A butterfly flaps its wings, and tuition fees remain frozen

By common understanding, an incoming Labour government won’t be making any major early changes to tuition fees and maintenance loans – either their rates or their repayment terms.

But soon enough, it will need to announce what the confirmed rates are for maximum fees and maintenance loans for 2025/26.

And there could well be a major problem coming for anyone hoping even for a gentle increase in either.

The cold never bothered me anyway

As it stands, “official” government policy is that the cap on undergraduate tuition fees in England is frozen at £9,250 for 2024/25.

That therefore also means that as it stands, we’re due an increase in 2025/26. Government policy has never said what the unfreezing would result in – but in today’s official Department for Education (DfE) Student loan forecasts for England, the assumption that civil servants are working with is that this would mean that the 2025/26 fee cap would be 2.1 per cent higher in cash terms at £9,450 – based on the OBR’s official RPI projections.

It’s also used those projections to cost up and forecast expected rises in maintenance loans, postgraduate loans and so on.

But don’t get too excited.

You’ll have noticed that Keir Starmer’s response to Martin “Money Saving Expert” Lewis on maintenance was that they would make changes to the system “without putting any more burdens on the taxpayer”.

You could read that sentence in all sorts of ways. Inflation can make the system cost more without doing anything. So can rising participation rates based on the number of 18-year olds knocking about – the DfE projections assume that UG entrant participation will rise from 526,000 in academic year 2022/23 to 567,000 in academic year 2028/29.

It also projects that postgraduate loan borrowing entrants are forecast to increase from 64,000 in academic year 2022/23 to 70,000 in academic year 2028/29, driven by an expectation that growth will return to pre-pandemic trends.

That all has a cost – but let’s assume that the commitment to not burden the taxpayer further prices that in.

The problem with today’s figures is that the subsidy the government is putting in is growing.

It’s funny how some distance makes everything seem small

Ever since the government was forced to book the estimated subsidies in the student loan system to the national accounts (rather than pretending that loans were all going to be paid in full), it’s been required to carry out calculations to that end.

Two bits matter. The RAB is the proportion of loan outlay expected to not be repaid when future repayments are valued in present terms, and the stock charge is the proportion of total outstanding loan balances that are not expected to be repaid.

It’s tricky enough to guess how much people will be earning in 30 years – and these days, now that 2023 entrants and onwards only get their loan written off after 40 years, it’s even trickier. DfE uses historical LEO data to have a stab at that.

Tiny changes to forecasts can make a big difference. If you think of 2010-era “Plan2” loans, if we get a 1 per cent fall in projected earnings growth to 2030, that can take the expected subsidy down by 7 percentage points – 7p in the pound.

That’s a lot when you’re loaning out about £20bn a year.

But of even more importance is the discount rate. This is how much the Treasury thinks that money will be worth in future years – and again, not only is that a guess 30 years out, it’s even more of a punt 40 years out.

I could go into the projections on all of this for each cohort, loan plan, year of entry and so on – but what matters is the upshots of the model for the RAB charge.

Let the storm rage on

The headline is that everything has become significantly more expensive as far as the Treasury is concerned.

This time last year it calculated the RAB for 2023 FT UG entrants at 27 per cent – think of that as a 27p subsidy on UG loans in real, long run terms – falling to 23 per cent by 2027/28.

The problem is that this year, the model has spat out a 29 per cent charge – and that is still 29 per cent by 2027/28.

If we’re loaning out about £18bn a year to FT UG students on that new “Plan 5”, that’s over a billion a year that’s gone.

But here’s a curious thing. The RAB for the old “Plan 2” loans (at least for anyone taking out their allocation this year) has actually gone down from 30p in the pound to 28p in the pound.

You’ll remember that when the government a) froze the repayment threshold, b) moved to no “real” interest on student loans and c) extended the repayment term to 40 years, the idea was to both a) make the system seem nicer by dealing with the politically deadly interest rate, and b) make the system more “sustainable” by recovering more money over time back to the Treasury.

But those little fluctuations in the discount rate, LEO data and earnings growth projections now mean that the RAB on this year’s issued loans for new style and old style terms is pretty much the same.

In other words, the ONLY thing that has changed is that rich men in their fifties will pay less, and pretty much everyone else will end up paying more.

As an aside, the subsidy in student loans in the devolved nations is managed by the Treasury – and they’re allowed to keep their own systems and terms as long as the costs are “broadly comparable”.

Until today, if I was in Wales I’d have been panicking a bit that the Treasury is closing in on demanding the sorts of savings that in theory 2023-England-change (ie Plan 5) students represent. These figures appear to take a bit of pressure off on that front.

One thought crystallizes like an icy blast

Importantly overall, the total theoretical “cost” the taxpayer of the student loan “system” has gone up quite a bit without anyone doing anything – and will continue to do so each year for the next 5 years.

And so here’s the thing. Bridget Phillipson and Rachel Reeves will have on their desks by now the draft announcements on maintenance and tuition fee loans for 2025/26.

If they take into account these extra taxpayer costs, look at the participation projections and stick to the Martin Lewis commitment, it’s hard to imagine that fee cap being unfrozen.

There’s really no time for a wider and more detailed review which ought to ask questions about whether this system of “loans” – the constraints around which now involve impossibly long-run guesswork – is a sensible way to fund things.

But a decision is going to have to be made soon on 2025/26 fee levels. On these projections, both the fee going up and the guesswork involved has a notional significant cost to the Treasury. Whether Reeves will count that as busting, or being within, the envelope is the killer question.