Flashback on a Cross-Country Survey on Education and Skills Development

An online cross-country study I conducted in 2014 and 2015 revealed that the cost of a programme, employability, and the influence of mentors were the key factors influencing the youth to choose particular university courses. Asked to reason from an objective position in terms of what they desire in a university, the respondents placed the highest value on:

• acquiring skills for self-employment (83%);
• the reputation of the university (76%);
• the technology and innovation opportunities offered by the university programme (74%); and
• recommendation by mentors (74%). Peer influence turned out to be the least factor, with a weighted mean score of 43%.

Almost a decade later, it is evident that these findings remain relevant as Kenyan students revise their degree choices through the Kenya Universities and Colleges Central Placement Service (KUCCPS). This is happening in the face of a new model of funding universities that lays bare the differentiated costs of university programmes and what they imply for the burden of government loans the students must bear.

The Conundrum of Complete Confession

Kenya has introduced a new university funding model, but the underpinning concept of differentiated unit cost per programme, introduced in 2014, has remained relevant as a benchmark for equity in higher education funding. Recent reaction from the public affirms that despite its dispassionate rationality, the concept of differentiated unit costs (DUC) has been less understood, if not totally misunderstood. The masses decry a massive increase in programme fees. In contrast, it is the mode of funding university education that has changed to embrace the full disclosure of the actual cost of running each programme and how this information should influence justice in allocating the maximum amount of scholarships and loans, differentiated by the programme and determined levels of student needs. This is a striking case we can refer to as the conundrum of complete confession and candid communication. There is a genuine temptation to be nostalgic about the old block funding model, where all university programmes seemed to share a similar cost, judging by the direct fees the students have been used to paying for more than three decades. This article demystifies the conundrum and uses simulated model results to put across a dispassionate critique of the radical shift.

Kenya is at a pivotal juncture in its journey to revolutionise the funding model for its public universities. Recent indications suggest that President William Ruto is resolute on a radical shift that is likely to score at either of two extremes: enhance quality and equity in university education for learners from diverse backgrounds or widen the existing quality and equity gaps with the disadvantaged in society bearing the brunt of a mismanaged transition. Read on to discover why.

Key Highlights

• Kenya has ushered in an extensive policy transformation in its approach towards funding public universities, increasing the allocation for universities by 57% (from KES 54 billion in 2022/23 to 84.6 billion in 2023/24) and transitioning from a block grant model to a student-centric and needs-based funding system. The increase in funding per student is from KES 152,000 to KES 208,000, an increase of 37% per student.
• The new model incorporates a diverse blend of grants, loans, and household contributions, closely aligning with the specific needs and circumstances of individual students.
• This radical shift in funding universities has left out private universities and focused solely on public universities, based on established programme-specific maximum differentiated unit costs (DUC) and a multi-variable means testing instrument that classifies students into four categories of need: vulnerable, extremely needy, needy, and less needy.
• A critical analysis of the proposed shift in funding model must be ventilated against the changing landscape of higher education in Kenya with less than 20% of the candidates (or 173,127 in 2022/23) qualifying to join public universities directly, cutting off a huge balance that needs to join Technical and Vocational Educations Training (TVET) centres. The cost of training TVET students will be KES 67,189 per year per trainee, down from KES 71,420 for the new intake of May 2023.

• Kenya is now witnessing numerical parity between private universities (30) and public universities (29), but the latter are the strongholds of high-cost STEM courses, such as Medicine and Engineering. Attaining superpower parity in terms of the more demanding courses is still a tall order for private universities.
• For the courses that public universities have declared higher fees than private universities, fierce competition is unavoidable, with private universities better placed to benefit in student numbers because of the uncertainty or apathy associated with high amounts of education loans.

The Mettle of a Means Testing Instrument

The University Funding Model envisaged in the DUC was expected to comprise 80% government funding, 10% student fees, and 10% university contribution through income-generating activities. However, the shortfall in government funding has been a key challenge, with contributions falling below 50% for government-sponsored students lately. A shift in the funding model is, therefore, imperative.

The government has exuded confidence in a means testing instrument, which is reportedly capable of deducing a student’s level of need based on various factors, such as household income, the nature of precious schools attended, the cost of the programme a student is going to take, and gender, among others. The instrument suggests that approximately 29% of the total student population in Kenya falls under the ‘vulnerable and extremely needy’ category. Additionally, 17% are classified as ‘needy’, while the remaining 54% are categorised as ‘less needy’. The neediest student category is supposed to receive 100% government scholarship as the least needy category gets allocated a higher share of loans (55%) with less in government scholarship (38%) and 7% to be met through household contribution. Justifiably, parents and candidates need solid assurance that this framework can dependably help discriminate the different levels of student needs – complete with integrity of the human agents taking action to deliver end-to-end transparency.

Cases like the ones in the early days of the Higher Education Loans Board (HELB), which featured many cases of the less needy students receiving more support in loans and the more needy ones receiving less and having to appeal through priests and Deans of Student, must not be allowed to repeat themselves. In my case, I received a loan of KES 40,000 and with no bursary (grant) for the first academic year at the University of Nairobi, The fact that I lost my father at the end of the first year and applied for the full loan of KES 42,000 and bursary for the second year, complete with a copy of the death certificate to prove I was more needy, only made me suffer more with an allocation of KES 25,000 and no bursary for that year. The influential people in society would call or walk to Anniversary Towers and get favours for their children or dependants, all at the expense of the helpless and deserving parties.

A shift to a needs-based and student-centred funding model has key implications. Needs-based funding refers to the allocation of resources according to the financial requirements of students, ensuring equity and access for all, irrespective of their socioeconomic status. Student-centred funding, on the other hand, allows funding to follow the student, rather than being tied to the institution. This encourages competition among universities, as they develop an incentive to attract and retain students. The new liberal approach in the educational enterprise has now allowed Kenyan universities to declare the cost of their programmes democratically, without exceeding the maximum DUC per programme. The cost declared per programme in a university is an all-inclusive education service package and must remain unchanged until further notice.

Demystifying the Differentiated Unit Cost Model

The maximum fees that can be levied annually for each public university programme have been meticulously calculated based on a ‘Differentiated Unit Cost’ model, a groundbreaking concept devised by Prof. Dr.-Ing. Francis Aduol, an esteemed geospatial engineer and my former lecturer. Navigating through the comprehensive 127-page Differentiated Unit Cost report can be a daunting task, but this became a necessity for me recently following the new directive to universities to declare the cost of their programmes in response to the newly proposed university funding model.

The DUC model considers a wide array of factors, including student workload, faculty remuneration, student-faculty ratio, and infrastructure and operation costs. The model has been benchmarked against international practices, providing valuable insights for its further evolution. The benchmarking conducted against international practices provides a robust foundation for the model’s development and refinement.

In a typical academic year, the baseline assumed from the benchmarking exercise amounted to 1,800 workload hours (see the European Credit Transfer System). The contact hours with a professor (say lecturers, whose average per capita pay in one academic year is F) was fixed at 40%, leaving 60% for self-study by the student. In Kenya’s case, it was observed that most STEM courses are allocated more hours than this standard reference, with 25% over and above the regular programme hours, R, reserved for industry-based learning. With 1,800 hours as the benchmark, the student workload factor, k, becomes 1.25R/(0.4×1800). For clinical and engineering studies, k is much higher than unity (1). Their student/lecturer ratio, r, is also lower than for the others, so as to ensure quality training with the instruments and laboratory facilities available. The case of the Medicine Programme at Maseno University and its appreciable student/lecturer ratio is worth mentioning here as a mark of quality, which the graduates of the programme already demonstrate in the marketplace. Agricultural and applied sciences follow and are ahead of social sciences and humanities on this score.

Recognising that students share infrastructure and operations for effective training, the DUC report also presented estimated ratios of infrastructure and operation costs to staff costs, I/S. Finally, the simulation used here to estimate the programme cost per student per academic year is f’ = k(F+F(I/S))/r. Based on experiences across Kenyan universities, the higher/upper constrained student/lecturer ratio, r, has been used to estimate the lowest programme cost while the lower limit has been used to estimate the maximum programme cost. The mean value of the two estimates has been used as the mid-point in each case. The table displayed has some model figures based on subject groups.

Against the mid-point value and the DUC, how various Kenyan universities have differentially priced their programmes can be better ventilated. The corresponding fees in private universities have been shown in the table with the United States International University in Africa as a reference point. The figures confirm that there will be fierce competition for students between public and private universities for some programmes because of the competitive pricing in the education market. This fierce competition is aggravated where public universities have posted much higher fees for programmes offered by private universities, the uncertainty or apathy associated with securing high amounts in education loans to finance training at public universities being a key contributor.

DUC and the Optimisation Challenge

The mathematical model, due to its veneer of complexity in the DUC report, may impress as an object of fascination for curious students. As a result, seasoned modellers may find it oversimplified, and even declare that it falls short of the advanced calculus required to guarantee optimality with time-varying cost parameters. The model’s existing formulae do not include optimisation functions, leading to potentially suboptimal resource allocation. The model does not currently account for the changing cost of living, which can impact the accuracy and relevance of funding allocations.

A robust formula for calculating the annual cost of a programme should take into account a myriad of factors. It needs to consider the broad spectrum of qualifications and experience of the teaching staff contributing to student training throughout the year. Each trainer has varying levels of time commitment, leading to differing per capita infrastructure costs. The ideal equation should, therefore, embrace more variables and account for the time value of money. This presents a practical instance of applying differential calculus to explore scenarios of minima and maxima.

The DUC model, despite its strengths, requires refinements to optimally serve the diverse needs of Kenyan public universities and their students. I look forward to being part of the review team, as a volunteer to whom modelling is a virtuous fascination, after all.

By Nashon Adero

The author is a geospatial and systems modelling expert, lecturer, author and youth mentor. He can be reached on nashon.adero@gmail.com.