Leaving Kent

Moving on from academic research into industrial research

I have recently handed in my resignation notice at Kent. I recall my former colleague (in the direct institutional sense) wrote a blog post when he left Kent, and I thought I would reflect on my own experiences in kind.

What was my job?

I was a Research Fellow at Kent. This is a fairly privileged academic job: I had a full-time open-ended contract with only research responsibilities. The dream. This meant I could lead my own research programmes, and apply for funding as a PI. Indeed, I did apply for funding and was successfully awarded a DASA grant to look at Rust on Morello for Embedded Systems. The draw back of this role is that I’m expected to find funding for most or all of my time, failure to do so would mean I quickly find myself redundant.

The dream job, but at the cost of job security.

Why am I leaving?

There are two major factors, the first is purely personal financial, the second environmental.

The house I live in has gone up in value, which is a boon in one sense and a curse in another. To buy my housemate out of his share I need an increasingly high wage to satisfy the bank about mortgage affordability. With the mortgage rate at about 6% the monthly repayments alone swallow a very large portion of my take-home pay. Energy, food, and services all going up by ~10-20% for food and services, and almost doubling for energy, in the last year add extra pressure.

The University of Kent, like many middling HE institutions in the UK, is starved for income. Student fees have stagnated for years, being stuck barely above the £9000 fees that were introduced in my first year of undergraduate back in 2012. Real-terms income to universities per-student has deteriorated, and wages along with it. My wage as a Research Fellow is tied indelibly to student income as it is locked to the wage of a Lecturer.

This sector-wide financial problem bites especially hard at Kent. The university is seeking second round of “automatic spine point increment” deferral until 2024, and is not implementing the per-point UCEA increases until September ’24 either. This in total defers about a 8% pay increase for a year, against a backdrop of about an 10% inflation rate.

The blind spot

I worry that there is a bit of a blind spot for existing staff within institutions, especially ones in the South East, to the pressures of housing vs. wage. In my area a modest 3-bedroom semi-detached house is around £350k. Mortgage affordability tests require a ~10% deposit (£35k) and an income-to-loan multiplier no greater than 4.5x (£70k/pa). Hitting this is currently only just in reach for a pair of academics, and looks very very hard if you try to factor in children as well. For me, my partner is doing a PhD, meaning our combined wage would not hit the £70k threshold. In the UK median wages are around £30k, and an academic is paid about £45k. The access to housing is fragile for the researchers getting permanent positions around now, and a few years head start makes all the difference.

Why do I call this a blind spot? Well even the potentially small head-start that’s held by existing faculty members makes a huge difference. Pre-pandemic the average 3-bed semi was around £300k, the deposit around £30k, and the income threshold at a more accessible £60k. 3 years has moved homeownership from ‘within reach’ to ‘maybe out-of-reach’ for a couple where one half is an academic. 3 more years might leave homeownership ‘solidly out-of-reach’ for people in a similar position.

Looking back to 2016 when a similar house was £250k, deposit £25k, income required £50k, and interest rates at 1.5% – and everything was a lot rosier.

Income has not kept up with inflation, and has nowhere near kept up with house price inflation in the last 10 years. This creates a new major barrier to people wanting to start an academic career. A new PhD graduate can either post-doc for 2-4 years, getting further behind the curve because of inflationary pressures, or they can give up and get an industrial role which permits them some stability.

Career progression

On a strongly related note… The DASA grant was a significant win pretty early on for me. It was awarded after about 6 months in-post, which is financially very good performance as far as the university is concerned. I applied for a spine-point uplift (a ~2% pay rise) on the basis of that success, but was declined and awarded a one-time 3% salary bonus – to be deferred until October 2024.

Broadly speaking, the criteria for career progression within my job was (perhaps callously) boiled down to “get a new investigator award”. Forging ahead with a big grant like this would be a lot of work, and apart from securing me a few more years of job security (admittedly laudable on its own) the university has failed to make it clear that such success is met with promotions.

Environment at Kent

In the REF results that were recently published, Computing at Kent scored very well on research outputs and impact, but suffered specifically on ‘Research Environment’. The Research Environment is a bit of a catch-all for how well the university supports its staff pursuing research aims. The number-one mechanism to set one university apart from others is funding for PhD students. PhD students make the world go round in academia, and the two-fold effect of higher research output and better environment is a double-whammy on REF outcomes.

As a Research Fellow I was able to supervise PhD students, but funding at Kent was severely lacking. With only 2 EPSRC scholarships and 1 GTA scholarship available for the whole department there is very little to go around. The department also ties its own hands, rewarding PIs who are successful with grants over some threshold of value to the university with an assured PhD scholarship.

This combination makes the ability to get a PhD student very challenging.

None of this even looks at the ethics of PhD funding in a climate of insane pressures on teaching staff (many PhD students are compelled to teach), and external financial pressures where the PhD stipend doesn’t cover costs nor does it allow students to make use of in-work benefits like free childcare. That’s a rant for a different time, though.


The University of Kent is going through its second serious organisational restructuring since 2019. The first was called Organising for Success (O4S), a euphemism for cutting per-School administrative and management functions back and coalescing resources into ‘Divisions’. The second that is ongoing is called Effectiveness and Efficiency (EE), a euphemism for cutting per-Division administrative and management functions back, and coalescing them into the centre.

These two projects have been a violent shake-up of university management and governance, and has left many business processes in disarray. It has also led to head-count loss through voluntary severance and redundancy to many people who worked in Professional Services within the institution. This head-count loss naturally has caused a significant loss of institutional knowledge, so many PS departments have been left picking up the pieces while the university re-learns how to carry out vital administrative functions. The loss of head-count has also pushed some administrative burden down onto academic staff, and especially with IT equipment has encouraged the centre to find one-size-fits-all policies that do not work well at least for Computing.

Reflecting on Stephen’s experience

As I said in the introduction, this article is really a response to Stephen’s article from when he left Kent. Something in Stephen’s post which resonates strongly with me is this remark:

The reasons for the move are primarily personal. Of course, that doesn’t mean they are unrelated to work… in academic jobs especially, work is personal.

Kent has seen some very noticeable attrition of staff recently (Stephen among them). Something very common in the emails we receive about our colleagues moving on to pastures-new is the phrase “leaving Kent for personal reasons”, and I don’t doubt that this is true. Something that I think comes from a good academic job is a desire to make it work. This might be a negative instinct, but in my opinion the people who forge a path as an academic are not normally doing so for financial gain, or for a good work/life balance. People become academics because in their hearts they believe in the power of research and of universities. When staff quit Kent for personal reasons I can’t help shake a feeling that these reasons are the polite way of saying that they couldn’t make it work at Kent. The factors outside of work that are pulling people away are now greatly outweighing the factors that make people want to stay.

Stephen’s blog is what really made this fact sink home for me, in fact. It is rare you get to have an exit interview with your colleagues when they move on, and there’s an air of mild unprofessionalism about ‘airing the dirty laundry’. But, if I wanted to be consummately professional then I wouldn’t be an academic in the first place, and I abhor the process of obscuring information behind professional facade because it is what is expected.

Two years later

Since Stephen wrote his post we’ve had the EE re-organisation, and a sentiment from the upper management of the university that they wish to pivot away from being research focussed to teaching focussed. In Computing we are still considered to be a research active part of the university, but we’ve also been recognised as a profitable teaching centre. Student recruitment targets have been raised, with an ambition to raise them again and again to dizzying numbers in subsequent years. This has not come with a significant effort or promise for recruiting to offset the teaching load that extra students bring. Indeed, a significant amount of teaching effort comes from PhDs who lead classes (GTAs), and reductions in PhD scholarships also has left a reduction in the GTAs available to lead seminars and mark coursework.

While I do not teach, I work with and care about my compatriots who do. Being a member of the academy allows me to build research ideas with my colleagues in a way where the sum is (hopefully) greater than the total of the individual parts. But that summation is not possible when fellow academics are snowed-under by non-research duties, or even simply burnt out even when their calendars do open up.

Reasons to be hopeful

While I’ve painted a picture of gloom, I think there are still some reasons to be optimistic about the future. For starters, providing a research-only role with a career path all the way to professor is unusual and to be commended. It’s a pity the path hasn’t worked out for me, but the reasons for that are largely out of the control of the university and won’t reflect everyone’s experience.

The whole sector is in a bad spot, and Kent is just about pulling through regardless. Outside of innovative post-92 institutions and the Russel Group there’s a big middle of “Just About Managing” Universities. In 2019 30% of universities were being run at a deficit (Kent not among them), in 2023 I suspect this proportion is far larger. The politics of 30% or more the entire HE sector collapse is clearly negative, and a change of government feels like it is not far out now.

The Director of Division does seem to get it. He’s fairly powerless to hold back the tide, but within his capability it does seem he’s pushing for the best for Computing. We were awarded the largest portion of PhD scholarships in the Division, and he appears to recognise the importance research has to the existing staff and is looking for that same research-drive in new staff that are being recruited.

The university as a whole sort of gets it too. Computing has just been moved to a more modern larger building. This is, of course, largely for marketing purposes to attract new undergraduate students, but the building is nice and it does improve the conditions everyone works in.

What next?

Leaving Kent is bitter-sweet: on one hand I’m giving up a dream job because of pressures (largely) outside the control of myself or my institution. On the other hand, the job I’m moving into is very exciting and I can continue to consider myself an academic, albeit industrial. I am going to be a Research Scientist working at NVIDIA in the Architecture Research Group of NVResearch. At first I will be continuing the work I did as an intern there in 2018, and working on their memory consistency model. All going well I’ll continue to participate in the ISO C++ standardisation process, and I will certainly continue to be an active researcher.