Who should pay for our failing student loan system?

YouGov polling found that of those 44 per cent who supported writing off the loans, more than a third (36 per cent) felt the loans should be forgiven entirely.

35 per cent supported writing off a portion, while 25 per cent agreed there should be a limit that means students don’t have to pay back more than they took out (which translates as no “real” interest).

In Plan 2, RPI+3 per cent was all about soaking the rich to pay for poorer graduates. Lowering interest makes the system less progressive. But Labour’s woeful defence of the progressive features in the system (it’s more like a progressive graduate tax) runs the risk of public opinion drifting towards the opposite (where everyone pays what they borrowed regardless of the benefit obtained).

There’s a follow-up comment piece in the paper tomorrow (published this evening) that quotes one of the architects of the system, Nick Hillman. Writing in the Guardian in 2014, he said:

Come with me to the election of 2030. Those who began university when fees went up to £9,000 in 2012 will be in their mid-thirties by then. That is the average age of a first-time homebuyer and the typical age for female graduates to have their first child. By then, there will be millions of voters who owe large sums to the Student Loans Company but who need money for nappies and toys, not to mention childcare and mortgages. So, however reasonable student loans look on paper now, the graduates of tomorrow could end up a powerful electoral force.

And so it came to pass. Hillman’s view was that if the funding model was to prove sustainable, policymakers would have to ensure a healthy graduate labour market in which higher-level skills earn higher-level rewards:

Those who want the main features of our current student loan system to survive must ensure today’s students get at least as much benefit from university as their predecessors did. But that is a real challenge as higher education is no longer a small elite system reserved for a favoured few.”

And it’s one that politicians are clearly finding it harder and harder to meet – if they intend to at all. Arguably the real issue is that in 2014, projections were more expensive to taxpayers than was predicted in 2010, with 45p of every £1 loaned out to students being written off.

But few policies are perfectly conceived and any design flaws are mild … because no one expects the scheme to be wholly self-financing.

Try telling that to 2022 entrants, whose loans IFS now thinks will deliver a profit for the taxpayer. That the Treasury appears to have spent the past decade getting its revenge on whoever did the initial modelling is the hardest bit of the system to spot.

The mis-selling debate

In the mood music that surrounds the current student loans (Plan 2) crisis, one of the allegations is that loans were mis-sold – accompanied by suggestions that the state can retrospectively vary the terms.

It’s a fair point insofar as the government actually has a broad statutory power to amend almost all of the repayment terms after borrowers have taken out the loan. The legal basis sits in the Teaching and Higher Education Act 1998 and the Education (Student Loans) Regulations made under it.

The courts have consistently treated Plan 2 loans as statutory loans rather than conventional consumer credit contracts, which very much limits the extent to which borrowers can rely on contractual certainty.

It’s an unfair point insofar as the only things that governments have actually done is freeze some thresholds that were intended to be uprated. In reality, most Plan 2 borrowers pay nine per cent above the given threshold for most of their lives. While legally that could be changed via regulations, an increase would be more exposed to challenge.

In other words, change for these borrowers requires an attempt to shift the Overton window such that government feels it has to either change something it previously has never done, or keep a political promise that it has previously abandoned.

As I say, the key thing governments can do – and have done – is fiddle with the threshold. The planned freeze in the threshold announced in the Budget is arguably the source of the current commentary. A group of Labour MPs are said to be organising to reverse it.

And it’s where, despite some commentators suggesting that there are “only three ways to tackle the current student loan crisis”, is where government could keep its promises.

Demo-lition

In the run-up to the £9,000 fee settlement, coalition partners Vince Cable and David Willetts presented the higher repayment threshold as a core “progressive” feature of the package.

The repeated headline was a rise in the income threshold from £15,000 to £21,000, with repayments at 9 per cent of earnings above the threshold, all with the threshold intended to rise over time rather than stay fixed.

At the time, Cable’s framing (Commons oral statement, 12 October 2010) was explicitly anchored to John Browne’s recommendations, describing “a £21,000 graduate income threshold before any payment is made” and stating it should be “linked to average earnings”. Simples. Higher earners would contribute more, lower earners would be protected by the higher starting point for repayment.

Willetts’ key formulation (Commons statement, 3 November 2010) was that graduates would begin repaying only when their annual income reached £21,000, and that the threshold would then be increased “periodically to reflect earnings”. In the same statement he tied this to lower monthly repayments for graduates and the overall progressivity of the repayment mechanism (including the 30-year write-off).

Even that was a bit of sleight of hand. By then in opposition, in the debates in 2010 John Denham noted that the £21,000 repayment threshold was being presented in “2016 prices”, which risked overstating its generosity when set against the £15,000 threshold under the previous system. He had wider concerns about whether and how the repayment threshold would in practice be uprated over time.

By December of that year, ministers had moved to reassure by promising annual uprating for earnings:

The £21,000 earnings threshold will also be uprated annually in line with earnings from April 2016 (when the majority of students who commence a three year degree course in September 2012 will become liable to repay).”

The 2015 threshold freeze

The government’s first run at breaking that promise came in July 2015, when a consultation appeared on freezing the post-2012 student loan repayment threshold at £21,000 from April 2016 until at least April 2021, rather than uprating it in line with earnings as borrowers had been promised when the system was sold.

The consultation and the response justified freezing the threshold (and the £41,000 “upper” interest threshold) primarily on the grounds of public cost and fiscal risk. It argued that when the post-2012 system was designed, forecasts assumed a higher proportion of graduates would be making repayments once the £21,000 threshold took effect in April 2016.

By 2015, updated evidence and modelling implied fewer borrowers than expected would be above the threshold, which increased the estimated long-run subsidy cost of the loan book. Freezing the threshold in cash terms from April 2016 to April 2021 was presented as a way to keep the higher education funding system “sustainable” by increasing expected repayments relative to the uprating plan.

It then justified freezing the £41,000 interest threshold alongside £21,000 as an internal coherence point. The argument was that if you freeze one threshold but not the other, you change the shape of the variable interest regime in ways that were not intended, because the interest rate steps between the lower and upper thresholds.

Martin “Money Saving Expert” Lewis, who’d been touring the studios on behalf of government to sell the features of the 9k fees system, was tamping.

He framed it all as a retrospective worsening of loan terms for the first cohorts under the £9,000 fee regime, arguing it undermined trust in the student finance system by changing a key parameter after students had taken out the loans.

No doubt drawing on his skills as an SU officer in his youth, he launched a public mobilisation via MoneySavingExpert, including a large petition, submitted a formal consultation response jointly by him and MoneySavingExpert that argued the freeze breached a clear promise and would materially increase repayments for affected graduates, and pushed explicit pressure tactics including threatening judicial review.

The core consumer-protection claim was that the policy was sold as “threshold rises with earnings”, so freezing it was equivalent to rewriting the deal after signature. The government did it anyway.

Theresa May’s partial reversal

The freeze was then partially reversed by Theresa May. In her party conference speech in 2017, there was a “we have listened and we have learned” pitch aimed at younger voters and “ordinary working families”.

In a section on “the British dream” and living standards, alongside commitments on housing, energy prices and conditions at work, she promised to increase the amount graduates could earn before they start repaying their fees to £25,000 straight away, alongside scrapping planned fee rise and freezing the maximum fee while the Augar review took place.

The rhetorical “burning injustices” move was that the party should show it is on the side of people who “work hard and do the right thing” but feel the system is not working for them, with student debt presented as one pressure among several on household finances and early adult life – “putting money back into the pockets of graduates with high levels of debt”.

So the threshold was lifted to £25,000 from April 2018, with annual uprating in line with average earnings restored thereafter. No changes to the tapered interest rates were put in.

Most think she got little political credit for it, and it came at a significant cost to the Treasury, who for reasons explained in other articles, soon had to start actually accounting for expected long-term losses in the system.

The Augar review

When her Augar review finally reported, it went for a version of the graduate premium argument – proposing that there should be a stronger expectation that student contributions will be made “once a financial benefit is secured”. It said:

For students in degree-level education we therefore recommend that the most suitable threshold is median non-graduate earnings.

When it came out in 2019 it was using 2018-19 prices, which meant reducing the threshold from £25,000 to £23,000. Later on Page 170 of Augar, we got this:

However, the panel would expect this change to be implemented… in academic year 2021/22. At this point – on current earnings forecasts – the recommended threshold would have risen to approximately £25,000, around the same nominal level as today…. Once introduced, that threshold should continue to increase with average earnings over time, as is currently the case.

David Willetts then appeared in the Telegraph to deploy a version of the who pays, grads v non grads thing:

So it is right to expect graduates to pay back so that their higher education is not funded by taxpayers earning less than they do.

In that article instead of taking the median non-graduate salary in general (£25,500), he quoted the median non-graduate salary for people in their twenties (£21,500) – and then as if by magic, £23,000 started to look generous.

But he did understand the interest rate issue:

The interest rate on debt could also be cut. It is probably the most hated feature of the system. But it is just part of the real problem which is the terrible combination of the low repayments because of the high threshold and then the high interest rate so that for too many graduates their debt is rising every year. That is depressing for graduates and their parents and politically toxic. Polling shows that graduates and their parents would rather get on with paying their loans back, even though they are nothing like commercial debt.

Technically, lowering interest rates would only benefit rich male grads – while raising them would both generate more back into the system and would only hit very well off male graduates. But it wasn’t the point.

The Donelan freeze

Scroll to when Rishi Sunak was at No.11, when higher education minister Michelle Donelan was sent out to front out some changes both for new loan holders (Plan 5), and a sneaky reversal of the Theresa May position.

Donelan’s “freeze” was the January 2022 decision to stop the automatic uprating of the Plan 2 repayment threshold – her ministerial statement said it would be kept at its 2021–22 level of £27,295 for 2022–23, rather than rising with earnings, and that the Plan 2 interest rate thresholds would also be held at their existing levels for the same year (with the lower interest threshold aligned to £27,295 and the upper interest threshold at £49,130).

The government justification in the statutory instrument material was again “sustainability” of the student loan system, and commentary described it as a tax rise by stealth – because it raised repayments without changing the 9 per cent rate.

Few seemed to notice, and so the government then confirmed it would be held at that cash level until April 2025. That meant the “freeze” became a multi-year policy rather than a one-year pause, with even more borrowers pulled into repayment and those already repaying paid more each month than they would have under an uprated threshold.

In April 2025, the Plan 2 threshold did actually rise – from £27,295 to £28,470, reflecting the resumption of annual uprating after the freeze. And uprating on interest resumed too. The lower interest threshold increased in line with the repayment threshold to £28,470, and the upper interest threshold was also increased (to maintain the same banded structure), rather than remaining frozen, as set out in government guidance.

Rachel Reeves and the 2025 Budget

It couldn’t last. Last November, Rachel Reeves announced that the Plan 2 repayment threshold would be held at its 2026 to 27 cash level for three years from April 2027 (so the threshold does not rise in April 2027, April 2028 or April 2029).

The Budget document framed this as a sustainability and distribution argument, saying graduates generally benefit from higher earnings and that requiring them to repay more is fair to workers who did not go to university, and also asserting that the change “does not increase the level of debt” for those graduates (because it changes repayments rather than the face value borrowed).

The IFS then highlighted that a parallel freeze was being applied to the Plan 2 lower and upper interest rate thresholds for three years as well, even though this was not clearly set out in the main Treasury Budget document. Freezing the cash values of those interest thresholds means that as earnings rise, more borrowers sit further up that taper for longer, increasing interest accrued and, for many borrowers who eventually clear their balance, increasing lifetime repayments by extending the repayment period rather than raising monthly deductions in the short run.

The IFS reports these interest thresholds were set to be frozen at their April 2026 levels of £29,385 (lower) and £52,885 (upper), as detailed in their annual report on education spending.

On impacts, the IFS estimated that millions of Plan 2 borrowers would repay around £93 more in 2027 to 28 and £259 more in 2029 to 30 because of the repayment-threshold freeze alone, and that the combined effect of the repayment-threshold freeze plus the interest-threshold freeze would increase expected lifetime repayments for the 2022 to 23 entry cohort by about £3,200 (around 6 per cent) on average (from £52,600 to £55,800 in today’s prices), with borrowers in the third decile of lifetime earnings repaying around £5,000 (around 14 per cent) more.

The IFS also emphasised that the interest-threshold freeze accounts for more than a third of the increase in average lifetime repayments for that cohort – and is particularly important for higher-earning graduates who spend much of their careers between the two interest thresholds and ultimately repay the additional interest rather than having it written off.

What would keeping promises mean?

So what? Well, we have two ways of looking at things on the “what can we change, what can’t we change” thing.

The Martin Lewis position is essentially – stick to the terms (even if only stated politically/morally) you offered. The government’s repeated position is – nah, this is more like a tax, and if our modelling on how much the cohort pays is undercooked, or if the wider economy tanks, we reverse the right to make you contribute more via fiscal drag.

Full disclosure – I’m with Martin on this. It’s not Plan 2 borrowers’ fault that the economy hasn’t been doing as well as intended, and nor is it their fault that the system overall isn’t seeing the repayments intended. If you want to pull off some measures that tackle that, that’s what the general taxation system’s for.

But what would that have meant, and mean?

On the threshold issue, if we operationalise the original promise as “£21,000 in April 2016, then uprated each April by whole-economy average weekly earnings, regular pay”, the counterfactual comes out at about £30,700, around £2,200 above the actual £28,470 now in force.

That would feel better. Not massively, but better. And a lot better by April 2029.

Then on the interest issue, most would be paying less of that as well – because the taper from RPI to RPI = 3 per cent would be aligned with that new threshold.

None of that would really shift the dial financially, mainly because it wouldn’t really fix the biggest “miss-sell” in the system – paying their graduate tax for a full 30 years then getting some loan written off is a design feature, not a bug. The only borrowers who get to stop paying their graduate tax are (very) high earners.

It reminds us that all (other) things being equal, reducing interest only means that the richest third get to stop paying their graduate tax a little earlier. It reminds us that to reduce repayment rates would probably require even higher interest – which although would only impact richest grads (by asking them to pay their graduate tax a little longer), would still look and feel bad.

But as I say, it mainly reminds us that the major problem with that Plan 2 system is that nobody said “most of you will never pay this back in full, and as such it’s a tax for 30 years”.

Considering it as a tax, and thinking about it in the round of tax burdens for the young would help, as would being honest about what the Treasury is expecting back when it fiddles. But at the very least the one thing the government could do is keep the promises made when students took out their loans.

On the HEPI blog today, Nick Hillman argues that the Plan 2 protests resemble the failing WASPI campaign by those who say the equalisation of State Pension Ages was so badly communicated that billions in compensation should now be paid.

He’s right that in both cases, part of the problem is crap communication. But when the state can alter the substance of the deal in ways that have increased lifetime repayments, increased interest exposure, made the system less progressive and make middle-low earning graduates pay for poor projections in the past, that’s a very different kind of policy failure.