Student number controls are back – with deadly penalties for over-recruitment

Controls on student numbers are returning for what is billed as a one year, one-off reappearance. David Kernohan finds the devils in the detail.

Sneaking out late on a Monday evening comes details of the first attempt to cap the number of students getting places at universities since 2012.

The policy is broadly following the lines set out by Universities UK’s proposed “gentlemen’s agreement” cap – last year’s recruitment of English domiciled undergraduate full-time students, plus the forecasted addition for this year, plus 5 per cent. But there numerous curious additions and tweaks to unpick.

In a nutshell

All English providers in the Approved (fee cap) registration category get their student numbers cap (SNC) as described above. There are a few exceptions – providers that don’t (FE colleges) or haven’t (new-ish providers) submitted their annual financial return to OfS get 1.5 per cent on last year’s numbers rather than their forecast. Brand new providers don’t get a cap at all, and neither do those in the Approved (no fee cap) registration category.

Each provider can merrily recruit up to their allocated cap – and if they already have more acceptances that’s fine too, they just can’t recruit any more students after that.

For providers in devolved nations, there’s a cap on the number on English domiciled students they can recruit (set as last year (from Student Loans Company data), plus 1.5 per cent as a forecast proxy, plus the 5 per cent). These are just the students that are funded from SLC.

What happens if a provider over recruits – well money is taken back. This is done in two hugely peculiar ways.

In England, if a provider goes over, there will be a reduction to the fee limit for that provider in 2021-22. That’s right – all full-time undergraduates would pay less than the £9,250 cap for that year, with penalties in three bands:

  • For a provider going beyond their SNC by >0% to 6%, there’s a 3% reduction in the fee limit (this presumably might include going over by a single student)
  • For a provider going beyond their SNC by >6% to 12%, there’s a 9% reduction in the fee limit
  • And for a provider going beyond their SNC by >12%, a 15% reduction in the fee limit

So, 2021-22 students may well pay less in fees because of the bad behaviour of their provider in a previous year. Which is… odd.

In the devolved nations, there is a variation – instead, we get a reduction to the amount paid to the provider per new student in 2021-22 by the Student Loans Company. There’s draft secondary legislation – “The Higher Education (Fee Limits and Student Support) (England) (Coronavirus) Regulations 2020” – to follow, primarily to make the punishments work in 2021-22.

Bidding wars

If a provider wants to recruit even more than their SNC, there are two additional student numbers competitions they could enter. The first allows them to bid for up to 250 places in an odd grab bag of subjects including the hard sciences, architecture, teaching, and social work – which at least gives us what might turn out to be a prescient sense of the courses that DfE ministers regard as “valuable”.

The second allows providers to bid for any number of courses in a selection of healthcare disciplines.

The bidding/eligibility process feels very TEF – to the extent it uses TEF data. You need a continuation rate greater than 90 per cent and a highly skilled employment rate greater than 75 per cent from last year’s TEF workbooks to qualify for the liquorice allsorts competition (other than for initial teacher training for some reason) – plus there’s a 500-word statement of assurance where you assure DfE that you have the capacity, kit, and commitment to teach those extra students.

For the health care competition, there are no metrics, but you have to prove that you’ve filled all your existing places up to the cap and that you have the capacity to teach and provide clinical placements for the extras. There’s no cap on the number of places you can apply for – even the total of 5,000 doesn’t appear to be a hard limit.

The bidding competition has a metrics element – the higher your percentages in the two indices of power, the more places you get. You can’t bid at all if you are subject to any conditions of registration that start with the letter B.

6 massive gaping holes in the overall idea

In devolved nations, the cap only applies to full-time undergraduate students who attract student loans company funding, in England it applies to all students. This may mean that the (roughly) 7.5 per cent of FT undergraduate students who pay their fees with a cheque would not be subject to these limits if they choose to study in Scotland, Wales, or Northern Ireland – providers there that recruit students from well-to-do backgrounds can, seemingly, fill their boots. For similar reasons, we anticipate a growth in part-time enrolments that later shift mode – the perfect antidote for those who don’t fancy year one online.

The clawbacks are staggeringly inept in design and execution. There is no reason at all not to cap the amount of loans available via SLC for English providers – it could, in fact, be argued that forcibly reducing fees for 2021-22 would make the cap-breakers more attractive to applicants. Why not pay less for year one, if you have the chance?

And while the principle of giving students a fee cut might be a good one, this feels like an odd way to do it – offering the lower prices to the first cohort to see their study not affected by Covid-19.

Those young people who chose to sit actual A level exams in the autumn are also not counted within the cap. You can see why this has been done – imagine delaying your university year to ensure that you get the right grades and then not having a place available? But this now becomes a perverse incentive not to enter clearing if you’ve missed your grades – why go to a less favoured provider if you can resit and then get a place pretty much anywhere you want in January?

Speaking of which, there isn’t any mechanism for students who might choose to apply to university for a January start for any other reason. So if you need to spend an extra few months taking care of your poorly Mum, you might want to resit an A level too. Otherwise, the cap will apply and you may not get a place.

What are these caps? I mean we know how they are calculated, but we don’t know how many students each provider can take this year, because neither OfS nor DfE deems it appropriate (commercial in confidence, apparently) to publish the student number forecasts. So we won’t know who has been naughty and who has been nice until, we suppose, DfE or OfS announces it in due course. Transparency is a basic component for a trustworthy system of student number allocation – HEFCE used to do it well.

6 massive gaping holes in the additional student numbers bidding process

The full details for the bidding element of the policy are not due until 17 June, so take these points in the spirit of formative assessment – in the hope some of these problems can be fixed.

First – why is there an arbitrary selection of subjects eligible under pot one (no, it isn’t “high cost” subjects assumed at a first glance). Architecture is seemingly a priority despite there being neither a shortage or a current immediate need. Business logistics is going to be a massive issue – pandemic resistant supply chains feel like an investment in the future, but there are no additional places made for it. Surely the sensible thing to do would be to allow these additional places to flow where needed – if more people than usual want to study sport at Loughborough then surely a case could be made.

Or do you even need that? There aren’t any mechanisms that would stop a provider bidding for 250 extra social workers and then enrolling 250 more fine arts students. Clearly you wouldn’t (or shouldn’t) get away with only recruiting 249 social workers from that allocation. But 251 would presumably be fine – you could say that 250 of them were the “extra” you’ve been given.

There are some odd omissions in the healthcare list. Who’s been running all these Covid-19 tests for us? Lab physiologists. Is there a huge shortage in the UK? Yes, there is. Are there more places on offer. No. Pharmacists have been doing a cracking job as the new frontline, advising people on health care and hygiene during the pandemic, but according to DfE, we don’t need any more of them. Spoiler – we do.

It’s an easy target, but we’ll shoot: what exactly is the relevance of the employment of 2013-14 graduates six months after graduation on the ability to manage additional places? Why should the number of students who started in 2012-13 and then dropped out hold any sway? We understand why you can’t use actual TEF (because you might accidentally look at which students actually reported getting something out of the course in the NSS metrics) but it is obvious to any data-literate observer that eight-year-old data will have no predictive utility in the After Times.

Maybe we could drop these banal, useless, pretence of data-driven regulation and just beef up the actual bids. You get more than 500 words about a course on the UCAS page – surely to make a nationally-informed strategic judgement on the investment of public funds we would need a few extra words and maybe some data tables or diagrams. We also note you reserve the right to prioritise applications by geography and subject – why not just do that? Decide what courses you want to beef up, and where in the UK you want them, and then run some tiny, localised competitions?

And meanwhile, if you have a condition of registration under category B you can’t bid. Just to be clear, if you have 14 pence in the bank and your campus is partially owned by your credit card supplier, then you can bid for as many places as you like. But if your rate of highly skilled employment was a little out of whack in 14-15, forget it.


We know this is difficult to believe from a government that has been marked from day one by an attention to detail and a commitment to transparency and consultation, but from what we can see of this proposal it is a bit of a dog’s breakfast. We’ve said before that only a handful of monster recruiters are likely to come anyway near last year plus forecast plus five per cent, especially with deferments widely forecast and a demographic low point for UK 18-year-olds. So expectations for this intervention were low.

But somehow it has managed to be worse even than expected. Needlessly complex – requiring a statutory instrument? – and using wildly inapplicable data in a bidding competition that has learned nothing from the clean and elegant way HEFCE used to allocate additional student number. We’ve not been this disappointed by something since the year-long delay in the publication of the TEF review.

Note: Some changes were made to this article during the course of 2 June as our understanding grew.

8 responses to “Student number controls are back – with deadly penalties for over-recruitment

  1. So the big universitys will be able to get a 5% increase in numbers at the expense of the rest. This exactly what a cap should be trying to avoid, but it will keep those big universitys with the loudest voices quiet so that those in power can claim everything is alright. It’s going to be a fun year indeed I think a few of the smaller institutions will not make it to Christmas without help.

  2. Just on a point of fact. DFE is using current student numbers rather than forecasts for the 176 colleges on the register because OfS didn’t collect a financial return from colleges in autumn 2019 while ESFA (their lead regulator) didn’t collect equivalent data. This leaves colleges with the 1.5% assumed growth rate (or 10) rather than a figure reflecting their actual circumstances or plans

    Some of the commentary on the temporary student number controls says they’ll be too tough (“deadly penalties”) other comments say they’ll be too loose (“increase in numbers at the expense of the rest”). These sort of funding polices generally need testing and trialling so need more than a year to work. Student behaviour is unpredictable so who knows? A topic that hasn’t had so much attention is that the fact that it’s Department for Education managing these rules rather than the Office for Students. Presumably the HE regulator felt it lacked the time and the legal authority to take quick action. Just two years after OfS started work and the department is stepping in to regulate where the regulator can’t. Given that student number controls are quite normal where public funds are involved, this suggests the 2017 HE legislation was incomplete

  3. It does seem a bit Swiss cheese, or half-hearted. But not clear if that is by design or oversight.

    The wording of the exemption for already ‘accepted’ applicants is unclear (despite the enthusiastic underlining). Rationally this would surely include any C offers made before the notification date, since the provider has zero control over whether someone enters into a binding contract for acceptance in response to that. If this is the case then providers can recruit to any number through confirmation of CF/CI, regardless of whether grades are met or not: a generous provision for many. If it doesn’t mean that, then in some cases providers may well have to renege on taking applicants who have met the terms of conditional offers that were made prior to the crisis. What DfE wants providers to do here seems an important point to clarify.

    Similarly, the exemption for students retaking A levels (or rather, taking them for the first time) appears broad. If the intention is not to disadvantage these students from seeing if they have been unfairly treated by calculated grades (and given the equality concerns here this seems reasonable), then rationally this extension would apply to the ‘insurance’ place they take up in the meantime. Otherwise they have to sit out the year on the off chance their suspicion of inequitable grading is correct. In which case it seems universities would only need to sponsor retakes for some of their intake to exempt those students from the limit?

    We don’t think fee discounts are the best response to the current situation but the rules do seem to be encouraging some kind of position where you could pre-announce a discount for year 2 fees in 21-22 to encourage people to enter 20-21 (and compensate continuing students who feel their experience is diminished). You can then recruit to 20-21 effectively exempt from controls?

    Even if these apparent holes get closed in clarifications, the penalties do not seem sufficient to deter the stated policy of increased variation in provider recruitment. For a provider they seem broadly break-even in terms of total teaching revenue (penalty applied to all years in 21-22, but you get three years of income from the 20-21 entry cohort), at the cost of reducing the unit of resource. Which with fixed costs (and vacant overseas places) might not be much of a deterrent, particularly if you consider accommodation income and eventual PG progression on top. And once you get to +15% up on intake it is all extra revenue (at a new lower unit of resource) from that point so you might as well go further if you can? This would again exaggerate divergence in provider recruitment, not restrict it.

    We do think there remain plausible scenarios where UK demand is strong this Autumn (as well as those where it is very weak) and universities feel compelled to help young people avoid unemployment. So, we do think it is possible these controls will bite into potential recruitment and the details do matter. As we’ve argued previously, overall we think student quotas are the wrong macro response and there are better ways to preserve temporarily vulnerable provision.

  4. “FT undergraduate students who pay their fees with a cheque are not subject to these limits – providers that recruit many students from well-to-do backgrounds can, seemingly, fill their boots“

    From my reading, the control is on students who would be theoretically eligible for tuition loans by way of nationality or domicile etc and thus not limited to just those who take out a loan, so you may be wrong on this. This would be consistent with previous SNC regime, although not clear from the current circular as to whether or not ELQ for example would lie outside the controls this time.

  5. The footnote on page 6 defines UK/EU dom students as (i) in relation to tuition fee limits, students falling within the Schedule to the Higher Education (Fee Limit Condition) (England) Regulations 2017, and (ii) in relation to tuition fee loan amounts, students falling within Schedule 1 to the Education (Student Support) Regulations 2011, which would seem to exclude ELQ students, as the previous SNC did.

  6. So if you are willing to accept Augar’s recommended fee (give or take some rounding) you are effectively exempt from SNC? Coincidence?

  7. I agree with Mark Corver’s comment: there needs to be (urgent!) clarification of the exemption for ‘accepted’ students. Institutions may have made conditional offers in good faith, perhaps before the virus was an issue, but may not have had all or most of those offers accepted. Such institutions will have no control over how many conditional offers will be met (and of course also face uncertainty due to A level changes etc this year), but will presumably have over-offered to some pre-determined cover ratio. To penalise such institutions seems bizarre, but, on the face of the published guidance here, the institutions will (possibly) have a choice of either resiling from offers or reducing fees next year. Surely that’s not what DfE wants?

  8. surely it is fundamental that if the offer terms are met as expressed ( ie whether as grades or as points) that the offer cannot be resiled from irrespective of any financial penalty on the institution ( whether direct or by fee reduction in the following cycle which may arise). This is fundamental to operation of the admissions scheme irrespective of what the DfE/OfS may latterly have decided it wants.

    The issue around the treatment of conditional offers is surely around whether the ‘exemption’ Mark refers to is extendable to include any margin for flexibility/mitigation institutions may traditionally have applied in previous cycles and would doubtless wish to retain the right to exercise in the peculiar specific circumstances they – and applicants – face this cycle in order to take in students holding conditional offers at slightly reduced grades or tariff points. these two instances are distinguishable and it may be that the latter case would be regarded as being outside the letter of any exemption but arguably – within some sort of recognised margin – should be held to be within its spirit.

    What is truly bizarre about the methodology selected is that the benefit of over-recruitment goes to the next cohort of entrants rather than as a penalty leviable on institutions through a reduction in their income flow from SLC – I would guess this route has been chosen less because of any thought about future cohorts of entrants and ensuing market behaviours in future cycles than to be seen by Vice Chancellors as not reducing income flows to institutions in the 2020 academic year where the financial pressure on institutions is already known to be intense and so to reduce the likelihood of a further UUK request for a sector bail-out.

Leave a Reply