Conversion Conversations

Alchemy: like getting a research grant,
only easier (image: Wellcome CC BY)

Last week Alex Hulkes, the Economic and Social Research Council’s strategic lead for the insights team, wrote an interesting report on the council’s conversion rate for grants. This is the percentage of fundable grants that it is able to fund. In November he provided a similar analysis of success rates. Taken together, the two reports suggest that while the quality of applications has increased (the proportion of ‘fundable’ applications has risen from roughly one-quarter to half of the total), the success rate is pretty much where it was five years ago.

I wrote about this phenomenon a couple of years ago, and it’s slightly disheartening that things haven’t changed in the intervening period. The most recent data for responsive-mode research grants suggest that the cut off for funding is still in the high eights, and even with Hulkes’ positive spin (‘if a proposal is fundable there is still a pretty good chance it will be funded’), one can’t help but think that demand management isn’t working. Applications have fallen marginally in five years (from around 1,800 to around 1,600), but the success rate has returned to the high teens for all grants, and is actually lower for responsive mode funding than it was five years ago.

To me, Hulkes’ report shows that the message has got through to universities that they need to increase the quality of their applications, but the implicit promise of a brighter, better-funded future has not materialised. This isn’t necessarily the fault of the ESRC: its budget has been effectively frozen for the past seven years, during which time academics are being increasingly pushed to apply for larger amounts of money.

However, I believe that two changes could make a huge difference to the council’s conversion rate statistics, and the waste of effort underlying them. They are to reintroduce funding for smaller projects and to introduce a two-step application process.

First, funding for smaller projects. Hulkes, in his success-rate report, points out that the average (mean) grant for all of the ESRC’s schemes has doubled in the past five years, from £200k to £400k last year. This was due in large part to a strategic push in its Delivery Plan 2011-15, to focus on “longer, larger grants that deliver ambitious social science”. The council cut its popular Small Grants scheme, and set the lower limit for responsive-mode funding to £200k, revising it again in July 2015 to £350k.

By doing so it hoped both to cut the administrative burden (small grants were notoriously bureaucratic in relation to the amount of funding they provided), but also to discourage all but the big hitters from applying. While the former has been achieved, the latter clearly hasn’t worked. Or perhaps it has: the increase in the proportion of fundable proposals suggests that applicants know what they’re doing, which implies that they are more seasoned investigators.

Although the ESRC does have some focused smaller-scale funding for schemes such as the Secondary Data Analysis Initiative, there is nothing for those seeking funding for projects that will cost less than £350k. This seems to be a real gap in the portfolio.

The popularity of the Small Grants, which was one of the reasons for the scheme’s downfall, suggests that they are what the social sciences want and need. Unpopular though it is to voice this, a lot of social science research doesn’t need huge funding. As the BA Small Grants have demonstrated, a little goes a long way.

If they were to reintroduce small grants, it should not be with the associated old, overloaded bureaucracy. There needs to be a light-touch early stage, but I think the whole of the ESRC’s portfolio would benefit from such a filtering of outlines.

Hulkes said: “It’s always going to be hard to tell a fundable proposal from one that is not fundable, without actually putting it through the system.” I’m not sure that this is true. The Leverhulme Trust, for instance, doesn’t have any trouble in assessing which of its outline bids is potentially fundable.

Moreover, Leverhulme aims to send 50 per cent of its outline applications to the full stage. Of those, it usually funds half. This chimes in with the same figures for unfundable, fundable and funded research at the ESRC. In other words the conversion rate for fundable proposals is about 50 per cent.

The ESRC is, then, running a de facto two-stage assessment process: it demands a full application from the outset, but essentially runs it through two filters: one to assess whether it’s fundable in principle, the other to assess whether it is fundable in practice.

It would be fairer and less costly for it to seek a simple 1,000 word summary (or similar) of an investigator’s plans, as Leverhulme does at its outline stage, and only invite those that are genuinely fundable to the second stage. There would, of course, still be frustration among those who made it to the second stage and were still turned away, but it would save a huge amount of time and effort among the 50 per cent who don’t have a hope.

Couple this with the reintroduction of the small grants and I believe that the effect on the council would be transformative. Not only would it be able to offer many more grants, but it would be nurturing a positive relationship with the disciplines it represents, and saving its academics from the frustration and subsequent disillusionment that has soured their relationship in recent years. Together the changes would overcome the growing problem of fundable but unfunded projects, and that would save both applicant, administrator and reviewer a huge amount of ultimately futile work.

This article first appeared in Funding Insight in February 2017 and is reproduced with kind permission of Research Professional. For more articles like this, visit www.researchprofessional.com