In the wake of new suggestions that student loan costs have increased, proposals for a graduate tax are re-emerging. There¡¯s some speculation that Labour may even be considering it as a manifesto pledge.

The primary reason to reject a graduate tax is that it replaces the finite and time-limited ¡°¡®debt burden¡± of tuition fee loans with a tax burden that is unlimited both in terms of the total amount due and the period over which it is to be paid. I have always struggled to understand why this would be better for students. Wouldn¡¯t most graduates prefer a time-limited repayment of a fixed amount?

Advocates of a graduate tax might respond that it is possible to create a tax that doesn¡¯t last for life, or that caps total payments - but then we¡¯re talking about something more like repaying a loan. Using the word ¡°tax¡± is then a purely rhetorical gesture ¨C and a strange one at that; I¡¯ve seen no evidence that graduates consider taxes to be more palatable than loans and, so far, the introduction of much higher loan-backed tuition fees has not prevented increases in participation, even among undergraduates from the lowest socio-economic groups.

If advocates of a graduate tax want a system that is more progressive, then let¡¯s talk about that. On the face of it, though, a graduate tax on current income tax thresholds would in one important way be less progressive than the current system, since everyone earning over ?10,500 would have to make a contribution; loan repayments don¡¯t begin until you¡¯re earning over ?21,000. You could play around with the details of the graduate tax to change that, but then you could play around with the details of loan repayments, too.

ÖйúAƬ

Ultimately, the priority for advocates of a graduate tax is probably to reduce marketisation in ÖйúAƬ. The core problem with fees, as they might see it, is variability. I might respond that there isn¡¯t much variability in fees at the moment, but that would be disingenuous. The 2010 Browne Review, for which I headed the supporting civil service team, did explicitly seek greater variability in fees. We wanted universities to be forced to think harder about how and what they taught, for those choices to show up in fees and for students to evaluate them when making application decisions.

Under a graduate tax regime, choices about what courses to back with funding would be made by central government instead. That might work out OK, but advocates of a graduate tax who care about the value of the public university should ask themselves this: is it likely that the government, over the course of several years and changes of administration, will continue to back a non-utilitarian idea of ÖйúAƬ? Or would the values of public education be more likely to be defended by students voting with their feet? Personally, I would back the students. But, as a former and probably future civil servant, I¡¯ll experience a curious sort of pleasure if opponents of fees want to construct paeans to central government instead.

ÖйúAƬ

There are very many practical problems with moving to a graduate tax, too. The main issue with student loans at the moment is uncertainty about how much will be repaid. But switching to a new payment regime with different rules for who pays and how much they pay would only increase the uncertainty; it is almost certain that in a few years¡¯ time graduate tax revenues would be different from those forecast and, even if the tax covered tuition fees, you would probably still need to give students loans to cover the cost of living. This would further multiply the uncertainty.

Of course you could replace maintenance loans with grants, but that would only increase the cost problem. Graduate tax revenues wouldn¡¯t arrive until students have graduated, and would build up gradually over time as graduates¡¯ earnings increased through their careers. In the meantime, government spending would have to fill the gap between the tax revenue and what universities need to operate.

Of course, there is also a time lag until loans begin to be repaid. But those future repayments can be directly set against the cost of filling the gap. Future tax revenues cannot be used to offset specific items of public spending in the same way and so the spending has to be scored in full when it takes place. In one sense, this is an accounting issue but it¡¯s deeper than that: it also reveals that there is no guarantee that future graduate tax revenues will be used to counterbalance current or even future spending on ÖйúAƬ. Switching from loans to a tax doesn¡¯t only make the system more expensive for the next 25 years or so, it also reduces certainty about funding for ever.

I¡¯m not a politician, but I can¡¯t imagine anyone being able to say with a straight face: ¡°The system introduced in 2012 is too expensive, so we¡¯re replacing it with one that costs even more at a time when there¡¯s even less public money to spend. Oh and, by the way, you can trust us not to divert future graduate tax revenues into something that seems a higher priority than ÖйúAƬ.¡±

ÖйúAƬ

If what we really care about is the sustainability of ÖйúAƬ funding, we should stick with the current system and apply what we learn from repayment patterns to adjust it over time. As well as being a terrible idea in principle, a graduate tax would solve none of the problems it faces.