How to fund higher education in South Africa: a public-private-university partnership may be the answer

Reads 4,973
Philippe Burger, 28 July 2021
Higher education is both a private and public good – for society, an investment in the future, for the individual, a chance to improve their lives. But the current model of student funding is neither financially nor fiscally sustainable. It is unlikely that government can fund the “missing middle” – students from households with incomes between R350 000 and R600 000 a year. What then to do? A loan scheme involving universities, government, and the private sector may enable students to graduate and to start re-paying only once their income reaches a certain level.

 With a Covid-hit, shrinking economy and a mounting public debt burden, the Minister of Finance, Tito Mboweni, announced a tight budget in February 2021. This budget also constrained its allocation to the Department of Higher Education and Training (DHET).[1]

Within the DHET budget, the allocation to the National Student Financial Aid Scheme (NSFAS) was set to increase from R34.8 billion in the 2020/21 fiscal year to R36.4 billion in 2023/24, a cumulative increase in nominal terms of 4.6% over the three-year period. This allocation covers NSFAS bursaries to university students and students at technical and vocational education and training (TVET) colleges.

However, the National Treasury’s Budget Review has projected inflation at 3.9%, 4.2%, and 4.4% in the three fiscal years from in 2021/22 to 2023/24. This means that the consumer price level over the three years is expected to cumulatively increase by 13%, well in excess of the 4.6% increase the government budgeted for NSFAS. Moreover, the government also expects the number of NSFAS students to increase.

Student organisations countrywide expressed their dissatisfaction, which led to protests and campus shutdowns in March 2021. Tragically, a bystander to the protests, Mthokozisi Ntumba, died during police action in Braamfontein.

Following the protests, the Minister of Higher Education, Innovation and Technology, Dr Blade Nzimande, announced a reallocation of the DHET budget, effectively allocating a further R6.3 billion to NSFAS. A total of R2.5 billion of this came from a reduction in the general allocation for universities, R3.3 billion from the National Skills Fund, and a further R500 million from the TVET college new accommodation construction budget.

The provision of university subsidies was already a concern before this reallocation, with the subsidy per student in real terms set to drop cumulatively by as much as 7% over the period 2020/21 to 2023/24.

In addition to the subsidy and bursary pressures, student organisations are also demanding a full write-off of student debt. Outstanding student debt at South African universities stands just shy of R14 billion. Much of this debt burden is carried by students from so-called missing-middle households - households with an income between R350 000 and R600 000 per year. 

With mounting financial pressure, it is clear that the current model of student funding in South Africa is not financially and fiscally sustainable, that is, financially for universities and students from missing-middle households, and fiscally for government. Deteriorating fiscal conditions also make it unlikely that the government can fully finance the missing middle. Minister Nzimande subsequently established a National Task Team, involving various stakeholders, to address the student funding challenge in a sustainable manner.

The National Task Team will have to revisit the recommendations that the Heher commission made in 2016. The commission recommended the implementation of an income-contingent student loan scheme. This would enable a student to obtain a loan to cover all or part of his or her tuition, accommodation, books, living costs, and transport. Once a student finishes studying and starts working, loan repayment can start, but only when income exceeds a set threshold. The amount paid per month is also linked to the graduate’s income. The loan repayment period can be capped, for instance, at 25 or 30 years. Whatever is not repaid after that, is written off.

Such a loan scheme can augment a revised NSFAS bursary scheme, and instead of the hard R350 000 family-income cut-off currently applied for NSFAS bursaries, be implemented with a sliding family income scale that allows for a combination of bursary and loan financing. Thus, poorer students will receive a bigger or full bursary, reducing their need for a loan, while better-off missing-middle students will need to obtain a partial or full loan.

Will students be able to afford the debt burden they will incur with such loans? In 2019 BusinessTech conducted a survey among eight large South African universities to ascertain the range of a tuition fees students face per year in BA, BCom, BSc, LLB, and BEng degrees.[2]

Annual tuition fees ranged from R32 560 to R68 135. In 2020 and 2021 universities applied a 5.4% and 4.7% tuition fee increase respectively, lifting the range to between R35 931 and R75 190 in 2021. Setting the allowance for transport, living costs, books, and personal care equal to the 2021 NSFAS allowance of up to R30 600 and assuming accommodation costs of R35 000 for ten months, means the total cost of studying at a university will range between R101 531 and R140 790 per year.

If this was the cost for the first year of study, allowing for further tuition fee increases of 4.7% per year for a second (2022) and third (2023) year, and 4% inflation for all other costs, the total cost over three years with a degree obtained at the end of 2023, will range between R317 716 and R441 113, to be repaid over 10 to 30 years. Note that this cost is the same order of magnitude as the R376 500 current retail price of a Corolla 1.2T Xs, a mid-size family car typically bought by middle-class (including graduate) families. The car, though, is repaid over only five years.

Given the limits on government finances, even to fund all income-contingent loans, there is a need for significant private-sector involvement (banks, pension funds) in funding the loan scheme. This could be done by the banks on their own books, whereby each bank commits an amount, or the banks and other investors can create a special purpose vehicle to which each institution commits a share of the capital needed to fund the loans.  

If 300 000 students each incur a loan averaging R120 000 per year, the cost would be R36 billion per year (and at a GDP of R5 trillion, be 0.7% of GDP), a quite feasible amount when combining the resources of government and the private sector. Universities are institutions that affect social change and are drivers of economic growth. Both the public and private sectors are key beneficiaries of the output of universities. Hence a solution towards sustainable student finance will need to involve an appropriate public-private-university partnership (PPUP). 

The loan scheme can include a sliding-scale of interest paid on the income-contingent loans, based on the student’s household income, coupled with a partial or full underwriting of the loan by government.

Commercial banks can administer the loan scheme, as they already have well-developed financial vetting systems and expertise (even if they channel all funding to an SPV to which all banks contribute a share of the capital – in a sense they will be acting as an office of the SPV). To reduce the risk of non-repayment, and because the loan repayment is linked to a worker’s income level, the South African Revenue Service can collect instalments and pay them over to the loan scheme.

There are, however, several obstacles that could derail the successful implementation of an income-contingent loan scheme and will require careful planning and analysis. By design, income-contingent loans will cater for the missing middle. Unlike traditional student loans from banks where a parent would put up collateral, missing-middle households are likely to be asset-poor, with little or no collateral. Thus, loans to students from missing-middle households will be unsecured loans. This calls for risk assessment and management practices that will minimise risk in the absence of collateral. Having SARS administering the repayment is one method that can lower such risk. Another method would be selection criteria based on a potential students’ school performance and for senior students, their academic progress at university.

A further problem an income-contingent loan scheme faces is the long lead time till repayment starts. A bachelor’s degree takes a minimum of three years to complete (but typically students take longer), whereafter the student enters the labour market. However, in general, entry-level salaries are relatively low, which means that it may take a few years before the graduate exceeds the income threshold at which repayment starts. And since the loan is income contingent, instalments might initially be small, leaving the bulk of the loan to accumulate interest to be paid in later years. Therefore, lending institutions will be required to see income-contingent loans as medium- to long-term investments.

To deal with interest payments that accumulate at compounded rates to later years of repayment and thus, to ensure affordability, there is a need to keep the repayment burden on ex-students manageable. As such, the government might have to partially subsidise the interest burden on these loans. So what degree of subsidisation can the government afford? This is an empirical and political question that depends on the fiscal space available to government as well as the prioritisation of its objectives, subject to its overall budget constraint. Financing and subsidising students is but one of government’s objectives – healthcare, basic education, social security, housing, among others, are all making claims on government funds.

Another problem is the risk of low total repayments. Although the rate of return on tertiary education is relatively high in South Africa, South African universities also see high dropout rates.[3] Students who leave university with a degree will likely join the middle class and earn salaries that will ultimately enable them to repay the loans. However, students who drop out might join the ranks of the unemployed, or, if they secure a job, will likely earn much lower incomes than graduates. As a result, they will repay little to nothing on their loans as their income levels are likely to remain below the threshold where repayment starts. Furthermore, the extent to which higher-income graduates’ repayments can cross subsidise ex-students who dropped out will be limited. Again, lenders might be unwilling to join an income-contingent loan scheme unless the government takes on some of the risk of low repayment rates. Again, the question is to what extent the government can afford such backstopping. This is an empirical and political question; the answer depends on the fiscal space available to government, as well as its prioritisation of objectives.

One method to reduce the risk for banks and other lenders is to securitise student loans. Banks can sell the income-contingent loans to a special purpose vehicle (SPV) they set up, which in turn sells securities backed by the loans to investors. In essence this means that a number of student loans are packaged together and then sold by the bank to an investor willing to carry the risk. By pooling loans into a single security, risk is diversified. And selling it to investors ensures that risk is carried by a party (i.e. investors) willing to do so. It also transforms illiquid assets in the form of income-contingent loans into liquid assets (the securities sold by the SPV). Banks in South Africa have been active in securitised markets for many years, securitising, among other, motor vehicle loans, home mortgages, and credit card debt. Thus, the expertise to securitise student loans exists.

The obstacles listed above that need to be overcome (e.g. long lead times, the need to subsidise interest payments so as to keep loans affordable, and the effect of high dropout rates lowering repayments) all require that government spends money to ensure the participation of banks and other funders. Such a scheme is a partnership involving government, the private sector and universities – a public-private-university partnership. Thus, not only do we need government finance, but also private sector investment. For this to work the private sector should realise that even though a student loan system inevitably involves risk, it is in the interest of its own long-term growth and profitability to fund such loans. A larger pool of graduates will ultimately bring more competition among skilled workseekers, making salaries more affordable to companies.

Government also needs to realise that higher education is both a private and public good. To contribute a component to student finance is an investment in the future, not merely an expenditure.





[1] This article is an expanded version of an article by Burger and Francis Petersen that appeared in Business Day of 8 April 2021.



[3] Van Broekhuizen, Van den Berg and Hofmeyr (2016) (also see Spaull) show that for every 100 learners who enter school, only 12 access university. Of those 12 only six complete some type of undergraduate university qualification within six years after matric, while only four obtain a bachelor’s degree within six years after matric. (For their full articles, see: and

Download article

PDF icon Burger, P_StudentFunding_July21.pdf
Post a commentary

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. To comment one must be registered and logged in.

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. Please view "Submitting a commentary" for more information.