Across August and September 2024, The Australian Government Treasury ran a consultation to assist with revitalising National Competition Policy. Members of the wider community were invited to make submissions pertaining to competition policy in Australia. This is the 1st of my two submissions. The 2nd is here.
I studied physics and maths at university and progressed into a career in software development. Across the last 27 years I have worked as CTO and then managing director of companies specialising in the development of brokerage and exchange trading platforms. Recently my interest has been drawn to ‘productivity’ and I developed the concept of a Productivity Exchange (https://prodxchg.com). My company, Xilytix (https://xilytix.com), is creating the productivity technology needed by such an exchange.
To build up my knowledge of productivity (and improve my writing skills), I write articles in areas of productivity that interest me. The intersection of knowledge and productivity is one of those areas of interest. This submission combines this interest with competition policy.
There is a perception1 that Australian universities have a business model focused on exporting education. They use their research reputation and ranking to compete with other universities to attract overseas full fee paying students. It is a very successful business model for universities. For several universities, it brings in over 30% of their revenue; in Sydney University, it brings in 44% of its revenue (2022)2. Income that universities can use to bolster their research and better compete in this market. Education is also is one of our top export industries. In 2022/23 it was the 4th largest export at $36.4 billion3.
However this industry exports skills. International students leave with skills supplied by Australian institutes. Skills that are highly valued in other countries. Skills that drive productivity. Considering that Australia (like many other countries) has a large problem with falling productivity4, this poses the following 2 questions:
This submission explores some of the undesirable impacts of the competition environment in which Australian universities currently operate. It then presents a proposal for using a competitive market to incentivise the leadership of universities to improve domestic skill levels by improving teacher training.
The skills of people in a country are the largest determinant of the long term growth of its economy5. Especially skills in mathematics and science. Skills could be considered as a countries most valuable resource. A perishable resource which constantly needs to be regenerated and improved. A resource that drives productivity.
While people can attain skills in many ways, the skills obtained during primary and secondary schooling education are foundational6. They lay the basis for the further skills needed to develop vocation and careers and participate in the workforce. So, the extent of foundational skills learnt during these schooling years, has a large impact on the level of skilling that an economy can achieve. Accordingly, they have a bit impact on productivity.
There are many factors which affect how well students in primary and secondary schools learn skills (student outcomes). However evidence from literature indicates the most influential factor is teaching efficacy (quality)7. Teaching efficacy can be separated into teacher attributes and teaching practice - both of which are developed as part of teacher training programs.
In Australia, teacher training is carried out in Initial Teacher Education (ITE) courses run (mostly) by Australian Universities. From the Universities Australia report in 20228 I estimate that ITE courses approximately contribute 5% to universities’ overall business9. The proportion of revenue from international students in Go8 universities range from 15% to 47%10 and in 2020, approximately 28% of students were international11. From this, I estimate international students are approximately 30% of university business12.
For universities, it makes a lot of sense to focus on a business model around exporting education and compete in this space. It is already a large part of their business, there is large demand with excellent growth potential, the margins are good13 and it is not dependant on government funding. The only significant constraint is Australian Government immigration policy.
The same does not apply to ITE courses. Competing in this space would provide universities with far lower growth potential14 compared with competing in the international space. ITE most likely provides a reliable revenue stream, the profits of which could be used to compete in the international space.
The fact that universities compete in exporting skills rather than compete in developing local skill, can lead to some incentives that are detrimental to improving productivity in Australia.
The export revenue currently generated by university exports is likely to be highly valued by the Australian Government. Not only does it reduce Australia’s current account deficit, it also reduces the government funding required by universities. However sometimes it is in the interest of the Government to restrict or regulate certain economic activity - despite these activities generating substantial revenue for the state. For example:
The above criticisms of the current university competition environment serve little purpose unless an alternative can be made to work. If not, we may as well accept the benefits of the export income and rely on other fixes to our productivity problems.
One alternative is for universities to compete based on the effectiveness of the teachers they train. Measuring teacher effectiveness is a difficult task. Various methodologies, using different data sources such as test scores, classroom observations, surveys have been developed - each with advantages and limitations. To get an accurate assessment requires using multiple methodologies23. This would be a costly endeavour.
However we do not need to calculate the effectiveness of individual teachers. We are looking for an aggregate effectiveness for all teachers from each ITE provider which allows us to rank ITE providers. Is there a combination of methodologies which could provide this at an affordable cost?
One potential candidate is Value-added modelling (VAM). It uses complex algorithms to predict a student’s future test scores and then compares these against the actual scores. The difference is considered to be due to the teacher24. While this methodology is not suitable for ‘high stake’ evaluations of teachers25, with aggregation it may provide results sufficiently consistent to evaluate ITE courses.
Research would be needed to investigate whether teacher effectiveness evaluation methodologies could be used to measure performance of ITE courses. The application of machine learning/AI techniques may assist with this endeavour.
Assuming that that it is possible to score and rank ITE courses for the purpose of competition, how could those rankings be used to foster competition? It is unlikely that the reputational impact by itself would be sufficient to focus management’s mind on improving ITE courses. ITE is not a major part of the business and its growth prospect is small. Adjusting government funding per student in these courses would be counter productive as it impacts the quality of the courses. However the following financial adjustments may have a suitable impact:
Changing the competition environment in which universities operate would would have huge ramifications on our education system, research capabilities and even the wider economy. Before contemplating such a change, much discussion, consultation and research would be required to determine if this is feasible, and if so, what options and approaches could be used. It would also be necessary to weigh the potential long term productivity benefits against likely short term export losses.
Some initial research which investigates feasibility together with approaches and options would assist with such community and stakeholder engagement.
Recommendation 1
Undertake research to investigate the feasibility of using a competition model based around performance of teacher training to better focus university management on improving Australian skills and productivity. Also investigate alternative competition models or options which could achieve this purpose. The outcome of the research should be used to stimulate and guide community discussion on the use of competition policy to encourage greater university management focus on improving skills and productivity within Australia.
Skills are the largest driver of productivity in a country. Universities are a key part of Australia’s skills development infrastructure, with their ITE courses training teachers, who in turn, develop critical foundational skills in students. However it seems that universities operate in competitive environment which focuses university management on the export of skills - possibly at the expense of Australian productivity.
Competition policy could potentially be used to refocus university management on parts of their business which improve Australian productivity. Introducing competition to university ITE courses could possibly achieve this. Research into teacher effectiveness methodologies to enable this, and options for setting up such competition, would facilitate community discussions that can further explore this possibility.
© Paul Klink 2024: No AI Training26
Nico Louw: MRC Report: Breaking Universities’ Addiction to International Students ↩
ABC News: Group of Eight universities warn they are reliant on international student fees as government looks to cap enrolments ↩
Australian Government Department of Education: Education export income - Financial Year ↩
Australian Government Productivity Commission: Annual productivity bulletin 2024 David Taylor: Australians are working as hard as ever, but output is going backwards. Who’s to blame? ↩
Eric A. Hanushek, Ludger Woessmann, Ludwig Maximilian: Education and Economic Growth ↩
Australian Government Department of Employment and Workplace Relations: Draft National Foundation Skills Framework 2022 to 2032 ↩
Deloitte Access Economics: School quality in Australia: Exploring the drivers of student outcomes and the links to practice and school quality ↩
UNIVERSITIES AUSTRALIA: 2022 HIGHER EDUCATION FACTS AND FIGURES ↩
Using the rounded numbers of total revenue, number enrolled in ITE courses, and total resourcing per student in education, we can estimate that ITE courses are approximately 5% of university business. We can also estimate based on student numbers (ignores research business) as being around 6%. Let’s assume 5%. ↩
Maani Truu, Jake Evans: The true scale of Australia’s international student industry — in four charts ↩
From above Universities Australia report ↩
The simple average of the proportion of revenue from international students in Go8 universities range is approximately 30%. ↩
Assumption based on reports that international students subsidise teaching of domestic students
Sydney Morning Herald: International student numbers not easy to cut without dire results (paywalled) ↩
UNIVERSITIES AUSTRALIA: 2022 HIGHER EDUCATION FACTS AND FIGURES - Figure 51 ↩
aitsl: Understand initial teacher education program accreditation ↩
Glenn Fahey, Rob Joseph: Starting off on the wrong foot: How to improve Initial Teacher Education in Australia ↩
Bridie Smith: Teachers told to ignore Victoria’s phonics push as union lashes deputy premier (paywalled) ↩
Rachel M. Cohen: The new “science of reading” movement, explained ↩
Parliament of Australia: Universities Accord (Student Support and Other Measures) Bill 2024 ↩
National library of Medicine: The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General. ↩
Australian Government, Department of Climate Change, Energy, the Environment and Water: Gas Markets ↩
Brett Cohen, Armin Fazely: Domestic gas reservation policies – are they a good thing or a bad thing? ↩
Wikipedia: Value-added modeling ↩
Hannah Putman: Teaching that goes beyond the test? How to measure the many accomplishments of great teachers ↩
No AI Training
Without in any way limiting the copyright holder’s exclusive rights under copyright, any use of this publication to “train” generative artificial intelligence (AI) technologies to generate text is expressly prohibited. The author reserves all rights to license uses of this work for generative AI training and development of machine learning language models. ↩