmed’s role in the decision-making, the court considered whether he had a personal friendship with Professor Stock that could give rise to apparent bias by reason of conflict of interest. Conflict of interest is not a separate legal category of bias, but rather a circumstance that can give rise to it. In Sussex, the OfS initially recognised Dr Ahmed's conflict of interest. Internal emails from the OfS’ then chief executive, Susan Lapworth, in June 2023 record her view in clear terms: "I know that he knows KS [Professor Stock] and expect he [Dr Ahmed] will have discussed relevant issues with her. So, I think he is conflicted.” That position seems to have been shared by others within the OfS and by Dr Ahmed himself. Even so, in October 2023, Ms Lapworth decided that Dr Ahmed should take on responsibility for the day-to-day conduct of the investigation, by which point the OfS had already issued its Provisional Decision.

The court did not conclude that Dr Ahmed’s involvement, taken on its own, rendered the decision unlawful on grounds of apparent bias. It gave two principal reasons. First, it was satisfied that Dr Ahmed’s relationship with Professor Stock was professional rather than personal. Secondly, and more significantly, Dr Ahmed was not the decision-maker and only became involved after the Provisional Decision had been issued. In those circumstances, the court considered that his role in the making of the Final Decision was not sufficiently central for any apparent bias on his part to invalidate the outcome.

However, although the apparent bias challenge failed, the court went on to hold that the Final Decision was unlawful and vitiated by predetermination because the decision-maker had approached the matter with a closed mind and had therefore unlawfully predetermined the result.

Predetermination in Sussex

The court identified five indicators of a closed mind:

1. The overarching strategy. Even before the investigation opened in October 2021, Ms Lapworth was actively seeking a "test case" that would send "a strong signal" to the sector. Although the principle of using a test case is not in itself wrong, the court concluded that Ms Lapworth's evidence pointed to "a fixed intent to find the University in breach of the conditions, in order for the case to have its requisite incentivising effect on the sector." As noted above, it is within the OfS' power to prioritise free speech, but such a policy must be applied in a fair manner.
2. The approach to settlement. The OfS refused to meet with the University to discuss settlement unless it accepted the alleged breaches in their entirety. Although placing conditions on a meeting is not determinative of a closed mind, this was, the court found, indicative of a fixed view to find breaches so as to further the investigation's objective.
3. The single-target approach. Multiple universities were using materially the same policy template, yet only Sussex was investigated and sanctioned. The OfS wrote to the other institutions only after the Final Decision was published, highlighting the breach and the fine. If the OfS’ principal concern had been to change the Policy Statement itself, and secure freedom of speech and academic freedom in other HE providers, then the court concluded that the quickest and most effective measure would have been to find out which HE providers were using the offending policy and communicate with them.
4. The failure to have regard to the University's Freedom of Speech Code of Practice, described in  OfS' guidance as the institution's "definitive and up to date statement" on the correct approach to free speech. 
5. The refusal to consider whether the University had remedied the alleged breaches by the time of the Final Decision, despite having had ten months to review the limited changes. The court concluded that acknowledging compliance "would have undermined the strategy that Ms Lapworth and the OfS had adopted in October 2021, pursued for three and a half years, and invested very significant resources into. The evidence supports a finding that the OfS had closed its mind to anything that would lead to not finding breaches and being unable to therefore sanction the University." 

Nor did the existence of an independent decision-making committee (the USCEC) prevent a finding of predetermination. The USCEC adopted the draft decision in its entirety, with no contemporaneous evidence of independent challenge. Both Ms Lapworth's and the investigation lead, Dr Arif Ahmed, were present when the Final Decision was taken.

Policy delivery disguised as adjudication

The Sussex decision highlights a structural vulnerability that goes beyond the regulation of higher education in England. Predetermination risk is heightened when a single regulator simultaneously sets policy, selects its enforcement target, runs the investigation and frames the evidence for its own decision-maker(s).

This concern is not unique to the OfS. Any regulator with a dual mandate (setting standards and enforcing them) faces the same tension. The risk materialises most acutely in cases that have become symbolic or where significant resources have been committed to a particular outcome, making it difficult to back down from a position.

Practical guidance: designing a lawful investigation

The judgment in Sussex offers clear lessons for regulators. First, there needs to be a separation between those setting regulatory policy and those making or shaping enforcement decisions. If the same people are involved in developing strategic priorities, steering the investigation and sitting in on the final decision, it becomes difficult to show that the outcome was reached with an open mind. That separation should not only exist in practice, but also be clearly recorded. The court made clear that settlement should remain a genuine possibility throughout the process. Taken with other flaws, conditioning discussions on full admission of breach before a provisional decision has been reached is an indicator of a closed mind. The same openness is needed in the treatment of evidence. A regulator must be willing to consider material that points away from a finding of breach, including up-to-date compliance evidence.

The judgment also highlights the risks of singling out one party for enforcement where the conduct in question is widespread across the sector. If only one respondent is pursued, there should be a clear and principled explanation for that choice. "We needed a test case" is not, without more, such a basis.

Just as importantly, the decision-making process must be demonstrably independent from the investigation. There should be a record proving that the decision maker(s) tested the investigation findings, considered the regulated party’s representations in full, and reached its own view without the investigation lead or senior sponsor being present when the decision was taken.

Finally, potential and actual conflicts of interest need to be handled consistently and in a forward-looking way. The OfS initially recognised Dr Ahmed's conflict of interest, then reversed that assessment. Although the court found that his role in the Final Decision was not sufficiently central to have contaminated the process with apparent bias, the OfS' inconsistency contributed to the overall picture of a process that lacked the balance and rigour required of a fair regulatory investigation. 

---

For further information or support for your organisation please contact Kris Robbetts or Liz Smillie in our Higher Education team.