Media attention has emphasised that the financial issues facing universities continue to worsen. While research is a cornerstone and strength of the sector, it is often regarded as a cost, which leads to scrutiny as part of institutional savings targets. Despite calls to acknowledge the value of research, the focus understandably remains on research costs.
The focus of universities on the volume and cost of unfunded research, or more accurately, internally funded research, is a question that must be addressed. Institutions are reflecting on and revising internal research allowances as part of their efforts to achieve a more sustainable financial position, as the cross-subsidy from international student fees is no longer as viable as it once was.
The question of funded research, however, is a different matter. For quite some time, there have been questions about what constitutes the full economic cost (FEC) and how these costs are recovered when projects are funded. Both issues have once again come to the forefront in the current climate, especially as institutions are failing to recover the eligible costs of funded projects.
As part of the Innovation & Research Caucus, an investment funded by UKRI, we have been investigating why the recovery of UKRI-funded research is often below the stated rates. To put it simply, if the official recovery rate is 80 per cent FEC, why is 80 per cent not being recovered on UKRI-funded projects?
Understanding under-recovery
We conducted a series of interviews with chief financial officers, pro vice chancellors for research, and directors of research services across mission groups, the Transparent Approach to Costing (TRAC) group, and various geographic regions. They identified several key reasons why universities are not recovering the funding to which they are entitled.
Before exploring the causes of under-recovery on UKRI-funded projects, the project aimed to establish the extent to which TRAC data was curated and utilised. Notably, the study found that the data collected for TRAC does not exist within research organisations and would not otherwise be collected in this form if it were not for the TRAC reporting requirement.
While scrutinising TRAC data was less of a priority when the financial situation was more stable, in many institutions, it is now of interest to the top table and serves as the basis for modelling, projections, and scenario planning. That said, such analysis did not always recognise TRAC’s limitations in terms of how it was compiled and, therefore, its comparability.
In many of the research organisations consulted, the responsibilities for TRAC, project costing, and project delivery are distinct. Given the growing significance of TRAC data in influencing resource allocation and strategic decision-making, it is essential for research organisations to adopt a more integrated approach to compiling and utilising TRAC data to achieve improved outcomes.
Drivers of under-recovery
A wide range of factors explains why the cost recovered at the end of a funding grant is less than anticipated at the point of submission and award. Almost all respondents highlighted three factors as significant in low cost recovery:
- Equipment and facilities costs were consistently cited as a factor, including issues associated with allocating and costing overheads and estates. Several institutions highlighted the difficulty in realistically costing equipment and facilities shared between research projects or between research projects and teaching.
- Staff under-costing was frequently mentioned, as principal investigators (PIs) underestimated their own and their colleagues’ time commitment to projects. This ineffective practice was driven by a (mis)perception that lower costs will likely improve success rates – despite the emphasis being on value rather than cost within a specific funding envelope.
- Inflation has been identified as a factor affecting all cost elements – from staff costs related to pay settlements and promotions to the rising expenses associated with consumables, equipment, and energy. This reveals a growing gap in applications, delivery, and reporting.
Beyond these top three, the report highlights the implications of the often “hidden” costs associated with supporting and administering UKRI grants, the perennial issues of match funding, and the often inevitable delays in starting and delivering projects – all of which add to the cost and increase the prospect of under-recovery.
In addition, an array of other contributing factors were also raised. These included the impact of exchange rates, eligibility criteria, the capital intensity of projects, cost recovery for partners, recruitment challenges, lack of contingency, and no cost extensions. While not pinpointing the importance of a single factor, the interplay and cumulative effect were considered to result in under-recovery.
Addressing under-recovery
Universities bear the cost of under-recovery, but funders and universities can take several actions to improve under-recovery – some of which are low- or no-cost, could be implemented in the short term, and would make a real difference.
Funders, such as UKRI, should provide clearer guidance for research organisations on how to cost facilities and equipment, as well as how to include these costs in research bids. Similarly, applicants and reviewers should receive clearer guidance regarding realistic expectations from PIs in leading projects, emphasising that value should be prioritised over cost. Another area that warrants clearer guidance is match funding, specifically for institutions regarding expectations and for reviewers on how match funding should be assessed. We are pleased to see that UKRI is already taking steps to address these points in its funding policies.
In the medium term, research funders could also review their approaches to indexation, which could help mitigate the impact of inflation in driving under-recovery, although this is, of course, not without cost. Another area worth exploring by both research organisations and funders is the provision of shared infrastructures and assets, both within and across institutions – again, a longer-term project.
We are already seeing institutions taking steps to manage and mitigate under-recovery, and there is scope to extend good practice. Perhaps the main challenge to improving cost recovery is better managing the link between project budgets – based on proposal costs – and project delivery costs. Ensuring a joined-up approach from project costing to reporting is important, but more important is developing a deeper understanding across these areas.
A final point is the need to ensure that academics vying for funding really understand the new realities of cost and recovery. This has not always been the case, and arguably still is not the case. These skills – from clarifying the importance of realistic staff costs to accurately costing the use of facilities to effectively managing project budgets – will help close the cost recovery gap.
The real FEC of research funding
The current project has focused on under-recovery in project delivery. The next step is to understand the real cost to research organisations of UKRI grant funding.
This means understanding the cost of developing, preparing and submitting a UKRI grant application – whether successful or not. It means understanding the costs associated with administering and reporting on a UKRI grant during and beyond the life of a project (think ResearchFish!).
For more information, please get in touch – or watch this space for further findings.
You can read the Innovation & Research Caucus report – Understanding low levels of FEC cost recovery on UKRI grants – as well as the UKRI policy changes referred to in the article.
In my experience there is a constant perception from PIs that overheads are excessive, and we regularly see comments about the excessive costs of admin. and management. The price of having HR, Finance, Estates, IT, etc. on tap never seems to be low enough.
The under-recovery of costs on funded research is about £3b. The cost of own-funded research is £2.8bn.
Improving cost recovery on funded research is challenging. As the article alludes there is often a cultural mis-perception that under-reporting costs will increase success rates, for example. Even industry sponsored research only recovers 77% of its costs, as providers under value and under price it. Changing that culture is not easy. But providers can more easily change the mix of research funding, focusing more on sponsors that generate higher cost recovery for example, but even that requires a strong degree of central policy intervention that many are reluctant to do.
However, the first part of the article mentions own-funded research. This is research effectively funded from non-research sources such as tuition fees. So a business school has far fewer opportunity to earn funded research but lots of opportunity earn premium tuition fees. Providers have huge scope to seriously challenge how much non-funded research is undertaken and whether this investment is worth it. A lot will have a high indirect ROI, but lots will not. It would seem natural to link any downturns in turion fee income to scaling back the amount of own-funded research undertaken. But I am willing to bet no one makes that connection let alone manage it. I am doubtful many providers even have visibility of those costs (despite that they are returned via TRAC(, let alone any grasp of their ROI.
If providers want to make a step change in research cost recovery, relatively easily and entirely within their own control, they should give some serious focus on one-funded activity and evaluate its returns.
Its disappointing that the article doesn’t acknowledge the role research councils have played in reducing cost recovery, by putting pressure on institutions to co-fund projects or grants and by challenging why fEC rates are above median, as if a normal distribution of costs must indicate inefficiency.
I completely agree that the pressure from UKRI to co-fund research through institutional contributions has ballooned in recent years and is underrecognised. There are instances, e.g. BBRC’s current call for Doctoral Focal Awards, where it is outright stated that the grant cannot coat for administrative support for the programme. Also, having sat on grant funding panels, investigators time is heavily scrutinised. For example, there is a perception that there is a “limit ” for PI time and if there are multiple coI than these should each be at a reduced level. Because of this, I disagree that the applicants bear the sole responsibility for under costing. This is an outcome of the system.
I think the panel behaviour is very discipline dependent. I worked for EPSRC for 20 years- more often on large grants the panel was querying why the PI time was so low….
While the cost of “unfunded” research is real, it is (for my department at least) fundamentally entangled with the seasonal nature of teaching, in a way that makes it non-trivial to cut, and arguably an inherent cost of teaching. We teach for 23 weeks of the year, and spend another maybe 6 weeks on exams etc. What substantial teaching work needs to be done outside termtime (e.g. curriculum development) occupies our minority of T-focussed staff year-round. Even allowing for 8 weeks annual leave (and in reality, we have trouble getting many of our T&R staff to take even their statutory minimum leave) that leaves 15 weeks when we wouldn’t have much meaningful non-research work for T&R staff to do. Short of a huge change to the structure of our degrees (not impossible but not clear that our students want this) or employing academics on US-style 9-month contracts (maybe fine if we also paid US-style salaries!), the choice is between letting them do unfunded research, paying them to do nothing, or inventing busy-work for them…..
Well, yes. But academics are not waiting for managers to give permission to ‘let them do unfunded research’. The notion that unfunded research is such a terrible sin that it could be better to be paid to do nothing is revealing of how far apart university leaderships’ and academics’ conceptions of what universities are actually for. If universities were to, as Undercover Accountant proposes, ‘seriously challenge how much non-funded research is undertaken and whether this investment is worth it’, they would discover that unfunded research is not a cost, but rather a subsidy to universities from staff. Most unfunded research – which in some subjects means most research full stop – is done, for free, in academics’ spare time (evenings, weekends, annual leave), on top of their day jobs of teaching preparation, student support, filling in forms, etc, etc. It is also typically done in a location other than their normal place of work, where university overheads don’t apply as costs are paid for by someone else, often from academics’ own pockets. The purpose of research is not to generate a surplus in order to fund endless layers of management to tell us how to do research (which experienced academics surely know already), it is to advance knowledge. Unfunded research is often a much more efficient way of producing research than wasting time applying for funding, which is itself a prime cause, amongst many others, of ‘invented busy-work’, another example being the scam accountancy trick known as ‘Full Economic Costing’. Any attempt to clamp down on unfunded research is doomed in an overmanaged sector with declining resources for its core activity, because the more that workloads for everything else increase, the more research becomes a spare time hobby that staff will rightly assert autonomy over.
As a researcher, who’s had multiple grants with different funders, my feeling (and that of many other researchers I speak to about this) is massive frustration with the amount of grants that is eaten up with overheads. This point rarely seems to be heard in commentary on this topic, which often reflects the views of university leaders and others, but rarely researchers themselves. I’m currently working o a UKRI grant for 2.5 year project with 2 co-Is and 1 post-doc, no expensive equipment, and half of the £600k budget is taken up with overheads.
Other models are available: Leverhulme (no overheads), ERC (20% contribution of value of grant to admin costs). My preference would be for grants to be funded at 100% of FEC but with no overheads; instead there could be some separate central pot to fund research infrastructure. The Australian Accord had some interesting ideas about this. Some will no doubt scoff that these ideas are impracticable and there’s no magic money tree etc. But given the current system is flawed, if not broken, is it not time in this case (and in others in HE) for some fresh thinking and new ideas beyond tinkering and patching up the status quo?
Overheads, or more precisely indirect costs such as the estate, research support services, IT, HR, finance, are costs. These costs have to be funded somehow.
Full economic costs = direct costs + indirect costs + margin for sustainability.
I think the current system is hopelessly complex, to the extent that most don’t understand it.
Senior leaders have ignored this important data for too long while times were good—about time they took it seriously. The financial tensions are clear: business schools generate surpluses through high SSRs and overseas PGT, while STEM subjects face higher costs due to lower SSRs and research intensity. Unfunded research isn’t managed at all in terms of volume, adding to the problem. It’s all about maintaining a balanced portfolio at the overall university level—but when that balance shifts, as it has now with the massive drop in international students, you need clear plans to respond.
Some of the comments above have pointed to supposedly excessive overheads, but these are real costs. Rather than cutting bid submissions to win grants, universities should be making the case to funders to pay for them properly. Cross-subsidies have always existed and probably always will, but they need to be managed carefully as part of the overall balance—not just left to happen by default.
Without strong leadership, a unifying narrative, and a clear sense of overall institutional purpose, things will only get worse. And crucially, leaders need the financial acumen to understand which levers they can pull to respond to balance shifts before it’s too late.
Yes, totally agree with that. Some HEIs gold plate their costings with the assumption that those that get funded will help balance the books, some vastly cut them back to try and get some funded at all.
100% of FEC with no overheads is an oxymoron, as full economic cost by definition includes all operating costs. If grants were funded at 100% DI costs, then there would be very little research conducted, as it would not be financially viable (as per the current Charity research recovery). Considering that overheads are too high is more due to lack of transparency and understanding of what the overhead recoveries support than the actual levels. If an academic has an EA for example, then the EA’s costs will, at least in part, be recovered in the overheads, but suggest to the academic that their grant overheads could be reduced by not having an EA and I suspect they would not be willing to cooperate.
Contrarian position (to the one taken here, and to the attitude of governments and uni management in general):
Research ought to be desirable. Uni managements should compete to get the best researchers. Institutions that succeed at doing the best research should be empowered to do more. If your 10 faculty are consistently producing above-average research, it is better for the UK if your faculty expands.
That means “cost recovery” is the wrong model. Research needs to be profitable. Funding at “80% of FEC” is idiotic, because it means any one piece of research is running at a 20% loss – we should fund at 120% of FEC, and expect the best unis to use the 20% to do even better at research.
I get a strong feeling of Deja Vu reading this. I believe it was Quentin Cooper of then Coopers & Lybrand who told a conference I was at 30 years ago that there wasn’t a university in the world that didn’t subsidise research in the interests of reputation and marginal income.
Plus ca change …
@Undercover Accountant: While agreeing that there are Indirect Costs that need to be paid, it is not clear how these are calculated. What guarantee is there that these are not being inflated? Also, some costs exist regardless of research activities. For example, the building lights and heating are kept on because of classes being taught. My understanding is that the formulas used to calculate such estates costs do not take this overlap into account, so in way the universities are charging multiple sources for the same costs. This is compounded by the opaque manner in which such costings are calculated by university accountants. One would think that this should be public knowledge but I have never seen the finance offices open up the books and explain the costing details in a transparent way. I would be interested if colleagues in other universities have had a different experience.
The costing process has become an overly complicated accounting exercise.
Your point about over-lapping costs (or sunk costs) isn’t quite right. fEC is a full costing process based on a static allocation of 100% of university costs to 100% of activities. So your buildings electricity bill, for example, will be crudely allocated between T and R time. However, the more granularity you want, the more costing precision you want, the more complex the costing process becomes, to the point it becomes even more impenetrable and spuriously accurate.
In my opinion the fEC aspect of TRAC has become a huge and overblown process to calculate annual indirect cost rates (ie inflation on last years rates). Because its a static whole-system calculation if you press down in one area (grow your medical school say) then, all other things being equal that sucks the distribution of costs from Law (say) to the medical school, while total costs may remain unchanged. So each year fEC results in net nil flows of indirect-cost allocations around the system.
It’s time fEC was reformed and simplified. Do the full calculation every 3 or 5-years to change/validate the base rates and uprate by a cost-index in between.
Undercover Accountant is correct on the allocation process and methodology. If the costs are not clear as to how they are calculated then this is a challenge for the finance team at each institution. Unfortunately in my experience there are often knowledge and understanding gaps within fEC teams that make suitable dissemination and explanations challenging to the non-financial community. In your specific estates example, this is not, or should not, be any overlap., and all calculations should be very transparent. If there are challenges to the size of the overhead costs, more focus should be directed to the overall operating budget of institutions and less on their derivation.
This conversation seems to have got stuck in the weeds of accounting practices, which supports my point that researchers own views often seem to be marginalised in these debates. I take the points made and agree that the system is overly complex and opaque. There must be a better way of getting more money to fund actual research and researchers time (so much of which is now eaten up with applying for grants) than this.