Long-term projections of the fiscal impacts of student loans

The current English student loans system is largely the result of reforms first introduced for students starting higher education in 2012. Since then, policy on student loans has frequently changed, with reforms generally reducing the long-term cost to the government, such as freezes to the tuition fee cap, changes to repayment thresholds, and the introduction of Plan 5 loans.

There are two channels through which student loans affect the long-term public finances. The first is through cash flows, reflecting the difference between total loan outlays and total repayments in any period, which affects the central government net cash requirement (CGNCR) and public sector net debt (PSND). The second is through the treatment in the accruals accounting framework, where public sector net borrowing (PSNB) and public sector net financial liabilities (PSNFL) are affected by the difference between the estimated capital transfer at outlay (an estimate of the portion of the loans that will eventually be written off), and modified interest (an estimate of the interest payments that will be received on the portion of the loan that will be repaid).^a

In our baseline scenario, government cash borrowing (measured by the CGNCR), used to finance student loans, declines as a share of GDP until the mid-2040s (Chart C). This reflects: a projected decline in loan outlays due to a fall in the number of young adults as a proportion of the population; increasing repayments from the cohort of students since 2012 moving up the earnings distribution; increasing numbers of borrowers in repayment until Plan 2 loans reach their write-off; and, the impact of the introduction of Plan 5 student loans. As Plan 2 student loans begin to be written off after their 30-year term, there is a brief period from around 2045 when cash borrowing is projected to increase again as repayments from these loans decline. Thereafter, cash borrowing is projected to be broadly stable through the rest of the projection period. These cash flows drive the net impact of student loans on PSND, which is pushed up as a percentage of GDP over the next 10 years due to the initially relatively high cash borrowing, before stabilising as cash borrowing falls.

Turning to the projected net impact of student loans on PSNB, initially, from the end of the medium-term forecast period, the interest receivable on the stock of student loans is projected to exceed the estimated capital transfers. Over the next decade or so, there is a further decline in PSNB as a share of GDP as modified interest is projected to grow as the stock of loans grows, while capital transfers are projected to remain relatively consistent as a share of GDP. From the mid-2040s, modified interest stabilises as the stock of loans stabilises, and thereafter changes in PSNB are relatively small and largely driven by projected demographic changes. The impact on PSNB shown here does not include the cost to government of the borrowing it uses to finance student loans. Based on our projections of the government’s long-run borrowing costs, the long-term annual cost to the government of financing the stock of PSND accumulated since 2012-13 will average 0.4 per cent of GDP, meaning that the system as a whole increases borrowing in most years.

Chart C: Baseline scenario for the impact of student loans on the public finances

Note: These reflect the impact on CGNCR, PSNB and PSND from student loans only, including from the devolved administrations. PSND reflects the estimated impact from student loans since 2012-13.

Source: Student Loans Company, OBR

These projections are very sensitive to uncertain assumptions on the number of students going to university, the path of graduate earnings, and long-term government policy. In the baseline scenario shown here, we assume that the tuition fee cap, maintenance loans and repayment thresholds are uprated by average earnings. This is consistent with our wider set of assumptions for unchanged long-term government policy (see Chapter 1 for details). If we were to instead assume that over the long term the tuition fee cap, maintenance loans and repayment thresholds were to rise with inflation, then the impact on PSNB and the CGNCR would converge towards zero, due to the decreasing value of loans and repayment thresholds relative to earnings and GDP. If this were to happen, it would create significant financial pressure for the higher education sector by reducing the value of income from tuition fees relative to average earnings.^b Conversely, any future increases to the generosity of the student loans system, such as above-earnings increases to repayment thresholds or the tuition fee cap, would be likely to increase the CGNCR, PSND and PSNB relative to the baseline scenario.