EPSRC — the grants, the data, the diversity
The Engineering and Physical Sciences Research Council (EPSRC) recently published a report on the gender balance of their grant portfolio. It presented a stark picture of under-funding of women scientists. In this post, we’ll summarise some of the findings of the report and discuss their implications. First though, it’s important to point out what isn’t in the report. The report acknowledges that “Currently UKRI collects gender data through the Je-S portal, across three options: Male, Female and Not Disclosed. While we recognise that a broader range of gender identities exist, we used our Je-S data for the analysis for this report.” Hence, the experiences of non-binary people aren’t acknowledged in the data. Furthermore, not all women experience our current academic systems in the same way. It’s tough to navigate the physical sciences as a straight, white, abled woman. It’s even tougher for women who are Black or of other minority ethnicities, disabled or LGBTQ+. The available data don’t address these differences, and true equity demands that these issues of intersectionality are brought to the forefront.
Nonetheless, the report is a big step forward in terms of analysis of EPSRC diversity data. The EPSRC have taken responsibility for analysing the data, not just reporting it. This is in stark contrast to the UKRI data release earlier in the year — where there is a lot of information, but almost no analysis or reflection. It seeks to reveal where the problems are not, hide them and has enough depth and detail to let us think about the origins of these problems. So let’s have a look at what the report does tell us.
1. Women are under-represented in EPSRC’s Principal Investigator applicant pool — and the under-representation is more marked at higher grant values.
The graph below shows how the proportion of women applicants to EPSRC at various grant values compares to the HESA data for the proportion of physical/engineering scientists that are women — about 18% of the academic population. For small grants up to about 15% of applications may be submitted by women, but for the largest grants (in excess of £10M) only ~6% of applicants are women.
2. If women apply to EPSRC for large grants, they are less likely than men to be successful. This is masked by data which considers success rates only considering numbers of grants, not their value.
The graph below, taken directly from the report, shows success rates for women and men diverge for grant values in excess of £2.5M, with men being more than twice as likely to be successful when applying for grants in excess of £10M. This problem is not reducing as time goes on. Calculating award rate (i.e. success rate) by value, there was little difference between men and women in 2007–2009, but women have experienced lower award rates by value than men in almost every year since.
3. Even when women do apply for large grants, the amounts they are applying for are not as large as the amounts that men apply for.
The table below shows the largest grant apple for by a man and by a woman in each of the last 12 years. The values for men are higher than the values for women in every year.
4. EPSRC claim that there are “No noteworthy differences by gender in costs added by applicants” — i.e. men and women apply for the same things. But they also say women are A FACTOR OF 3 less likely to apply for funds for new equipment.
5. The EPSRC’s analysis of the salaries which applicants request on their grants is a very effective illustration of the gender pay gap.
Using age as a proxy for career stage, we see that men get paid more than women at similar career stages, and this effect increases with seniority level.
We’ll discuss each of these key findings in a bit more detail.
Findings 1 and 3
The first key finding is that women are less likely to apply for grants than men, and are particularly less likely to apply for large grants. This is also reflected in the third key finding. It would be easy here to shrug our shoulders and blame women for a lack of ambition and confidence. Instead we must ask what hidden mechanisms restrict the pool of applicants? What is it about the academic system that prevents or discourages women from applying?
TIGERS have written a review of the barriers women face when applying to grants, many of which are relevant here. The review suggests women are less likely to be encouraged to apply for funding by their line managers or research deans. It also provides ample evidence that women carry greater administrative and teaching loads than men, reducing the time they have available to apply for grants. This can foster a vicious cycle, according to Prof Carole Mundell at Bath: “The knock-on effect of not winning funding can also lead to an increased teaching load on women in male-dominated departments where men are winning grants and buying themselves out of teaching and/or admin duties, thereby further aggravating the problem.”
Another major concern is that many large grant processes involve some form of internal triage — a need to get internal approval from your University before you have the opportunity to put forward an application. We know of no data assessing how women fare in this process. Prof Athene Donald, Master of Churchill College, Cambridge says “Based on my experience in university politics, I can see only too easily how the man may look more forceful and command more favour than the woman, regardless of intrinsic merit of either the specific application or the particular woman who wants to lead it. Bias may well creep in at this point, unconscious though it may be.” Differences between men’s and women’s track records, which reflect personal circumstances rather than actual ability may feed into this problem, since women with caring responsibilities may be less likely to, e.g., be able to travel to give prestigious invited talks. Unfortunately, track records may also be affected by straightforward bias as well — whether conscious or unconscious, which diminishes the opportunities available to women.
Funding bodies and Universities need to consider how to break down the barriers which prevent women from applying for large grants, to give women the opportunities they deserve and to raise up a generation of women leaders who will be the role models for the next generation.
The second key finding in the report is that women have lower success rates than men when applying for high value grants. In many ways, what’s striking in the data is not the low success rate for women applying for very large amounts of money, but the very high success rate for men. For example, roughly three quarters of men applying for >£10M get awarded the money they asked for. That compares to a typical ~30% success rate for lower grant values, with some recent responsive mode rounds having success rates <20%. Are these very high success rates for men applying for very large grants really consistent with a picture of open competition, where everyone in the community has equal opportunity to apply for large sums of money?
Let’s have a look at an example of a grant greater than £10M: EP/R00661X/1 — awarded to Prof Phil Withers and one of several large grants awarded to establish the Sir Henry Royce Institute — “The UK’s National Institute for Advanced Materials Research and Innovation”. The grant has a male PI and nine men as co-investigators. To the best of our knowledge all of these men are white. All of them lead or have led “Royce” activity in their Institutes — a position of considerable power and influence. We can find no record of an open competition for this funding and grants on the web does not give any details for any panel at which applications for this funding were discussed. The decision making process around founding and funding the Sir Henry Royce Institute was a political one — it was announced by George Osborne in his Autumn Statement in 2014, who sited the Institute in Manchester as part of the “Northern Powerhouse”. This illustrates the political complexities involved in the decisions to award these large sums of money, and the fact that EPSRC and more broadly UKRI may not always have the power to implement equity policies.
Award rate data for the £10–200M range may be complicated by the inclusion of national institutes. However, at the £5–10M range, roughly half of men are awarded the money they asked for, but only a third of women. These grants span the EPSRC portfolio. EPSRC strategy funnels a substantial amount of funding into “longer larger” grants, when the data shows stark differences in outcome between men and women. This raises difficult questions about whether the proportion of the budget spent on large grants should be reduced to allow more funding to be channeled into lower value grant processes with more equitable outcomes.
The report reveals that while women apply for smaller grants than men at every level, there are few clear patterns in what specific things they ask for less of — except for one point: women ask for less money for new equipment. There has been no investigation of the reasons for this yet, but we can make a hypothesis, based on one specific feature of the EPSRC application process: the EPSRC currently only fund 50% of most equipment requests, with Universities having to make up the other 50%. Given we regularly see women complaining of institutional gatekeeping and a lack of university support for their applications, we hypothesise that getting a promise of funding for 50% of a large piece of equipment is more difficult for women than men. Women within TIGERS report that they either don’t succeed in obtaining such a promise or have stopped bothering to even try.
Because equipment funding is difficult to access, women have to think round the problem, and find other ways to get their research done — e.g. accessing shared equipment, or building collaborations. This suggests that when women do get grants, they have to work harder to get their research done, using the lower budgets available to them, whilst still stuck with more teaching, administration etc than men. This will have long term career impacts.
Of course this is only a hypothesis, and one which we hope will receive further investigation.
The final finding which we will discuss is that the salaries requested by men are higher than those requested by women, and this gets more marked with the age of the applicant. The report is not entirely clear on this point. We think it is referring to the salary rates which PIs request for their own contribution to the research project. The numbers in the report include the pension and oncosts in these salary rates. Assuming we’re interpreting this correctly, it’s important to realise that the salaries PIs request on their grants are not reflective of any kind of self-evaluation. These are the salaries set by the Institutions who employ the PIs, who are paying men more than women of a similar age. This is reflective of what is already known about the gender paygap in Universities: according to UCU the mean gender paygap in Universities in 2019 was a whopping 15.1%.
Although these data are basically just a presentation of the known information on the gender paygap, there is a link between pay and grant success. Evidence of winning funding is often required to get a promotion or pay rise, and if women experience greater challenges than men in winning funding, they will stay stuck at low pay grades. This is just one aspect of the broad range of impacts that inequalities in funding have on women’s careers, opportunities and scientific work.
Overall, the report tells us that women are less likely to apply for grants across the board, and are also less likely to be successful when they apply for large grants. How will this impact women’s careers and the research ecosystem?
We know that in research — just as in pretty much all else in life- success breeds success. This report gives evidence of systematic biases and barriers which will inevitably have a huge impact on the careers of women in science. Research is the one key metric needed to climb the Academic ladder, and lack of access to grant funding is keeping women out of senior roles. We need more visible women at the top of our profession if we are serious about tackling lack of representation.
What used to be ‘anecdotal’ evidence is now backed up by numbers: men are awarded more, women less. It would be lazy to think this is due to lack of ambition. Women go through a painful process of learning that teaches them that in asking for more, they are just reducing their chances to get funded. Systems designed by men for men, and the “old boys network” conspire to keep women ‘in their place’. By ‘asking for less’, women effectively work more over their careers for less money overall, just to keep their research going. The cumulative effect in a career is huge. We must also consider the psychological strain of engaging with institutional demand management processes in which women are often confronted with opposition not support; of a constant need to prove your worth and compete with men who have more privilege and access to resources.
Our current research funding system, of very low success rates and intense competition, breeds a culture where desperation can prevail over fairness, and where the opportunity to a short cut route to the large probability of a high value grant is never going to be turned down. What we need is major systemic change: Political will to properly fund the research ecosystem; leadership committed to equity and a recognition that policies and processes which ensure fairness and transparency are key to achieving the most from our talented researchers.
With the system as it stands, biased and flawed, everyone loses out, not just women and other marginalised folk who suffer directly from the in-built unfairness. Research is poorer because of these inequities. Innovations go unrealised, great ideas remain on the drawing board. The potential impact of the ingenuity of thousands of unique minds is lost. This report on gender only represents the tip of the iceberg in uncovering the impact of our funding systems on the diversity of our scientific community, and issues affecting racial minorities, disabled people and LGBTQ+ people are far more profound.
EPSRC are at the very beginning of taking action — perhaps not to achieve the kind of major systemic change we envision to address the broad swathe of equity issues, but at least to tackle the specific problems facing women applying for large grants. A consultation has been launched to help try and understand the data in the report better. We’d strongly encourage everyone, of all genders, to fill it in, and to amplify it to your networks. It will be available until 3rd December. Here’s the link: https://www.surveymonkey.co.uk/r/KGN98VK