Skip to content

Student loans – proposed form is not a fix

May 15, 2019

You’ve probably seen the coverage of the Money Saving Expert / Russell Group team-up and their proposals for overhauling the annual statements provided by Student Loans Company. Although the intentions are good, the “revamp” misfires and they’ve based a key part of their statements on an elementary mathematical blunder.

MSE and RG are keen to emphasise the “tax-like” behaviour of ICR loans by turning borrowers’ attention away from interest accruing and the outstanding balance (which is more likely to be written off than cleared). Instead, they want your focus to be on the current and future repayments.

The problem is that income contingent student loans are complex and have long lives.

Not only is it hard to give a sensible indicative level of cash repayments, it is extremely difficult to translate that projected cash total into something meaningful today. (MSE are most concerned by individuals who make additional voluntary repayments, which in most cases is throwing away money). Even if you have some familiarity with financial concepts like present value and discounting, the unusual nature of ICR loans makes such methods less reliable.

Unfortunately what’s proposed by MSE and RG makes a hash of this crucial issue: how should you understand today what repayments your student loan commits you to.

At this point, I should say that I haven’t read the accompanying report. I have only looked at the example statement. I will read the report tomorrow and blog a bit more.

The problems are located on the final page of the form: “Predictions of your likely future contributions”.

Various key assumptions are set out: repayment threshold, average earnings, salary growth, inflation and your working life. Average earnings growth of 2.7% is assumed for the next 30 years (government loan models assume AEG returns to over 4% in the long-run); the repayment threshold increases in line with that as does salary growth (no promotions or pay rises are factored in); your working life is assumed to be 30 years with no career breaks. We might quibble over these figures; we might also refer back to the previous work on this blog that forced the SLC to take down its “repayment calculator” owing to “everyday sexism”.

The first mistake seems to come with respect to inflation.

inflation

Student loans are linked to RPI, but the Bank of England targets CPI. RPI is higher than CPI and the spread in recent years has been as much as 1 percentage point. This suggests that the MSE/RG model is underestimating interest.

The second mistake is more fundamental. It comes when the statement tries to present its example of a £39,000 “future contributions” estimate (a cash total) in “today’s money”.

discounting

It can be seen quite quickly that there’s a blunder here. The form treats that £39,000 as if it were a single lump sum paid in 2048 and derives an equivalent figure in today’s money by removing 29 years’ inflation (at 2%).

£39,000 / (1.02 ^ 29) = £22,000
(where “^” means “to the power of”)

It doesn’t really need saying that student loans don’t work like this! You aren’t asked to pay up in 2048!

That £39,000 results from different payments made at different times and so each payment needs to be “discounted” to reflect when it was paid. One made today isn’t discounted, one made in ten years is divided by 1.22 (1.02 ^10), and only those sums paid in 2048 should be discounted with a factor of 1.78 (1.02^29).

So it’s immediately apparent that the MSE/RG approach underestimates interest and undervalues the repayments made, leaving borrowers with a misleading view of what their projected repayments are and what they are worth.

[Update – I’ve now had the chance to play around with the figures and think I have recreated something very close to the flow of repayments in the example. Using the suggested discount rate of 2%, the “equivalent” figure should be about £29,000 – £7000 higher than claimed.]

In the proffered example, this may not appear to be an issue as estimated cash repayments are lower than the outstanding balance of £50,422. But this will not be the case for everyone. In many ways MSE/RG have chosen an easy example: it is much harder if you have set out a statement for higher earners, where estimated cash repayments exceed the outstanding balance and borrowers want to consider whether additional repayment will save money. In such a case, the MSE/RG will misinform borrowers.

There is a further issue: a discount rate matched to CPI is completely inappropriate for an individual here. (The government’s financial reporting discount rate is RPI +0.7%). I will say more about this in tomorrow’s post. My own feeling is that discounting itself is inappropriate here (and that this means that real interest rates are a poor design feature in ICR loans) but if you must discount, then you should set the rate to match annual increases in average earnings as that is the main driver for estimating cash repayments.

(Note that this problem of discounting also bedevils the analyses of those academics who assume that paying fees upfront is necessarily cheaper than taking out loans. They misunderstand the time value of money).

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: