An OBR update on student loan accounting changes

While today’s Spring Statement promised that the Augar review would appear “shortly” and that a government response would be forthcoming later in the year, the real HE action was located in Annex B to the OBR’s latest Economic & Fiscal Outlook.

This section – “Accounting for Student Loans” – outlined the state of play regarding the ONS’s decision to reclassify student loans in the national accounts. Although the OBR “does not yet have sufficient clarity” on the detail of the new method to include it in its forecasts (a number of issues remain to be resolved), it does now understand the difference between the approach that is proposed (“the partition model”) and its own Hybrid model (published last summer).

The fundamental concepts in the two models are similar: the current fiscal illusions in the student loan accounting treatment are reduced by recording as upfront spending an estimate of the write-offs associated with the loans issued that year. At the same time, the interest accruing against loan balances is only recorded as income if it is expected to be repaid. Thus the spending on loans increases in the year they are issued and the income being generated from all extant loans is reduced. Spending goes up and income comes down leading to a large change in Public Sector Net Borrowing Requirement – the “deficit”.

Previously the OBR had estimated the impact on the deficit of implementing this new treatment at £12billion in 2018/19, rising to £17bn by 2023/24.  Their new – albeit provisional – estimate is that £10billion would be added this year and £14bn in 2023/24. That said, “this estimate is subject to considerable uncertainty”. Currently, the deficit is estimated to be £22.8bn this year and £13.5bn in 2023/24. In both cases, the key target measure is moved adversely and significantly; it is more than doubled in the later year.

OBR are also to be commended for giving attention to the problem of revising estimates. Estimates of write-offs depend on forecasts of repayments and the economic variables baked into the income contingent repayment scheme. What should be done when these vary from original estimates?

OBR concludes that Eurostat, the body responsible for official government statistics in the EU, will push the ONS towards using one-off lump sums to record revisions. This has the risk that

“changes in long-term expectations [might have] a distorting effect on the year-to-year path of the deficit that would make it less meaningful and less suitable as a target aggregate for policy”.

As we can see in the figures above, it deficits are generally small then revisions to student loan spending estimates could blow PSNB around.

Finally, OBR also refer to Eurostat guidance on the new treatment that was published in December. Though brief, it contains a key reference to the accounting treatment for student loan sales.

“further capital transfers should be recorded in case of the sale of loans or when significant changes of law, having an impact on the amounts which will be repaid, occur.” (my emphasis)

OBR do not comment on this aspect of the guidance. ONS have so far only indicated that the treatment of sales is an area where conceptual issues remain to be determined; they have provided no indicative treatment for future.

Eurostat’s reference to “capital transfers” here though points to a revision to that treatment too. This blog has previously argued that ONS has mistakenly treated recent loan sales as “revaluations” rather than “capital transfers”. It looks like Eurostat will vindicate that opinion and now require the losses made through sales to be recorded against the deficit.  (Don’t ask me how leaving the EU might affect Eurostat’s authority!)