Category: Featured

  • Dennis A. Mitchell | The EDU Ledger

    Dennis A. Mitchell | The EDU Ledger

    Dr. Dennis A. MitchellDennis A. Mitchell has been appointed Dean of the Columbia University College of Dental Medicine and Senior Vice President of the Columbia University Irving Medical Center. 

    A renowned clinician, researcher, and administrator,  Mitchell joined the Columbia faculty in 1991 and has since devoted much of his career to strengthening both the faculty and student experience and fostering an inclusive campus climate. He is a professor of Dental Medicine (Community Health and Periodontics) at CUIMC and recently served as Senior Advisor for Inclusion and Belonging. 

    Mitchell earned his DDS from the Howard University College of Dentistry and completed his general dentistry residency training at Harlem Hospital. He obtained his MPH degree in executive health services management at the Columbia Mailman School of Public Health. 
     

    Source link

  • Specialist arts institutions are not a luxury; they are the crucibles of Britain’s creative future

    Specialist arts institutions are not a luxury; they are the crucibles of Britain’s creative future

    This blog was kindly authored by Professor Randall S Whittaker Principal and CEO Rose Bruford College.

    London’s creative industries are not a cultural accessory; they are an economic engine. Around one in seven jobs in the capital sits within the creative industries, and if you include creative roles embedded across other sectors, that figure rises to nearly one in five. Almost a third of all UK creative businesses are based in London.

    The UK’s creative success is no accident. It rests on a delicate, interdependent education ecosystem: specialist arts institutions; research hubs; and universities that together generate not only talent but innovation, identity and national soft power.

    That ecosystem is under pressure. Rising costs, uneven funding, and the new fashion for mergers, the proposed “super university” being the latest example, are driving a wave of consolidation.

    Why “super universities” miss the point

    When two generalist universities merge, their academic portfolios may blend. When a small, practice-led arts institution is absorbed, it rarely blends; it dissolves. Studios become seminar rooms. Ensemble training becomes optional. Niche disciplines disappear in the name of efficiency. Scale rewards the generic; creativity thrives in the specific.

    The Kent–Greenwich merger, planned for 2026, is being hailed as a pragmatic response to sector-wide financial stress. On paper, such consolidations look neat: shared back-office functions, pooled estates, a single regional brand. But higher education is not a spreadsheet exercise.

    It’s understandable that, given Rose Bruford College’s geography — located between Kent and Greenwich — and a financial position that has been challenging but is now improving, some might assume that joining a “super university” is the logical next step.

    Yet that assumption misunderstands what specialist colleges contribute. Rose Bruford’s strength lies precisely in what cannot be merged: its scale, its agility, its ensemble ethos, its craft-specific research culture, and its proven industry connectivity. The College’s recovery — from stabilised finances to a UKRI-funded research project and multiple national awards for both performance and technical excellence — shows that independence is not indulgence; it is impact.

    The question is not whether Bruford can survive outside the merger, but whether the creative industries can afford to lose what institutions like Bruford uniquely provide. When specialist institutions disappear, we do not gain efficiency; we lose an entire mode of creativity.

    There are, of course, examples where partnership has protected identity: the Royal Birmingham Conservatoire operates as an associate faculty of Birmingham City University, retaining its governance and character while sharing infrastructure. That balance, autonomy with alignment, is the exception not the rule. For most specialist creative institutions, a merger could mean absorption, not collaboration.

    From curtain call to crucible

    It remains true that it is a curtain call for the old, exclusionary model of time-intensive training that shuts out those without privilege or flexibility.
    What must be defended now is the right of specialist institutions to re-imagine rigorous training on equitable, sustainable terms.

    Specialist creative higher education is not a conveyor belt. It is a crucible.
    To mistake it for a “skills pipeline” is to misunderstand its purpose. Specialist higher education institutions are not service departments for the creative industries; they are cultural forces — sites of disruption, experimentation and social imagination.

    Graduates from these environments do not merely enter the creative industries; they redefine them. They found new companies, invent formats, challenge hierarchies, and expand who gets to tell Britain’s stories.

    Research, re-imagined

    Specialist arts institutions do not reject research; they redefine it. Practice is their laboratory. Performance, design and experimentation are their methodologies. Rose Bruford’s recently UKRI-funded research project exemplifies how specialist providers drive national innovation, producing knowledge that moves from rehearsal rooms to public discourse, from artistic experiment to policy impact.

    The power of the specific

    The reach of this work is visible every night on screens and stages.

    • Jessica Gunning, BAFTA, Emmy and Golden Globe winner for Baby Reindeer, trained at Rose Bruford.
    • Bernardine Evaristo, Bruford alumna and Booker Prize winner, saw her novel Mr Loverman adapted for television and a Women’s Prize Outstanding Contribution Award, recognising her “transformative impact on literature and her unwavering dedication to uplifting under-represented voices”.
    • Stephen Graham and Hannah Walters, who met as Bruford students, co-starred in Adolescence — proof that specialist institutions forge lifelong creative partnerships.
    • Sir Gary Oldman, Slow Horses, began his journey at Bruford and continues to define British performance worldwide.

    Excellence extends far beyond the spotlight. At the Profile Awards, lighting design alumni Jessica Hung Han Yun, Sarah Readman, and Joshua Pharo, together with Joshie Harriette, all received national recognition. Hung Han Yun — also an Olivier Award winner for My Neighbour Totoro — shows how specialist training produces innovators whose artistry is both technical and conceptual. These achievements prove that excellence in production crafts is not ancillary to the arts; it is integral to Britain’s creative leadership.

    Diversity and student choice

    A healthy higher-education system depends on difference, in mission, in method, in who it serves.

    If independent specialist higher education institutions disappear, the UK’s higher-education landscape flattens. The sector loses, not only training for performers and designers, but the pedagogical diversity that keeps higher education alive, the alternative modes of learning that reach students who may not thrive in traditional university structures.

    For students, the consequences are immediate. Choice collapses from a landscape of craft pathways to a handful of broad “creative-arts” degrees. The student who might have trained as a lighting designer, scenographer or community-theatre facilitator is left with a single, generic option. In a system obsessed with “student choice”, consolidation removes the very choices that matter most — about identity, craft and form.

    GuildHE’s recent Championing a Diverse Higher Education Sector manifesto underscores this point. It highlights the extra costs of small-class teaching and industry-standard facilities that specialist colleges cannot cross-subsidise, and calls for direct funding, reform of research and knowledge-exchange thresholds, and capital investment to secure the sector’s future. These are not indulgences; they are the practical conditions for diversity itself.

    Funding reform is an investment in inclusion

    What specialist institutions seek is not indulgence — and not simply more money to do the same thing. They seek resources that enable transformation: sustainable workloads, flexible modules, hybrid teaching, and equitable access, without sacrificing rigour.

    As GuildHE notes, funding architecture must recognise that small specialist colleges cannot offset studio-based costs in the way comprehensive universities can. Reforming those systems is how government can genuinely champion diversity rather than merely declare it.

    Starving specialist institutions into mergers is not efficiency; it is slow erasure.

    A national imperative

    Britain’s creative industries are a cornerstone of the economy and of international reputation. Yet the institutions that make that possible are treated as optional extras.

    If independent, practice-led institutions vanish, we lose not only talent pipelines but the laboratories of imagination, the incubators of diversity, and the ability to renew what British creativity means.

    Specialist creative institutions are not relics of the past. They are the crucibles of the future — where risk is rehearsed, difference made visible, and new worlds imagined into being. Fold them into super universities, and the loss will not be obvious at first.
    But over time, our screens, our stages and our stories will all start to look the same. And by then, it will be too late.

    Source link

  • Unlocking Learning Potential with Concept Maps – Faculty Focus

    Unlocking Learning Potential with Concept Maps – Faculty Focus

    Source link

  • Unlocking Learning Potential with Concept Maps – Faculty Focus

    Unlocking Learning Potential with Concept Maps – Faculty Focus

    Source link

  • A university system reliant on international students has an obligation to understand them

    A university system reliant on international students has an obligation to understand them

    It is becoming difficult to ignore potential tension between the internationalisation of higher education and plans to cut net migration. Recent UK government policies, such as the reduction of the graduate visa from two years to 18 months, could have severe consequences for universities in Scotland.

    Scottish government funding per home student has not kept pace with inflation. To compensate for the subsequent gap in resources, universities have become more dependent on international enrolments.

    In addition, Scotland faces specific demographic challenges. By 2075, the number of working aged Scots is predicted to fall by 14.7 per cent and, without migration, the population would be in decline. Encouraging young people to remain after graduation could help to balance the ageing population. However, although the Scottish government favours a more generous post-study visa route, this is not supported by Westminster.

    Ability to adjust

    Rhetoric around internationalisation tends to emphasise positive factors such as increased diversity and cross-cultural exchange. Yet, as an English for Academic Purposes (EAP) practitioner, I have long been concerned that learners from diverse linguistic backgrounds are often viewed through a lens of deficiency. There is also a risk that their own needs will be overlooked in the midst of political and economic debate.

    To better understand how students’ sense of identity is affected by moving into new educational and social settings, I carried out interview-based research at a Scottish university. Like other “prestigious” institutions, it attracts a large number of applicants from abroad. In particular, some taught master’s degrees (such as those in the field of language education) are dominated by Chinese nationals. Indeed, when recruiting postgraduate interviewees, I was not surprised when only two (out of 11) came from other countries (Thailand and Japan).

    My analysis of data revealed typical reasons for choosing the university: ranking, reputation and the shorter duration of master’s courses. Participants described being met with unfamiliar expectations on arrival, especially as regards writing essays and contributing to discussion. For some, this challenged their previous identities as competent individuals with advanced English skills. These issues were exacerbated in “all-white” classes, where being in the minority heightened linguistic anxiety and the fear of being judged. They had varied experiences of group work: several reported – not necessarily intentional but nonetheless problematic – segregation of students by nationality, undermining the notion that a multi-national population results in close mixing on campus.

    In a survey administered to a wider cohort of respondents on a pre-sessional EAP programme, the majority agreed or strongly agreed when asked if they would befriend British people while at university.

    However, making such connections is far from straightforward. International students are sometimes criticised for socialising in monocultural groups and failing to fully “fit in”. However, the fatigue of living one’s life in another language and simultaneously coping with academic demands means that getting to know locals is not a priority. At the same time, research participants expressed regret at the lack of opportunity to interact with other nationalities, with one remarking, “if everyone around me is Chinese, why did I choose to study abroad?” Some encountered prejudice or marginalisation, reporting that they felt ignored by “fluent” speakers of English. Understandably, this had a detrimental effect on their ability to adjust.

    Different ways to belong

    To gain different perspectives, I also spoke with teachers who work with international students. EAP tutors believed that their classes offer a safe space for them to gain confidence and become used to a new way of working. However, they wondered whether there would be a similarly supportive atmosphere in mainstream university settings. Subject lecturers did not invoke phrases such as “dumbing down”, but several had altered their teaching methods to better suit learners from non-Anglophone backgrounds.

    In addition, they questioned whether internationalisation always equated to diversity. One commented on the advantages of having a “multicultural quality”, but added that it “has to be a mix” – something which is not possible if, like on her course, there are no Scottish students. Another mentioned that the propensity to “stick with your own people” is not a uniquely Chinese phenomenon, but common behaviour regardless of background.

    A few academics had noticed that most Chinese students take an attitude of, “I’m doing my (one-year) master’s and maybe then I have to move back to China.” Chinese students are less likely than some other nationalities to apply for a graduate visa, suggesting that their investment in a degree abroad is of a transactional nature.

    The majority of survey respondents indicated that they would adapt to a new way of life while living abroad. However, during my last conversation with focal interviewees, I uncovered different levels of belonging, ranging from, “I feel like I’m from Scotland”, to “my heart was always in China”, to “I don’t have any home.” Participants generally viewed their stay as temporary: in fact, all but the Japanese student (who accepted a job in the US) returned to their home country after graduation. Although they described their time in Scotland in mostly positive terms, some were disappointed that it had not provided a truly intercultural experience.

    Meltdown

    It is clear that universities in Scotland have become overly reliant on international tuition for their financial sustainability. At the same time, there is conflict between the devolved administration’s depiction of Scotland as outward looking and welcoming, and the reality of stricter migration policies over which it has no control.

    Discourses which position international students as outsiders who add to high immigration numbers could deter some from coming. If they are seen only as economic assets, their own cultural capital and agency might be neglected. It is also important to problematise the notion of “integration”: even my small study suggests that there are different ways of belonging. No group of learners is homogeneous: even if they come from the same country, individual experiences will differ.

    To navigate the current financial crisis, Scottish universities need to do everything possible to maintain their appeal. With elections being held next year, higher education policy will continue to be a key area of discussion. At present, there are no plans to introduce fees for home students, making revenue from international tuition all the more essential.

    However, at a time of global uncertainty, taking overseas students for granted feels enormously unwise. Instead, it is crucial to ask how they can be made to feel like valued members of the academic community. The answer to this question might be different for everyone, but engaging with students themselves, rather than relying on unhelpful assumptions, would be a start.

    Source link

  • Higher education postcard: London South Bank University

    Higher education postcard: London South Bank University

    On 12 May 1888 the London Evening Standard reported as follows:

    A meeting is to be held at the Mansion House at twelve o’clock, on June 8, to consider the projected South London Polytechnic Institutes. It is stated that Lord Salisbury, Lord Rosebery, and Sir Lyon Playfair have agreed to be present in order to lend their support to the scheme.

    We’ve met Sir Lyon Playfair before – sometime secretary to the Department of Science, he advised on the question of a maritime school for Southampton, which ultimately became Southampton Solent University. Lord Rosebery and Lord Salisbury were both eminent politicians, Salisbury a Tory, Rosebery a Liberal. In 1888 Salisbury was Prime Minister. Rosebery would be Prime Minister soon too – he succeeded Gladstone in 1894, the following year being replaced himself by Salisbury. Clearly the support of these figures was significant. But what was going on?

    Enter Mr Edric Bayley. Bayley was a solicitor living in Southwark: partner at a local practice, he was becoming a man of some substance. In 1892 and 1895 he was elected as a member of the London County Council, representing Southwark West for the Progressive Party. Prior to that, in 1887, he had established a group – the South London Polytechnic Institutes Council or, in some accounts, Committee. This sought to use funds under the control of the Charity Commissioners to create technical and recreative institutes in New Cross, Borough and Battersea. This seems to have been the scheme referred to in the Evening Standard article. In 1888 the Charity Commissioners agreed to match funds up to £150,000 for this scheme. And do the game was most definitely afoot.

    The New Cross institute became Goldsmiths College; the Battersea one became Battersea Polytechnic and in due course the University of Surrey.

    The Borough story goes like this. In 1890, anticipating success, buildings were purchased: these had previously been the base of the British and Foreign School Society; the South London Polytechnic Institutes (Borough Road Site) Act 1890 authorised the purchase. In 1891 sufficient funds had been raised to proceed with the overall scheme, and an act of Parliament passed to create a legal basis for the new institutions. And in 1892 Lord Rosebery opened the polytechnic. His speech was notable for suggesting that by forbidding smoking in the new polytechnic, they would be unable to compete favourably with public houses. And that the structures against dancing and dramatic performances similarly might be too severe.

    It’s worth looking at this extract from his speech – reported in The Globe of Friday 30 September 1892. Not only because it gives a lovely flavour of Rosebery’s speech-making, but also for the slight hint, maybe, of Johnsonian populism.

    The polytechnic was a technical and recreative institute, which means that as well as technical courses, it also had a gymnasium, and offered facilities for clubs and so on. Obviously as long as they weren’t dramatic or involved dancing. The model was the People’s Palace in the east end, which became Queen Mary College. And that’s a story for another time.

    And so the Borough Polytechnic Institute started to do what it did, which was to educate people. Very successfully too, with the National Bakery School, for example, being an early innovation.

    In 1970 the Borough Polytechnic Institute became the Polytechnic of the South Bank, and incorporated a number of other institutions: the Brixton School of Building, the City of Westminster College, and the National College of Heating, Ventilating, Refrigeration and Fan Engineering. In 1975, when education colleges were being brought into existing HEIs, the Battersea College of Education and some of the provision at Rachel MacMillan College of Education joined the polytechnic.

    In 1987 the polytechnic shuffled its name, becoming the South Bank Polytechnic. In 1992 it became South Bank University and in 2003 it became London South Bank University.

    Finally, here’s a jigsaw of the card. It’s unposted, which means I can’t be sure of the date, but it looks to be pre-World War One.

    Source link

  • What’s in the post-16 white paper for postgraduate study?

    What’s in the post-16 white paper for postgraduate study?

    The term “postgraduate” appears exactly 14 times in the Post-16 Education and Skills white paper : a clear improvement over 2021s Skills for Jobs (a mere two mentions in the appendix) and one more than 2016s Success as a Knowledge Economy (the one with the actual postgraduate student loans).

    So it looks like, more so than I’d personally have expected, “post-16” really does include “postgraduate” for once. But what does that mean?

    Fees & maintenance

    The obvious place to start is with the big headline: the planned increase to maximum tuition fees in the next two parliaments and the legislation to enable this to happen automatically thereafter. There’s also the (re)commitment to increase maintenance loans each year (I’ll leave Jim to explain what’s wrong with that).

    None of this really impacts postgraduate Masters and PhD students. Masters loans will also continue to increase with forecast inflation, but fees aren’t regulated and still aren’t properly monitored, despite promises to do so in 2016. PhD fees are largely shaped by the size of UKRI studentships, about which more below.

    I do think it’s interesting to consider what these undergraduate changes will do to perceptions of postgraduate fees. Will the cost of an MA or MSc provoke less sticker shock once BA and BSc fees (very quickly) cross the £10k rubicon? Or will greater undergraduate student loan debt make another £13k or so for a Masters feel less palatable?

    Personally, I think this stuff could end up mattering a lot, particularly if the government wants to improve postgraduate participation. Which apparently it does.

    Postgraduate participation

    We’re told that the government will “for the first time seek to address the barriers faced by disadvantaged students in accessing and succeeding at postgraduate level.”

    It’s fascinating to think about what this actually means.

    On access right now, little is known and less is done. There are some pockets of committed good work, often spotlighted here on Wonkhe and often supported by organisations like UKCGE (who I’m pleased to see will be funded to develop their involvement). But I can still point to data across our platforms demonstrating that postgraduate participation often looks very different to postgraduate interest.

    Postgraduate success, meanwhile, isn’t included in OfS B3 metrics and there’s still no postgraduate TEF. That means that, whilst continuation to a Masters records a good outcome for a university, progression from there isn’t really evaluated. The closest we have is LEO, which, though cited as “one of the best data sources” to drive informed student choices, is a crude and lagged metric taking no account of someone’s background.

    But what’s most intriguing is that all of this appears in relation to Access and Participation Plans.

    APPs determine a university’s ability to charge the higher undergraduate fee level. Postgraduate fees aren’t regulated, which leads to some of the mess around postgraduate funding. What’s here clearly isn’t a proposal to start scrutinising and intervening around PG fees but – like several other parts of the white paper – talking in this way is a potential step towards fundamental change.

    Home-grown PGR

    The white paper actually has a lot more to say about PhDs than it does about Masters degrees. Here’s where we find the most specific references to barriers faced by disadvantaged students and to challenges faced within specific subject areas.

    Here’s also where we find repeated references to a ‘home grown’ pipeline for UK research talent. Again, this is an interesting distinction to make. One of the few major interventions in PhD funding in recent years was the decision to open 30 per cent of UKRI studentships to international applicants from 2021. It hasn’t had a big impact on enrolments but it has meant more students – of all origins – competing for the same broad pot.

    The specific policy is light here (lots of verbs like ‘explore’ and ‘consider’) but prioritising domestic PhDs leads naturally to thinking about interventions around domestic funding.

    Elsewhere there are much clearer and very positive changes to medical and parental leave for UKRI-funded PhD students. This is explicitly framed as bringing conditions in line with employment law and therefore a step towards recognising that PhD students aren’t just students. Of course, this only applies directly to the relatively small proportion of students funded by UKRI.

    Post 16 postgraduate

    This is the first white paper in around ten years with a meaningful amount to say about postgraduate study. It does seem to understand what some of the key problems are and it seems to appreciate that PG is part of a joined-up system.

    There are other questions to ask – there’s little on Masters study and the perverse quirk of the international fee levy robbing PG to pay for UG feels worth scrutinising – but for once the government is asking questions about PG too. That is a good thing.

    Source link

  • A generation of students has been failed and forgotten again

    A generation of students has been failed and forgotten again

    Back in April 2020, with all the lovely weather we’d been having, I was yearning for a holiday.

    But I wasn’t sure the offer I was reading from a tour operator really stacked up.

    “Because of Covid-19”, said the brochure, “by the time your holiday due date arrives we might not be able to get you there and even if you get there yourself, we might not have any of the facilities open. If that’s the case and you book anyway, please note you’ll neither be entitled to a refund nor a discount, but you will get to watch some people swimming in a pool on YouTube. You’ll also be able to join in with our holiday chat on Zoom. And you jolly well better risk booking it – because holidays will be much harder to come by next year!”

    It sounded pretty unreasonable to me. And probably unlawful. But someone somewhere in central government appeared to disagree.

    In the impact assessment that accompanied the emergency Coronavirus Bill a month previously, officials had explained the implications of the power to close campuses – and as well as discussion on how compensation for universities might work, the section contained this ominous line on students:

    Where there are concerns to protect HEPs from being sued for reneging on their consumer protection (and/or contractual) obligations in the event of course closure we believe force majeure would be relevant.

    For me, it was that that set the tone for the entire way in which the pandemic was handled when it came to students and universities. The understanding was that students would pay. And they very much did – in more ways than one.

    As such, ever since March 2020, I’ve been imagining the moment when a minister or senior civil servant might be called by an inquiry to account for the catalogue of catastrophes that were the handling of higher education during Covid.

    I wanted to know if then-universities minister Michelle Donelan really believed that students would succeed if they did as she suggested and made a complaint about quality. I was keen to understand why campuses were being told to operate at 30 per cent capacity while the halls on their edges operated at 100 per cent.

    I wanted to know whether DfE’s plan in the run into September really was to get everyone to turn up, stick it out for a few weeks so that the fee liability kicks in, and then blame external factors when the sector inevitably had to shut campuses back down.

    I wanted to know why students were consistently left out of financial compensation measures, ignored in guidance and policies, told to pay the rent on properties they were told not to occupy, gaslit over quality and “the student experience”, and repeatedly blamed for transmitting the virus when they’d been threatened with losing their place if they deferred.

    Above all, I wanted to know why anyone, at any point, really thought that they should be paying full fees for the isolating omnishambles that they eventually experienced.

    I had reason to be optimistic. The UK COVID-19 Inquiry was set up as an independent public inquiry examining the UK’s preparedness and response to the pandemic through multiple modules.

    Module 8, which opened in May 2024, was to focus specifically on the impact of the pandemic on children and young people across all four UK nations. Universities and students were included in Module 8 because the inquiry defines “young people” to include those aged 18-25 who attended higher education or training during the pandemic.

    It was supposed to examine the full spectrum of educational disruption from early years through to universities, investigating decisions about closures, reopenings, exam cancellations, online learning adaptations, and issues like student isolation in halls of residence, as well as the broader impact on mental health, wellbeing, and long-term consequences for this generation.

    Well guess what. Yesterday was the final day of three weeks of testimony from ministers, civil servants, regulators, charities and other experts. And the total number of times that university students have been mentioned? Zero.

    Hope springs eternal

    If I think back to that March 2020 impact assessment, the logic appeared straightforward – with DfE struggling to extract funding from the Treasury for education, let alone for universities or students, the simplest solution to an assumption that Covid would bankrupt the sector was to ensure fee income continued flowing.

    Universities wouldn’t go bust as long as students could be persuaded not to defer en masse, which would have created a difficult demand spike the following year.

    The problem was how that was done. By the end of March, all sorts of people were calling for a Big Freeze – a pause to September enrolment. The case was straightfoward – universities couldn’t deliver what they were promising, students couldn’t make informed decisions, and the public health risks were unknown.

    But the government’s response was Education Secretary Gavin Williamson announcing in June that there was “absolutely no need for students to defer”, while universities Minister Michelle Donelan told students not to defer because “the labour market will be precarious” and that universities had “perfected their online offer” – despite having only weeks to prepare.

    Donelan repeatedly pushed a “business as usual” message, telling prospective students that September would go ahead with perhaps some online teaching but no real disruption worth worrying about. The Department for Education (DfE) published generic FAQs and social media messages encouraging students to enrol as normal, despite universities having no idea what they could actually deliver. Later in the year, Universities UK joined in with a social media campaign that told students going to university that September would show “strength” and that they “refuse to be beaten”.

    Meanwhile the Office for Students (OfS) imposed and then repeatedly extended a moratorium on unconditional offers – ostensibly to protect students from being rushed into decisions – but without addressing the fundamental problem that universities couldn’t provide the “material information” required by consumer law about what students would actually receive for their money.

    The regulatory system proved entirely inadequate for the crisis. Student Protection Plans, which were supposed to set out risks to continuation of study and mitigation measures, had never been properly enforced and became instantly obsolete. When COVID hit, the regulator made no move to require universities to update their risk assessments, even as financial viability, course delivery, access to facilities and support for disabled students all became dramatically more precarious.

    The Competition and Markets Authority’s consumer protection requirements – which clearly stated students must receive what they expected – were essentially ignored, with DfE telling students that as long as “reasonable efforts” and “adequate” alternatives were provided, they shouldn’t expect refunds. Nobody defined what “reasonable” or “adequate” meant in a pandemic, leaving students with no meaningful protections despite paying £9,250 in fees.

    The absence of any coordinated national approach left universities making contradictory announcements and students facing impossible choices. Cambridge declared all lectures would be online until summer 2021, while Bolton promised a fully “COVID-secure” campus with temperature scanners and face masks. Both still expected students to move house and live in halls of residence, with no clear government guidance on whether mass student migration was safe or legal under social distancing rules.

    There was no extension of maintenance loans despite part-time work vanishing, no student furlough scheme to allow pausing or extending studies, no support for those whose courses couldn’t meaningfully continue, and no plan for accommodation contracts due to start on 1 July.

    Students faced being charged full fees for a drastically reduced experience, with no agency over whether to accept major changes to their courses, no compensation mechanisms, and no honest assessment from universities about what would actually be available – all while being told they should “take time to consider their options” and make “well-informed choices” without any reliable information to base those choices on.

    Summertime scramble

    By June, the absence of any meaningful government planning or support became intolerable. The Office for Students told a student conference that provision needed to be “different but good – not different but bad” but when pressed on what “good” meant, admitted it was a “very good question” and had no answer.

    Its baseline standards were expressed only as outcomes-based metrics that would take 18 months to show up in surveys, leaving universities and students with no clarity about minimum acceptable standards for the pandemic era. Its position appeared to be that it would only intervene if there was a major risk to provider finances, not student welfare – representing “provider interest” rather than “student interest”.

    The government’s approach to student hardship was even worse. After Chancellor Rishi Sunak repeatedly claimed the government was doing “whatever it takes” for people affected by COVID-19, the Department for Education announced on 4 May a £46 million “support package” for students in financial difficulty.

    But this wasn’t new money at all – it was just two months’ worth of existing “student premium” funding that universities were being told they could “repurpose” for hardship funds. The actual sum available was far smaller once you accounted for the fact that much of this funding was already committed or spent on staff costs for widening participation work that hadn’t stopped. In Canada, by contrast, the government announced the equivalent of £736 per month per student until the end of the summer.

    As August arrived with weeks to go before term started, the sector lurched towards chaos. Despite OfS CEO Nicola Dandridge demanding in May that students receive “absolute clarity” about what was on offer before confirmation and clearing, and OfS Chair Michael Barber saying on results day that “universities should set out as clearly as possible what prospective students can expect”, the reality was almost total opacity. Most universities were still offering vague “blended” mood music rather than specifics, with course pages unchanged from before the pandemic.

    On August 30th, the University and College Union called on universities to “scrap plans” to reopen campuses, warning they could become “the care homes of a second wave”. Independent SAGE had published detailed recommendations on 21 August including mandatory testing, online-only teaching except for essential practical work, and COVID-safe workplace charters – but with students already arriving to quarantine, any changes were impossibly late.

    Fresh guidance

    Scotland’s guidance emerged on 1 September, but contained no provision for mandatory testing despite weeks of signals that it would be required, and placed responsibility for enforcement of quarantine and disciplinary measures firmly on universities without clarity on their jurisdiction over private accommodation.

    Then at precisely 1.18am on Thursday 10 September, the Department for Education published guidance for higher education on reopening buildings and campuses in England – and in many ways, it symbolised handling throughout. It insisted on “provider autonomy” and made clear that “it is for an HE provider, as an autonomous institution” to manage risks, while simultaneously dictating detailed operational requirements.

    The responsibility allocation framework seemed systematically designed to deflect blame from government onto universities and students. Despite acknowledging that “mass movement” of students created risk and that Public Health England and local authorities logically should lead outbreak response, the guidance repeatedly assigned universities responsibility for “ensuring students are safe and well looked after” – an impossible mandate for commuter students and adults in private accommodation.

    The testing infrastructure was inadequate and contradictory – walk-through test sites weren’t going to be ready until late October, asymptomatic testing was restricted due to “national strain on testing capacity,” all while universities were warned against running their own testing programs due to complex legal obligations. Contact tracing guidance had obvious holes, particularly compared to Scottish requirements, and did nothing to address the myriad disincentives students faced in getting tested or self-isolating.

    The most damning part was arguably the government’s fatalistic acceptance that outbreaks would occur while simultaneously allowing the “mass migration” to proceed. Both SAGE analysis and the guidance itself acknowledged that increased infections and outbreaks were “highly likely,” yet the entire year could have been “default online.” The government was “doing all it can” while creating the very conditions guaranteed to spread the virus, then pre-allocating blame to students and universities.

    Mental health support amounted to telling autonomous institutions they were “best placed” to handle it themselves, backed by that £256 million that was actually a cut from the previous year’s £277 million – only accessible by making swingeing cuts elsewhere. The guidance exempted halls of residence from gathering limits without explaining why publicly, mentioned nothing about the 200,000 students in private halls, and dumped international student hardship onto universities despite “no recourse to public funds” rules.

    Overall, the government’s handling of the student return to campus in September 2020 was marked by confusion, contradictions and last-minute changes that left universities and students scrambling. Key regulations like the “rule of six” were published at 23:44 on 13 September – just 16 minutes before they came into force – giving institutions no time to communicate changes to students despite DfE saying providers were “responsible” for ensuring awareness.

    The lack of clarity extended to fundamental questions like what constituted a “household” in student accommodation – with profound implications for who could socialise with whom, who had to self-isolate and whether students could return home at weekends – yet no clear legal definition was provided despite fines of up to £10,000 being threatened.

    Different ministers gave contradictory guidance on who was responsible for refunds and financial support – with Nicola Dandridge saying it was a matter for government, Number 10 saying it was a matter for universities and Gavin Williamson suggesting students could complain through the Office for Students.

    By now, the Secretary of State was repeatedly claiming that £256 million was available for hardship funding when this was actually existing student premium funding that had been cut in May, and trumpeted £100 million for digital access that turned out to be a schools fund covering perhaps 650 care leavers worth around £195,000 in total.

    SAGE advice to government warned that a national coordinated outbreak response strategy was urgently needed, that universities should provide dedicated accommodation for isolation and that enhanced testing would be required for clusters. None of this made it into DfE guidance. Instead public health teams were left to improvise, leading to entire buildings of 500 students being locked down when the “household” concept collapsed, while students who actually had Covid were legally allowed to travel home to self-isolate there.

    The quality of data and assumptions going into SAGE about student behaviour was poor – minutes from September meetings showed “low confidence and poor evidence base” yet still concluded students posed risks of transmission, particularly at Christmas when they would “return home” – a framing that completely ignored commuter students and those who travelled home every weekend.

    And we’re off

    As term began, research from Canada and the US suggested reopening campuses substantially increased community infections, yet the government pushed ahead without adequate testing infrastructure in place – with even Dido Harding admitting to Parliament that Test and Trace had failed to predict demand from students despite the obvious likelihood of “Freshers’ Flu” generating testing requests.

    Students were singled out for treatment that was stricter than other citizens yet without equivalent support – expected to self-isolate without access to the financial cushions available to others, unable to claim Universal Credit, threatened with £10,000 fines and faced with university disciplinary action including expulsion.

    Universities Scotland announced students must avoid all socialising outside their households and not go to bars or hospitality venues – a “voluntary lockdown” that was later rowed back as merely a “request” after it caused uproar. Meanwhile Exeter implemented a “soft lockdown” asking only students not to meet indoors with other households, and Scotland threatened to prevent students returning home at weekends – measures that had never appeared in any government guidance or playbook yet were being invented on the fly by local officials.

    The sector was essentially told to operate minimum security prisons while still charging full fees for an experience that bore no resemblance to what had been promised, with no clarity on legal rights to refunds despite the Competition and Markets Authority having provided detailed guidance on partial refunds for other sectors like weddings and nurseries. The fundamental question of what students were supposed to do all week beyond online lectures was never answered – and the predictable result was isolation, mental health crises and outbreaks that vindicated warnings the sector had been making since summer.

    By October, a cascade of evidence had emerged showing fundamental government failures in managing the return of students to universities. The central mistake was allowing student accommodation to operate at 100 per cent capacity whilst carefully reducing every other setting to around 30 per cent, including lecture theatres, libraries, corridors, catering outlets, shops, bars, public transport and social spaces. SAGE had warned as early as September that housing posed major transmission risks due to large household sizes, high density occupancy, poor quality housing and poor ventilation.

    Despite commissioning advice from SAGE on the role of housing in transmission, neither the Department for Education nor the Ministry of Housing, Communities and Local Government acted on recommendations to reduce occupied density. DfE had asked for specific guidance on student accommodation as early as 8 July, yet the resulting advice failed to implement SAGE’s core recommendation that “we really ought to have been reducing occupied density”.

    Ministers instead expected students to spend carefully distanced time on campus for three or four hours weekly, but for the rest of the week to remain in spaces “never designed to be used this intensively, with inevitable results”.

    The government rejected crucial scientific advice at multiple points. On 21 September, SAGE recommended that “all university and college teaching to be online unless face-to-face teaching is absolutely essential” as part of a package to reverse exponential rises in cases, estimating this would reduce the R number by 0.3. Ministers explicitly rejected this recommendation, with universities minister Michelle Donelan maintaining the line that “we decided to prioritise education”, despite students being confined to online learning in tiny rooms anyway.

    The government also failed to implement mass testing strategies recommended by both SAGE and the CDC, despite promising repeatedly to deliver testing capacity. Students requiring self-isolation received no financial support, being excluded from the £500 Test and Trace Support Payment available to other citizens on low incomes, with most local authorities refusing to include students in discretionary schemes.

    SAGE had explicitly warned that “students in quarantine require substantial support from their institution during the period” and that “failing to provide support will lead to distress, poor adherence and loss of trust”, yet the government provided no funding beyond expecting universities to redirect existing hardship funds.

    Jingle bells

    As the country looked to Christmas, no UK-wide coordination existed for managing student migration, with four-nations discussions chaired by Michael Gove only beginning in late October. Guidance on safely managing returns remained unpublished throughout October despite ministers knowing since March that student migration posed transmission risks.

    Students fell through gaps between departments, with DfE, MHCLG and DWP unable to agree responsibility for financial support, testing strategies or housing standards. Working students in hospitality and retail lost employment income but found themselves excluded from benefits, discretionary payments and adequate hardship fund support, with some resorting to commercial credit to pay rent arrears.

    As Christmas got closer, the government’s handling of higher education descended into a familiar pattern of late guidance, ignored scientific advice and policy decisions designed to avoid financial liability. DfE repeatedly issued vague guidance that left universities to make difficult decisions about face-to-face teaching – explicitly to avoid being drawn into fee refund claims – while ministers like Michelle Donelan insisted that “blended learning” must continue despite SAGE’s repeated recommendations to move teaching online.

    When guidance on Christmas travel and January’s return finally emerged, it came so late that universities couldn’t meaningfully prepare, and students faced impossible choices between physical and mental health. Office for National Statistics data revealed that 65 per cent of students in mid-November were receiving zero hours of face-to-face teaching – despite government rhetoric about prioritising in-person education – while student anxiety levels (6.5 out of 10) dramatically exceeded the general population (4.3 out of 10).

    The government’s failure to address accommodation density became undeniable when Public Health Scotland published evidence that traditional halls of residence with “households” of up to 30 people had fuelled transmission, with almost 3,000 cases associated with student accommodation and the majority occurring in a three-week period in September-October.

    Students described having to remain in small rooms without access to food, fresh air or exercise during self-isolation, with some international students arriving to empty buildings with no support, no SIM card to order food and no meaningful mental health provision despite contractual promises.

    The government’s solution for Christmas was to create complex household bubble rules that meant some families with multiple students away from home couldn’t all gather together, while hundreds of international students and care leavers faced spending the entire break isolated in university towns with no meaningful support beyond opened canteens where they still couldn’t mix with others.

    Auld lang syne

    As the new year approached, the government announced a “staggered return” that would see some students barred from returning to paid-for accommodation for up to nine weeks, with no rent rebates offered – and Minister Donelan explicitly telling universities that “any issue about accommodation is for the universities to address with students”. The mass testing programme was revealed to have serious efficacy problems – with one study suggesting the lateral flow tests detected just over 3 per cent of cases – and participation rates of around 7 per cent in Scotland.

    Despite mounting evidence about the failure to deliver “blended learning”, worsening mental health and the concentration of infection in accommodation rather than teaching spaces, ministers continued to frame policy around getting students “back to campus” while simultaneously telling them to stay away, creating a chaotic situation where students couldn’t plan, universities couldn’t prepare and the January term faced collapse before it began.

    By January 2021, the government had announced that some students could return to campus, but the policy was riddled with problems from the start. Michelle Donelan’s attempt to specify eligible subjects using HeCOS vocabulary was marred by missing code numbers and duplications, and ignored both professional bodies’ actual requirements and basic public health principles. The Department for Education essentially determined which campuses would see large student numbers purely by virtue of subject mix – meaning some universities like King’s College London expected over 6,500 students while others like LSE had virtually none.

    Worse still, the policy showed no consideration of local Covid case rates, with the data showing numerous providers expecting large numbers of “tranche one” students in areas with high infection prevalence. No professional body was actually insisting students be on campus during a “terrifying third phase of a pandemic” – yet the government was.

    The legal restrictions that came into force on 5th January created even more chaos. Nothing in the regulations actually changed specifically for higher education, yet guidance from DfE contradicted existing legal exemptions – for example, insisting that campus catering had to be takeaway only when the law explicitly allowed “cafes or canteens at a higher education provider” to open.

    The movement rules allowed students studying away from home to move back to their student housing before 8th February, yet DfE guidance urged them to stay away – creating a situation where students were being told they couldn’t use properties they were legally entitled to occupy and were being forced to pay rent on.

    Meanwhile, emerging research painted a terrifying picture that government policy largely ignored. Cambridge’s genomics work suggested little transmission between students and the wider community in their specific context, but a separate study of 30 universities found that 17 campus outbreaks translated directly into peaks of infection in their home counties within two weeks.

    The government’s lateral flow testing programme was riddled with problems. Data from Scotland’s pre-Christmas exercise showed that 28.5 per cent of positive lateral flow tests were false positives when PCR-confirmed – yet the government then removed the requirement for PCR confirmation in England in February, potentially causing thousands to self-isolate unnecessarily while eliminating the data needed to assess test accuracy.

    Research showed the type of test mattered enormously and that high participation rates were key to any testing programme’s success, yet participation in university testing remained abysmal – in one week in February just 100,000 tests were conducted across English higher education, suggesting only around 2.5 per cent of students were being tested despite government insistence on twice-weekly screening. The cost per positive result in community testing was around £20,000 when prevalence was low, and the government appeared to have learned nothing from the pre-Christmas pilot about why students weren’t engaging.

    On accommodation, ministers merely “urged” landlords to offer rent rebates while keeping students away from campuses, refusing to underwrite rebates or mandate them. And the government’s handling of vaccination access created additional problems – students studying away from home could only access vaccines where their GP was registered, but the NHS insisted students could only be registered with one GP at a time. It meant that students isolating at home due to health conditions were offered vaccines at their university address hundreds of miles away, creating impossible choices.

    Summertime sadness

    As the term continued, universities minister Michelle Donelan repeatedly told parliament that the Office for Students was “actively monitoring” the quality of provision and that students dissatisfied with teaching could complain to the OIA for potential refunds. But OfS confirmed it wasn’t monitoring quality in any meaningful sense – it had merely made some calls to universities in tier 3 areas – and the OIA re-clarified that it couldn’t adjudicate on academic quality matters. The disconnect between ministerial assurances and regulatory reality left students without the redress mechanisms they had been promised.

    Similarly, the government’s approach to student return dates proved chaotic – initially planning for staggered returns from mid-February, then pushing back to March 8th for only “practical” courses, then suggesting a review “by the end of Easter holidays” for everyone else. By May, the government confirmed remaining students could return from Step 3 on May 17th – by which point teaching had largely finished for most.

    The maintenance loan system created further problems – initially, officials planned to reassess loans downward for students no longer in their term-time accommodation, despite many still paying rent. Only after intervention was this policy reversed, with “overpayments” instead added to loan balances. Meanwhile, students on courses requiring practical work faced significant challenges – many would be “unable to complete” without access to facilities yet received no extensions to maintenance support or additional hardship funding.

    The graduate employment situation was similarly neglected, with unemployment among recent graduates rising to 18 per cent for both graduates and non-graduates, reaching 33.7 per cent for young Black graduates. The government’s response was merely to create a six-step collection of weblinks hosted on the OfS website rather than meaningful job guarantee schemes or funded postgraduate study.

    A failure to learn

    There are so many other aspects I haven’t covered here, and so many more issues I’d have loved to see ministers account for. The international student experience deserves its own inquiry – from those trapped abroad unable to access online teaching due to time zones and VPNs, to those who arrived to empty campuses having paid flights and deposits they couldn’t recover.

    The impact on disabled students who lost access to essential support services. The postgraduate researchers who saw years of lab work destroyed or delayed. The creative arts students paying full fees for courses they literally couldn’t do from their bedrooms. The teaching quality collapse that was dismissed as “blended learning”. The mental health crisis that was met with a cut to support funding dressed up as additional help. The list is almost endless.

    But even if all these issues had been put to ministers at the inquiry, even if we’d heard testimony about employment, housing, transport, testing, health, isolation support, and student finances, we’d likely have faced the same problem we always do with students and higher education. Just as we see outside of pandemics, other government departments simply don’t think about students.

    They’re not on the radar at the Treasury, at MHCLG, at DWP, at DHSC. Students fall through the gaps between departmental responsibilities, and the Department for Education – already stretched covering schools, further education, and early years – doesn’t have the influence, the power, or the capacity to force the student interest to be considered in decisions made elsewhere.

    When Test and Trace payments were designed, nobody thought about students. When furlough was created, nobody thought about students. When housing regulations were written, nobody thought about students. And DfE either didn’t notice – or couldn’t do anything about it.

    The problem is that if the decisions aren’t examined, the lessons won’t be learned. And there are crucial lessons that go far beyond the specifics of this pandemic.

    The first is about responsibility and coordination. Throughout the entire saga, nobody seemed to be in charge. Was it DfE setting the policy? Was it universities exercising their autonomy? Was it local public health teams managing outbreaks? Was it MHCLG regulating housing? The answer appeared to be all of them and none of them, with each able to blame the others when things went wrong.

    The tension between treating universities as autonomous institutions who should be “free” to make decisions, and then dictating detailed operational requirements while disclaiming responsibility for the outcomes, was never resolved. The sector demanded provider autonomy and then was sore when society blamed providers when they made autonomous decisions that some didn’t like.

    Second, there’s the fundamental question of whether universities are part of the education system or something else entirely. They were conspicuously left out of education prioritisation decisions, treated neither as essential education that should continue (like schools) nor as adults who could make their own choices (like the general population).

    Instead they occupied a weird middle ground where they were expected to operate like big schools – following government guidance, prioritising in-person education, being “responsible” for students – but with none of the government support or coordination that schools received. Students were simultaneously infantilised (being told they couldn’t be trusted to socialise responsibly) and abandoned (being told they were autonomous adults who should sort out their own problems).

    Third, the pattern of guidance arriving impossibly late – or not at all – for me was a systematic failure to (scenario) plan. Regulations published 16 minutes before coming into force. Christmas travel guidance arriving when students had already booked transport. January return policies announced in December. Testing strategies finalised after students arrived. It wasn’t just poor planning – it was policy-making that appeared designed to avoid being pinned down, to maintain government flexibility at the expense of everyone else’s ability to prepare. And crucially, it was lateness that consistently disadvantaged students while protecting government and institutions from liability.

    Fourth, the entire approach suggested to me an implicit hierarchy where protecting university finances trumped protecting students. That impact assessment in March 2020 said the quiet part out loud – force majeure would be relevant because universities had to be protected from being sued for breaking their consumer obligations.

    Everything else flowed from that – encouraging enrolment even when delivery was uncertain, maintaining full fees regardless of quality, shifting hardship costs onto universities through repurposed funding, avoiding clear guidance that might create refund obligations, and telling students to complain to regulators who couldn’t actually help them. The system was designed to keep money flowing, not to deliver value or protect students.

    And fifth, regulatory failure was baked in from the start. In England, the Office for Students proved entirely unfit for purpose in a crisis – unable to define quality standards, unwilling to intervene on student welfare, focused on provider finances rather than student interests, and apparently content to let ministers mislead Parliament about the protections it was providing. The gap between what ministers promised (monitoring, complaints routes, refunds) and what regulators could actually deliver left students without redress at the precise moment they needed it most.

    None of these lessons are being learned because none of these decisions are being examined. And perhaps that absence reflects something deeper. Everyone involved – universities, the Office for Students, the Department for Education, ministers – were invested in suggesting that what was planned would work, was working, and did work. Partly that was about reputation. Partly it was about liability. Partly it was about avoiding the cost of refunds.

    But the effect was the same – it prevented and continues to prevent any honest reckoning about what didn’t work, who was failed, who was damaged, and who was let down. That, on Radio 4’s More or Less a few weeks ago, I couldn’t answer the question how much learning had been lost, or what the impacts were on students, should be a source of deep national shame.

    Admitting the scale of the failure would’ve meant admitting the scale of what was owed. But instead we got a collective fiction that “blended learning” was comparable to what students had signed up for, that online education from bedroom prisons was a reasonable substitute for the university experience, that students complaining about quality were just being difficult. The investment in that fiction has been so total that even now, when the statutory inquiry is supposed to be learning lessons, it can’t examine what actually happened – because the evidence on impacts is so thin.

    Again and again

    It’s worth saying that everyone almost certainly did their best in impossible circumstances. From university staff pivoting to online teaching overnight, to SU officers running Zoom pub quizzes to combat isolation, to professional services teams working around the clock to support students in crisis – there were countless unsung heroes who made it all a little less miserable. Nobody wanted this. Nobody planned for it. Everyone was doing what they could with the resources and guidance available. And that deserves recognition.

    But the truth is that once students are paying individually for higher education – once it became a consumer market with £9,250 fees and interest-bearing loans and the language of student choice and value for money – doing your best isn’t enough. If you can’t deliver what was promised, “we tried our best” doesn’t cut it when someone is personally £50,000 in debt for the privilege.

    The hyper-marketisation of higher education came with consumer protections and Competition and Markets Authority guidelines and material information requirements precisely because students weren’t just participating in education anymore – they were purchasing it. Neither the sector nor the government can have it both ways. You can’t charge consumer prices and then claim education is special and different when it comes to consumer rights.

    You can make a reasonable case in other policy areas about whether central government or local authorities or devolved administrations or individual departments should be in charge of this budget or that decision. There are legitimate debates about the boundaries of autonomy and accountability, about who’s best placed to make decisions about housing or testing or public health interventions.

    But what I can’t get over – and what I don’t think we should let anyone forget – is the assumption that was there from the very beginning, baked into that March 2020 impact assessment and every decision that followed – that students would pay for pretty much everything that was done to them, and everything that wasn’t done for them.

    They’d pay full fees for reduced teaching. They’d pay rent on accommodation they were told not to use. They’d pay for support services that were closed. They’d pay for facilities they couldn’t access. They’d pay for an experience that bore no resemblance to what they’d signed up for. They’d pay for the privilege of being locked in their rooms, blamed for spreading a virus, threatened with fines and expulsion, excluded from the financial support available to everyone else, and told to be grateful that university was happening at all.

    And pay they did. While being gaslit about quality, lied to about protections, and blamed for their own misery.

    As Kate Ansty of the Child Poverty Action Group said in her oral evidence, to protect inside a pandemic, we must protect outside a pandemic. And as Sir John Cole said in his oral evidence, young people have played their part – society owes them a debt. Not the other way around.

    If another pandemic happens, or another national emergency that affects higher education, I fear that exactly the same mistakes will be made. We still won’t know who’s in charge. Government will still treat universities as simultaneously autonomous and controlled. It will still issue guidance too late for anyone to use it. It will still prioritise institutional finances over student welfare. It will still leave students excluded from support available to everyone else. We’ll still have regulators who can’t regulate.

    And students will still be blamed for the consequences of decisions made for them, about them, and without them.

    Source link

  • Michigan State University lays off 99 employees

    Michigan State University lays off 99 employees

    This audio is auto-generated. Please let us know if you have feedback.

    Dive Brief:

    • Michigan State University is laying off 99 faculty, staff and executives this month amid rising costs and other budget pressures, President Kevin Guskiewicz said in a public message Wednesday
    • The university had cut an additional 83 employees since March because of the Trump administration’s revocation of federal research funding. Taken together, the job cuts represent 1.3% of Michigan State’s workforce.
    • Officials expect the job eliminations to save the university $50 million annually. “Today we expect the overall general fund budget to be largely on target,” Guskiewicz said.

    Dive Insight:

    In explaining the layoffs, Guskiewicz pointed to rising costs, including significant increases in employee healthcare costs. The institution’s operating expenses generally have risen as well, a sectorwide trend playing out across the country.

    Between the fiscal years 2023 and 2024, Michigan State’s operating expenses rose 10% to $3.2 billion, according to its latest financial report. They rose 17.9% between fiscal years 2020 and 2024.

    Revenue hasn’t kept pace, and the university’s operating loss widened by nearly a quarter to $840 million in fiscal 2024. However, with state funding and other outside revenue sources factored in, the university’s total net position more than tripled during the same period.

    Along with inflation, federal funding disruptions have weighed on Michigan State’s budget. 

    By Oct. 1, the Trump administration had terminated 74 federally funded projects at the university, totaling $104 million in multiyear grants and contracts, according to Guskiewicz. Those include grants from the National Science Foundation, the U.S. Agency for International Development, the National Endowment for the Humanities, the National Institutes of Health and the U.S. Department of Agriculture.

    Another 86 research projects, at minimum, have been hit with stop-work orders, pauses on future funding or conditional terminations, Guskiewicz said. 

    Enrollment has been a bright spot for the university, with its fall headcount hitting 51,838 students, according to institutional data. 

    This total is close to our predictions and will keep us on the budget path we have laid out,” Guskiewicz said. Michigan State’s undergraduate class this semester hit a record 41,415 students.

    However, he noted recent declines in international enrollment have weighed on tuition revenue. International students made up 8.2% of Michigan State’s student body in fall 2025, down from 8.5% last year. Their share of the university’s enrollment has almost halved since 2015, when they made up 15% of its students.

    Michigan State’s colleges and administrative units have been working since the summer to cut their budgets by 9%. 

    I am proud of units achieving as many savings as possible through non-personnel actions and evaluating vacancies before filling open roles,” Guskiewicz said. “Nearly two-thirds of the reductions, in fact, were proposed across supplies, services and other non-personnel expenses.”

    But, he added, Michigan State wasn’t entirely able to avoid layoffs, hence the round of workforce cuts announced this week. 

    These colleagues are valuable parts of our community, and their loss, for any reason, is still felt by colleges and programs,” Guskiewicz said.  

    The layoff numbers don’t include those whose employment classification changed, or faculty whose contracts were not renewed. However, Guskiewicz said it was difficult to quantify how many of those contract nonrenewals were related to the budget cuts or other factors such as enrollment levels or course demand.

    Source link

  • Feds launch site for employers to pay controversial H-1B fee, clarify exemptions

    Feds launch site for employers to pay controversial H-1B fee, clarify exemptions

    This audio is auto-generated. Please let us know if you have feedback.

    Dive Brief:

    • The U.S. Treasury Department launched an online payment website for employers to pay President Donald Trump’s $100,000 fee on new H-1B visa petitions, according to an update Monday from the U.S. Citizenship and Immigration Services.
    • USCIS said the fee applies to new H-1B petitions filed on or after Sept. 21 on behalf of beneficiaries who are outside the U.S. and do not have a valid H-1B visa, or whose petitions request consular notification, port of entry notification or pre-flight inspection. Payment must be made prior to filing a petition with USCIS, per the agency.
    • Separately, USCIS’ update clarified that the fee requirement does not apply to petitions requesting an amendment, change of status or extension of stay for noncitizens who are inside the U.S., if that request is granted by USCIS. If it is not granted, then the fee applies.

    Dive Insight:

    Trump’s proclamation announcing the H-1B fee left employers with plenty of unanswered questions. While Monday’s update provides some clarity, the policy’s future is still uncertain in part because business groups, employers, unions, lawmakers and other stakeholders oppose it.

    At least two lawsuits have been filed seeking to enjoin the fee proclamation — one by the U.S. Chamber of Commerce in Washington, D.C., and another by a group of plaintiffs in California. Both similarly alleged that the H-1B fee violates the constitutional separation of powers as well as the Administrative Procedure Act. The complaints also warned of negative effects on U.S. employers that depend on the H-1B program to attract skilled foreign workers.

    In a letter to Trump and Secretary of Commerce Howard Lutnick, a bipartisan group of congressional lawmakers agreed to the need for reform of the H-1B program while expressing concerns about the potential effects of the fee on U.S. employers’ ability to compete with their global counterparts for talent.

    “The recently announced H-1B visa changes will undermine the efforts of the very catalysts of our innovation economy — startups and small technology firms — that cannot absorb costs at the same level as larger firms,” the lawmakers wrote.

    Trump and the White House have said the fee is necessary to combat “systemic abuse” of the H-1B program by employers that seek to artificially suppress wages at the cost of reduced job opportunities for U.S. citizens. In addition to the fee imposed on new visa petitions, the administration issued a proposed rule to change its selection process for H-1B visas to be weighted in favor of higher-paying offers.

    USCIS’ guidance noted that the Secretary of Homeland Security may grant other exceptions to the H-1B fee in “extraordinarily rare” circumstances where:

    • A beneficiary’s presence is in the national interest.
    • No American worker is available to fill the role.
    • The beneficiary does not pose a threat to U.S. security or welfare.
    • Requiring payment from the employer would significantly undermine U.S. interests.

    The agency provided an email address to which employers could send requests for fee exemption along with supporting evidence.

    Employers planning to file for new H-1B visas should plan to pay the fee unless litigation results in some kind of change, Akshat Divatia, attorney at law firm Harris Sliwoski, wrote in an article Tuesday. Divatia noted that some of the criteria for exemptions outlined by USCIS may conflict with congressional design of the H-1B program, and that employers “should watch closely how the courts respond” to such arguments.

    Source link