The TEF is OfS’ Brexit moment

There's a version of the Teaching Excellence Framework's (TEF) compatibility with the European Standards and Guidelines for Quality Assurance (ESG) story that writes itself as a Brussels bureaucracy piece.

Jim Dickinson

Jim is an Associate Editor (SUs) at Wonkhe

How we got here

The cause is partly the 2023 de-designation of QAA as England’s Designated Quality Body. OfS took the relevant functions in-house. QAA retains its ENQA membership and EQAR registration – which is why the UK national action plan refers to “OfS and QAA” in the same breath on quality – but QAA no longer has the statutory English role. OfS has the statutory role, and OfS isn’t EQAR-registered.

The difference between Scotland and England isn’t a difference of quality assurance philosophy. It’s a difference of which body is doing it. QAA continues to operate as the responsible body in Scotland, Wales and Northern Ireland, with its EQAR registration intact. England decided to bring things in-house instead, and that decision created the gap.

The government’s own 2023 consultation response acknowledged concerns about ESG compliance and international reputation, treated future quality assessment arrangements as outside the scope of that consultation, and moved on.

The House of Lords Industry and Regulators Committee later identified specific ESG issues – publication of assessment reports, student involvement in assessment teams, and whether OfS’s risk-based approach could satisfy EHEA expectations of cyclical review. The Quality Council has been even more direct, characterising England as “partially compliant” with Bologna quality commitments and identifying four operative gaps: independent cyclical review, public quality reports covering all providers, student involvement on review panels, and oversight by an EQAR-registered agency.

OfS’s September 2025 consultation document said “at this stage we consider that our proposed overall approach broadly includes all the elements required by these standards and guidelines.” The stage one outcomes document, published this spring, walks that back to “continue to explore how the system adheres to European standards and guidelines for quality assurance, and work towards applying to join the European Quality Assurance Register for Higher Education at the appropriate time.”

So broadly includes all the elements required has become explore how the system adheres. It’s worth noting that the Bologna Implementation Report had already recorded England’s non-compliance before that original, bolder sentence was written.

The UK’s national action plan – published in May 2025, covering 2024 to 2027 – committed to “moving towards compliance with ESG” and a “plan for compliance with ESG 2027.” The 2027 EHEA ministerial conference, where ministers will assess national action plan progress and adopt the revised ESG, is eighteen months away.

Part 1: what the TEF can’t see

The ESG has three parts. Part 1 covers what institutions should have in place internally. Part 2 covers what external quality assurance agencies should do. Part 3 covers the standards the agencies themselves must meet. The TEF has problems in all three.

ESG 2.1 requires external quality assurance to address the effectiveness of all the Part 1 internal standards. The TEF’s alignment with OfS’ B conditions means it covers some of them. But the TEF operates at provider level, and several Part 1 standards are inherently about programme design, programme review, and the student experience within specific courses – things that institutional averages can’t capture.

ESG 1.2: programme design and review

ESG 1.2 requires that institutions have processes for programme design, approval, monitoring and review that involve students and graduates as “equal partners,” keep programmes informed by the latest research, and – this is new in the 2027 draft – design programmes that support “active participation of students in democratic societies.” The TEF’s student experience indicators can’t tell you whether students are equal partners in programme review. They can tell you whether final-year students were satisfied with their course. Those aren’t the same thing.

ESG 1.3: pedagogical methods and self-evaluation

ESG 1.3 requires quality assurance processes to verify that institutions use a variety of pedagogical methods, evaluate and adjust modes of delivery based on evidence and feedback, and foster student autonomy. The NSS captures student satisfaction with teaching. It doesn’t verify whether the methods being used are pedagogically sound or whether the institution is genuinely evaluating its own approaches. OfS’s desk-based model can’t do this. A visit can.

ESG 1.5: teaching staff

ESG 1.5 requires institutions to assure themselves of the competence of teaching staff through fair and transparent recruitment, development and evaluation processes. The 2027 draft adds that student feedback should feed into teacher evaluation. There’s nothing in the TEF that assesses whether institutions have effective staff development frameworks, and no mechanism for checking whether student feedback is actually used to develop teaching practice rather than just collected and filed.

ESG 1.6: learning environment and student consultation

ESG 1.6 requires institutions to take into account the needs of a diverse student population and consult students in the planning and evaluation of resources. The TEF’s B2 alignment gets some way here, but “consulting students” in ESG terms means demonstrable ongoing involvement in resource planning, not a satisfaction survey score.

ESG 1.8: programme-level public information

ESG 1.8 requires clear, accurate, up-to-date information about programmes – their learning outcomes, assessment procedures, selection criteria, pass rates, graduate employment information – accessible to different target groups. The TEF publishes provider-level ratings and benchmarked data. It doesn’t assess whether individual programme information meets ESG standards. OfS’s own applicant research consistently finds students want course-level information. The system produces institution-level averages.

Part 2: what the external assessment fails to do

ESG 2.3: self-assessment and site visits

ESG 2.3 says external quality assurance should include a self-assessment or equivalent, followed by an external assessment normally including an in-person site visit. The TEF does neither in the ESG sense. Provider submissions are not structured self-assessments – institutions don’t evaluate their own internal quality systems against the standards. And the TEF is desk-based as a matter of design. OfS has decided that sample visits and targeted investigations can happen where concerns arise, but these are exceptional rather than normal. ESG says the opposite: visits are normal, with exceptions for cases where strong internal quality assurance has been demonstrated.

Draft ESG 2027 tightens this further: “an external assessment normally including an in-person site visit by peer review experts, complemented with stakeholder interviews, ensuring input from various perspectives.” A desk-based system that relies on NSS data and provider submissions assesses quality at a distance.

You learn that something went wrong after the fact, when the indicators have moved. A visit lets assessors see what the library looks like at nine in the evening, ask students questions that weren’t pre-approved by the institution, and talk to professional services staff whose experience never appears in a dataset.

ESG 2.4: peer review experts

ESG 2.4 requires peer review experts that include student members and bring perspectives from institutions, academics, students and employers. OfS does have student assessors on panels, which satisfies the narrow requirement for student membership. But two things – first, employers and professional practitioners are not a systematic part of the TEF assessor pool, which the standard expects. Second, and more importantly, student assessors sitting on panels is not the same as students being meaningfully involved in the design and governance of the methodology itself – which is what ESG 2.2 and 3.1 require, and which a public consultation exercise doesn’t satisfy.

There’s also the specific problem that student submissions – the main vehicle for student evidence – are restricted to the experience aspect only, with student voice structurally excluded from outcomes assessment. Even with student assessors on the panel, the evidence base they’re working from has a systematic gap built into it.

ESG 2.5 and 2.7: outcomes criteria and appeals

ESG 2.5 requires outcomes to be based on explicit published criteria applied consistently. The TEF’s rating criteria will be published, which satisfies the first part. But there’s a consistency problem: representations are restricted to Bronze and Requires Improvement providers only, on factual accuracy grounds only.

ESG 2.7 is explicit that appeals should allow institutions to demonstrate that “the outcome is not based on sound evidence, or that criteria have not been correctly applied” – and that appeals should be considered by a different entity from the one whose decision is appealed against. Neither of those conditions is met.

ESG 2.6: reporting

ESG 2.6 requires full reports to be published covering context, evidence and analysis, findings, conclusions, good practice and recommendations – and that all evidence used in decision-making be published alongside the report. OfS has confirmed that reports in the new scheme will be “more succinct” than TEF 2023. That sits awkwardly against a standard that requires substantive, structured reporting including the full evidence base.

Part 3: the agency itself

ESG 3.2: independence of formal outcomes

ESG 3.2 requires agencies to be independent in three specific ways: organisational independence, operational independence, and independence of formal outcomes. The third is where the TEF has a direct problem. OfS has still not resolved whether its own staff will serve as assessors for rating decisions – particularly on whether providers meet minimum student outcomes requirements. If OfS staff make the rating, OfS staff defend it through representations, and OfS staff make the enforcement decision, that is not independence of formal outcomes. The stage one outcomes document says OfS “wants to further develop our thinking” on this. Stage two will settle a question that is a direct EQAR compliance test.

ESG 3.1 and 3.3: stakeholder involvement and enhancement

ESG 3.1 requires agencies to ensure “meaningful involvement” of stakeholders including students in their governance and work. ESG 3.3 requires agencies to regularly engage in activities that support quality enhancement and publish findings on trends, good practice and persistent difficulties. OfS has committed to thematic reporting, which addresses 3.3 in principle.

But meaningful student involvement in agency governance is a higher bar than having students respond to a public consultation or serve as assessors in individual reviews. There is no standing mechanism by which students are involved in the ongoing design, review, and governance of the TEF methodology. A consultation every few years, with response analysis showing that student concerns didn’t make it into the formal thematic coding, is not what 3.1 has in mind.

None of this is new, by the way. The ESG 2015 already required site visits, independent peer review, and meaningful stakeholder involvement. ESG 2027 sharpens the language – “in-person,” “equal partners,” “complemented with stakeholder interviews,” “meaningful” – but doesn’t introduce obligations that didn’t previously exist. England’s compliance gaps are failures against standards in force since 2005. The revision makes them marginally harder to argue around, but they were always the standards.

What’s at stake

The most immediate practical risk isn’t EHEA sanctions. It’s partner preference – and in one case it’s harder than that.

The European degree label, which the EU adopted by Council Recommendation in 2025, explicitly requires quality assurance through EQAR-registered agencies or the European Approach for Quality Assurance of Joint Programmes. English providers can’t simply point to their TEF rating as evidence of ESG-compliant quality assurance for European degree label purposes, because OfS isn’t EQAR-registered. To participate in a label-bearing consortium, they’d need QAA involvement or a partner institution’s EQAR-registered agency to cover that function. That’s a concrete structural limitation, not a reputational risk.

More broadly, the DEQAR database – which gives partners visibility of which providers have been externally quality assured and where the reports are – will show a gap for England that doesn’t exist for Scotland, Wales, or Northern Ireland. Erasmus Mundus partners, strategic European alliance institutions, and anyone building a programme that needs ESG-legible credentials across multiple jurisdictions will notice.

For ordinary semester Erasmus+ exchange the risk is upstream rather than downstream – partner institutions asking more questions before signing agreements, or preferring providers with cleaner visibility – rather than students discovering their credits are invalid. The ECHE application process for 2027, whose deadline passed in March 2026, is separate from national ESG compliance, and individual providers’ Erasmus+ standing doesn’t turn on a switch that says “England is fully compliant.” But for the more valuable and strategically significant European partnerships, the gap matters considerably more than for a semester exchange.

There’s also a political incoherence that’s increasingly hard to ignore. The government is splashing on Erasmus+ association. Ministers are making speeches about closer UK-Europe relationships in higher education. The same DfE that recorded the ESG compliance problem in 2023 is now championing European partnership ambitions while the underlying infrastructure remains unresolved. There is a difference between political signalling and the technical architecture needed for higher education cooperation to actually work.

At the appropriate time

The second stage TEF consultation is due in autumn 2026, with first assessments in 2027–28. The decisions that will determine whether England is ESG-compliant are being made in that process – right now – without a clear ministerial instruction that ESG compliance is the required destination, and with a cost-reduction imperative pointing in the other direction.

The site visit question, the self-assessment question, the assessor independence question, the appeals question, the programme-level coverage question, and the meaningful student involvement in governance question aren’t going to resolve themselves through goodwill and aspiration. They require design decisions with cost implications that OfS has so far been given every incentive to avoid.

The ESG’s authors didn’t invent those requirements to make life difficult for English regulators. They put them there because quality assurance done at a distance, by the agency doing the enforcing, without structured self-assessment, without visits, with an appeals process that only covers some ratings on narrow grounds, and with student involvement that stops at the assessment room door – is not quality assurance you can trust.

It turns out the eurocrats had a point.

Wonk Corner | 11/06/26