The year is 2007, the Office for Science and Innovation (OSI) has been abolished and a new Department for Innovation, Universities and Skills (DIUS) has been created. There are significant concerns about reductions in the STFC budget and a House of Commons report notes that “the STFC have recognised that it could have communicated its plans better”
So far, so familiar.
There is also an agreement, as part of the 2007 Comprehensive Spending Review that the Research Councils, under the umbrella of Research Councils UK (RCUK), will achieve back-office budget efficiencies including significantly reducing the cost of grant making. So, when UKRI recently announced a new time-based reduction for grant making, application to decision reduced to within 90 calendar days for applicant-led calls and within 60 calendar days for targeted calls, I thought: so far, so familiar. This is just another corner of the basic project management triangle of cost-time-quality constraints. Once again the wrong corner has been chosen.
Lessons Learned
I understand why the cost and speed of grant making are important but neither are a proxy for quality, and lessons from the 2007 RCUK Shared Service Centre experiment seem long forgotten.
In the early 2000s there was an explosion of shared services across the UK and the public sector that promised lower costs, standardised processes, economies of scale etc. At a policy level it looked compelling. But for those of us in operational teams based in Polaris House in Swindon, the home of the research councils, it looked anything but compelling. Not because operational teams hate change, but because operational teams want to do the best for the research and the research communities that they support, and the wrong measure of success had been put into play.
And the operational teams were right. By 2011, the National Audit Office published a damning report on the research councils’ shared service project “The project has cost £130 million against an original budget of £79 million, a 65 per cent overspend, and projected savings are significantly less than expected.”
The report is well worth a read but one thing that the report missed was the real human costs within the operational teams. Colleagues knew the academics, institutions, disciplinary norms and emerging areas of research, and they understood that the success of a grant-making system depends not only on process compliance but on trust, judgement and relationships. These relationships help reduce misunderstandings, solve problems before they escalate, improve applicant experience and build confidence in funding decisions. But in the shared service model this tacit knowledge ceased to be seen as an asset. Instead, many colleagues felt treated as just another interchangeable part of a process, no longer valued or part of a community. Many experienced colleagues just left.
If you reduce operational teams down to their cost, or indeed the time they take, then you are missing something essential to the grant making process.
The business case doesn’t add up
One of the things that the NAO report didn’t miss is that the business case for the Research Council Shared Service Centre fell short “in terms of identified success factors and lacked detail about the benefits to different stakeholders and how they would be delivered.” That observation feels remarkably relevant today.
UKRI has announced a significant reduction in grant-making times, but I have yet to see a clear articulation of the ‘why’. What problem will it solve, how will different stakeholders benefit from the change and what evidence suggests that a reduction to 60 or 90 calendar days will deliver meaningful benefits for each of them?
To be fair to Ian Chapman, I can understand why this issue has risen to the top of his agenda. He arrived as UKRI’s CEO during a period of significant operational change, with the transition from and old grants making system, Je-S, to a new one, The Funding Service (TFS). He joined at the cut-over period with teams simultaneously trying to run a functioning grants operation across both systems and then change to the new system. Unsurprisingly, researchers, university research offices and operational teams have all felt the effects of delays, uncertainty and the complexity of operating in two systems at once, including some burdensome and lengthy manual workarounds. This has created some of the longest grant-processing times I can remember with a median time for applicant-led funding of 194.5 days. Of course this is being raised as an issue. But raising out-of-the-ordinary grant making delays is not the same as asking the turn around to be as quick as possible.
I think a symptom is being mistaken for cause. Researchers and universities are rarely frustrated simply because a decision took more than six months. The frustration normally comes when processes feel unpredictable, when communication is poor, when systems don’t work as expected, or when they lose confidence in funding decisions. In other words, what people are reacting to is not time itself but their experience of the system.
Researchers typically plan their future funding needs 9-18 months in advance. Not because funding decisions take that long, but because funding outcomes are uncertain and there are competing demands on their time that require careful planning. The challenge for most researchers is finding the time to develop a strong proposal alongside their existing responsibilities. Reducing decision times to three months does not obviously deliver a meaningful benefit to researchers, particularly within university planning systems that operate on annual cycles and often have limited flexibility to mobilise new projects quickly.
Unintended consequences
As well, as the ‘why’ do you want to compress timelines so far, I also want to know how it is possible, and what the potential knock-on implications are?
Almost certainly, one solution being discussed is drastically reducing the number of applications UKRI receives by limiting the number that organisations or individuals can submit. This does nothing to reduce the overall time being spent on grant making within the system, it just moves the time and cost to individual research organisations with limited control on the quality.
Distributed Peer Review must also be being discussed. This effectively creates a closed system in which applicants become the reviewers for their fellow applicants within the call they’ve applied for. Requiring applicants to review as a condition of participation. It may generate efficiencies, but it does not create reviewer expertise or time. If anything, it formalises the transfer of time pressure onto those already competing for scarce resources. A recent UKRI call that used distributed peer review required applicants to submit up to 10 reviews within a four-week period. If it takes on average half a day to do a fair review, then that’s 25 per cent of someone’s time if they work full-time. But what if they work part-time or have obligations (teaching, marking, caring responsibilities) within that period? The rules are strict if you can’t complete the reviews you can’t apply or your application will be rejected.
UKRI must confront the reality that time is not distributed equally across the research ecosystem. Compressed timelines will not affect everyone in the same way. Those working part-time, balancing caring responsibilities, health conditions, or coordinating work around school holidays are already navigate a system designed around assumptions of continuous availability. Who gets to participate in these new conditions? Compressed timelines privilege those with greater flexibility and fewer competing demands.
And what about the operational teams? They are a source of expertise that shape our research system, that should not be measured by the speed of administrative process. This undervalues their work and removes one of the most important sources of insights in the system. What does their work feel like if it is valued by the time it takes to run a call rather than by the value that call creates by being designed effectively? Who leaves the organisation when speed is a primary metric?
And that is where the lessons from the Shared Service Centre become relevant. Back then, the focus was on reducing cost. Today, the focus is on reducing time. Both are entirely rational management objectives. Time matters. Cost matters. But neither is the purpose of grant making. Research councils do not exist to process applications quickly or cheaply. Research councils should exist to make good funding decisions and to maintain the confidence of the research communities they serve.
Before UKRI commits itself to another efficiency programme, it should be able to answer a simple question: what evidence is there that a 60- or 90-day decision cycle will improve the quality, fairness and effectiveness of research funding? If it cannot answer that question clearly, then we may once again be too focused on the process while losing sight of the purpose.
