Transparency in Higher Education Expenditure

Transparency in Higher Education Expenditure

With Covid-19 striking the university sector hard (and suddenly) last year, the Commonwealth Government postponed the need for the university sector to provide their 2019 ‘Transparency in Higher Education Expenditure’ submission.  However, despite the ongoing impact to the sector in 2021, the decision was made to not only ask for the 2019 data, but also the 2020 data as well.  The deadline to provide the 2019 data was extended to early August, whilst the date hasn’t been set for the 2020 submission yet – other than it will be due within 6 weeks of the templates being provided to each institution.

So, with most universities now heavily involved with their annual budgeting process, the requirement to undertake this exercise in the 4th quarter of the year is a distraction that many institutions just don’t need.

But for 11 Australian universities, the pain associated with producing the complex data set required to populate the templates has been significantly reduced.


All 11 institutions have a detailed cost model that was easily tweaked to incorporate the submission’s requirements and automate the population of the Deloitte template.  As their models already contained 99% of the data required, this was a free service that Pilbara Group provided (and continues to do so).

What had to be changed?

There were two small Deloitte-specific data elements missing from the cost models:

  • the required categories for salary and non-salary expenditure; and
  • the required FOE ‘slices’, which were at a variety of 2-digit, 4-digit and 6-digit levels.

Both specific requirements were easy to incorporate into the model – the first just required each expense to be tagged with the appropriate ‘Deloitte expenditure’ category (at an account code level), whilst the second just required each 6-digit FoE (already held against each Subject in the model) to be tagged with the ‘Deloitte FoE’ category.

The only other thing that needed to be done was to create a separate reporting cube to hold the data required for the Deloitte template.  As the standard reporting cubes could contain dozens of dimensions and measures, and millions of rows of data, a separate mini-cube was created for each institution that only contained the data elements needed to populate the Deloitte spreadsheet, plus Subject and Degree name (to allow for detailed analysis / cost investigation if necessary).

The annual process

As each annual cost model is developed, the additional reporting dimensions and measures needed for the submission are automatically loaded into the model.  The Deloitte spreadsheet is augmented with two additional tabs that link directly to the reporting cubes – one reports on EFTSL and the other contains a detailed pivot that shows each of the reporting cost elements (Casual Salary, Other Non-Salary Expenses etc) by each of the reporting Survey FoEs (0101 Mathematical Science, 02 Information Technology, etc).  Assuming the Deloitte template doesn’t change the cost elements or FoEs year-to-year, the process to update the template takes around 1 minute – the time it takes to refresh these pivot tables to view the new ‘year’.  And if Deloitte changes the required ‘buckets’, it’s a simple re-mapping exercise that would take less than 30 minutes to complete.  Likewise, if Deloitte’s breakdown of FoEs change, it’s an even easier re-mapping process.

From here, depending on each university, further reconciliation tasks may need to be undertaken:

  • ensuring that the ‘survey’ dollars in the model equate to the latest Annual Report (this may require gains/losses in the FY to be reported as a negative expense rather than a revenue);
  • ensuring that the model EFTSL reconciles to that prepopulated in the template by Deloitte by career level – this may involve reviewing the career level associated with each degree as held in the institution’s student systems compared to that reported to the Commonwealth as part of their EFTSL reporting framework (i.e. Cross-Institutional courses etc); or
  • adding in or removing costs that are traditionally included or excluded from their model but are allowed / not allowed in the submission.

The added bonus!

As each of these institutions use a similar methodology to develop their respective cost models, it is then reasonably easy to benchmark their data.  These benchmarks are sent out a few weeks prior to the Deloitte submission deadline so that institutions have time to review their results compared to the other 10 universities and query anything that doesn’t look quite right.  The benchmarking is conducted each year, thereby allowing year-on-year trends to be reviewed as required.

What’s Next?

The Commonwealth Grant Scheme (CGS) funding for Commonwealth supported students doesn’t just cover teaching and scholarship, it may also cover activities such as internally funded research and community engagement.  Research can potentially be a significant ‘consumer’ of these funds as academic workload models allow internal or departmentally funded research to take place in addition to any externally or grant funded research.

It is therefore logical that the next step in DET’s costing exercise is to ask universities to provide the ‘Cost of Research’ in addition to teaching at some point.  Will this occur in 2022?  Other than some rumours that it’s ‘being discussed’, nothing formal has come out, so my bet is that it is at least a year away, if not two years.

So, what could this involve?

  • Will there be a need to cost the internal research effort into multiple Field of Research (FoR) buckets like is being done with Teaching (currently 22 FoE ‘buckets’).
  • Will overheads need to be included? Like teaching, almost definitely.
  • Will any externally funded research need to be included? Maybe as a balancing item (i.e. not at a FoR level), especially noting that it too will ‘consume’ overheads.
  • Will any other buckets be included? Possibly…in my view, it would make sense to ask universities to reconcile back to their operating statements, so to do that Teaching, Research (both internally and externally funded), and any commercial (i.e. student accommodation) or out-of-scope costs (gain/loss on sale of assets) would need to be accounted for.

This could lead to some significant impacts on the current methodologies being used to answer the ‘Transparency in Higher Education Expenditure’:

  • Some universities may realise that they’ve been overloading teaching with overheads, thereby misjudging the cost to deliver subjects and programs.
  • Likewise, others may realise the opposite – the remaining ‘central overheads’ to be distributed over research may be too high, thereby currently underestimating the cost of teaching.

Both issues would therefore alter the ‘trend’ of the cost of teaching that the Department is currently obtaining…a good reason to incorporate research into this exercise sooner rather than later.

Turbo-Charging Insights

So what could we uncover if we combined and then analysed the detailed data from these 11 institutions?  Between them, they cover 28 percent of the Australian university student population and teach over 34,000 subjects across 174 campuses…what insights would we see by FoE or by location?

Stand-by for the soon-to-be-released report from the University of Melbourne Centre for the Study of Higher Education!